CUMPING
MACHINERY
BARR
B 429023
TJ
900
B27
PUMPING MACHINERY
BARR
J.B.LIPPINCOTT CO.
ཀ'' "ཏྟཱ་་་་་་་

!
1837
ARTES
LIBRARY
SCIENTIA
VERITAS
OF THE
UNIVERSITY OF MICHIGAN
mmun
PERINT
DEPARTMENT
OF
ENGINEERING
TJ
900
B2?
PUMPING MACHINERY.
A PRACTICAL HAND-BOOK
RELATING TO THE
CONSTRUCTION AND MANAGEMENT
OF
STEAM AND POWER PUMPING MACHINES.
BY
illes-
2.
1
WILLIAM M. BARR,
MEMBER AMERICAN SOCIETY MECHANICAL ENGINEERS.
WITH UPWARDS OF TWO HUNDRED AND SIXTY ENGRAVINGS, COV-
ERING EVERY ESSENTIAL DETAIL IN PUMP CONSTRUCTION.
PHILADELPHIA:
J. B. LIPPINCOTT COMPANY.
1893.
COPYRIGHT, 1893,
BY
WILLIAM M. BARR.
PRINTED BY J. B. LIPPINCOTT COMPANY, PHILADELPHIA, U. S. A.
}
الام
лвие
11-13-34
Reclass
PREFACE.
THE author has long felt the need, for his personal use, of
a book similar to the one he now publishes, and had there
been available a work covering approximately the same
ground as the one now offered, the preparation of this
volume would not have been undertaken. No apology is
thought to be necessary for its appearance at this time, for, so
far as I am aware, no other book similar in scope has been
published in this country.
This book, as will be seen upon examination, is essentially
descriptive of pump detail; no attempt has been made to
enter into the theory and mathematics of pump-construction.
The number of leading pump-makers in this country is com-
paratively few, and such are well equipped for the designing
and building of pumping machinery; but there is another
and larger body, comprising engineers, architects, contractors,
plumbers, etc., who have occasion to recommend and use
pumping machinery, and who wish to inform themselves
regarding pump construction,-it is for the benefit of this lat-
ter class that this volume has been prepared. This book
being largely descriptive, its illustrations have been made a
prominent feature, the writer regarding a single suggestive
sketch of more real value than a page of reading-matter.
With few exceptions, the illustrations are from pumping
machinery actually constructed and in use, and those familiar
with the author's work will not fail to recognize how large a
proportion of the whole has been transferred directly from his
own practice. Inasmuch as a large portion of this book has
been in constant use by the writer for years, it is believed that
in its present convenient form it will prove valuable, or at least
suggestive, to others also. In addition to my own experience,
3
4
PREFACE.
I have been particularly fortunate in having the friendly advice
and co-operation of other engineers well known as successful
designers of pumping machinery; these have contributed much
valuable information, accompanied by drawings, here repro-
duced. Interesting and valuable extracts have also been made
from foreign publications not generally accessible except in the
larger cities.
Chapters XIII. and XVII. were originally intended to be
added to this volume as an appendix, but are now incorporated
with it for the following reasons:
The specifications of the Underwriters' pump are so valu-
able and complete that they ought not to be omitted, and
to this end the chapter on duplex pumps was purposely
abridged that the two might not parallel each other. The
design and construction of duplex pumps has engaged almost
the whole attention of the writer for many years, and it was
with some hesitancy that he relinquished his own subject-
matter for the insertion of what might be regarded as an
Underwriters' circular; however that may be, the fact remains
that it is a valuable contribution to the literature of the subject,
and well worth the space given it.
The first part of the chapter on duty-trials is an abstract of
an accepted report of the Duty-Trial Committee of the Ameri-
can Society of Mechanical Engineers, replacing my rewritten
lecture on that subject delivered before the Franklin Institute.
The substitution of the committee's report for my own subject-
matter is in deference to the excellence of the committee's
work, and to contribute to the earnest desire that for the sake
of comparison and uniformity of records it should receive the
support of the Society's membership.
The merit of a hand-book like this consists largely in the
judicious selection and arrangement of its contents rather than
upon strict originality, so that whatever selections have been
useful to the writer he now transfers to the reader with the
added results of his experience.
PHILADELPHIA, January, 1893.
WILLIAM M. BARR.
•
!
i
CONTENTS.
CHAPTER
I.—INTRODUCTION
II.-WATER-PISTONS AND PLUNGERS
III.-PISTON- AND PLUNGER-RODS
IV.-WATER-VALVES AND SEATS
V.-AIR- AND VACUUM-CHAMBERS
VI. SUCTION- AND DELIVERY-PIPES.
VII.-WATER-END DESIGN.
•
VIII.-HYDRAULIC-PRESSURE PUMPS
IX. STEAM AND POWER CRANK-PUMPS
X.-DIRECT-ACTING STEAM-PUMPS
XI. THE DUPLEX PUMP
XII.-COMPOUND DIRECT-ACTING STEAM-PUMPS
XIII. FIRE-PUMPS.
PAGE
7
15
37
47
90
98
126
•
152
173
200
226
•
235
•
271
XIV. MINING-PUMPS
289
XV.-ROTARY PUMPS
XVI.-CENTRIFUGAL PUMPS.
XVII. DUTY-TRIALS OF PUMPING ENGINES.
XVIII. HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING
XIX.-HIGH-DUTY PUMPING ENGINES-FLY-WHEEL .
325
334
361
387
411
I*
5
را
•
PUMPING MACHINERY.
CHAPTER I.
INTRODUCTION.
THE art of raising water must of necessity have been one
of the first of the mechanic arts to engage the attention of
man, for no progress in civilization can be had without a con-
venient and ample supply of good and wholesome water.
The earliest water-supply must have been the permanent
springs and water-courses; but the growth of population, the
increase in wealth, and a higher civilization required a broader
development of the land, and it was during this period of de-
velopment that the ingenuity of man was exercised in origi-
nating schemes and appliances for the lifting and distribution
of water; but of the origin and early history of this art we
know nothing.
It is not probable that any satisfactory device for raising
water is lost to us, although its history has long since been
forgotten, the fact of utility has been the very means of its
preservation; nor did the great invasions and conquests of the
ancient world affect unfavorably the development of this useful
art, for water was alike essential to the conqueror and the con-
quered.
Machines for raising water admit of a great variety of
forms, all depending upon the conditions of supply and
delivery. The sources of supply are usually streams and
wells; the ordinary delivery ranges from the simple lifting for
irrigation to that of high service water-supply for cities.
7
8
PUMPING MACHINERY.
A water-elevator has been defined (Knight) as a device for
raising buckets from wells, and a pump as a device for lifting
water by the motion of a piston in a cylinder.
Among the more important water-elevators are the fol-
lowing:
Archimedian screw,
Mental,
Baling-machine,
Noria,
Bascule,
Persian-wheel,
Bucket-wheel,
Picotah,
Chapelet,
Scoop,
Dutch scoop,
Scoop-wheel,
Ejector,
Shadûf,
Flash-wheel,
Swape,
Flush-wheel,
Turbine,
Hydraulic belt,
Tympanum,
Hydraulic ram,
Water-screw.
Jantu,
A partial list of pumps will include the following:
Bellows-pump,
Piston,
Centrifugal,
Plunger,
Chain,
Rope,
Chapelet,
Rotary,
Diaphragm-plunger,
Spiral,
Draining-pump,
Steam jet pump,
Eccentric,
Steam-vacuum pump,
Ejector,
Syringe,
Elastic-piston,
Vacuum,
Hydrapult,
Water-ram,
Injector,
Water-screw,
Pendulum,
Water-snail.
By reason of the limited scope of the present work it will
be impossible to illustrate and describe so formidable an array
of water-raising machines included in the partial lists given
above. The reader is referred to Knight's "Mechanical Dic-
tionary" for definitions, and especially to Ewbank's "Hy-
INTRODUCTION.
9
draulics" for illustrations, description, and history of early
and curious water-raising devices.
Atmospheric Pressure.-It must not be inferred that the
ancients were unacquainted with the physical properties of
the atmosphere, and that they did not take it into account
in the development of their hydraulic machines. There is
every reason to believe that they understood and applied
certain principles relating to the atmosphere; for example, the
ancient Egyptians understood and used the siphon at least
fourteen hundred and fifty years before the Christian era,
which clearly indicates that they were acquainted with some
facts regarding the expansibility, as well as the compressibility,
of the air; but this was only a partial knowledge, for it is not
clear that the exact data regarding atmospheric pressure were
known until the middle of the seventeenth century. So also
the suspension of a liquid in inverted vessels by the atmos-
phere, such as the atmospheric sprinkling-pot, was known in
the earliest historic times, or, at least, was well known in the
fifteenth century B.C.
The Syringe.-Few ancient devices could be pointed out
that have given rise to more important improvements in the
arts than the primitive syringe. Its modifications exert an
extensive and beneficial influence in society. As a piston-
bellows it is still extensively used in Oriental smitheries. It
may be considered as the immediate parent of the forcing
if not of the atmospheric pump, in both of which it has
greatly increased the comforts and conveniences of civilized
life.
Suction is a word which has come down to us from a vast
antiquity. The operation of sucking, as in the case of an
infant, the sucking of poison from a wounded part by the
application of the lips, are well-known illustrations. So also
the raising of a liquid through a tube into the mouth. This
operation has long been known as suction, and it was formerly
ΙΟ
PUMPING MACHINERY.
L
!
believed that it was effected by some power or faculty of the
mouth independently of any other influence.
Suction is simply a term used to denote the absence or the
removal of the atmosphere, so as to permit the flow of the
liquid; suction does not raise the liquid, nor does it help to
raise it. The term sucker for the valve attached to the
pump-
rod in an ordinary lift-pump no doubt had its origin in the
fancied similarity of its action as compared with that of the
mouth.
An atmospheric pump is merely a contrivance placed at
the upper end of a pipe to remove the pressure of the atmos-
phere there, while it is left free to act on the liquid in which
the lower end is immersed. It is immaterial what the sub-
stance of the machine is, or what figure it is made to assume,
for any device by which air can be removed from the interior
of a vessel is or may be used as a pump to raise water; there
will be required, however, two valves, one opening upwards
and placed in any part of the pipe or in the machine itself, to
allow the water to pass up through it, but none to descend;
the other valve placed over an aperture opening outwards,
through which the contents of the vessel can be discharged,
and at the same time prevent the entrance of external air. Just
how long it took the earlier inventors to determine the "limit
of suction" is not known, but the exact weight or pressure of
the atmosphere was not authoritatively announced until after
the experiments of Torricelli, in 1608, and subsequently con-
firmed by Pascal forty years later. The fact was then fully
established that an atmospheric pump must be placed within
twenty-six or twenty-eight feet of the surface of the water to
be lifted; but, owing to the difficulty in getting tight joints in
the suction-pipe, this distance was gradually shortened until
twenty-two to twenty-five feet was regarded as the practical
or ordinary limit of suction.
Ewbank records a singular incident of a tinman of Seville,
who undertook to raise water from a well sixty feet deep by
a common pump. Instead of making the sucker play within
INTRODUCTION.
II
thirty feet of the water, he made the rod so short that it did
not reach within fifty feet of it. As a necessary consequence
he could not raise any. Being greatly disappointed, he de-
scended the well to examine the pipe, while a person above
was employed in working the pump; and at last, in a fit of
despair at his want of success, he dashed the hatchet or ham-
mer in his hand violently against the pipe. By this act a small
opening was made in the pipe about ten feet above the water,
when, what must have been his surprise! the water instantly
ascended, and was discharged at the spout.
This fact being published (1776) led to a reinvestigation of
the subject, and instead of overthrowing the received doctrine
of atmospheric pressure, more fully confirmed it. It was ascer-
tained that the air on entering the pipe became mixed with
water, and which, therefore, instead of being carried up in an
unbroken column, was raised in disjointed portions, or in the
form of thick rain. This mixture being much lighter than
water alone, a longer column of it could be supported by the
atmosphere; and by proportioning the quantity of air admitted,
a column of the compound fluid may be elevated one hundred
or two hundred feet by the atmospheric pump.
CLASSIFICATION OF PUMPS.
The easy and natural classification of pumps would be to
divide them into three classes:
I. Lift-pumps,
II. Force-pumps,
III. Lift- and force-pumps,
reciprocating
or
rotary.
These may again be sub-classified into-
Single-acting pumps,
Double-acting pumps.
And still further into-
Vertical pumps,
Horizontal pumps.
12
PUMPING MACHINERY.
If pumps be classified according to their details of construc-
tion, the list would be still further extended into-
Bucket-pumps,
Piston-pumps,
Plunger-pumps,
Bucket- and plunger-pumps,
Bucket- and piston-pumps,
Piston- and plunger-pumps (known as the differential
plunger-pump),
Rotary pumps,
Centrifugal pumps.
These names indicate a particular form of construction, and
not a new or distinct classification, for each of these latter
pumps must necessarily be included in the former. This latter
classification is a convenient one, and has been adopted by the
writer for his present use. It may be said that it is not an
exact or scientific arrangement, this much is admitted at the
outset, but it is the commercial one, and, therefore, in the
direct line of every-day use.
The increasing subdivision in business enterprises, and the
growing importance of pumping machinery as a part of the
plant, would seem to call for another classification of pumps
adapted for special uses; for example, acids, alkalies, ammonia,
beer, bilge-water, bleacheries, breweries, dye-works, drainage,
fire-pumps, gas-works, etc. A mere catalogue of names, with
suggestions regarding suitable pumping machinery for each,
would occupy more space than could be given it in the pres-
ent work, and it is doubtful even then if such a presentation
would prove satisfactory because of the repetitions which must
inevitably occur.
Pumps for General Service.-There is no subject in
which it is so difficult to give advice in a general way as in
pumping machinery, because each pumping plant has its own
special peculiarities which must be considered, and which may
not apply to any other pumping plant. There are two things,
INTRODUCTION.
13
however, which come within ordinary practice, and if designs
be made to accord with either or both, the greater part of
pump-service will have been fully met.
The first one is, that if water-ends be made sufficiently
strong to handle water at one hundred and fifty pounds press-
ure, fully eight-tenths of the ordinary run of pump require-
ments can be supplied. It is the common practice in designing
water-ends for trade pumps to make the details of suitable size
and form for this pressure. This will cover the highest fire-
pressure, which is usually the severest test to which an ordi-
nary trade pump is put. For small water-works the pressures
rarely ever reach the one-hundred-and-fifty-pound limit, even
when on direct service. For hydraulic elevator service the
pressure seldom exceeds one hundred pounds per square inch,
except in steel-works and other places where there is a general
hydraulic system using very high pressures.
Tank-service usually calls for lighter pressures, ranging
from twenty-five to fifty pounds, but it is not customary among
steam-pump-makers to make any difference in the weight of
the water-end; the size remaining the same, will require the
same detail and workmanship, so that nothing but a small
amount of cast iron would be saved, and that is not worth the
cost of altering or making new core-boxes.
The second relates to steam-pressure, which does not in
ordinary practice exceed eighty or ninety pounds, so that if a
steam-end of a pump be designed for one hundred pounds
pressure, factory, water-works, and other service will be amply
provided for.
Combinations of such steam- and water-ends will, therefore,
meet almost every requirement in ordinary hydraulic oper-
ations.
Pumps for special service for higher steam- and water-
pressures, such as doubling either or both of them, will re-
quire new proportions. An increased steam-pressure will,
in general, require nothing more than thicker castings and
stronger bolting, the size of the ports, the distribution of
2
14
PUMPING MACHINERY.
steam, and other details remaining much the same. For the
water-end it is often best to entirely change its form, and this
usually occurs. Coupled with this higher water-pressure is,
with few exceptions, a smaller quantity of water to be deliv-
ered. This is especially so in the case of hydraulic-pressure
pumps, but is not generally true of pumping for mines.
Whatever the service, pumps must be able to work contin-
uously without danger of breakage, and this with as little
expenditure of power as possible. But there are other con-
siderations than the mere saving of coal: a pump must be
simple, easily managed, and certain in its operation, or it will
fail to meet the requirements of its owner, who seldom knows
anything about pumping machinery. The very large busi-
ness now annually done in direct-acting single and duplex
pumps can probably be traced directly to their meeting the
above conditions.
WATER-PISTONS AND PLUNGERS.
15
CHAPTER II.
WATER-PISTONS AND PLUNGERS.
FIG. I.
Pistons.-The piston shown in section in Fig. I is the one
in most common use. It consists of an iron or brass cast-
ing bored to fit the piston-
rod, and turned on the
outer flange to a loose fit
in the bore of the cylinder;
and also turned to a diam-
eter suitable to the thick-
ness of the packing to be
used. The length of the
piston should be such as
to admit not less than
three, or, better still, four
rings of packing; to the
end of the piston should be fitted a follower-plate and ring
for compressing the packing. In the above illustration, which
is suited to pistons of small and medium diameter, say eight
inches or less, the piston-rod is turned down to allow a shoul-
der for driving the piston, and supplied with a nut for holding
the piston firmly in position; a second nut, re-enforced by a
jamb-nut, permits an adjustment of the follower and packing.
Each follower should be fitted with two tapped holes for
screwing in eye-bolts, to facilitate removal when the piston
requires repacking.
For larger pistons the design, Fig. 2, is used; it is in all
respects the same as the former, except in the method of se-
curing the follower.
The packing may be of any one of the numerous kinds now

16
PUMPING MACHINERY.
offered to the trade, such as square-plaited flax, hemp, cotton
with rubber core, etc. Another variety, made up of layers of
FIG. 2.

cotton-cloth and sheet-rubber, known to the trade as Tuck's
packing, can be had in great variety in both width and thick-
ness, but usually square, as ½" x ½", 5%" × 5%", 3/4" x 3/4",
etc.
In packing a water-piston a word of caution may not be
out of place here, and this caution applies particularly to the
use of Tuck's packing, and all other packings of similar con-
struction, and that is,—the packing must be cut not less than
one width shorter than will permit the ends touching when
wrapped around the piston; for example, 5%" packing should
be cut 5%" shorter than the circumference of the outer rim
of the piston, 3/4" packing to be cut 3/4" shorter, and in like
manner for any other size. The reason for shortening the
packing is that when new it is perfectly dry; but as it imme-
diately absorbs moisture in the pump, it will soon swell tight
enough to wholly prevent the movement of the piston if
the ends of the packing are allowed to touch each other; by
shortening the packing an end movement is permitted, instead
of compelling a radial one.
Cup-leather-packed pistons are largely in use, espe-
cially for small pumps. Fig. 3 represents a sectional elevation
WATER-PISTONS AND PLUNGERS.
17
FIG. 3.
of such a piston; it is made up of three parts,—the piston-head,
or that portion secured to the rod; the chunk-ring, or central-
distance piece, and the
follower; together with
the two cup-leathers,
as shown. Inasmuch
as this type of piston
does not require an
adjustable follower, the
piston-rod extends clear
through, and fitted with
a nut and jamb-nut, as
shown.
The leather must be
of a solid oak-tanned
5
16
quality, without soft spots or spongy places, and must be
uniform in thickness; 4 and of an inch being the ordinary
or average thickness in selected hides. Nearly all leather
dealers, especially those in the larger cities, have a specially-
prepared leather for hydraulic work, at a price but little in
advance over first-quality sole-leather.
A mould for making cup-leathers is shown in Fig. 4.
The outside diameter, A, corresponds to that of the cylinder it
FIG. 4.
--- A-·
B-
た
​

is intended to fit. The width of the groove must be suited to
the thickness of the leather, together with a slight reduction
b
2*
18
PUMPING MACHINERY.
1
2
on the inside, B, say of an inch; the corners must have a
radius sufficiently large, say ½ to 34 of an inch, to prevent
injuring the leather; the depth of the cup, C, should not be
more than sufficient to make a tight joint. Leathers for a
four-inch piston need not be cut more than 14 inches larger
in diameter, and those for an eight-inch pump need not be
more than 15% inches larger. As the wear is confined to that
portion of the cup-leather which touches the bore of the water-
cylinder, there is no advantage in increasing its depth over
the size given above.
To prepare a Set of Leathers.-After cutting out the
disk of leather, with a central hole for the bolt to pass through,
and making sure of its uniform thickness, it is then soaked in
water until quite soft and pliable, after which it is placed in the
mould with the grain or hair side down, so that that shall be
the working side. The central part of the mould must now be
very gradually tightened until the leather is in firm contact
between the two faces of the mould. There will be more or
less of a ragged edge above the top of the mould, which can
be trimmed off with a knife or chisel. After the leather has
taken a permanent set, as by remaining over-
night in the mould, it may then be removed,
after which it should be well greased with
tallow, and is then ready for use.
As cup-
leathers thus pressed will permanently retain
their shape, they may be made up in advance
of requirements.
FIG. 5.

Pistons with Metal Rings.—It is not a
common practice in this country to fit water-
pistons with metal rings; but if for any rea-
son such a packing is thought to be more
desirable than those just described, a piston made with a single
broad ring, as shown in Fig. 5, will be found quite satisfactory.
This ring must be turned slightly larger than the bore of
the cylinder in which it is to work, say % of an inch for
WATER-PISTONS AND PLUNGERS.
19
a piston twelve inches in diameter. The ring is to be split
diagonally and a piece taken out the same as for steam-
pistons; the ring should touch on the inclined edges when
sprung down to its proper diameter. If made of cast iron or
hard brass, the ring will be sufficiently elastic to fit the barrel
of the pump until either or both are worn out.
There is a strong prejudice in this country against such a
piston-packing for water, and in consequence pistons of this
construction are not plentiful, the preference being given to
those adapted for the use of fibrous packing.
FIG. 6.
A solid piston with grooves, as shown in Fig. 6, is not
largely in use in this country; but it is a kind of piston which
possesses some merit. It is simply a
plain piston turned to fit the cylinder
in which it is intended to work, and
after finishing to size several grooves
are turned to a moderate depth, say 3%
to ½ of an inch on a twelve-inch piston;
the width of the grooves may be about
1/4 to 1 of an inch.

5
16
The efficiency of such a piston depends
on the fact that sudden enlargements in
any cavity through which water is to pass
under pressure induces certain currents
in the cavity which seriously impede, if they do not prevent,
a direct passage of water. Now, in the case of a rapidly-
moving piston this interference, as would be the case where
several grooves occur, would nearly, if not entirely, prevent
a flow of water in any direction before the reversal of the
piston at the end of its stroke, at which time the direction
of pressure would also be reversed, the result being a tight
piston, water-packed.
The piston fitted with wood, shown in Fig. 7, is repro-
duced from Burgh's "Condensation of Steam." This piston
was introduced in 1860 for use in circulating pumps for marine
20
PUMPING MACHINERY.
engines. It is a metal-grooved disk, with blocks of wood
(lignum-vitæ) fitted in the groove. The only feature worthy
FIG. 7.


BB
P
of comment is the fitting of the blocks with each other, which
is shown in the top view.
Linings for Piston-Pumps.-In designing a water-end
for a piston-pump it is considered good practice, but by no
means a universal one, to line the barrel with a brass bushing.
For small pumps, say less than six inches in diameter of piston,
a seamless drawn tube pressed into a bored hole, and slightly
upset or riveted over each end, as in Fig. 8, or made long
enough to touch the heads, as in Fig. 9, is practised by pump-
WATER-PISTONS AND PLUNGERS.
21
makers. The second is much the best arrangement of the
two, because the lining can be more easily removed and a new
one substituted; the objection to it, if any, being mainly one
of first cost in manufacture, there being a longer hole to bore
FIG. 8.

in the pump-barrel and a longer tube to supply, which means
more weight of brass; in addition to which is the cost of cutting
a port through each end of the lining for the water. The
objection to the first of the two examples is the great difficulty
in supplying a distant customer with a tube which shall exactly
fit the cylinder of his pump. The difference between a tight
FIG. 9.

fit and a loose fit is here made more apparent than almost
any other portion of the pump; for example, if the tube be
10 of an inch over size, it will be almost, if not quite, im-
possible to get it in place without special appliances; on the
other hand, should it be 10 of an inch less in diameter than
the bore of the water-cylinder, it will not fit, and cannot be
used.
00
22
PUMFING MACHINERY.
So small a dimension as that given above, when taken in
connection with diameters of four to six inches, is almost im-
possible to register and make duplicate work to fit, except the
pump be returned to the shop where it was built, and where
are special appliances for forcing in the larger tube. If, how
ever, the tube extend from end to end, as in Fig. 9, the smaller
tube may be used, the heads preventing end-motion, and a
film of water will make the tube tight in its place.
A removable lining, as shown in Fig. 10, is a better form
than either of the above. In this case the barrel of the pump
FIG. 10.

is bored and the lining turned to fit. This lining is provided
with a flange for securing it in place, a strong and well-de-
signed arrangement. Incidentally, another advantage is had
in the fact that the flange being circular admits of a regular
spacing of bolt-holes, so that, in the event of the bottom of
the bore of the lining becoming worn by the action of the
piston, accompanied by sand or grit, and thus become scored
or worn out, it is only necessary to smooth down the rough
ridges and partially turn the lining in place so as to bring
a new and unworn surface at the bottom, the fibrous pack-
ing adapting itself to the uneven or irregular circumference.
These linings when worn out can readily be replaced by new
ones.
WATER-PISTONS AND PLUNGERS.
23
BUCKET-PUMPS.
A bucket-pump may be described as a vertical, single-
acting piston-pump, with one or more valves fitted to the top
of the piston, opening upwards. These valves may be of any
one of several varieties best suited to the work. For cistern-
pumps, and usually for all pumps of small diameter and in-
tended only for low lifts, leather valves weighted on the back,
FIG. II.
FIG. 12.


as illustrated in Fig. 11, are in general use, and give but
little trouble, as wet leather makes an excellent hinge.
If a bucket-pump is to lift the water to a considerable eleva-
tion, say one hundred feet, and the diameter of the pump-barrel
is sufficiently large to permit the use of a rubber valve, as
shown in Fig. 12, satisfactory results are usually had. The
valve-seat is made with four or more grids about 4 of
:
24
PUMPING MACHINERY.
an inch wide; the clear opening for the passage of water
through the seat between the grids may be of any convenient
shape, but should not present a larger area than two square
inches, and as much less as circumstances will permit. The
rubber disk may be ½ of an inch thick for diameters not
exceeding three inches, and 3/4 of an inch thick for valves
six inches in diameter. A brass plate, of an inch thick
and nearly the diameter of the valve, should be added to the
top of the valve to distribute the pressure of the spiral spring
over a larger area, and to prevent the spring wearing a circular
groove in the back of the valve.
FIG. 13.
1
16
A pump-bucket for high pressure is shown in Fig. 13.
The valve and seat are made of
hard gun-metal, and designed for
continuous service under one hun-
dred and fifty pounds pressure.
The valve and seat are flat, and
will require to be carefully scraped
to each other; the best way is to
scrape the valve true to a surface-
plate, and then scrape the seat to
the valve. The engraving does
not show it, but it is the common
practice when facing the valve in
a lathe to hollow out that por-
tion of the valve between the
outer and inner bearings to a dis-
tance of say of an inch below
32
the face; this is to make the fit-
ting easier, as only the surfaces
in contact have then to be dealt
with.

1
The bucket type of pump is
used exclusively in drilled and
oil-well service, and often at great
depths. A variety of buckets and valves have been designed.
WATER-PISTONS AND PLUNGERS.
25
for this service, but none have been so entirely satisfactory as
those fitted with ball-valves and cup-leathers, as in Fig. 14.
FIG. 14.
It is the practice to make these
buckets and valves of the toughest
gun-metal only; and as they are
sometimes required to work under
pressures which range from four

FIG. 15.

ހ
hundred to eight hundred pounds per square inch, the greatest
care must be exercised in designing, that a proper valve area
and strength of parts are secured.
Instead of using a ball-valve, an adaptation of the miter-
valve, as shown in Fig. 15, may be used. This is an excellent
form of valve for clear water, but will not work so well if there
be much sand or grit in the water, as it gets between the valve
and its spindle, and often prevents proper seating. It is for this
reason that the ball-valve is given the preference for deep-
B
3
26
PUMPING MACHINERY.
well pumps. The engraving includes a design for a water-
packed bucket or piston, a style not much in use because of
the liability of occurrence of sand in suspension in the water
at the bottom of the well; it does not work satisfactorily
except in clear water.
Air-Pump Bucket-Piston.-The bucket-pump is gener-
ally liked for both air and circulating pumps in vertical marine
engines. Fig. 16 represents a type of bucket in very general use
FIG. 16.

until within a few years. The bucket was usually, though not
always, arranged for the use of fibrous packing; the top of the
piston was made with suitable openings for the passage of
the water through it; over these openings was placed a
single rubber disk, and on the top of that a curved guard,
a central bolt securing all together. The bucket descending
into the water forced the rubber disk off the seat at its outer
WATER-PISTONS AND PLUNGERS.
27
edge only, it being firmly
secured at the centre; this
required of the valve a
somewhat complicated se-
ries of internal movements
at each stroke of the bucket,
the result being destruc-
tive, as well as requir-
ing considerable time for
the recovery of the shape
and proper seating of the
valve.
The more recently-de-
signed marine engines run
at much higher speeds
than was the practice ten
years ago, and among
the other changes in de-
tail was that of the air
pump-
and circulating
circulating
buckets; so that it is now
the common practice to
put in a number of small
valves instead of one large
one. (See Fig. 17.)
FIG. 17.

PLUNGER- PUMPS.
A plunger-pump is one in which a turned plunger passes
through a stuffing-box, ring, or barrel of a pump, so as to
alternately produce a vacuum for the water to flow into the
pump and a pressure when forcing the water out of the pump.
Fig. 18 is an illustration of a design of pump much used for
feeding steam-boilers; the lower is the suction-valve, and the
upper the delivery-valve. The plunger need not fit the barrel
except at the stuffing-box. This pump is single-acting only.
It is not necessary that the valves be one exactly above the
28
PUMPING MACHINERY.
FIG. 18.
other, as shown; but if the suction-valve
or the delivery-valve be located on the
barrel near the stuffing-box, the barrel
must then be increased in diameter, so
that a clear space of fifty per cent. of the
plunger area is had between the plunger
and the barrel in which it works. A pump
of this description is easily packed and
kept in order; any leakage is made ap-
parent at the stuffing-box, and thus with
little care the pump can be kept at a high
state of efficiency.
An arrangement of plunger and ring,
as shown in Fig. 19, is largely employed
in direct - acting steam- pumps.
The
plunger is usually made of cast iron, and

.
the ring of brass; the two are accurately
bored and turned to fit each other; no pro-
vision is made for wear, which, unless the
At first thought one would be
water is gritty, is very slight.
FIG. 19.

WATER-PISTONS AND PLUNGERS.
29
inclined to condemn this arrangement as faulty in design; but
many years of service, covering thousands of steam-pumps,
has completely demonstrated its excellence as a pump detail,
when used in handling clear water, free from gritty matter.
Inside-Packed Plunger-Pump.-If the water to be
pumped is gritty it is likely to bring extraordinary wear upon
both the plunger and the ring when made solid. To obviate
this an internal stuffing-box, as shown in Fig. 20, can be used
FIG. 20.

BOOOD
with advantage. This form of plunger-pump is to be pre-
ferred over a piston-pump for gritty water, inasmuch as the
cost of a new plunger is much less than that of a new lining.
There is a further advantage in the fact that slight reductions
can be made in the diameter of the plunger by turning in a
lathe to remove the scoring incident to the service in which it
may be employed, the difference in diameter being made good
by the use of a thicker packing. This arrangement of pack-
ing a plunger makes a very compact design for a water-end,
but the packing is troublesome to adjust or renew, because
the back-head of the pump must be taken off and the pump
drained before the stuffing-box can be reached.
To obviate this somewhat troublesome detail in manage-
3*
30
PUMPING MACHINERY.
ment, glands for internal stuffing-boxes have been furnished
with a bale extending alongside, beyond the plunger and
through the back-head, with suitable adjusting screws or
other device on the outside of the pump. This greatly facili-
tates the adjustment of the packing, but does not make its
renewal any the less difficult or disagreeable.
Outside-Packed Plunger-Pump with Central Dia-
phragm.—Another design for a plunger-pump is shown in
Fig. 21. This is an adaptation of the water-end described
FIG. 21.

in the preceding section. The heads at both ends of the
water-cylinder are provided with stuffing-boxes, as shown;
the plunger-rod connects the two plungers so that their
movements are coincident. The plunger-rod passes through
a rigid bearing, which is bolted to an annular ring included
in the water-end casting. This bearing may be bushed
with brass, but generally this is not done. The leakage past
the bearing is quite trivial, and is not seriously taken into
account.
An outside-packed plunger-pump with tie-rods is
shown in Fig. 22. It is a favorite one for tank-service or gen-
eral supply in rolling-mills, steel-works, iron-furnaces, etc.
The water-end has a central partition which divides the cyl-
inder into two chambers, each having its own plunger and
WATER-PISTONS AND PLUNGERS.
31
stuffing-box. The plungers are fitted with tie-rods, which
serve to transmit the power from one plunger to the other,
and to make their strokes coincident.
FIG. 22.

A centrally-packed plunger-pump is shown in Fig.
23. The principle of operation is the same as other plunger-
pumps. One advantage which a pump of this kind has over
the one described in the preceding section is that less room is
required for the same displacement. The stuffing-boxes are
not so accessible in this design as in the former one, but no
FIG. 23.

G
difficulty is experienced in adjusting or putting in new pack-
ing. An additional stuffing-box is required for the rod neces-
sary to drive the plunger. This is an excellent design for a
packed plunger water-end, and one which rarely fails to give
complete satisfaction.
32
PUMPING MACHINERY.
Bucket- and Plunger-Pump.-For any other purpose
than simply that of lifting water the bucket-pump is not sat-
FIG. 24.

isfactory, because of its intermittent delivery. If, instead of
discharging into the atmosphere immediately above the level
of the bucket at the top of its stroke, as such pumps usually
2
WATER-PISTONS AND PLUNGERS.
33
do, the delivery-chamber be lengthened sufficiently to fix a
stuffing-box through which a plunger shall work, as illus-
trated in Fig. 24, it will be shown that a double delivery can
be had with a single-acting pump. To begin with, the area
of the plunger must be one-half that of the bucket, and the
stroke of each must be the same, and coincident; the oper-
ation will then be as below:
We will assume the bucket to be at the bottom of its stroke
and the pump fully charged with water; then by its upward
movement a vacuum will be formed underneath the bucket,
the water will flow into the pump-barrel from below and fill
the empty space; let us assume that this volume be one gal-
lon; on the return-stroke of the bucket all the water which
flowed into the pump-barrel underneath the bucket (one gal-
lon) passes through the bucket, past the valve, into the cham-
ber above it; if this were an ordinary bucket-pump no over-
flow would occur on this its downward stroke, but the pump
we are now considering is fitted with a plunger one-half the
area of the bucket, and having a stroke coincident with that
of the bucket, the plunger has by its downward movement dis-
placed its volume of water (one-half gallon) in the upper cham-
ber, this displaced water passing off through the delivery-pipe.
The next stroke, upward, brings another volume of water into
the lower chamber equal to one gallon, but at the same time
that this water is being lifted the upper plunger is being with-
drawn, and instead of one gallon of water passing through the
delivery-pipe, only one-half gallon has been delivered at the
completion of the upward stroke; now the downward move-
ment of the bucket transfers the water (one gallon) from
underneath it to the chamber above, the plunger descending
at the same time with the bucket forces its volume (one-half
gallon) out of the upper chamber; the effect of which is to
convert a one-gallon, single-acting pump into a half-gallon,
double-acting pump.
Bucket- and Piston-Pump.—A pump such as just de-
scribed need not of necessity have a plunger; the same effect
C
34
PUMPING MACHINERY.
can be had if there be two cylinders, one above the other, as
shown in Fig. 25. These cylinders must have a ratio of area
of two to one as in the former case, the larger or bottom
FIG. 25.

cylinder to be fitted with a bucket, and the smaller one with a
piston. The operation of this pump is precisely similar to
the bucket- and plunger-pump, already described.
Piston- and Plunger-Pump. This is commonly known
as the differential plunger-pump. It is, however, but a modi-
fication of the bucket- and plunger-pump, already described.
The piston in this pump is not fitted with a valve, but is solid
WATER-PISTONS AND PLUNGERS.
35
and furnished with layers of packing or not, as the circum-
stances may require. This design differs from others of its
class in the fact that a water-passage, or port, must be had
FIG. 26.

leading from the delivery valve-chamber to the end of the stroke
of the piston in its barrel. Fig. 26 is a representation of such
a pump arranged for use vertically. All the lifting of the
36
PUMFING MACHINERY.
water is done at the upward movement of the piston; when
the piston descends all the water below it is forced into the
delivery-chamber; thus far it is but a single-acting pump. If
we now consider the action of the plunger, it will be seen that
during the movement of the piston downwards the plunger is
entering the upper part of the pump-cylinder, through a stuff-
ing-box, from the atmosphere; this plunger, being half the
area of the piston, and of the same stroke, displaces only one-
half the water forced into the delivery-chamber upon the next
upward stroke of the piston; the other half of the water is dis-
charged because the displacement of the piston is made com-
plete by the withdrawal of the plunger, the delivery-valves being
shut; these valves, as well as the suction-valves, open but once
during a double stroke of the pump. It needs no further
description to show that the lower part of this pump is single-
acting, while the upper part containing the differential plunger
is double-acting when operating in combination with its piston.
:
2
PISTON- AND PLUNGER-RODS.
37
CHAPTER III.
PISTON- AND PLUNGER-RODS.
Rods which enter a water-cylinder through a stuffing-box
must be round, straight, and well finished, and must also be
free from seams and ridges, as both of these are likely to tear
the packing.
For clear, cold, and fresh water, there is probably no better
material for pump-rods than cold-rolled steel, and for this
reason pumps are usually supplied with rods made of this
material, unless for certain reasons some other material is to
be preferred.
The tensile strength of cold-rolled steel piston-rods
will average about 75,000 pounds per square inch of section.
The elastic limit is increased somewhat by the process of
finishing, and more nearly approaches the ultimate strength
of the steel than is the case in ordinary bars. The superiority
of cold-rolled rods consists in their almost absolute uniformity
to size from end to end, much more so than can be had
by ordinary turning and finishing. The hard, smooth sur-
face produced by the finishing-rolls is the very best surface
for the packing.
Piston- and plunger-rods are usually made much larger in
diameter than simply to furnish the requisite strength for
doing the work; if a rod breaks, the pump is useless until a
new one can be supplied. This is an accident which now rarely
occurs, so that it is not a common practice to keep spare rods
on hand. Pump-rods must also be large enough to do the
4
38
PUMPING MACHINERY.
work without vibration or tremor, as it would be next to im-
possible to keep a stuffing-box tight if it had a lateral motion
in any portion of its stroke.
The size of a rod in a trade pump is usually selected for
150 pounds water-pressure upon the piston or plunger; in
addition to this there is always an uncertain quantity to be
added for water-hammer and other unusual strains. The
diameters given in Table I. average equal to those in com-
mon use for the corresponding water-pistons or plungers at
150 pounds water-pressure and for pumps of 12 to 18 inches
stroke.
TABLE I.
DIAMETERS OF COLD-ROLLED STEEL PISTON- AND PLUNGER-RODS OF 65,000
POUNDS TENSILE STRENGTH, FOR PUMPS OF 12 AND 18 INCHES STROKE.
WATER-PISTON or PLUNGER.
Area.
Area.
Diameter.
Inches. Square
Inches.
Pressure at
150 Pounds.
Diameter.
Pressure per
Square
Square Inch on
Cold-Rolled Steel Rods.


Strength in
Pounds of
Inches.
Inches.
Rod. Pounds.
Rod at 1-10.
4
12.57
1,885
138
1.49
1265
9,865
5
19.64
2,946
11/2
1.77
1664
11,505
6.
28:27
4,241
15%
2.07
2063
I3,455
7
38.49
5,774
134
2.4I
2698
15,665
50 27
7,541
18
2.76
2732
17,940
9
63.62
8,543
2
3.14
2721
20,410
IO
78.54
11,781
24
3.98
2960
25,870
12
113.
16,950
238
4.43
3826
28,759
14
154.
23,100
2/2
4.91
4704
31,915
It is customary to assume 65,000 pounds as the average
tensile strength of mild steel, and this figure is taken instead
of the higher one given in the preceding paragraph relating to
cold-rolled steel. A factor of safety of ten is allowed, as it is
the one usually employed in all pump calculations.
PISTON- AND PLUNGER-RODS.
39
TABLE II.
DIAMETERS TO WHICH COLD-ROLLED STEEL RODS MAY BE TURNED FOR BOTH
PISTON AND PLUNGERS.
DIAMETERS IN INCHES.
Water-Pis-
ton or
Pump-Rod.
Bore of
Piston or
Root of
Thread.
Area at Root
of Thread.
Square Inches.
Strength at Root
of Thread in
Pounds at 1-10.
Plunger.
Plunger.
9
ΙΟ
456 78
122
13%
11/8
.94
1/2
14
1.06
888889
.69
4,485
5,700
15/8
13/8
1.16
1.06
6,890
134
1½
1.28
1.29
8,385
1%
1½
1.28
1.29
8,385
2*
15%
1.39
1.52
9,880
24
134
1.49
1.74
11,370
12
238
134
1.49
1.74
11,370
14
22
2
1.71
•
2.30
14,950
Tobin bronze, so named after its inventor, is a somewhat
recent alloy, of which copper is the principal ingredient. This
material has been used by the author in pump construction,
for piston-rods and other details, and, so far as his experience
goes, it fully bears out the claims of Mr. Tobin, as possessing
great strength, toughness, and uniform texture. It can, when
heated, be forged into any required shape as readily as steel.
It works well in a lathe, and is susceptible of a high polish.
Tensile tests, by Professor R. H. Thurston, of cast speci-
mens .798 inch diameter, 5 inches long, showed a tenacity
per square inch of original section of 67,600 pounds; te-
nacity per square inch of fractured section, 73,160 pounds.
The diameter of the fractured section = .767.
Samples of hot-rolled metal prepared from a 34-inch cylin-
drical rod gave the following results when tested by Professor
J. E. Denton under torsional strain: Tensile strength, 94,550
pounds, with a stretch of 36.44 per cent. in 4½ inches length
of specimen, having a diameter of 5% inch.
The factor of safety given in the above tables, together
with the net strength, has made it appear that the rods were
figured on too liberal a scale; but we have not yet considered
40
PUMPING MACHINERY.
the method of securing the piston or plunger to the rod; there
will be in all cases a reduction in diameter, which will also
reduce the margin of strength. Several kinds of fastenings
are shown in connection with a water-plunger, but it is ob-
vious that the same fastenings will be equally applicable for
water-pistons.
Pump-Rod Details.-To those more accustomed to steam
than to hydraulic work, their first suggestion would probably
be to taper the pump-rod as in Fig. 27, and, indeed, many
FIG. 27.

pump-rods are so fitted, but this method of fastening, so favor-
ably known and practised in steam-work, has for one reason
or another proven so troublesome in pump-work that it is now
seldom employed. Any corrosion, dirt, or want of perfect
alignment with the pump-rod will throw the piston or plunger
out of line, and thus produce not only an uneven wear in a
piston which is quite short for its diameter, but especially so
in the case of plungers which have considerable more length
than diameter; the friction first on one side of the plunger
and then on the other side, as it passes the centre, makes a
pump work badly, and is, moreover, a very difficult thing to
correct. When a pump is new this difficulty is not usually
experienced, especially in such shops as make it the practice
to finish each plunger on its own rod; but it is afterwards,
when repairs are needed, when the rods have become corroded,
or when new pistons or plungers are required to fit an old rod,
PISTON- AND PLUNGER-RODS.
41
or new rods to fit an old piston, that the trouble is usually
had.
A common and a better method of securing a piston or
a plunger to a rod is shown in Fig. 28, which simply con-
FIG. 28.

sists in turning down the rod to a diameter suited to the load
on a pump. It does not matter whether the piston or plunger
fits the rod or not; the rod being turned to a standard
diameter to fit a nut, the hole in the piston or plunger being
enough larger to slide on easily, a reamer of an inch larger
4
will give an ample allowance.
A word of caution is needed in fitting rods with a shoulder:
a sharp corner, as shown in Fig. 28, is wrong, and if the rod
be so fitted it will be only a question of time, and usually not
a long one, when the rod will break in the sharp corner; this
FIG. 29.

is especially true in the event that a steel rod be used. A
fibrous iron rod will last longer than a steel one under the
above conditions. There should be a liberal round or fillet in
the corner, as shown in Fig. 29, and a corresponding curve in
the piston or plunger.
Rods of the diameters given in Table I. may be turned
4*
42
PUMPING MACHINERY.
MPING
..
down to the diameters given in Table II. for their correspond-
ing pistons or plungers.
If a larger area against the end of the piston or plunger be
required, it can be had by turning a taper back of the plunger
and fitting a collar, as shown in Fig. 30. The taper should be
FIG. 30.

made on a lathe having a taper-attachment, and not by setting
over the tail-centre. The collar should be reamed, and the
rod carefully fitted to the collar, which latter should be faced
off true in place to receive the end-thrust of the piston or
plunger.
The same result may be secured by fitting the rod with the
threads, as in Fig. 31. The larger thread must have its root
FIG. 31.

above the diameter intended for the plunger, and threaded for
a hexagon nut, which should be faced off true on the rod-
centres; the end of the rod is to be fitted with a thread below
the diameter intended for the piston or plunger, as are all the
others.
PISTON- AND PLUNGER-RODS.
43
If the end of the rod be threaded, as in Fig. 32, no reduc-
tion in diameter will be required, and if thought advisable its
diameter could be reduced at least one size from those given
FIG. 32.

in Table II. In this arrangement the piston or plunger will
be held by the faces of the nuts, which offer an ample bearing,
all the strain coming on the thread.
In water-ends of large size, and others in which the rod is
not continuous between the steam-piston and the water-
plunger, the pump-rod may have a collar welded on, as shown
in Fig. 33.
FIG. 33.

In such construction the rod is to be uncoupled at the cross-
head between the steam- and water-cylinders; the plunger-rod
cannot pass through the stuffing-box, therefore the plunger
and rod are to be withdrawn entire through the rear end of
the water-cylinder; provision must be made for such with-
drawal when locating the pump in the building in which it is
to be used. In regard to the size of collars, it is a common
practice to make them once and a half the diameter of the
rod, the length of the collar to be one-half its diameter.
44
PUMPING MACHINERY.
1
FIG. 34.
Jamb-nuts are shown in the several illustrations for securing
the nuts which fasten plungers to the rods. This is a simple
and reliable device; a further precaution may be had by drill-
ing a hole in the end of the rod and inserting a split pin to
prevent the nut working off in case it should loosen.
A special nut, as shown in Fig. 34, is sometimes used. It
has an hexagonal body
tapped to fit the end of
the rod, and an extension
piece smaller in diam-
eter, drilled, tapped, and
fitted with a set-screw
for tightening against the
end of the rod, this set-
screw being fitted with
a jamb-nut as shown.
It possesses no advantage over the two nuts, and is much
more expensive to make.

Stuffing-Boxes.-The design of stuffing-box shown in
FIG. 35.
Fig. 35 is the one com-
monly used in small
pumps; that is to say,
for rods 14 inches in
diameter, or less; but
it is sometimes used
for rods as large as 3
inches in diameter.
These stuffing - boxes
are usually made of
brass, though some
pump - makers

make
them of iron; they
screw into a tapped
hole in the cylinder-
head or steam-chest.
The cavity for the packing in stuffing-boxes of this design
PISTON- AND PLUNGER-RODS.
45
is not usually as liberal as in the design given in the next
paragraph, therefore the packing will oftener require renewal.
The commonest form of stuffing-box for sizes for 1½ inches
in diameter and larger is shown in Fig. 36. The stuffing-
box is cast together
with the cylinder-
head, and is bored
concentric with the
opening for the rod.
The stuffing-box
should always be
deep enough to take
in four strands of
packing; the gland
should be made to
compress the pack-
ing, as it wears to
FIG. 36.

nearly one-half its original bulk. The stuffing-box and gland
here shown are intended to be of cast iron, which for ordi-
nary service, and fresh water, will answer every purpose.
Stuffing-boxes and glands bushed with brass are shown in
Fig. 37. For diam-
eters of rod less
than 2½ inches the
be
gland should
made wholly of
brass instead of in-
serting a bushing.
The bottom of the
stuffing-box shows
the bushing extend-
ing through the
thickness of the cyl-
inder-head; this ar-
FIG. 37.

rangement is a good one, but stuffing-boxes are frequently
made with a brass washer, bevelled for the packing, but not
extending through the head.
46
PUMPING MACHINERY.
The question of bevelling the bottom of the stuffing-box and
gland is one about which there is considerable difference of
opinion; the writer favors the bevel for round packings, and
thinks that even in the case of a machine-made square flax
packing a tighter joint is had against the rod than is the case
when the gland and bottom of the stuffing-box are flat; but
for packings made up of layers of cotton cloth and india-
rubber the bottoms of the stuffing-box and the gland should
be flat.
Metallic packings, so commonly used for steam-cylinder
piston-rods, have not been found to be altogether satisfactory
for pump-rods, and in consequence are but little used.
WATER-VALVES AND SEATS.
47
CHAPTER IV.
WATER-VALVES AND SEATS.
Clack-Valves.-The simplest form of valve for pumps of
small size and low lift is shown in Fig. 38. It is largely used
in hand-pumps and for
small power-pumps for
agricultural and house
services. The lower
illustration shows a
plan of the valve. It
is cut from a piece of
leather about three-
sixteenths of an inch
thick, the outer diam-
eter corresponding to
that of the flange by
which it is held, the
inner diameter being
enough larger than
the opening into the
pump to make a good
joint under pressure;
the clearance around
the valve may be % to
1/4 inch, depending on
the size of the pump
and the thickness of
the leather. The al-
FIG. 38.

О
lowance for hinging may be one-half the diameter of the valve.
A metal plate under the valve nearly the size of the opening,
48
PUMPING MACHINERY.
and another plate on the top as large as the diameter of the
valve, should be securely fastened to each side of the leather
by rivets or bolts. Nearly all English books on pumps men-
tion the use of hippopotamus and rhinoceros hides for large
valves; the same material is also referred to as being satis-
factorily used for packing pump-pistons. Leather of this
description is very sparingly used for pumps in this country,
if at all.
A piece of india-rubber held at one edge, as in Fig. 39, is
FIG. 39.
often used in large
pumping- and sew-

شت
erage-engines. The
openings are usu-
ally rectangular,
having a length
several times that
of the breadth; this
is a simple and dur-
able valve. The
rubber may be 5%
to I inch thick, de-
pending on the ser-
vice and the general dimensions of the pumps. The guard
should have a generous curve to prevent breaking the rubber
at the line of flexure.
An interesting experience with clack-valves is furnished by
Mr. Henry Teague (1887), in a paper read before the Institute
of Mechanical Engineers, England. Mr. Teague had occa-
sion to remove a 15-inch clack in 1863, substituting a
new one with one-third of its area cut out in the centre of
the flap; the aperture so made was covered with a sup-
plementary leather clack hinged upon the main clack at
the edge opposite the main hinge, as shown in Fig. 40. By
this means the concussion, which was the cause of making
the change, was so completely prevented that, by placing
the hand or ear in contact with the clack-box, not the least
tremor or sound was perceptible, and the time of closing
WATER-VALVES AND SEATS.
49
could not be detected. The leather of the small supple-
mentary clack continued in constant action for seven years'
before requiring to be
changed, and that of
the main clack for thir-
teen years.
Another
experience
occurred in 1866 with
a 14-inch pump, in
which Mr. Teague sub-
stituted for double-beat
valves others similar to
the illustrations in the
preceding paragraph,
whereby the action of
the pump was greatly
improved, as shown by
the two pairs of dia-
grams which are repro-
duced in Figs. 41 and
42, which were both
taken under similar cir-
cumstances when the
FIG. 40.

pump was travelling at 160 feet per minute. These dia-
grams were taken with an indicator having a spring of 10
pounds per inch of range. Fig. 41 shows the vacuum above
the clacks, which, at the commencement of the stroke, is 5½
pounds per square inch for the break-clack, as shown by the
lower line, and 7 pounds for the double-beat valve, as shown
by the upper line. Fig. 42 shows the vacuum below the
clacks, which began at 4 pounds per square inch in each case.
The suction-pipe was 13 inches in diameter, and ran 40 feet
horizontally from the pump before dipping into the tank, in
which the water-level at the time of taking the diagrams was
6 feet below the bottom clack. In taking a number of
diagrams similar to those shown in Figs. 41 and 42, it was
found that scarcely any two were exactly alike. This was
c d
5
50
PUMPING MACHINERY.
attributed, in most instances, to the undulatory motion of the
water in the tank from which it was being pumped. The tank
was 70 feet long by 40 feet wide, and the undulation was
ts
2
4
FIG. 41.
Vacuum above foot valre.
10 20 30 40 50 60 70 80
Line

JATROS.
8
20
2
4
6
#
Break-clack.
FIG. 42. Double-beat Valve.
Vacuum below foot valve
Atmospheric Line

+
10 20 30 40 50 60 70 80 90 100
Percentage of Stroke,
sufficient at times to raise the indicated line above the atmos-
pheric line at the finish, as shown by the dotted line in Fig. 42.
In some pumps with a vacuum of about 7 pounds per
square inch, or 16 feet head of water at the commencement
of the up-stroke, the indicated line has been raised several
feet above the atmospheric line at the finish. To obviate this
evil with a long horizontal suction-pipe, it occurred to him to
interpose at the foot of the pump a vacuum-vessel for the long
horizontal pipe to deliver into, the vessel containing sufficient
vacant space to spare for taking up the fluctuations of the water,
which are thereby prevented from affecting the working of the
pump. This addition answered the purpose thoroughly, the
pumps working with the greatest ease when so supplemented.
The break-clacks (Fig. 40) were found to work incessantly
!
WATER-VALVES AND SEATS.
51
!
for five years without changing a leather and without showing
the least sign of leakage, under 350 feet head of water, or 150
pounds per square inch, and there is not the slightest con-
cussion, the time of closing not being perceptible. For a
velocity of 160 feet per minute of the pump Mr. Teague found
the weight of the flap should be about 2 pounds per square
inch. On the top of the second or supplementary clack a
third of still smaller diameter can be added, and even a fourth
or more hung alternately on opposite sides.
FIG. 43.
A hinged clack-valve made of metal with a leather face
is shown in Fig.
43; this does not
differ essentially in
its action from the
valve described in
a preceding para-
graph (Fig. 39), ex-
cept that a fulcrum
is usually above the
valve, and should

be as far removed
from it as the design will permit; the greater the distance
the better will be the flow through the valve-seat and the less
will be the angle of lift.
FIG. 44.
A butterfly-valve is shown in Fig. 44. It is a modifica-
tion of a pair of clack-
valves hinged usually at
a common centre. These
may be of any convenient
shape, and arranged in any
manner best suited to the
requirements of the pump.
The valves are usually of
metal, faced with leather
or india-rubber; it is not

52
PUMPING MACHINERY.
often that the valve and face close metal to metal,—probably
more on account of the noise than anything else.
Clack-valves and butterfly-valves should be provided with
stops to prevent their rising any higher than is actually neces-
sary for the proper admission of the water into the pump; the
angle of the valve should be about 30°, but must not in any
case exceed 45°, above the seat.
Fulcrums for Clack-Valves.-The central pin, or ful-
crum, should be of brass, and of liberal diameter to prevent
FIG. 45.
FIG. 46.

breakage. The wings, or valves, should have the holes oblong,
at right angles to the face of the valve, so as to permit the
valve to lift from 1% to 4 inch vertically at the hinge before
it begins to turn on its fulcrum. (Figs. 45 and 46.) This
kind of valve is suited only to slow-moving pumps by reason
of the long time it requires a valve to seat, and the loss of
water occasioned by slip during the interval of seating.
A wing-valve, as shown in Fig. 47, is largely used in
FIG. 47.
pumps for feeding boilers, and in
hydraulic pumps for high pressures.
The valve and seat should be made
of hard gun-metal. The seat may
be driven in straight, and is held
by the friction of its sides in the
hole in which it is driven; some-
times the seats are fitted with an
external taper, but this is not the
usual practice. The face of the
valve and seat are tapered 45°, and
are readily fitted and ground together. This valve may have

WATER-VALVES AND SEATS.
53
a spring fitted to its back to hasten the seating, but this is not
always done.
Wing-valves sometimes have the lower portion of each wing
on a curve, as shown somewhat exaggerated in Fig. 48, the
object being to give the valve a partial rotation at each stroke
of the pump, and thus compel the valve to seat itself in a new
place at each stroke.
FIG. 48.
FIG. 49.


A wing-valve with a flat seat and a leather face, as shown
in Fig. 49, is sometimes used in hydraulic-pressure pumps
where the pressure does not exceed 400 pounds per square
inch. It works fairly well, but possesses no marked advan-
tages over similar valves fitted with a bevel and seating metal
to metal.
A conical mitre-valve, as shown in Fig. 50, is recom-
mended by Björling, but, so far as the writer knows, it has not
been adopted by any pump-manufacturer in this country. It
consists of a conical shell, the sides of the cone being at an
angle of 45°; the top part is turned to fit the bevel in the top
of the seat; the lower part of the cone is furnished with three
or more wings or guides, according to the size of the valve;
these guides should be put in at such an angle that the valve
at every rise will partially rotate so as to wear the seat evenly
and prevent any grooves being formed. Inside the cone is
cast a short spindle, which beats against a stop provided in
"
5*
54
PUMPING MACHINERY.
the pump-valve box. This design of mitre-valve is con-
sidered by him the best, the flow of water being brought into
FIG. 50.

Nh
the proper direction by reason of the cone causing less ob-
struction to its passage and producing less contraction of the
fluid vein than the one referred to in the preceding section.
A spindle-valve with a flat seat, such as shown in Fig. 51,
is not often used in pumps. It is, however, occasionally met
FIG. 51.
FIG. 52.


with in small pumps intended for light service, a disk of
leather being sometimes attached to the valve to insure tight-
ness when seated, as in Fig. 52.
WATER-VALVES AND SEATS.
55
It
FIG. 53.
A bell-valve for impure or thick fluids, such as tar,
molasses, paper pulp, tan liquor, etc., is shown in Fig. 53. It
has a perpendicular lift, but
is not fitted with guides.
receives its name from a re-
semblance to an inverted
bell. An inspection of the
engraving will show its con-
struction, and especially the

+
concentration of weight below the valve-seat to insure its
proper seating. The valve is suitable for slow-moving pumps
only, and, so far as the writer is aware, they are not in use in
this country.
Ball-valves are used in considerable numbers, but the gen-
eral impression is that they are not a regular, but a special, valve.
They are quite commonly in use in sugar-house pumps,
filter-press pumps, and especially in artesian- or deep-well
pumps. Fig. 14 shows the arrangement of a bucket of a
deep-well pump, and Fig. 54 a design of a ball-valve and
seat, for general service.
The ball and seat should
be made of tough gun-metal,
the guard should screw to
the seat, and be so designed
that it will have three or four
ribs to guide the ball verti-
cally as well as to limit the
height of its lift, which should
not exceed 1/4 of an inch. The
width of the guides should be
as little as consistent with
safety, so as not to obstruct
the flow of the liquid above
the valve-seat.
Ball-valves are sometimes
FIG. 54.

made with an iron or lead core and covered with india-rubber
56
PUMFING MACHINERY.
to a depth of 3% to 5% of an inch thick, depending on the
size of the ball. This arrangement is not favored by the
writer except for very light service. In his experience balls
of this kind never gave satisfaction, but did give a great deal
of annoyance at an average pressure of 100 pounds per square
inch, the trouble being that the rubber covering would split
and allow the metal core to pass through the seat or wedge
fast in it under pressure. For cold water and light service,
say 25 pounds pressure, the ball may be wholly of rubber, or
weighted with a metal core to make it seat with greater
promptness; but for higher pressures only gun-metal should
be used. Ball-valves must be firm in order to retain their
shape, but when made of gun-metal they need not exceed
one-half of an inch in thickness for pressures up to 1000
pounds per square inch; care must be exercised in lightening
balls that they are not made so thin as to prevent seating
promptly.
Disk-valves made of india-rubber are in more general use
than any other type of valve in this country. Valves for
cold water should be of the best quality of india-rubber and
vulcanized sufficiently to give the valve firmness, yet be
sufficiently elastic to permit bending at right angles and regain
its shape. For hot water a composition of india-rubber and
graphite makes an excellent valve. A valve of this com-
position is hard and inelastic, and will not adapt itself to
inequalities; it is necessary, therefore, to see that the valve-
seat is true and flat. The valve should be scraped true to
a surface-plate, and then the seat carefully fitted to the valve.
These two valves are usually distinguished as a cold-water
valve, meaning the india-rubber valve, and a hot-water valve,
meaning the hard composition.
Disk-valves are usually made from 2 to 6 inches in diam-
eter; sometimes larger diameters are employed in mine and
other special pumps, but the general conclusion reached by
builders and users alike is that rubber valves should not
exceed 4½ inches in diameter. The four sizes more in use
WATER-VALVES AND SEATS.
57
than any other are 3, 32, 4, and 4½ inches, and probably
one-half of all the india-rubber valves in use of this type are
3 inches in diameter. The following proportions for india-
rubber valves have given satisfactory results:
2 inches diameter Xinch thick Xinch hole.
2골
​3
3 1/2/10
4
4를
​5
(6
X 16
X 1/
66
66
9 "
16
66
(6
X
(C
11
16
CC.
66
13 4
(
66
3
6
FIG. 55.
Valve-Seat and Spindle combined.-The valve-seat
shown in vertical section in Fig. 55 is one which has long
been in use by the writer, and has given such excellent satis-
faction that he gives it his
fullest endorsement. The
seat is threaded to screw
into a corresponding taper-
hole tapped in the valve-
deck of the water-end of
the pump; a taper of one
inch to the foot will an-
swer; the threads may be
8 to the inch for all sizes
up to 4½ inches diameter.
The valve-stem is included
in the same casting with
the seat; it is turned and

16
polished so as to offer no resistance to the movement of the
valve; an allowance of of an inch is given for clearance
between the valve and stem. The top of the stem is threaded
with a taper-thread, the guard-nut being tapped with a corre-
sponding taper-tap so as to screw down firmly, making a
much tighter joint, and one less likely to jar or work loose
than when parallel threads are used. As a precautionary
measure a hole is drilled immediately above the guard-nut,
58
PUMPING MACHINERY.
16
and a split pin inserted to prevent the nut working off and
the valve getting adrift in the water-end. A brass plate of
an inch thick for a 2-inch valve, and of an inch thick for
a 4½-inch valve, is put on the back of each valve to dis-
tribute the pressure of the spring over a larger area, and to
prevent the spring wearing a groove in the back of the valve.
This plate need not be more than three-fourths of the diameter
of the valve.
FIG. 56.
Valve-Cap.-A brass casting, as shown in Fig. 56, extend-
ing over the top and down the sides of a valve, has been used
by the writer in a num-
ber of special cases with
excellent results. There
is no other objection to
using it on all disk-valves
of india-rubber except that of increased cost.

Springs. No general directions can be given for the
stiffness of springs, but if the 2- and 2½-inch valves be fitted
with springs of No. 12 brass spring wire, the 3- and 3½-inch
valves with No. 10 wire, and the 4- and 4½-inch valves with
No. 8 wire, good results may be expected for the ordinary run
of service. The diameter of the spring may be one-half that
of the valve, and if made with five coils will have the proper
elasticity.
A metal disk-valve, as shown in Fig. 57, will inter-
change on the same seat with the valve, spring, etc., just de-
FIG. 57.

scribed. These metal valves are frequently used instead of
the hard-rubber valves for hot water.
WATER-VALVES AND SEATS.
59
Valve-Seat with inserted Spindle or Guard.—A modi-
fied form of the design illustrated on page 57 is shown in Fig.
58. It does not differ ma-
terially from the seat and
stem already described ex-
cept that the seat is tapped
and the guard screws into
it, both threads being
tapered as in the one pre-
viously described. This
seat and guard is in quite
general use, but is not so
good a design as the
former, because the guard
is likely to, and often does,
FIG. 58.

work loose, there being no practical method of preventing it.
FIG. 59.
Valve Seat with threaded Spindle and Nut.-An-
other design for a valve-seat, guard, etc., is shown in Fig.
59. In this figure, it will
be observed, the valve-
seat is drilled with a
straight hole, the guard
has a shoulder, and is
held down and in place
by a nut underneath; this
nut can be secured by a
split pin if desired. The
objection to this design
is that the valve cannot
be removed without un-
screwing the nut, and in
almost every pump the
under side of the valve-
seats are inaccessible in the suction-chamber, and it frequently
happens that those in the force- or delivery-chamber are
equally so.

60
PUMPING MACHINERY.
i
Conical Springs.—Attention is directed to the conical
spring used in this illustration; this is a style of spring for-
merly in almost universal use, but it has been superseded
within the past five or six years by the parallel spring, which
has already been described. Some makers of pumps continue
to use it, but the spring is defective, in the fact that all the
strain of the spring centres in the two upper coils, and break-
ages constantly occur at that point. In a parallel spring, as
shown in Fig. 58, the tension is throughout its whole length,
and breakages seldom occur.
Disk-Valve with Hemispherical Guard.-An india-
rubber disk-valve with hemispherical guard, as shown in Fig.
60, is largely employed in pumping machinery in Europe, and
FIG. 60.

is often of large size, having long been in favor with pump-
makers and marine-engine-builders. It is seldom met with in
this country except in mining-pumps. It possesses no advan-
WATER-VALVES AND SEATS.
61
tage over the flat disk-valve with spiral spring, as shown in
Fig. 55.
Securing Valve-Seats in Place.-A design of pump-
valve shown in Fig. 61 is one which has long been in use by
the makers of the Cameron
pump. It consists of a brass
shell filled with vulcanized in-
dia-rubber. The rubber forms
the valve-face, and the metallic
casing resists the pressure and
protects the rubber from injury.
It will be observed that the
metal casing extends down-
wards and surrounds the stem.
The valve-seats are driven in on
a taper, one exactly above the
other. One valve-stem is com-
mon to both valves. By un-
screwing the plug on the top
of the delivery-chamber the
valve-stem can be withdrawn
and the valves and springs
removed for examination or
repairs through the hand-hole
plates.
Another method of securing
valve-seats and valves is shown
in Fig. 62, which represents the
practice of the makers of the
FIG. 61.

Davidson pump. The suction-valve seat is screwed into the
pump-casting, and is tapped to receive a spindle or valve-
guard; the delivery-valve seat is centred by the upper hole,
which is bored in line with the lower one. A shoulder is
provided on the upper valve-seat. The spindle, or valve-guard,
passes through the upper valve-seat, and is made to screw
into the lower seat; a shoulder is provided on the valve-spindle
for holding the upper valve-seat down upon its bearing when
6
62
PUMPING MACHINERY.
the spindle is screwed sufficiently into the lower valve-seat.
The examination or removal of the valves is easily had by the
withdrawal of the valve-spindle.
Large valve-seats are often made with projecting wings or
flanges for bolting the seat in place, somewhat after the style.
FIG. 62.

of Fig. 71. Standing-bolts are recommended rather than tap-
bolts. Muntz-metal makes a good bolt; the nuts may be
of cast brass.
For mine-pumps the valve-seats are often flanged and
WATER-VALVES AND SEATS.
63
inserted between the pump-flanges, as shown in Fig. 44, as
well as in several illustrations in the chapter on mine-pumps.
In designing a pump for high pressures with such a valve-
seat, the lower pump-flange should have a diameter of
opening the same as that of the valve-seat above it; this will
give the latter a better support than if held by the flange
alone.
An easy-seating water-valve is illustrated in Fig. 62,
A. It consists of a cup-shaped valve with central guiding-
pin, and a valve-seat and disk cast solid, as shown in cut.'
FIG. 62, A.

This disk is slightly larger than the aperture below it in the
valve-seat, and causes the fluids pumped to be deflected at
right angles, just the same as an ordinary water-valve does
with its lift limited by a stop. The fluids, therefore, have
power to raise the valve as high as the disk only, and for that
reason a stop to limit the lift of the water-valve is not
necessary. This water-valve does not strike against a stop in
its upward movement. When it closes it does not seat with a
harsh, abrasive action, but, being partially cushioned on the
water between the disk and valve, it seats softly, and there is
reason to believe it a durable and satisfactory water-valve.
+
64
PUMPING MACHINERY.
The water-ways are large, and the valve has capacity equal to
This valve is secured by a recent patent, and is
any other.
used in the Marsh steam-pump.
An annular valve is shown in Fig. 63. The seat has
grids to give support to the central spindle, or guard, but which
FIG. 63.
do not extend up to
the valve-face. The
valve is an annular
ring fitted with wings
which attach to a
bored guide sliding on
the valve-spindle. A
spring, with nut, etc.,
is furnished as shown.
In this valve - seat
the area is somewhat
reduced under that of
an ordinary valve of
the same diameter, but
having two circumfer-
ences for the discharge,
the quantity of water
which may be deliv-
ered is very much in-

creased. Assuming the outer diameter of the opening
through the valve-seat to be four inches and the inner one
to be two and a half inches, there is then :
4 inches diameter
21/
12.57 inches area and 12.56 inches circumference.
66
Net area =
4.91
7.66 inches,
66
66
7.85
66
66
+
and 20.41 inches total circumference.
÷
The lift required would be 7.66 20.41.375 of an inch
to give the full area.
The annular valve shown in sectional elevation and plan in
Fig. 64 is exceedingly simple and has much to commend it for
WATER-VALVES AND SEATS.
65
FIG. 64.
FÜ
valves of medium size, and for pumps running at a moderate
rate of speed. Weighted valves do not seat as promptly as
those in which the in-
itial impulse is given
by a spring. The valve
in this design is simply
a brass ring guided by
four wings, which are
included in the same
casting with the seat.
The lift of the valve
not being excessive,
there will be little or no
tendency to stick or to
prevent prompt seating.


The Troy Valve.-
Designers of pumping
machinery have always
endeavored to keep
down the lift of a valve,
and to get rid of the
grids in the valve-seat;
some of these efforts
have been successful,
others less so. Fig. 65
I
is an illustration of a valve and seat by the Holly Manufac-
turing Company; it has a 14-inch clear way of opening
through the valve-seat, a flat rubber valve enclosed in a metal
shell, which is top-guided; it is not furnished with a spring
to assist it in seating. This may be said to fairly represent
the practical limit to which the reduction in diameter of valves
for large pumping engines has been carried out.
Valve-Guides.-Guiding a valve by a spindle issuing from
the top of the valve and extending into a drilled hole in the
cover of the valve-chamber, and at the same time having
e
6*
66
PUMPING MACHINERY.
7.
wings to the valve fitting in a bored valve-seat below, as
shown in Fig. 66, is occasionally met with, and is a practice
FIG. 65.
which the writer does not
consider a good one. The
fitting of a valve and seat
together is in itself an
operation of the greatest
nicety, and requires the
utmost precision and free-
dom of movement to make
its operation satisfactory.
No such precision is had
in fitting valve-covers, so
there can be no certainty
that the centres of the
valve-seat and the drilled
hole in the cover are in line
with each other; if not, the
operation of the valve will
be faulty, and in conse-
quence the action of the
pump uncertain. Valves

of this type should never be guided in two places, and as a
matter of choice between the two guides, as shown in the
engraving, the wings fitting the seat is one much to be pre-
ferred to that of having the spindle guided in the cover.
A top-guided valve without the lower wings is shown in
Fig. 67; the top-guide is, as in the former case, the cover of
the valve-box. This design is objectionable also, and for the
same reasons as given in the preceding paragraph, there being
no certainty that the guide for the valve-stem will be at right
angles to the valve-seat, or that it will be exactly in line with
the valve movement.
In all cases where it is thought to be necessary to have a
mitred valve guided from the top, the valve-seat and guide
should be self-contained, so that the seat, valve, and guide may
all be fitted together before its insertion in place in the pump.
WATER-VALVES AND SEATS.
67
For valves having a leather or rubber face coming down
upon a flat valve-seat the conditions are more favorable to
satisfactory working, but on general principles the writer
advocates having valve-seats and guides self-contained.
FIG. 66.
Π
A built-up wing-valve of large diameter (360 mm.) is
shown in Fig. 68, in section, with a partial plan showing the
arrangement of the wings or guides. The valve-stem has
a welded collar
against which the
flat disk for taking
the pressure is fit-
ted; underneath
this disk is another
one of rubber for
making the joint
on the valve-seat;
still another metal
disk is added, after
which the wings or
guides are placed
on the same spindle,
the whole being
held in place by a
through-going key,
as shown. The de-
tails of this valve
are good, the only
objection being the unusual number of parts which compose
it. An examination of the engraving will make clear that
no derangement is likely to occur which would interfere
with its proper working, except that of the backing out of
the key; a split pin would make secure against such a con-
tingency.

A rectangular valve with double seating is shown in
Fig. 69. This is a form of valve not much used in this coun-
68
PUMPING MACHINERY.
try; it is open to the objection that the seating of the valve
on the grid always occurs at the same place, causing a rapid
wear of the valve under what are usually thought to be very
moderate pressures.
The writer does not recommend this
design except for light service, say for pressures not exceed-
ing 25 pounds per square inch, although such valves are
in use at press-
FIG. 67.
ures many times
greater than the
limit placed upon
them above; nev-
ertheless, they
have not shown
themselves dur-
able under high
pressure, and are
quite inferior in
many respects to
circular valves,
which are free to
rotate about the
central spindle or
guide. Rectan-

gular valves
should always be
of the very best
quality of india-rubber to insure satisfaction, as they are sub-
ject to shearing action over the openings in the valve-seat,
as well as the bending movement along the central line of
fastening.
A multiple beat-valve with four beats or valves is
shown in Fig. 70. This form of valve is not anywhere in very
general use; in England, probably more than elsewhere; but,
so far as the writer is aware, it is not used in this country in
any prominent pumping-station. The main valve-seat may be
secured to the valve-chamber casting by any one of several
1
WATER-VALVES AND SEATS.
69
methods; the central spindle has a length suited to the com-
bined thickness of the four valves, together with a proper
allowance for the lift of each.
The bottom valve has two faces, the lower one fitted to the
valve-seat and the upper one fitted to receive the second valve,
which is similarly fitted for the third valve, and thus for the
fourth valve, which completes the series. There are four open-
ings for the passage of water, so that after making allowance
for the difference in the circumferences of the several openings,
there will then be
required a lift but
little more than one-
fourth over that, if
one valve had to do
the work.
FIG. 68.

There does not
seem to be an agree-
WANKAE
ment of opinion
among English en-
gineers regarding
this valve, some
thinking highly of
it, while others have
taken them out and
substituted valves of
a different design.
The principal ob-
jection seems to be
that by reason of
the large diameter
as compared to the
height the rings are
liable to stick; the

valve is also reported as being noisy in operation, which
would seem to indicate intermittent or irregular seating,
with the attendant jar and hammer in the water end of the
pump.
70
PUMPING MACHINERY.
A double-beat india-rubber valve is shown in Fig. 71.
The main valve-seat is bolted over a suitable opening in the
valve-chamber; this valve-seat casting includes the grid for
the lower annular ring, of india-rubber, loosely fitting around
О
FIG. 69.
MK
о
O
LILIHIL
о
A
a central cylinder also in-
cluded in the same casting,
and forming the water-way
to the upper valve.
curved guard controls the
rise of the lower valve, the
latter having a vertical lift
of about 3/8 inch before com-
ing in contact with the
guard. A brass grid forms
the upper valve-seat, and by
a slight projection down-
ward also serves to keep the
large guard in place. A col-
lared bolt and nut secure
this upper grid in place,
and forms the central spin-
dle for the upper valve;


the curved guard for this valve permits the latter to rise about
1/4 inch before contact, the guard being secured to the vertical
spindle as shown. This is a better form than a single large
valve of the same diameter as the lower ring, because the
circumferences of the two valves are much greater than the
one large valve at the bottom, thus permitting the passage
of the same volume of water with a greatly reduced lift,
and in consequence smoother working than if all the water™
were required to pass the circumference of the larger valve
only.
Cornish Double-Beat Valves.-These are largely used
in England, and to a considerable extent in this country, but
as for ourselves the general tendency is towards employing a
number of small valves rather than one large one in water-
WATER-VALVES AND SEATS.
71
ends, so that, for the present at least, they may be regarded as
out of style.
Through the courtesy of Mr. A. F. Nagle I am able to
present a copious extract from his paper on
Cornish or
FIG. 70.

Double-Beat Valves," contributed to the American Society of
Mechanical Engineers, and published in Vol. X. of its Trans-
actions.
The principal features to be considered are,—
1. The width of seat.
2. The unbalanced area.
3. Its weight.
4. Its lift.
5. Its form of body.
72
PUMPING MACHINERY.
1. The Width of Seat.-What should be the width of a valve-
seat? Theoretically, a knife-edge, so that the same area
FIG. 71.

should be presented to the water-pressure before as after it is
lifted. Practically, only sufficiently wide to sustain the pressure
FIG. 72.
12'dia.
9 dia.
FIG. 73.


12'dia.
9 dia
brought to bear upon it without injury to the metal. Brass
should sustain a pressure of at least 1000 pounds per square
inch of surface with safety and permanency. This is less
WATER-VALVES AND SEATS.
73
than one-thirtieth of its crushing strength, and only about
two-thirds the pressure brought upon crank-pin journals.
"2. The Unbalanced Area.-With a wide seat it is impossible
to know exactly what the unbalanced area of a valve really
is. It may be that of either extreme between the inside or
outside diameters, as indicated in Figs. 72 and 73, or it may
be a yet worse case if the bearing should be perfect over its
entire surface, like Fig. 74, where it may approximate to a
vacuum between the faces.
66
Even if we do not assume a vacuum to exist between the
faces, it is still certain that something less than the normal
pressure must be between the faces, or the valve would be in
a leaky condition; and hence there must inevitably be
required a greater pressure per square inch to start the valve
than exists outside of it, and this condition is one which
accounts for the shocks and noise of these valves.
FIG. 74.
g'dia.
3. The Weight of the Valve.-I thought it was the weight
of the valve, if free to move, which
determined the velocity of discharge
through it. If the valve is large in
diameter compared with its lift, so
that the velocity of approach be-
comes so small that it could be
ignored, and its form of such gentle
curves that no violent impingement
occurs, then it would seem that the
weight per square inch of unbal-
anced area must govern the flow
or velocity; for it is this weight
which is the equivalent of a press-
ure upon the water within the
valve which causes the outward
flow. And if this theory were correct, then the flow through
it would have the same velocity at any position it might be in,
and the valve should rise and fall in exact proportion to the
changing velocity of the plunger. If, on the other hand, the
valve be of irregular form, very light, and the velocity of

12 dia.
D
7
74
PUMPING MACHINERY.
approach very great, then the calculation for its action would
be very complicated.
"4. Its Lift.-This is practically answered in the last section.
The velocity being determined by the weight, and always the
same for the same weight, then its lift would naturally adjust
itself to the changing speed of the plunger, so that the
requisite water might be supplied to it or discharged from it.
If, for example, a valve weighed one pound per square inch
of its inside unbalanced area, it was reasoned that the velocity
through the valve would be that due to this pressure, or,
applying the well-known formula:
v = 8.03 X Vñ•
v = 12.20 feet per second.
"The size of the plunger and its velocity, and the number of
valves, now determine the lift of each valve.
"5. The Form of the Valve.-First of all, there should be no
air-pockets such as are possible in such forms as are shown in
Figs. 72, 73, or 74. The curves should all be of easy lines in
order to avoid impact, it being reasoned that flat surfaces,
particularly at the upper bend of the valve, would cause an
impact which would make the valve rise more than that due
to the pressure produced by the velocity.
"With these theories in mind I constructed the pump-valves
for the High Service Pumping Engine at Providence, R. I.
The engine is of the vertical compound type, with cranks
exactly opposite each other (the first instance of the kind in
this country, I believe, 1874) and geared 1 to 5, driving two
horizontal double-acting plunger-pumps. For full description
of this engine, see Franklin Institute Journal for September,
1876. The plungers are 17 inches in diameter and 4 feet
stroke, and the greatest speed about 20 revolutions per minute.
All the valves were of the same size, and only one for each
inlet or outlet, and that was 12 inches in diameter at the
lower seat, and 9% inches at the upper. The seats were
designed to be three-eighths of an inch wide, but the seats,
WATER-VALVES AND SEATS.
75
Sq. Head Fronze
&-Bolts-
not the valves, were actually chamfered so that only one-
eighth of an inch bearing-surface remained.
Center Force (276) Nozzle.
FIG 75.
----+-

"
"
‹-12-
"
-10 ~ +
Center Suction (15) Nozzle,
SUCTION & FORCE VALVES
-FOR-
-
"HOPE"' PUMPING ENGINE
Providence, R. I. November, 12th. 1874.
Bottom of Force
Bottom of Suction
1-6 SIZE
Suction
Force
1-R.
3%
16
5%
"
11 Diam.
-1234 Diam.-
4
3 d.
9 Dia. Inside
10 Dia.Outside
-d.
Bronze
}
"
1% Diam.!
"
13 Diam.Outside
-13-
T!
1½ Diam.
"
-17-Diam-
-13-Diam.-
-16- Diam.-
18
"Fig. 75 is a vertical section of the valve, and Fig. 76 a full-
size section of the seat.
"The weight was 53.44 pounds in water, one-seventh less
than in the air.
Mean net water-pressure
Lower seat outside diameter 1234"
52 pounds.
127.68 square inches.
"inside
CC
12½" =122.72
Suction
Force
76
PUMPING MACHINERY.
Upper seat outside diameter 9%"
65.40 square inches.
❝ inside
93811
= 69.03
Net outside unbalanced area
= 62.28
inside
Seat area
53.69
8.59
"The seat area is only 16 per cent. of inside area, and only
52 × 62.28÷8.59=377 pounds pressure per square inch of
FIG. 76.
-
"
11% afa
12'dia.
surface.
Upon the theory of a perfect
seating, the pressure required to
open the valve would be 62.28 X
(52 +15)÷53.69=77.31 pounds.
"I confess that I am somewhat
sceptical as to the possibility of such
perfect seating of a ground valve as
to produce the condition of a
vacuum, but somewhere between a
vacuum and the water-pressure it
must be, and I have assumed this
extreme condition the better to illus-
trate a theory.
"These valves worked noiselessly at the greatest speed, and
after six months' run the grinding-marks were not worn away.
"It will be observed that the valve weighed just about one
pound per square inch of inside unbalanced area, and hence,
if the theory advanced in section 3 were correct, the velocity
of the water through it should have been 12.20 feet per second.
It proved to be fully 20 feet per second.
"In order to learn how much truth there was in these the-
ories, I took several indicator-cards directly from the valve
itself.
"These cards are reproduced at full size in Figs. 77, 78, and
79, and the dotted lines are added to represent the line of
velocities of the plunger at all points of the stroke.

“The greatest lift attained at
II revolutions per minute was inch = .032 foot.
13
18
(6
5.66
2
.039
r
9
(6
16
= .047 "
WATER-VALVES AND SEATS.
77
"It is not possible to know the exact diameters at which
the discharge may be considered to take place, but I as-
FIG. 77.
11 Revolutions per minute.

1 ft.
2 ft.
8 ft.
4 ft.
Position of piston.
Heavy line traced by indicator-piston attached to valve.
Dotted lines velocity of pump-plunger.
FIG. 78.

13 revolutions per minute.
FIG. 79.

18 revolutions per minute.
sumed it for the lower seat at 12 inches, and the upper at
9% inches.
"Circumference of discharge at
12 inches diameter = 37.70 inches.
9%"
Total
28 27
= 65.97
"
= 5.50 feet.
7*
78
PUMPING MACHINERY.
"Area of discharge of valve at
II revolutions = 5.50 X .032
13
18 (C
5.50 X .039
.1760 square feet.
.2145
= 5.50 X .047 = .2585
Area of 17 inches plunger
1.576
"Maximum velocity of plunger at
4 × 2 × 11 × 1.57 ÷ 60 = 2.30 feet per second.
II revolutions
13
18
66
= 4 X 2 X 13 X
66
= 4 X 2 X 18 X
1.57 ÷ 60
1.57 ÷ 60
2.72
66
3.77
CC
CC
""
'Displacement of plunger at
II revolutions = 1.576 X 2.30
13
18
(6
66
1.576 X 2.72
3.6248 cubic feet per second.
4.2867
1.576 X 377 = 5 9415
Velocity through valve at
CC
(C
II revolutions
= 3.6248÷.1760
20.60 feet per second.
13
= 4.2867 ÷.2175
20.00
(C
18
= 5.9415 ÷ .2585 = 23.00
CC
It was calculated to be
12.20
Head due to velocity of 20.60 feet
20.00
6.60 feet, or 2.87 pounds.
6.20 66 2.70
(C
(C
23.00
66.
8.20
3.57
Weight of valve per square inch of unbalanced area, I pound.
Ratio of weight of valve to pressure due to flow through the valve, about 1 to 3.
"The diagrams, as well as experience, showed :
(C
First, that the width of a valve-seat could safely be brought
to a very narrow surface, probably much less than I made it
(%"); for the pressure in this case was but 377 pounds per
square inch of surface.
"Secondly, that the lift of a valve is exactly proportioned to
the velocity of the plunger, if it is not too light so as to be
brought to its stop before the maximum velocity of plunger is
attained. The deviation from this theoretical curve, as shown
in the cards, is attributable to the friction of the stem running
to the indicator, and possibly somewhat to seat area, small as
it is.
WATER-VALVES AND SEATS.
79
Thirdly, that in the form of valve shown, the theory that
the velocity of the water through the valve is that due to the
head corresponding to the weight of the valve per square inch
of unbalanced area, did not prove to be very near the truth.
I can conceive of but one reason for this great variation from
FIG. 80.

the theory assumed, and that is, the effect of the horizontal
issuing stream diminished the vertical pressure. I think it is
not improbable that there is a mathematical demonstration for
the resultant vertical force due to an issuing horizontal stream
from a curved aperture, and express correctly the relation of
8B
PUMPING MACHINERY.
lift to weight, but I have not attempted to thus solve that
problem. I do not think that the friction of stem or force of
spring is sufficient to account for the deviation.
"The valves were symmetrical and round in form, and were
afterwards turned down and reduced in weight to 35 pounds
in water, or .66 pound per square inch of inside unbalanced
area, but I regret that I took no further diagrams. The nar-
row seats, and the quiet action, and the synchronous motion
with the plunger, were the more important features in my
mind at that time, and the question of weight of valve was left
to experiment after all."
The four-seated valve shown in Fig. 80 is reproduced
from an engraving in Engineering of a pump designed by
Edward Easton & Co., London. It represents the valve as
being attached to a rod forming a bucket- and plunger-pump,
but it is obvious that the same valve may be differently
employed by simply omitting that portion relating to the
operation of driving. The pump from which this detail is
taken has a plunger 20 inches in diameter, the bucket being
28 inches in diameter, and a stroke of 33 inches. This pump
min
FIG. 81.
Down Stroke-
Scale 1-60 th
Atmos. line.
Up Stroke
works against a head
of 72 pounds per
square inch. The
revolutions of the en-
gine were intended
to be 18 per minute,
but during the trial
averaged but little
more than 13 revolu-
tions; the efficiency

of the pump at that rate of speed was found to be 97.3 per
cent.
The indicator diagram, Fig. 81, taken from the above pump
at the time of the trial, shows how nearly a large valve, and
one of the above design, meets the exacting requirements of
water-works service. The indicated agreed with the calculated
WATER-VALVES AND SEATS.
81
power required of the pump, which was found to be 47 horse-
power. In the copying of the drawing the packing-box and
ring at the top of the bucket was omitted for the sake of
clearness.
FIG. 82.
Perreaux's Valve.-A novel form of valve is shown in
Fig. 82; it is not used on this side of the Atlantic, and
probably not extensively in Europe.
The valve is made of india-rubber in
the form of a tube flattened at one end,
like the mouth-piece of a clarionet.
The thickness of the sides of the upper
part diminishes gradually to the top,
where the two sides meet and form two
lips, which, when the valve is in a state
of rest, are in close contact and prevent
the downward passage of the fluid.
With any upward pressure the lips
separate and allow of the upward pas-
sage of the fluid. The gradual diminu-
tion in thickness, or tapering of the
sides forming the lips of the passage,
enables the valve to open and close with
the slightest variation of pressure, and,
by properly proportioning, to resist any
required amount of downward pressure.
The passage for the fluid is larger in
these valves than in any others of equal dimensions; they also
possess the advantage of having a clear way, there being
nothing whatever to retard the passage or flow of the fluid,
and they close perfectly and instantaneously the moment the
pressure from below ceases. The illustration shows two valves,
one in the plunger and the other held between flanges forming
the inlet-valve to the pump. The action is the same as in
ordinary pump: raising the piston producing an opening of
the lower or suction-valve, whilst lowering it closes the valve.
This valve is chiefly suitable for the raising and moving of

f
any
82.
PUMFING MACHINERY.
semi-fluid masses, like paper pulp, etc. These valves
These valves may be
used singly for small pumps, or combined in sets for large
pumps. But it may be remarked that, with the resistance of
the material and the small dimensions of the orifices, an excess
of motive-power is required, which may be an important
element in large pumps. This valve is not to be recommended
for heavy pressures; the ball-valve is believed to be much
better for pressures greater than say 25 pounds per square
inch.
These pumps have been made with rubber valves and piston,
working in a glass barrel for handling acid.
Bushings for valve-seats are not often required except
for repairs, but it sometimes happens that such a detail must
be carried out in a new design. Fig 83 is a suggestion based
FIG. 83.

upon a successful experience. The cast-iron deck of the
pump is reamed, counterbored, and tapped with a straight tap.
The bushing is of gun-metal, threaded to correspond to the
size of the tap, the collar being turned to fit the counterbored
hole, and projecting slightly above the valve-deck of the pump
to save the trouble of facing off any slight irregularities of
the casting. The inside thread has the same taper as that of
the valve-seat, say one inch to the foot. A steel driver made
to fit the inside thread, with an adjustable screw-collar to
tighten down upon the flat surface of the upper side of the
bushing, makes a good tool for screwing these bushings in
place. When the bushing is down to where it belongs, the
adjustable screw-collar on the driver can be slackened and the
driver easily and quickly withdrawn, after which the bushing
is ready to receive the valve-seat.
+
WATER-VALVES AND SEATS.
83
Dovetail and Lead Joints for Valves.-Securing valve-
seats in place by dovetail joints, calked with lead, as shown
in Fig. 84, was a common practice a few years ago, when larger
valves than are now commonly employed in large pumping
engines were in general use. This arrangement is not well
suited to valves less than 6 inches in diameter, and has but
little to recommend it for valves of any size, although valve-
FIG. 84.

LEAD
seats of 8 and 9 inches in diameter have been in use for
many years having this kind of fastening. Now that the
use of smaller valves which permit of screwed seats is the
common practice, the above detail is not likely to be carried
out in any important pumping engine contracts.
A weighted valve, such as that shown in Fig. 85, was
much in use in this country twenty years ago for large pump-
ing engines, but has been almost entirely displaced by the
smaller valves with springs instead of weights. The valve-
seat in the illustration is a cast-iron grid, faced with brass, the
latter metal cast in a dovetail recess as shown, and afterwards
faced off true to receive the valve. The spindle is of tough
brass or Muntz-metal, fitted into a tapering hole in the seat.
The weight is cast iron, and is made to slide loosely on the
spindle, and to prevent noisy contact a small rubber buffer is
included in the spindle-cap as shown. The valve-seat is held
in place by lead calking, as shown in Fig. 84. Valves of this
design, from 6 to 12 inches in diameter, are in use, but they
84
PUMPING MACHINERY.
are somewhat sluggish in their movement and do not seat as
quickly as similar valves furnished with spiral springs. They
are only suited to large pumps and slow piston speeds.
FIG. 85.

Size of Valves.-It was formerly, say twenty-five years
ago, the practice to fit water-ends with valves much larger in
diameter than at present; valves from 5 to 8 inches in diame-
ter being quite common in water-works practice.
A reaction afterwards set in, and the diameters of water-
valves were then reduced so much that on some large pump-
ing engines, built perhaps twelve or fifteen years ago, thin
metal disks were employed less than 2 inches in diameter.
This may be regarded as an error in judgment, which has since
been very generally corrected, so that it may be said the com-
mon practice is now to confine the diameters of rubber disk-
valves to between 3 and 4½ inches.
The area of clear water-way through a set of valves in a
water-end should be not less than forty per cent. of the plunger
area for pumps having a speed of 100 feet per minute; and if
WATER-VALVES AND SEATS.
85
that speed be increased to say 125 feet per minute, then the
combined water areas through the valve-seats should be fifty
per cent. of the plunger area; and in like manner 150 feet per
minute would require sixty per cent. valve area; 175 feet per
minute would require seventy-five per cent. valve area; and
200 feet per minute should have a valve area equal to the
plunger area.
In order to get this valve area there is the temptation to
make the valves as few in number and as large in diameter as
possible, consistent with quiet action when the pump is
working at its highest speed. It will be borne in mind that
the two properties of the circle are to be considered when
designing valves and seats. The area increases as the square
of the diameter; the circumference varies directly as the
diameter, The flow of water through the valve-seats has to
do with the area only; the escape of the water under the
valve into the chamber in which the water is to flow has to do
with the circumference only.
Let us assume, by way of illustration, that a pump would
require a single valve area, corresponding to 9 inches in
diameter:
Area of 9 inches 63.62 square inches.
Circumference of 9 inches
28 27 inches.
Then 63.62 ÷ 28.27 2.25 inches lift required to make the circumferential
opening equal to the area.
If, now, we try four valves, each 4½ inches in diameter,
precisely the same area is had, but the circumference will be
doubled, thus:
Area of 4½ inches 15.90 X 4 valves = 63.60 inches total area.
Circumference of 4½ inches 14.13 X 4 valves
circumferences.
56.52 inches combined
Then 63.60 ÷ 56.52 = 1.125 inches lift required of each valve to give a
circumferential opening equal to the area, or one-half that required by the single
valve.
It will be understood that both of the above illustrations
have reference to the diameter of the opening, and not that
8
86
PUMPING MACHINERY.
of the valves; to be exact, we must deduct the area of the
grids, taking only the clear water-way through the seat, but
the application would be the same.
TABLE III.
SHOWING THE PROPERTIES OF CIRCLES APPLICABLE TO VALVE CALCULATIONS.
Diameter.
Inches.
Area,
Square Inches.
Circumference,
Inches.
Lift required to equal
the Area. Inches.
1 2
1½
1.76
4.7I
.375
3.14
6.28
.500
22
4.91
7.85
.625
3
7.07
9.42
.750
3½
9.62
10.99
.875
4
12.57
12.56
1.000
42
15.90
14.13
1.125
5
19.64
15.70
1.250
5/2
23.76
17.27
1.375
66778
28.27
18.84
1.500
6½
33.18
20.42
1.625
38.48
21.99
1.756
72
اله
44.18
23.56
1.875
50.26
25.13
2.000
8½
56.75
26.70
2.125
9
63.62
28.27
2.250
9½
70.88
29.84
2.375
10
78.54
31.41
2.500
10%
86.59
32.98
2.625
II
95.03
34.55
2.750
11%
103.86
36.12
2.875
12
113.09
37.69
3.000
The speed at which a pump will run noiselessly,
assuming that an ample water-supply is had, will probably
depend more on the relation of its valve area to that of
its plunger area than almost anything else.
In a quick-running pump with too small a valve area an
excessive lift is required of the valves, so that in the interval
of seating a portion of the water in the pump-cylinder passes
under the valves and back again into the suction-chamber; at
the moment when the pressure overtakes the valves in their
downward movement the velocity is so greatly accelerated as
to force them violently down upon their seats, the pump be-
comes noisy, and nothing will relieve the pump but a reduc-
WATER-VALVES AND SEATS.
87
tion in the speed of plunger, suited to the proper and noiseless
action of the valves.
Noisy action is not always confined to quick-running pumps ;
it is a common fault with nearly all low-priced pumps; the
temptation evidently being to put in larger water-plungers
than the valve area can supply at the common rating of 100
feet piston-speed per minute. The additional cost of a larger
plunger in any properly-designed pump is a small one com-
pared with that of an entire water-end, which would have to
be supplied if a larger valve area must be furnished. The
writer has known of pumps which barely measured 25 per
cent. of valve area, when the least measurement should have
been 40 per cent., a species of fraud rarely detected until too
late to remedy it.
A concentric ring-valve is shown in Fig. 86; it has but
recently been introduced in this country.
FIG. 86.

After a very exhaustive research in the matter of valve
design, Professor Riedler, of Berlin, has given this the pref-
88
PUMPING MACHINERY.
erence over all others for large valves, for quick-running
engines. An inspection of the drawing will show that this
design offers a large valve area for a comparatively small diam-
eter of valve-chamber. This valve is intended to be operated
by a suitable mechanism, which insures opening and closing
at periods exactly corresponding to the changes in piston or
plunger movement. This positive movement permits a higher
piston speed than can be had in designs depending upon either
gravity or springs for closing the valves. Its efficiency has
certainly never been surpassed, and in point of simplicity,
strength, and consequent durability it seems to leave but little
more to be desired
Mechanically-operated water-valves are among the
very earliest devices connected with the development of the
direct-acting steam-pump. For one reason or another they
never became popular in this country, and finally dropped out
altogether, so that it was something of a surprise to American
engineers when it was learned that several large, high-speed
pumping engines had been built, and exhaustive experimental
tests made, in Germany, in which mechanically-operated valves
proved so highly satisfactory that their future adoption in all
large and important high-speed pumping engines seems now
to be nearly, if not altogether, certain.
Professor Riedler, of Berlin, has probably undertaken and
accomplished more in the matter of indicator research in
pumping engines than any other person now living. His
designs for valves and valve-moving mechanism are the result
of the thorough investigation of pumping machinery actually
at work in mines and water-works stations. The valve-gear
invented by him is the result of this investigation, and not the
development of a chance thought. Large pumping engines
made on his system are at work in Europe at piston-speeds of
more than 300 feet per minute, and at high pressures such as
usually, attach to mining operations. Several designs, with
descriptive matter, have been published by Professor Riedler
in Germany, but so far as the writer knows they have never
1
WATER-VALVES AND SEATS.
89
been translated into English. It was the intention to have
had translations made, in part, of Professor Riedler's papers,
and certain engravings reproduced, but it was found that to
make proper presentation would occupy more space than
could be given the subject in this book. Whatever the writer
may do in the future regarding this subject, he now refers the
reader to:
I. Indicator-Versuche an Pumpen und Wasserhaltungs-
Maschinen. Von A. Riedler, Professor des Maschinenbaues
an der kgl. Technischen Hochschule in München. Mit 21
Tafeln und 24 Text Figuren. 1881.
II. Illustrated papers by Professor Riedler, published in the
Zeitschrift des Vereines Deutscher Ingenieure :
Band xxvii., 1883, " Die unterirdischen Compound-Wasser-
haltungs Maschinen am Mayrau-Schachte der Prager-Eisen-
industrie-Gesellschaft zu Kladno."
Band xxix., 1885, " Constructions-Grundlagen der Pumpen-
und Gebläse-Ventile."
Band xxxii., 1888, "Pumpen mit gesteuerten Ventilen.”
Band xxxiv., 1890, " Neuere Wasserwerksmaschinen."
90
PUMPING MACHINERY.
CHAPTER V.
AIR- AND VACUUM-CHAMBERS.
Air-Chambers.-All single pumps must be fitted with air-
chambers if a smooth flowing and continuous delivery of
water is desired. Duplex pumps have a discharge much more
uniform than is the case in single
pumps; but even they should have
air-chambers to insure a perfectly uni-
form delivery.
If the discharge of water from a
pump be intermittent, each successive
FIG. 87.
FIG. 88.


impulse is equivalent to a blow; and this effect is not local-
ized, but is distributed throughout the whole system. The
final reaction, however, comes upon the pump, and if the
pressure be a heavy one, leaky joints or broken castings are
AIR- AND VACUUM-CHAMBERS.
91
the almost inevitable result. A properly-designed air-chamber
corrects this to a certain extent, but not wholly so. The
largest air-chambers are required for single-acting pumps, and
the smallest size for double-acting duplex pumps.
The form of an air-chamber admits of considerable
latitude in design. For small pumps in which it is the
common practice to use copper air-chambers, the design
shown in Fig. 87 is seldom deviated from, because pump-
FIG. 89.
makers with scarcely any exception
buy them from certain copper-
smiths who make their manufac-

FIG. 90.

ture a leading specialty, and have adopted certain dimensions
which meet all the ordinary requirements of the trade. These
dimensions are approximately as below:
Diameter, 6 inches. Height, 10 inches. Base threaded for 1 inch pipe tap.
66
8 CC
66
14
CC
15
16
9
ΙΟ
(C
<<
66
(C
114
1/½ 66
CC
66
CC
*
30
2
ક
92
PUMPING MACHINERY.
For highly-finished pumps the air-chambers are sometimes
made of polished copper, riveted and soldered to a cast-iron
or brass base, as in Fig. 88. The design of the base admits
of a wide variety of treatment, but the whole does not
materially differ in proportions from those of cast iron, except
that they are usually smaller in diameter and made enough
higher to get the proper cubic capacity.
Where larger air-chambers are required, it is customary to
make them of cast iron, and usually of the form in Fig. 89 for
medium-sized pumps, and Fig. 90 for large pumps. A table
of dimensions and capacities suited to all ordinary require-
ments is here given:
TABLE IV.
ORDINARY DIMENSIONS AND CAPACITIES OF AIR-CHAMBERS.
Inside Measurement.
CAPACITY.
Diameter. Height
Inches.
Inches.
Cubic Feet.
Gallons.
12
14
16
18
20
22
24
26
28
30
60
32
34
68
36
72
*♡♡♡ & INDOJON
24
1.57
11.75
28
2.49
18.64
32
3.72
27.87
36
5.30
39.66
40
7.16
54.34
44
9.68
72.48
48
12.56
94.00
52
15.98
119.42
56
19.96
149.12
24.55
183.60
64
29.79
222.69
35.76
266.59
42.42
317.28
REMARKS.
The diameter of the
neck and flange of an
air-chamber will be
governed somewhat
by its intended loca-
tion, but it will usually
approximate one-third
that of the air-cham-
ber for the neck, with
its corresponding di-
ameter of flange.
Size of an Air-Chamber.-For single pumps, double-
acting, the cubic contents of the air-chamber should be not
less than three times that of the pump displacement; that is,
a single stroke. This will be for moderate pressure and speed
only, such as ordinary boiler-feeding, tank-service, etc. If for
a pressure of 100 pounds per square inch and upwards, to-
gether with a very rapid piston movement, as in the case of
AIR- AND VACUUM-CHAMBERS.
93
fire-pumps, then the air-chambers should be not less than six
times the displacement of a single stroke of the pump-piston.
For double-acting duplex pumps the cubic capacity of an air-
chamber need not be more than one-half to two-thirds of the
size given above. The position of the air-chamber should be
on the highest portion of the pump, and always above the
highest portion of the delivery opening.
The diameter of the neck of an air-chamber should be no
larger than that necessary to give stability to it and insure
proper strength of connection to the flange by which it is
bolted to the pump. The larger the neck the greater will be
the disturbance of the water-level in the air-chamber, and the
less efficient will the air-chamber be in correcting the inequal-
ities of flow.
It is generally known that water contains air in suspension
or solution, and in improperly-designed hydraulic machinery
is likely to make trouble if there be air-pockets or cavities in
which it may collect. Having this fact in mind one would
naturally conclude that an air-chamber would always have its
proper supply of air; that there would be no decrease in
quantity at any time, even though it were not augmented by
the liberation of air from the water when passing through the
pump; but the reverse of this is true, and it is a common
experience that air in large air-chambers almost wholly disap-
pears, passing off, probably, with the water by absorption.
An automatic air-pump for supplying the air-cham-
ber with air is shown in Fig. 91. It is exceedingly simple,
and does its work in a very satisfactory manner. It consists
of a piece of, say, 2½-inch wrought-iron pipe, about 30 inches.
long, which we shall name the air-cylinder, on the top of
which is screwed a 2½-inch tee, one end having a 14-inch
check-valve opening inward, and on the other end of the tee
a 3/4-inch check-valve, opening outward; a pipe from this
small check-valve leads to the air-chamber. On the bottom
of this 2½ inch pipe, which we have named the air-cylin-
der, is screwed a cap from which a 1-inch pipe, which we
94
PUMPING MACHINERY.
shall name the supply-pipe, leads to the water-cylinder head;
intermediate between the two is inserted a gate-valve or cock,
which completes the whole apparatus, and which is suitable
for pumps from 750,000 to 1,500,000 gallons in 24 hours.
To start this air-pump, it is necessary that
the pump to which it is attached shall be in
operation. First open the valve in the supply-
pipe leading from the water-cylinder head to
FIG. 91.
To Air Chamber

个​委
​Pump Cylinder
No Scale
the air-cylinder to charge
the latter with water; then
partially close the valve or
cock in the supply-pipe
until the check-valves be-
gin to work; this is easily
determined by the sound
or click of the valves in
seating.
Its operation may be thus
described: Once the air-
cylinder is full of water, any
lowering of the water will
be the cause of a vacuum
above its level; the air enter-
ing through the inlet check-
valve immediately fills this
space above the water at
atmospheric pressure. The distance to which this water shall
be lowered may be controlled by the valve or cock in the sup-
ply-pipe under the air-cylinder. The water is lowered in the
air-cylinder by reason of its open connection with the water-
cylinder of the pump, for each time the water-piston draws its
AIR- AND VACUUM-CHAMBERS.
95
#
water the same vacuum is had in the air-cylinder as that in
the suction-chamber of the pump. The air-cylinder being of
so much larger area than the supply-pipe leading to the water-
cylinder, there is not sufficient time for the air-cylinder to
completely empty itself before the change in the direction of
motion of the water-piston, which at once changes the direc-
tion of the flow in the supply-pipe, arresting the downward
movement of water in the air-cylinder, forcing it upwards and
filling the air-cylinder completely with water from the water-
cylinder of the pump, thus driving the contained air in the air-
cylinder through the delivery check-valve into the upper por-
tion of the air-chamber. It will be seen that this operation is
precisely similar to that of a reciprocating air-pump, the rise
and fall of the water being precisely similar in effect to that of
a reciprocating piston. There will be a greater velocity of
water upwards in the air-cylinder than in its downward move-
ment, because the downward movement will have a velocity due
only to the pressure of the atmosphere, whereas the upward
movement has a velocity due to the pressure given it by the ac-
tion of the piston of the main pump. There will always be a little
water carried over with the air into the top of the air-chamber;
this excess of water serves a useful purpose in the fact that by
it the clearance in the air-pump is absolutely eliminated.
There is no uniformity of opinion among managers of mines
in general as to the utility of air-chambers on mining pumps,
the drift of opinion being against them, mainly because they
soon fill with water and become useless for the purpose
intended. Owing to the indifference regarding this important
detail of a pump, there has rarely been, until within a few
years past, any device by which the pressure, and especially
the volume of air, is maintained in the air-chamber, thereby
failing to secure the benefits of an elastic cushion in the
delivery-chamber of the pump. In the case of a single mine-
pump a large air-chamber is of very great advantage, and
under no circumstances should it be omitted.
If a duplex pump be used, the air-chamber is less essential
to smooth working; but even then it should not be omitted.
96
PUMPING MACHINERY.
VACUUM-CHAMBERS.
Definition.-Properly speaking, there is no such thing as
a vacuum-chamber belonging to a pump; it is an air-vessel
attached to the suction side of the pump to steady the flow.
The use of a vacuum-chamber is precisely the reverse of that
of the air-chamber, which has for its function the changing
of an intermittent flow into a continuous one; whereas the
vacuum-chamber has for its function the conversion of a
continuous flow into an intermittent one.
The flow of water into a pump is that due to the action
of the atmosphere alone, and once the column of water is
started its flow must be continuous if the best results are to
be expected. The effect of the vacuum-chamber is to take
away from the suction-chamber of the pump the water-
hammer and other disturbing influences consequent upon a
continuous flow into it, and from which the withdrawal of
water is intermittent. The air in the vacuum-chamber forms
an elastic cushion which will receive the excess of flow without
noise, and give it out again as silently as it received it. The
air is thus partially expanded and compressed at each wave or
impulse of the water flowing into the pump-chamber.
The size of a vacuum-chamber need not ordinarily be
more than one-half that of the air-chamber; a good practice
is to make the cubic capacity twice that of a single displace-
ment of a water-cylinder for a single pump. A less size
would answer for a duplex pump, but should one be needed
at all it ought to be a liberal one, and the above proportions
will probably be none too large.
Suction-pipes are not always short and straight, but quite as
likely to be long and crooked; it is for these latter that a
vacuum-chamber becomes a necessity, and to get the best
results it should be placed as near the pump as possible. The
form of a vacuum-chamber should be similar to that of the
AIR- AND VACCUM-CHAMBERS.
97
air-chamber; that is, it should have a much greater height
than diameter. The table of air-chambers will give good
proportions for vacuum-chambers also,
and to which the reader is referred.
It
A good design for a vacuum-
chamber is shown in Fig. 92. Its
form permits of application imme-
diately below the pump, and continue
the suction-pipe down to the water-
supply. There is nothing about the
design that will interfere with its being
made in any size, even for large
water-works pumping engines.
consists of two castings flanged, faced,
and bolted together, as shown in the
engraving. The enlarged mouth-
piece is advantageous in the fact that
a full supply can be had without the
formation of an eddy about the inner
edge of the opening when the pump
is working up to nearly the capacity
of the suction-pipe.
The introduction of the suction-
pipe into the air-vessel diminishes
the volume of the latter to that
FIG. 92.

extent; an increased height rather than diameter ought to be
given it to make good the space thus occupied.
9
E g
98
PUMPING MACHINERY.
:
CHAPTER VI.
SUCTION- AND DELIVERY-PIPES.
Suction-pipes should be as short and direct as possible.
In ordinary trade pumps the area of suction-pipes is approx-
imately two-thirds that of the water-cylinder, but if the con-
ditions are in any respect unusual, then the suction-pipe should
nearly, if not entirely, equal that of the water-cylinder of the
pump.
The suction side of the pump is subject to atmospheric
conditions wholly, and everything which would tend to re-
strain the free flow of water into the pump should be carefully
guarded against. It is not possible to dispense with angles
and bends in a suction-pipe, but much of the increased resist-
ance to the flow by reason of these bends can be overcome by
increasing the size of the suction-pipe; this permits a slower
movement of the water flowing towards the pump, so that the
resistance is largely reduced.
The velocity of flow in a suction-pipe should not exceed
200 feet per minute. It is not an uncommon practice to make
the suction-pipes for water-works pumps of the same diameter
as that of the water-cylinders, and as 100 feet piston-speed per
minute is the common speed for direct-acting pumps, it will
be seen that the flow is a very moderate one indeed. In the
specifications recently prepared by Mr. Freeman for the Un-
derwriter Pump, the suction-pipes were made unusually large;
for example, a
6 X 12 pump is to have a 6-inch suction-pipe.
7 X 12
9 X 12
IO X 12
8 "
CC
66
IO
66
66
12
66
66
པ
SUCTION- AND DELIVERY-PIPES.
99
These pumps are to be of the duplex type only, double-acting,
and run at 70 “revolutions" per minute. By revolution is
meant one complete circuit of the motion of any of the recip-
rocating parts of the pump, and for a duplex pump is equiva-
lent to four single strokes; a somewhat unusual speed except
for fire-pumps, as it is the equivalent to a piston-speed of 140
feet
per
minute.
The following table gives the relative proportions of water-
cylinders to that of suction- and delivery-pipes as used by the
writer with very satisfactory results:
TABLE V.
DIAMETERS SUITABLE FOR SUCTION- AND DELIVERY-PIPES FOR DUPLEX DIRECT-
ACTING PUMPS RUNNING AT 100 FEET PISTON-SPEED PER MINUTE.
WATER-CYLINDER.
SUCTION-PIPE.
DELIVERY-PIPE.
Diam-
eter.
Area.
Diam-
ter.
Area.
Velocity of
Flow at 100
Feet.
Diam-
eter.
Area.
Velocity of
Flow at 100
Feet.
Inches.
Inches.
Feet.
Inches.
Feet.
456 7∞ a
12.57
19.64
28.27
38.48
3456
7.07
12.57
178
156
19.64
143
28.27
136
8
50.27
6
28.27
180
63.62
8
50.27
126
IO
78.54
8
50.27 156.
I2
14
113.09
153.93
ΙΟ
12
78.54
113.09
2 3 4 5 6 7∞
3.14
400
7.07
277
12.57
224
19.64
196
19.64
256
28.27
225
38.48
204
144
50.27
224
136 IO 78.54
196
Suction-Pipes should be of One Diameter only from
End to End.-Enlargements are always to be avoided, be-
cause they interfere with that uniform rate of flow so essential
to the proper filling of a pump. The pipe should have a
continuous rise from the water-level of the source of supply
to the pump; any irregularity in the laying which would result
in the formation of an air-pocket in the pipe must be corrected
if good results are to be expected.
If the suction-pipe is larger in diameter than the opening
into the pump, a short conical piece of pipe, one end fitting
Uor M
100
PUMPING MACHINERY.
P
the pump and the other end fitting the suction-pipe, should
be made and attached either to the pump or as close to it as
possible.
Suction-pipes must be tight, absolutely tight; about
this there must be no mistake, as anything short of it means
uncertainty and loss of efficiency in working, if not a com-
plete failure of the pump to perform the service for which it
was intended.
A leaky suction-pipe must be tested joint by joint until
the leak is discovered and corrected. If the pipe has been
laid in a trench and covered over with earth, it may seem like
FIG. 93.
FIG. 94.


00
100
O
100
100
100 O
00
100
0 0 0 0
00
O
a great deal of labor and expense to lay it bare, but there is
no other way out of it. A practical method for testing a
suction-pipe is to put a blank flange over the lower joint in
the water, a similar one on the delivery-flange of the pump,
Mou
SUCTION AND DELIVERY-PIPES.
ΙΟΙ
fill the suction-pipe and pump with water, insert in any con-
venient place in the suction-pipe a pressure-gauge, then with
a small hand force-pump, unless another source of pressure
is had, get up a pressure of say 50 to 60 pounds per square
inch, and then shut off the force-pump connection with this
pressure on the suction-pipe and pump. Now watch the
pointer on the dial of the pressure-gauge, and if it loses
pressure it is certain that a leak exists somewhere; nothing
now remains but to search it out and fix it.
If the suction-pipe is not in a covered trench, but accessible
throughout its length to the water-supply, a leak can readily
FIG. 95.

»
be discovered when the pump is in use by the sound of the
air rushing into the pipe; having thus located the joint, for
that is where the leak is likely to occur, the exact location of
the hole or leak can generally be found by means of a lighted
candle, the in-rushing air diverting the flame towards the hole
in the pipe, if held near enough to be so influenced. Suction-
pipes should be provided with a strainer, to keep out floating
matter, such as twigs, leaves, fish, etc., if the supply be had
from a stream or a pond.
A strainer may be made in any one of a variety of ways,
and answer the purpose for which it is intended. A wooden
or iron skeleton framing covered with wire-cloth of say ½-inch
or 5%-inch mesh, carefully fitted around the bottom of the
9*
102
PUMPING MACHINERY.
FIG. 96.
suction-pipe and securely fixed in the stream, makes a good
strainer. A simple and effective strainer for small pipes is
shown in Fig. 93. It is nothing more than a thin pipe,
enough larger in diameter than the suction-pipe as to present
a sufficient area of opening through the holes in its outer
surface. There should be no holes in the
bottom. The combined area of these
holes should in no case be less than three
times the area of the suction-pipe; and if
the holes are likely to be covered by leaves,
the area should be four times that of the
suction-pipe. A vertical or slanting side
to a strainer is always to be preferred, be-
cause it is less likely to retain leaves, etc.,
on the strainer after the flow of water had
ceased at the stoppage of the pump.
It is frequently the case that a strainer
cannot be conveniently reached for cleaning
if attached to the end of the suction-pipe.
There would, then, be an advantage in
placing it near or attaching it to the pump.
The arrangement shown in Fig. 94 is a
good one for small pipes, but for large
pipes the strainer should be horizontal,
and may be placed at any convenient
location between the source of water-
supply and the pump.
Such a strainer is

illustrated in Fig. 95.
taken out and cleaned.
A combined strainer and vacuum-cham-
ber is shown in Fig. 96. Like the former,
the strainer-basket inside can be readily
As the outlet attaches directly to the
pump, no better place could be selected for the vacuum-
chamber than the top flange of the side-pipe which holds the
strainer; this chamber is provided at the top with a screw
plug, so that, should it be necessary to charge the pump
with water, as is often the case with long suction-pipes not
SUCTION- AND DELIVERY-PIPES.
103
wholly tight, it can easily be done through the opening thus
provided.
The strainer shown in Fig. 97 has a semicylindric vessel
located on one side of the side-pipe.
FIG. 97.
Holes are drilled

源
​through the flat side extending across the diameter of the
side-pipe; any floating matter which will not pass through
the holes collects in the strainer-vessel and can be quickly
removed. This is a more substantial arrangement than those
illustrated above, but it is more expensive to make, because a
larger diameter of side-pipe is necessary to get the same area of
opening that could be
secured through the
meshes of a wire-cloth
basket.
A foot-valve
should be attached to
all suction-pipes when-
FIG. 98.


ever of unusual length, or lift of water. In selecting a valve,
be sure that a clear water-way is had through the seat of at
least the area of the pipe with which it is to be used. Fig. 98
104
PUMPING MACHINERY.
is the common design, as sold in all supply stores, for com-
bined foot-valves and strainers in all sizes up to 4-inch wrought-
iron pipe.
A sectional elevation of a common form of butterfly-valve
is shown in Fig. 99.
Foot-valves of this type are usually
FIG. 99.
fitted with leather-faced
clack-valves seating on a
tool-finished cast-iron bot-

tom or seat.
The hinged
valves are usually secured
by the same centre-pin, or
through-going bolt, stops
being provided so that
when a full water-way is
had the valves cannot fall
back, but must return,
each to its own seat.
Foot-valve castings
should be amply strong,
so that any delivery leak-
age past the valves in the pump into the suction-pipe would
not split the foot-valve.
On large water-works valves it is quite a common practice
to put a safety- or relief-valve on the side of the foot-valve
casing or shell; this relief-valve being loaded or set for a
pressure less than that which would endanger either the foot-
valve or the suction-pipe, both of which are generally made
lighter than pressure pipes or valves for the same size or diam-
eter. This relief-valve is not intended to be anything else than
simply a leak in the suction-pipe, set to a certain known press-
ure, to prevent an accumulation of a higher pressure danger-
ous to the suction-pipe or foot-valve. This relief-valve need
not be a large one; an inch-and-a-half valve will answer for
an 8- or 10-inch pipe, and a two-inch valve for a 12- or 14-
inch pipe.
A foot-valve with a single india-rubber disk held between
metal plates is shown in Fig. 100. This design is suitable for
SUCTION AND DELIVERY-PIPES.
105
pipes up to 8 or 10 inches in diameter. The drawing shows
a spindle extending both above and below the valve passing
through bored
guides. This
is a method
of construction
which the writer
does not wholly
favor, as it is
difficult to get
the upper and
lower holes ex-
actly in line with
each other; but
for a foot-valve
having a rub-
ber face, as
shown in the
engraving, the
fitting need not
be very exact,
SO that play
FIG. 100.

L
enough can be had in the guides to allow the valve to seat
without any binding stress on the spindle.
The design
shown in Fig.
IOI Consists of
a number of
small valves
similar in con-
struction to
those used in
the water-end
of a pump.
This design
FIG. IOI.

MAAN
is not in so common use as the valve shown in Fig. 99,
nor does the writer regard it as favorably. Foot-valves
106
PUMPING MACHINERY.
should preferably be so designed that the seating would be by
gravity alone, and wholly unassisted by springs. The action
1
FIG. 102.
L
of a foot-valve
is different from
that of a pump-
valve, which al-
ternately rises
and falls at each
stroke of the
plunger. In a
properly-de-
signed pump the
foot-valve ought
not close from
the moment the
pump starts until
it is stopped, and
as
are
pumps
often in continu-
ous operation for
several hours or days, the springs, if any, would be under a
tension, the extent of which would be measured by the lift of
the valve.
By the use of a clack- or disk-valve seating by gravity no
such effects can follow, no matter to what height of lift the
valve may be subjected, and is wholly uninfluenced as to time.
The writer does not recommend the use of a single disk of
india-rubber, as illustrated in Fig. 102, for use as a foot-valve,
because the flow of the water being continuous and always in
the same direction, the valve is forced upwards against the
curved guard, and is likely to receive a permanent set or dish-
ing, and is in consequence less likely to fit the flat seat tight
enough to prevent leakage around its outer edge.
The combined foot-valve and strainer shown in Fig.
103 is from a design executed for the water-works at La
Chaux de Fonds. The flat plate on the bottom of the
་

SUCTION- AND DELIVERY-PIPES.
107
FIG. 103.
strainer rests upon a stone foundation. The upper flange
contains the rubber seating upon which the metal valve rests
when the pump is not in operation. This valve has a central
spindle passing through bored
guides, one each in the casting
above and that below the valve, a
method of construction not wholly
free of objections. In order to in-
sure perfect alignment the upper and
lower flanges should, after the proper
adjustment of each other to the valve,
be further fitted with dowel-pins, in
order that the upper and lower
guides may be exactly brought in
line, if for any reason it should be
necessary to take them apart.
Delivery - pipes need not be
more than one-half the area of the
water-cylinder for any ordinary ser-
vice. The movement of the water
in the delivery being subject to that
of the pump by which it is forced,
and not to atmospheric conditions,
makes the problem one of friction.
only, and this amounts to so small
a fraction of the total friction that it is seldom taken into
account. In fact, it is quite an unusual thing to ascertain
experimentally what the friction-loss in the water-end of a
pump really is. The allowance usually made to cover the
friction-loss in the distributing system is large enough to
include that of the pump also.
The sizes for delivery-pipes as given in Table V. are those in
very common use, and as the permissible velocity of discharge
is 400 feet per minute, it will be seen that for all except the
first size the dimensions are very liberal. For use in a
factory system the diameters of the delivery-pipes in all pumps

108
PUMPING MACHINERY.
TABLE VI.
FRICTION-LOSS IN POUNDS PRESSURE
For each 100 Feet of Length in Different Size clean Iron Pipes discharging Given Quantities of Water per Minute. Also Velocity of Flow in Pipe in Feet
per Second.-G. A. Ellis, C.E.
2-Inch.
3/4-Inch.
I-Inch.
1/4-Inch. 1½-Inch.
2-Inch.
22-Inch.
3-Inch.
4-Inch.
6-Inch.
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Gallons discharged
per Minute.
5 8.17 24.6 3.63 3.3 2.04 0.84 1.31 0.31 0.91 0.12
ΙΟ
15
20
25
30
35
40
45
50
75
100
125
150
175
200
250
300
350
400
450
500
•
•
16.3 96.0 7.25 13.0 4.08 3.16 2.61 1.05 1.82 0.47 1.02
10.9 28.7 6.13 6.98 3.92 2.38 2.73 0.97
14.5
50.4 8.17 12.3
18.1
78.0 10.2 19.0
12.3 27.5
•
•
•
0.12
5.22 4.07 3.63 1.66 2.04 0.42|
6 53 6.4 4.54 2.62
7 84 9.15 5.45 3.75 3.06 0.91
6.36 5.05
7.26 6.52 4.09 1.6
8.17 8.15
9.08 10.0
14.3 37 O 9.14 12.4
16.3 48.0 10.4 16.1
II.7 20.2
13.1 24.9
19.6 56.1 13.6 22.4
18.2 39.0
•
1.63 0.21 1.13 0.10
•
5.11 6.44 3.26 0.81| 2.27 0.35 1.28 0.09
7.66 5.32 4.9. 1.8 3.4 0.74
10.2 9.46 6.53 3.2
12.8 14.9 8.16 4.89
0.33
4.54 1.31 2.55 0.33 1.13 0.05
5.67 1.99
•
•
15.3 21.2 9.8 7.0 6.81 2.85 3.83 0.69 1.7
•
· 5
นา
9
ΙΟ
15
20
25
30
35
40
45
50
75
100
125
O.I
150
17.1 28.1 11.4
9.46 7.94 3.85
•
•
175
20.4 37.5 13.1
16.3
19.6
12.48 9.08 5.02 5.11 1.22 2.27 0.17
19.66 11.3 7.76 6.39 1.89 2.84 0.26)
28.06 13.6 11.2 7.66 2.66 3.4 0.37
15.9 15.2
200
250
300
8.94 3.65 3.97 0.5
350
4.73 4.54 0.65
400
6.01 5.11 0.81 450
18.2 19.5 10 2
20.4 25.0 11.5
22.7 30.8 12.8
30.8
7.43 5.67 0.96 500
SUCTION- AND DELIVERY-PIPES.
109

JO
Gallons discharged
per Minute.
Velocity in Pipe
per Second.
Friction Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
Pounds.
Velocity in Pipe
per Second.
Friction-Loss in
TABLE VI.-Continued.
FRICTION-LOSS IN POUNDS PRESSURE
For each 100 Feet of Length in Different Size clean Iron Pipes discharging Given Quantities of Water per Minute Also Velocity of Flow in Pipe in Feet
per Second.-G. A. Ellis, C.E.
8-Inch.
JO-Inch.
12-Inch.
14-Inch.
16-Inch.
18-Inch.
20-Inch.
24-Inch.
30-Inch.
250 1.59 0.07 1.02 0.03
0.71
0.01
250
500 3.19 0.25 2.04 0.09 1.42 0.04 1.04 0.017
750 4.79 0.53 3.06 0.18 2.13 0.08
0.8 0.009
0.63 0.005
500
750
1,000
6.38 0.94 4.08 0.32 2.84 0.13
2.08 0.062
1.6 0.036
1.26 0.02
1.02 0.012 0.72 0.005 0.45 0.002
1,000
1,250
7.97 1.46 5.1 0.49 3.55 0.2
1,250
1,500
1,750
9.57 2.09 6 12 0.7 4.26 0.29 3.13 0.135
7.15 0.95 4.96 0.38|
2.39 0.071
1.89 0.04
1,500
1,750
2,000
8.17 1.23 5.67 0.49 4.17 0.234
3.19 0.123
2.52 0.071 2.04 0.042 1.44 0.02
0.91 0.006
2,000
2,250
6.38 0.63
•
2,250
2,500
7.09 0.77 5.21 0.362
3.99 0.188
3.150.107
3.15 0.107
·
2,500
3,000
8.51 1.11 6.25 0.515
4.79 0.267
3.78 0.15
3.06 0.091
2.16 0.047
1.36 0.012
3,000
3,500
7.29 0.697
5.59 0.365
4.41 0.204
4,000
8.34 0.91
6.38 0.472
5.04 0.263
4.08 0.158
2.88 0.067
4.500
7.18 0.593
5.67 0.333
1.820.03
•
3,500
1.82 0.022
4,000
4,500
5,000
7.98 0.73
6.3 0.408
5.11 0.244
3.6 0.102
2.27 0.035
5,000
6,000
7.56 0.585
6.13 0.348
4.32 0.146
2.72 0.048|
6,000
7,000
7.15 0.472
5.04 0.196
3.18 0.065
7,000
8,000
8.17 0.612
8.170.612
5.76 0.255
3.63 0.083
8,000
9,000
6.47 0.323
4.08 0.105.
9,000
10,000
7.19 0.398
4.54 0.131
10,000
Friction-Loss in
Pounds.
Gallons discharged
per Minute.

110
PUMPING MACHINERY.
having water-cylinders 6 to 14 inches diameter, both inclusive,
could safely be reduced, beginning with a 3½-inch delivery-
pipe on the 6-inch pump, and ending with an 8-inch delivery-
pipe for the 14-inch pump.
Duplex pumps with water-cylinders 8, 10, 12, and 14 inches.
diameter are in common use in small water-works, and there
should be no reduction in the diameters of the delivery-pipes
for such a service.
How far vertically and horizontally can water be
sucked at a definite speed by a common bucket- or ram-
pump? This question was answered by Mr. James McCreath
in a paper read before the Mining Institute of Scotland in
1883, from which the following abstract is made:
The sucking of water being dependent on the pressure of
air, we may, for all practical purposes, consider air as a perfect
gas. The pressure which the atmosphere exerts is never
much less, at the level of the sea, than will support a column
of water 32 feet high, and for every 262 feet above sea-level
the pressure is about one-hundredth of itself less, according
to Rankine.
The proportion of the height of 32 feet to which water can
be raised in a pump depends upon the proportion of air within
the pump which the pump can extract, and this depends upon
its proportions. For example: A pump the bucket of which
at top of stroke is 20 feet above the level of the water, the
length of stroke being 2 feet and the clack being 16 feet above
the water, can never suck the water up to the clack, because
at the bottom of the stroke the air-pressure must be slightly
more than that of the atmosphere, and at the top of the stroke,
the air there occupying double the space, the pressure must
be slightly more than half of that, and therefore the water
will not rise to quite the half of 32 feet, which it would require
to do to reach the clack. If the pump be filled with water
from the outside, it will raise water until air again collects in it.
The following are examples of what the utmost rarefaction
of air producible by various lengths of stroke and distances
SUCTION- AND DELIVERY-PIPES.
III
between bucket and clack can do, 32 feet of water being
taken as equivalent to the atmospheric pressure:
Length of Stroke
in Feet.
I X 32 = 32
32.
2 X 32 = 64.
WITH THE LOW CLACK.
Greatest Height of Bucket at
Top of Stroke above Water
in Well which will admit of
Water being raised.
WITH THE HIGH CLACK.
Distance between Bucket
and Clack-Valves
Bottom of Stroke.
I.
I
•
•
at
2 √ 32 = 11.3 feet.
2 √ 64 = 16
Height Water will raise
to Clack.
Length of Stroke
in Feet.
I.
2
•
1/2 X 32 16
2/3 X 32
21/3
Ft
FIG. 104.
Top
of
Stroke
This raises water only to the clack. Very little of a longer
stroke would raise it to the bucket, for the air-space to be
exhausted lessens as the water rises above the clack; but some
allowance must be made for the water
returning through the valve while it is
closing, provided a sufficient pause at
end of stroke be not made to let it fall.
If the water get above the bucket, 20
either by suction or by being filled from
the outside, and the bucket be steadily
raised, the water will in time follow the
bucket until it reaches 32 feet above
the water in the well; but if it does
not reach the bucket, of course no
water will be pumped.

The height to which water will
rise in a ram-pump may be calcu-
lated in the same way, the space under
and around the ram being reduced to an
equivalent length of pipe of the diam-
18Ft
Bottom
of
Stroke
16F格
​Clack
eter of the ram between the bucket and the clack.
Provision
ought also to be made for the escape of air remaining around
the ram above the level of the exit-pipe.
II2
PUMPING MACHINERY.
It will thus be seen that the capacity of an ordinary pump
to clear itself of air depends directly upon the length of the
stroke, and, within certain limits, inversely as the distance
between the bucket and the clack.
Weisbach gives the following rules obtained by
experiment for the loss of energy of flowing water:
1. When flowing through short tubes whose diameter is
equal to their length, the water is contracted to 0.61 of the
area of the tube.
2. When passing round an elbow at right angles, 0.984 of
(or very nearly the whole) head corresponding to the velocity
is lost.
3. When passing round a circular bend, whose radius of
curvature is to the radius of the cross-section of the pipe as IO
is to 4, the loss of head is equal to about of that correspond-
ing to the velocity.
4. When passing through a valve similar to that shown in
Fig. 104, where the cross-section of the aperture to that of the
pipe is 0.535, the co-efficients of resistance are as follows:
•
Angle of opening.
Co-efficients of resistance.
Angle of opening .
Co-efficients of resistance.
·
15° 20° 25° 30° 35°
30 20
•
90 62 42
40° 45° 50° 55° 60° 65° 70°
14 9.5 6.6 4.6 3.2 2.3 1.7
Let the foregoing rules be applied to a bucket-pump having
bends under the clack, as shown in Fig. 105, the length of
pipe from the bottom of the suction-pipe to the bucket at half-
stroke = 100 feet, the height of the bucket at the top of the
stroke above the surface of the water in the well = 20 feet,
and above the clack at the bottom of the stroke 4 feet, the
=
stroke 5 feet, and speed required, 12 strokes a minute, the
pump being driven by a crank whose rotating motion is uni-
form.
In order to find whether a pump can clear itself of
air, assume that the weight of the bucket-valve be such as to
SUCTION- AND DELIVERY-PIPES.
113
reduce the effective stroke for air to 434 feet, and the weight of
the clack-valve and the slip of water past the valve to take 6
inches from the column of water capable of being raised. The
length of the stroke,
434 feet divided by 9
feet, the distance be-
tween the clack and
the bucket at top of
stroke, gives extraction
434
of
9
X 32 = 17
feet; and deducting 6
inches for weight of
clack - valve and for
slip, leavės 16 feet 6
inches which the water
would rise to the clack.
But the bucket at bot-
tom of stroke is only
15 feet above the water
in the well, and there
is, therefore, a large
margin to cover im-
perfections in the
pump.
In order to ascer-
5 Ft.-
Aft
11 Ft:
FIG. 105.

-Top_of_
Top of stroke.
tain whether the water will
follow the bucket at its speed,
let the resistances be first ascer-
tained, and then the time required
to fill the working-barrel with the
remaining pressure.
-Bottom of stroke.
Clack
The mean velocity of the pump assumed is 2 feet per
second, and the resistances at this velocity are,-
h
10*
114
PUMFING MACHINERY.
Loss of Head
in Feet.
Snore-holes and turn into pipe when sum of \
(2 x 100) 3
2
area of snore-holes = area of suction of pipe
S
0.167
64.4
22
Circular bend.
= 0.012
64.4
22
Square elbow..
= 0.062
64.4
22
Valve at 20°
X 62 = 3.857
64.4
Friction of the water in the pipes, say of 12 inches diameter = 0.165
Total resistance due to square of mean velocity.
4.263
Weisbach shows that the resistance is as the mean square
of velocity, and that in a uniformly accelerated or uniformly
retarded motion this is twice the square of the mean velocity,
and in a crank motion 1.645 times the square of the mean
velocity. Assume, meanwhile, that the water follows the
bucket, the total resistance will be equivalent to 4.2635 X
1.645, equivalent to a head of about 7 feet. The bucket at
bottom of stroke being 15 feet above the water in the well,
the pressure at first for raising the water and overcoming
resistances is equivalent to 17 feet; but deducting 7 feet for
resistances, the pressure available for giving motion to the
water will be only 10 feet.
Taking the following approximate formula for time of fill-
ing the working-barrel (with the acceleration modified), suit-
able where the extremes of acceleration do not vary greatly,
we have,-
✓ height
10--2.5
Time = 2
64.4 X
twice the mean acceleration 2
100
2.034
seconds.
But the time the bucket takes to make the half-stroke is
22 seconds, so that there is nearly half a second to cover the
imperfections of the pump and any hinderance there may be
at the beginning of the stroke from the interference of the
bucket with the motion of the water. If an attempt be made
to drive the pump at 17 strokes, the water will lag behind the
SUCTION- AND DELIVERY-PIPES.
SUCTION
115
bucket, and they will meet on the return stroke, the water
moving with a velocity of about 4.8 feet per second and the
bucket with a velocity of about 1.66 feet per second, together
about 6.46 feet per second.
If the water were so far behind the bucket as to meet it
at half-stroke, a much more serious shock would arise, because
in addition to the momentum of the water the momentum
of machinery at a higher velocity, backed by the steam-
pressure, would have to be met.
The diagram, Fig. 106, will show the relative motions of
the bucket and the water, supposing the average resistance to
FIG. 106.
12 Strokes per min.
Time=2.5 sec.
17 Strokes per min.
Time-1.76 sec.


2
-Space 5 feet
2
1.76
be uniform. The speed of the water would, however, if un-
retarded, be actually greater at the beginning of the stroke,
and it will be less at the middle of the stroke; the top will
therefore be reached a little later than shown.
The foregoing calculations show that the largest resistance
is due to the clack-valve, and therefore special attention
requires to be given to find a valve giving little resistance
116
PUMPING MACHINERY.
when either the height or distance which the water has to be
sucked is such as to raise a doubt of the pump working well.
What would be still more effective would be to double the
area of the pump and drive at half the speed, and so reduce
both the resistances and the work lost in the motion of the
water to a fourth. With the pump already instanced the same
quantity of water could be sucked 680 feet as freely with a
17-inch bucket and pipes going 6 strokes a minute as could
be done 100 feet with a 12-inch bucket and pipes going 12
strokes a minute; and if, in addition, the stroke could be
lengthened downwards 4 feet, retaining the same velocity, the
water could easily be drawn 1415 feet.
The calculation of pressure caused by a shock in
pumps may be obtained as follows: Calculate the
amount of work in compressing the water and extending the
pipe, which would be equivalent to the work due to the fall of
the water, keeping in view that a load suddenly applied to an
elastic substance will produce double the strain that the same
load would do if applied gradually.
This may be illustrated in the case of a spring-balance and
a half-pound weight. If the weight be put suddenly in the
balance it drops down until the index shows nearly a pound
of pressure, and then rebounds, and, as it oscillates alternately
above and below the position of the half a pound of steady
load, gives a very good illustration of what is called a "live
load."
Example from Actual Experience.-The height of a
column of water is 339 feet; diameter of pipe, 26 inches. A
bucket-piece was broken by the water being suddenly stopped
when descending with a velocity of 5 feet per second.
Thickness of pipe, 134 inches. We may take the compressi-
bility of water at 294.000 of itself for each pound of steady
load; extensibility of iron 17,000,000 of itself for each pound
of steady load. Weight of water = 147 pounds per square
inch.
1
SUCTION- AND DELIVERY-PIPES.
117
Let it be assumed first that the work due to the fall has to
be inet wholly by the compression of the water, then
= half of ultimate pressure X amount
of compression.
The work due to velocity of 5 feet
V2
X weight of I foot of water
2g
}
25
X
64.4
X.433 =
2
41 x
294,000
x=314 pounds per square inch.
This is the ultimate pressure required to meet the momen-
tum of the water if moving horizontally, but the column
above referred to being vertical, it has upon contact the
additional compressing force due to a sudden application of a
load of 147 pounds,—viz. :
Double the weight applied
Add pressure due to stopping 5 feet velocity
Total pressure if work be met only by water
294 pounds.
314
= 608
(C
The work, however, is partially met by the stretching of the
iron pipe, and the proportions borne by each are:
26" x 26" X
Work on Water.
3.1416
Х
4
I
294,000
: 26" X 3.1416 X -X
Work on Iron.
26
32/ 17,000,000
I
::.00085:
.0000168; and 8668: 608 :: 168: 12. The total actual theoretical pressure is
therefore 608 12 = 596 pounds per square inch.
We have thus, required to stop velocity of 5 feet
And double the weight of column .
Total pressure due to the fall..
Proportion borne by Iron.
Pounds. Pounds. Pounds.
•
314
294
-
6
6
308
288
•
596
This is the simplest form of shock. When a bucket or
ram is suddenly stopped in descending, the calculation becomes
more complicated from the working-barrel being partly
occupied by iron or wood,* which are respectively 60 and 6
times less compressible than water, and from the momentum
of the machinery backed by the steam-pressure having also to
be met. The calculation, however, of this simple shock
* In mining-pumps only.
118
PUMPING MACHINERY.
makes more intelligible the thickness of iron practically found
necessary in pumps, keeping in view that repeated shocks.
weaken cast iron by a half. It likewise shows that the press-
ure due to the shock depends upon the velocity of the water
when suddenly stopped and the resistance to compression of
itself and the resistance to extension of the vessel containing
it, and as in an ordinary pressure-gauge the velocity of the
water and the resistance to extension of the vessel containing
it are both less than in the working-barrel, the pressure
indicated must also be less.
Air-vessels will be found useful in lessening shocks
in pumps; but it would appear that within the working-
barrel they cannot do so to the extent generally supposed.
The time taken to compress water sufficiently to stop a 5-feet
velocity and raise 314 pounds is only part of a second.
This time must elapse before the water below the air-vessel
can begin to flow back, and the extent to which the water
above the air-vessel can be relieved of the shock due to its
velocity must depend on the quantity of water in the air-
vessel and the direction and rate of its motion at the instant,
as well as the quantity of air, because the quicker the change
of motion and the larger the mass of the water the greater
the force required, and an abrupt change of motion requires
an infinite force. If there be one foot in length of water be-
tween the main column and the vessel, and that at rest at the
beginning of the shock, it will attain the same velocity as the
descending water when .866 of the full compression due to
the shock has been reached; and the pressure will probably not
much exceed this proportion thereafter at or above the entrance
to the vessel, provided there be sufficient air above the water.
About 1½ feet of air would suffice for a shock capable of
producing a pressure of 600 pounds per square inch. No
substantial relief would be given if two feet of water inter-
venes between the main column and the air in the vessel.
Water may be much more readily driven in or out of the air-
vessel to or from the bucket than can the whole column of
SUCTION- AND DELIVERY-PIPES.
119
water be moved, because the mass of water to be moved by
the same pressure is much less, and probably the air-vessel is
as useful in driving the water downwards to fill a vacancy
and so permit the fall of the column above as in forming a
cushion for the fall.
Pumping Hot Water.-Experiences are not wanting in
which the difficulty of pumping hot water has been made
clearly manifest, but recorded experiments as to its practical
limitations are quite rare. It has been the author's practice
to always insist that water having a temperature higher than
120° shall flow into the suction-chamber of the pump; if this
be done the matter then becomes simply one of a proper
selection of pump and valves.
Mr. H. J. Coles, Inst. C. E., London, made some experi-
ments, the results of which were embodied in a paper presented
to the above institution, an abstract of which is here given:
The depth from which hot water of a given temperature
may be pumped can be theoretically deduced from the formula
"6
B C
log. p = A
and the inverse of the above
A
14
log. p
C
2
4
+P) - D
in which p absolute pressure, t = absolute temperature,
and A B C are constants. Both formulas are quoted from
page 283 of Rankine's ' Rules and Tables;' or the same results
may be more readily obtained from Regnault's Tables given
at page 263, 'Ganot's Physics' (third edition).
"Having had, however, frequent inquiries as to what could
be done in actual practice by donkey-pumps for feeding
boilers from hot-wells, etc., and supposing that possibly suf-
ficient vapor might be evolved from the water at lower tem-
peratures than those ascertained by the above rules to dimin-
ish seriously the quantity pumped, the author carried out a
series of experiments to obtain actual results.
The donkey-pump employed was single-acting, having a
:
120
PUMPING MACHINERY.
ram 3 inches in diameter, with a length of stroke of 7 inches.
The pump was elevated to various heights; but the results
being so nearly alike, allowing for difference in height and
temperature, the table given below for 15 feet may be taken as
typical of all. The supply-tank stood on the ground, the
water in it being heated by a jet of steam. The suction-pipe
was led direct to the valve-box with only one bend, and the
delivery-tank was elevated to about the same level as the
pump, the water being discharged through a valve loaded to
60 pounds per square inch. A large cock, fitted to the bottom
of the delivery-tank, was kept open while the speed of the
pump was being regulated, and was shut as soon as the trial
commenced. A certain depth of water always existed in the
tank while the cock was open; this was carefully gauged and
deducted at the end of the trial. The speed of the pump was
regulated as nearly as possible to that given in the first column,
and on each trial, as soon as the exact number of strokes was
completed, the pump was stopped.
"It will be seen that the results agree closely with those
given by the above rules, the falling off in the quantity at the
higher temperatures being most probably due to the friction
of the water in passing through the pipes, valves, etc. It will
also be observed that the speed of the pump had to be reduced
for the higher temperatures, the speeds stated in the list being
found to give the best results.
TABLE VII.
RESULTS OF EXPERIMENT NO. 3 WITH THE PUMP 15 FEET ABOVE THE WATER-
Revolutions
per Minute.
ន ន ន ន ន ន ទ
50
LEVEL.
Temperature, Fahrenheit. Hot Water pumped per Minute.
Degrees.
70
100
Cubic Inches.
3430
3430
I 20
3430
140
3430
160
3286
170
2682
180
2180
SUCTION- AND DELIVERY-PIPES.
121
"Each quantity stated is the mean of several trials. Above
180° Fahrenheit scarcely any water could be pumped. Ac-
cording to Regnault, 185° would be about the limiting tem-
perature at 15 feet."
FIG. 107.

101
RELIEF PIPE.
H
CHECK VALVE
CHARGING PIPE,
YACUUM CHAMBER
SUCTION PIPE
FOOT VALVE.
1
| DELIVERY PIPC,
•
A general arrangement for piping a pump is shown
in Fig. 107, subject, of course, to such modifications as are
necessary to adapt it to any particular location.
F
II
122
PUMPING MACHINERY.
The suction-pipe should be as short and direct as possible,
and never less in diameter than the opening leading into the
pump. A foot-valve should be used, and if a strainer is not
attached to it there should be a strainer-box built around it
to keep out floating-matter, fish, etc. The check-valve should
bolt directly against the flange of the delivery-elbow.
A charging-pipe should lead from the pressure-side of the
check-valve into the suction-pipe. By opening the valve
shown in the charging-pipe, the water from the reservoir or
tank will flow into and fill the suction-pipe and the pump, thus
expelling all the air.
A relief-valve and pipe is shown immediately back of the
check-valve. This valve should be opened when filling the
suction-pipe and pump with water, in order to allow the air to
escape. The relief-pipe should lead to a drain or any place
where the overflow would not cause inconvenience.
A vacuum-chamber is shown in dotted lines, together with
a tee in the suction-pipe. This is not always necessary, but
when it is it should be placed near the pump. If the suction-
pipe goes into a deep well, and the distance from the well to
the pump is short, the delivery-elbow may be changed to a
tee, and the vacuum-chamber placed directly over the high
lift.
The steam-pipe should be so arranged that the water of con-
FIG. 108.
densation, when the pump is not
running, will drain back into the
boiler. The exhaust may be led
to any convenient point for escape
into the atmosphere, or it may be
used in steam coils for heating.

Check-valves for pumps should
always be straight way and fitted.
with hinged or swinging valves, and
ought to be attached directly to the
delivery-flange of the pump, if practicable. The diameter
should equal in size that of the delivery-opening of the pump.
SUCTION- AND DELIVERY-PIPES.
123
:
There are several good straight-way valves now furnished the
trade in the smaller sizes, so that there is less need than
formerly for the employment of spindle check-valves fitted
to seats at right angles to the flow. Fig. 108 is a sectional
elevation of a Pratt & Cady check-valve which the writer
has found to be quite satisfactory. Check-valves for diam-
eters suited to 3-inch wrought-iron pipe and less should
always be made of tough brass. For larger sizes a good de-
sign is shown in Fig. 109. It will be observed that the valve-
face is nearly at
right angles to the
flow; the position
of the fulcrum and
the angle of the
valve-face insures
prompt return to
its seat. Such
valves are regu-
larly made in all
sizes, from 4 inches
to 24 inches diam-
eter of opening;
larger sizes are
made to order, and
have usually two
or more rectangu-
FIG. 109.

lar hinged valves, rather than one large one. The shell and
valve are of cast iron; the valve is faced with leather and held
in place by a brass ring or plate underneath; throughgoing
rivets securely fasten all three together, a detail not shown in
the drawing.
!
124
PUMPING MACHINERY.
Feet Head.
per
Square Inch.
Pressure
Feet Head
per
Pressure
Square Inch.
TABLE VIII.
PRESSURE OF WATER.
The Pressure of Water in Pounds per Square Inch for every Foot in Ieight to
300 Feet; and then by Intervals, to 1000 Feet Head.




Feet Head.
per
Pressure
Square Inch.
Feet Head.
per
Pressure
Square Inch.
123456 7∞ a
I
0.43
1.30
1.73
2.16
$4079
44
19.05 87 37.68
0.86 45 19.49
38.12
46
19.92
38.55
47 20.35 90 38.98
130 56.31
173 74.94
88
131
56.74
174
75.37
89
132 57.18
175
75.80
133
57.61
176
76.23
48
2.59 49 21.22
3.03 50 21.65
20.79
91 39.42
134
58.04
177
76.67
92 39.85
135
58.48
178
77.10
93 40.28
136
58.91
179
77.53
9
3.46
389
51 22.09
94 40.72
137 59.34
180
77.97
52 22.52
95
41.15
138
59.77
181
78.40
IO
II
12
4.76
13
5.63
14
6.06
15
16
17
6.49
6.93 59
7.36 60
18
7.79
61
19
20
8.66 63
21
22
9.09
24
9.53 65 28.15
23 9.96 66 28.58
10.39 67
25
10.82
26
Imtmo no aOENMJKONO
4.33 53 22.95
96| 41.58
139
60.21
182
78.84
54 23.39
97 42.0I
140 60.64
183
79.27
5.20 55 23.82
98 42.45
141 61.07
184
79.70
56
24.26 99 42.88 142 61.51
185
80.14
57
24.69 100 43.31
143 61.94
186
80.57
58
25.12 ΙΟΙ 43.75
144
62.37
187
81.00
25.55
102 44.18
145 62.81
188
81.43
25.99
103
44.61
146 63.24
189
81.87
26.42
104
45 05
147
63.67
190
82.30
8.22 62 26.85 105
45.48
148
64.10 191
82.73
27.29 106
45.91
149
64.54
192
83.17
64 27.72
107 46.34
150 64.97 193
83.60.
108 46.78
151 65.40 194
84.03
68
29.02
29.45
109 47.21
IIO
47.64
III 48.08 154
152 65.84 195
84.47
153 66.27 196
84.90
6670 197
85.33
II.26 69 29.88 112 48.51
155
67.14 198
85.76
27
11.69
70 30.32 113 48.94
156
67.57
199
86.20
28
12 12
71 30.75
114 49.38
157 68.00
200
86.63
29
12 55
72 31.18
115 49.81
158
68.43 201 87.07
30
12.99
31
13.42
35
wwww.
32
15.16
36
15.59
37
38
16 02
16.45
39
16.89
777∞ ∞ ∞ ∞
73
31.62 116 50.24
74 32.05 117 50.68
13.86 75 32.48 118 51.11 161
33 14.29 76
32.92 119 51.54
34 14.72 77 33.35 I20 51.98
78 33.78
121 52.41 164 71.04
79 34.2I 122 52.84 165
159
68.87
202
87.50
160
69 31
203
87.93
69.74
204
88.36
162
70.17
205
88.80
163 70 61
206
89.23
207
89.66
71.47 208
90.10
80
81
34.65 123 53.28 166 71.91
35.08 124 53.71 167 72.34
209 90.53
210
90.96
82
40
17.32
83
4I
17.75
42
18.19
43
18 62
86
85
86
37.25
35.52 125 54.15
35.95 126 54.58
84 36.39 127 55.01
36.82 128
129
168
72.77
169
211 91.39
73.20 212 91.83
55.44
55.88
170 73.64 213
171 74:07 214
172 74.50
92.26
92 69
215 93.13
Feet Head.
Pressure
per
Square Inch.
SUCTION- AND DELIVERY-PIPES.
125
Feet Head.
Pressure
per
Square Inch.
Feet Head.
per
Pressure
Square Inch.
Feet Head.
TABLE VIII.—(Continued.)


per
Pressure
Square Inch.
216
93.56
237 102.60
258 111.76
279 120.85 300
I29.95
217
219
218 94.43 239 103.53
94.86
93.99 238 103.09
259 112.19
280 121.29
310
134.28
260 112.62
281 121.72
320
138.62
240 103.96
261 113.06
282 122.15
330
142.95
220
221
222
95.30 241 104.39
262 113.49
283
122.59
340
147.28
95.73
242 104.83
263 113.92
284 123.02
350
151.61
96.16 243 105.26
264 114.36
285 | 123.45
360
155.94
223
96.60
244 105.69
265
265 114.79
286 123.89
370
160.27
224 97.03
245 106.13
266 115.22
287 124.32
380
164.61
225
97.46
246 106.56
267 115.66
288 124.75
390
168.94
226 97.90 247 106.99
268 116.09
289 125.18
400
173.27
227
98.33 248 107.43
228 98.76 249 107.86
269 116.52
290 125.62
500
216.58
270 116.96
291 126.05
600
259.90
233
229 99.20 250 108.29 271 117.39
230 99.63 251 108.73
231 100.06 252 109.16
232 100.49 253 109.59
100.93 254 110.03
234 101.36 255 110.46
235 101.79 256 110.89
236 100.23 257 111.32
292 126.48
700
303.22
272 117.82
293 126.92
800
346.54
273 118.26
294 127.35 900
389.86
274 118.69
295 127.781000
433.18
275 119.12
296 128.22
276| 119.56
277 119.99
278 120.42 299 129.51
297 128.65
298 129.08
Feet Head.
per
Pressure
Square Inch.
Feet Head.
Pressure
per
Square Inch.
126
PUMPING MACHINERY.
CHAPTER VII
WATER-END DESIGN.
THIS subject has already been touched upon, and especially
in the chapter on valves, as the latter detail is so intimately
connected with water-end design that it is not easy to sepa-
rate them. The illustrations given in this chapter are so
nearly self-explanatory that little descriptive matter will be
necessary: The writer regrets that the small size of the en-
gravings necessary to their insertion in a printed page of the
dimensions selected for this publication would not permit of
both scale and dimensioned drawings to be given, but the
illustrations themselves represent designs now in successful
use, so that the mere fact of such presentation in convenient
form will probably not fail to be useful to engineers and others
not interested in the design and manufacture of pumping
machinery, but who are desirous of knowing the interior ar-
rangement of the pumps now offered by builders to the
general public.
Length of Stroke and Piston-Speed.-In ordinary
short-stroke pumps the capacity of the water-end is limited
not so much to the piston-speed in feet per minute as by the
number of times a valve can safely and noiselessly open and
close in a given time. To assume 100 feet per minute as an
ordinary speed for pumps has been a time-honored practice;
it is obvious, however, that for short strokes it imposes an
injurious rate of speed, to which a pump should not, in regular
service, be subjected. For example, a pump having
WATER-END DESIGN.
127
3-inches stroke must make 400 strokes per minute.
4
66
66
56
78
66
*
339
300
66
240
66
66
35
200
171 +
150
66
66
IO
12
120
(6
66
66
66
CC
100
66
As the above list of strokes represents lengths commonly
in use, it needs no argument to show the impracticability of the
100-feet basis of comparison; for pumps having a stroke of
six inches and less, the number of strokes as given above is
too great for continuous service.
TABLE IX.
SPEED AND CAPACITY OF PUMPS.
•
A piston-speed of 100 feet per minute is considered an ordinary speed for direct-
acting pumps.
But in boiler-feeding, pumping under heavy pressure, or moving
hot liquids, a slower speed is advisable. In fire-pumps, on the contrary, where a
high velocity and large volume are imperative, the speed may exceed 200 feet
per minute, if the valve area of the pump is sufficiently large.
THEORETICAL CAPACITY OF 100 FEET SPEED OF PISTON OR PLUNGER PER
MINUTE.
I
4.08
3
36.75
6
147
14
800
1%
5.16
3/4
43.13
6½
172
15
917
11/4 6.38 3½
50.02
7
200
16
1044
138
7.71
334
57.42
7/2
229
18
1321
1½
9.18
4
65.34
8
261
20
1631
15%
10.78
44
73.76
82
295
22
1974
134 12.58
42
82.7
9
330
24
2350
18 14.35
434
92.14
9%
368
26
2757
2
1633
5
102.0
ΙΟ
408
28
3270
24
20.67
54
II2.0
10½
450
30
3670
234
2/2 25.52
30.88
52 123.0
534 135.0
II
494
32
4176
12
587
34
4715
This is the theoretical performance; practically the strokes will slightly exceed
that number for the quantity stated. In a duplex pump the number of gallons
delivered per minute is found by multiplying the displacement of one plunger by
twice the number of strokes.
128
PUMPING MACHINERY.
TABLE X.
SHOWING THE NUMBER OF STROKES REQUIRED TO ATTAIN A PISTON-SPEED
FROM 50 TO 125 FEET PER MINUTE FOR PUMPS HAVING STROKES FROM 3
TO 18 INCHES IN LENGTH.
Length of STROKE IN Inches.
Speed of PIS-
TON IN FEET
PER MINUTE.
3 4 5 6 | 7 | 8 | 10 | 12 | 15 | 18
NUMBER OF Strokes PER MINUTE.
50
200 150
120 100 86 75.0 60 50
55
220 165
132 IIO 94
82.5 66
55
60
240
180
144 120
103
90.0
72
60
65
260 195
156
130
III
97.5 78
65
70
280 210
168
140
I 20
105.0
84 70
75
300 225
180
150
128
112.5
75
80
320 240
192
160
137
I20.0
96
8
85
340. 255
204
170
146
127.5 102
85
90
360 270
216
180 154
135.0 108
90
95
380
285
228
190
163
142.5
114
95
100
400 300
240 200
171
150.0
I 20
100
105
4.20 315
252 210 180
157.5
126 105
IIO
440 330 264 220 188
165.0
132
110
115
120
460 345
125
276 230 197
138 115
480 360
288 240 206 180.0 144 120
500 375 300 250
172.5
214 187.5 150 125 100 83
GIFMLSJONRIJ8288
33
37
40
43
56
47
60
64
76
84
96
FOMNOMDR~KO♡
50
53
60
63
67
70
73
77
80
Note.-To find the number of plunger displacements in a duplex pump multi-
ply the number of strokes as given above by 2.
Piston Water-End with Wing-Valves.-Fig. 110 is a
sectional elevation of a duplex pump for feeding steam-boilers,
or for any other service where a pressure not exceeding 150
pounds per square inch may be required; it is of the piston
pattern, and lined with a drawn brass tube, carefully fitted and
then forced into the bored casting, after which the ends of the
tube are expanded by calking or riveting over a slight depres-
sion included in the casting. The holes for the valve-seats
may be bored with a straight taper of say I inch to the foot, as
shown in the engraving, or a taper tap may be used and the
valve-seat screwed in. This water-end represents a pump of
2 inches bore by 4 inches stroke, and for this size it has been
the practice to drive the valve-seats in on a taper. This draw-
ing shows a joint immediately above the water-cylinder, and
WATER-END DESIGN.
129
another joint immediately above the top of the delivery-valve
seats, the delivery- or force-chamber completing the main parts.
The holding-down bolts extend through from the top of the
lugs included in the force-chamber casting, and screw into
•
FIG. IIO.

SUC.
tapped holes in the water-cylinders below. The piston may
be for fibrous packing, or fitted with metal rings. The valves
are of gun-metal, as are also the seats. This drawing is
shown with wing-valves with mitre joints.
Piston Water-End with Ball-Valves.—Fig. III is the
same water-end as described above, except that ball-valves
and seats are fitted for the handling of thick stuffs like
molasses, etc.
Plunger-Pump.-A sectional elevation of a water-end.
having a plunger and ring is shown in Fig. 112. The ring
DEL'Y
i
130
PUMPING MACHINERY.
slides into a bored cavity in the water-cylinder and is held in
place by bolts, one of which is shown in the drawing. The
plunger slides through the ring, and is not provided with any
means of adjustment for wear. The drawing is from a pump
having plungers 4 inches in diameter by 6 inches stroke. The
FIG. III.

SUC.
valve-seats are screwed in; the valves are of india-rubber for
cold water, and may be fitted with metal valves or vulcanite
composition, if so desired, for hot water. The delivery-valve
seats are screwed into a valve-plate held between the water-
end casting and the force-chamber.
A piston water-end with a removable lining is shown
in Fig. 113. The water-cylinder is bored at the three or more
points of support, and the lining is turned to fit; the support
or rib at the rear end is machine-faced, as is also the lining,
so as to make true surfaces for bolting together. Linings
DELIVERY
WATER-END DESIGN.
131
should be made of gun-metal, and must be sound castings,
free from imperfections and accurately bored. By a proper
spacing of the drilling the lining can be arranged for turning
FIG. 112.

10000
around in position so as to present a new wearing surface
on the bottom, should it ever become necessary through the
abrasive action of gritty water passing through the pump.
132
PUMPING MACHINERY.
The suction- and delivery-valves are both above the piston,
so that once the pump is charged with water it will always
remain so. The valve-seats are intended to be screwed in.
This drawing partially represents a duplex water-end having
7 inches bore by 12 inches stroke.
FIG. 113.

Worthington Plunger-Pump.-The sectional elevation
of a Worthington plunger-pump, shown in Fig. 114, is thus
described in their catalogue: The double-acting plunger
shown at B works through a deep metallic packing-ring,
bored to an accurate fit, being neither elastic nor adjustable.
Both the ring and the plunger can be quickly taken out, and
either refitted or exchanged for new ones at small cost, and
1
WATER-END DESIGN.
133
FIG. 114.

D
B
C
if it be desired at any time to change the proportions be-
tween the steam-pistons and pumps, a plunger of somewhat
larger size, or decreased to any smaller diameter, can be
12
134
PUMPING MACHINERY.
readily substituted. As exact proportions between the power
and work are always desirable, if not necessary, this is a very
important advantage.
This system of renewal of the working parts has proved by
long experience to be the least expensive and most satisfac-
tory for ordinary work. The plunger is located some inches.
above the suction-valves to form a subsiding-chamber, into
which any foreign substances may fall below the wearing-
surfaces. This enables it to work longer without renewal
than the usual form of piston-pump, especially in water con-
taining grit or other solid material. The water enters the
pump through the suction-chamber C, through the suction-
valves, then passes partly around and partly by the end of
the plunger, through the force-valves, nearly in a straight
course, into the delivery-chamber D, thus traversing in a very
direct and ample water-way. The bottom and top plates fur-
nish a large area for the accommodation of the valves. These
consist of several small disks of rubber, or other suitable
material, easy to examine and inexpensive to replace.
The drawing shows the usual central partition separating
the two ends of the pump. It is the practice to bore out this
central partition and then insert a permanent collared brass
bushing, into which is fitted the plunger-ring, also provided
with a collar, one side of which abuts against the bushing,
and the other side adapted to receive a cast-iron ring for
holding in place, this latter ring being securely held by bolts
not shown in the drawing.
The accumulated facts of a large and successful experience
with plungers simply fitted into a bored ring, as compared
with packed plungers, so clearly set forth by the late Mr.
Worthington, cannot fail to interest the reader. His close
study of the peculiarities of action of his duplex pump led
him very early in his practice to adopt a metal packing-ring,
without elasticity or provision for adjustment, a detail which,
after forty years of continuous trial, is the only form of pack-
ing ordinarily recommended by his successors, because noth-
ing better has been found to take its place. Let it be noticed
WATER-END DESIGN.
135
that engines of the duplex type have little or no momentum
to help out the stroke; they therefore will not allow of any
inequalities or tight places in the packing. A cylinder or a
plunger always wears fastest in the middle of the stroke; if
the packing were adjusted to fit the middle, it would bind
enough at the ends to embarrass the motion. He detected a
tendency at times in elastic pump-packings to collapse upon à
plunger or expand against a cylinder with such force as to
produce much unnecessary friction.
The wear of the plungers in the non-adjustable metallic
rings is not as great as one might be led to suppose, and for
reasons that will appear obvious on examination. The rings
are made deep, and thus afford ample bearing surfaces. The
water-ways surrounding the plungers constitute subsiding-
chambers into which hurtful material can settle away from the
moving parts. (See Fig. 114.) The protrusion into the forcing
side of the pump of the plunger while in motion tends to
carry away from the ring any such material that would other-
wise be forced under it. The thin film of fluid that may enter
the space between the ring and the plunger is only sufficient.
for proper lubrication of the parts, and reduces their frictional
resistance to a minimum. As the pressures are reversed when
the pump changes its stroke, little or none of this water leaks
past the ring.
In such cases, however, as seem, by reason of unusually
gritty water or of excessive pressure, to demand adjustable
packing, any one of the several designs shown in this chapter
will give satisfaction, if the details are properly worked out.
All of these packed plunger water-ends, except one design,
are arranged with exterior stuffing-boxes, so that the packing
is readily inserted and adjusted to compensate for wear. The
stuffing-box glands act as sufficient guides or bearings for
the plungers. All leakage in the water-cylinders is in this
arrangement prevented, as the working parts are displayed to
the engineer at all times. This form of water-end is largely
used, especially in the Western States, for water-works sup-
ply; it is the type almost always selected for mines and for
136
PUMPING MACHINERY.
iron- and steel-works. Great care must be taken in the design
and construction of pumps of this type to reduce as much as
practicable the large loss inseparable from packed plungers
of all forms. This loss, even under the best conditions of con-
struction and careful packing, consumes a large percentage
of the power applied, especially when the pump is running
against heavy pressures. In order to diminish the friction as
much as possible, it is recommended that the stuffing-box
glands be screwed up no tighter than is actually necessary to
prevent leakage.
A piston water-end by the Erste Brünner Maschi-
FIG. 115.
nen-Fabriks,
Brünn, having
some of the
characteristics
of the Worth-
ington, water-
end, is shown
in Fig. 115. By
reason of the
working-barrel
being included
in the

the main
casting of the
water-end, it is
less difficult to
construct if the
latter is made
in two pieces,
as shown in the
drawing; it is
true an addi-
tional joint is
required,
but
the facility with which the work can be done in the machine-
shop will probably not increase the cost of the water-end.
WATER-END DESIGN.
137
This illustration is from one of a pair of pumps placed side
by side, each pump being driven by an automatic cut-off
engine, both of which are coupled to the same shaft, and
have one fly-wheel in common, but each engine may with
its own pump be operated singly if so required. The pumps
are 11.9 inches diameter by 43.55 inches stroke. Each end
of each pump has two suction- and two discharge-valves as
shown; the valves are 10.7 inches diameter, are of india-
rubber, and have wrought-iron back-plates held down by
conical spiral springs. The valves are fitted with brass bush-
ings to reduce the wear in the central hole. The valve-gratings
are of cast iron. The air-chamber seems very small for a
pump as large as the one now under consideration, but it
is supplemented by a wrought-iron "wind-chest," about 36
inches diameter by 10 feet in height, situated to the rear of
and central to the two water-ends. This vessel is divided into
two chambers by a wrought-iron diaphragm, one of which acts
as an air-vessel for the suction, the other half acting for the
delivery.
The plunger-pump shown in Fig. 116 is from designs by
the author, who recognizes fully the value of a larger number
of small valves rather than a less number of large valves, pro-
vided the latter are of a diameter exceeding 4½ inches. The
plunger-ring is supported at each end, in a bored recess in the
water-cylinder, and is securely bolted to a flange which is cast
in and forms a part of the main cylinder casting, insuring
not only a perfect alignment, but the utmost rigidity. The
plungers and rings are easily removed from the pump when-
ever desired, either for examination or repairs.
The discharge- and suction-valves are above the plungers;
the pump can never, therefore, lose its charge. This feature
is believed to be of value to the user in securing prompt action
in emergency and avoiding the danger of breakage that is in-
curred with pumps running dry through leakage past the
suction-valves, when the latter are located below the plungers.
Hand-hole openings permit an easy examination of both the
12*
138
PUMPING MACHINERY.
suction- and discharge-valves, and without drawing off the
water from the main cylinders of the pump.
FIG. 116.

conn
The inside-packed plunger-pump illustrated in Fig.
117 is a modified form of the water-end just described; it
preserves the distinctive features of the above pump, the ad-
dition of a stuffing-box and gland, adapted for packing the
plunger with a flexible or a fibrous packing, constituting the
only change. This pump is chiefly desirable in localities where
the water-supply is muddy or contains gritty matter.
This form of pump is to be preferred over a piston-pump
for gritty water, inasmuch as the cost of a new plunger is very
much less than that of a new lining. There is the further
advantage in the fact that slight reductions can be made in the
diameter of the plunger by turning in a lathe to remove the
WATER END DESIGN.
139
scoring incident to the service in which it may be employed,
the difference in diameter being made good by the use of a
thicker packing.
H
J
FIG. 117.

100%
The outside-packed plunger-pump shown in sectional
elevation in Fig. 118 is still another modification of the plunger-
pump (Fig. 116). Instead of the plunger-ring as there shown,
a central diaphragm, bored to receive the rod connecting the
two plungers, is secured by bolting in the bored recess con-
tained in the water-cylinder. This diaphragm is not fitted
with a stuffing-box, nor is it the common practice to include a
brass bushing, the cast iron forming of itself a suitable mate-
rial for resisting the very slight wear of the rod passing
140
PUMPING MACHINERY.
through it. Instead of the ordinary heads, those fitted with a
stuffing-box and gland, as shown in the drawing, are substi-
tuted, through which the plungers slide. The plunger-rod is
connected by a screw-joint immediately inside of the inside
plunger. The other details are in no respect different from
those already described.
Outside-packed plunger-pumps are a more recent invention
than the packed piston, their first employment being, in all
probability, an expedient by which to obtain a weighted, ver-
FIG. 118.

tical, single-acting, elongated piston. This was made to pass
through a stuffing-box having leather packing not unlike that
used in hydraulic presses at the present time. As this device.
did not require the boring of the water-cylinder, as must be
done in all piston-pumps, together with the fact that the op-
eration f turning is a much simpler and cheaper method
of manufacture, it led to the early adoption of the packed
plunger on purely commercial principles. Aside from this,
there is a certain satisfaction in seeing a plunger pass out and
WATER-END DESIGN.
141
C
in a water-end, without leakage, through an adjustable stuffing-
box; and this, no doubt, has had its effect in giving direction
favorable to this detail in pump design.
The outside-packed plunger-pump with parallel
rods, as shown in Fig. 119, is a type of pump in very
general favor among furnace, rolling-mill, and steel-works
managers. These pumps usually have strokes of 18 to
FIG. 119.

I
24 inches, and are seldom made with plungers less than 12
inches in diameter. The sectional elevation shows clearly
the internal arrangement. An end elevation of this pump is
shown in Fig. 120. The suction-pipe distributes its flow of
water to the two sides of the pump into the space below the
142
PUMFING MACHINERY.
lower or suction-valves. The delivery is central between the
two pumps. The air-chamber is located immediately above
FIG. 120.
the delivery-opening.
These pumps are

generally used at
pressures seldom ex-
ceeding 75 pounds
per square inch.
An outside cen-
trally-packed
plunger-pump is
shown in sectional
elevation in Fig. 121
and in end elevation
in Fig. 122. For
ordinary service,
pumps of this design
range in size from
small boiler feed-
pumps to sizes suit-
able for water-works.
A modification of this
design is employed
in mines and other
heavier pressures than occur in domestic water-supply. This
design includes two separate or single water-cylinders placed
side by side, and adapted for duplex steam-cylinders and
valve-motion. The suction-opening is at the rear end of each
water-cylinder, the two openings being connected by a dis-
tributing-pipe, as shown in the drawing. In this design the
suction-valves are placed below the plunger, and the delivery-
valves above. A delivery-pipe, connecting the two water-ends
with air-chamber above, is shown with the flanged opening
leading to the rear of the pumps, but it is evident that, by a
suitable arrangement of the pattern, it can lead over either
side if so desired.
WATER-END DESIGN.
143
0
Differential Plunger-Pump.-The combined piston- and
plunger-pump, commonly known as the differential plunger-
pump, is shown in one of its various forms, in sectional ele-
vation, in Fig. 26. The description of the working of a
bucket- and plunger-pump, as given on page 34, is applicable
to this design and need not be duplicated. The pump there
FIG. 121.

金​金​金
​金​金
​illustrated is of 12 inches stroke, and as it was designed for
use as a sinking-pump in mining operations in which acid
water is almost certain to be encountered, the valves were all
included in the bottom section of the pump. This piece may
be made wholly of gun-metal if thought desirable; in any
Borm
144
PUMPING MACHINERY.
event it is easily detached from the pump if any repairs or
renewals are rendered necessary through the chemical action
of acidulated water.
FIG. 122.
in
Vertical Plunger-Pump.-A sectional elevation of a ver-
tical plunger-pump, with internal stuffing-box, is shown in
Fig. 123, and presents several peculiarities of design which
may be consid-
ered good: its
compactness,
which a suction
air-chamber, suc-
tion - pipe, and
suction-valve
plate are included
in a single unit,
as it were; then
the distance-piece
to the delivery-
valve
through

plate
which
the plunger
works; the cast-
ing containing
the delivery-
opening, on top
of which is placed
the air-chamber,
makes a good
arrangement of
detail for some
certain positions
in which height is available rather than floor-space. The
pump is single-acting. The delivery-valve plate shows but
two valves, but it will be understood that other valves ex-
tend around the whole circumference on properly-spaced
centres.
Mau
WATER-END DESIGN.
145
FIG. 123.
A duplex piston-pump with metal clack-valves is
shown in longitudinal sectional elevation in Fig. 124 and in
cross-section in Fig. 125. This pump was designed by the
writer for handling thick stuff,
such as mash in brewing opera-
tions. The main castings are of
iron; the piston, piston-rod, and
cylinder-linings are of gun-metal;
the valve-seats and valves are
hinged together, and are easily
removable from the pump through
the hand-hole plates opposite
each. The lower hand-hole
plates may be removed without
disturbing the tension of the
spring by a simple device secured
by hooked bolts shown in the
drawing.

Piston-Pump with Wing-
Valves on the Side. The
cross-sectional elevation of a
piston-pump with valves on the
side, as shown in Fig. 126, is of
English design, and represents a
type of water-end not often met
with in this country. Aside from
what the writer believes to be a
defective method of guiding the
valves by spindles projecting from
££
the top of each, the design is a good one, but is not so com-
pact and symmetrical as the single water-ends furnished by
Knowles and other well-known builders.
A by-pass is an attachment to the water-end of a double-
acting pump, consisting of a pipe and gate-valve, as shown in
the sketch, Fig. 127, the object of which is to make a direct
Gk
13
146
PUMPING MACHINERY.
water connection from one end of the water-cylinder to the
other. This attachment is useful in several respects, for
example, in a compound pumping engine in which the high-
pressure cylinder may not be large enough to start the load
from a state of rest, no provision having been made for directly
FIG. 124.

10
'
admitting steam of boiler-pressure in the low-pressure steam-
chest. If, now, the by-pass valve be opened, the effect will
be to immediately reduce the resistance to the movement of
the plunger by allowing the water to flow through from the
WATER-END DESIGN.
147
pressure side into the one which has no pressure, and thereby
enable the engine to complete a sufficient number of strokes
to bring the low-pressure cylinder into service; as soon as
FIG. 125.

D
this is accomplished, the by-pass valve should then be closed
and the engine allowed to take up its full load.
In crank and fly-wheel engines, especially those operating
a direct service, there occur times when the speed of the pump
cannot be brought sufficiently low to enable the engine to
148
PUMPING MACHINERY.
properly pass the centres, but if the by-pass valve be opened,
the speed of the engine may be increased somewhat without
increasing the small quantity of water to be delivered.
It will be understood that a by-pass is simply a leak under
the control of the engineer in charge; the size of the opening
FIG. 126.
is not subject to
any fixed ratio to
plunger area, but
it ought to be lib-
eral, and the writer
suggests

1/2
inches for a 12-
inch water-cylin-
der, and 2½ inches
for one 24 inches
diameter. A gate-
valve only should
be used in a by-
pass unless it is
combined with a
charging - pipe, in
which case three
valves will be re-
quired, as shown
in the sketch, Fig.
127 A. The valve
leading from the
supply should be a
gate-valve, and the
two end ones may
be angle-valves.
That a by-pass
is wasteful of steam is admitted, but its use is entirely one
of expediency, and is only resorted to in cases of emer-
gency; so far, then, its application is justified, and should be
included in all crank and fly-wheel water-ends intended for
direct service. Direct-acting duplex pumps do not require
WATER-END DESIGN.
149
a by-pass, as they can be run at any rate of speed, however
slow.
#
FIG. 127.
No Scale.
A water-end with bored valve-seats, as shown in
Fig. 128, has been adopted by the Woodward Steam Pump
Company. The arrows in-
dicate the flow of the water
through the pump when in
action, on its forward stroke.
A valve, the two heads cov-
ering the end openings, the
bolt for securing in place, are
all so clearly shown in Fig.
128 A as to need no further
description.

The loss of efficiency
FIG. 127 A.
No Scale.
in pumps under certain conditions, in which the delivery
is much less than was to have been expected from the calcu-
lated displacement of the plunger and its rate of speed, is at
best a difficult thing to trace. So also the loss of power in
pumping, in which a
large percentage of the
energy of the steam-end
cannot be accounted for
in the delivery of water
under a given head.
Some experiments wit-
nessed by the writer in a
mine of moderate depth
(about 450 feet) showed
by the indicator diagrams
that nearly twenty-five
per cent. more power was developed in the steam-end than
could be accounted for by the indicator diagrams of the
water-end. The pump was direct-acting, and one of many
of the same type, this especial one having been selected in

•
13*
150
PUMPING MACHINERY.
order to get a high average performance. In this case the
suction-pipe was not of excessive length, was of large di-
ameter, and all the conditions apparently favorable to good
performance.
FIG. 128.
Losses occur in pumping if the suction-valves are of such
size or shape as not to readily admit the water into the pump-
barrel so as to completely
fill it by the time the
plunger begins its return
stroke. The plunger re-
ceives its first shock when
the body of water is en-
countered and made to fill
the space instantly under
full pressure, and its second
shock almost immediately
thereafter, when the deliv-
ery-valves must be raised
from their seats.

B
How
much power is lost by these two faulty actions is not known,
but it is considerable.
The frictional resistance retarding the water is subject to
three laws, quite the reverse of friction between rigid bodies:
I. It is proportioned to the amount of surface in contact.
FIG. 128 A.
2. It is independent
of the pressure.
3. It is proportional
to the square of the
velocity.
From the first of
these laws it will be
seen that the minimum

amount of surface must be exposed to the flowing fluid; this
surface, for a given sectional area, will be the least when the
section is circular.
From the third law a greater amount of efficiency may be
expected by allowing the water to move with a small velocity.
WATER-END DESIGN.
151
In decreasing the velocity at the same time to maintain a
given supply, we must increase the surface in contact with the
water, and thus the friction due to this cause of increased sur-
face will be augmented; but since the resistance of friction is
proportional to the square of the velocity and only as the
surface, a greater advantage in this respect will be gained.
From this it follows that it will be advantageous to make the
pipes as large as possible, limiting their size by other con-
siderations, such as space, expense, etc. With regard to the
direction of the water, any change in the direction of the flow
must be gradually introduced. The more sudden the change,
and the greater the angle through which it is directed, the
greater will be the opposing resistances.
Drainage-pipes should be attached to all water-ends
which, by reason of exposed situation, are likely to freeze
when not in use. Such pipes must be fitted to every portion
of the pump in which there are water-pockets or cavities, and
not to the two ends of the water-cylinder alone. Pumps in
quarries and other exposed situations may have small holes
drilled through the valve-deck plates and other portions of
the pump, so there will be a drainage from the delivery-
chamber through the pump to each end of the water-cylinder,
and thus dispense with the outside pipes and cocks. These
holes will take off a small percentage of pump efficiency, but
their usefulness and automatic drainage will probably fully
compensate for that. A pump so fitted should have either
suitable priming-pipes, or a funnel attachment for filling the
pump before starting.
152
PUMPING MACHINERY.
CHAPTER VIII.
HYDRAULIC-PRESSURE PUMPS.
THE transmission of power by means of water under a high
pressure is now generally employed in riveting, shearing,
punching, bending, and flanging machines, as well as for the
handling of Bessemer converters, lifts, cranes, and other
machines in and around steel-works, ship-building, boiler-
making, and other kindred establishments.
There is little or no uniformity in the pressures employed
for doing the same work in different establishments using
hydraulic power, probably for the reason that each manage-
ment had independently worked out the necessary hydraulic
detail, in most cases originating the machines; and in fixing
upon a pressure to do the work it was generally without any
reference to what was being done in other, and perhaps rival,
establishments. It may be said that in steel-works and other
metal-working plants the pressures employed for operating
the hydraulic machinery are not often less than 500 pounds.
per square inch, and do not generally exceed 2000 pounds,
although double this pressure is occasionally employed for
special work.
Some memoranda relating to pressures collected in the
ordinary course of business show the following wide diver-
gencies:
Ordinary hydraulic lifts for warehouses
Foundry hydraulic lifts
Hydraulic cranes
Flanging, punching, and riveting machines
Hydraulic shears
Special work ranging up to
75 to 150 pounds.
100 " 300
300
1000
<<
<<
500 " 1500
1500" 2500
5000 pounds per square inch.
HYDRAULIC-PRESSURE PUMPS.
153
Pressures of 1200 to 1500 pounds per square inch in
metal-working establishments are probably oftener used than
higher pressures, for the reason that many of the details of
hydraulic machines can be better worked out for 1500 pounds.
pressure than for 3000 pounds, size being an important factor
in many details, and especially in valve-gear, so that the ability
to double the area of certain parts may be, and often is, of
the utmost importance in the construction and successful
operation of hydraulic machines. An instance once came
under the notice of the writer in which a reduction of pressure
from a proposed 4000 pounds to 2000 pounds per square inch
determined the practicability of certain devices through no
other reason than that a better mechanical construction could
be had by making certain important and somewhat intricate
detail double the area. The writer is not to be understood as
saying that high pressures are not practicable, but rather that
moderate pressures should be first considered before pressures
of more than 2500 pounds are adopted.
In reference to the relative advantage of using a moderate
pressure with a large ram in hydraulic machines, or a high
pressure with a small diameter, Mr. Tweddell, an hydraulic
engineer of large experience, observes that "it must be borne.
in mind that the friction of the water was not perceptibly
increased with the increase of pressure, and consequently there
was no loss in using the higher pressure, which allowed of a
more compact machine; and practical experience has shown
that there was no trouble in keeping pipe-joints tight at
pressures of 1500 or 2000 pounds per square inch. At the
same time he did not think there was any occasion for carry-
ing such high pressures throughout the whole of a works,
and a much lower pressure would be sufficient for working
foundry and other hydraulic cranes; in such cases there was
certainly a limit to the extension of a high pressure."
The above has reference to what may be termed live press-
ures; that is to say, continuous pressures from the pump
through an accumulator to a system of piping leading to the
several hydraulic machines, cranes, etc., and not to accumu-
154
PUMPING MACHINERY.
lated pressures practically without motion, such as hydraulic
presses and the like.
The service required of water-ends for hydraulic-pressure
pumps is so entirely different from that described in the pre-
ceding chapter that it was thought best to separate the two,
as they have little or nothing in common either in design or
method of construction.
Plungers.-Hydraulic-pressure pumps are almost always
fitted with double-acting plungers working through stuffing-
boxes at each end of the pump-barrel; parallel rods couple
the outside plungers to a central cross-head, to which are
also secured both the inside plungers and the steam piston-
rod or other driving mechanism. The stuffing-boxes should
always be deep, so that a liberal quantity of packing can be
inserted at one time. Plungers for small pumps working
under high pressures should be made of machinery or cast
steel, and finished by grinding, or by means of a lead lap, so
as to insure their being perfectly round and straight. Care
should be exercised in the selection of bar steel to see that it
is entirely free from seams.
Plungers 3 inches in diameter and larger may be made
of cast iron, turned and polished; in the event of blow-holes
or other defects, they should be repaired by drilling into the
plunger and the insertion of iron plugs well riveted or ex-
panded by hammering, and afterwards finished to size. For
cast-iron plungers it is customary to include the cross-head for
the parallel rods in the same casting with the plunger.
Materials.—Water-ends for pressures up to 500 pounds
per square inch may be made of ordinary cast iron, but for
pressures greater than that, and not exceeding 1000 pounds.
pressure, air-furnace castings should be used; the tensile
strength of well-mixed air-furnace cast iron will average not
much below 45,000 pounds per square inch of section. For
pressures more than 1000 pounds per square inch, open-hearth
HYDRAULIC-PRESSURE PUMPS.
155
steel castings are recommended; not that such castings are
stronger on an average than air-furnace iron, but the quality
of the metal is entirely different, and partakes more of the
nature of wrought iron. Whatever the material used, very
great care must be exercised in design that lumps and unu-
sually thick portions do not occur, for every such occurrence
means a defective casting in the central line of the junction of
two or more parts. The writer is not partial to the use of
brass water-ends except for small sizes, say for plungers not
exceeding 1½ inches in diameter; the metal should not be
much, if any, less than for cast iron to work under the same.
pressure, experience having shown that ordinary brass cast-
ings are liable to change shape under a continued high
pressure. If the metal chosen be phosphor-bronze, it will add
but little to the cost, and a much stiffer casting will be secured.
Double-Acting Pressure-Pump.-The sectional eleva-
tion shown in Fig. 129 represents the usual design for a small
pressure-pump; it shows an objectionable detail in a neces-
sity for putting the largest valves on the delivery side of the
pump, where they are not needed, instead of the suction
side, where they are needed. This fact does not prevent the
pump doing good work, and is not considered as absolutely
bad. The valve-seats are usually fitted and driven tightly into
straight holes, from which they may be removed whenever
desired by means of a hook bolt-head underneath the bush-
ing, with a strap and nut above. The diameter of the upper
valve-seat must be large enough to permit the passage of the
lower valve through it. The valves and seats are usually of
hard gun-metal or phosphor-bronze. The caps over the
valves are also of gun-metal or hard brass. For small pumps
no objection exists to tapping the water-end for insertion of
the caps as shown, but for larger pumps, say those having
plungers of 3 inches diameter and larger, square or round
caps, held in place by bolts and nuts, are to be preferred.
A pressure-pump water-end having four separate valve-
openings, as shown in Fig. 130, is one which admits of a
156
PUMPING MACHINERY.

FIG. 129.
HYDRAULIC-PRESSURE PUMPS.
157

FIG. 130.
14
158
PUMPING MACHINERY
FIG. 131.
1
0
readier examination in
case the pump should
work irregularly, as only
the valve at fault is dis-
turbed in its seating.
A water-end with
pot-valve chamber,
as shown in Fig. 131, is
recommended for all sizes
having plungers larger
than 3 inches in di-
ameter. In this design
the water-cylinder is a
plain barrel, having a
central diaphragm in-
cluded in the main cast-
ing, with one flanged
neck on each side of this
partition for connecting
the pot-valve chambers;
the working-barrel also
includes the stuffing-
boxes at each end, and
the connections for the
tie-rods for securing to
the steam-end.
A sectional eleva-
tion of a pot-valve
chamber is shown in
Fig. 132. It consists of
two chambers, in one of
which is placed the suc-
tion-valves and in the
other the delivery-valves.
This combined chamber
་

HYDRAULIC-PRESSURE PUMPS.
159
bolts to the pump-barrel at A. The suction-pipe bolts at B.
The delivery of the water under pressure is at C. A plan of
this pot-valve chamber is shown immediately below the sec-
tional elevation. The valve-seat for the suction-valves is
shown in its place; the ribs for supporting the delivery-valve
plate is shown in the plan, the arrangement for the suction-
valve plate being precisely like it on the opposite side.
An end sectional elevation through the delivery-valve
chamber and
seat is shown
in Fig. 133,
and an en-
larged view of
a section of
one of the
valves, to-
gether with
its cage and a
C
FIG. 132.

portion of the
A
seat, is shown
in Fig. 134.
The method
FIG. 132 A.
C
of securing
the
several
cages to the
valve-seat by
means of a
A
plate is clear-
ly shown in
B

O
the several
preceding en-
gravings and
Plan of Fig 132.
in Fig. 135. The above details are from a 9 × 36 duplex
pressure-pump, designed by the author for handling a portion
of a Bessemer steel plant, the pressure being 400 pounds per
square inch, the operation being continuous throughout the
year. This arrangement of seats and valves permits the
160
PUMPING MACHINERY.
removal of one set of valves and the substitution of another
without stopping the pump but the few minutes necessary
C
FIG. 133.
A
to make the change. In fitting
pot-valve chambers, valves, and
valve-seats, everything should be
made to gauges, so that similar
parts will interchange throughout
the series; there will then be re-
quired but one extra set of valves
and one extra seat to replace any
one of the eight sets included in
the pump. The valves and seats
are made of hard gun-metal, the
springs of phosphor-bronze.
Valve-Plate and Valves for 2000 Pounds Pressure.
-The valve-seat and valves shown in Fig. 136 are from a

FIG. 134.

smaller pump of the same general design, working under a
pressure of 2000 pounds per square inch. The valves are 2
HYDRAULIC-PRESSURE PUMPS.
161
inches in diameter, mitred as shown in the drawing, and
provided with four wings each; this same arrangement has
been applied to a pump working
occasionally, but not continuously,
under 3000 pounds per square inch.
FIG. 135.

O
Pot-valve chambers are some-
times made to contain but one suc-
tion- and delivery-valve each; if the
size of the pump is such that but
one pot-valve chamber is required
for each end of the working-barrel,
no objection exists to applying it
instead of the insertion of the valve-
seats in the working - barrel, as
shown in Fig. 130; but when two or more such pot-valve
chambers are required, the writer recommends that the pot-
valve chamber be made to include seats adapted to whatever
FIG. 136.

number of valves may be necessary, and thus keep down
the number of separate parts requiring to be bolted to the
working-barrel.
14*
162
PUMPING MACHINERY.
;
Pressure-pump water-ends of large size ought to be
cast in two pieces and bolted together with a blank flange
between the two ends, as shown in Fig. 137. This particular
water-end is the one referred to in the description of the pot-
valve chamber details given above. In large water-ends there
is always a chance that the metal diaphragm between the two
working ends of the barrel will be defective by reason of the
FIG. 137.

difficulty in getting a proper vent for the cores at that point;
there will also be required chaplets for securing the core in
place, and these of themselves interfere more or less in securing
sound castings. It is a very serious matter to lose a casting
of a large water-end, especially if the metal is 2/2 to 3 inches
in thickness, as the cost of breaking is sometimes quite as
great as would purchase the same weight of pig-iron.
Accumulator.-It is the common practice in hydraulic
work to have somewhere in the system an accumulator for
storing up a convenient volume of water under pressure.
This accumulator serves a useful purpose in the fact that it
renders the action of the pump less irregular than in a direct
service, and in many operations such a service would not be
permissible at all; therefore an accumulator may be con-
"+
HYDRAULIC-PRESSURE PUMPS.
163
sidered as a necessary adjunct to a pressure-pump, and its in-
fluence on the pump must be taken into account.
The use of
an accumulator is a direct benefit in this: it is more economical
to employ a small pump working under a uniform pressure all
the time than a larger pump only a portion of the time. The
draft upon the water-supply in any hydraulic system is always
sudden, and sometimes very great if several machines be
thrown into operation at the same time. It is this sudden
draft upon the water-supply that causes the accumulator to
descend rapidly, and as the flow is now as suddenly stopped
as it was originally started, the effect of its sudden arrest may
be easily imagined.
In regard to the pressure in an hydraulic riveter, shear, or
other machine having a direct connection with an accumulator,
it is not uniform throughout the stroke; in some respects this
is advantageous,-in riveting, for example, advantage has been
taken of the impulse of the falling accumulator at the end of
the stroke, so that a dead pressure of 40 tons was increased to
nearly 60 tons at the time of closing the rivet; the machine
thus starting with the lower pressure arrived at the higher
pressure at the point at which it was required.
This fact has a very important bearing upon pump-design,
and is one often overlooked. If it were a gradual increase of
pressure from end to end it would not be so bad, but it comes
more in the nature of a blow than that of gradual increase,
because it is the sudden arrest of the falling accumulator at
the bottom of its fall, the effect being precisely the same as
any other falling weight moving at the same velocity. Water
being practically incompressible, the shock is transmitted
throughout the whole system, including the pump.
It is just such experiences that lead one to distrust designs
of pumping machinery based on ordinary factors of safety,
especially if the water-end of the pump be made of cast iron,
the accumulator one of small diameter and considerable
height, coupled with the not unusual circumstance of being
scarcely large enough for the work. There is no certainty
whatever as to the solidity of cast iron at the centre of the
•
164
PUMPING MACHINERY.
juncture of two or more pieces, and especially if these pieces
are of unequal thickness. The common defect being a drawing
away of the outer skin of iron from the centre and forming a
cavity within. If this cavity have no outlet it is, nevertheless,
a weak spot in the pump because of the reduction in area;
if, however, an opening extend from this cavity into the
working-barrel, or into the delivery-chamber, then every stroke
of the pump and every arrest of the fall of the accumulator is
an effort to split the pump open, which in time is almost sure
to occur by cumulative effect, the granular structure of cast
iron favoring such rupture.
Air-Chambers for Hydraulic-Pressure Pumps.-It is
not a common practice to supply pressure-pumps, or systems
of hydraulic piping, with air-chambers instead of accumulators,
but in special cases air-chambers have been used working
under a pressure of 850 pounds per square inch, and supply-
ing a large plant of hydraulic machinery in a situation where
it was not convenient to introduce an accumulator on account
of its weight or size. An air-compressor will be required for
keeping the air-vessel properly charged. This is in the nature
of a complication, it is true, but it is essential to the proper
working of such a plant.
Hydraulic Transmission of Power.-Water under
pressure may be transmitted from a central source of supply
to outlying points in a large manufacturing establishment at
very reasonable cost, so that for driving punching and shear-
ing machines, or other hydraulic machine tools in out-build-
ing or in a yard, the power can be furnished at a cost much
less than if a direct line of shafting were used.
As to the power saved by hydraulic transmission, Mr. Tweddell
obtained, experimentally, the following results. The indicated
horse-power of the engine at 50 revolutions per minute was
6.65 I. H. P. when driving the transmitting shafting alone,
3.51 I. H. P. when running light without the shafting,
leaving 3.14 I. H. P. as the lowest power required for the transmitting shafting of
about 300 feet in length.
HYDRAULIC-PRESSURE PUMPS.
165
This shows that when running without doing work forty-
seven per cent. of the engine-power was spent in transmission
by the shafting, and a further loss has to be added for the
friction of the gearing employed in getting up or reducing
the speed of the several tools.
In the hydraulic transmission of power, on the contrary,
the loss of useful effect between the pumps and the accumu-
lator is very little at the usual speed of working; for with one
pump working, 1694 cubic inches was the theoretical delivery
of the pump for 20 strokes, and 1614 cubic inches was the
actual quantity pumped into the accumulator, showing a loss
of only 434 per cent., and with both pumps working, the cor-
responding quantities for 20 strokes were 3388 and 3278 cubic
inches, showing a loss of only 34 per cent. These experi-
ments were carefully made, the accumulator registering the
exact distance it travelled for a certain number of strokes of
the pump.
The loss from friction in the accumulator was found to be as
follows:
1250 pounds per square inch, ascending pressure in the accumulator;
1225
66
descending
66
therefore, as in ascending the friction had to be overcome by
the pump in addition to lifting the load, and in descending the
friction has to be overcome by the load itself, the amount of
friction will be half the difference of pressure in the two cases,
or 12½ pounds per square inch, being only one per cent. of
the power. This result applies equally to the friction in each
of the hydraulic machines; and as the power is generally
applied to the cutting tools direct from the ram, two per cent.
represents all the loss from friction, and there remains only
to add the loss due to the friction of the water in the pipes,
which cannot amount to much if proper care is taken in pro-
portioning their area and in avoiding bends.
Power for Hydraulic Engines.-The increasing variety
of uses to which hydraulic power is now being satisfactorily
employed shows that it is not only rapidly growing into
166
PUMPING MACHINERY.
present favor, but promises well for the future. Recent papers
and discussions on the application of electric power to drive
drilling machinery at the best only leave the reader somewhat
sceptical as to the economy, to say nothing of the practical
feasibility of thus finishing in situ the great amount of drilling
which cannot be done during the earlier stages of the con-
struction of ships, bridges, boilers, etc.
No doubt, however, exists as to the practical success and
the economy in working results when this drilling is done by
hydraulic power. For some years past a great amount of
work of this kind has been done in the French naval dock-
yards by very neat hydraulic drilling-engines. The type of
engine used is the Brotherhood three-cylinder, with which we
are all tolerably familiar; the design, at any rate, if not with
the machine itself. In one arrangement the three-cylinder
engine shaft carried a bevel-pinion, gearing direct into a larger
bevel-wheel driving the drill-spindle, together with the usual
feed apparatus. This can be attached temporarily to a bracket
for drilling work-for example, on a vise bench-away from
the shops. The whole affair is so light, weighing but 60 or
70 pounds, that it can be used in connection with an ordinary
hand-ratchet drill-head. Hydraulic pressure at 1500 pounds
per square inch (but also sometimes at 750 pounds or 1000
pounds per square inch) is conveyed to the drilling-engine by
means of special flexible copper tubing. There is no more
difficulty in attaching this gear to the plating on a ship's side
than is experienced in the case of hand-drilling, while the
speed of working, when once fixed, is nearly 10 to 1 in favor
of hydraulic work. In another application the drill-head is
still lighter, the power being transmitted from the engine to
the drill by means of a Stow flexible shaft. The necessary
speed for the flexible shaft is obtained by gearing up at the
engine-end and reducing it at the drill-end.
The working parts of the Brotherhood three-cylinder hy-
draulic engine consist only of the three pistons and connecting
rods, one crank, and one rotating balanced valve and spindle
which fits into the driver and is turned direct from the crank-
HYDRAULIC-PRESSURE PUMPS.
167
pin. There are no glands, stuffing-boxes, or oscillating joints,
and the wear of all the parts is taken up automatically. The
engine occupies very little space, and will work up to 200
revolutions per minute.
The duplex steam-end is the one in general use for
operating hydraulic-pressure pumps. Geared power-pumps
are seldom used in iron- or steel-works, but are largely used in
other lines of manufacture. Pressure-pumps driven by a steam-
engine in which the pump-plunger is attached directly to the
engine-piston are occasionally met with; it is a good form,
but its use is not general.
The especial fitness of the duplex method of operating the
plungers in hydraulic-pressure pumps was probably never
better stated than by the late Alexander L. Holly, in his
affidavit at the time of the expiration of the Worthington
patents, an abstract of which is here given:
"In all Bessemer works pumping engines for throwing
large volumes of water, under 300 to 400 pounds per square
inch, are required to actuate cranes, hoists, converters, and
other hydraulic machinery. The hydraulic machinery is the
most expensive and the hardest-worked part of the plant, and
the constancy and steadiness of the pumping power is the
most vital feature of the whole system of machinery. Any
delay or serious fluctuation in its operation in handling fluid
masses of iron and steel is fatal to the commercial success of
the Bessemer process.
66
The pumping engine runs from twenty to twenty-four hours
per day, rarely excepting Sundays. The strain on the engine
is constant, but the velocity is momentarily and suddenly
varied by the demand for water from a low to the highest
speed. Yet the pressure on the cranes, etc., is, and for the
safe working must be, nearly constant. The heavy pressure
by itself puts the machinery under a severe stress; but when
this pressure is also associated with great volume, requiring
very large parts, the conditions of service are more severe than
any other with which I am acquainted.
168
PUMPING MACHINERY.
"The reason why this pumping engine stands this severe
service without extraordinary repairs, and the reason why it
imposes no extraordinary stress on the machinery it actuates,
is, that the system on which it is constructed—the movement
of the steam-valve of one engine by the piston of the other
engine-permits the water-pistons to stop momentarily at the
ends of their stroke, thus allowing the water-valves time to
seat without slamming; all this being associated with a uni-
form velocity of piston, and hence a uniform pressure on the
water pumped instead of a varying pressure, such as is caused
by an irregularly-moving piston attached to a fly-wheel. The
water flows through the pump and to the cranes in a constant
and noiseless stream. In all other pumps with which I am
acquainted, when applied to such service, there is a violent
concussion of the water, slamming of the valves, and jarring
and straining of all the parts, and consequently frequent
break-downs of the pump and of hydraulic pipes and
machinery."
Compound Pressure-Pumps.-If the conditions of ser-
vice are favorable (and this is not always the case), compound
steam-cylinders, either with or without condensing apparatus,
may be applied to any direct-acting pressure-pump, and there-
by effect a considerable saving by getting a certain amount of
work out of the exhaust steam, which would otherwise escape
into the atmosphere.
Compounding is not recommended for pressure-pumps
where the service is irregular; that is, where the pump works
with great violence for a few minutes and then comes to a
state of rest, an effect following some kinds of direct service,
and always likely to occur if the accumulator is too small for
the work, for the usual method of controlling the steam throt-
tle-valve is by the rise and fall of the accumulator ram. If
the action of the pump can be made continuous over long
intervals of time, say several hours, then compounding is
recommended, provided the initial steam-pressure in the high-
pressure cylinder is not less than 75 pounds.
HYDRAULIC-PRESSURE PUMPS.
169
Pressure-pumps driven by power are often arranged
with two or more plungers driven from a crank-shaft; a large
plunger for filling the machine or ram quickly, at a pressure
say one-fourth of that ultimately required, a pressure which
will do enough work in many operations to use probably four-
fifths of the whole volume of water, there remains then but
one-fifth more water to be pumped at the higher pressure, if
each pump acted singly, one after the other, which is some-
times, but not always, the case.
It is the usual practice in constructing such pumps to have
both large and small plungers in continuous movement.
Such pumps are usually provided with an automatic device,
so that when the limit of pressure for the large plunger is
reached its suction-valve is lifted from the seat, and continues
off the seat so long as this pressure is maintained; the large
plunger during this time is simply playing back and forth in
its cylinder, without doing any work; meanwhile, the smaller
pump is forcing the water at the higher pressure to complete
the hydraulic operation.
As a practical example, let us suppose a crank-pump fitted
with two single-acting plungers of different diameters, both
working at the same time, the larger one 3 inches diameter,
the smaller one 1½ inches diameter, both of 6 inches stroke.
Each revolution of the crank-shaft will cause the delivery of
42.4 cubic inches of water for the large plunger, and 10.6 cubic
inches of water for the small plunger. If the crank-shaft
make 60 revolutions per minute, 3180 cubic inches of water
will have been delivered by both plungers, we will say at 500
pounds per square inch, at which pressure the large plunger-
pump ceases to act by the automatic lifting of the suction-
valve. The operation of the smaller plunger-pump goes on,
and takes up the task of increasing the water-pressure from
500 pounds to the higher one necessary to do the work; this
smaller plunger being only one-fourth the area of the larger
one, will have power sufficient to force the water-pressure up
to 2000 per square inch from the same crank-shaft, and at a
somewhat less expenditure of power per revolution, because
H
15
170
PUMPING MACHINERY.
the large plunger is thrown out of service; the quantity of
water delivered would be only one-fifth as much; that is, 636
cubic inches per minute as against 3180 when both plungers
were at work. This example assumes that two minutes are
given the hydraulic operation requiring the active use of the
pumps, but generally the second operation requires less time
than the first, except when pressing bales, etc.
The power pressure-pump illustrated in Fig. 138 was
designed by Watson & Stillman, New York, and represents
one type of the medium size of hydraulic pumps built by
FIG. 138.

PHOTO ELECTROTYPE ENG C
them.
This particular pump is fitted with four plungers
driven from the cranks and cross-heads; two of these pumps
are high pressure, having plungers 3/4 inch diameter, working
VERE
HYDRAULIC PRESSURE PUMPS.
171
under a pressure of 6000 pounds, the other two plungers
being 1½ inches in diameter, and working under the lower
pressure of 1500 pounds per square inch; the four pump de-
liveries are all connected to one common discharge-pipe.
The two low-pressure pumps are provided with an automatic
trip, so that when any desired pressure less than the above
is reached they will automatically stop their delivery, in addi-
tion to which safety-valves are also provided for each of the
above pressures; the pressure-valve of the low-pressure pumps
acting as a check-valve to prevent the water of the high-
pressure pumps from being driven back into the low-pressure
pumps.
Although differing in design, the details of the automatic
trip shown in the sectional elevation, Fig. 139, will illustrate
FIG. 139.

•Suction Valve
Delivery Valve
the principle of the method employed to stop the action of
the pump. For the sake of clearness the position of the
172
PUMPING MACHINERY.
valves were changed from that in the pump; it was also neces-
sary to leave off the piston connecting with the high-pressure
system, as shown upon the front trips in Fig. 138, but which
can be clearly understood at a glance; the action being that
when the required pressure has been reached the weighted
lever will be lifted directly by the piston, or, in the case of an
accumulator, by an accumulator trip, as in the beginning or
middle of a stroke; the valve would then be seated with a
heavy pressure upon it, and it would be impossible to raise it,
so that a slotted cross-head must be placed upon the connec-
tion; furthermore, the spring which had been kept in a state
of tension by the weight and lever is now kept in a state of
tension by the pressure upon the valve only when this valve
is raised from its seat by the action of the succeeding stroke.
The spring prevents its reseating again by means of the lever
and rod extending up through the suction-pipe, which also
allows the water to be driven backwards and forwards through
the suction-pipe without any loss of power.
!
STEAM AND POWER CRANK-PUMPS.
173
CHAPTER IX.
STEAM AND POWER CRANK-PUMPS.
PUMPS of this type are very much in favor in Europe, and
more largely made and used there than here; a circum-
stance probably due to the invention and development of the
steam-thrown valve in this country, and now so universally
employed in single steam-pumps, its use here having preceded
by many years its general manufacture and sale abroad.
Pumps with steam-thrown valves are compact, efficient, and
offered at a lower price than is possible with fly-wheel pumps;
the result is they meet with a ready sale, so that for twenty-
five years past fly-wheel pumps have been offered at a great
disadvantage because of its cheaper rival.
The question is sometimes now asked, though less fre-
quently than formerly, as to the comparative merits of crank
and fly-wheel as against direct-acting duplex pumps for small
and medium sizes, say those having a capacity less than one
million gallons in twenty-four hours. This is not an easy
question to answer, because there are so many things enter-
ing into the problem which lie wholly outside of the relative
merits or demerits of the two designs when brought in direct
comparison.
Without entering into detail, the general conclusion reached
by engineers and users of steam-pumps is that, in general, the
small and medium sizes of direct-acting single and duplex
pumps compare favorably as an investment, admitting the fact
that they are less economical in the use of steam.
The circular motion of a crank-pin must of necessity
be continuous and nearly uniform if the crank-shaft be fur-
15*
174
PUMPING MACHINERY.
nished with a fly-wheel of proper proportions. An analysis
of the relative motions of a crank-pin and its corresponding
piston, assuming the connecting-rod to be three times the
length of stroke, is shown in Fig. 140. The piston movement
FIG. 140.

Piston Movement
45
Regular Stroke Divisions
is not so objectionable on the first and last quarters of the
crank movement, but the rapid acceleration of the piston just
before the quarter stroke is reached, continuing to the half
stroke, and then of its retardation until the three-quarters
stroke is passed, brings some very complicated strains upon
all the reciprocating parts of the pump and their connections.
The best position for the cranks on a double pump-
ing engine is at right angles to each other; this equalizes
somewhat the operation of the pumping engine as a whole,
but does not remedy the faulty operation of each water-end
singly.
Crank-pumps for this reason require to be stronger than
direct-acting pumps of the same size; the energy stored up
in the rim of the revolving fly-wheel, together with its nearly
uniform rate of revolution, as well as its inability to suddenly
change either of the above conditions in a single stroke of
the pump, taken in connection with so inelastic a substance
as water, make the conditions difficult to provide for. The
difficulties which attend the practical solution of such a prob-
lem lead one easily to the conclusion that a crank and fly-
wheel are not only imperfect but undesirable in a train of
pump mechanism, and it is largely for this reason, as well
as for cheaper construction, that direct-acting steam-pumps
STEAM AND POWER CRANK-PUMPS.
175
have largely displaced the crank and fly-wheel pump for the
commercial or small sizes.
The advantages claimed for the crank over the direct-
acting movement is that the length of stroke is fixed, and no
short strokes can occur, thereby securing a measured dis-
placement for each stroke; thus crank-pumps require less
clearance at each end of the steam-cylinders, and it follows
also that less steam is used per stroke and with greater
economy. This is true in part, and becomes so only when
the engineer in charge of a direct-acting engine fails to adjust
the dash-relief, or cushioning valves, so as to make a full
stroke, and yet prevent the pistons striking the heads, a thing
not at all difficult to do.
Mr. H. P. M. Birkenbine says, with reference to and in favor
of crank and fly-wheel pumps, "A higher piston-speed can
be had with a crank and fly-wheel pump than if the pump
were direct-acting, for the reason that in the latter type the
termination of each stroke is defined and secured by steam
acting as a cushion to counteract the force of the moving
parts and of the water and bring them to rest. In large
steam-pumps 100 feet per minute may be considered as a
limit to safe piston-speed. With pumping engines having
cranks, connecting-rods, and fly-wheels to terminate and de-
fine the stroke of the piston any piston-speed possible to the
pump can be secured with safety. The power stored in the
moving mass of the fly-wheel at the termination of the stroke
is carried to the beginning of the next stroke, without any
loss but that due to the friction of the moving parts and the
resistance of the air to the motion of the fly-wheel. Then
the practically uniform speed of the rim of the fly-wheel se-
cures the desired motion for the piston through the connect-
ing-rod and crank of the pump by gradually retarding the
motion until the point of rest is reached, and accelerating it
after the piston has passed that point."
Valve Areas.-Crank and fly-wheel pumps, by reason of
the irregular plunger movement, should have slightly larger
176
PUMFING MACHINERY.
valve areas than direct-acting pumps of the same size; they
should have large air-chambers, and, preferably, each water-
end should have its own air-chamber. As has been already
stated, the best arrangement is to place two double-acting
pumps side by side, to operate as a pair, with a crank move-
ment at right angles to each other. When properly designed,
double-acting crank-pumps thus arranged have given excel-
lent satisfaction, and have shown exceptionally high economy
in the use of steam by reason of the high initial steam-press-
ure, and the expansion following an early cut-off, a result
which has been quite impossible to obtain in direct-acting
pumping engines until within the past few years, during which
time ingenious and efficient high-duty attachments have been
developed.
The want of approximation to continuous effort in
a crank and fly-wheel pump was shown by Mr. John G. Mair,
FIG. 141.

Line
of
mean
Velocity

Path
of
Crank
10°
pin
90°
180°
270°
360
One Revolution.
M. Inst. C. E., London, in a paper contributed to that institu-
tion in 1886, in which he remarks upon the delivery from a
compound rotative engine, with cranks at right angles, work-
STEAM AND POWER CRANK-PUMPS.
177
ing two double-acting pumps, supposing the connecting-rod
to be indefinitely long, to be similar to that shown in Fig.
141. The deliveries are added together and shown in full
lines; the variation of flow in this case is sufficient to make
the pressures fluctuate to such an extent that accidents are
very liable to occur when working without air. Mr. Mair
states that in his own practice he has met with many cases
where accidents have happened to the pump-work and rising
mains when through carelessness no air was in the vessel.
Such a result might have been expected. The importance of
a large and properly charged air-chamber attached to the de-
livery side of a crank-pump is now so generally understood
that bad effects are sure to follow any neglect in so important
a detail.
Using Steam Expansively. In a mill engine using a
high pressure of steam, cutting off early in the stroke and
expanding below the average pressure necessary to do the
work, the fly-wheel serves a useful purpose in absorbing the
surplus energy at the beginning of the stroke, where it is not
needed, and giving it out during the latter half of the stroke,
where it is needed. In this manner a properly-proportioned
wheel, working in conjunction with a good governor con-
trolling the point of cut-off, will give a very even rate of
rotation to the crank-shaft, varying, of course, within the limits
necessary to secure a proper action on the part of the governor.
In the case of a crank and fly-wheel pumping engine as
usually designed, the case is so entirely different as to almost
prevent the use or application of ordinary fly-wheel formula
such as would apply in the preceding paragraph, and for the
reason that the plunger-rod of the water-end passes into the
steam-end, connecting directly with either the high- or low-
pressure piston, or if a tandem engine, with both. A pump-
ing engine of this type is but a modified form of a direct-
acting engine, because the effort of the steam-end is transmitted
directly through its piston-rod, which, by continuation through
a cross-head or otherwise, extends into and becomes the
m
178
PUMPING MACHINERY.
plunger-rod of the water-end, controlling the movement of
the plunger, making it coincident with that of the steam-pis-
ton. In this case the power of the engine does not pass
through the crank-shaft and fly-wheel, as was the case in the
mill engine; the connection between the steam-end and the
water-end being direct, the fly-wheel absorbs and gives off
only so much of the energy of the steam-end as comes
through the irregular movement of the pump-plunger due to
that of the crank, together with another irregular impulse
upon the plunger due to the method of steam distribution.
If a pair of double-acting pumping engines were placed
side by side cutting off steam at say five-eighths of the stroke,
such as would be the case with the ordinary slide-valve, the
machine would work without any fly-wheel at all; and such
pumping engines are now in use in mines and for other
pur-
poses. Pumping engines of this type are more economical
in the use of steam than direct-acting pumps, but less so than
the same type of pump furnished with a fly-wheel and a
better steam distribution, together with perhaps twice or three
times the initial steam-pressure.
To better illustrate the best method of using steam for any
purpose three diagrams are given in Fig. 142, of which two
are intended to show how a saving in steam is had over a
direct acting engine following full stroke, by increasing the
steam-pressure and cutting off at half stroke and at quarter
stroke, using the same cylinder throughout; the engine to be
non-condensing, and exhausting against a back pressure of
18 pounds absolute, or about three pounds above the atmos-
phere. The initial pressures shown in the diagrams are 65,
80, and 120 pounds respectively.
In estimating the value of the diagrams, let us assume that
the cylinder is of four cubic feet capacity, and the diagrams
showing 65 pounds pressure to represent an ordinary direct-
acting steam-cylinder; the initial and terminal pressures are,
of course, the same, or a mean pressure of 65
18 = 47
pounds. Four cubic feet of steam are used, and as one cubic
foot of steam at 65 pounds pressure weighs .1569 pound, four
STEAM AND POWER CRANK-PUMPS.
179
cubic feet would be .6276 pound of steam required to move
the load one stroke.
If this same work could be done employing a higher press-
ure, say 80 pounds, and cutting off at half stroke, a saving
FIG. 142.
1
Atmos. Eine
would be had as follows, using the formula given on page 181,
and the table of hyperbolic logarithms on page 180.
I + .6931
80 X
2
- 18: 49.7 pounds mean pressure.
Instead of using four cubic feet of steam, only two are re-
quired. Steam at 80 pounds pressure weighs .1901 pound
per cubic foot; then 2 X .1901 = .3802 pound of steam for
each stroke, or .2474 pound less than in the preceding ex-
ample, a saving of about 39 per cent.
The third diagram is that corresponding to 120 pounds ab-
solute pressure and cutting off at quarter stroke, which repre-
sents a further saving, thus:
I + 1.3863
120 X
18
53 6 pounds mean pressure.
་
T

4
180
PUMPING MACHINERY.
Only one cubic foot of steam is required, and this weighs
.2742 pound, so that a saving of .3534 pound is had at 120
pounds pressure following quarter stroke, over 65 pounds fol-
lowing full stroke, a saving of 56 per cent. in weight of steam.
It will be understood that these are theoretical deductions,
and that losses of several kinds which occur in practice will
reduce these percentages somewhat, but they are approxi-
mately true, and will serve our present purpose, which is to
show the wasteful action of direct-acting non-expansion engines
as ordinarily applied to steam pumps.
Hyperbolic logarithms are seldom required in direct-
acting pump calculations because the steam is not used ex-
pansively, but for convenience in such cases as involve the
expansion of steam, as in the case of crank and fly-wheel, and
other high-duty engines, they are exceedingly useful.
TABLE XI.
HYPERBOLIC LOGARITHMS.


No.
Log.
No.
Log.
No.
Log.
No.
Log.
I.I
.0953
1.2 .1823
1.3
.2624
1.4
.3365
1.5
.4055
حب حب حب حب حب
3.4
3.5
3.6
3.7
3.8
1.6
.4700
3.9
1.7
•5306
4.0
456 7890
1.2238
5.7
1.2528
1.2809
5.8
78
1.7405
8.0
2.0794
I 7579
8.I
2.0919
5.9
1.7750
8.2
2.1041
1.3083 6.0
1.3350
6.I
1.7918
1.8083
1.3610
6.2
1 8245
1.3863
6.3
1.8405
∞ ∞ ∞ ∞ ∞
8.3
2.1163
8.4
2.1282
85
8.6
2.140I
2.1518
1.8
.5878
4.I
1.4110
6.4
1.8563
8.7
2 1633
1.9
.6418
4.2
1.4351
6.5 1.8718
8.8
2.1748
2:0
.6931
4.3
1.4586
6.6 1.8871
8.9
2.1861
2.I
.7419
4.4
1.4816
2.2
.7884
4.5
1.5041
69
6.7
1.9021
9.0
2.1972
6.8
1.9169 9.1
2.2083
2.3
.8329
4.6
1.5261 6.9
1.9315 9.2
2.2192
2.4
.8755
4.7
1.5476
7.0
1.9459 9.3
2.2300
2.5
.9163
4.8
1.5686
7 I
1.9601
9.4
2.2407
2.6
.9555
4.9
1.5892
7.2
1.974I
9.5
2.2513
2.7
.9933
5.0
1.6094 7.3
2.8
1.0296
5.I
1.6292
7.4
2.9
1.0647
5.2
1.6487 7.5
3.0
1.0986
53
www
3.1 1.1314
5.4
1.6677
1.6864
7.6
7.7
3.2
3.3
23
1.1632
1.1939
in in
5.5
5.6
56
1.7047
7.8
3456 7∞
1.9879 9.6
2.2618
2.0015
9.7
2.2721
2.0149 9.8
2.2824
2.0281
9.9
2.2925
2.0412
IO O
2.3026
1.7228 7.9
2.0541
2.0668
STEAM AND POWER CRANK-PUMPS.
181
Table XI. contains all numbers from 1.1 to 10.0, varying by
.1, which will probably meet any ordinary requirement, but
should a more extended table be required, the reader is re-
ferred to Cotterill's "The Steam-Engine considered as a Heat-
Engine," which contains a very elaborate table, from which
the one inserted was compiled. The use of such a table in
expansion of steam calculations is to facilitate and shorten the
work. The hyperbolic expansion curve of steam is used on
the assumption that the volumes are inversely as the pressures.
This is not exactly the case, but there are so many circum-
stances which affect all steam-engine calculations that it is near
enough true, and by reason of this close approximation and
extreme simplicity, its use in connection with steam expansion
calculations may be said to be universal.
The mean pressure of steam may be obtained by using the
following formula:
Let p
P:
R
= mean pressure of steam per square inch.
initial pre-sure, or pressure on admission to cylinder.
range of expansion, or ratio of volume at end of stroke to volume at
point of cut-off.
་
Then p = PX
I hyp. log. R
R
back pressure.
In other words, the work done before expansion is always
reckoned as I, to which must be added the hyperbolic loga-
rithm of the range of expansion; this sum must be divided
by the number of times the steam expanded, the quotient so
obtained must be multiplied by the initial steam-pressure, and
from the product subtract the back pressure; this will give
the mean pressure throughout the stroke.
Example.-What will be the mean pressure if steam of 100 pounds follow
quarter stroke or four expansions, assuming a back pressure of 18 pounds?
Work done before expansion
Hyperbolic logarithm of 4
I.
= I
1.386 2.386.
Then 2.386
4
.5965.
.5965 X 100 = 59.65
Back pressure
Answer =
18.00
41.65 pounds.
15
182
PUMPING MACHINERY.
The details of crank-pumps need not differ essentially
from those of other types. It is only necessary in designing
FIG. 143.

طاقة النظرية
A
་་་་་
D
THE SAWANO (
4
12
16
20
24
28
32
a crank-pump to select the best size and kind of water-end,
and combine it with a suitable steam-engine. The completed
machine may be either horizontal or vertical, depending on
STEAM AND POWER CRANK-PUMPS.
183
the nature of the service. The examples selected for illustra-
tion show how wide a range of general design may be in-
dulged in; some of the illustrations are much out of the
FIG. 144.

B
Steam per stroke, eight square units.
White lines, square of units.
White stripes, end of piston stroke.
A, cut-off or momentum; B, steam service.
general trend of pump design, whilst others were designed to
meet certain conditions of service for which the ordinary
designs were not thought to be suitable.
The Woodward steam-pump with double-acting piston
water-ends is shown in Fig. 143. It consists of two complete
pumps placed side by side with cranks at right angles to each
other, connecting by
means of
suitable
rods to the piston-
FIG. 145.

rod of each engine.
The cylinders are of
the ordinary three-
ported variety, the
distribution of steam
being effected by the
ordinary D slide-
valve driven by an
eccentric for each en-
gine. Two fly-wheels
are used by which a
regular rate of rota-
tion is secured. The manufacturers of this pump publish
a diagram, which is here reproduced in Fig. 144, showing
184
PUMPING MACHINERY.
the steam consumption and distribution for the double pump
illustrated above.
The cut-off is fixed at 5% of the stroke from the beginning,
S
S
OPENING
CLOSIN
ALTNY
FIG. 146.
ECCENTRIC DIAL
VALVE SPEED
REGULATION,
FAT.
18A9:
CUT OFF
CLOSEING
OPENING
Ε
Σ
C.
5
and by a proper adjust-
ment of the slide-valves
no live steam is needed
for cushioning.
A perspective view of
this pump is given in
Fig. 145, in which it will
be seen that the steam-
cylinders, water - cylin-
ders, and crank - shaft
are all on the same
horizontal plane.
For changing the
point of cut-off, or the
speed of the pump, an

adjustable eccentric is employed, as shown in Fig. 146.
Slotted Cross-Head.-A crank and fly-wheel pump fitted
with a slotted cross-head for operating the fly-wheel is shown
in Fig. 147. The crank-shaft has two bearings, one on either
side of the piston-rod; the slotted cross-head is fitted with a
sliding block, and any adjustment for wear may be had by
reducing the distance-pieces through which the bolts pass.
This is the most compact arrangement yet devised for pumps,
the whole length required being but little more than that
necessary for the path of the crank-pin and the boss of the
crank to which it is attached.
If two such pumps be placed side by side with the cranks
at 90° to each other they will work without a fly-wheel.
An excellent design for a crank and fly-wheel pump by
Guild & Garrison is shown in Fig. 148. The steam- and
water-cylinders are mounted on a bed-plate with the bear-
ings for the crank-shaft back of the water-cylinder. The
steam- and water-pistons are connected by a rod to which is
STEAM AND POWER CRANK-PUMPS.
185
+
also secured the cross-head for operating the crank. The dis-
tribution of steam is effected by an eccentric operating a plain
slide-valve. The fly-wheel in a single pump requires to be
heavier than when two pumps work together at right angles
any benefit is to be derived from the expansion of steam.
if
FIG. 147.

0
0.
A double-acting vertical pump of English design (Good-
brand & Co., Manchester) is shown in Fig. 149. The method
of operating the crank is a good one, because there are no
strains outside of the central line of the steam- and water-
pistons. The length of the connecting-rod, together with the
distance required for the yoke connecting the steam and water
piston-rods, gives the pump considerable height; the fly-wheel
shaft is also rather high because of the yoke connection.
This increased distance is less objectionable in a vertical than
in a horizontal pump. No power is transmitted through the
crank-pins to and from the fly-wheel other than that necessary
to equalize the speed after the point of cut-off, so that the
work required of the fly-wheel is very light.
16*
186
PUMPING MACHINERY.

FIG. 148.
ཀ
อย
GUILD & GARRISON.
BOOKHOUT BROS N
STEAM AND POWER CRANK-PUMPS.
187
This pump has been designed to do light as well as heavy
duty, and is arranged with a movable eccentric to vary the
FIG. 149.

}
cut-off in the steam-cylinders, ranging from one-eighth of the
stroke for light duty, to following the piston to about eight-
188
PUMPING MACHINERY.
tenths of the stroke. The eccentric as described by the
makers, unaccompanied by drawings, is constructed as follows:
A disk, having ears as guides for the eccentric, is keyed to the
FIG. 150.

K
Kill
crank-shaft, and the eccentric itself is simply a ring to receive
the clips for coupling up to the valve rod; it is made with two
lugs to work freely in the eccentric-groove, turned on a hand-
wheel, which is free to turn round on the crank-shaft, and
when turned in a direction indicated by the arrow on the
STEAM AND POWER CRANK-PUMPS.
189
disk, will alter the throw of the eccentric, suiting the duty of
the pump with an economical use of steam.
Slurry-Pump.-A pumping machine for the somewhat
unusual service of handling brick earth is shown in Fig. 150,
which is reproduced from Engineering. There are three
pump-barrels, each 10½ inches in diameter by 15 inches
stroke. The earth is moved through a 6-inch pipe nearly a
mile and three-quarters long, and is mixed with double its
volume of water. A series of trials have been made to find
the power required for the transportation of the earth, and the
results are given in Table XII.
TABLE XII.
SHOWING THE POWER REQUIRED FOR PUMPING BRICK EARTH. PUMP BY
TAYLOR & NEATE, ROCHESTER. EXPERIMENTS AT BRICK-YARDS OF
SMEED, DEAN & CO., SITTINGBOURNE, ENGLAND.
OBSERVED DATA.
PROBABLE DISTRIBUTION
OF POWER.
PROBABLE DISTRIBU-
TION OF POWER RE-
DUCED TO A MEAN
SPEED OF 130 REVOLU
TIONS PER MINUTE BY
DIRECT PROPORTION.

I
123456 780
114
133
124
C
3′.0"
30
122'Off. Off.
•
66
3′.0" 5.5 5 5
Off. Off. Off.
6.24 Off. Off. Off. 6.24
66
143
143
4.4"
66
66
31
7
37
4'.4"
26
66
7
28
66
35
143
136
3'.6″
8
23
18
7 8
23 18
56
22
15
7
8
104 16.670 + 5
141
25
12
7
151
3.0" 44
23
II
7
130 20.882 + 5
129 20.6 70 +5.
88, 14.1 75+5
mom mom 0 $
5777765
0865
♥ ♥ ∞ ∞ ∞ ∞
22 15 52
8 31 15 61
8
34 16 65
Note.-As the horse-power at 130 revolutions is calculated in direct proportion
only, it is probably considerably under the mark where the speeds were low. If
the power increased as the square of the speed, 90 I. H. P. would have been
required instead of 65, and this power ought to be available for emergencies.
190
PUMPING MACHINERY.
At these trials the entire plant consisted of two wash-mills,
well-pumps, and slurry-pumps, and the slurry-pumps driven
by a 20 H. P. portable engine with reduced boiler-pressure.
This engine was not powerful enough for the work, and was
usually assisted by a 12 H. P. portable engine driving on to
its fly-wheel; during the experiments, however, the 20 H. P.
engine was forced and the auxiliary disconnected.
The quantity of earth delivered through the 6-inch pipe
line, 2782 yards long, was equal to 1600 cubic yards per week
of 60 hours, mixed with about double this volume of water.
The power averaged through a number of diagrams was about
60 indicated horse-power, of which the slurry-pumps took
approximately half and the wash-mills the other half, the well-
pumps, which only lifted the water about 10 feet, being quite
unimportant. As only one man is required to attend to
the engine, the wash-mill, and the pumps, and coal is worth
$4.00 per ton delivered to the clay-pit, it is obvious that the
cost for current expenses (taking the consumption at 4 pounds
per I. H. P. per hour) will not exceed two cents per yard of
earth transported, or about one cent per ton per mile; and
when it is considered that the clay would have to be washed
in any case, the cost for transport comes to less than one-half
cent per ton per mile, which is probably the cheapest example
of land transport on record, considering the distance. Unfor-
tunately, the conditions of depth of slurry in the wash-mill,
proportion of water in the slurry, and the speed of working
varied so considerably that the results were very divergent.
The above may be taken, however, as fairly approximate, as
eight sets of diagrams were taken, and reduced by direct pro-
portion (for want of a better rule) to a mean speed of the
average of which has been given above.
Single-Crank Engine without Fly-Wheel.-—A steam-
end and valve-motion for a single-cylinder crank-pump without
a fly-wheel was ingeniously worked out by Shand, Mason &
Co., London. This engine was designed by them to meet the
objections taken to the fire-engines of this firm which carry a
STEAM AND POWER CRANK-PUMPS.
191
fly-wheel, and in consequence of this objectionable feature the
pump cannot have the valves so arranged as to be as easy of
access as they should
be.
Referring to
the steam-end only,
Figs. 151, 152, and
153 illustrate the man-
ner in which the im-
proved valve arrange- .
have
ments
been
carried out by which
a continuous rotary
motion of the crank-
shaft is obtained.
The crank is assisted
past the dead centres
without the aid of a
fly-wheel by means of
a piston fitted on the
slide-rod, which latter
is made to perform
alternately the duties
of a slide-rod and a
connecting - rod. As
a slide-rod it is im-
pelled by the crank-
shaft, whilst as a
connecting - rod the
crank-shaft is impelled
by it. Fig. 151 is an
elevation in part sec-
tion, Fig. 152 is a
sectional plan, and
m
M
N
E
FIG 1.
FIG. 151.
Fig. 153 an end elevation, partly in section, of the arrange-
ment for obtaining this result.
On the crank-shaft of engines in which the steam-supply is
regulated by a slide-valve an eccentric is fixed for the purpose
1

192
PUMPING MACHINERY.
of giving motion to that valve. The position of the eccentric
on the crank-shaft is considerably in advance of the crank,
FIG 152.
.FIG
日
​D
and usually above
90°. Therefore,
when the crank is at
either of the dead
centres, or at zero,
the eccentric is in a
favorable position
for assisting the
crank if auxiliary
power were supplied
to the slide when in
this position. The
slide would then be-
come the motor of
the crank-shaft, the
eccentric would act
as the crank, and the
eccentric-rod would
act as a connecting-
rod. Thus, when the
engine-crank is on
either of its dead cen-
tres, the slide auxil-
iary would be in full
action, and when the
slide - eccentric (or
crank) was on either
of its dead centres,
the
engine - crank
would be in full ac-
tion. By turning
these facts to ac-

count, the designers obtain a continuous rotary motion of the
crank-shaft, not requiring the moving of the crank past its
dead centres for starting nor the momentum of a fly-wheel
STEAM AND POWER CRANK-PUMPS.
193
.
while in motion, as there are no dead points on which the
engine would stop, and no part of its revolution at which it
would not start. To obtain this auxiliary power the slide-rod
is prolonged through the steam-chest into a small cylinder.
On the slide-rod
and in the small
cylinder is a pis-
ton, while there
is provided in the
steam - chest
a
small auxiliary
slide, which is
actuated by pro-
jections on it for
D
B
FIG. 153.

Τ
С
coming into con-
tact with a pro-
jection on the
main slide - valve
frame. The auxiliary valve is thus moved so as to admit the
steam to, or to release it from, the small cylinder, thus
actuating the piston fixed to the slide-rod, and through the
slide connecting-rod or cross-head conveying motion to the
small crank, and thus applying power for assisting the main
crank over its dead centres.
Referring to Figs. 151 and 152, A is the steam-valve; B, the
frame enclosing that valve; C, a small auxiliary valve; D, the
steam-chest; E, a small auxiliary cylinder; F, its piston; G, a
small crank for actuating the slide; H, the engine-crank work-
ing in a cross-head; I, the engine-cylinder; and J, its piston.
The piston is shown at the beginning of its stroke, the crank
H on the dead centre, and the small crank G in position for
opening the steam-port K, and at about 100° in advance of the
crank H. During the time the crank His on the centre, steam
is passing through the small port L and along the pipe M to
the end of the auxiliary cylinder E, and pressing the piston
F in the direction of the arrow will move the crank H by
the crank G past its dead centre into a position in which it
Ι
I 12
17
194
PUMPING MACHINERY.
1
has power. The crank H will then carry the crank G to its
dead centre, and the slide A to the end of its stroke. Mean-
FIG. 154.
D
while, the pro-
jection N on
the sling B will
touch the pro-
jection on the
slide Cand push
it across the
port L, opening
the same to the
exhaust, and
opening the
port S at the
other end of
the auxiliary
slide, and ad-
mitting steam
to the other side
of the small
piston F, to re-
peat the action
on the opposite
dead centres, the
two cranks thus
continuing to
assist each
other, and main-
taining a con-
tinuous rotation

of the
crank-
shaft P. The
small crank G
represents the
usual eccentric for working the slide-valve, its action being the
same; but, to avoid friction, its diameter has been reduced,
and it is called a crank.'
STEAM AND POWER CRANK-PUMPS.
195
.
A geared pump with eccentrics adapted to a duplex
water-end is shown in elevation in Fig. 154, and in plan in Fig.
155. This particular
pump has a water-end 8
inches bore by 12 inches
stroke, similar to the
sectional elevation, Fig.
116. This pump has
been working continu-
ously for three years
past at a pressure aver-
aging more than 100
pounds per square inch.
It was with some
hesitancy that the
writer adopted eccen-
trics when designing
this pump, but the suc-
cess attending the prac-
tical working of a num-
ber of power - pumps
thus operated has com-
pletely dispelled any
first doubts had regard-
ing the employment of
eccentrics for pumps of
12 inches stroke.
Pump-plungers of short
stroke operated by
eccentrics have long
been in use in this
country for boiler-feed-
ing and other purposes,
and have rendered good

doll
гото
FIG. 155.
and economical service. An advantage which such a train
of mechanism offers is its compactness. In this design the
eccentrics are bolted to the main gear, no power being trans-
196
PUMFING MACHINERY.

FIG. 156.
:
FOL
STEAM AND POWER CRANK-PUMPS.
197
:
mitted through the shaft, which serves only to take the thrust
of the water-plungers.
A geared crank-pump, such as shown in side elevation
in Fig. 156, and in end elevation in Fig. 157, is the kind recom-
mended
for
large pumps;
that is to say,
water-ends
having plung-
ers 14 inches
in diameter by
18 inches
stroke, and
larger. Pumps
of this kind
must be made
to suit the
conditions of
FIG. 157.

Ο
service for
each installa-
tion.
Greater
strength is
commonly giv-
en geared pow-
er-pumps than
O
is thought to suffice for steam-pumps, and especially is this
the case when a pump is to be driven by a train of gearing
attached to a water-wheel. In direct-acting steam-pumps
there is always an elastic cushion of steam in the cylinders.
with which to resist sudden jars or vibrations in the water
column, but if a water-wheel be employed, its rate of revolu-
tion being so much greater than that of the pump crank-
shaft, together with the fact that there is an inelastic medium
at both ends of the pumping machine, unusual strains occur,
for which ample provision must be made in the strength of
parts; it is also the common practice to operate such pumps
17*
198
PUMPING MACHINERY.
FIG. 158.
sizes, in such localities as do not readily afford either steam
or belt power. The experimental introduction of electrical
at lower piston-speeds than in the case of steam-pumps. For
ordinary steam-pumps the writer uses a factor of safety of ten,
but for pumps of this class he does not consider it prudent to
employ a factor of less than fifteen.
Electrical pumps, so called because driven by an electric
motor, are now being introduced, especially in the smaller

1
ROBERT WARNER & CO LONDON & WALTON ON THE NAZE
STEAM AND POWER CRANK-PUMPS.
199
pumps in mining operations promises well, but thus far no
reliable data has been published in which a comparison of the
best type of direct-acting engines has been brought into
direct competition with electric-motor pumps when handling
any considerable quantity of water, or when working under
heads similar to that called for in mines. No distinct type or
design of pump for this use has yet been adopted or followed
by leading pump-makers, the common practice being to simply
attach the electric motor to the pinion-shaft of any power-
pump of a size and type best suited to the work to be done.
The illustration, Fig. 158, represents a compact and well-
designed arrangement of a three-throw plunger-pump and
electric motor, with double gearing to adapt the two speeds to
each other. The pump is not a large one, it having a capacity.
somewhat less than 100 gallons per minute, the ratio of
gearing being suited to a delivery of 600 feet in height.
200
PUMPING MACHINERY.
CHAPTER X.
DIRECT-ACTING STEAM-PUMPS.
A direct-acting steam-pump is one in which the water
and the steam-ends are centrally in line with each other, so
that the water-plunger and the steam-piston can be attached
to the same piston-rod, both working together without the
intervention of a crank. This makes the simplest and most
compact form of steam pumping engine, and in its application
to all the ordinary operations of pumping outnumbers by far
all other varieties of pumping machines combined, and this
notwithstanding the fact that the ordinary direct-acting steam-
pump is, perhaps, the most wasteful and extravagant form of
steam-engine in use at this time. The remarkable growth of
the direct-acting pump business may be explained in part by
the commercial practice, which most buyers closely adhere to,
and that is to pay the least possible price for a pump, just as
they do for everything else. The direct-acting pump can be
more cheaply built than can a crank-pump for any given
capacity, and is, therefore, able to meet any possible compe-
tition in that direction; they also occupy less space than a
crank and fly-wheel pump, and thus compete sharply where
space is valuable enough to take that fact into account; they
weigh much less than crank-pumps, so that for use on ship-
board the lighter pump is generally chosen.
The practical workings of the leading designs of direct-
acting pumps have, in the main, been satisfactory from the
beginning; it is true that many and important changes have
been made in the valve-gear from time to time, so as to give
the steam-end the appearance of an entirely different design
from the original, but a close examination of the fundamental
DIRECT-ACTING STEAM-PUMPS.
201
:
•
principles would show that the departure was much less than
at first imagined. Two distinct types appeared about the
same time nearly fifty years ago; one in which the main
steam-valve was mechanically moved by a tappet-arm secured
to the piston-rod, operating adjustable stops attached to the
valve-rod; of this type the Worthington single pump is a
familiar example to the older engineers, the manufacture of
this particular design of pump having been discontinued for
several years. The other type was fitted with an auxiliary
steam-cylinder for operating the main valve by steam-pressure,
the pistons in the auxiliary cylinder being controlled by a
separate driving mechanism worked from the main piston-rod
of the pump, of this type the Knowles pump is a familiar
example.
The Knowles Valve Motion.-Fig. 159 is a sectional
view of the steam-cylinder and valve motion of the Knowles
pump. The construction is described as follows: An auxiliary
piston works in the steam-chest and drives the main valve.
This auxiliary, or "chest-piston," as it is called, is driven
backwards and forwards by the pressure of steam, carrying
with it the main valve, which valve, in turn, gives steam to
the main steam-piston that operates the pump. The main
valve is a plain slide-valve of the B form, working on a flat
seat.
The chest-piston is slightly rotated by the valve motion;
this rotative movement places the small steam-ports (which
are located in the under side of the chest-piston) in proper
contact with corresponding ports cut in the steam-chest. The
steam entering through the port at one end and filling the
space between the chest-piston and the head, drives the said
piston to the end of its stroke and, as before mentioned,
carries the main slide-valve with it. When the chest-piston
has travelled a certain distance, a port on the opposite end is
uncovered and steam there enters, stopping its farther travel
by giving it the necessary cushion. In other words, when
the rotative motion is given to the auxiliary or valve-driving
202
PUMPING MACHINERY.
PUM
piston by the mechanism outside, it opens the port to steam
admission on one end, and at the same time opens the port
on the other end to the exhaust. Thus instant and positive
motion is secured with but few working parts. There is no
point in the stroke at which either the chest-piston or the
main piston is not open to direct steam-pressure. A descrip-
tion of the operation of the pump is as follows: The piston-
rod, with its tappet-arm, moves backwards and forwards from
FIG. 159.
CHEST PISTON

Junny
F
ய
STEAM VALVE

und!
the impulse given by the steam-piston. At the lower part of
this tappet-arm is attached a stud or bolt on which there is a
friction-roller. This roller, coming in contact with the "rocker-
bar" at the end of each stroke, operates the latter. The mo-
tion given the "rocker-bar" is transmitted to the valve-rod by
means of the connection between, causing the valve-rod to
partially rotate. This action operates the chest-piston, which
carries with it the main slide-valve. The said valve giving
DIRECT-ACTING STEAM-PUMPS.
203
ניון
steam to the main piston, the operation of the pump is com-
plete and continuous. The upper end of the tappet-arm does
not come in contact with the tappets on the valve-rod, except
the steam-pressure from any cause fail to move the chest-piston,
in which case the tappet-arm moves it mechanically.
The friction-roller on the tappet-arm may be lowered or
raised when desired, to adjust the pump for a longer or
shorter stroke.
The Cameron Valve Motion.-A sectional view of the
steam-cylinder of the Cameron pump with its valve motion is
shown in Fig. 160
A is the steam-cylinder; C, the piston; D, the piston-rod;
L, the steam-chest; F, the chest-piston or plunger, the right-
FIG. 160.

ल
R
A
hand end of which is shown in section; G, the slide-valve;
H, a starting-bar connected with a handle on the outside; II
are reversing-valves; K K are the bonnets over reversing-
valve chambers; and E E are exhaust ports leading from the
ends of steam-chest direct to the main exhaust and closed by
204
PUMPING MACHINERY.
the reversing-valves II; N is the body piece connecting the
steam- and water-cylinders.
The operation of the pump is as follows: Steam is admitted
to the steam-chest, and, through small holes in the ends of
the plunger F, fills the spaces at the ends and the ports E E as
far as the reversing-valves II. With the plunger F and slide-
valve G in position to the right (as shown in the cut), steam
would be admitted to the right-hand end of the steam-cylin-
der A, and the piston C would be moved to the left. When
it reaches the reversing-valve I, it opens it and exhausts the
space at the left-hand end of the plunger F, through the
passage E; the expansion of steam at the right-hand end
changes the position of the plunger F, and with it the slide-
valve G, and the motion of the piston C is instantly reversed.
The same operation repeated makes the motion continuous.
In its movement the plunger F acts as a slide-valve to shut
off the ports E E, and is cushioned on the confined steam
between the ports and steam-chest cover. The reversing-
valves II are closed immediately the piston C leaves them,
by a pressure of steam on their outer ends, conveyed direct
from the steam-chest.
The Blake Valve Motion.-Fig. 161 shows sectional views
of the steam-cylinder, valves, etc., of Blake's steam-pump.
The following is the description of the operation of the
valves given in the company's catalogue:
The main or pump driving-piston A could not be made to
work slowly were the main valve to derive its movement
solely from this piston; for when this valve had reached the
centre of its stroke, in which position the ports leading to the
main cylinder would be closed, no steam could enter the
cylinder to act on said piston, consequently the latter would
come to rest, since its momentum would be insufficient to
keep it in motion, and the main valve would remain in its
central position or "dead centre."
dead centre." To shift this valve from
its central position and admit steam in front of the main
piston (whereby the motion of the piston is reversed and its
DIRECT-ACTING STEAM-PUMPS.
205
action continued), some agent independent of the main piston
must be used. In the Blake pump this independent agent is
the supplemental or valve driving-piston B.
The main valve, which controls the admission of steam to
and the escape of steam from the main cylinder, is divided
FIG. 161.

EXHAUST
K
STEAM

PPLEMENTAL
Ꮴ
Ꭻ
J
E
K
TRAZI
H
H
into two parts, one of which, C, slides upon a seat on the
main cylinder, and at the same time affords a seat for the other
part, D, which slides upon the upper face of C. As shown in
18
206
PUMPING MACHINERY.
the engravings, D is at the left-hand end of its stroke and C
at the opposite or right-hand end of its stroke. Steam from
the steam-chest J is therefore entering the right-hand end of
the main cylinder through the ports E and H, and the exhaust
is escaping through the ports H¹, E¹, K, and M, which
causes the main piston A to move from right to left. When
this piston has nearly reached the left-hand end of its cylinder
the valve motion (not shown) moves the valve-rod P, and thus
causes C, together with its supplemental valves R and S S'
(which form, with C, one casting), to be moved from right to
left. This movement causes steam to be admitted to the left-
hand end of the supplemental cylinder, whereby its piston B
will be forced towards the right, carrying D with it to the
opposite or right-hand end of its stroke; for the movement
of S closes N (the steam-port leading to the right-hand end),
and the movement of S' opens N¹ (the steam-port leading to
the opposite or left-hand end), at the same time the movement
of Vopens the right-hand end of this cylinder to the exhaust,
through the exhaust-ports X and Z. The parts C and D now
have positions opposite to those shown in the engravings, and
steam is therefore entering the main cylinder through the
ports E¹ and H¹, and escaping through the ports H, E, K,
and M, which will cause the main piston A to move in the
opposite direction, or from left to right, and operations simi-
lar to those already described will follow, when the piston
approaches the right-hand end of its cylinder. By this simple
arrangement the pump is rendered positive in its action; that
is, it will instantly start and continue working the moment
steam is admitted to the steam-chest.
The main piston A cannot strike the heads of its cylinder;
for the main valve has a lead, or, in other words, steam is
always admitted in front of said piston just before it reaches
either end of its cylinder, even should the supplemental piston
B be tardy in its action and remain with D at that end
towards which the piston A is moving, for C would be moved
far enough to open the steam-port leading to the main cylinder,
since the possible travel of C is greater than that of D.
.
DIRECT-ACTING STEAM-PUMPS.
207
The supplemental piston B cannot strike the heads of its
cylinder, for in its alternate passage beyond the exhaust-ports
X and X', it cushions on the vapor entrapped in the ends of
this cylinder.
The Clarkson Valve Motion.-The longitudinal and
transverse section of a

direct - acting
steam-
、 cylinder, shown in Fig.
162, presents some
peculiarities which are
quite interesting.
is
This pump is one of
a large class in which
the motion of the steam
distributing - valve
made quite automatic,
all eccentrics, tappets,
or valve-gear of any
kind being dispensed
with. The means by
which the valve is made
to move are here very
simple. The valve is
cylindrical, and capable
of an endlong motion,
without rotation, steam-
tight, in a cylindrical
box placed on the top
of the steam-cylinder.
By means of closed
ends and the two trans-
verse partitions the
valve is divided into
three separate compart-
O
FIG. 162.
ments. The steam-ports of the cylinder terminate, or rather
commence, in openings into the bottom of the valve-chamber,
208
PUMPING MACHINERY.
while the steam-pipe terminates in two openings in the top of
the same chamber. The end compartments of the valve have
holes top and bottom so arranged that in certain positions
they allow free communication between the cylinder-port and
the steam-pipe; the steam on being admitted to the cylinder,
therefore, passes through these compartments. The middle
compartment is open on the under side, and serves just the
same purpose as the hollow under an ordinary slide-valve,
making the communication between the steam- and exhaust-
ports. The steam-piston is double-ended, the space between
the two ends being always in free communication with the
exhaust by means of a hole in the cylinder-wall. The clear
distance between what are virtually the two pistons must
therefore be equal to the stroke of the engine plus the width
of the hole just mentioned. Two small holes about ½ inch
diameter are also made in the side of the cylinder, and pass-
ages from these communicate with the two ends of the valve-
chamber, each hole communicating with the end which is
farthest from it. These holes are so placed that when the
piston is at or near either end of its stroke one of them
opens into the space between the piston (and consequently
into the exhaust), and the other into the cylinder beyond one.
of the pistons.
%
We can now describe the action of the valve. When it is
in the position shown, the steam passing downwards through
the right-hand compartment of the valve forces the piston
from right to left, the steam on the left of the piston exhaust-
ing freely through the centre of the valve. As soon as the
right-hand piston passes the small hole, however, steam rushes
through it, and throws the valve suddenly over from left to
right (for the other end of the valve is in communication with
the exhaust), and so admits fresh steam against the left-hand
piston, and allows that which has just been doing the work to
be discharged. At the end of the stroke the piston is again
in the position shown in the engraving, the valve is thrown
over from right to left, and so the operations go on ad
infinitum.
DIRECT-ACTING STEAM-PUMPS.
209
The object of the small brass handle seen in the cross-
section is to start the engine if the valves should happen to
have stopped in the wrong place when the pump last ceased
working.
Dean Brothers' Valve-Gear.-The motion of the auxil-
iary steam slide-valve is continuous like that derived from an
eccentric. All other steam-pumps have an intermittent motion.
The ports leading to the chest-piston are closed, except at the
moment the main piston is being reversed; hence there can
be no "blow through"
or waste of steam in
case the chest-piston
becomes worn. The
stroke, of pump can be
instantly regulated by
moving the stud in slot
at the upper end of
lever. If raised, the
pump will make shorter
strokes; if lowered, it
will make longer
strokes.
The advantages
claimed for this steam-
valve gear are: It is
00
வ
FIG. 163.

noiseless. The auxiliary valve, having a long stroke and a
rapid motion, insures a prompt reversal of the piston at the
proper time. There can be no waste of steam or poor working
incident to the wearing of chest-piston. It is durable and
positive. The stroke of pump can be instantly changed.
Action of Valve-Gear.-The auxiliary valve F slides on
the valve-seat E2, and is provided on its under side with
diagonal exhaust-cavities d d. The ports bb' and exhaust-
port c are arranged in the shape of a triangle, and the diagonal
cavities diverge from each other, whereby the cavity d connects
18*
210
PUMPING MACHINERY.
the ports b and c, and cavity d¹ connects the ports b¹ and c
when the valve F is in extreme positions.
The operation is as follows: When the main piston moves
from left to right, the valve F is moved in an opposite direction,
opening the port b', admitting steam to the sub-cylinder E¹ at
the moment the main piston has reached the limit of its stroke,
whereby the auxiliary piston E is forced to the left, opening
FIG. 163 A.
H
F'
CE
d
20
H
G
the main port and
admitting steam to
the steam-cylinder,
consequently rever-
sing the movement
of the main piston.
On the return stroke
of the main piston
the movement of
the auxiliary valve
is reversed, whereby
the port b' is closed,
and at the moment
the main piston has
reached the limit of
its outer stroke, the
port b is opened by
the valve F, causing
the auxiliary piston
E to reverse its mo-
tion, opening the
main port and re-
versing the motion
of the main piston.
By this continu-

ous movement of the auxiliary slide-valve the ports b and b¹
are kept closed, except at the end of each stroke, at the
moment the main piston is reversed. This prevents any
waste of steam in case the auxiliary piston is worn enough
to leak as the ports b and b' are closed. The arrangement
DIRECT-ACTING STEAM-PUMPS.
211

الراد
H
K
FIG. 164.
212
PUMPING MACHINERY.
of ports admits of a short valve with a long travel. The
length of stroke of pump can be regulated by moving the
stud G2 up or down in the segmental slot G¹, as this varies.
the travel of the auxiliary valve, thereby reversing the stroke
of main piston earlier or later. The action is noiseless.
Davidson Valve Motion.-The mechanism of this valve
is illustrated in the accompanying drawing, Fig. 164. Its
principal peculiarity consists in the fact that it is moved both
by direct steam-pressure and by mechanical connection with
the main piston-rod of the engine. The cylindrical steam-
chest Mis bored out to make a face for the valve A and the
pistons B and B', which assist in operating the valve. Suffi-
cient space is allowed between the pistons for the valve and
the steam-ports. In the exhaust-steam passage is placed a
cylindrical cam, which is rocked back and forth by an arm
connecting with a cross-head clamped to the piston-rod. This
cam engages with a steel pin in the centre of the valve, and is
of such shape that it moves the valve back and forth, and also
oscillates the valve on its axis. By oscillating the valve small
passages are opened and closed, opening either end of the
valve-case alternately to steam and to exhaust. This gives a
quick opening of the ports and largely relieves the valve
motion of friction and wear.
When one of the main steam-ports ƒ is completely open,
admitting steam to the cylinder and driving main piston, cam,
and valve in direction shown by arrows, the first movement of
the cam will be to oscillate the valve preparatory to bringing
it in proper position for the opening of the auxiliary steam-
port e to live steam, and e' to exhaust. It will then bring the
valve to its closure (mechanically) slightly before the end of the
stroke of main piston, thereby causing slight cut-off and com-
pression. It will then fully open the auxiliary port e to steam,
and e' to exhaust. The admission of steam to one end of the
valve-piston, the other being open to exhaust, throws the valve
in the direction shown by the arrow, admitting and exhausting
steam to and from the main cylinder for the return stroke.
DIRECT-ACTING STEAM-PUMPS.
213
The advantages of this arrangement are that the pump has
no dead centre, the auxiliary ports e and e' being opened
whenever the main steam-ports ƒ and ƒ' are closed. The
closing of the valve before the piston reaches the end of the
stroke eases the action of the pump and prevents the piston
from striking the cylinder- heads. The fact that the valve is
moved mechanically as well as by steam also insures a uniform
length of stroke.
The Marsh valve-gear is shown in the sectional engrav-
ing, Fig. 165, which engraving also includes the water-valves
already described. This valve movement has no external
mechanism, the movements of the steam-actuated valve being
controlled by the movement of the steam-piston, and con-
FIG. 165.
T

versely. A section of the steam-chest on a larger scale is
shown in Fig. 165 A, and another detail of the valve in Fig.
165 B. The valve is solid and all in one piece, except that
one head is screwed on, as shown in the engraving.
214
PUMPING MACHINERY.
The steam-valve does not require setting. It has no dead
centre, and will always start when the steam is admitted. The
steam-piston, as shown, is double, and each head is provided
with a metal packing-ring, the interior space constituting a
reservoir for live-steam pressure, supplied by the live-steam
pipe through a drilled hole shown by dotted lines. At each
FIG. 165 A.

end of the steam-cylinder are similar holes leading to each
end of steam-chest, which, together with the centrally-drilled
hole and the space between the piston-heads, constitute posi-
tive means for tripping or reversing the valve with live steam.
The governing element of the valve is the enlarged heads,
which present differential areas to the action of steam.
FIG. 165 B.
The

inner area of the valve-head is reduced by an amount equal
to the cross-sectional area of the valve body, and is acted
upon by live-steam pressure, which varies with the annular
opening of the valve due to linear movement. The outer area
of the valve-head is larger, and is acted upon by the pressure
that moves the main piston. The working pressure in the
DIRECT-ACTING STEAM PUMPS.
215
main cylinder acting upon the larger area of the valve tends
to increase the annular opening of the valve, while the in-
coming steam acting upon the smaller area of the valve-head
tends to reduce the annular opening. Each movement finds
its limit in the preponderance of the opposing force. If the
piston moves easily much pressure cannot exist in the steam-
cylinder, as the piston would rapidly fly away from it, and as
the pressure in the steam-cylinder and outer end of valve-
chamber is always the same, the larger area of the valve-head
would have slight force to enlarge the portage when the piston
moved easily. On the other hand, if the piston moved hard,
or against much resistance, the pressure in the cylinder would
readily augment, and the larger area of the valve-head would
overcome the resistance of the inner area and increase the
portage of the valve.
The Guild and Garrison valve motion differs mate-
FIG 166.

F
C
ID
rially in its arrangement from other steam-thrown valves, in
that while the final stroke of the main valve is effected by a
216
PUMPING MACHINERY.
i
difference in steam-pressures, the greater part of its movement
is mechanically controlled by the steam-piston. A sectional
elevation of this valve-motion is shown in Fig. 166, and in
longitudinal elevation showing the exterior moving parts in
Fig. 167.
FIG. 167.

The steam-chest is a rectangular chamber; in its interior
are two small cylindrical openings, one at each end; there is
also at the side a raised seat for the auxiliary valve; these are
part of the casting. Small ports run from the ends of the
small cylinders to the seat of the auxiliary valve and to the
main exhaust-port. A plain cylindrical casting, each end con-
stituting a piston, fits into the small cylinders. This piece,
called the valve-driver, has two slots or openings at its centre,
the lower one just large enough to receive a lug on the back
of the main steam-valve, the top one being intended to receive
the large toe of the rock-shaft. The rock-shaft crosses the
steam-chest at right angles to the movement of the valves; it
is made of steel and has two toes, the larger one engaging
DIRECT-ACTING STEAM-PUMPS.
217
with the valve-driver by means of the slot just mentioned, and
the smaller toe fitting into the back of the auxiliary valve.
There is no lost motion in the latter. Both the main steam-
valve and the auxiliary valve are plain flat slide-valves, and
there are no ports whatever in the valve-driver. By means of
a lever and link connected with the piston-rod, any motion of
the rod causes the rock-shaft to rotate and the steam-valves
to move in unison.
The office of the main steam-valve, is, of course, to admit
steam into, or permit it to exhaust from, the main steam-cylin-
der of the pump; the auxiliary valve performs the same office
for the valve-driver cylinders in the chest, both being actuated
by the toes upon the rock-shaft. The auxiliary valve is a D-
valve, and its action is precisely the same as a D slide-valve
in a steam-engine, and its effect upon the valve-driver is exactly
the same as if it were a piston of a steam-engine.
In operation, the piston being at the end of its stroke and
the pump about to be started, steam is admitted, and the main
steam-piston moves forward; simultaneously motion is com-
municated to the rock-shaft by means of its connection with
the piston-rod, the valve-driver and the auxiliary valve are
mechanically moved at the same time and in the same direction
as the steam-piston. This action continues until the piston
has nearly completed its stroke, when the auxiliary valve un-
covers the small steam passage leading to one of the chest-
cylinders, steam enters the latter and exhausts from the oppo-
site chest-cylinder; the valve-driver is thereby driven ahead,
carrying the main steam-valve with it. The travel of the
main valve is thus completed, and it is brought in position to
reverse the stroke of the main steam-piston. If the pump
were now stopped, the valves would be found to occupy the
same relative position as at the beginning of the stroke, the
valve-driver having been carried forward ready to meet the
contact of the large toe on the rock-shaft upon its return.
It will be seen from this description that the valve-driver,
or in effect the main steam-valve, is not dependent for the
greater part of its movement upon differences of steam-press-
K
19
218
PUMPING MACHINERY.
ure, but that so long as the main pistons act it is compelled
to respond to their motion; and that when steam is finally
admitted to and exhausted from the chest-cylinders, this valve
is already in motion, and requires but a slight additional im-
pulse to finish its throw. It is also apparent that as the
motion of the auxiliary valve is practically continuous, there
can be no dead point, and that the pump will start from any
part of its stroke and run with a uniform motion.
The advantages claimed for this arrangement are positive
action and good wearing qualities. The construction is simple
and easily understood upon inspection.
The isochronal valve-gear, by the Gordon Steam-Pump
Company. Fig. 168 represents a steam pumping engine with
Lag
d'
C
FIG. 168.
F
e
B
D"
I
H
J
an ordinary steam-moved valve, the auxiliary valve being
moved by a lever driven by the main piston-rod. This lever
takes hold of a sliding-cylinder, H, whose piston, G, is fixed
?

DIRECT ACTING STEAM PUMPS.
219
upon a prolongation of the rod of the auxiliary piston. A
cock, L, places the two ends of this cataract-cylinder H in
communication, and makes the passage free or restricted as
required by circumstances.
Assume the cataract-cylinder to be empty. The machine
will then act as usual. At the end of a stroke the auxiliary
valve will be opened, and the auxiliary piston will open the
main steam-valve the full width, and so on for each stroke,
the cataract-piston not interfering at all with the usual move-
ment of the auxiliary piston. If something happens to the
discharge system, such, for instance, as the bursting of a main,
it is obvious that the unresisted pump-piston will be shot for-
ward with the full force of the steam, and damage is liable to
ensue. It is the office of the cataract-cylinder to prevent ex-
cessive motion of the main parts in such an emergency and
to insure a uniform piston-speed.
The cataract-cylinder is filled with liquid, and it is obvious
that the auxiliary piston in making its usual stroke must
needs pass the cataract liquid from one end of the cataract-
cylinder to the other, the cock L being so adjusted that this
transfer of the liquid can take place just so fast and no faster.
It follows that the auxiliary piston, at proper pumping
speeds, is uninterfered with by the cataract-piston, the liquid
simply passing from one end of the cylinder to the other.
The cataract-cylinder is always moving, and in case the
pump-piston should, by reason of resistance being removed,
attempt to jump ahead, the liquid in the cataract-cylinder
would have to pass much more rapidly through its restricted
passages. This it refuses to do; therefore the cataract-cylin-
der imparts, through the medium of its liquid, more or less
of its motion to the auxiliary piston and to the main valve,
thus closing the main valve more or less. Briefly, if the
pump seeks to move too fast it automatically affects a closure
of its valve, and if the pump moves too slowly the steam-
pressure upon the auxiliary piston preponderates over the
cataract resistance and an opening of the main valve is
effected. By this simple device complete control is had over
220
PUMPING MACHINERY.
•
the main parts. With it we do not think it possible that
serious damage could occur to a direct-acting steam pumping
engine. Cases have occurred in large machines where a dis-
charge-valve has been blown out, thus letting the full head of
water against the plunger on the back stroke. This would
mean destruction to an ordinary machine. The isochronal
machine, as its makers aver, "continues the even tenor of its
way under such circumstances." The makers test all pump-
ing machines under steam and full load, and one of the tests
of the isochronal is to restrict the discharge to the merest trifle
of an outlet. While the machine is working against this im-
mense resistance they suddenly open the discharge-valve full
width, thus removing all of the resistance. The cataract-
cylinder asserts itself and no running away occurs; in fact,
hardly a perceptible change occurs.
Valve-Gear with Cataract.-The steam-end illustrated
in Fig. 169 was designed by the writer, and exhibits the some-
what unusual feature in a steam-cylinder for a single direct-
acting pump in its having five ports; but it may be explained
that this valve motion was so designed that the same steam-
cylinder, such as used for duplex pumps, might be used for
single pumps also, the principal business at the time being
the manufacture of duplex pumps. The economic advantages
secured by this design in the factory consisted in the employ-
ment in common, by both styles of pumps, of the steam-cylin-
der, slide-valve (with slight alterations for driving it), piston,
piston-rod, cylinder-heads, and stuffing-box. Instead of the
ordinary steam-chest, such as usually furnished duplex pumps,
a special one was made not unlike those in common use for
single pumps, and which in our description we will call the
auxiliary cylinder. This cylinder has a double piston, one
fitted to each end, and midway between these two pistons are
collars for driving the main slide-valve. This auxiliary cylin-
der, like the main steam-cylinder, is provided with five ports,
and covering these is a plain flat slide-valve receiving its
motion from a rocker-arm, the shaft of which is operated by
DIRECT-ACTING STEAM-PUMPS.
221
the swinging movement of a lever driven by the cross-head
attached to the main piston-rod,
The two outer ports are for steam, the three inner ones are
for the exhaust. The slide-valve has no lap or lead on either
the steam or exhaust sides, consequently a steam- and its cor-
responding exhaust-port must be open from end to end of the
stroke. When the piston travels past an exhaust-port it
thereby cuts off all escape of exhaust steam from the cylin-
FIG. 169.
:

der, and compresses the remaining portion, which is a suffi-
cient cushion to prevent the piston striking the heads.
The small slide-valve above the auxiliary piston distributes
the steam so as to give motion to the piston underneath by
admitting steam at one end of the auxiliary cylinder, and un-
covering the exhaust-port leading from the other end, the
effect of which is to cause the auxiliary piston to move with
the pressure, a movement which continues until its forward
edge closes the exhaust-port, cushions the pent-up exhaust
steam, and with the increasing pressure thus obtained the
19*
222
PUMPING MACHINERY.
piston gradually comes to a state of rest; at the same time
that this movement is going on the main slide-valve has been
carried by the auxiliary piston in the same direction, effecting
a corresponding opening of the steam- and exhaust-ports in
the main cylinder, and producing a similar movement of the
main piston. When the main piston approaches and finally
reaches the end of its stroke the vibrating lever driven by the
main cross-head gives the segmental tappet a partial rotation,
which through the intervention of a rocker-arm and valve-rod
carries the small slide-valve over to the opposite end of the
valve-chest, and the whole operation is then reversed.
The steam from the boiler is admitted to the main steam-
chest; that is to say, to that space between the two auxiliary
pistons in which the main slide-valve is located; as the pistons
are both of the same area, they are not influenced in any way
by this pressure. There is a small hole (not shown in the
drawing) drilled from the steam-chest for the small slide-valve
over the auxiliary pistons, down into the steam space below,
so that boiler-pressure is also had for operating the auxiliary
pistons. The exhaust from the small slide-valve may lead
down into the main exhaust cavity, or into the atmosphere.
The adjustable cataract shown in the engraving was for the
purpose of steadying the movement of the auxiliary piston.
It consists of a piston and rod connecting directly to the aux-
iliary piston, so that their movements may be identical. The
rods pass through both ends of both cylinders, so that no
unbalanced pressures occur in the auxiliary steam-cylinder,
and no difference in areas exists in the two ends of the cataract-
cylinder. This latter cylinder has but one port extending
from end to end, as shown in the drawing; between the two
ends a slotted plug is inserted, by the partial rotation of which
any amount of opening from a full port to absolute closure
may be secured. This cylinder is to be filled with any con-
venient fluid, mineral oil, for example, and the plug inserted;
it is then ready for service. The function of the cataract is
to control the movement of the auxiliary piston and prevent
a violent movement by requiring the displacement of the oil
DIRECT-ACTING STEAM-PUMFS.
223
FIG. 170.
0
on one side of the cataract-
piston, forcing it through the
slotted plug and emptying
into the opposite end of the
cylinder; by a partial rota-
tion of this plug a greater
or less resistance is secured,
by which a quicker or slower
movement of the auxiliary
piston and main slide-valve
is also secured. Once this
plug is adjusted to the work
to be done by the main
pump, no further attention
need be given it, as the
pump will continue to op-
erate at this fixed number of
strokes per minute whether
there is any load on the
pump or not.

The deep-well pump-
ing engine, or artesian-well
pump, as commonly called,
illustrated in Fig. 170, is an
adaptation of the same kind
of valve motion as that just
described, differing only in a
few minor points of detail.
No cataract is supplied this
224
PUMPING MACHINERY.
steam-end, as another method of governing is employed. The
ports, main valve, auxiliary piston, and small slide-valve are
substantially the same in design. The method of moving the
tappets is different, and is probably the best device of the two,
as it is easier to make any needed adjustments than with the
segmental tappets.
Two dangerous accidents are likely to occur in deep-well
pumping,-one, the breakage of a pump-rod, the other, a fail-
ure of water-supply. Either of these is liable to damage the
steam-end, if not to wholly wreck it, especially when pumping
from great depths, unless some provision be made in advance
to meet such a contingency should it occur.
The cushioning of the exhaust is at once a convenient and
satisfactory method of controlling the movement of the main
piston. By reference to the engraving two semi-cylindrical
plugs are shown in the exhaust-ports, the function of which
is to control the flow of the exhaust steam from the cylinder
by giving these plugs a partial rotation so as to choke the
exhaust; a sufficient back pressure can be had to secure any
desired rate of piston-speed, fast or slow, with a full load or
without any load. As each plug may be separately adjusted,
a separate movement may be secured to the upward or to
the downward movement of the piston, adapting it to the
depth of the well or to any other conditions of supply.
Compounding Single Direct-Acting Pumps.—When
so desired, single direct-acting steam-ends may be arranged
for using steam expansively by the addition of a low-press-
ure cylinder, tandem to that of the high. The slide-valve-
faces for the high- and low-pressure cylinders should be in
the same plane by carrying up that of the smaller cylinder
to the level of the larger. The slide-valve in the high-press-
ure cylinder need not differ from that which would be used if
no compounding were attempted. The auxiliary piston would
differ only in having a rod passing through its cylinder-head
into the steam-chest of the low-pressure cylinder. To balance
the pressure, an extension of the same diameter should pass
DIRECT-ACTING STEAM-PUMPS.
225
through the opposite cylinder-head; and if a similar extension
be provided at the opposite end of the low-pressure steam-
chest, the balance would then be complete. The auxiliary
cylinder must be large enough to easily and promptly handle
both slide-valves; if this cylinder be one-half the diameter of
the high-pressure cylinder, ample power ought to be secured
for this purpose. The low-pressure valve may be a plain
slide-valve, with a suitable adjusting device for fixing its rela-
tion to the high-pressure valve, as these two valves must work
together.
Р
226
PUMPING MACHINERY.
4
CHAPTER XI.
THE DUPLEX PUMP.
THE credit for the invention of what is commonly known
as the duplex pump, distinguished for its almost ideal sim-
plicity, with its peculiarly efficient valve motion, to which it
owes its complete exemption from noise or concussive action,
is due the late Henry R. Worthington.
It consists of two steam-pumps, of equal dimensions,
placed side by side, with the valve motion so designed that
the movement of the steam-piston of each pump shall have
the controlling movement of the slide-valve of its opposite
pump, the effect of which is to allow one piston to proceed to
the end of the stroke, and gradually come to a state of rest;
during the latter part of this movement the opposite piston
then moves forward in its stroke, and also gradually comes to
a state of rest; but in its movement forward, and before
reaching the end of its stroke, the slide-valve controlling the
first piston is reversed, and in consequence the first piston re-
turns to its original position, and in nearing the end of its
stroke it, in a similar manner, reverses the slide-valve con-
trolling the second piston; these movements are both uniform
and continuous so long as steam is supplied to the pistons.
A sectional elevation through one of the steam-cylinders
of a duplex pump is shown in Fig. 171. To those not familiar
with the construction of duplex pumps, a noticeable feature
will be that the cylinder has five ports instead of three as in
ordinary engines. The two end ports are for the admission
of steam, the two inner ones are for its exhaust. The slide-
valve has neither lap nor lead on either the steam or exhaust
sides. The drawing shows the two ends of the steam-valve
to be exactly on line with the outer edges of the steam-ports;
THE DUPLEX PUMP.
227
the face of the valve extends over the steam-port, the bridge,
and to the inner edge of the exhaust-port, so that in its
FIG. 171.

E
present position no steam can enter or leave the cylinder. It
will be understood that the piston movement of one engine
controls the slide-valve not of its own cylinder, but the slide-
valve belonging
to the opposite
engine; this is
accomplished
by means of a
cross-head se-
curely fastened
to the piston-
rod operating a
lever, with
a
shaft extending
across the fram-
ing of the pump,
as shown in
Fig. 172. The
rocker-arms for
moving the
slide-valves are
FIG. 172.

O
arranged one above the shaft and the other below it; this
is for the purpose of securing for one engine a slide-valve
228
PUMFING MACHINERY.
!
movement with its driving piston, and the other one in an
opposite direction, otherwise no reversals at the end of the
stroke could occur.
In a direct-acting engine there is no controlling mechanism,
such as a crank and connecting-rod, to prevent the piston
overrunning its stroke; the arrangement of five ports in the
steam-cylinder effectually prevents the piston coming in con-
tact with the heads, and accomplishes it in the following
manner: The forward movement of the piston is due to
unbalanced pressure; so long as the exhaust is open its for-
ward movement is assured, but when the piston passes over
the inner or exhaust port there can be no further escape of
exhaust steam, and whatever quantity of steam should be
then remaining in the cylinder will be compressed between
the tightly-fitting piston and the cylinder-head, forming an
elastic cushion and effectually preventing direct contact be-
tween the two. It is the practice to allow a considerable
clearance in duplex steam-cylinders, for a 12-inch-stroke
pump the allowance being usually half an inch at each end
before contact; but before this limit is reached the slide-valve
belonging to this cylinder will have been carried over to its
opposite position, the steam-pressure admitted, and the reversal
of the piston movement occurs.
Dash-Relief Valves.- In the larger sizes of cylinders, say
14 inches or more in diameter, a drilled and reamed opening
FIG. 173.

EXHAUST
.STEAM
is formed at each end of the steam-cylinder between the
steam- and exhaust-ports; this opening is fitted with an
THE DUPLEX PUMP.
229
adjustable valve so that a greater or less opening may be
secured between these two ports. This arrangement, in one
These are called dash-
regulate the extent of
of its forms, is shown in Fig. 173.
relief valves, and are intended to
cushioning, adapting it to the peculiarities of the piston
movement, and to slightly lengthen the stroke after the piston
has closed the exhaust-port, especially when the pump is
working slowly and with a heavy load. This communication
between the two ports, with its adjustable valve, is simply a
controlled leak, useful in lengthening out the stroke in case
of excessive cushioning.
Lost motion between each valve and the nut by which it
is driven is almost always necessary in the final adjustment of
the slide-valve travel, it being a convenient method by which
to equalize the length of stroke of the steam-pistons on the
two sides of a duplex pump. The amount of lost motion
necessary to the proper working of a pump cannot always be
determined in advance, but after testing a few pumps from
new patterns of any given size, an amount of lost motion
necessary to that particular size or pattern of duplex pump is
had, and from that point forward the same quantity of lost
motion in duplicate pumps will secure substantially the same
results as to piston travel. This is not absolute, but it is near
enough to manufacture the parts, to assemble the pumps from
such patterns and send them to the testing room for final
adjustment. Small
such as
FIG. 174.

pumps,
boiler-feed and tank
pumps, having a
stroke of say 9
inches and less, re-
quire a lost motion
varying from % to
3% of an inch. The usual method of construction of the
slide-valve and the nut by which it is driven for small pumps
is shown in Fig. 174. The valve-nut is simply a square
20
230
PUMPING MACHINERY.
block of iron tapped and screwed on the end of the valve-
rod as shown. The method of adjustment may be as follows:
The steam-pistons are to be placed in the centre of the stroke;
the cross-heads are to be fastened to the piston-rods midway
of their stroke; the levers connected to the rock-shafts so that
the rocker-arms carrying the pins giving motion to the slide-
valves are in a vertical line, one rock-shaft pin above the
centre, and the other below it; the slide-valves are now to be
placed on the valve-seats and centred so as to cover all the
ports; the valve-nut must now be centred as shown in the
engraving; that is, there must be an equal amount of lost
motion on either side between the projections on the valve and
the nut by which it is to be driven. This adjustment must
be made in both valves in the manner indicated above, after
which this detail of the pump is ready for steam.
Duplex pumps of 10 or 12 inches stroke are usually fitted
with nuts at each end of the valve, as shown in Fig. 175.
FIG. 175.
The method of ad-
justment is the
same as described
in the preceding
paragraph. The
nuts offer a greater
facility in making
adjustments during
the operation of

testing, and as it sometimes happens that one side of the
pump needs an amount of lost motion greater or less than
the other, it can be had quickly and without disconnecting the
valve-rod from the rock-shaft pin. If care is exercised in
tightening the nuts after their adjustment, the probability of
their ever working loose is quite remote and rarely ever
happens.
Lost-motion links for pumps of 18 to 24 inches stroke
are commonly made as shown in Fig. 176. The rock-shaft
pin is fitted to the bored hole in the sliding-block. The link
THE DUPLEX PUMP.
231
to which the sliding-block is fitted has an end motion suited
to the requirements of the pump with which it is to be used.
The amount of lost mo-
tion may vary from ½ an
inch to I inch, depending
somewhat on the design of
the pump as well as the
conditions of service.
When testing a new
pump at the works with
such a link, a sliding-block
much shorter than is likely
to be needed is sometimes
used, and then by inserting
pieces of thin sheet-metal
Ø
FIG. 176.

о
into each end of the link opening, while the pump is in oper-
ation, an accurately determined quantity of lost motion can
be experimentally secured; then the new sliding-block in-
tended for the pump can be made to an exact length by
simply making it to the combined length of the short block
and the several thicknesses of sheet-metal used at each end in
the slot necessary to give the slide-valves their proper travel.
An adjustable lost-motion link, such as shown in Fig. 177,
is the kind usually included in the design and construction of
large pumping engines. The drawing is nearly self-explana-
tory; the sliding-block
is made with a shallow
recess at each end, in
FIG. 177.

which are fitted pieces of
園
​raw hide, against which,
the round heads of the
adjusting-screws at each end of the link may touch and com-
municate motion without noise. The screws are easily ad-
justable even when the pump is in motion. When the proper
amount of lost motion is had, the jam-nuts firmly fix the
adjusting-screws in place. This style of lost-motion link
permits a very wide range of adjustment.
232
PUMPING MACHINERY.
The general features of the valve-gear shown in Fig. 182
are reproduced in Fig. 178 on a somewhat larger scale, and
sufficiently sectioned to make the operation more easily un-
derstood. A is the piston-rod; B, a cross-head secured to the
piston-rod; C, a link connecting the cross-head with the lever
D; E, a rock-shaft operated by D, by which is driven the
rocker-arm F; G, a lost-motion link by which the valve-rod
cross-head His driven; I is the valve-rod by which the slide-
FIG. 178.

I
H
B
C
J
A
K
E
F
G
D
valve is moved; Jis an outer bearing for the valve-rod; K
is the upper portion of the cross-stand. This description in-
cludes only one set of valve-gear; the duplicate set for oper-
ating the other engine is partly shown, and is in all respects
the same except one rocker-arm must always be down below
the centre of the shaft, and the other one must always be
above it; that is to say, one valve-rod must always travel with
the piston by which it is driven, while that of the other engine
must travel in an opposite direction to that of the piston which
THE DUPLEX PUMP.
233
gives it motion. It will doubtless be understood that the
piston-rod A and the valve-rod I are not on the same, but are
on opposite, engines. The lever, the lost-motion link, and the
valve-rod cross-head, not lettered but shown in the drawing,
belong to the opposite engine, and were simply included to
show the relation of each to the other.
The action of a duplex valve-movement such as just
described must not be confounded with that of two crank
pumping engines arranged to work at right angles to each
other. It is not the same kind of movement, and the two
inethods of propulsion have nothing in common. A velocity
diagram showing the action of two double-acting pumps with
cranks at right angles has already been given in a previous
chapter (see Fig. 141). The flow of water from a duplex
FIG. 179.
Mean Velocity

One Stroke
-One Stroke
pump is so entirely different that the illustration, Fig. 179, will
hardly be recognized as one performing a similar service, but
as a matter of fact it represents approximately the flow from
a Worthington pump at each point of the stroke. As soon
as one pump begins to slow down at the end of the stroke
the other pump starts, so that by combining the flow it will be
seen how uniform it is.
In pumping engines of this type the weight of the moving
parts is reduced to a minimum, so that the elastic force of the
steam practically acts upon the water column directly, the
smoothness of working being well illustrated in the above
diagram.
The very great success attending the introduction of the
duplex pumping engine shows that it well provides the means
of pumping heavy columns of water with ease and safety to
20*
234
PUMPING MACHINERY.
·
the machinery employed, permitting the application of any
amount of power required to lift the water column without
violent or abrupt action upon the water, thus meeting an
acknowledged demand that the rate of movement of the water
column through the forcing-main shall be, as nearly as pos-
sible, uniform, so that no considerable alteration of pressure
shall be shown at any time while the pump is working. It
also meets the requirement that the propulsion of the water
shall be produced by the use of the smallest practicable
amount of moving material for transmitting the force of the
steam to the column of water in order to reduce to the lowest
point the momentum of moving parts, and the hurtful effects
due thereto in case of derangement of the valves or pipes.
The time allowed at the end of each stroke before the piston
takes up its return motion is sufficient to permit the water-
valves to seat quietly, and to allow the incoming supply to
completely fill the water-cylinder.
COMPOUND DIRECT-ACTING STEAM-PUMPS. 235
CHAPTER XII.
COMPOUND DIRECT-ACTING STEAM-PUMPS.
THE Compounding of direct-acting pumping engines is
made necessary if the heat wasted by the exhaust from such
engines is to be turned to a useful account by directly assist-
ing in the work of the engine. Single-cylinder direct-acting
pumping engines cannot of themselves, as ordinarily con-
structed, use steam expansively, because the terminal pressure
must equal the pump-load, which is, or ought to be, continuous
throughout the stroke; no greater steam-pressure is required
at the beginning than at its termination. The steam-pressure
is, therefore, the same from end to end of the stroke. Com-
pound steam-pumps are now made in considerable quantities,
with cylinders adapted to a range of expansion usually from
2 to 4 volumes, depending upon the initial steam-pressure, and
whether they are to be operated non-condensing or condensing.
The compounding of steam-cylinders for a direct-acting
pumping engine is a different problem to that of an ordinary
steam-engine. Practically there is no momentum of moving
parts to aid in expansion; there are no fly-wheels or other
devices for storing up energy in the early part of the stroke to
be given out at the end. The terminal pressure must be
sufficiently high to complete the stroke against the water-
pressure; the engine is, therefore, working at a disadvantage,
and the same degree of economy cannot be expected as if it
were an ordinary engine.
Gain by Compounding.-Non condensing compound
steam-ends are effective and economical when from 65 to 100
pounds boiler-pressure is used; the gain over the ordinary
direct-acting cylinders being from 20 to 35 per cent., depend-
236
PUMPING MACHINERY.
ing on the initial pressure and the ratio of expansion. Non-
condensing steam-ends are not recommended for pressures
below 50 pounds, as there will not be a saving sufficient to
warrant the additional cost of the machinery.
Tandem Direct-Acting Compound Steam-End.—The
high-pressure cylinder is supposed to take its steam directly
from the boiler and admitted at the same pressure. Practi-
cally this is seldom the case, and from 5 to 10 pounds less
than boiler-pressure is the ordinary available pressure.
It is not a general practice to supply receivers or tanks with
ordinary compound duplex pumps. The high- and low-
pressure cylinders are placed tandem to each other, with both
pistons attached to the same rod, as shown in Fig. 180. The
exhaust from the high-pressure cylinder passing directly into
the low-pressure steam-chest, and from thence into the low-
pressure cylinder.
The slide-valves of both cylinders have the same movement,
the ports being of equal dimensions in the direction of valve
travel. The valves having neither lap nor lead, the steam and
exhaust are operated at full stroke in both cylinders. The
pressure of steam in the low-pressure steam-chest will be
variable, depending on the distance which the pistons have
travelled towards the end of the stroke.
In this design the low-pressure cylinder is fitted with tie-
rods connecting it with the water-end; a substantial cast-iron
foot or base under the cylinder affords support for carrying
the weight of the entire steam-end, as it is not a common
practice to provide the high-pressure cylinder with any means
of support other than the end flanges connecting it with the
intermediate head. For small pumping engines it is the
practice to have the high- and low-pressure valve-seats in the
same plane, both valves being driven by the same valve-rod.
The lost motion may be had in each steam-chest by a proper
adjustment of the nuts at each end of the valve. The double
nut shown between the two steam-chests is simply to provide
a convenient means of connecting or disconnecting the valve-
COMPOUND DIRECT-ACTING STEAM-PUMPS.
237
rod, should it be necessary at any time to remove the high-
pressure cylinder; it has nothing to do with the valve
adjustment.
The valve-rod movement is constant, the variation in valve
travel being secured by the valve-nuts. The steam- and
exhaust-ports must be of the same dimensions in the direction
of valve travel, any difference in area that may be required can
be had by a proper width of port, as this detail is not affected
by the valve travel. The steam-ports in pumping engines are
usually two per cent. of the cylinder area, and in compounding
these same proportions are used in each cylinder respectively,
so that the width of the ports will vary for each diameter.
Dash-relief valves are fitted to the low-pressure cylinders only
if the engine is to be operated non-condensing, but if con-
densing, and if the high-pressure cylinder be of a size larger
than 12 inches diameter, they may be applied to each end
of each cylinder. These valves are not shown in the drawing,
but are similar to the one illustrated in Fig. 173. A side-pipe,
shown partly in dotted lines and in elevation, connects the
high-pressure exhaust-cavity with the low-pressure steam-
chest, each side of a duplex engine having its own side-pipe.
The steam-pipe shown at the end of the high-pressure steam-
chest extends across and connects to the two high-pressure
chests, but there must be no connection between low-pressure
steam-chests. The exhaust from the low-pressure cylinder
may lead directly into the air or into a condenser.
High-Service Attachment.-A compound pump is
sometimes required to work temporarily against a water-
pressure which cannot be had by the use of the ordinary
steam-pressure acting upon the high-pressure piston. Pas-
senger-elevators in office and public buildings are sometimes
used to elevate safes and other heavy weights, requiring a
water-pressure much greater than is employed regularly.
Small water-works pumping engines delivering into a reser-
voir, stand-pipe, or in direct system, frequently require that
the ordinary domestic pressure be doubled for fire-service.
238
PUMPING MACHINERY.
These and other reasons make it desirable that a compound
pump be quickly changed to a high-service non-compound
when emergencies arise.
To accomplish this requires nothing more than a direct
steam connection to the side-pipe (on each side of the pump)
leading from the exhaust of the high-pressure cylinder into
the steam-chest of the low-pressure cylinder. This pipe must
be fitted with a globe-valve for shutting off the high-pressure
steam when it is not wanted.
Its action is this: Suppose a non-condensing compound
pump to be regularly at work and a fire-alarm be sounded;
nothing requires to be done except to open the globe-valve
leading from the main steam-pipe into each exhaust side-pipe;
by this act the high-pressure cylinders will be thrown out of
service, because the exhaust pressure would balance the live-
steam pressure and no work could be done. The boiler-
pressure being now transformed directly to the low-pressure
pistons, which are probably 2½ times the area of the high-
pressure cylinder, will so increase the water-pressure that fire-
hose connections may be made directly to the fire-plugs.
When the fire or other emergency is over this valve may be
closed, and the engine goes on with its work as before.
The intermediate head connecting the high- and low-
pressure cylinders, as shown in Fig. 180, is one which the
writer believes to be superior to all others with which he is
acquainted. A loosely-fitting sleeve, preferably of cast iron,
is held in place by its flange, and so fitted that it may move
easily in any radial direction in the recess between the inter-
mediate casting and the bolted flange back of it. This joint
should be well made by scraping to a steam-tight surface on
both sides of the flange, otherwise there will be a steam leak
between the two cylinders. There may be a reasonable
allowance, as shown in the clearance lines, for radial move-
ment of this cast-iron sleeve to allow for any want of original
alignment; but as all the work is machine-fitted, there ought
not to be any considerable deviation from absolute truth.
COMPOUND DIRECT-ACTING STEAM-PUMPS.
239

FIG. 180.
240
PUMPING MACHINERY.
The high-pressure piston-rod passing through this sleeve will
in a short time, through such lubrication as the cylinder gets,
glaze the inside of the sleeve, producing a surface-finish not
subject to abrasion, and of great hardness combined with a
high polish.
For steam-ends having a stroke of 18 to 24 inches the
high-pressure cylinder is sometimes placed between the low-
pressure cylinder and the water-end, as shown in Fig. 181.
The high-pressure cylinder-head is fitted to receive the tie-
bars, but in other respects the design is substantially the
same, except that a lost-motion link is included in the valve-
gear and no lost motion is given the valves in the steam-chests.
Compound Steam-End for Large Engines. — For
pumping engines larger than those described, say from 24 to
48 inches stroke, it is customary to make them with the low-
pressure cylinder outside, the high-pressure cylinder inside, as
shown in Fig. 182, but without any intermediate head as pre-
viously described. In this drawing the usual method of con-
struction is shown, which is to include the flange for bolting the
high-pressure cylinder to the low in the same casting with the
former, and providing an intermediate cover to the smaller
cylinder as shown. The high-pressure cylinder-head through
which the piston-rod passes does not receive the tie-bars as in
smaller sizes, lugs being included in the cylinder-casting for
taking the strain of the engine. Steam-ends of this design
have three piston-rods, one for the high- and two for the low-
pressure piston,-a detail shown in the half-plan, Fig. 183, in
which a cross-head is common to the three piston-rods, as
well as the plunger-rod connection leading into the water-
end. The low-pressure rods have each a stuffing-box, which
is located well forward, almost to the high-pressure cylinder-
head, a pipe connection with flange securing each to the low-
pressure cylinder.
An advantage which this design offers over the ones pre-
viously described is that the high- and low-pressure pistons
may be removed, if necessary, without disturbing the main
COMPOUND DIRECT-ACTING STEAM-PUMPS.
241

FIG. 181.
L 9
21
242
PUMPING MACHINERY.

FIG. 182.
0
0
COMPOUND DIRECT-ACTING STEAM-PUMPS. 243

>
portions of the en-
re-
gine. For small
engines, the
moval of the high-
pressure cylinder
and intermediate
head is not much
of an undertaking,
but for large water-
works engines it
would be a serious
matter, which the
present design
makes wholly un-
necessary. The
valve motion as
shown in this en-
graving has been
previously illus-
trated and de-
scribed. (See Fig.
178.)
The cross-head
for large engines is
usually made
made of
forged open-hearth
steel; for the small-
er sizes a good qual-
ity of steel casting
has been found to
answer
the pur-
pose, but cast iron
is seldom, if ever,
used. The rods.
should be of a
good quality of
mild steel; the
1
0
0
Ο
о
O
FIG. 183.
244
PUMPING MACHINERY.
FIG. 184.
·

COMPOUND DIRECT-ACTING STEAM-PUMPS.
245
writer has used cold-rolled steel in such engines with excellent
satisfaction.
Intermediate Head with Stuffing-Boxes.-A com-
pound steam-end with an intermediate head furnished with
stuffing-boxes is shown in Fig 184. It does not often occur
that such a head must be used to the exclusion of the one
previously described, but it has this one advantage, that any
leakage around the rod may be detected and remedied by the
adjustment of, or the insertion of, new packing. Any one of
the several metallic piston-rod packings now offered to the
trade may be used in these stuffing-boxes in lieu of the fibrous
packing if desired.
Intermediate Head with Protected Rod.-A combi-
nation of a cast-iron sleeve with that of a stuffing-box is shown
in the sketch,
Fig. 185. This
is simply a pre-
liminary sketch
by the writer,
having
never
•
been used by
him, nor is he
sure whether or
not it may in-
terfere with an-
other's secured
rights. The in-
FIG. 185.

tent is to protect the high-pressure rod from contact with the
atmosphere, by the use of a long sleeve, as well as to prevent
leakage from one cylinder to another by means of the stuffing-
box and gland shown in the low-pressure head.
Action of Steam in Compound Direct-Acting En-
gines.-Let the upper part of Fig. 186 represent one side of a
tandem compound duplex pumping engine, but in the consid-
}
21*
246
PUMPING MACHINERY.
eration of the subject it is not necessary to take the duplex
feature into account, for the two engines are entirely separate
except in the single feature of valve-movement for the dis-
tribution of steam, therefore whatever remarks apply to one
FIG. 186.

10 9 8 7 6 5 4 3 2 4
90
ONE
YOLUME
80-
60
50
40
30-
THREE
YOLUMES.

g
20
10-
Atmospheric Line.
engine apply equally well to the other. The lower diagram
is drawn for 90 pounds absolute pressure; the low-pressure
cylinder has an area three times greater than that of the high;
both pistons are connected to the same rod, therefore the
same length of stroke. This diagram illustrates the action of
the steam between the high-pressure piston and that of the
low for each one-tenth of one stroke. For convenience we
have lengthened the low-pressure cylinder diagram corre-
sponding to three volumes, and we may further suppose the
high-pressure cylinder to have a cubic capacity of one foot, or
any other volume, real or assumed. The left-hand side of the
diagram represents the action of the steam in the high-press-
ure cylinder, and the right-hand side that of the low. The
exact detail of these two cylinders may be the same as the
sectional elevation, Fig. 180. No receiver is employed in this
engine, but the volume of ports, pipe, and low-pressure steam-
COMPOUND DIRECT-ACTING STEAM-PUMPS.
247
chest is taken to be 50 per cent. of the high-pressure cylinder,
the exhaust from the high-pressure cylinder being led directly
into the low-pressure steam-chest as shown in the drawing.
The engine is represented as non-condensing in the diagram,
but as all pressures are reckoned from a vacuum, a slight
change in the diagram will make it apply equally well to con-
densing-engines.
When steam is admitted at A the pressure immediately
rises to B, 90 pounds above vacuum, the piston begins its for-
ward movement, and will continue to the end of the stroke at
C; the inner edge of the high-pressure piston is marked 10,
and the inner edge of the low-pressure piston is marked o;
when, therefore, the one piston has travelled to 9, the other is
at a; but before considering the expansion curve in the dia-
gram it is necessary to take account of the clearance space
between the two cylinders. In engines of this type the steam
is exhausted from the high-pressure cylinder at the same
pressure used in doing its work, which we have placed at 90
pounds in the diagram. The clearance volume in the ports,
side-pipe, and low-pressure steam-chest has been assumed to
be 50 per cent. of the high-pressure cylinder. If this space
were empty the initial pressure for the low-pressure piston
would be one volume of steam at 90 pounds expanded into
12½ volumes=60 pounds; but there was remaining in this
space from the last stroke a volume of steam having a tension
equal to the terminal pressure upon the low-pressure piston
at the completion of its stroke, say 24 pounds; this must be
taken into account, so that the pressure in the low-pressure
chest would be increased thus:
66
12
I vol. at 90 pounds 90 pounds from the H. P. cylinder.
½ vol. at 24
terminal pressure in L. P. cylinder.
1.5 volumes
) 102 ( 68 pounds.
This 68 pounds becomes, then, the initial pressure for the
low-pressure cylinder, as it is also a back pressure against the
high-pressure piston. This reduction of pressure occurs at
the moment of reversal of the slide-valve in the high-pressure
248
PUMPING MACHINERY.
|
steam-chest, and at a time when the reversal of pistons is
about to occur, but the movement is not yet begun.
Returning now to the consideration of the diagram, Fig.
186, when the high-pressure piston is at 9, and that of the
low is at a, a reduction of pressure is shown, and it is that
due to the expanded volume of steam, or difference between a
contraction of one-tenth of the high-pressure cylinder volume
and the enlargement by three-tenths in the low-pressure
cylinder volume. When the pistons reach 8 and 6 respec-
tively we have the difference between a contraction of two-
tenths in the high-pressure cylinder and an enlargement of
six-tenths in the low, which reduces the pressure to b, and in
like manner for other pressures.
It will be a near enough approximation, if we consider the
back-pressure line of the high-pressure cylinder, to follow the
same curve as the expansion in the low-pressure cylinder, but
a slight variation occurs in practice, due to the resistance of
the steam flowing through the passages between the two
cylinders. The effect of the clearance space between the two
cylinders is shown first in the fall of pressure at c, in Fig.
186, at which point the highest pressure in the low-pressure
cylinder is had and the greatest back pressure in the high-
pressure cylinder. The total effect being observable in the
difference between the extremes of the curved line terminating
at O.
The exhaust side of the high-pressure cylinder and the
pressure side of the low-pressure cylinder are always in direct
communication; the pressure must, therefore, be the same in
both cylinders. The left-hand diagram shows the back press-
ure in the high-pressure cylinder until o is reached, a similar
line is traced in the right-hand diagram until k is reached,
and the low-pressure exhaust-valve opened to the atmosphere,
causing an immediate drop in the pressure to the back-
pressure line, which is drawn at 18 pounds, or about three
pounds above the atmosphere. The high-pressure cylinder
not being in communication with the atmosphere, its terminal
pressure is never below that in the low-pressure steam-chest,
COMPOUND DIRECT-ACTING STEAM-PUMPS.
249
Diameter.
Area.
HIGH-
PRESSURE
CYLINDERS.
Nearest
Diameter.
a pressure we have estimated to be 24 pounds. The pistons
having reached the end of the stroke, the slide-valves moved
to the opposite throw, the operation of the engine is now
reversed, and as its action is precisely similar to that just
described, no further comment is necessary.
This tracing
of the valve movement of a single pair of compound cylinders
is equally applicable to the duplex valve movement, because
such steam-cylinders work independently of each other.
Ratios of High- and Low-Pressure Cylinders.-Table
XIII. gives the actual dimensions for the several ratios of
compounding, but in practice these figures are seldom adhered
¡
The areas are calculated, and opposite each is the corresponding diameter to the
nearest inch.
Area.
Nearest
Diameter.
Area.
TABLE XIII.
SHOWING THE RATIOS OF HIGH- AND LOW-PRESSURE STEAM-CYLINDers.

Low-PRESSure CylindERS.

2 to 1. 2½ to 1.
3 to 1. 3% to 1. 4 to 1.

Nearest
Diamet r.
Area.
678 9
28.3
38.5
812- 56.6
98+
IO
50.3 11+ 100.6
63.6 1234- 127.2 144-
78.5 148+ 157.0 15%-
9½-
TI
77.0
+
1258-
70 7 10%+
96.2 1218+
125.7 1334+
159.0 1558-
84.9 114
99.1
12
113.1
115.5 13%
150.9 15
190.8 16%-
134.8
14
153.9
176.1
16
201.1
222.6
18
254.5
196.3 174+
235-5 1834-
274.8
20
314.2
II
95.0 152+
190.0 1738-
237.5 19 +
285.0 20½ +
332.5
22
380.1
12
113.1 17
226.2 19
282.8 2034+
339.3 222-
396.0 24
452.4
13
132.7 18%+
18%+ 265.4
265.4 2012+ 331.8 222+
398.12438-
464 5
26
530.9
14
153.91934+
307.8 2218+ 384.8 244- 461.7 268+
538 6
28
615 8
15
176.7 214
353.4 2334- 441.8 26
530.1 2818-
618.5
30
706.9
16
201.I 2258
402.2 254+
502.8 2734-603.3 30
703.9 32
804.3
17
227.0 24 +
454.0 26% +
567.5 292 681.0 31%-
794.5 34
907.9
18
254.5 252-
509.0 282-
19
283.5 268-
567.0 30 +
20
314.2 284
628.4 315%
21
346.4 2934
692.8 334-
22
23
24
636.3 318+ 763.5 33%+ 890.8 36
708.8 32%+ 850.5 3558- 992.3
785.5 345%+ 942.6 373% 1099.7
866.0 363% 1039.2 394+ 1212.4
380.1 3118- 760.2 3434+ 950.3 388-1140.3 4118+ 1330.4
|
415.5 322+ 831.0 363% 1038.8 398- 1246.5 43 + 1454.3
452.4 33% 904.8 38 1131.0 415%- 1357 2 448 + 1583.4
|
1017.9
38
1134.I
40
1256.6
42
44
1385.4
1520.5
46
1661 9
1809.6
Nearest
Diameter.
Area.
Nearest
Diameter.
Area.
250.
PUMPING MACHINERY.
to, as fractional parts of an inch are always avoided in steam-
cylinders if whole numbers can be used instead; for example,
if a high-pressure cylinder 16 inches in diameter be selected
for a given pumping engine, and a ratio of expansion of 3 to
I was desired, the table gives 2734 inches as a suitable
diameter; the probability is that 28 inches would be selected
instead, but the general tendency is towards smaller ratios of
expansion. The ordinary or trade sizes of compound steam-
ends for engines of 10 to 18 inches stroke are given in Table
XIV.
TABLE XIV.
COMMERCIAL SIZES FOR COMPOUND STEAM-ENDS, OR THOSE ORDINARILY
FURNISHED DUPLEX PUMPS FROM IO TO 18 INCHES STROKE.

DIAMETERS.
AREAS.
RATIO OF
EXPANSION.
High Pressure.
Low Pressure. High Pressure.
Low Pressure.
678a
ΙΟ
28.3
78.5
2.78
12
38.5
113.1
2.94
12
50.3
113.1
2.25
9
14
63.6
153.9
2.42
10
16
78.5
201.0
2.56
12
18
113.1
254.5
2.25
14
20
153.9
314.2
2.04
16
24
201.0
452.4
2.25
18
30
254.5
706.9
2.77
A noticeable feature in this table is the comparatively low
ratio of expansion, but it should be borne in mind that these
are nearly all small sizes, and are used at pressures rarely
exceeding 75 pounds boiler-pressure, and often not more than
60 pounds; the last three or four sizes being, in general, the
only ones fitted with condensing apparatus.
COMPOUND DIRECT-ACTING STEAM-PUMPS.
251
*.
TABLE XV.
EXAMPLES OF COMPOUND DUPLEX ENGINE PERFORMANCE.
Size of Engine.
At Rest.
Pressure.
Speed.
Pressure.
Speed.
Boiler.
At Speed in
Strokes per
Minute.
H. P. Cylinder,
L. P. Cylinder.
WATER-PRessure.
STEAM-
Pressure.
Exhaust BACK PEESSUre,
Suction.
Delivery.
Steam.
Exhaust.
Water.
Steam.
SIZE OF PIPES.
STATIC HEAD IN FEET.


8 and 12 X 7 X 10
48
8 and 12 X 7 X 10
8 and 12 X 8 X 10
8 and 12 X 8½ X 10
10 and 16 × 104 X 10
10 and 16 X 10 X 10
12 and 182 X 12 X 10
12 and 181½ X 12 X 10
18½ and 29 X 16 X 18
бо
69 100
•
38
50 106 52 200
72 200
52 150 57 180
•
29
37 160
•
36
44 160 47
200
52
52 110
56 | 160 | 60
52
888
62
70
160
68
54 100 57 180 70 бо 15
65 100
75 60 80 92
21 and 36% 62 X 36 200 1125
бо
•
•
100
8 32UNG 28 78
56
14
60
57
60
75
94
ཁྫོ*#84833
50
13
18
53
13
17
15
7
3½
7
55 15
3
7
47 9
25
14 12
3
432 2 343 m a∞
2 2 2 2 2 2 2 245
5534 in 70 6
65666
2230 H H ( Mom
4
32
4
III
128
88
212
67
83
120
120
143
157
462
The last pump tabulated was performing the very unusual
service of pumping through a delivery 24 miles long.
The others represent fair average performance of compound
duplex non-condensing pumping engines.
Steam-Jackets. It is not a common practice to jacket
the smaller sizes of steam-cylinders, whether in single or
duplex pumping engines; that is to say, for sizes smaller than
18- and 30-inch compound. It is not believed, for the smaller
sizes at least, that the future saving will repay the original
cost; but if a partial jacketing be thought desirable, it should
be the low-pressure cylinders rather than the high. If the
high-pressure cylinder only is jacketed, the exhaust from it
into the low-pressure cylinder suffers by excessive condensa-
tion because of its lower temperature; but if the latter cylinder
is jacketed, this condensation is less in quantity, and in so far
a step towards more economical working.
The usual practice is to jacket the barrel of the cylinder
only; the heads are seldom jacketed except in very large
engines. A steam-jacket must be designed with reference to
proper circulation and drainage, otherwise it may prove
252
PUMPING MACHINERY.
worthless, and defeat the very end to which it was applied.
The steam used in the jacket should be at full boiler-pressure,
and provision made for returning any water of condensation
back to the boiler.
The degree of expansion in ordinary pumping engines is
not sufficiently great to require the many refinements which
are thought to be necessary in the case of large engines
where high pressures are employed and expansion carried to
the extreme limit; in the latter engines jackets are always
employed.
The actual value of the steam-jacket, first applied by Watt,
has always been called in question. Tredgold, the earliest
writer on the steam-engine, condemned it, but it steadily grew
into favor with engineers, and it is now the general belief that
expansive engines cannot work to any good purpose without
one. A few years ago it was thought that its application
effected a saving from 10 to 30 per cent., but the lowest of
these two figures is not now believed to be possible even
under the best conditions.
The best series of experiments undertaken up to this time
to determine the exact truth in relation to steam-jackets were
those undertaken and conducted by Professor J. E. Denton
in 1889, the high-duty Corliss engine of the Pawtucket water-
works having been placed at his disposal.
The details of this engine are tolerably well known, but it
may be well to repeat that it is a horizontal cross compound
engine, steam-cylinders 15 and 30% inches bore; water-
cylinders, 10.52 inches; stroke of all pistons, 30 inches;
clearance, high-pressure cylinder, 4 per cent.; low, 3.7 per
cent. Diameter of rods, 234 inches. Ratio of volumes of
cylinders, 4.085. Average cut-off in high-pressure cylinders,
one-fourth, and in low, one-third. Jackets envelop the barrels
but not the heads of both cylinders, and steam of full boiler-
pressure is used in each. The heads are not jacketed, but
contain passages leading to and from the ports. The condensed
steam from the jackets is pumped into the feed-pipe at a point
between the boiler and hot-well.
*
COMPOUND DIRECT-ACTING STEAM-PUMPS.
253
The experiments were conducted along the lines of every-
day usage and service, and have an especial value for this
reason. The influence of the steam-jackets on the Pawtucket
pumping engine form the subject of an elaborate and con-
vincing paper by Professor Denton, published in Vol. XI.,
"Transactions of the American Society of Mechanical En-
gineers." The conclusions reached by him are as follows:
1. That the averages of results of indicator-cards, taken in
the most careful manner with the best modern indicators,
show a possible saving from the use of jackets amounting to
from 0.13 to 0.35 pound of steam per hour per horse-power,
but that these amounts are within the limit of error to which
the determination of indicated horse-power and cut-offs are
subjected, so that
2. The most that can be claimed for the jacket is that it
probably caused no loss, and may possibly cause a saving not
exceeding 3 per cent. of the total steam consumption.
Lagging steam-cylinders with wood or other non-con-
ducting material in lieu of jackets is very common. This does
not, of course, take the place of a steam-jacket, nor is it the
intention to arrest cylinder condensation by its application; its
only function is to prevent as far as practicable external radia-
tion from the steam-cylinder. The best non-conductors of
heat are hair felt, cork, fossil meal, magnesia, charcoal, and
rice chaff. A cylinder clothed with hair felt, and afterwards
covered with reeded black walnut staves, held in place by a
polished brass band about two inches wide at each end of the
cylinder, makes a good and at the same time a very neat finish
for a steam-end.
Circular slide-valves are sometimes used, as shown in
Fig. 187. The arrangement of the ports and of the valve
does not essentially differ from that of the ordinary duplex
steam-cylinder, except in the matter of change of form of the
valve and its seat from flat to circular. The necessary lost
motion for the slide-valves is secured by the employment of
22
254
PUMPING MACHINERY.
..
/
adjustable lost-motion links at the rocker-arms; the method
of operating the valve within the steam-chest being similar to
that of a Corliss steam-valve makes it inconvenient to have
any lost motion there. The centring of the two valves is
secured through the use of a right- and left-hand nut attached
to the valve connecting-rod, by which it may be lengthened
or shortened to get any suitable distance from centre to centre.
FIG. 187.

THER
划
​This nut is shown between the two cylinders. It is not neces-
sary that the two valve-seats be bored to the same diameter,
but care must be exercised in designing that the valve-move-
ments be precisely alike in the matter of admission and port
closure: this is easily secured by making the high-pressure
ports shorter, or by so proportioning the high-pressure
valve-lever that the two valves shall open and close exactly
alike.
A tank-engine is a duplex pumping engine in which there
is one high-pressure cylinder and one low-pressure cylinder
placed side by side as in the construction of an ordinary
COMPOUND DIRECT-ACTING STEAM-PUMPS.
255
duplex pump, with a large receiver or tank, into which the
high-pressure cylinder exhausts, and from which the low-
pressure cylinder draws its supply. This type of engine
probably originated with Ernest Wolff,* who patented it in
England in 1834. The essential features of his patent were
thus described by him:
"The invention consists of the combination of two or more
engines, each complete in all its parts, and so disposed that
while the first receives steam at one, two, or more atmos-
pheres of pressure, the next engine is moved by the steam
that escapes from the first. In the last engine the steam is
condensed in the ordinary way, or escapes in the atmos-
phere. The work supplied by the several engines is applied
to the same shaft, or to several combined, or to independent
shafts.
"It is sometimes useful to have between the cylinders an
intermediate reservoir to regulate the pressure; this may be
placed with advantage at the base of the chimney, so as to
maintain or raise the temperature and the pressure of the
steam in its passage from one cylinder to the other. Indeed,
if necessary, the heat may be supplied by a special fire-box.
"It is often necessary to employ a special pipe with a stop-
cock to admit the steam from the boiler to an intermediate
reservoir in order to give the machine the power of starting
any crank. This direct introduction may be employed to
increase for a time the power of the engine."
Certain modifications of this engine were patented by
Worthington in 1871, and engines for situations favorable to
its use have been built from time to time, but its introduction
has not been general. In these engines a large receiver or
tank is a necessity, not less than say ten times the volume of
the high-pressure cylinder; it is from this tank that the low-
pressure cylinder draws its supply as if from a steam-boiler.
This large tank is necessary that the contribution of the high-
* This invention must not be confounded with another engine patented in 1804
by Woolf. Notice the difference in spelling.
256
PUMPING MACHINERY.
5
pressure cylinder to its contents, or the withdrawal of steam
from it by the low-pressure cylinder, is not sufficient to vary
its pressure in any considerable or troublesome degree. The
condensing apparatus is attached, of course, only to the low-
pressure cylinder. A careful consideration of the working
conditions under which this engine is found will show that it
must be accurately proportioned as well to its work as to the
proposed pressure of steam, in order to secure anything like
an equal division of the resistance between the two pistons.
And it is necessary to do this with some reasonable degree
of exactness, as upon it depend that uniformity of water-flow
and stability of pressure which give to this engine its advan-
tages as an hydraulic motor. Economically this engine holds
about the same position as the four-cylinder engine, and costs
somewhat less. It is not, however, universally applicable, and
is not likely ever to supersede the present types of duplex,
either ordinary or compound.
Power of Compound Pumping Engines.-The paper
on the "Power of Compound Pumping Engines,” by John W.
Hill, in Engineering News, vol. xxi., is laid under contribution
for the subjoined tables and formulas.
Table XVI. contains the general data for the steam-ends
of four proportions or classes of compound, duplex, direct-
acting pumping engines; thus,—
Class A embraces engines where the diameter of the low-
pressure piston is 1.5 times the diameter of the high-pressure
piston, as 8 and 12, 12 and 18, 16 and 24, etc.
Class B embraces engines where the diameter of the low-
pressure piston is 1.6 times the diameter of the high-pressure
piston, as 10 and 16, 15 and 24, etc.
Class C embraces engines where the diameter of the low-
pressure piston is 1.57 times the diameter of the high-pressure
piston, as 14 and 22, 21 and 33, etc.
Class D embraces engines where the diameter of the low-
pressure piston is two times the diameter of the high-pressure
piston, as 12 and 24, 16 and 32, 18 and 36, etc.
COMPOUND DIRECT-ACTING STEAM-PUMPS.
257
The table contains the piston displacements, clearances, and
expansions, based upon volume (1) and expansion (1) in high-
pressure cylinders, the hyperbolic logarithms for the effective
expansions, and the reciprocals of total expansion.
TABLE XVI.
COMPOUND DUPLEX DIRECT-ACTING ENGINES.
Cylinder Volumes and Clearances stated in Terms of High-Pressure Cylinders and
Steam Full-Stroke, High-Pressure Cylinders.
TERMS.
CLASS OF Engine.
A.
B.
C.
D.
Intermediate space
•
Diameter C. P. cylinder (H. P. = = 1.00). 1.5
High-pressure cylinder, area
•
clearance
1.6
1.57
2.0
1.000
1.000
1.000
1.000
0.0630 | 0.0630
0.3388 0.3388
0.0630 0.0600
0.3388
I.1200
Low-pressure cylinder, area
clearance
Expansion, high-pressure cylinder
intermediate space
•
Expansion, low-pressure cylinder
2.2500
2.5600 2.4700
4.0000
0.1125
0.1280
0.1235
0.2040
I.0000
1.000
I 0000
•
•
1.3190
1.8254
2.6000
I.4391
2.0200
1.4350
I 9640
1.0000
2.2500
2.2584
2 9067
2.8182
5.0814
1.8254
2.0200
1.9640
2.2584
0.6018
0.3846
Total expansion
Effective
Hyperbolic Logarithm.
Reciprocal total expansion .
0.7031 0.6750 | 0.8146
0.3440 | 0.3548 o 1968
Table XVII. is calculated for compound duplex direct-
acting pumping engines, when worked condensing, and gives
for each of the four classes the mean effective pressure reduced
to work of low-pressure piston for the absolute pressure of
column I; for example:
The steam-end of a compound duplex direct-acting con-
densing engine has cylinders 16 and 24 inches diameter, and
the steam-pressure is 70 pounds. What size or diameter of
pump will this work against a pressure or head of 80 pounds
(184.64 feet)?
The data will be found in Class A.
r
22*
258
PUMPING MACHINERY.
TABLE XVII.
MEAN EFFECTIVE PRESSURES, ENGINE WORKED CONDENSING.
Compound Duplex Direct-Acting Pumping Engines; Steam Full-Stroke, High-
Pressure Cylinders; Counter-Pressure, 3.5 Pounds Absolute.

Class of Engine.
A.
B.
C.
D.
INITIAL
PRESSURE
ABSOLUTE.
Constants.
0 72869
0.68252
0.69542
0.46587
65
43.864
40.864
41.702
26.781
70
47.508
44.277
45.679
29.III
75
51.152
47.689
48.656
31.440
80
54.795
51.102
52.133
33.769
85
58.438
54.515
55.611
36.098
90
62.082
57.927
59.088
38.428
95
62.725
61.340
62.565
40.659
100
69.369
64.753
66.042
43.087
105
73.012
68.165
69.519
45.418
ΙΙΟ
76.656
71.578
72.996
47.746
115
80.299
74.990
76.473
50.077
I 20
83.943
78.403
79.950
52.407
125
87.586
81.816
83.427
54.736
130
91.229
85.228
86.904
57.063
135
94.873
88.641
90.382
59.392
The area of the low-pressure piston (24 inches diameter)
is 452.39 square inches, and mean effective pressure for abso-
lute pressure of 85 pounds (70 by gauge) is 58.438, and
moment of water-load-.
452.39 X 58.438 X .85 22471.24 pounds.
The head is 80 pounds, and area of plunger which a steam-
end containing high-pressure piston 16 inches diameter, and
low-pressure piston 24 inches diameter, with steam-pressure
of 70 pounds, will drive-
22471.24
80
280.89 square inches,
corresponding to a diameter of 19 inches nearly. Two such
pumps, double-acting, will deliver at 100 feet piston-speed per
minute 4,034,309.9 gallons per day of 24 hours.
•
COMPOUND DIRECT-ACTING STEAM-PUMPS. 259
Table XVIII. contains the mean effective pressures for
compound duplex direct-acting pumping engines worked
non-condensing.
TABLE XVIII.
MEAN EFFECTIVE PRESSURES, ENGINE WORKED NON-CONDENSING.
Compound Duplex Direct-Acting Pumping Engines; Steam Full-Stroke, High-
Pressure Cylinders; Counter-Pressure, 17 Pounds Absolute.
CLASS OF Engine.
A.
B.
C.
D.
INITIAL
PRESSURE
ABSOLUTE.
Constants.

0.72869
0.68252
0.69542
0.46587
65
30.364
27.364
28.202
13.281
70
34.008
30.777
31.679
15.611
75
37.652
34.189
35.156
17.946
80
41.295
37.602
38.633
20.269
85
44.938
41.015
42.III
22.599
90
48.582
44.427
45.588
24.925
95
52.225
47.840
49.065
27.159
100
55.869
51.253
52.542
29.587
105
59.512
54.665
56.019
31.978
IIO
63.156
57.078
59.496
34.246
115
66.799
61.490
62.973
36.577
120
70.443
64.903
66.450
38.907
125
74.086
68.316
69.927
41.236
130
77.729
71.728
73 404
43.563
135
81.373
75.141
76.881
45.892
Tangyes's Compound Pump.-The Floyd and Morton
direct-acting pumping engine, owned by Tangyes, Soho, Eng-
land, is illustrated in sectional elevation in Fig. 188, which is
a reproduction of the drawing attached to their American
patent, and is thus described by them:
"The high-pressure steam-cylinder and its appliances may
be of any ordinary construction suited for the application of this
invention, such, for instance, as that used in the pump known
as 'Tangyes's Special Pump,'-and we will describe, with ref-
erence to the accompanying drawing (which is a longitudinal
260
PUMPING MACHINERY.
FIG. 188.
LONDON
vertical section), the application of the invention to a pump
of that kind, from which its application to pumps of other
descriptions will also be understood.

A
D
A is the high-pressure cylinder and B is the low-pressure
cylinder. The slide-valve C of the low-pressure cylinder B is
not mechanically connected with, but is independent of, the
valve of the high-pressure cylinder, and is operated by a
COMPOUND DIRECT ACTING STEAM-PUMPS.
261
piston, D, working in an auxiliary steam-cylinder, E, prefer-
ably in line with the valve C of the said low-pressure cylinder
B, which may receive its steam from the high-pressure cylin-
der in the usual or any convenient way. The ends of the said
auxiliary steam-cylinder E are connected by means of pipes, F
and G, or equivalent steam-ways to the steam-ports or ends of
the high-pressure steam-cylinder A, as shown. The slide-valve
C in the steam-chest of the low-pressure steam-cylinder B
may be of the D or other type, and the said slide-valve is
fitted with piston-valves, HH, the pistons of which work in
short cylinders, JJ, at opposite ends of the steam-chest, and
the piston D is connected to one of the said piston-valves, H,
by the rod d.
"In the arrangement illustrated we use the piston-valves HH
in the steam-chest of the low-pressure cylinder B for control-
ling the motion of the piston D of the auxiliary cylinder E in
the following manner: Each end of the steam-chest where
the pistons H H of the slide-valve C work is provided with an
auxiliary port, K, which auxiliary ports connect the ends of
the steam-chest JJ in which the slide-valve pistons H H work
with the interior of the steam-chest in which the slide-valve
C works. The auxiliary ports K K described may be capable
of being opened and closed and regulated by means of screwed
plugs or valves, L L. By means of the said auxiliary ports
KK in the steam-chest the space between the back of each
of the piston-valves H H of the slide-valve C and the end of
the steam-chest JJ in which it works is charged with steam,
and when the piston D of the auxiliary cylinder E is moved
by the steam entering its cylinder by the pipes FG, or steam-
ways in connection with the high-pressure cylinder A, the
steam is compressed in one or the other of the spaces JJ at
the ends of the low-pressure steam-chest, so as thereby to
cushion the slide-valve piston and retard the motion of the
auxiliary piston D. By these means the main pistons in the
high- and low-pressure cylinders A and B are caused to make
a pause at the ends of their strokes, thereby permitting the
pump-valves to seat themselves quietly on their seats.
262
PUMPING MACHINER Y.
"We do not limit ourselves to this arrangement for control-
ling or retarding the action of the auxiliary piston D, as in
place thereof an oil-, water-, or air-cylinder may be used, the
displacement of the oil, water, or air in the said cylinder by
the motion of the piston effecting the retardation of the slide-
valve C, as hereinbefore described; or the retardation of the
motion of the low-pressure slide-valve C may be effected by
regulating the supply of steam to the auxiliary cylinder by
placing stop-valves, MM, in the pipes FG, by which the said
auxiliary cylinder E is connected with the steam-ports or ends
of the high-pressure steam-cylinder.
"In a compound pump or pumping engine with more than
two cylinders the low-pressure cylinder aforesaid would be-
come an intermediate cylinder having its valve actuated by
steam, as described, and the slide-valve of the third cylinder
would be connected with the ports or ends of the interme-
diate cylinder and be operated by steam, as described, with
respect to the low-pressure cylinder of a two-cylinder com-
pound pumping engine. In the same manner the invention
may be applied to a fourth cylinder, if desired.”
Compound Direct-Acting Engine with Isochronal
Valve-Gear.-The isochronal valve-gear by the Gordon
Steam-Pump Company, illustrated in Fig. 168, is further
shown in its adaptation to a compound engine in Fig. 189.
The working of the valve-gear in the compound engine is in
no respect different to that already described, excepting, of
course, in that due to its different application.
In this design the condenser and air-pump are placed in a
pit between the steam and water-ends. The necessary motion
for operating the air-pumps is obtained through a vibrating
beam receiving its motion from a suitable cross-head attached
to the main piston-rod; brackets are bolted to the under side
of the girders joining the steam- and water-ends, in which are
included the bearings for the vibrating-shaft. From this same
vibrating-shaft motion is given the sliding-cylinder by means
of connecting-rods, the centres of which are indicated by

FIG. 189.
263
264
PUMPING MACHINERY.
dotted lines; the movement of the sliding-cylinder and its
influence upon the auxiliary cylinder is the same as described
on page 218.
The illustration, Fig. 189, represents a single compound
engine, but this valve-gear permits the placing of two engines
side by side, and so arranged that they may work together as
a pair, or independently as single engines, if necessary.
The folding plate, Fig. 190, gives an illustration of a pair
of the same machines arranged to work independent of each
other as independent isochronal engines, and also provided
with duplex gear, by which they can, at will, be run as a com-
pound duplex engine. In this construction all of the ad-
vantages of the varying systems are combined, and while the
machine may be run as a duplex machine it is also capable of
running independently, so as to permit one side to do its work
while the other side is at rest or undergoing repairs.
Davidson Direct-Acting Triple-Expansion Pumping
Engine. This is a horizontal direct-acting steam-pump, with
the three steam-cylinders in line with the water-cylinder. Fig.
191 is an elevation and plan of the engine. The intermediate
cylinder is placed nearest the water-cylinder, and then come
the low-pressure and the high-pressure cylinders. Steam is
led by pipes from each cylinder to the next. The valves for
the high-pressure and the low-pressure cylinders are plain
slide-valves, worked by direct connections with the main valve
on the intermediate cylinder, which is similar in design to the
valve used by Mr. Davidson on compound and single-cylinder
pumps. (See Fig. 164)
With a view to saving space, the intermediate and the low-
pressure cylinders are set close together, the rear end of the
first forming the head of the second. This made it impossible
to use a central piston-rod between these cylinders, there being
no room for a stuffing-box. The low-pressure piston, there-
fore, has two piston-rods, which pass outside the intermediate
cylinder, but underneath its lagging, and are connected to a
cross-head fastened to the main piston-rod. The stuffing-boxes
10
ارف لري
用户
​11
FOL
#AL.
FIG. 190.

D
AI
141
in
งอม
НЯ
药
​101
ומו
กษ
MULTITEEN
انار
$15
17
PROBEREKKELISTELEMEN
10
13
1140
'nd
10 ----10
CLEGG, GOESER: COCINO
:
1
1
14

D
COMPOUND DIRECT-ACTING STEAM-PUMPS.
265
2
D
FIG. 191.
*

M
23
266
PUMPING MACHINERY.
for these rods are placed on the end of the intermediate cyl-
inder farthest from the low-pressure cylinder, and the rods work
through brass bushings, thus making it as easy to care for
these stuffing-boxes as for those in the ordinary position. To
facilitate starting the engine, a by-pass leads from the main
steam-pipe to the pipe connecting the high-pressure and inter-
mediate cylinders, by which both these cylinders can be worked
with live steam until the engine is under way.
A sectional elevation of the steam-end is shown in Fig. 192.
The valve-cylinder attached to cylinder A is 12 inches in
diameter, giving a large area for wear, and steam-ports of
ample size. The area of the valve-piston is 113 square inches,
so that the steam-pressure furnishes ample force to work the
valves of the three cylinders.
The low-pressure cylinder and steam-chest are shown at B.
The steam-ports are 14 by 14 inches. The high-pressure
cylinder, shown at C, has its left head cast in one piece with
the right head of the low-pressure cylinder, and the pistons
of the two cylinders are connected by a central piston-rod.
The steam-ports for this cylinder are ½ by 54 inches.
The steam-cylinders are 11½ inches, 20 inches, and 36
inches diameter, with a common stroke of 36 inches. They
are made of hard cast iron and jacketed with live steam, the
condensation being returned to the boiler by a small inde-
pendent pump. Steam-pistons are in one piece, with packing-
rings sprung in, and re-enforced by brass springs pressing
outwards. The cylinders are covered with approved non-
conductor, handsomely lagged with walnut strips, held in
place by finished brass bands.
The water-cylinder is of cast iron, 36 inches diameter by
36 inches stroke, with removable lining of hard cast brass
and brass valve-seats, guards, and springs. It is arranged for
suction connection on either side, as convenient, and has
ample openings for removal or examination of valves. The
capacity of the cylinder is such that at 45 strokes per minute
a delivery of 7000 gallons per minute is obtained.
The water-piston is packed with square fibrous packing for

FIG. 192.
267
268
PUMPING MACHINERY.
!
taking wear and with cupped leather rings for preventing
leakage.
The air-vessels are of cast iron, with brass water-gauges
and air-charging device whereby air is automatically main-
tained at proper height.
The piston-rods are of machinery steel, secured to the
pistons with brass nuts and jam-nuts and to the cross-head
by steel keys. The tie-rods for uniting the steam- and water-
cylinders are of iron, neatly turned and finished.
Oil-cups are placed at all points where lubrication is re-
quired, and a brass one-quart sight-feed lubricator is placed
on the high-pressure steam-cylinder.
With each engine is furnished a Davidson independent
condenser, with steam-cylinder 8 inches and air-cylinder 14
inches in diameter and a stroke of 16 inches. The air-cylin-
FIG. 193.
A
der is lined with brass, and has
brass valve-seats and guards and
sheet-metal valves. Condensers
have vacuum-breaking attachments
to insure against flooding the
steam - cylinder of main engines
and possible damage to them by
reason of inattention or negligence
on the part of those in charge.
It is of interest to compare this
pumping engine with other well-
known pumping engines. Of
course the first question which
arises is as to comparative duty or
fuel economy. In a power-plant
working under the conditions pres-
ent in most water-works pumping-
stations, the matter of fuel economy
is rightly considered of the first importance. A high-duty
pumping engine, very expensive in first cost, may be much
more economical in the long run than a cheap engine giving
only moderate fuel economy.

Cross-section through valve-cylinder.
;
COMPOUND DIRECT-ACTING STEAM-PUMPS. 269
So far no direct-acting steam-pump (with the exception of
the Worthington engine with the high-duty attachment) has
been able to approach in fuel economy the fly-wheel engines,
which store up the force exerted by the steam during the first
part of the stroke and give it out again on the last part of the
stroke when the steam is expanded and is pressing with less
force upon the piston.
As to the question how the pump illustrated may be ex-
pected to compare in duty with the best types of compound
duplex pumping engines, it is difficult to form any opinion
without a comparison of indicator diagrams. The duplex
compound can cut off the stroke and secure considerable
expansion of steam in the first cylinder, and has the great
advantage of securing a steady movement of the water at the
pump end by the mutual action of the two cylinders.
2
The triple-expansion engine here illustrated has the advan-
tage of carrying on the expansion in three cylinders. If we
assume that in this engine the valves of the three cylinders
open and close in exact unison, and each valve remains open
for the whole stroke, which we believe is practically correct,
then the number of expansions obtained is the ratio between
the areas of the high-pressure and the low-pressure pistons.
Since these are to each other as the square of their diameters,
or as 11½²: 362, the number of expansions will be 1296÷
132.25 10, nearly. While the highest duties recorded by
pumping engines have been obtained with greater ratios of
expansion than the above, many excellent records have been
made with about this number of expansions. The pumping
engine at Lawrence, Mass., designed by E. D. Leavitt, expands
the steam about 22 times, and the Pawtucket engine, designed
by Geo. H. Corliss, expands the steam about 18 times. The
ordinary duplex direct-acting engines have, as a rule, much
smaller ratios of expansions.
The position of the high-pressure cylinder in this engine
seems rather unfortunate for the attainment of a high duty, on
account of the long pipe necessary to carry the steam to the
intermediate cylinder. The clearance of the intermediate
23*
270
PUMPING MACHINERY.
cylinder is practically increased by the volume of this pipe.
This arrangement of the steam-cylinders was adopted to per-
mit the examination and packing or readjustment of each of
the three steam-pistons with the least delay and inconvenience.
The piston-speed when working at the contract capacity is
135 feet per minute. This again is much lower than that of
the fly-wheel engines. The Lawrence engine works at about
216 feet piston-speed, and the Pawtucket engine at about 250
feet per minute. The speed of 135 feet per minute, however,
is certainly high for direct-acting pumps, and we understand
that the engine is capable of as high speeds even as 200 feet,
or 67 strokes per minute. It is remarkable that smooth
motion and freedom from shock can be obtained from so large
a pump moving its own valves.
It is to be hoped that a thorough duty test may be made of
these engines after they are in place, that it may be known
what economy can be attained with engines of this class.
FIRE-PUMPS.
271
CHAPTER XIII.
FIRE-PUMPS.
THE Underwriter Pump is the name adopted by The Asso-
ciated Factory Mutual Insurance Companies to designate a
steam fire-pump built in strict accordance with certain specifi-
cations, the salient points of which are given below, taken
from a special circular prepared for the use of their agents
and patrons. The writer regards this circular as containing
the most intelligent and carefully-prepared specifications for
steam fire-pumps that have come to his knowledge.
INTRODUCTION.
Fire-pumps at the factories which we insure are so very
frequently found incapable of being started promptly when
tried for our inspectors, particularly at the spring inspection,
and are found in so very many cases to be incapable of deliv-
ering anywhere near their alleged or rated capacity without
violent "hammering," as to make it plainly evident that some
improvement in fire-pumps is greatly needed at the average
mill in one or all three of the following particulars:
The construction of the pump.
The erection and fitting up of the pump.
The care of the pump.
Steam fire-pumps are ordinarily sold, rated, and expected
to run at double the speed of a pump for boiler-feeding or
any ordinary water-supply; they do not ordinarily receive
such care or attention as a steam-engine or other machine on
which the product of the mill depends. Being constantly
subject to dampness, warmth, and contact with condensed
steam, their parts are especially subject to rust. After weeks.
272
PUMPING MACHINERY.
and months of disuse a fire-pump should be capable of being
instantly started and run at full speed by a man who is
excited and perhaps unskilful.
Therefore a fire-pump needs more strength, better work-
manship, and better protection against rust than does a
pump for boiler-feeding or ordinary miscellaneous use; and
the high speed at which it is expected to run demands the
large steam valve-ports and large suction-valve areas specified
below.
These specifications cover only certain general features of
the design, and may be viewed as an effort to indicate to those
whom we insure the kind and quality of pump which we
believe is needed.
Beyond these features which are specified, the various pump-
makers are free to follow each his own judgment and experi-
ence in details of shop practice, to the end that competition
in excellence may be encouraged. On the pump-maker rests
the responsibility for first-class workmanship, material, and
strength of castings, bolting, etc.
SPECIFICATIONS.
ARTICLE I. Only "duplex pumps" are acceptable for steam
fire-pumps.
(So-called "duplex" pumps, consisting of a pair of pumps with "steam-thrown
valves" actuated by supplemental pistons, are not acceptable.)
The experience of our inspectors goes to show that duplex pumps are more
certain of starting after long disuse. The whole power of the main cylinder is
available for moving a corroded valve or valve-rod, whereas on a single pump
with a steam-thrown valve no such surplus of power is available.
Further, the direct-acting duplex has the great advantage over a fly-wheel pump
of not suffering breakage if water gets into the steam-cylinder.
SIZE OF PUMP.
ART. 2. The present multiplicity of odd sizes is confusing,
and different makers estimate the capacity in gallons according
to different arbitrary standards.
FIRE-PUMPS.
273
Capacity to fur-
nish Standard
Fire-Streams.
(To simplify matters we recommend that only the four
different sizes given below be recognized or considered.)

Capacity in Gal-
lons per Min-
ute delivered
at Full Speed.
250
4
to
to
One stream.
320
I
Ratio of Piston
Areas.
(See Article 4 for commercial sizes.)
Too Small for a Fire-Pump
except as an auxiliary. Convenient for
regular use in extensive boiler-feeding
or domestic supply.
At speed suitable for continuous use delivery
would be but about half this, or about 150 gal-
lons per minute.
4
Two
streams.
500
to
I
3
Three
750
to
streams.
I
3
Four
streams.
1000
to
I
Ordinary Size for Small Mills.
This needs steam-boilers of at least 100
horse-power capacity to drive pump at rated
full speed and maintain 100 pounds water-
pressure. A steam-pressure of at least 40
pounds is needed at the pump to give this
water pressure of 100 pounds at full speed.
(The foregoing is for pumps in the best of
order.)
Ordinary Size for General Use.
This needs steam-boilers of at least 115
horse-power and a steam-pressure of at least
48 pounds square inch at the pump to drive at
full speed and at 100 pounds square inch water-
pressure.
Pressure at boiler must be a little more to
allow for loss of steam-pressure between boiler
and pump.
For cases where boiler-pressure is unusually
low (less than 50 pounds), use a pump with
larger (4 to 1) steam-cylinder.
Size for Large Factories.
Requires 150 horse power boiler capacity
and at least 44 pounds square inch steam-press-
ure.
(The boiler horse-powers above are reckoned
on the A. S. M E. basis of 30 pounds of water
evaporated or consumed per I. H. P. per hour,
and are for pumps in the best order, and with
parts as given in these specifications.)
(l'umps in poor order or too tightly
packed will require more steam than
stated above.)
Where more than 1000 gallons capacity is needed it is
generally best to provide two pumps.
S
}
274
PUMPING MACHINERY.
Two hundred and fifty gallons per minute is our standard allowance for a good
1%-inch (smooth-nozzle) fire-stream.
(From 15 to 20 automatic sprinklers may be reckoned as discharging about the
same quantity as a 1%-inch hose-stream under the ordinary practical conditions
as to pipes supplying sprinkler and hose systems respectively.)
It is not expected to provide pump capacity sufficient to supply all the sprink-
lers in a large room for operation all at once; it being assumed that if controllable
by sprinklers the fire will be controlled before spreading outside the area covered
by from 25 to 75 heads.
CAPACITY, HOW COMPUTED.
ART. 3. The capacity of a pump depends on the speed at
which it can be run.
To compare different pumps, some definite speed must be
agreed on as a basis.
We have adopted,-
For direct acting 10-inch stroke,--Full speed
steam-pumps of
For direct-acting} 12-inch stroke,—Full speed
steam-pumps of
=
75 revolutions per minute.
70 revolutions per minute.
The rated capacity is, therefore, to be computed as the
product of the speed just stated, by the plunger displacement
for one revolution, corrected for piston-rod volume and for
full nominal length of stroke, and with 10 per cent. deduction
for slip and loss of action.
(One revolution means one complete circuit of the motion of any of the recip-
rocating parts of the pump, and for a duplex pump is equivalent to four single
strokes. "Revolutions" is a term less liable to be misunderstood than "strokes.")
It is all right to run fire-pumps at the highest speed that is
possible without causing violent jar or hammering within the
cylinders.
Although a 12-inch stroke is 20 per cent. longer than a 10-inch stroke, the
maximum delivery of a 12-inch-stroke pump will not be so much as 20 per cent.
more than the delivery of a 10-inch-stroke pump of same diameter, for the reason
that, having to travel farther, it cannot make quite so many revolutions per minute
before jar will begin.
On the other hand, it will deliver somewhat more, perhaps 12 per cent. more,
than the 10-inch-stroke pump, because it can get equal delivery with a less number
of reversals of motion per minute.
FIRE-PUMPS.
275
7
i
ART. 4.
COMMERCIAL SIZES.
Nearest Commercial Size.
Number of
Standard
Ratio
Nominal
of
18-Inch
Fire-Streams.
(Each
Areas
Steam
to
Capacity.
Gallons
10-Inch Stroke.
Rating.
per
Diam.
Diam. Nominal
Gallons per
Minute.
Minute
Steam-
Water-
250 Gillons
Water-
Full
Cylinder.
per Minute.)
Piston.
Speed.
Inches..
Inches.
Cylinder. of Stroke.
Inches.
Length (Including 10 per
cent. deduction
for slip, etc.)
to be seldom de-
These two pumps being so
manded for fire-pumps, mak-
ers are hardly expected to pre-
pare special patterns for them
small as
at present.
One
stream.
Too small
for
a
fire-
pump alone.
4 to I
320
Two streams.
4 to I
500
Three streams.
3 to I
750
Four streams.
3 to I
1000

10 X 5 X 10
Exact
Capacity
Underwriters'
Nearest Commercial Size.
12-Inch Stroke.
Diam. Diam.
Steam-
Water-
Cylinder. Cylinder.
Inches. Inches.
Nominal
Length
of Stroke.
Inches.
Exact
Capacity
Underwriters'
Rating.
(Including 10
per cent.
deduction for
slip, etc.)
220
10 X 5 X 12
246
As between these two pumps, a somewhat greater price for the second is reasonable on account of its 12 per
cent. greater capacity.
12 X 6 X 10
321
12 X 6 X 12
359
As between these two pumps just given above, the second rates at 12 per cent. greater capacity than the
first, and a price 12 per cent. greater would therefore not be unreasonable.
16 X 8 X 10
569
14 X 7 X 12
As between these two pumps just given above, the second rates at 15 per cent. less capacity than the first,
therefore the second should be sold about so much cheaper.
16 X 94 X 10
767
16 X 9 X 12
484
807
As between these two pumps just given above, the second rates at 5 per cent. greater capacity than the first.
18½ X 10½ X 10
989
18 X 10 X 12
1007
As between these two pumps last mentioned, the second rates at 2 per cent. greater than the first.
276
PUMPING MACHINERY.
No pump with areas less than 3 to I will, except under special circumstances,
be accepted as a fire pump.
It has been common to make all fire-pumps with water-plunger of only one-
fourth the area of steam-piston, with the idea that the pump could thereby be
more readily run at night, when steam was low.
The capacity in gallons is thus reduced 25 per cent. as compared with a 3 to I
plunger on the same steam-cylinders.
Often, especially with large pumps, this is a mistake, for though the pump might
start and give a few puffs, it withdraws so much steam that it can run effectively
for but an extremely short time without the boiler fires are first aroused to make
fresh steam to replace that withdrawn.
A steam-piston relatively larger than necessary is a source of weakness. It takes
more volume of steam, and gives more power with which to burst something if the
throttle is opened wide suddenly during excitement.
ART. 5. The inside-plunger pattern of pump is preferred to
the water-piston pattern for all situations where water is rea-
sonably free from grit or mud, and will generally be best for
all New England situations.
For pumps in the West or South, or for water like Ohio
river water, a water-piston with packing or a packed plunger
gives much better wear.
By means of the priming-tank and pipes hereinafter speci-
fied, the plunger pattern is enabled to "get hold of its water"
as quickly as the piston pattern even under extreme lifts, and
our experience goes to show that the plunger pattern is much
better adapted to start free after long disuse or unskilful treat-
ment.
SIZE-PLATE.
ART. 6. Every steam fire-pump must bear a conspicuous
statement of its capacity, thus:

16 X 9 X 10.
Capacity 750 gallons per minute, or three
good 18-inch smooth nozzle streams.
Full speed, 75 revolutions per minute.
For fire purposes never let steam get below
50 pounds, nights or Sundays.
This plate to be of porcelain-like enamel, bearing black
FIRE-PUMPS.
277
letters ½ inch high on a white ground. The letters being of
enamel and burned in.
The plate is to be securely attached to the inboard side of
the air-chamber.
BRASS FITTING.
ART. 7. Fire-pumps are to be brass fitted throughout,―viz.,
both piston-rods for their whole length are to be preferably
of solid rods of special bronze, equal in quality to Tobin
bronze, but rods of steel covered with a brass shell % inch
thick throughout both water- and steam-ends are acceptable.
(Common foundry bronze composition is not dense or
strong enough for this purpose.)
Valve-rods are to be preferably of solid Tobin bronze, or
may be brass covered through their bearings in stuffing-boxes.
Water-plungers are to be of solid brass or bronze, and the
ring in which they slide is also to be of brass or bronze. The
composition of the plunger and its ring should be of very
dissimilar alloys to insure good wearing qualities.
All six stuffing-boxes are to be bushed at the bottom with
a brass ring with suitable neck and flange, and the follower or
gland is to be either of solid brass, or is to be lined with a
brass shell 3-16 inch thick, united with a flange covering the
end next the packing.
STRENGTH OF PARTS.
ART. 8. The maker is to be understood to warrant each
pump built under these specifications, to be at time of delivery,
in all its parts, strong enough to admit of closing all valves on
water outlet pipes while steam-valve is wide open and steam-
pressure 80 pounds, and agree to test it before shipment
from his works.
(In other words, although these pumps are not expected to be designed for a
regular working water-pressure of 240 or 320 pounds, it is expected that bolts,
shells, rods, etc., will be figured to stand this comparatively quiet, temporary, high
pressure exclusive of further allowance for initial strain due setting up of bolts,
with a factor of safety of at least four.)
24
278
PUMPING MACHINERY.
(This high test pressure is analogous to the custom of proving all common cast-
iron water-pipes to 300 pounds and all common lap-welded steam-pipe to 500
pounds per square inch, even though it is to be regularly used at much less
pressure.)
We are assured that castings no heavier than at present used by the best makers
will stand this test, if properly shaped and liberally bolted.
ART. 9. The pump is to be warranted so designed, with
such arrangement of thickness of metal that it shall be safe to
instantly turn a full head of steam on to a cold pump without
cracking or breaking the same by unequal expansion, due to
this cause.
SUCTION-VALVES.
ART. 10. The net areas of ports in suction-valves for each
end of each plunger is to be not less than 50 per cent. of area
of plunger, where normal stroke is 10 inches, and for pumps.
of 12-inch stroke, valve area is to be 56 per cent. of plunger
area.
(Experience indicates that it is preferable to secure this area by a larger num-
ber of 3-, 3½-, or 4-inch valves rather than by increasing diameter of valve.
The valves probably rarely lift so high in operation that area of cylinder under
their edge half equals the area of ports; hence, strictly speaking, it is probably a
question of securing the greatest convenient development of total valve circum-
ference rather than merely large port area.)
FORCE-VALVES.
ART. II. The force-valves need have but two-thirds the
total port area specified for the suction-valves.
SPRINGS.
ART. 12. All valve-springs to be coiled on a cylindrical
arbor (conical springs are not approved). The spring to be
properly guided and held from drifting sidewise out of centre
by ends resting in grooves, and a suitable metal plate to be
interposed between the rubber valve and the spring.
Valve-spring and valve-spindle or stud are to permit an
extreme lift equal to one-fifth diameter of rubber valve, and
are to be so formed by screwing in on a taper or otherwise,
that this amount of free lift cannot be restricted.
FIRE PUMPS.
279
Stiffness of springs is for the present not specified, but left to the discretion of
each maker. Preferably one rather stiff standard spring had best be uniformly
used for all force-valves, while for the suction-valves three grades of stiffness can
be used with advantage.
Thus the standard will serve for all lifts up to 18 feet.
An extra mild spring can be used for higher lifts, and an extra stiff spring where
pump takes its suction under 5 pounds or more pressure may aid pump in run-
ning smoothly at high speed.
VALVE-SEATS.
ART. 13. Valve-seats to be all of U. S. gun-metal compo-
sition and firmly secured into valve-deck by forcing in on a
screw-threaded taper, or by forcing in on a smooth taper and
expanding out its lower edge below the valve-deck.
STICKING OF VALVES.
ART. 14. A serious objectionable feature of fire-pump valves, and one for which
no satisfactory remedy has been put in practice as yet, is that if the pump is left
standing or unused for several weeks or months the rubber valve-disks are some-
times liable to become stuck to their seats, and if suction has a high lift there may
not be vacuum enough to tear all the suction-valves open.
At present we can only repeatedly urge that steam fire-pumps be started enough
to limber them up, at least once a fortnight.
Perhaps some non-corrosive alloy of nickel or aluminum can be found which
rubber vulcanized with sulphur would not corrode and adhere to, and experience
in other arts indicates that rubber will not adhere to a block-tin face even with
extreme long standing. This is a subject well worth investigation by pump-
makers.
STEAM-PORTS.
ART. 15. The minimum area of each exhaust-steam passage,
at its smallest section, is to be not less than 4 per cent. of the
area of the piston from which it leads.
This is a large increase over the size heretofore common, but indicator-cards
which we have taken from pumps of several different makes indicate this to be
one of the points in which improvement is most needed to accommodate the high
speeds at which fire-pumps are always supposed to run, and this unrestricted
exhaust aids very materially in giving steadiness to the jet of water.
The admission-ports may be not less than 2½ per cent. of
piston area.
Although from structural reasons the steam-port has usually heretofore been
made of same size as exhaust-port, it can perhaps be advantageously made shorter.
•
•
譬
​280
PUMPING MACHINERY.
STEAM-CLEARANCE SPACE.
The clearance space between face of piston and cylinder-head must be reduced
to smallest possible amount, and these contacting surfaces be flat, without projec-
tions or recesses, excepting the one for the piston-rod nut
Some makers, with the idea that a fire-pump need not be economical, have not
taken pains to keep these waste spaces small.
Securing small clearance costs almost nothing but care in design, and is often
of value, since at many factories boiler capacity is scanty for the large quantity of
steam taken by a fire-pump of proper size.
The edges of the steam valve-ports and likewise of the
valve are to be accurately milled, or chipped and exactly filed
to templets, true to line, and the face of valve and its seat
accurately scraped to a plane surface, all in a most thorough
and workmanlike manner and equal to high-grade steam-
engine work.
STEAM SLIDE-VALVE ADJUSTMENTS.
(Adjustable valves are not recommended.)
It is recognized that the practice of making adjustable
valve-tappets located outside the steam-chest is a good thing
on a large pump in constant service and operated by a skilled
engine-driver, but for the infrequently used ordinary fire-
pump the utmost simplicity is desirable, and it is best not to
tempt the ordinary man in charge to "readjust" the valve-
gear.
Since the number of different sizes of pumps called for in these specifications
are few, it is recommended that gauges and templets be prepared by which the
valve position can be set once for all without necessity or ready means for re-
adjustments.
CUSHION-VALVES.
ART. 16. Cushion-valves regulating the amount of steam-
cushion at ends of stroke, by controlling a passage commu-
nicating between steam- and exhaust-ports, are recommended
for the 750-gallon and 1000-gallon pumps. They may, if the
maker thinks it desirable for simplicity, be omitted on the
pumps of 500 gallons per minute or less capacity.
In order to obviate the possibility of the pump pounding itself to pieces in case
of a sudden release of load, as by a break in suction- or delivery-mains, we recom-
FIRE-PUMPS.
281
mend that the steam-cushion release be through independent ports terminating ½
inch back from cylinder-head, instead of through the induction-port opening, as
at present common.
This makes the pump safer in case cushion-valves are unskilfully left open too
wide.
STROKE GAUGE.
ART. 17. A length-of-stroke index or gauge will be insisted
on in all cases. These are to be of satisfactory strong and
simple form for at all times rendering obvious the exact length
of stroke which both pistons are making, and thus calling
attention to improper adjustment of cushion-valves or stuffing-
boxes.
PIPE SIZES.
ART. 18. Suction- and discharge-pipe connections must
have standard flanges to connect with pipes of the sizes given
below.
Size of Pump. Diameter of Suction-
Gallons per
Min-
Pipe. Inches.
ute.
Diameter of Dis-
charge-Pipe.
Inches.
320
500
750
1000
8
56 7∞
68
1Ο
12
Steam-Pipe. Exhaust Pipe.
21/2
3
3445
3/2
4
5
These suction-pipe sizes, although larger than heretofore common, are believed
to be amply justified by our experience, and exert a powerful influence towards
enabling the pump to run smoothly at high speed with water-cylinders filling per-
fectly at each stroke. No defect is more common than restricted suction-pipes.
There should preferably be three suction entrances,—viz.,
one each side and one at end of pump. One of these
openings is designed for the attachment of the suction air-
chamber, and the providing of two others is often a great
convenience in adapting pump to any particular location, or
in permitting drafting from two different sources of supply.
A single central end suction-opening is, however, acceptable.
VACUUM-CHAMBER.
ART. 19. The suction air-chamber (often called a vacuum-
chamber), or its equivalent mentioned below, is to regularly
form a part of every pump made under these specifications.
24*
282
PUMPING MACHINERY.
This may, however, be omitted under special agreement for cases where suction
is from an open reservoir by a pipe of full specified size less than twenty feet in
total length, but flanges must be arranged so it can be readily added afterwards if
found necessary.
If so desired, the vacuum-chamber may consist of a closed vertical pipe of
same diameter as suction, and 6 feet long.
This vacuum-chamber must be attached to the pump in the most direct way
practicable, but provision must be made for attaching it in such manner as not to
prevent readily taking off the cylinder-heads.
For pumps taking feed under a head, a vertical open pipe of same diameter as
suction, reaching up to five feet above level of feed-water, is much preferable to a
closed suction-chamber, since the latter is liable to become filled with water.
If
a great head or other cause makes this impracticable, the vacuum chamber should
be so connected and fitted with gates that it may readily have water emptied and
be filled with air while pump is in motion.
AIR-CHAMBER.
A large air-chamber is more necessary on a fire pump than on other pumps;
without it the hose vibrates so that holes are quickly worn through.
An air-chamber of hammered copper, and warranted tested
under a hydraulic pressure of not less than 300 pounds per
square inch, is preferable to cast iron, as, holding the air better
and being lighter, it wrenches and strains the pump less when
running fast and shaking.
The air-chamber will be acceptable, however, if made of
cast iron; but then it is to be warranted subjected to a hy-
draulic test of 400 pounds per square inch before connection
to pump, and is to be thoroughly painted inside and out to
diminish its porosity.
The form of air-chamber combined with the elbow beneath it, and also the
relief-valve and hose-valve connections, should all be carefully designed to make
the whole height the very least possible. Keeping this weight low makes the pump
run steadier and brings less strain on the flanges at high speeds.
SIZE OF VACUUM- AND AIR-CHAMBERS.
Vacuum-Chamber is to con-
Air-Chamber is to con-
tain:
tain:
"6
320-gallon pump.
500-
750-
1000-
8 gallons.
13
18
"
10 gallons.
17
25
(6
24
("
30
(6
+
FIRE-PUMPS.
283
These volumes are based on making suction-chamber for long pipes six times,
and pressure-chamber eight times one displacement of one plunger.
Where the suction will never be from a pipe under pressure, and where the suc-
tion-pipe has a total length of less than 25 feet, vacuum-chambers of from half to
two-thirds the above will be acceptable.
PRESSURE-GAUGES.
ART. 20. A water-pressure gauge provided with 4-inch
cock with lever handle is to be provided with the pump, and
connected close to the air-chamber.
The gauge itself is to be of the 5-inch iron case duplex spring pattern (such as
used on locomotives), this kind of gauge being believed to be the best for with-
standing the vibrations which cause gauges on fire-pumps so often to be unreliable.
A steam pressure-gauge of the same kind is to be attached
to the steam-chest inside the throttle-valve.
SAFETY-VALVE.
ART. 21. A safety- or relief-valve of the Ashton, Crosby, or
other similar and accepted pattern is to be regularly included
in the price, and is to be attached to each pump; preferably
extending horizontally inboard from base of air-chamber, so
that its hand-wheel for regulating pressure is within easy reach.
This valve is to be set ordinarily at a working pressure of
100 pounds to the square inch, and is to be of such capacity
that it can discharge the full throw of the pump at 3 speed,
at a pressure not exceeding 125 pounds per square inch. It
is to be provided with a hand-wheel, marked very conspicuously
to show direction of turning, thus :-OPEN →→→
It should not have a locked adjustment, and does not need a device for lifting
valve from seat by hand.
Until further experiment it will be assumed that—
For 320-gallon pump a 2½-inch Ashton valve (or equivalent) is suitable.
500-
(6
750-
"C
(6
3-
3½-66
(C
ff
"
1000-
(6
(6
4-
(6
66
CC
66
(C
This should discharge by a vertical, downward pipe, about
2½ feet long, opening into a cone or tunnel, fitted to a 6-, 8-,
or 10-inch waste-pipe.
284
PUMPING MACHINERY.
This very short open part at the cone being for rendering
it obvious to pump operator whether water is wasting through
relief-valve.
This cone and its pipes are not included in price of pump.
DRIP-COCKS.
ART. 22. Three eighths-inch brass drip-cocks with a lever-
handle are to be provided on both ends of each water-cylinder
and each steam-cylinder, for effectually draining the same. A
similar drain-cock is also to be provided for drawing off the
water above the upper valve-deck.
A 1/4-inch air-cock with lever-handle is to be attached to
the cover over water-cylinders.
(Cocks with lever-handle are used on account of showing to every passer-by
whether they are open or shut.)
PRIMING-PIPES.
ART. 23. Each pump is to be regularly fitted with 1-inch
brass priming-pipes before leaving the shop, as follows (these
being included in its price), beginning at a 2 x I x 1-inch brass
tee close to pump beneath delivery-flange, and thence extend-
ing to four 3/4-inch valves, one of which leads into each of the
four plunger-chambers. These priming-pipes should not con-
nect into suction-pipe, lest priming-water be thereby wasted,
and should not connect with chamber above force-valves.
For cases where pump only takes its suction under a head, these priming-pipes
may be omitted, but openings for them into the pump-shell must be provided and
fitted with screw-plugs.
A priming-tank (not included in pump contract unless at an
extra charge) will in all cases be a requirement in the instruc-
tions for properly setting up a fire-pump to which the water
does not flow under a head. This tank must be situated with
its bottom not lower than 5 feet above the pump, and con-
taining a volume of water equal to half the nominal capacity
of the pump in gallons per minute, said priming-tank to be
for the sole and exclusive use of the fire-pump, and connected
with nothing else whatever.
FIRE-PUMPS.
285
Having this provision for priming quickly and surely, a
foot-valve at bottom of suction-pipe is not needed.
ART. 24. Each pump is to be regularly fitted, not as an
extra, but as a part included in the cost of the pump, with
hose connections, consisting of Chapman Valve Company's
2½-inch Straight-way Hose Valves (not including cap and
chain), attached to the neck of air-chamber. The number of
these is to be as follows:
For the 300-gallon pump
500-
750-
1000-
(C
One.
Two.
Three.
Four.
The screw-head at end of these valves for connecting to hose is to be either
fitted to a hose-coupling furnished by the customer, or left with the thread uncut.
ART. 25. Tests for acceptance after pump is set up at its
destination:
1. Pump being set up in good order and properly packed,
is to run smoothly without slamming or jumping or hammer-
ing, at its full rated speed (of 70 or 75 revolutions per minute),
and make full normal stroke, meanwhile maintaining a water-
pressure of 100 pounds per square inch while furnished with
500 gallons (4 to 1).
45 pounds square inch steam-pressure for the
50
45
(C
(C
66
(C
66
750
<<
1000
(3 to 1).
(3 to 1).
Water-pressure being measured at base of air-chamber.
Steam-pressure being measured close to steam-chest.
Start slow. Gradually and alternately open throttle to give speed required,
and if hose is short or discharge too free, close or adjust outlet-valve to give back
pressure enough to raise water-gauge to about 100 pounds.
Pump is to run at full speed free from thumping without necessity for "snifting"
air into suction.
2. During this trial it is preferable to discharge the water
through two, three, or four 1%-inch smooth nozzles (accord-
ing to the number the pump is rated to supply, as per Art. 4).
Each nozzle being connected to the hose outlet on the pump
by 150 feet of 2½-inch rubber-lined hose, and meanwhile
note the pressure of water.
286
PUMPING MACHINERY.
The 100 pounds water-pressure is intended to allow for friction in hose lines
150 feet long. If shorter lines are used, the same nozzle delivery will be
obtained with less pressure at the pump.
3. While thus playing through the nozzles the hose near
the pump should lie quiet, or with but little sliding to and fro
on the ground near the pump. One of the nicest points
about building a pump is to so design its valves and gear
that the pulsations of the pump will not quickly wear holes
in the hose by causing it to rub back and forth at each beat
of the piston, and the quietness with which the hose lies is a
good index of the pump-maker's skill in securing uniform
delivery.
4. An experiment may then be made to determine the
greatest speed at which pump can be run, at 100 pounds
water-pressure, before beginning to jar or pound seriously,
cushion-valves meanwhile being adjusted.
This may be tried by adding another line of hose at some.
hydrant near, and seeing how well the pump will fill this
extra nozzle in addition to those it is rated to deliver.
Count the revolutions, and note the steam- and water-press-
ures meanwhile.
5. With all water outlets closed and with sufficient steam
admitted to give 80 pounds average water-pressure, the internal
leakage is to be so small that pump will not make more than
one revolution per minute.
The safety-valve should be on during this test, lest pressure be accidentally
carried too high.
6. With all water outlets nearly, but not completely, closed
and safety-valve screwed down and steam admitted sufficient
to give a water-pressure of 240 pounds per square inch, pump
moving very slowly meanwhile, all joints about the pump are
to remain substantially tight.
7. With all other outlets closed and the safety-valve ad-
justed to 100 pounds, then on starting the pump this safety-
valve should be sufficient to discharge the full delivery of the
pump at fifty revolutions per minute, with a pressure of not
over 125 pounds.
FIRE-PUMPS.
287
8. A brief experiment may then be made, with cushion-
valves wide open, to determine greatest speed at which pump
can be run in this condition without injurious hammering.
FINALLY, COST.
ART. 26. A pump built as specified above will cost more
than a fire-pump of same size built as heretofore, but we be-
lieve it well worth the extra cost; and the tests on the few
pumps already made indicate that the first cost per gallon
actual capacity is, by reason of smooth running and higher
speed, practicable, even less than for the ordinary style.
The main points of difference are:
Pump has brass plungers instead of cast-iron ones.
Pump has bronze piston-rods and valve-rods instead of iron.
Pump has brass-lined stuffing-boxes instead of iron.
Area of water-valves 25 to 50 per cent. greater.
Steam- and exhaust-passages 20 to 50 per cent. greater.
Suction-pipe connections 2 to 4 inches greater diameter.
Cushion-valves better arranged.
Air-chamber is made much larger.
Shells and bolting are warranted especially strong.
The following necessary fittings, heretofore charged for as
extras, are included in the price, and regularly furnished as a
part of this pump,—viz. :
A capacity-plate is added.
A stroke-gauge is added.
A vacuum-chamber is added.
Two best quality pressure-gauges.
A water-relief valve of large capacity.
A set of brass priming-pipes and valves.
From two to four Chapman hose-valves.
A sight-feed cylinder-lubricator is added.
Respectfully submitted,
JOHN R. FREEMAN,
Engineer Assoc. Mutual Ins. Co.
June 1, 1891.
288
PUMPING MACHINERY.
The following vote was passed at the monthly conference
of the Associated Companies, June 15, 1891:
"Voted: To recommend that the so-called 'Underwriter's Pattern' of steam
fire-pump, complete with all attachments as per specification of June 1, be recog-
nized as the approved type, and that pumps built and fitted up less perfectly than
per this specification be not approved in future installations, except under special
circumstances and by special agreement with some executive officer of the in-
surance companies."
;
•
MINING-PUMPS.
289
CHAPTER XIV.
MINING-PUMPS.
Pumping engines for mines are generally placed under-
ground; there are instances in this country in which the
engines are located on the surface, but by far the larger num-
ber are down in the mines, and usually making the delivery
in a single lift. The service required of a mine-pump is con-
stant, and as pumps are not always duplicated, it is a matter
of first importance that they be properly proportioned as to
strength, and furnished with a valve mechanism not liable to
get out of order.
There was formerly no uniformity in the type of pump used
in mines; piston-pumps and several varieties of packed
plunger-pumps were employed, but, all things considered, the
use of plunger-pumps is preferable to that of piston-pumps,
especially in deep mining operations. The plunger-pump is
almost exclusively used at this time, for in the old plan of
working under a heavy pressure with a piston, if dirty water
were met with, the pump is less likely to give satisfaction; the
cylinder cuts out quickly, and if a lining is employed, it is a
costly operation in a large pump to put in another.
The metal used for mine-pumps is almost always hard
cast iron; occasionally, however, a gun-metal water-end is
used, but such examples are rare, especially for large pumps.
It is customary to give water-end castings for mine-pumps an
extra thickness of metal apart from that necessary for work-
ing strength, including any jar or water hammer, the pump
castings under pressure having from half an inch to an inch
and a half added to the thickness for deterioration. The metal
•
N
t
25
290
PUMPING MACHINERY.
itself is made very hard; in fact, as hard as can be worked
with special steel tools; ordinary machine castings have been
found to be entirely too soft to long withstand the action of
bad water. A water-end of a pump was recently shown the
writer, in which, by a sudden change in the character of the
water in a mine from sweet to bad, the pump was utterly
ruined in less than a month's service. The pump was brass
fitted, the castings soft gray iron; the action of the acid water
was to loosen the valve-seats and attack every interior por-
tion of the pump which came in contact with the brass, as well
as corroding such portions as had been machine fitted.
The action of mine water is very capricious, and does not
affect all portions of a pump alike; that is to say, all parts are
subject to the destructive action of the acid water, but certain
localized points in the interior are often found to deteriorate
much more rapidly than others. For such portions of a pump
special provision may be made in advance by having duplicate
parts, if the water-end be made up of sections, easily attached
whenever needed.
Sinking-pumps are usually vertical direct-acting single
pumps with steam-thrown valve-gear. The water-ends are
of either the piston, differential plunger, or fitted with double-
acting plungers centrally packed; the latter, being a better
type of pump for handling gritty water, is more favorably
thought of by mine-operators than either of the former.
Such pumps are fitted with an eye-bolt in the upper cylinder-
head, or with three eye-bolts attached at points of support,
which will enable the pump to hang vertically when sus-
pended by a chain. In addition to this, wrought-iron dogs
or clamps are provided by which the pump may be fastened
to suitable timbers on the side of a shaft; the pump may,
therefore, be either held in suspension from above, and raised
or lowered as may be required in the case of a flooded mine,
or it may be temporarily fixed to the side of the shaft, to be
lowered at such intervals as the progress of sinking a shaft
may require.
MINING-PUMPS.
291
A sinking-pump must be simple, compact, strongly made,
and the valve-gear well protected. Probably no other type of
pump is subjected to the rough usage which sinking-pumps
invariably get. It is a condition which cannot be overcome
in the preliminary stages of mining, on account of the lia-
bility of accident from blasting operations, the necessity at
all times for pumping gritty and often acid water, and the
fact that there is always more or less muddy water trick-
ling down the shaft from the water-bearing strata overhead,
which so completely covers a pump with mud that if it were
not in motion it could scarcely be distinguished from any
other mass of débris. Notwithstanding all this, a sinking-
pump must work continuously night and day, and often up to
its limit of capacity. The failure of such a pump at a critical
period, even for a day, may flood a shaft, which would require
a week or more to recover.
The Cameron sinking-pump, by reason of its having
no external valve-gear, has met the above conditions, and as a
result it is very popular with mine-operators. The steam-end
of this pump has been previously described and illustrated in
the sectional engraving, Fig. 160, which shows the mechanism
by which the steam-piston is operated for horizontal mine-
pumps; for vertical sinking-pumps the valve-chest is placed
at right angles to the stroke, the port-openings being arranged
to correspond, a detail easily understood, and does not require
special illustration. A sectional elevation of a double-plunger
pump, showing the arrangement of stuffing boxes, suction-
and delivery-valves, and suction-pipe are clearly shown in
Fig. 194.
A vertical sinking-pump by the Deane Steam-Pump Com-
pany is shown in sectional elevation in Figs. 195 and 196. This
design is intended to meet the demands for such a machine in
the mining districts of Colorado. These machines are for sink-
ing or unwatering mines or for any use where a varying water-
level must be accommodated. The plungers are in one piece,
are double-acting and centre-packed; they can be very easily
292
PUMPING MACHINERY.
and quickly packed by the removal of only two glands. The
valves are of rubber, of a special texture for heavy duty and
to resist the action of bad mine water, with brass covers.
FIG. 194.

#
酸​白
​Ο
Hinged bolts are used on the hand-hole plates; stuffing-box
glands and all removable parts and all nuts throughout the
machine are secured by split pins, so that no parts may be
lost while the machine is being packed or adjusted. The
MINING-PUMPS.
293
machine is fully equipped with drip- and sand-valves. The
suction is taken in at the bottom of the pump, thus saving a
FIG. 195.
length of pipe required by most
machines of this class. The

FIG. 196.

valve motion is the well-known Deane patent. It is absolutely
positive, and will start at any point of its stroke, and is thor-
25*
294
PUMPING MACHINERY.
oughly protected by a cast-iron shield, which also serves as
the yoke between the steam- and water-ends. The steam-
cylinder is equipped with cushion-valves to regulate the up
and down strokes of the engine. Each machine is made with
O
BANG
FIG. 197.
(WORTHINGTON)
1
גן
段
​a bale attached to the
steam-cylinder, and
also with shoes and
dogs to secure easy
handling and control
while in the shaft.

A duplex sink-
ing-pump by Wor-
thington is shown in
Fig. 197. It is not
often that duplex
pumps are arranged
to work vertically,
but when properly
designed for such
a position they will
be quite as satisfac-
tory in their opera-
tion as if employed
horizontally.
A
close inspection of
the engraving of this
sinking-pump shows
that in the several
details of the water-
end the best practice
in mining - pump
-
design has been car-
ried out. The water-
plungers are double-
acting, working through exterior stuffing-boxes and adjust-
able packing. Means are provided, as shown in the engraving,
MINING-PUMPS.
295
for either suspending the machine at the eye-bolt shown be-
tween the steam-cylinders, or for hanging it on suitable timbers
on the sides of the shaft. The suction opening is at the lower
end of the pump, convenient for attaching the suction-pipe or
hose. The discharge connection to column-pipe is on the
side. The water-valves are enclosed in heavy pot-chambers,
and are made accessible for examination or repairs by means
of swing-bolt covers on the valve-pots. The duplex valve-
movement being positive, the pump is always ready to start,
and when running there is an entire absence of concussive
action; it is, therefore, well fitted to withstand the rough usage
to which sinking-pumps are subjected.
Single direct-acting mine-pumps occupy less room
in a mine than a duplex pump of the same capacity, but the
delivery from a single pump is at times so faulty that a very
large percentage of the steam used is expended in starting
anew the column of water at each stroke. This fault is due,
in part, to the construction of the pump, by which an interval
occurs at the end of each stroke, during which no water is
being delivered, the flow is suspended for the time, and thus
permits the water column to come to a state of rest. But the
fault does not lie wholly with the pump as such, but to neglect
in the failure to keep the air-chamber of the water-end prop-
erly charged with air. If this latter detail were properly
looked after there would be less trouble with pumps and valves
than has heretofore been the case.
The capacity of an air-chamber in a single, direct-acting
mine-pump should never be less than four volumes of the
pump capacity per stroke. This air-chamber should connect
directly with the highest point of the delivery-chamber; it
should be furnished with a glass water-gauge so that the quan-
tity of air in the vessel can be determined at a glance. The
height of the water in the air-chamber should never exceed
a level due to one-third of its cubic content. A device similar
to that shown in Fig. 91 can be applied to any pump, and the
deeper the mine the greater the necessity for its application.
296
PUMPING MACHINERY.
Lubricating Pistons of Mining-Pumps.-The sectional
elevation, Fig. 198, represents one of the Knowles piston
mining-pumps, fitted with Simpson's patent lubricating water-
piston. This device consists of an arrangement whereby a
lubricant composed of pitch and tallow is sent from the inside
of the pump-piston through its rings to the surface of the
pump-cylinder, thus lubricating the water-
piston and cylinder in a satisfactory manner.
The lubricant not only insures the freest
possible movement of the piston, but soon
coats the entire inside of the pump and the
inner surface of the pump-column; thus
acting as a preventive against the destruc-
tive effects of bad mine water. The nature
of the compound of pitch and tallow secures
a tight piston.

FIG. 198.
B
The action of this device is as follows: That part of the
piston-rod which enters the water-end of the pump has
through its centre a long passage-way, B, which connects the
small hand-pump A holding the lubricant with the holes Cin
the pump-piston, as shown in the engraving. The holes Care
drilled in the water-piston at several points, ending under the
packing-rings D. The packing-rings are made of composition
copper and tin, and are perforated with small holes to permit
the compound of pitch and tallow to ooze out and lubricate the
cylinder and piston. The lubricant is by this little hand-pump
forced through the hollow piston-rod and the holes in the
MINING-PUMPS.
297
water-piston, thus filling them and setting up the rings in the
water-piston. When the lubricant is pumped up solid, cock
E is closed to retain the pressure. In the action of the pump
a small amount of pitch and tallow oozes out through the
holes of the packing-rings, lubricating the cylinder and pro-
ducing the benefits above named. It is only necessary to
charge the pump with this lubricant about every eight hours.
Duplex mine-pumps are steadily growing in favor, be-
cause the duplex valve-movement is particularly well suited
to handling water at high pressures. The advantages of this
valve-movement have already been set forth in the chapter on
hydraulic-pressure pumps, and need not, therefore, be repeated
here.
Pumps of this type are always placed at or near the bottom
of the mine, and no difficulty is had in making the delivery
in a single lift, wholly irrespective of the depth of the mine.
Air-chambers are not usually supplied duplex mine-pumps;
effective and quiet working having always accompanied the
practical operation of pumps of this type, it was believed that
the addition of an air-vessel was unnecessary. This is true
in part, but the duplex action is not wholly continuous, or
perhaps a better way to put it is that the pressure in the de-
livery-pipe is not absolutely uniform, but it is more nearly so
than in any other type of pump now employed in mining
operations. The addition of an air-chamber, together with
an automatic charging device, as shown in Fig. 102, will fully
meet all the conditions of this exacting service.
The superiority of the duplex movement was strikingly ex-
hibited in the development of the pipe-line system of the
National Transit Company, the service being analogous to
mine-pumping except as to material. When single direct-
acting engines were used the jarring effect produced by their
irregular delivery was very destructive on the pipes and
fittings, causing leakage, and was altogether a constant source
of expense in the maintenance of the pipes, to say nothing
of the losses occasioned by the delay in making oil deliveries.
298
PUMPING MACHINERY.
The substitution of duplex for single pumping engines cor-
rected at once this troublesome detail in the pipe-line business.
A 12″ × 48″ duplex pumping engine was built by Worth-
ington for the National Transit Company some six years ago,
which has ever since its erection been at work, continuously
delivering 25,000 barrels of oil (50 gallons) in 24 hours at a
pressure of 1500 pounds per square inch, pumping the oil over
a mountain of a height corresponding to the above pressure.
So general is the acknowledged superiority of this type of
pumping engine for mining operations over the single direct-
acting pump, that many firms whose reputation and success
were wholly built up in the single pump business are now
supplying their patrons with pumps having duplex steam-ends,
either simple or compound.
A piston-pump with gun-metal lining, such as shown
in Fig. 113, is often used in mines where the depth does not
exceed two hundred feet and the water is not acid. The
fibrous packing can be readily renewed whenever necessary.
Removable gun-metal linings are commonly arranged so that
they can be partially turned in the water-cylinder to present
a new or uneven surface underneath the piston. As the bolt-
holes are spaced on either three or four divisions of the cir-
cumference, a corresponding new surface may be brought
under the piston until it is so completely worn as to need re-
newal, in which case the substitution of a new lining for the
one worn out can be accomplished in a few minutes.
Piston-pumps occupy less space in a mine than plunger-
pumps of the same capacity. If the water is not bad, the
pump may be of the same general construction as for any
other pumping service for which a piston-pump would be se-
lected. It is a wise precaution, however, to have the pump
brass-fitted throughout; it adds but little to the cost and is
well worth the difference.
Piston-pumps for acid water should have the water-
ends made up of sections, any one of which may be easily
MINING-PUMPS.
299
detached from the assembled water-end and replaced by a new
one, in case such renewal should become necessary through
the corrosive action of the mine water. Fig. 199 represents
partly in section and partly in elevation one side of a 7" X 12"
duplex water-end, from designs by the writer. The suction-
pipe underneath is made in two pieces, with a flanged joint in
the centre; these two castings are in all respects alike; they ex-
FIG. 199.

о
O
tend across the pump similar to those in Fig. 201. There are
four water-cylinders precisely alike. The working barrel of
the pump is made of gun-metal, and fits into an opening, as
shown in the drawing; the flange included in the gun-metal
barrel is faced and drilled similarly to those of the two water-
cylinders to which it is applied, through-going bolts securing
300
PUMPING MACHINERY.
all together. The valve-seat is shown in elevation in the
above drawing; a section of a similar valve-seat is shown in
Fig. 200. The width of the bottom flange is only sufficient
to make a good joint. It will be seen that the water-cylinders,
suction- and delivery-pipes are simply faced castings, held
together by through-going bolts.
Centrally-packed plunger-pumps, similar in general
design to the 8" X 12" duplex shown in Fig. 200, are largely
used in anthracite coal-mines. The water-cylinders are con-
structed with special reference to pumping acid water. The
shape is made up of curved surfaces as far as practicable.
There is no fitting to these chambers other than the facing
and drilling of the flanges, and the boring of the stuffing-
box; the interior surface or skin of the casting not being dis-
turbed offers a better resistance against the action of bad
water than such parts as have had this surface removed.
The plungers are made of cast iron, working through a
stuffing-box included in the water-cylinder casting. These
stuffing-boxes and glands are sometimes lined with gun-metal
bushings, as shown in Fig. 37, but not as a general practice.
The plunger is always kept well lubricated, which greatly
retards the corrosive action of the water upon it. This lubri-
cant also covers the interior surfaces of the gland and the
bored guide at the bottom of the stuffing-box, so that corro-
sion proceeds less rapidly there than elsewhere in the cylinder.
In this design the plunger is driven by a rod passing through
a stuffing-box in the inside head, as must be the case in cen-
trally-packed pumps.
This rod should be of some copper and tin composition, and
never of Muntz-metal. The writer has used Tobin and phos-
phor-bronze with satisfactory results. This gland and stuffing-
box should be fitted with gun-metal bushings as shown in
Fig. 37.
The valves are usually one single rubber disk for plungers
up to 12 inches diameter, and it is this diameter which usually
fixes that of the valve, both being alike. The valve-seats being
MINING-PUMPS.
301
FIG. 200.

FIG. 201.
made of hard gun-metal will outlast several water-ends; the
flange by which a seat is held in place need not be wider
than will make a
safe joint. The de-
livery-valve cham-
ber is of a form well
calculated to resist
the action of the
current of water
passing through it
to the central deliv-
ery-pipe shown in
the end elevation,
Fig. 201.

•
A duplex cen-
trally-packed
water-end for a 15"
X 36" mine-pump,
from designs by the
author, are shown in Figs. 202 and 203. It does not differ
.
26
302
PUMPING MACHINERY.

101
FIG. 202.
FIG. 203.
O
O
materially from the
pump just described
except in the fact of
its having a gun-
metal valve-plate fit-
ted with as many
small valves of the
design shown
shown in
Fig. 55 as the plate
will contain. The
diameter of this plate
can usually be made
to accommodate
from 60 to 75 per
cent. of plunger area,
through valve-seats
32 to 4 inches in
diameter, without
enlarging the water-
chamber to an un-
reasonable propor-
tion. By reason of
the uncertain de-
mands upon mine-
pumps it is recom-
mended that for
strokes of 18 to 36
inches the combined
area of valve open-
ings be not less than
60 per cent. for the
shorter stroke, and
from 75 to 100 per
cent. for the long
stroke, as high.
plunger speed
should always be
}
MINING-PUMPS.
303
accompanied by a correspondingly large valve area to get

the best results.
An outside-packed
plunger - pump with
end stuffing-boxes and
central diaphragm, as
shown in Fig. 204, is an
excellent form of mine-
The absence
pump.
of a piston-rod work-
ing through one of
the water-cylinder ends
is an advantage. The
illustration represents
one side of a 9" X 18"
water-end; by reason
of the projecting
plungers this pump re-
quires a greater length
of floor space for the
same working stroke
than those having cen-
trally-packed plungers.
The end stuffing-boxes
are more convenient for
filling and for adjust-
ment than when placed
centrally. Each
plunger has a cross-
head included in the
same casting; wrought-
iron or steel tie-rods
connect each pair of
plungers belonging to
one side, as shown in
0
0
I
FIG. 204.
the drawing. The valves, valve-seats, valve-chambers, suction-
1
304
PUMPING MACHINERY.
and delivery-pipes are substantially the same as for the two
preceding pumps. Each water-cylinder casting includes two
plunger-chambers; the lugs for attaching the steam-end being
included also, it will be seen that there is much less fitting
required in this than is called for in centrally-packed water-
ends; there is one objection, however, which must not be
overlooked, and that is, in the event of any portion of a water-
cylinder of this design being sufficiently affected by bad water
as to require renewal, a complete water-cylinder must be
furnished.
A compound mine-pump by the Buffalo Steam-Pump Com-
pany is shown in perspective view in Fig. 205, a side eleva-
FIG. 205.

SCONANT-CO."
tion in Fig. 206, and an end elevation in Fig. 207. The lat-
ter two illustrations are sufficiently sectioned to show the
arrangement of plunger, pot-valve chamber, suction- and
delivery-valves.
The high-pressure steam-cylinders of the pump here shown
are 12 inches diameter; low-pressure cylinders, 22 inches
diameter; stroke, 18 inches; the plungers are 7 inches diam-
eter. The rated capacity is 300 to 350 gallons per minute,
delivered 600 feet above the station with a steam-pressure of
65 pounds in the initial steam-chest. The plungers are tied
together with steel tie-rods coupled into cast-steel cross-heads.
The plungers are carried through bronze-lined stuffing-box
glands. The water passages are 50 per cent. of the plunger
MINING-PUMPS.
305
area. Each valve-chamber contains three valves of a size that
also equals 50 per cent. of the plunger area; both suction- and
discharge-valves are overhead, so that the plungers are always
water-packed. The caps of the valve-chambers are held by
swing-bolts. The pump is supplied through a 7-inch suction-
pipe, and the discharge is driven
through a 6-inch column. This
pump was furnished with an im-
proved single air-pump and con-
denser designed for the most
severe duty. The valves of the
steam-cylinder are of the ordi-
nary plain slide type, and ar-
ranged to run by the ordinary
duplex movement already de-
scribed.
A Knowles compound con-
densing duplex plunger-pump,
illustrated in Fig. 208, is from a
drawing representing the 800-
foot station of a mine, with the
pumping engine in place and the
sinking-pump in position below
the station. This pump has a
capacity of 1000 gallons of water
per minute. The water-end is
fitted with pot-valve chambers
similar to those described in
the chapter on hydraulic-press-
ure pumps. This pump is
arranged to draw its supply from
the sump located underneath the

t
FIG. 206.
engine, one or more sinking-pumps raising the water from
a lower level. From this same tank the condensing appa-
ratus draws its supply of injection water necessary for con-
densing the exhaust steam as it issues from the low-press-
ure cylinder. The water from the hot-well delivery of the
น
26*
306
PUMPING MACHINERY.
condensing apparatus flows back into the sump from which it
was taken, an arrangement of piping making this delivery
near to the main pump-suction, so that the hot-well discharge
is at once taken up by the main suction instead of communi-
cating its heat to the whole body of water; by this means a
large volume of water is available for condensation which has
received no heat from the hot-well discharge.
The engraving shows a float in the sump; this float oper-
ates a balanced throttle-valve on the steam supply-pipe. The
movement of this float, up or down, regulates the speed of
FIG. 207.

the pumping engine, and will stop the engine automatically
should the sump be emptied of water.
In deepening the shaft the sinking-pump is lowered at in-
tervals as the excavating progresses. The water at the
bottom of the shaft is taken by the sinking-pump and deliv-
ered to the tank or sump that supplies the pumping engine in
the station. A single steam-pipe down the shaft supplies
both the pumping engine and sinking-pump. The exhaust
steam from the pumping engine, as before explained, is con-
densed by the independent air-pump and jet-condenser. The
MINING-PUMPS.
307
TRAP
SUCTION
exhaust steam from the sinking-pump, however, is usually
carried to the sump or into a special connection arranged on
FIG. 208.

STEAM PIPE
SUCTION CHAMBER
the suction-nozzle of the sinking-pump.
This connection is a neat and inexpensive
condensing arrangement (not shown in the
cut) and does away with the exhaust steam-
pipe in the shaft. With a pumping-plant
similar to that described, mines can be
sunk to any depth desired or flooded mines
recovered with the greatest facility and
security.
The location of the sump does not
always fix that of the pumping engine in
a mine. The former must be placed
wherever the conditions of the mine re-
quire it, and this may not be the best loca-
EXHAUST PIPE
SUCTION HOSE
DISCHARGE
tion for the pump, which latter should be located not more
than 20 feet above the bottom of the sump; and as for the
STEAM PIPE
1
308
PUMPING MACHINERY.
length of the suction-pipe, it is well understood that it should
be as short as possible. This is a detail in mine engineering
which we cannot enter into here, but the main pumping
engine, if more than one is required, ought to be placed as
near the shaft as possible, and if water has to be pumped
from several distant points, and perhaps at greatly varying
levels, small, low-service pumps may advantageously be used
at any convenient place in the mine, the combined deliveries.
of these pumps being into a sump at a convenient level for
the main engine; an arrangement such as that shown in Fig.
208, reproduced from Knowles's catalogue, will make our
meaning clear.
Lining a Mine-Pump with Gun-Metal.-The experi-
ment has been tried of lining pump-barrels and other interior
portions of mine-pumps with gun-metal, but it has proved a
costly thing to do. The intricate details of a pump interior
prevents lining the entire surface with gun-metal. The only
practical method by which it can be done is to have all por-
tions cylindrical, so that the fitting would be reduced to the
simple operations of boring and turning, and then forcing
them in place. By the time all this is accomplished the cost
will fully reach that of a properly-designed water-end con-
structed wholly of gun-metal.
Lining a mine-pump with wood is better and cheaper
than with gun-metal. The pump details must have been pre-
viously arranged with reference to the insertion of the wood
lining. Fig. 209 represents a water-cylinder of a horizontal
plunger-pump with the wood lining in place. The staves are
of soft pine, machine-dressed radially and to the outer and
inner curves, suited to their respective diameters; they are cut
to exact length and arranged in place. Two of the staves
are then bevelled to admit a third stave between them, also
bevelled, and which shall act as a wedge; this middle stave is
then driven home with a maul; all the staves will now be firmly
fixed in their places in the interior of the cylinder. Any
MINING-PUMPS.
309
openings into this cylinder may be lined in the same manner,
care being taken that the inner ends of such staves closely fit
the curvature of the main staves; after these are securely
wedged in place the opening may then be cut through into
the working-barrel. A gun-metal bushing is shown inserted
FIG. 209.

in the bottom of the stuffing-box, through which a cast-iron
plunger works.
A valve-chamber, shown in Fig. 210, is similarly lined with
wood. It will be observed that this interior is an inverted
frustrum or a cone with a cylindrical intersection. The inser-
FIG. 210.


tion of this lining is not so simple a matter as that of a plain
cylinder, because each stave requires to be fitted to the two
top and bottom diameters of the valve-chambers. By making
a wooden templet, having the proper curves and angles accu-
rately representing one stave, duplicates can be quickly made
310
PUMPING MACHINERY.
by a wood-worker, each piece exactly fitting into its place.
Three staves must be fitted for wedging, as in the preceding
paragraph. After the conical portion has been lined, the cylin-
FIG. 211.
•
drical intersection can then be
fitted, after which the branch
opening may be cut through.
This chamber is fitted with a
gun-metal lining below the
valve-seat. The valve-seat is
inserted on a taper, and extends
down, forming a lining to pro-
tect the metal at the top of the
bushing. It will be seen that
this chamber is very thor-
oughly protected.
An air-chamber with wood
lining is shown in sectional
elevation in Fig. 211.

Only
that portion is protected that comes in contact with the water.
The duplex water-end, Fig. 212, is lined throughout with
wood, as shown in the preceding illustrations, and in end ele-
vation in Fig. 213. This pump is fitted with plungers 12 inches
diameter by 48 inches stroke; it is located in an anthracite
coal-mine nearly 1000 feet deep. The whole design is good,
and may be said to fairly represent the best modern design and
construction, except in two points, about which engineers are
not fully agreed: one, in regard to a centrally-packed plunger,
as against end-plungers; and the other, in the employment of
one large valve instead of several smaller ones. The thick-
ness of metal in this water-end subject to pressure is two
inches; the wood lining one inch thick. The valves are 15
inches diameter and 14 inches thick. No air-chambers are
provided except the ones over each delivery valve-chamber.
The Cornish pumping engine was among the first
pumps used in the anthracite coal regions of Pennsylvania.
The late Howell Green, whose experience with mining-pumps
1
MINING-PUMPS.
311

де
مالا
O
FIG. 212.
ollo
FIG. 213.
312
PUMPING MACHINERY.
was both varied and extensive, was of the opinion that "if
the working-barrel was made of hard iron and the length of
rod nicely adjusted so that the leather would come slightly
above the bore in the upstroke, and if the water was clean and
not acid, and the lift not too heavy-then, it is still the best
mining-pump ever made." This view is not now entertained
by mining engineers, and the types of pumping engines al-
ready described and illustrated in this chapter are, for many
reasons, to be preferred.
That type of pumping engine known as the Cornish may
be described as a single-acting, high-pressure, expansive con-
densing engine, working single-acting pumps through the
medium of a beam. Cornish pumps are usually of the
plunger pattern, the plungers being loaded with iron weights.
sufficient to counterpoise the pressure of the water column.
The engine may be considered as consisting of two parts; the
power of the engine is used to lift the loaded plunger, after
which the steam-engine part of the machine is detached and
the weighted plunger is allowed to descend by gravity at a
speed depending on the quantity of engine-power in action
and the rate at which the water is being drawn away. The
chamber of the pump becomes filled when the plunger is raised.
and the act of inhaling the full charge through the suction-
valve is a portion of the work which the steam has to perform,
and a portion also much subject to variation.
The speed of the engine is regulated by an adjustable cata-
ract; the exhaust-valve first and then the steam-valve are
thrown open by treadle-weights, as soon as the catches are
detached by the cataract. The valves are closed by tappets
on a plug-rod, first the steam-valve and then the exhaust-
valve, the former at a period of the stroke varying in practice
between one-third and one-fifth from the commencement, and
the latter at the end of the stroke.
In engines working on this principle, as also in all recipro-
cating engines pumping without cranks, there is nothing to
limit the strokes of the engine to any exact length. It is
necessary, therefore, that bumpers or catch-pieces be provided
MINING-PUMPS.
313
to restrain the engine at both ends from an undue length of
stroke; and thick plates of india-rubber under hard-wood
blocks are now used for this purpose in place of the spring
beams formerly employed. An engine thus arranged, work-
ing alone, lifting water from one fixed level to another, would
work continuously with one length of stroke and at one speed,
at whatever it might be set.
The single-acting engine on the Cornish principle was
thought to have some advantages over a pumping engine with
crank and fly-wheel, in the fact that no power is required in
the Cornish engine for keeping gearing in motion at each end
of the stroke; a certain amount of percussion action is in-
deed necessary to overcome the inertia of the engine at the
beginning of the stroke; but, on the other hand, the whole
engine is brought to a dead stand at the end of every stroke
by the whole effective power being completely absorbed in the
work done in pumping.
The Jeanesville Mine-Pump.-The compound duplex
mine-pump illustrated in Fig. 214 is from designs by Mr.
Vernon H. Rood, and constructed by the Jeanesville Iron-
Works, Jeanesville, Pa. It represents their standard design
of compound condensing, duplex, outside-packed plunger
mine-pump, designed especially for the anthracite and bitumi-
nous coal-mines of Pennsylvania.
The pump selected for illustration is one having-
High-pressure cylinders, 25 inches diameter.
Low-pressure cylinders, 42 inches diameter.
Pump-plungers, 14 inches diameter.
All of 48 inches stroke.
7
Its rated capacity is 2000 gallons of water per minute
against a head of 425 feet.
The steam-end shown in sectional elevation in Fig. 215 is
of the ordinary duplex form. One rod is common to both
pistons, and passes through stuffing-boxes in the high- and
low-pressure cylinder-heads. The valve-faces of both cylin-
ders are in the same plane, the valves being operated by a
27

LEGO GOESER-CUN.
314
----
FIG. 214.
MINING-PUMPS.
315

BOD
EXOTH
FIG. 215.
316
PUMPING MACHINERY.
driving mechanism similar to that shown in Fig. 178. The
dash-relief valves are fitted to the high-pressure cylinder
only. This detail is similar to that of Fig. 173. An end
elevation of the steam-cylinders is shown in Fig. 216. The
Ο
FIG. 216.
Ο
O
O
Ο
customary practice
on all large com-
pound duplex
steam-ends in hav-
ing one main and
two auxiliary throt-
tle-valves is here
shown. The object
of the auxiliary
throttle-valves is to
give each engine a
separate adjust-
ment should one
side appear to work
a little slower than
the other; in start-
ing and stopping
the engines the
upper or main valve

only is used, the auxiliary valves not being disturbed in their
adjustment. This steam-end, in common with the general
practice in mining-pumps, is not steam-jacketed, but when in
position in the mine and thoroughly tested, the cylinders and.
chests are covered with a non-conducting material, such as
magnesia, which can be laid on with a trowel. This prevents
radiation and a protection against the water in a "drowned-
out" mine. The steam-ends of mine-pumps are generally
made of unusual strength, with extra large wearing surfaces
and connections to withstand the continuous and severe duty
they are called upon to perform in times of high water, or
possibly in a "drowned-out" mine, when it will be forced to
work submerged to a depth of from 10 to 6o feet, in which
case lubrication or any other care is impossible.
MINING-PUMPS.
317

The water-end is shown in
partial sectional elevation in
Fig. 217. It consists of four
cylinders, or working-barrels,
as they are generally called,
bolted together in pairs, with
a recessed blank flange be-
tween them; these working-
barrels rest on two cross-
feet, which serve to tie the
whole four together as if
they were one casting. To
reduce the number of spare
working-barrels necessary to
keep on hand, and also to
facilitate their removal when
worn out, the bosses for
receiving the tie-rods which
connect the pump to the
steam-cylinders are not cast
to the water-cylinder, but to
a strongly-bracketed flange,
which is bored and faced to
fit over the outside of the
stuffing-box and against a
flange cast on the working-
barrel from 8 inches to 10
inches from its end, as shown
in the drawing.
These pumps, in accord-
ance with the best modern
practice, are built to gauges,
so that only two spare work-
ing-barrels are required for
a whole pump, the left front
being exactly the same as
the right back, and vice
T010
FIG 217.
27*
318
PUMPING MACHINERY.
versa. This arrangement also serves to preserve the align-
ment of pump, as this tie-rod flange, as it is called, being a
separate casting from the working-barrel, and not exposed to
the action of the acid water, will last as long as the steam-
end, so that in renewing a front or inside working-barrel it
is only necessary to place it in position and draw up the bolts
to insure the pump being in perfect line. To still further
facilitate the renewal of inside working-barrels, the tie-rods are
turned to the same size as that portion which goes through the
bosses on the tie-rod flange; cast-iron split clamps or collars,
shown in the drawing, are bolted thereon of sufficient length
to allow of the tie-rod flange being drawn towards the steam-
end far enough to clear the working-barrel, which can then be
removed by lateral movement without disturbing the steam-
end.
The pump has four single-acting, outside-packed plungers
connected in pairs by two steel parallel rods coupled to strong
cast-iron cross-heads, which, in turn, are fastened to plungers
by T-headed bolts let into cored recesses in the plungers.
The plungers are supported on the outboard end by adjust-
able shoes or gibs on the bottom of each cross-head. These
run in slides, as shown. The writer considers this outside
form of connection the best, even for very moderate lifts, in
mine-work, as it does away with the necessity of bronze rods
and connections, which are always a source of more or less
trouble and uncertainty when used in acid or gritty water.
The plungers are made of hard, close-grained iron, which
will, if properly packed and lubricated, last fully as long as
brass ones. The plunger gland-bolts have T-heads, and are
let into cored recesses on the outside of the stuffing-box, the
same as the bolts securing the plungers to the cross-heads.
There are no studs whatever in the water-end.
The water-end has 16 valve-chambers with a single valve
in each. A sectional elevation of the valve-chamber is shown
in the longitudinal section, Fig. 217, and in the end elevation,
Fig. 218.
The use of a single large valve to each chamber instead of
MINING-PUMPS.
319
a number of small ones, Mr. Rood thinks, has many advan-
tages when applied to mine-work. It may be well to state
that the mine water in the anthracite coal-fields not only
attacks iron but brass and phosphor-bronze as well, so that
the life of a valve-seating is at best a limited one. This fact
alone makes it imperative that pump-valves for mines be
simple, strong, easily examined, and quickly replaced. It
must be admitted that 8 or at the most 16 large valves are
FIG. 218.

O
O
O
much easier kept in repair, and can be removed much more
rapidly than from 50 to 100 small ones. The several details,
such as valve-stems, springs, etc., are always much larger and
stronger for large valves, and are better able to withstand the
over-effort of an energetic man pulling on a long wrench;
those who have had small valve-stems twisted off in this man-
ner with the water gaining on the pumps will fully appreciate
the importance of this detail. An enlarged illustration of
the valve and its seat is shown in plan and sectional elevation
in Fig. 219. The valve-seat is held in place by a flanged joint
between the valve-chamber and the pump or pipe-flange under-
neath. The valve is made of india-rubber with a gun-metal
cap; a phosphor-bronze spring and bolt completes the arrange-
ment as shown.
While 8 chambers and valves would have answered as
well as the 16 small ones, so far as the proper and smooth
320
PUMPING MACHINERY.
working of pump is concerned, the desirability of keeping the
chambers as small as possible for ease in handling in case of
renewal, etc., determined in this case the use of the smaller
pattern.
FIG. 219.
ㅁ
​о
ㅁ
​оо
N N N
о
コロ
​This form of valve-chamber, while differing radically from
the usual style, has after a thorough trial proved itself so well.
fitted for the work as regards durability, etc., that they have
•


MINING-PUMPS.
321
adopted it as a standard for their larger permanent station
mine-pumps, as combining the greatest strength with the least
weight of metal, the largest clearance around the periphery
of the valve without any waste room elsewhere, and finally as
costing very little to fit up and consequently not expensive to
renew.
The question of room around the periphery of a valve is
an important one in mine-pumps, as the water, in most cases
being acid and carrying large quantities of fine coal and grit,
if allowed to impinge against the walls of a chamber as it
rushes out from under the valve, will cut the metal away so
rapidly as to ruin the chamber in a short time. In the cham-
ber illustrated it will be noticed that in addition to giving an
extra amount of room at this point, its shape is such as to
turn the course of the water upwards and inwards instead of
against the walls. It is to this feature that is attributed the
long life of this style of chamber, which invariably outlasts
all the other parts of the water-end.
The method of securing the seatings between flanges of a
chamber and the water-cylinder, while not new, is particularly
adapted to mine-pumps, as it admits of very easy and quick
renewals, there being no fits to make in the chamber, and it
has the further advantage that no leak past the seat can occur
without being instantly detected and stopped.
Steam-Pipes for Collieries.-An excellent paper, pre-
pared by Mr. E F. C. Davis, and included in Vol. XI., "Trans-
actions of the American Society of Mechanical Engineers,"
is here reproduced:
"The most common and the cheapest method of carrying
steam, taking the world at large, is probably through wrought-
iron 'gas-pipe' joined by the taper-thread, screwed into
sockets or ferules.
"This answers admirably for small pipes, and even for com-
paratively large pipes where the conditions are favorable for
screwing up the joints, and where the threads are not sub-
jected to any serious corrosive action. Many of the steam-
ย
322
PUMPING MACHINERY.
pipe lines in the anthracite coal-regions, however, run for great
distances underground, through contracted slopes and head-
ings where it is almost impossible to make the screwed joint.
In the screwed socket-joint there is always some space be-
tween the ends of the pipes, and the condensed steam from
the best available feed-water is so corrosive that a cutting or
furrowing action takes place between the ends of the pipes
and the ferule, which sooner or later causes leakage. It is
then impossible to tighten up these screwed joints without
screwing up the whole pipe-line.
"Some of these difficulties are avoided by the use of
'flange-unions.' With these the pipe-line can be more con-
veniently put together underground, and in the event of a
leaky thread the flange can be screwed on tighter, or a de-
fective pipe can readily be replaced by a new one of the
same length. But in the ordinary flange-union there is a space
between the ends of the pipes, and the above-mentioned cor-
rosive action is so destructive to the threads that cast-iron
pipes have generally been considered necessary for reliable
and durable steam-pipe lines; though the first cost is about
double that of wrought-iron pipe.
"In view of the foregoing, the Philadelphia and Reading
Coal and Iron Company has adopted the flange-joint shown
in annexed cut for all colliery steam-pipes. These flanges are
screwed tightly on the pipe,-the pipe carried in a steady rest,
-and the end of pipe and flange faced off flush with each
other. The lugs are at the same time bored out, and the pro-
jection turned off concentric with the bore of the pipe. This
insures perfect continuity in the pipes, and the lugs also centre
the gum-joint rings accurately, so that a gum-joint is obtained
between the abutting ends of the wrought-iron pipes. The
continuity of the bore of the pipe insures a free flow of steam
and condensed water, so that all liability to furrowing at the
joints is avoided and the gum-joint formed between the ends
of the wrought-iron pipes protects the thread from all danger
of corrosion. If an odd length of pipe needs to be made at
a colliery, the pipe, if not over four inches, can be threaded
MINING-323
PUMPS.
.
Size of Pipe.
Diameter of
Flange.
Ө
Ө
PUMPSwith a hand-stock and die, and a finished flange screwed on
until the pipe projects through. The pipe must then be filed
off flush with face of flange.
'In moulding these flanges it is best to have the pat-
tern arranged to leave its own cores.
This insures accu-
FIG. 220.

* F *
<E Turned
KE Bored
Faced
Faced
G

of Flange.
Length of Pipe.
A.
B.
C.
D. E. F.
G.
H.
J.
K.
L.
44
3″
734"
3½"
81½"
9½"
57/2
5" 102'
6" 12"
8" 7/8"
7"
6
7"
13"
IO" 7/8"
8"
6
IT"
8"
7/8"
14"
6
I2"
10"
164"
7/8"
10
8
12"
6"
634" 3/4
7/2" 7/8"
3/4"
5"
9"
3"
I
11410100
m\m\^\^\S S
3/4"
I!!
3/11
7/8"
72
72
72
1/2"
444
7/8" 3/4"
Feet.
1/2"
7/2" 3/4"
11/4" 16
7/8"
1/2"
16
3/4"
4
7/8"
1/2"
13/8"
16
3/
72
4
72
IA!!
7/8" 3/11
1/2"
13811
16
I/!!
72
7/8'
7/8" 3/11 1/2"
1/2"1
11½"
20
11/2"
FO
11"
1/2" 6 7/8"
5/8"
13/4"
20
5/8" 13/4"
20
Pipe must be screwed through flange so as to be steam-tight, and flange and end of pipe
faced off flush at one operation.
racy in the positions of the bolt-holes and the large central
hole, relative to each other and to the other parts of the
flange.
"Several thousand feet of steam-pipe fitted with these
flanges have been put in service, and have all proved perfectly
satisfactory."
324
PUMPING MACHINERY.
Water-pipe for mines should be of hard cast iron made
in uniform lengths, so that in the event of having to replace
FIG. 221.
a section of pipe it can
be easily done, and
without having to take
the time necessary to
face it off to an exact
length. The pipes for
the anthracite coal-
mines of this State are
generally cast with
what corresponds to a
chipping-piece in ordi-
nary work, this projec-

tion being one-fourth of an inch above the body of the flange,
by one and a half inches wide. This chipping-piece is cast
against a cast-iron chill, thereby insuring a smooth and true
face for making a joint. The chill is so arranged as to include
the bolt-hole cores for the flanges, so that the pipes require no
machine-work whatever. Fig. 221 represents a section of a
FIG. 222.
joint made up. For
column or delivery-
pipes a wrought-iron
welded ring 14 XI
inch iron is made with
an inside diameter half
an inch larger than the
bore of the pipe; this
ring is spirally wound
with strips of cotton
cloth 1½ inches wide,
previously boiled in
coal-tar until thor-

oughly saturated. For making slight bends in a pipe-line, a
bevelled ring of either wrought or cast iron similarly wrapped
as above is substituted for the ring having parallel sides. (See
Fig 222.)
ROTARY PUMPS.
325
!
CHAPTER XV.
ROTARY PUMPS.
Rotary motions were, as Ewbank suggests, favorite ones
with ancient philosophers; they considered a circle as the
most perfect of all figures, and erroneously concluded that a
body in motion would naturally revolve in one. In later
times, as the problem of pump design began to press itself
upon the attention of mechanics, it was a favorite one with
many to convert the intermittent action of the pump-piston
into a continuous circular movement, that the power expended
in constantly bringing all the
water in the cylinder and suc-
tion-pipe alternately to a state of
rest and motion was saved, be-
cause the liquid would be kept
in constant motion in passing
through the pump.
One of the oldest and at the
same time one of the best rotary
pumps is shown in Fig. 223. No
precise date can be assigned for
its invention, but it existed in its
present form probably early in
the seventeenth century. Two
cog-wheels, the teeth of which
are fitted to work accurately into
FIG. 223.
LILD (NITORING!
JATIAN
each other, are enclosed in an elliptical case. The sides of these
wheels turn close to those of the case, so that water cannot
enter between them. The axle of one of the wheels is con-
:

28
326
PUMPING MACHINERY.
L
•
tinued through one side of the case, and the opening made
tight by a stuffing-box or collar of leather. A crank is
applied to the end to turn it, and as one wheel revolves it
necessarily turns the other; the direction of their motions
being indicated by arrows. The water that enters the lower
part of the case is swept up the ends by each cog in rotation,
and as it cannot return between the wheels in consequence of
the cogs being there always in contact, it must necessarily rise
in the ascending or forcing-pipe. The machine is therefore
both a sucking and a forcing one, as indeed all rotary pumps
are.
Rotary pumps may be divided into classes, according to the
forms of and methods of working the pistons, or those parts
that act as such, and according to the various modes by
which the butment is obtained. It is this last that receives the
force of the water when impelled forward by the piston; it
also prevents the liquid from being swept by the latter entirely
round the cylinder or exterior case, and compels it to enter
the discharging-pipe. In these particulars consist all the
essential differences in rotary pumps. In some the butments
are movable pieces that are made to draw back to allow the
piston to pass, when they are again protruded till its return;
in others they are fixed and the pistons themselves give way.
It is the same with the latter; they are sometimes perma-
nently connected to the axles by which they are turned, and
sometimes they are loose and drawn into recesses till the
butments pass by.
An objection to rotary pumps is that they are noisy, wear
out rapidly, and are costly to repair. They do not have a
high efficiency, to begin with, and this grows constantly less
with wear. They are, nevertheless, very useful pumps, and in
the smaller sizes, at least, are used in large numbers throughout
the country in manufacturing and mercantile establishments.
The engraving, Fig. 224, shows the internal arrangement
of rotary pumps as made by Rumsey & Co., who claim that
this form of revolving cam or piston has been demonstrated
after many years of experiment and successful practice by the
ROTARY PUMPS.
327
most prominent mechanics and manufacturers to be the best
in use, and of the proper shape to insure the very minimum of
friction and wear.
The arrows shown in the
cut indicate the direction in
which the cams should be
run,-i.e., inwardly, or to-
wards each other at the top
or discharging orifice.
They have built and placed
these pumps for many years,
claiming they have given
uniform and unequalled sat-
isfaction in all their situations
FIG. 224.

and uses, while comparatively cheaper and stronger than any
other first-class rotary. They have heavy cast-steel mandrels,
and babbit-metal boxes, with two pairs of heavy gears to
relieve the cams. They discharge a large and constant stream
without the aid of an air-chamber, can be run at a very high
rate of speed without injury, and will operate for months at
a time without cessation, having no packing or valves to be-
come obstructed or worn out. There is nothing about these
pumps but metal. They are simple, and all essential parts can
be got at without trouble. They will not clog in pumping
thick fluid or matter, and they are made non-freezing when not
The motion being rotary or circular, the strain upon
the belt or gear is equal at all points. Their qualities well
adapt them for establishments, including distilleries and brew-
eries, for pumping hot liquors; in oil- and sugar refineries,
paper-mills, and starch-factories; in slate-, stone-, and marble-
mills, where they serve the double purpose of supply and
freeing the quarries of muddy and gritty water; at railway-
stations, mines, salt-wells, water-works, and also for wrecking-
pumps, as they will not become clogged with wet grain, etc.
in use.
Rotary-pump design has engaged the attention of in-
ventors for many years; unfortunately, most of them have
328
PUMPING MACHINERY.
་
been more or less badly designed, and in consequence the
mere use of the word " rotary" in connection with any pump
seems at the present time a sort of disparagement which,
according to M. Poillon, the principle does not merit.
On looking into the cause of this state of things, the mis-
chief is readily perceived to have arisen from neglect, on the
part of inventors and makers of rotary pumps, of some very
simple facts in connection with the subject. In most cases
the following points have practically been lost sight of:
1. That water is practically incompressible, and that in a
liquid mass the pressure and the velocity at different points
bear an invariable relation to each other.
2. That in any machine which has attained its regular
working speed the driving power expended is equal to the
sum of the useful work done plus the useless resistances
overcome, and that, when proper care is not taken to prevent
such a result, the useless work may absorb actually the
greater part of the power.
3. That, whatever care be taken in the construction of a
machine, two metal surfaces cannot be made to rub one upon
the other in water without occasioning their rapid destruction.
by friction, but that if simply in contiguity without friction
they will work without injury.
4. That, as the speed increases, the friction of the work-
ing parts in a pump very quickly becomes a matter of great
importance and absorbs a considerable amount of work,
while, on the contrary, absolute contact between parts in
motion is a matter of no practical importance, the leakage
becoming relatively less at higher speeds.
Hence may be explained the failures attending so many
designs of rotary pumps, due equally to the want of appre-
ciation of theory and to neglect of the teachings of practice.
An examination into the efficiency of many of the rotary
pumps now in use show that it is very low. The great excess
of driving-power over useful work done arises from two
principal causes :
1. The inertia of the water, or difficulty of putting it into
ROTARY PUMPS.
329
motion again after it has been brought to rest, and the conse-
quent reduction of the effective pressure.
2. The necessity of imparting at certain moments a high
velocity to a considerable mass of water, the production of
this velocity requiring the expenditure of a great amount of
power, of which only a small portion is given out again as
useful effect.
It is clear that if these two sources of difficulty are got rid
of, a near approach is made to perfect efficiency; that is, to
an equality between the theoretical driving-power required
and that which is utilized in the work done.
The Greindl pump, illustrated in Fig. 225, consists of a
chamber within which work two cylindrical drums, A and B,
of equal diameter, running in contact with each other on
Madle
FIG. 225.

parallel shafts. One of these drums, A, carries two radial
vanes or blades acting as pistons, which as they revolve enter
alternately into a recess of epicycloidal section extending
along the whole length of the other drum, B. The shafts of
28*
330
PUMPING MACHINERY.
the drums are geared so that the recessed drum, B, makes
two revolutions to one of the bladed drum, A, thereby en-
abling the single recess in the quick drum to serve for the
two blades on the slower. The inlet and outlet passages are
arranged in such a manner as to present everywhere the same
sectional area throughout the entire course of the water, in
order not to impede its movement in any way during its pas-
sage through the pump. In consequence of the continuous
motion of the stream of water, any foreign solid substance
can pass through the pump without occasioning either a
stoppage or a breakage. The blades of the slower drum
strike the water without any perceptible shock. Lateral
pockets in the end-cover plates afford ample space for the
water to escape through at the moment when the space left
between the blade and the recess threatens to be insufficient
for that purpose.
In rotary pumps, by simply changing the speed of work-
ing, the actual quantity of water delivered can be varied
within certain limits without producing any change in the
FIG. 226.
proportionate efficiency of the
pump, whatever its size, an
advantage which cannot be
obtained with centrifugal
pumps.

The Berrenberg rotary
pump is shown in sectional
elevation in Fig. 226; each
piston is a circle, from which
four semicircular sections have
been cut. In two of the semi-
circles, opposite each other,
tubes are bolted through from
the semicircular spaces left
vacant. It is so constructed that when the pistons revolve, the
tubes (or boxes, as they are called) of one piston fit into the
vacant spaces of the other piston with perfect accuracy; and
ROTARY PUMPS.
331
the circumferences of the pistons being always in contact with
each other, it will be seen that there is no time in the course
of their revolution when there is much chance for leakage.
In case of wear, it can be made good by removing the tubes
(or boxes) and putting a few thicknesses of paper packing in
the seat, or by turning the tubes a little so as to present a
new surface to the point of contact with the case.
The pump
is double-geared, and has inside and outside bearings; these
latter are tapered, so that any wear in the bearings or shaft
may be taken up and the shafts kept true.
TABLE XIX.
EXPERIMENTS ON THE EFFICIENCY OF A 2-INCH BERRENBERG ROTARY PUMP,
CONDUCTED BY R. A. HALE, HYDRAULIC ENGINEER, APRIL,
1888.


Number of Experiment.
Revolutions of the Pump
per Minute.
Gallons pumped per Min-
ute.
Height of Centre of Press-
ure-Gauge above Water
in Pit. Feet.
Reading of Pressure-Gauge.
Pounds.
Total Lift in Feet.
Depth of Water in Feet on
Weir.
Cubic Feet of Water dis-
charged per Second.
driving Pump.
Horse - Power
Actual
used
in
Work performed
pumping against
Horse-Power.
Efficiency of Pump
Lift.
Ratio
of Usefl Effect to Power
expended.
I
1 2 3 4 DO 78 9
10
690 120
7.00
722 127 7.03
10.6
25.2
31.49 .233
65.36
.268
4.62
0.96 .207
.242
.284
6.05
2.II •350
698 126 7.04
21.4
56.47
.241
.282
5.03
1,80 .359
697 124 7.05
2.49 .428
8.65 27.08
580 107 7.10
593 | 104
583 99
7.10
31.5 79.81 .237 .275 5.81
24.0 62.54
40.3 100.24
7.15
573 93
7.20 55.15 134.60
575 96
591 ΙΟΙ
II 530 83
12 478 86
13 479 81
14 468 76
15 451 70
16 462 69
475 81
*17
.215 .239 2.36 0.73 .311
.210 .231 3.43
1.64 .478
.203 .221 4.88 2.51 .515
.194 .206 6.47 3.14 .486
.215 5.04 2.51 .499
.207 .226 3.71 1.80 .485
.185 8.61 3.68
.427
1.76 0.48
.200
7.25 41.5 103.11
7.33 27.2 70.16
7.43 72.9 175.83 .180
7.35 6.3 21.90 .185
7.40 22.9
60.30 .177
7.40 33.I 83.86 .170 .170 3.14
7.40 47.6 117.36 .161 .156 4.01
7.40 65.2 158.00 .159 .154 5.98
2.76 .462
7.40 23.5 61.70 .177 .181
2.43 1.27 .521
.192
.180 2.42
.270
I 23
.510
1.62
.514
2.08 .518
* 3-inch belt in this experiment.
The power was measured by a Webber dynamometer of
10-horse-power capacity. The pressures were measured by a
gauge attached to the discharge-pipe, about seven feet above
332
PUMPING MACHINERY.
گھر
the level of the water in the tank from which the water was
taken.
The water was measured over a rectangular, sharp-crested
weir, with two end contractions. The length of the weir was
.763 feet, and the depth was measured by a hook-gauge, and
quantity of water passing was measured by the formula of
J. B. Francis, C.E.: Q 3.33 (LH) H 3, thus giving
the quantity in cubic feet per second.
The pulley for the 3-inch belt proving too small to convey
the power for the range of pressures which it was desired to
run, the loose and tight pulleys were blocked together, and a
6-inch belt was used. The accompanying table comprises
seventeen experiments, divided in three groups of varying
speeds obtained by shifting the size of the pulleys on the
counter-shaft; each group covers a range of various pressures,
from which can be determined the pressures and speeds to
give the best efficiency in the working of the pump. The first
group was run at a much higher speed than the pump was
intended to run, and a wide range of pressures was not taken.
The second and third groups were more nearly alike in
efficiency; the third one, with the lowest speed, showing the
best result.
The varying pressures were produced by partly closing the
valve to the discharge-pipe, which, in general, decreased the
amount of water pumped.
.30
.45
Pressures in Pounds per
Square Inch.
Efficiency of Pump
=
Ratio of Useful Effect to Power
expended.
GROUP 2.
.30
GROUP 3.
20.
30.
.48) Mean,
Mean,
40.
.50
.493
50.
.515
.515
.517
60.
.50
70.
.455
80.
.4I
.52
TFF
.50
.485
.45
It will be seen that, between 30 and 50 pounds pressure, the average effi-
ciency is .493 with 575 revolutions of the pump, and .516 efficiency with 460
revolutions per minute, the efficiency falling off above and below that point.
ROTARY PUMPS.
333
The above may be taken as a fair sample of the performance
of rotary pumps generally, and the few illustrations given will
suffice our present purpose. To enter upon a description of
the various designs for rotary pumps would require too much
space in this publication; the writer suggests to those in-
terested in rotary-pump design the perusal of Kennedy's
translation of Reuleaux's "Kinematics of Machinery." The
elementary details are systematically classified; sectional ele-
vations are given by which any particular construction may
be quickly identified.
334
PUMPING MACHINERY.
CHAPTER XVI.
CENTRIFUGAL PUMPS.
A DEMAND exists for a pump with which large volumes of
water may be quickly handled, such as in tanneries, paper-
mills, print-works, dry-docks, etc. This water being liable to
contain chips, bark, and other floating matter, makes it espe-
cially desirable that the pump be valveless,--a problem for
which the centrifugal pump offers an almost complete solution.
The construction of the centrifugal pump is exceedingly
simple, consisting of a revolving fan having two or more.
blades, either straight or curved, attached to a revolving
spindle, and fitted in a case or shell so constructed that the
suction shall enter at the centre of the wheel, and the delivery
placed tangent to the outer path of the revolving blades.
The recorded experiments relating to centrifugal
pumps are few. Among these, two are favorably known to
the writer,-one by Mr. R. C. Parsons,* England, and the
other by Mr. W. O. Webber,† Lawrence, Mass, the first re-
lating more especially to the theory of centrifugal pumps, and
the latter to their efficiency as compared with reciprocating
pumps; the writer acknowledges his indebtedness to both of
these experimenters for subject-matter used in this chapter.
Centrifugal pumps are by no means a modern in-
vention, the crude idea of which probably dates as far back
as the middle of the last century, when the mathematician
* Proceedings of the Institution of Civil Engineers. London, 1876. Vol. XLVII.
† Trans. Am. Soc. Mechanical Engineers. New York. Vols. VII. and IX.
CENTRIFUGAL PUMPS.
335
i.
Euler brought out a primitive form of centrifugal pump, an
account of which he published in the Proceedings of the
Academy of Berlin for 1754, but which never came into prac-
tical use.
From that period many rotary and centrifugal
pumps were invented, principally by French engineers, but
none of them seemed to have yielded even a reasonably good
efficiency. The first mention of a centrifugal pump at all to
be compared with those of the present day is in the year
1830, when one was erected by Mr. McCarty in the navy-
yard at New York, and some improvements were patented by
him in the following year. The next epoch in the history of
the centrifugal pump is the Exhibition of the year 1851,
London, when the late Mr. J. G. Appold achieved a great
success with his pump of trebling the efficiency obtained by
any other exhibitor.
The experiments made with Appold's centrifugal
pump showed that its efficiency mainly depended upon the
form of the blades of the fan; and, further, that the best form
was a curved blade pointing in the opposite direction to that
in which the fan revolved. These were tried in comparison
with two other forms.
Height
of
Lift in
per
Feet.
Minute.
Discharge Revolu-
in Gallons tions
per
Minute.
TABLE XX.
APPOLD'S CENTRIFUGAL PUMP.
Velocity of Percentage
Circumfer- of Effect
10
Power.
ence in Feet
per Minute.
With radial arms
18.0
474
720
2262
24
With straight inclined
arms
18.0
736
690
2168
43
With curved arms
∞ oi
8.2
2100
828
26ɔI
59
9.0
1664
620
1948
65
18.8
1164
792
2988
65
19.4
1236
788
2476
68
27.6
681
876
2751
46
From the above table it will be seen that as between radial
blades and curved blades the increase in efficiency was more
1
336
PUMPING MACHINERY.
than doubled in favor of the latter; subsequent calculations
and more exhaustive experiments show that nothing in con-
nection with centrifugal pumps has been more clearly proved,
both theoretically and experimentally, than that the blades.
should curve backward, and according to Thomson the gen-
eral form of the curved blade is an element of great impor-
tance, and it is a necessary consequence of the arms being
thus shaped that the water is driven through the wheel, partly
by the action of centrifugal force and partly by the oblique
pressure of the blades on the water. The ratio which each
of these forces bears to the other varies even in the same
pump, according to the proportion the speed of the pump
bears to the height of the lift. When the lift is very small
and the speed great, the water is expelled without having any
considerable rotary motion imparted to it. In such a case
the resistance to the outward motion of the water is so small
that the oblique action of the blades is sufficient to expel it
without giving it a speed of rotation at all approaching that
of the fan.
In centrifugal pumps the main object is to pass the water
through the pump with as little "whirling" velocity as possible,
because the power thus absorbed is not again given out, and
becomes a more or less hinderance to the flow. Designs vary
so much that it is clear no settled opinion has yet been reached
as to the best proportions for blades, nor yet as to the best
proportions and best shape for the case which contains the re-
volving fan. Mr. Thomson recognizes it as a favorite idea.
with many designers of centrifugal pumps to contract the
space between the sides of the fan at the circumference so as
to make the area of opening through the fan uniform, and
thus preserve a uniform radial motion of the water during the
whole passage through the fan. The effect of this as well
as of many other modifications was tested experimentally by
Mr. Appold, and he found that it was productive of no in-
crease of duty. Mr. Parsons, on the other hand, considers it
a detail which affects the efficiency of the pump, but not to
the same extent as the form of the blades, and says that "the
CENTRIFUGAL PUMPS.
337
old theory which Morin, Appold, and many others held, and
which is still held in some recent books,-viz., that as long as
the casing outside the fan is large enough it is immaterial what
shape it is, can be proved to be false both by theory and
experiment.'
The passages throughout the pump must be so pro-
portioned as to have a gradually increasing velocity in the
water until it arrives at the circumference of the fan, and then
to have a gradually decreasing velocity until it issues from the
discharge-pipe. This condition is effected by having a conical
end to the suction-pipe, and, what is much more important, is
to have a spiral casing surrounding the fan. The importance of
this last detail, Mr. Parsons says, is shown most conclusively
by experiments with both circular and spiral cases. The form
of the casing should be an Archimedian spiral, which has the
property that the water flowing round the case moves with
the same velocity as that issuing from the fan. The casing
should then gradually open out into the discharge-pipe.
TABLE XXI.
EXPERIMENTAL EFFICIENCIES OBTAINED BY R. C. PARSONS ON A 14-INCH RE-
VOLVING FAN, 10-INCH SUCTION, AND IO-INCH DISCHARGE, MADE ON THE
APPOLD PRINCIPLE.

FOOT-POUNDS.
Revolutions Gallons
per Minute. per Minute.
Lift in Feet.
Efficiencies.
Per cent.
Water Raised. Indicated Power.
392
1012
14.67
148,461
298,438
49.74
394
1108
14.70
162,875
317,158
51.35
395
1197
14.65
175,364
332,136
52.80
400
1431
14.75
211,073
374,954
56.20
405
1695
14.75
251,987
419,790
60.17
425
I 108
17.20
190,576
388,316
48.97
431
1431
17.40
248,994
447,552
53.63
435
1695
17.60
298,310
486 050
61.37
This pump was placed on a floating scow to obtain as
nearly as possible a constant lift; it was driven by a separate
steam-engine, and the power was measured by a dynamometer.
There are two totally different conditions in which a cen-
P
29
W
338
PUMPING MACHINERY.
trifugal pump may be situated while it is rotating,—one in
which it is revolving just fast enough to raise the water up
to the discharge-pipe and no farther, and another in which
it is revolving slightly faster, and is discharging water out
of this pipe. In the first case there is only centrifugal force,
which is produced by the water in the fan rotating, that
maintains the column of water in the discharge-pipe. In the
second case, this force is still produced, but in addition to it
another, which may be called the force of impact, or in other
words, the force with which the blades of the fan impinge
against the water discharged by the pump.
The centrifugal force in the first case was calculated by
Mr. Parsons to be as follows: Assuming that the fan is a
cylinder of water; every particle of this water as it rotates
exerts a force outwards from the centre; consequently the
force exerted at the circumference, or that which maintains
the head in the discharge-pipe, is the sum of the forces of all
the particles from the centre of the fan to the circumference.
This force is given in pounds per square inch by the formula
R
F=
SP xd x
2002
08
Integrating this expression
F
Þ R2 w2
2g
(1)
(2)
weight in pounds of a column of water I inch square in section and I foot
long.
Rradius of fan in feet.
ข
g
angular velocity of fan.
dynamical force of gravity.
Now, since R wv, where v equals velocity of circumference of fan, by re-
placing R2 w2 by z2 in equation 2 it becomes
F=
pv2
2g
(3)
Now, supposing that the head supported by the fan, while it is rotating with a
tangential velocity v, be h, then the pressure at the base of the column is ph, but
by the ordinary formula of dynamics
22
h- therefore ph—prz
2g
thus by equations 3 and 4 F = ph.
2g
(4)
CENTRIFUGAL PUMPS.
339
7
Therefore a fan, when rotating, will support a column of
water the velocity due to whose height is equal to the tangen-
tial velocity of the circumference of the fan. This conclusion
is fully borne out by experiments, where corrections are made
for the axle of the pump displacing a small quantity of water,
and thus reducing to a slight extent the centrifugal force.
The Second Force exerted in this Case is that of
Impact. It is estimated by the maximum tangential velocity
generated in the water passing through the fan, which takes
place just as it is escaping at the circumference. The reason
advanced that this force can be estimated by the tangential
velocity produced is that no other force can produce this
velocity.
The centrifugal force can only produce a radial force
or radial velocity, but can in no case produce a tangential force
or tangential velocity. This latter force can only be made use
of by gradually reducing the velocity of the water issuing
from the fan, and this condition is effected by the spiral casing
and conical discharge-pipe, which can easily be calculated by
multiplying v by cosine ', the angle made by the blade of the
fan at its outer extremity with the tangent to the fan; and
subtracting this from V, the velocity of the circumference, the
absolute tangential velocity of the water leaving the fan is
obtained,-viz.:
ارح
<
VI v cos A
(5)
The head then due to this velocity is given by the formula
H₂
7/2
2g
(6)
This, in other words, is the height that the water would rise
supposing that there was no friction to impede it. Now, the
circumferential force has been estimated in pounds per square
inch; but by dividing it by 0.434 it is reduced to feet head of
water. Then, by adding these two heads together, the theo-
retical height to which the pump lifts the water is obtained,
—i.e.,—
·
F
212
H+ H₂=
+
.434 2g
(7)
340
PUMPING MACHINERY.
This theoretical lift is always greater than that deduced by
experiment, and it is only in a perfect pump that these two lifts
would coincide. Consequently, if the practical lift be divided
by the theoretical lift, and the result multiplied by 100, the
percentage efficiency of the pump is obtained.
Effect of High Speed of Rotation.—The faster the fan
rotates, the lift remaining constant, the smaller is the centrifu-
gal force. This seems to be a paradox at first sight, but the
reason is evident. As the discharge increases, the velocity
of the water in the casing more nearly approaches that of the
water leaving the fan; consequently the efficiency of the pump
improves, and the theoretical lift diminishes, and with it the
centrifugal force.
A remarkable property of centrifugal pumps may be men-
tioned, which has been clearly shown by experiment, and that
is, a small increase in the number of revolutions of the pump,
when it has begun to discharge, produces a very large increase
in the delivery. Thus, in Table XXI. the difference in dis-
charge between experiments at 392 revolutions per minute
and 405 revolutions per minute is 683 gallons per minute, and
this with the small increase of only 13 revolutions.
In the table of the Appold centrifugal pumps the highest
efficiency given is 68 per cent., and in the table of efficiencies
resulting from Mr. Parsons's experiments the highest efficiency
is given at 61.37 per cent., which is practically the same as
that given by Mr. Webber in his reference to the Gwinne
pump, tested in 1883, under 14.7 feet lift.
Experimental Tests of Centrifugal Pumps by Mr.
W. O. Webber.-Mr. Webber's use of the term efficiency
Water H. P.
Indicated H. P.
he explains as indicating the value of
for such
pumps as are driven by an engine direct, and does not, there-
fore, show the full efficiency of the pump, but that of the com-
bined pump and engine. It is, however, a very simple way
of showing the relative values of different kinds of pumping
CENTRIFUGAL PUMPS.
341
engines having their motive-power forming a part of the plant.
Mr. Webber's tests were made with ordinary centrifugal pumps,

m
TUZLAKERRITSA
PETENTNÍ
Ο
FIG. 227.
O
ness.
such as are supplied to the trade in the regular course of busi-
The general elevation of a 5-inch, class B pump is
given in Fig. 227, and in sectional elevation in Fig. 228. In
29*
342
PUMPING MACHINERY.
FIG. 228.
calculating the efficiency of the pump, the cubic feet of water
passing over the weir, measured by the hook-gauge, being
converted into pounds by multiplying by 62.5, is again multi-

plied by the height from the level of water in the tank when
the pump is running to the centre of the discharge-pipe, and
the foot-pound so obtained, divided by 33,000, equals the
water horse-power being developed.
CENTRIFUGAL PUMPS.
343
The power used to do this work is measured by the dyna-
mometer and equals the dynamometer horse-power; the water
horse-power being divided by the dynamometer horse-power
equals the efficiency of the pump being tested; or to formu-
late:
Water H. P.
Dynamometer H. P.
the efficiency of such pumps as are
driven by a belt.
The diagram, Fig. 229, shows the efficiency curves for dif-
ferent velocities, plotted from tests made of two pumps with
70
VELOCIT
FIG. 229.

"SEÇONE
110 *
INCH APERTURE
60
50
WHP
E-
BHP
80
a
20
10
100
Ο
200
1100
17
1000
16-
15+
900
14
13-
12
GALLONS OF
ONS OF WATER PER MINUTE TO IT FEET ELEVATION
900
400
5-inch apertures known on makers' lists as a No. 5, class B
pump, and here illustrated.
These tests were made under an average elevation of 17
feet, the pumps in both cases draughting about 9 feet and dis-
charging 8 feet higher. The upper curve a b was the result
of tests made by a pump that was very clean and smooth inside.
The lower curve c d was made by a pump in which, through
carelessness in the foundry, the core sand had been allowed
to burn into the inside face of volute or casing and water-
passages. The difference between these two curves (which,
by the way, are remarkably uniform) shows the absolute
necessity of having the inside of all such pump-castings very
344
PUMPING MACHINERY.
smooth and free from the slightest roughness. Both of these
pumps were taken at random from stock, and were in no wise
especially prepared for these tests. These tests seem to show
that the efficiency rises very gradually and uniformly until the
water reaches a velocity equal to 11½ feet per second. The
highest efficiency with this size pump seems to equal a
velocity of 12 feet per second, after which point the efficiency
falls very rapidly.
A selection covering a considerable range of efficiencies is
taken from Mr. Webber's table of data, as follows:
TABLE XXII.
TESTS OF CENTRIFUGAL PUMPS (NO. 5, CLASS B)-WEBBER.
Revolutions of pump
Height of lift.
Net load on dynamo
•
•
•
390
410
445 460
16'8" 14'5" 17'10" 17'2"
14,850 18,200 23,200 25,905
29.78 43.35 63.51 87.14
224.4 324.3 475.I 651.8
470
17'2"
28,036
510
17/2"
34,188
585
17'10"
590
17'2"
32,790
40,000
100.42 123.0
111.83 122.37
749.4 920.6
•
•948 1.19
31,294 39,181 70,585 93,458 107,250 136,731
840.4 917.8
124,820 131,295
2.14 2.83
3.25
4.08
2.7
3.78 3.98
2.4
4.0
4.7
5.3
7.4
6.5
8.6
W. H. P
Efficiency.
•
.35
.43
D. H. P.J
.53
.60
.61
.55
.59
.47
Water cubic feet per minute
Water in gallons per minute
Water in foot-pounds
Water horse-power
H. P. for dynamo
In another series of tests undertaken by Mr. Webber the
results seemed to show that the efficiency of centrifugal pumps
increases as the size of the pump increases, and which might
be approximately stated as follows: that a 2-inch pump (this
designation meaning always the size of discharge outlet in
inches of diameter) giving an efficiency of 38 per cent., a 3-
inch pump giving 45 per cent., and a 4-inch pump giving 52
per cent., were giving as good results, proportionally, as a 5-
inch pump at 60 per cent., and a 6-inch pump at 64 per cent.
of efficiency.
Table XXIII. gives a list of sizes of centrifugal pumps as
made by the Lawrence Machine Company, together with the
deliveries at various heights and revolutions per minute.
CENTRIFUGAL PUMPS.
345
TABLE XXIII.
LAWRENCE MACHINE COMPANY.
Centrifugal Pumps for Lifts up to 35 Feet, known on Maker's List as Class "B." Table of Revolutions per
to raise Minimum Quantity of Water to Different Heights with Different Sizes of Pumps.
Minute necessary
SIZE OF PIPES.
Suction.
Discharge.
Economical Capacity
in Gallons per Minute.
Horse-Power per
Foot-Lift
Smaller
tity.
for
Quan-
for
Horse-Power per
Foot - Lift
Larger
tity.
Quan
Diameter of Pul-
ley.
Width of Pulley.
Width of Belt.
2//
1½"
20 to
50
.024
.03
23406∞
22
2
60 to
80
.035
.05
32
3
80 to
160
.055
.08
350
4½
4
160 to
350
.075
.176
330 to
600
.175
.30
12
500 to
900
.22
.378
14
8
8
1,100 to
2,000
.45
.84
ΙΟ
ΙΟ
1,600 to
3,000
.62
I.40 18
55772 40 ∞
7/2
72
ㄓ​ˇˇ
II
II
557NHH
8
9
16 12
IO
215 240
12
IO
170 190
12
12
2,000 to
4,000
1.00
1.90
22
14 12
150 165
346 7∞
6'.
HEIGHT IN FEET AND REVOLUTIONS PER MINUTE.
8'. 10'. 12'. 16'. 20' 25'. 30'. 35'.
3 465 515 560 605| 680 745 820 885 945
425 475 515 560 625 680 750 810 870
390 435 475 510 575 630 695 750 800
365 405 445 475 535 590 645 700 745
320 | 355
355 390 415 470 520 570 610 650
285 315 345 370 415 460 500 540 580
260 280 310 340 375 410 435
|
210 225 250
250 275 300 325 350
185 195 220 240 265
Minimum
Quantity
of Water.
20
60
80
160
330
500
1,100
1,600
285 | 310
2,000
14
14
3,000 to
5,000
1.25
2.10
24
15
15
3,500 to 7,000
1.40
2.94
18
18
6,000 to 11,000
2.40
4.70
2 2
28
28
4∞∞
ниН
15
456
12
14
135
150
165 175
195 215 240 255 275
3,000
14
125 145
155 165
16 14 ΙΙΟ 120 130 135
190
210 231 245 260
3,500
160 | 175
160 175 190 250
| || 220
6,000
24
24
8,500 to 18,000
3.60
7.60
8,500
30
30
12,000 to 25,000
4.95
10.50

12,000
346
PUMPING MACHINERY.
Efficiency test of a 24-inch centrifugal pump at the
sewage station, Montreal. This test, conducted by George
FIG. 230.

ELEVATION.

10
No.2.
DI-H
PLAN.
·Guage
No.3.
No. I.
No. 4.
GUAGE
CRIFICE
Dil-
→→→→→→→ TO MAIN
SEWER.
H. Barrus, was with one of four centrifugal pumps made by
the Lawrence Machine Company. The general arrangement
of the pumps and engines is shown in Fig. 230.
CENTRIFUGAL PUMPS.
347
The pumps are arranged in pairs, two being attached to
one shaft, and two to another shaft. A pair of engines, with
cranks 90 degrees apart, is placed in a central position be-
tween the two pumps on each shaft, the whole plant thus em-
bracing four engines. The line-shaft for the two pumps and
two engines is continuous, and the various sections are joined
together by means of flange-couplings with recessed joints.
Each pump is fitted with a steam-ejector, by means of
which the pump is primed, and the outlet end of the dis-
charge-pipe is provided with a plain flap-valve, made of cast
iron. The opening through the valve is the full size of the
pipe. Each discharge-pipe is 27 inches in diameter, although
the outlet of the pump is only 24 inches in diameter, and the
pipe is made of wrought-iron plates, with riveted joints. As
seen in the sketch, the water, on leaving the pump, is de-
flected from a straight course, a short bend being introduced
connecting the pump to the discharge-pipe. With this
exception, the pipe is straight.
The test was made on No. 3 pump. It embraced an indi-
cator test of the quantity of power used, and a capacity test
of the quantity of sewage discharged.
The pump was driven by the two engines on its shaft, and the
No. 2 pump was allowed to run without being supplied with
sewage. The character of the coupling between the two
engines made it difficult to separate one engine from the
other, and it was impracticable, for this reason, to make the
test on a single engine and pump; but as the object of the
test was to find the quantity of power expended upon the
sewage, the use of the two engines and the supply of power
for the friction of the whole shaft did not prove objectionable.
To determine the power expended upon the sewage, indi-
cator diagrams were taken from each engine when the No. 3
pump was at work in a normal manner, and again when no
sewage was being discharged, the pump running empty. The
difference in the quantities obtained is taken for the desired
result.
348
PUMPING MACHINERY.
To determine the quantity of sewage discharged, an orifice
was provided at the outlet end of the discharge-pipe, and the
necessary data were observed for obtaining the result by cal-
culation. The data used for the purpose were the head of
water back of the orifice, shown by a mercury gauge, and the
known velocity of discharge from an orifice of this kind under
the observed head.
The orifice consisted of a 4-inch iron plate, with an open-
ing 20 inches in diameter, which was introduced in the flange-
joint, between the flap-valve and the main discharge-pipe.
The edges of the orifice were at right angles to the surface
of the plate.
The nature of the sewage, according to a sample dipped up
from the pump-basin, was but little different, in the matter
of consistency, from that of pure water, although it was con-
siderably discolored. All the solid material is removed before
its entrance to the basins. The weight of a gallon of the
sewage was not determined, but for the purpose of calcula-
tion, it has been assumed to be 8.4 pounds.
The corrected indications of the gauges averaged:
Mercury gauge. Actual head measured above centre of orifice
Suction gauge. Suction-lift, including all corrections
The total lift.
Average revolutions of pump per minute, 157.8.
Boiler pressure varied from 75 to 91 pounds per square inch.
TABLE XXIV.
Feet.
•
•
4.41
•
15.59
20.00
POWER TESTS.
No. 3 Pump
in Operation.
No. 3 Pump
Running Empty.
I. Mean effective pressure, right-hand engine,
in pounds . .
2. Ditto left-hand engine, in pounds.
3. Ditto both cylinders, average, in pounds.
•
•
20.52
7.41
19.50
1.16
20.01
•
3.12
4. Revolutions per minute
•
5. Indicated horse-power
•
•
157 40
161.0
126.18
20.16
The difference between the results of the two tests, which is 126.18—20.16
=106.02, is the quantity of power expended upon the sewage.
CENTRIFUGAL PUMPS.
349
The quantity of sewage discharged through the orifice, ex-
pressed in cubic feet per second, equals V, the velocity of the
current in feet per second, multiplied by a, the area of the
orifice in square feet, multiplied by c, the coefficient of dis-
charge. The first quantity in this equation, V, is determined
by the formula, V=8×
I
h
'a \2, in which h is the head
(a)²,
in feet, measured from the centre of the orifice, a the area of
orifice, and A the area of the discharge-pipe. The quantity
c, for orifices giving a perfect contraction, is, according to the
best authorities, 0.6. The contraction in this case was not
perfect, and the coefficient is assumed to be 0.65.
•
Substituting in the formula for velocity, h=4.41 feet, a =
area of a circle 20 inches in diameter 314.16, and A, area of
a circle 27 inches in diameter 572.55, the velocity works
out 20.8 feet per second. The quantity discharged per second
is, therefore, area of orifice (2.18 square feet) X velocity (20.8)
X coefficient of discharge (0.65) = 28.45 cubic feet per second.
This corresponds to 212.8 gallons per second, or 12,768 gal-
lons per minute.
The quantity of work performed, which is represented by
the raising of this quantity of sewage through the range of
lift noted,-viz., 20 feet,-is 12,768 gallons X 8.4 (pounds per
gallon) 20 2,145,024 foot-pounds per minute. Reduced to
2,145,024
horse-power, this becomes.
65.01 horse-power.
33,000
The efficiency of the pump can now be calculated; that is,
the relation borne by what may be called the useful work of
the pump to the power expended upon it by the motor. The
useful work, as noted, is 65.01 horse-power, and the horse-
power expended by the engine is 106.02. The percentage of
65.01
efficiency, is, therefore, × 100=61.3 per cent.
106.02
Efficiency of Large Centrifugal Pumps.-There is
published in Engineering, Vol. LII., an abstract of a report by
!
30
350
PUMPING MACHINERY.
Mr. A. Elink Sterk, Haarlem, on the efficiency of centrifu-
gal pumps tested by him, October, 23, 1891, for the “Groote
Ypolder," Holland. These pumps were constructed by Messrs.
J. & H. Gwinne, London, but the dimensions are not given in
the report, only the record of efficiency, which averages as
follows:
Water horse-power.
Coal consumption per hour per water horse-power
Steam per pound of coal
•
Steam per water horse power per hour.
Efficiency
W. H. P.
I. H. P.
104.8
.
3.89
7.59
29.7
0.656
The test lasted 4 hours and 2 minutes, the pump making
in all 25,754 revolutions, or an average of 106.5 per minute.
The steam-pressure varied from 93 to 96½ pounds. The
mean discharge of water in cubic metres per minute was
116.7. The mean water horse-power was 104.8, the indicated
horse-power being 159.8; therefore, 104.8 ÷ 159.8 = 0.656
efficiency.
The coal consumption per hour and per water horse-power
was, according to contract, to be 4.97 pounds; but it was
actually found to be 3.89 pounds, or 1.075 pounds less than
the contract figure.
The efficiency, or the relation of water horse-power and
indicated horse-power, was for the whole duration of the trial
a mean of 0.656, but, as appears now, the engine did not work
during the whole trial at the most advantageous speed. If it
had run at its best speed the efficiency would certainly not
have fallen below 0.67, or fully 2 per cent. higher.
This high efficiency, Mr. Sterk says, is higher than that of
any centrifugal pumping engine known to him. The next in
this respect are the large centrifugal pumps at Hatakbeh, in
Egypt, made by Farcot, at Paris, each of which is capable
of raising 400 cubic metres per minute.. The efficiency of
these pumps was, according to trials made by Mr. Brühl,
0.651, which figure Mr. Sterk has no hesitation in saying is
flattered.
CENTRIFUGAL PUMPS.
351
Morris Machine-Works Centrifugal Pump.—A sec-
tional elevation of a centrifugal pump by the Morris Machine-
Works, Baldwinsville, N. Y., is shown in
Fig. 231. This pump has a central suction-
opening, the revolving disk is furnished.

FIG. 231.
A
with curved blades for imparting to the water a proper direc-
tion and velocity into the spiral chamber which surrounds
FIG. 232.
FIG. 233.


the circumference of the revolving blades. The makers have
adopted this design in common with other first-class builders,
352
PUMPING MACHINERY.
as it was experimentally shown long ago that spiral castings
closely surrounding the fan gave the best results.
Several designs for pistons, as these revolving blades are
sometimes called, have been tested in actual service, from
which three have been selected for illustration.
Fig. 232 represents a hollow arm piston, which is used in
their standard pumps in size No. 4 and above. It is the one
on which the fame of the Heald & Sisco pump is mainly
FIG. 234.
CONANT COBCSICH
based, and is their special favor-
ite for raising water or any thin
fluid not too much encumbered
with stringy or tenacious matter.
For over a quarter of a century
it has held the lead against all

comers.
Fig. 233 represents a concave
arm wing, which they use in
their No. 3 standard pump and
smaller sizes. It has proved
itself very efficient. For raising
half stuff in paper-mills and
stringy material often found in
tan liquor, they recommend the
wing as being the best.
pumping very thick material
they use a wing with two arms
only.
In
The piston shown in Fig. 234
is of their own invention, and
has been used upwards of six years in their special sand-
and dredging-pumps. It is very heavy, has large openings,
and is very efficient. By its use material taken in it is de-
posited on pump-scroll, thereby preventing wear on the pump-
sides. All sand- and dredging-pumps of their make are fitted
with it.
CENTRIFUGAL PUMPS.
353
TABLE XXVI.
MORRIS MACHINE-WORKS-STANDARD HORIZONTAL CENTRIFUGAL PUMP.
Actual Ca-
Economical Ca-
No.
pacity in Gallons,
per Minute.
pacity in
Gallons, per
Minute.
Horse-Power
required for
each Foot
of Lift.
Minimum
Quautity.
Diameter and
Face of
Pulleys, in
Inches.
Floor Space
required,
in Inches.
HHN
1/2
20 to 40
160
.OI
X
5
19 X 30
134
40 to 60
225
.016
6
21 X 33
2
60 to 80
325
.019
2½
80 to 100
400
.039
7 X 6
3456∞
120 to 180
675
.047
7 X
669
23 X 37
24 X 38
7
25 X 39
200 to 300
1,300
.078
8 X 10
30 X 40
350 to 500
1,900
.14
500 to 700
2,700
.22
900 to 1,300
4,800
.34
IO
1,600 to 2,200
7,500
.64
12
15
18
22
250 2
2,000 to 3,000
10,500
.88
+248
IO X IO
34 X 54
12 X 12
37 X 55
18 X 12
45 X 63
20 X 12
51 X 71
24 X 14
62 X 78
3,000 to 5,000
16,500
I.20
30 X 16
77 X 80
5,000 to 7,000
22,000
1.80
36 X 18
93 X 103
7,000 to 10,000
35,000
2.90
48 X 18
126 X 130
The number of the pump is also the diameter of the dis-
charge opening in inches. Where more than 25 feet of dis-
charge-pipe is attached to the pump, they recommend using
one or two sizes larger than the pump-discharge.
The horizontal right-hand pump is the one preferred by
manufacturers generally. To use it without a priming appa-
ratus the water must flow to the pump and rise as high as
the top of the shell. They are the favorite for drainage and
irrigation work, and when so used are generally fitted with a
foot-valve. In point of execution there is no difference be-
tween the horizontal and vertical pump. The former has the
advantage of being more readily examined in case of acci-
dent. All of their pumps are made to run right-handed,
unless otherwise ordered. In their horizontal pumps the
discharge, as usually made, will be in a vertical position, as
shown in Fig. 235, unless another delivery is desired.
Double suction-pumps are to be preferred to standard or
side suction-pumps, where there is a suction-lift of several feet;
-!
1
x
30*
354
PUMPING MACHINERY.
but they can be recommended only for raising clear water,
whereas the standard pumps will pass anything that will enter
the suction.
Combined Pump and Engine.-When a pump cannot
be conveniently driven by a belt on account of the situation
of shafting, distance from power, etc., a combined pump and
engine may be quite satisfactorily employed, especially for the
larger sizes. Such a combination is shown in Fig. 235, which
is not only very compact, but economical in the use of steam,
making them especially desirable for use on shipboard, circu-
lating pumps for ice-factories, etc.
These
For dredging operations a centrifugal pump fitted with
a piston as shown in Fig. 234, has been found to give a high
efficiency, the makers claiming that it will wear longer and
cost less for repairs than the forms heretofore in use.
pistons are very heavy, twice as much as in ordinary pumps,
and are fitted with a water-bearing that will prevent the sand
from cutting out the journal and the shaft, thereby prolonging
the life of the pump at least double over one fitted with a
piston similar to Fig. 233; the latter being good for water
pumping, has proven itself extravagant in the use of power,
but wanting in durability on account of the rapid wear, because
the sand is dragged at all times between it and the pump-shell,
which increases the clearance between the shell and the wing,
quickly impairing its efficiency.
In this piston the sand is taken in at the centre and is
deposited in the pump-shell, which has a clearance at the
periphery of the same equal to the diameter of the pump's
discharge. When the pump is in operation this is stated by
the makers to form a water-cushion equal at all points on the
outer edge of the piston, which is not found in any other
pump.
It is said that sand, mud, gravel, or marl can be dredged
with a centrifugal pump at a much less cost than with the old-
style dipper- or scoop-dredge, for the following reasons:
CENTRIFUGAL PUMPS.
355
I. The plant costs about one-twentieth as much as a dipper-
dredge.
2. Less help required to run it, two or three men being all
that are necessary.
3. Material that will pass through
raised much more
pumps can be raised much
rapidly than with a dipper or scoop.

MORRIS MACHINE WORKS
BALDWINSVILLE NY.
FIG. 235.
10
– p
4. That a pump-dredge can work in weather and sea where
a dipper-dredge could not live.
5. That a centrifugal plant is not liable to the many break-
ages that occur in the dipper machines.
6. When working in rivers or bays, material can be dredged
and piped ashore at one operation.
356
PUMPING MACHINERY.
འ
TABLE XXVII.
SPEED AT WHICH MORRIS MACHINE-WORKS CENTRIFUGAL PUMPS SHOULD RUN TO RAISE WATER TO DIFFERENT HEIGHTS.
Number.
Feet. Feet.
5.
Feet. Feet. Feet.
Feet.
Feet.
Feet.
Feet.
Feet.
Feet.
Feet.
Feet.
Feet.
IO.
15.
20.
25.
30.
35.
40.
50.
60.
70.
80.
go.
100.
1/2
426 603
740
850
955
1046
1130
1211 1350
1478
1597
1705
1812
1995
134
341 484
592 683
765
837
905
977
1081
1216
1282
1368
1451
1530
2
323 460 564 650 727 796
886
920
1028
1127
1217
1302
1381
1456
સ
21/2 314 446 546
630
705
772
835
892
997
1092
1180
1262
1337
1411
3
300 426
521
602 660
737
796
852
952
1043
1127
1205
1277
1322
4
+
283 398
480
557 621
693
752
808
895
992
1054
1138
1220
1275
5
250
357
437 505 565
619
692
715
800
876
946
1012
1073
1128
6
215 300 370 426 485
522
562
609
675
745
790
860
940
965
8
182 258 316 365
408
447
483
518
577
633
686
730
776
835
10
168 237 291 336
375
4II
443
475
531
582
626
671
712
751
12
133 189 231 267
300
327
353
378
423
463
501
536
568
600
15
108 154
189 216 242
266
287
307
343
377
407
436
462
487
18
92 135
160
180 203
231
246
277
304
323
352
374
400
413
22
80 112 139
160
178
197
212
227
253
277
300
321
340
358
}
CENTRIFUGAL PUMPS.
357
The application of a centrifugal pump to dredging
operations was tested by General Q. A. Gillmore, U. S. A.,
in deepening the channel over the bar at the mouth of the St.
John's River, Florida. A No. 9 Andrews centrifugal drainage-
pump was used on the deck of a small steamer. This pump
had suction- and discharge-openings 9 inches in diameter.
To the suction-opening there are connected by a 2-way
'branch-pipe two 6-inch suction-pipes, instead of one 9-inch as
usual, the object being not only to work on both sides of the
boat simultaneously, but to render the necessary handling of
the pipes as easy and prompt as possible. There is, on the
other hand, considerable disadvantage in operating with two
suction-pipes instead of one, on account of the greater amount
of friction for an equivalent suction-capacity; for while a 9-
inch pipe has an area of 81 circular inches, two 6-inch pipes
have an aggregate area of only 72 circular inches. The fric-
tional surface is therefore increased as 27 to 36, making the
disadvantage from or loss by friction from this cause as
2 to 3.
As a partial compensation for this increased amount of fric-
tion, an increased velocity is given to water in suction-pipes
of less aggregate area than the discharge-pipe, and a larger
proportion of sand is thereby carried up.
It was necessary also to encounter another disadvantage by
using several bends, of which there were two in each of the
suction-pipes, and one in the discharge-pipe, those in the
suction-pipe being each one-eighth of a circle, and that in the
discharge-pipe one-fourth of a circle. These bends reduce
the delivery at the rate of 10 per cent. for each turn of 90
degrees, and about 6 per cent. for each turn of 45 degrees,
the reductions in each case being calculated upon the quantity
passing the preceding bend.
Thus, the first one-eighth bend in the suction reduces the
quantity to 94 per cent., the second to 88 per cent., and the
one-fourth bend in the discharge to 79 per cent. The disad-
vantages, therefore, under which the apparatus labored may
be briefly summed up as follows:
358
PUMPING MACHINERY.
of
I. The loss by friction, due to the use of two 6-inch instead
one 9-inch suction-pipe, is increased 50 per cent.
2. The unestimated loss by friction, due to the use of suc-
tion-pipes three times as long as the height to which the
material is to be raised.
3. The loss of 21 per cent. by bends in the suction- and
discharge-pipes.
Although 200 revolutions of the pump-disk per minute will
easily raise 3000 gallons of clear water 12 feet high through
a straight vertical 9-inch pipe, 300 revolutions are required to
raise 2500 gallons of sand and water II feet high, through
the two inclined suction-pipes having two turns each, dis-
charged through a pipe having one turn. To prevent the ends
of the suction-pipes being lifted off the bottom by the pitch-
ing of the boat, and as a precaution against accident, a portion
of each pipe is made flexible, being composed of 6-inch rub-
ber hose stretched over a coil of wire. In addition, the ends
are loaded with an iron frame or drag, each weighing about
250 pounds, which is intended to move flat along the bottom
during the operation of dredging. To the under surface of
this frame, directly below the mouth of the pipe, a number of
teeth or knives are attached to stir up the sand and aid its
entrance into the pipes. A chain attached to each drag, and
leading to the deck of the steamer on either side, takes the
strain from the pipe when the drag is down and the steamer
in motion.
The proportion of sand that can be pumped de-
pends greatly upon its specific gravity and fineness. The
calcareous and argillaceous sands flow more freely than the
silicious, and fine sands are less liable to choke the pipe than
those that are coarse. When working at high speed, 50 to 55
per cent. of sand can easily be raised through a straight ver-
tical pipe, giving for every 10 cubic yards of material dis-
charged 5 to 5½ cubic yards of compact sand. With the
appliances used on the St. John's bar the proportion of sand
seldom exceeded 45 per cent., generally ranging from 30 to
CENTRIFUGAL PUMPS.
359
35 per cent. when working under the most favorable con-
ditions.
In pumping 2500 gallons, or 12.6 cubic yards, of sand and
water per minute, we would therefore get from 3.7 to 4.3 cubic
yards of sand. During the early stages of the work, before
the teeth under the drag had been properly arranged to aid
the flow of sand into the pipes, the yield was considerably
below this average, not often exceeding, and frequently falling
below, 2 cubic yards of sand per minute during the time.
actually occupied in pumping.
With a centrifugal drainage-pump sand can be easily dis-
charged at a height of 30 feet above the level of the water; and
when the distance to which it has to be conveyed is so great that
open troughs from the discharge-pipe to the dumping-ground
cannot have sufficient inclination to secure a free flow of the
sand and water, it would be necessary to make the discharge
through pipes, increasing the power expended in proportion
to their length, so as to insure a velocity that will transport
the sand and prevent choking. The pump itself should, in
all cases, be placed as low as possible, and it would generally
be practicable to locate it from 3 to 5 feet above the surface
of the water.
The cost of dredging with a 9-inch pump would prob-
ably not exceed 10 to 11 cents per cubic yard, inclusive of
running expenses, wear and tear of machinery, and all stop-
pages for repairs and other contingencies. Indeed, assuming
the pump on St. John's Bar to have worked continuously in
raising sand 10 hours per day, except Sundays, with the same
average results per hour actually attained while pumping,
thus charging the six working-days of each week with the
expense of seven, the cost of raising the sand into the bins
would have been only 8 cents per cubic yard, and if it could,
at the same time, have also been continuously discharged to
the dumping-ground, through either open troughs or pipes, no
additional expense, except a trifle for increased power, would
have been incurred. There were, moreover, constant losses
:
360
PUMPING MACHINERY.
encountered on the bar while actually pumping, which would
not occur in still water, and of which no account has been
taken, due to the pitching of the boat, which frequently lifted
the ends of the suction-pipes from the bottom. It is there-
fore considered safe to estimate the cost of removing sand at
IO to II cents per cubic yard, when the conditions are such
that the work of raising the sand and discharging to the
dumping-ground can be carried on simultaneously and con-
tinuously.
Centrifugal Pumps in Series.-Such a method of
pumping water is not often carried out, but a recent example
of such an application was shown in an instance where it was
necessary to raise water 150 feet, the only appliances available
for this purpose being a pair of centrifugal pumps, either of
which was not equal to raising the water to a height above 75
feet. It should be remembered that the operation of a cen-
trifugal pump is due entirely to the centrifugal force of the
water as it is revolving in the pump-case, but the water was
raised to the required height by attaching the two pumps in
series, the second one adding the necessary increment to the
pressure which was received, and working in that way very
satisfactorily.
The reader is referred to an interesting and valuable paper
on the "Irrigating Machinery of the Pacific Coast," by Mr.
John Richards, published in the Proceedings of the Institution
of Mechanical Engineers, London, 1888. This paper was also
published in Engineering, Vol. XLIV. The paper is fully
illustrated, and the writer only regrets that want of space
prevents its reproduction here, as Mr. Richards has kindly
placed it at his disposal.
DUTY-TRIALS OF PUMPING ENGINES.
361
CHAPTER XVII.
DUTY-TRIALS OF PUMPING ENGINES.
THE duty of a pumping engine is a term used to denote the
number of pounds of water lifted I foot high by the combustion
of 100 pounds of coal. This is a convenient method by which
to compare the results of pumping engines one with another,
because a measure of duty in so many millions of foot pounds
of work done fixes at once the relative position which such an
engine occupies when compared with other pumping engines.
The method of calculation usually employed in ascertaining
the duty of a pumping engine may be expressed in the fol-
lowing formula:
A X PX VX 100
C
Duty.
Where A = area of plunger, in square inches.
P-pressure in pounds per square inch.
V= total plunger-travel during the trial, in feet.
C=coal burnt during the trial, in pounds.
Example based on a 10 hours' trial:
Then
A =201 square inches, area of plunger.
P=80 pounds pressure, including the suction-lift.
V=66,000 feet of plunger-travel in 10 hours.
C= 1600 pounds of coal burnt in 10 hours.
201 X 80 X 66,000 X 100
1600
66,330,000 foot-pounds of duty.
Another method of determining the duty of the same
pumping engine may be expressed thus:
GX W X H X 100
C
Duty.
Where G = gallons of water actually delivered.
W= weight of water per gallon.
H= height in feet to which the water is pumped.
C = coal burnt during the trial, in pounds.
Q
31
362
PUMPING MACHINERY.
Example based on a 10 hours' trial:
Then
G= 689,040 gallons in 10 hours.
W=8.33 pounds per gallon.
H=184 feet height to which the water is pumped.
C= 1600 pounds of coal burnt in 10 hours.
689,040 X 8.33 × 184 × 100
1600
=66,006,587 foot-pounds of duty.
A high-duty pumping engine is understood to be one
in which nearly or more than 100,000,000 foot-pounds of
work are accomplished with an expenditure of 100 pounds of
coal burnt in the furnace. This definition, it will be seen,
couples the boiler and pumping engine as one machine, the
coal being charged against one end of it, and the foot-pounds
of work done are credited at the other; and so far as the
average water-works management goes, it is about all they
care to know. To the engineer there are, however, interme-
diate problems which are not only interesting in themselves,
but useful to know, that a correct estimate may be reached
regarding the efficiency of the plant in its several details.
The boiler part of the pumping plant should always be tested
separately from that of the pumping engine. In duty-trials
it has been customary to assume the evaporation of 10 pounds.
of water from and at 212° Fahr. per pound of coal, without
any allowance for ashes and clinkers. This assumption is not
always realized in practice, and a deduction of from 10 to 20
per cent. must be made to have it accord with ordinary boiler
evaporation. This fictitious standard of boiler performance
probably originated in a divided responsibility, one contractor
furnishing the boilers and another the pumping engine.
Some contracts have been made to read 1000 pounds of dry
steam delivered to the engines, rather than the more usual
one of 100 pounds of coal coupled with a boiler performance
of 10 pounds of water evaporated per pound of coal.
The boiler-test should be conducted independently of the
engine, because a low efficiency is more likely to occur there
than in the engine; the ordinary performance of the latter being
such that calculations may be made in advance with close ap-
DUTY-TRIALS OF PUMPING ENGINES.
363
proximation to experimental results. There are many things
which affect boiler performance: the boiler may not be prop-
erly proportioned, the circulation in it not good, the furnace
may not be the best for the fuel to be used, the draft may be
deficient on the one hand, and too much air may be admitted
for the weight of fuel burnt on the other. The heating-power
of coal must of necessity be variable, because no two mines
yield exactly the same quality of coal, to say nothing of such
wide divergencies as exist between the properties of anthracite
and bituminous coals; it is clear, therefore, that the coal stand-
ard is at best a very uncertain one in estimating the efficiency
of a pumping-plant.
The Duty-Trial Committee in their report to the American
Society of Mechanical Engineers, published in Vol. XI., Trans-
actions, recommend that in all reports on duty-trials of pump-
ing engines, the existing unit of 100 pounds of coal be abol-
ished, and that in its place a new basis of 1,000,000 heat-units
be established. This new unit will not seriously affect com-
parison with the one abolished, because it is expected that I
pound of coal will yield at least 10,000 heat-units; therefore
10,000 10.35 pounds of water evaporated from and at 212°
966
Fahr., the denominator 966 being the latent heat of evapora-
tion.
The final results of a boiler-trial, for the sake of uni-
formity and comparison with other trials should, as far as
practicable, be tabulated in the form recommended by the
American Society of Mechanical Engineers, Transactions, Vol.
XI., which is here reproduced.
•
I. Date of trial
2. Duration of trial
Hours.
3. Grate surface
DIMENSIONS AND PROPORTIONS.
wide
long
Area
•
4. Water-heating surface
5. Superheating surface .
•
6. Ratio of water-heating surface to grate surface
(Give also complete description of boilers.)
•
Square feet.
66
364
PUMPING MACHINERY.
•
AVERAGE PRESSURES.
7. Steam-pressure in boiler by gauge
8. Atmospheric pressure by barometer
9. Force of draught in inches of water
10. Of steam
11. Of escaping gases
12. Of feed water
AVERAGE TEMPERATURES.
13. Total amount of coal consumed*
14. Moisture in coal
FUEL.
Pounds.
Inches.
•
Degrees.
66
Pounds.
Per cent.
Pounds.
•
15. Dry coal consumed
16. Total refuse (dry)
•
17. Total combustible (dry weight of coal, item 15, less refuse, item 16)
18. Dry coal consumed per hour
RESULTS OF CALORIMETRIC TEST.
19 Quality of steam, dry steam being taken as unity.
20. Percentage of moisture in steam.
•
•
21. Number of degrees superheated
•
•
•
Per cent.
Degrees.
WATER.
22. Total weight of water pumped into boiler and apparently evapo-
rated †
23. Water actually evaporated corrected for quality of steam
24. Equivalent water evaporated into dry steam from and at 212°
Fahr.‡
•
Pounds.
25. Equivalent total heat derived from fuel, in British Thermal Units B.T.U.
26. Equivalent water evaporated into dry steam from and at 212°
Fahr. per hour
ECONOMIC EVAPORATION.
Pounds.
27. Water actually evaporated per pound of dry coal from actual
pressure and temperature.
28. Equivalent water evaporated per pound of dry coal from and at
212° Fahr.
Pounds.
•
* Including equivalent of wood used in lighting fire. One pound of wood
equals 0.4 of a pound of coal, not including unburned coal withdrawn from fire
at end of test.
† Corrected for inequality of water-level and of steam-pressure at beginning
and end of test.
Factor of evaporation
H-h
965.7
H and h being respectively the total heat-
units in steam of the average observed pressure, and in water of the average ob-
served temperature of feed.
DUTY-TRIALS OF PUMPING ENGINES.
365
29. Equivalent water evaporated per pound of combustible from and
at 212° Fahr.
30. Number of pounds of coal required to supply 1,000,000 British
Thermal Units
RATE OF COMBUSTION.
31. Dry coal actually burned per square foot of grate surface per hour
RATE OF EVAPORATION.
Pounds.
•
Pounds.
32. Water evaporated from and at 212° Fahr. per square foot of heat-
ing surface per hour.
Pounds.
[Note.-To determine the percentage of surface-moisture in the coal a sample
of the coal should be dried for a period of twenty-four hours, being subjected to
a temperature of not more than 212°. The quantity of unconsumed coal con-
tained in the refuse withdrawn from the furnace and ash-pit at the end of the test
may be found by sifting either the whole of the refuse or a sample of the same, in
a screen having 3%-inch meshes. This, deducted from the weight of dry coal fired,
gives the weight of dry coal consumed, for line 15.-Duty-Trial Committee.]
The pumping engine duty-trial should be conducted
on the lines proposed by the American Society of Mechanical
Engineers, previously referred to.
The heat-unit basis of computing duty is proposed by
the above society, and to make the computation from the
quantity of heat supplied to the complete plant; using not
only that supplied to the engine-cylinders, but that supplied to
all the accessory parts of the engine, such as the steam-jack-
ets, the donkey feed-pump, the independent air-pump, if this
be driven with steam, and any other apparatus using steam
which is necessary to the operation of the engine. It is
recommended that the scope of the test be made so broad
that, for the sake of completeness, the quantity of steam
which passes through the cylinders of the engine be deter-
mined independently of that used for other purposes, and like-
wise, that the quantity of steam used by each accessory part
of the engine be also determined. In contract tests, if a steam-
pump be used for the boiler feed-pump, the quantity of heat.
supplied for operating this apparatus is to be included in the
31*
366
PUMPING MACHINERY.
total quantity, not only in cases where both boiler and engine
are supplied by one party, but also where the boiler is fur-
nished by a separate contractor.
The heat-unit method requires that the actual total heat of
the steam shall be known, and for this purpose allowance will
necessarily be made for any moisture or superheat contained
by the steam furnished to the engine.
Measuring the Work done.-In determining upon a
suitable method of measuring the amount of work done, which
involves a measure of the quantity of water discharged into
the force main, the committee have endeavored to find one
which may be employed universally, and which may, in a
reasonable manner, serve the ends of the builder, purchaser,
and all interested parties.
It is proposed that the plunger-displacement system of
measurement be employed, and that the purchaser's interest
be protected by the determination of the amount of slip in
the pump, so far as slip is produced by leakage of the plunger,
which is probably its main factor, and leakage of valves, if
this occurs through faulty design. A satisfactory determina-
tion of the approximate extent to which leakage occurs does
not present serious difficulty.
In deciding upon plunger displacement, accompanied by a
determination of the leakage, as the best mode of measure-
ment for the purposes in view, the committee does not for a
moment underrate the importance and desirability of measure-
ment by weir, tube, or nozzle, whenever either of these can be
employed to advantage. It is strongly recommended that
these additional measurements be undertaken in all cases
where it is practicable to do so, that the results of the test may
be supplemented by the additional data thus obtained.
Quantity of Work done.-In determining the quantity
of work done by the pump, the committee recommends that
the work of overcoming the friction of the water in passing
through the passages and valves in the pump should not be
DUTY-TRIALS OF PUMPING ENGINES.
367
included in the desired total; but that the work expended in
friction of both the force- and suction-mains be included in
that on which the duty is computed. It is held that the
efficiency of the engine should not be made dependent upon
any condition which is foreign to itself, and that the builder
of the engine should be held responsible only for the work
done from the time when the water enters the pump to the
time when it leaves it. The purchaser, it should be observed,
should guard his interest in the matter by having the mains
furnished of such capacity as to reduce their friction to a
minimum.
To carry out these provisions, the data to be determined,
apart from that relating to the plunger displacement, are the
indication of a pressure-gauge attached to the force-main,
that of a vacuum-gauge attached to the suction-main, and the
vertical distance between the centres of the two gauges.
It is recommended that no air be allowed to enter the pump-
cylinders during the progress of the test, thereby removing
all possibility of imperfect filling. If it is necessary, in special
cases, for air to be "snifted in," this should be regarded as a
defect in the action of the pump, which should be noted by
the expert in his report, and such allowance should be made
for the imperfect filling, due to the presence of air, as may be
determined upon by an examination of the indicator diagrams.
taken from the pump-cylinders, or from other data which may
be secured.
Engine Performance.-The necessary data having been
obtained in accordance with these recommendations, the duty
of an engine, and other quantities relating to its performance,
may be computed by the use of the following formulæ :
1. Duty
Foot-pounds of work done
Total number of heat-units consumed
A (P±p+s) X L X N
H
X 1,000,000
X 1,000,000 (foot-pounds).
CX 144
2. Percentage of leakage
X 100 (per cent).
A X L X N
368
PUMPING MACHINERY.
3. Capacity
= number of gallons of water discharged in 24 hours
A X L X N X 7.4805 X 24
DX 144
A X L X NX 1.24675
D
(gallons).
4. Percentage of total frictions
-[
I.H.P.
A (P+p+s) X L X N
DX 60 X 33,000
I.H.P.
X 100
]
= [1 - A (P ± p + s ) X L X N
A (P + + s X L X N ]
= [1–
As × M.E.P. X Ls X Ns
X 100 (per cent.);
or, in the usual case, where the length of the stroke and number of strokes of
the plunger are the same as that of the steam piston, this last formula becomes :
Percentage of total frictions
= [1 -
A (P±p+s)
As X M.E.P.
X
2] × 100 (per cent.).
In these formulæ the letters refer to the following quantities:
A = Area, in square inches, of pump-plunger or piston, corrected for area of
piston-rod. (When one rod is used at one end only the correction is one-
half the area of the rod. If there is more than one rod the correction
is multiplied accordingly.)
P = Pressure, in pounds per square inch, indicated by the gauge on the force-
main.
Pressure, in pounds per square inch, corresponding to indication of the
vacuum-gauge on suction-main (or pressure-gauge if the suction-pipe is
under a head). The indication of the vacuum-gauge, in inches of
mercury, may be converted into pounds by dividing it by 2.035.
Pressure, in pounds per square inch, corresponding to distance between the
centres of the two gauges. The computation for this pressure is made by
multiplying the distance, expressed in feet, by the weight of one cubic
foot of water at the temperature of the pump-well, and dividing the
product by 144, or by multiplying the distance in feet by the appropriate
quantity found in the following table. The quantities in this table are
computed from the weights of one cubic foot of water at the various
temperatures, as given by D. K. Clark in his "Rules and Tables,"
which also correspond to Charles T. Porter's figures in his work on the
"Richards's Steam-Engine Indicator.”
DUTY-TRIALS OF PUMPING ENGINES.
369
:
:
Temperature of
Water in
Pump-Well.
Weight of 1 Cubic
I
Foot of Water
divided by 144.
Temperature of
Water in
Pump-Well.
I
Weight of 1 Cubic
Foot of Water
divided by 144.
Degrees Fahrenheit.
Degrees Fahrenheit.
45
3 m of one 50
32
.4335
75
.4325
35
.4335
80
.4322
40
.4335
85
.4319
.4334
90
.4315
50
.4333
95
.43II
55
.4332
100
.4307
60
.433I
105
.4303
65
.4329
IIO
.4298
70
.4327
L
Average length of stroke of pump-plunger in feet.
N= Total number of single strokes of pump-plunger made during the trial.
As Area of steam-cylinder, in square inches, corrected for area of piston rod.
The quantity, As × M.E.P., in an engine having more than one cylinder,
is the sum of the various quantities relating to the respective cylinders.
Ls = Average length of stroke of steam-piston, in feet.
N's
Total number of single strokes of steam-piston during trial.
M.E.P. Average mean effective pressure in pounds per square inch, measured
from the indicator diagrams taken from the steam-cylinder.
I.H.P.
Indicated horse-power developed by the steam-cylinder.
C = Total number of cubic feet of water which leaked by the pump-plunger
during the trial, estimated from the results of the leakage test.
D= Duration of trial, in hours.
H = Total number of heat-units [B. T.U.] consumed by engine weight of
water supplied to boiler by main feed-pump X total heat of steam of
boiler-pressure reckoned from temperature of main feed-water + weight
of water supplied by jacket-pump X total heat of steam of boiler-pressure
reckoned from temperature of jacket-water + weight of any other water
supplied total heat of steam reckoned from its temperature of supply.
The total heat of the steam is corrected for the moisture or superheat
which the steam may contain. For moisture, the correction is subtracted,
and is found by multiplying the latent heat of the steam by the percent-
age of moisture, and dividing the product by 100. For superheat, the
correction is added, and is found by multiplying the number of degrees
of superheating (i.e., the excess of the temperature of the steam above
the normal temperature of saturated steam) by 0.48. No allowance is
made for heat added to the feed-water, which is derived from any source,
except the engine or some accessory of the engine. Heat added to the
water by the use of a flue heater at the boiler is not to be deducted.
Should heat be abstracted from the flue by means of a steam reheater con-
nected with the intermediate receiver of the engine, this heat must be in-
cluded in the total quantity supplied by the boiler.
y
370
PUMPING MACHINERY.
The total and latent heats may be found by reference to the Tables of the Prop
erties of Saturated Steam, given in Charles T. Porter's treatise on the "Richards's
Steam-Engine Indicator."
The two examples following, (1) compound direct-acting
duplex pumping engine, and (2) compound crank and fly-
wheel pumping engine, are given to illustrate the method of
computation. The figures are not obtained from tests actually
made, but they correspond in round numbers with those
which were so obtained.
Compound Tandem Direct-Acting Duplex Pump-
ing Engine.—Both high-pressure and low-pressure cylinders
jacketed with live steam. Jet condenser used, with air-pump
driven by main engine. Boiler feed-pump also driven by
main engine. Jacket-water returned to boiler by gravity.
Main supply of feed-water drawn from hot well.
DIMENSIONS.
Diameter of each high-pressure cylinder (two)
•
Diameter of each low-pressure cylinder (two)
Diameter of piston-rod, each cylinder (one at each end high-
pressure, two at one end low-pressure)
Diameter of pump-plungers (two)
•
Diameter of piston-rod, each plunger (one at one end)
Nominal stroke
15 inches.
30
(6
3.5
15
3.5
(6
18
(6
GENERAL DATA.
1. Duration of test (D)
2. Boiler - pressure by gauge (barometric - pressure, 14.7
pounds)
3. Temperature of water in pump-well
•
•
4. Temperature of main supply of feed-water
5. Temperature of jacket-water.
•
6. Percentage of moisture in steam
•
7. Weight of water supplied to boiler by main feed-pump.
8. Weight of water supplied to boiler by jackets.
DATA RELATING TO WORK OF PUMP.
9. Area of plunger minus ½ area of rod (A)
10. Average length of stroke (L and Ls)
11. Total number of single strokes during trial (N and Ns)
12. Pressure by gauge on force-main (P)
12 hours.
120 pounds.
80 deg.
100
280
o per ct.
•
22,400 pounds.
2560
66
171.9 sq. in.
1.572 feet.
76,000
100 pounds.
DUTY-TRIALS OF PUMPING ENGINES.
37I
13. Vacuum by gauge on suction-main
14. Pressure corresponding to vacuum given in preceding
line (p)
•
15. Vertical distance between gauges
•
16. Pressure corresponding to distance given in preceding
line (s). .
17. Volume of water which leaked through the plungers
computed from results of leakage test (C)
•
DATA RELATING TO WORK OF STEAM-CYLINDERS.
18. Area of high-pressure piston minus area of one rod (As) .
19. Mean effective pressure high-pressure cylinder (M.E.P.,)
20. Area of low-pressure piston minus ½ area of two rods
(As₂)
•
21. Mean effective pressure low-pressure cylinder (M.E.P.₂)
22. Number of double strokes each side per minute
23. Indicated horse-power developed by steam-cylinders
24. Feed-water consumed per indicated horse-power per
hour.
25. Heat-units consumed per indicated horse-power per
9.3 inches.
4.57 pounds.
8 feet.
3.46 pounds.
5900 cu. ft.
167.09 sq. in.
61.31 pounds.
697.24 sq. in.
13.72 pounds.
26.39
99.61 I.H.P.
20.88 pounds.
hour..
= 383 B.T.U. per minute.
23,088 B.T.U.
TOTAL HEAT OF STEAM RECKONED FROM THE VARIOUS TEMPERATURES OF
FEED-WATER AND COMPUTATIONS BASED THEREON.
26. Total heat of pound of dry steam at 120 pounds gauge-
pressure reckoned from o° Fahr.
27. Ditto, reckoned from temperature of main feed-water
(100°)
•
28. Ditto, reckoned from temperature of jacket-water (280°)
29. Heat consumed by engine (H) (22,400 X 1120.5) +
(2560 × 938.5) .
1220.6 B.T.U.
1120.5
938.5
27,501,760
RESULTS.
Substituting these quantities in the formulæ, we have
1. Duty
A
P
P
S
L
N
171.9 X (100 + 4.57 + 3.46) × 1.572 X 76,000
H
27,501,760
= 80,671,622 foot-pounds.
2. Percentage of leakage
C
5,900 X 144
A
L
N
171.9 X 1.572 X 76,000
X 1,000,000
X 100=4.1 per cent.
372
PUMPING MACHINERY.
3. Capacity
A
L.
N
171.9 X 1.572 X 76,000 X 1.24675
4. Percentage of total frictions
-C-
I
A
P
D
12
p
S
171.9 × (100 +- 4.57 + 3.46)
As₂
2
M.E.P., As₂ M.E.P.₂
(167.09 X 61.31) + (697.24 X 13.72 )
2,133,735 gallons.
X100 = 9.4 per cent.
Compound Fly - Wheel Pumping Engine.—High-
pressure cylinder jacketed with live steam from the boiler.
Low-pressure cylinder jacketed with steam from the interme-
diate receiver, the condensed water from which is returned to
the boiler by means of a pump operated by the engine. Main
steam-pipe fitted with a separator. The intermediate receiver
provided with a reheater supplied with boiler steam. Water
drained from high-pressure jacket, separator, and reheater, col-
lected in a closed tank under boiler-pressure, and from this
point fed to the boiler direct by an independent steam-pump.
Jet condenser used operated by an independent air-pump.
Main supply of feed-water drawn from hot well and fed to
the boiler by donkey steam-pump, which discharges through
a feed-water heater. All the steam-pumps, together with the
independent air-pump, exhaust through the heater to the
atmosphere.
DIMENSIONS.
Diameter of high-pressure steam-cylinder (one).
Diameter of low-pressure steam-cylinder (one)
Diameter of plunger (one)
Diameter of each piston-rod
•
Stroke of steam-pistons and pump-plunger
20 inches.
40
20
4
(C
3 feet.
GENERAL DATA.
1. Duration of trial (D)
•
2. Boiler-pressure indicated by gauge (barometric pressure,
14.7 pounds)
3. Temperature of water in pump-well
4. Temperature of water supplied to boiler by main feed-
pump, leaving heater
5. Temperature of water supplied by low-pressure jacket-
pump
10 hours.
120 pounds.
60 deg.
215
225
DUTY-TRIALS OF PUMPING ENGINES.
"
373
6. Temperature of water supplied by high-pressure jacket,
separator, and reheater pump, that derived from sepa-
rator being 340°, and that from jackets 290°
7. Weight of water supplied to boiler by main feed-pump
8. Weight of water supplied by low-pressure jacket-pumy
9. Weight of water supplied by pump for high-pressure
jacket, separator, and reheater tank, of which 210
pounds is derived from separator
10. Total weight of feed-water supplied from all sources
11. Percentage of moisture in steam after leaving separator.
DATA RELATING TO WORK OF PUMP.
300 deg.
18,863 pounds.
615
"L
1025
20,503
"
66
1.5 per ct.
12. Area of plunger minus ½ area of piston-rod (A)
13. Average length of stroke (L and Ls)
14. Total number of single strokes during trial (N and Ns) .
15. Pressure by gauge on force-main (P)
16. Vacuum by gauge on suction-main
17. Pressure corresponding to vacuum given in preceding line
(p)
•
18. Vertical distance between centres of two gauges.
19. Pressure equivalent to distance between two gauges (s)
20. Total leakage of pump during trial, determined from
results of leakage test (C)
21. Number of double strokes of pump per minute
22. Mean effective pressure measured from pump diagrams
23. Indicated horse-power exerted in pump cylinders
•
307.88 sq. in.
3 feet.
24,000
95 pounds.
7.5 inches.
3.69 pounds.
10 feet.
•
4.33 pounds.
DATA RELATING TO WORK OF STEAM-CYLINDERS.
24. Area of high-pressure piston minus ½ area of rod (As,) .
25. Area of low-pressure piston minus ½ area of rod (As₂) ·
26. Average length of stroke, each
•
27. Mean effective pressure measured from high-pressure dia-
grams (M.E.P.,)
•
•
28. Mean effective pressure measured from low-pressure dia-
grams (M.E.P₂)
29. Number of double strokes per minute (line 21)
30. Indicated horse-power developed by high-pressure cylin-
der
•
31. Indicated horse-power developed by low-pressure cylinder
32. Indicated horse-power developed by both cylinders
33. Feed-water consumed by plant per indicated horse-power
per hour, corrected for separator water and for moist-
ure in steam
34. Number of heat-units consumed per indicated horse-power
per hour
•
35. Number of heat-units consumed per indicated horse-power
3078 cu. feet.
20
105 pounds.
117.55 I.H.P.
307.88 sq. ins.
1250.36
"
3 feet.
59.25 pounds.
13.60
20
"
66.33 I.H.P.
61.82 66
128.15
15.60 pounds.
15,652.1 B.T.U.
per minute
260.9
افر
32
374
PUMPING MACHINERY.
TOTAL HEAT OF STEAM RECKONED FROM THE VARIOUS TEMPERATURES OF
FEED-WATER AND COMPUTATIONS BASED THEREON.
36. Total heat of I pound of steam at 120 pounds gauge-
pressure, containing 1.5 per cent. of moisture, reckoned
1220.6 (1.5 per cent. of 866.7) .
from o° Fahr.
=
37. Ditto, reckoned from 215°, temperature of main feed-
water = 1207.6-215.9
•
38. Ditto, reckoned from 225°, temperature of low-pressure
jacket-water = 1207.6 — 226.1
•
•
39. Ditto, reckoned from 290°, temperature of high-pressure
jacket and reheater water = 1207.6 — 292.3.
40. Heat of separator water reckoned from 340° = 353.9-
343.8
1207.6 B.T.U.
991.7
981.5
915 3
10.1
41. Heat consumed by engine (H) = (18,863 × 991.7) +
(615 × 981.5) + (815 × 915.3) + (210 × 10.1) . . 20,058,150
RESULTS.
Substituting these quantities in the formulæ, we have:
1. Duty =
A
P P
S
L
N
307.88 × (95 + 3.69 + 4.33) × 3 × 24,000
X 1,000,000
113,853,044 foot-pounds.
H
20,058,150
C
3078 × 144
2. Percentage of leakage
A
3. Capacity
A
L
L N
307.88 × 3 × 24,000
N
307.88 × 3 × 24,000 X 1.24675
4. Percentage of total frictions
I
As₁
A
D
IO
Р
Þ
s
307.88 × (95 + 3.69 + 4.33)
M.E P.1
× 100 = 2 per cent.
= 2,763,716 gallons.
As₂ ME.P
× 100 = 9 per ct.
(307.88 x 59 25) + (1250.36 × 13.6)
Guaranteed Performance.-In order that the contract
between builder and purchaser of a pumping engine may
conform to the proposed standard, the guarantee as to per-
formance should be expressed in the following terms:
I. The engine shall perform a duty, based upon plunger displacement, equiva-
lent to not less than .. foot-pounds of work for each one million heat-units
consumed.
•
DUTY-TRIALS OF PUMPING ENGINES.
375
噜
​=
་
•
per cent. of the total
2. The leakage of the pump shall not exceed
plunger displacement when the engine is working at its rated capacity.
3. The boiler shall supply one million heat-units to the engine on a con-
sumption of . . . pounds of . . . coal, or it shall evaporate not less than
•
pounds of water from and at 212 degrees per pound of the combustible portion of
the coal named.
4. The mode of determining these quantities is to conform to the standard
method of conducting duty-trials recommended by the Committee of the American
Society of Mechanical Engineers.
Should one contractor furnish the engine and another the
boiler, separate guarantees will be made, the individual re-
quirements of which are the same as those noted.
It is desirable, where both parties concur therein, to intro-
duce into the contract the following additional provision
regarding friction,-viz.:
"The friction of the engine shall not exceed .
power developed in the steam-cylinders."
•
•
per cent. of the indicated
The general mode of operation is to first subject the plant
to a preliminary run under the working conditions, for the
purpose of determining the temperature of the feed-water or
the several temperatures where there is more than one supply.
It is usually impracticable to weigh the main supply of water,
derived, as it generally is, from a low-placed hot well, and
the test of the main quantity of feed-water used must, as a
rule, be made with cold water drawn from the service-main.
The changed conditions in the working of the plant thus in-
troduced and the arrangement of apparatus which is frequently
needed to measure the additional supplies of feed-water, make
it desirable to obtain the working temperatures as a prelimi-
nary to the main duty-trial. Hence the preliminary run is
made, as noted, merely for securing the temperatures. The
main test of the boiler and engine is then carried forward,
and during this test the weights of the various supplies of
feed-water are determined, and the remaining data needed for
making the computations. Finally, as soon as practicable
after these tests are completed, the rate of leakage through
the pump is measured with the engine at rest.
As to the duration of the test, it appears to the committee
376
PUMPING MACHINERY.
that, so far as the main trial is concerned, which is practically
a feed-water test, it need not be prolonged more than ten
hours, unless, in that time, appreciable errors should be pro-
duced by inaccuracies in the observations of the height of
water in the gauge-glass. The duration of the boiler-trial
might, with good reason, be made longer were it not that the
results of the boiler-test are independent of those of the duty-
trial. It is desirable to reduce, if possible, the number of
hours of the trial to such a point that the time expended
upon the work, including that required in preparation for the
beginning of the test and that spent in bringing the test to a
close, shall be such that the same expert, without undue
physical exertion, may have the test under his continuous
supervision from beginning to end. This is feasible where
the length of the duty-trial, according to the plan proposed,
does not exceed ten hours.
An English method of determining the duty of a pump-
ing engine is shown in the annexed abstract of a report on a
Worthington high-duty pumping engine. The test was con-
ducted by Professor W. C. Unwin; the engine was constructed
by Messrs. James Simpson & Co., London. The report is in-
teresting for two reasons,-first, as showing the economy of
a large pumping engine on a low lift, using a steam-pressure
much lower than is common with us in high-duty engines;
and, second, as showing Professor Unwin's method of conduct-
ing a duty-trial, which differs in some respects from the
methods which obtain in this country.
The engines are of the compound Worthington duplex
type, fitted with high-duty attachment, the main dimensions
being as follows:
Diameter high-pressure pistons
low-pressure pistons
pump-plungers
All of 44 inches stroke.
•
•
27 inches.
54
40
(6
+1
DUTY-TRIALS OF PUMPING ENGINES.
377
The valves are of india-rubber and spring-loaded. The
compensating cylinders are II inches diameter, and loaded by
air-pressure to about 120 pounds per square inch. The pumps
lift water from a well communicating with the river and deliver
it through two 3-foot mains to the reservoir about nine miles
distant. The head during these trials, measured by the differ-
ence of pressure in the suction- and discharge-pipes, was from
50 to 65 feet; a head almost entirely expended in overcoming
the friction of the main.
The engine-cylinders were completely jacketed, and the
steam was also taken through a jacketed reservoir between
the cylinders. The jacket-water was discharged through a
pipe regulated by a stop-valve and weighed. The condensers
are injection-condensers with horizontal air-pumps.
The actual dimensions of cylinders, pumps, and rods were
as tabulated below:
DIAMETERS AND AREAS OF CYLINDERS AND PUMPS.

Diam-
eter at
Diam-
eter at
Area of Area of Effective
Piston.
Means.
Rod..
Area.
60° F. 316° F.
Inches. Inches.
Sq. in.
H. P. cylinder A. Back.
26.98 27.02
"C B.
L. P. cylinder A.
A. Front
B. Back.
Front
Back.
A. Front
(C B. Back.
B. Front
26.98 27.02
27.02 27.06 575.1
27.02 27.06 575.I
573-4
573.4 23.8
Sq. in.
17.7
Sq. in.
555-7
549.6
5535
+
17.7 557.4
23.8
551.3
•
53.99 54.07 2296.2
7.0
2289.2
•
53.99 54.07 2296.2
17.7
2278.5
•
54.02 54.10
2298.7
2285.1
7.0
2291.7
54.02 54.10 2298.7
17.7
2281 0
Pump-plungers.
66
Back.
Front
•
•
39.90
39.90
1250.0
16.8
1233.2
1241.6
1250.0
0. 1250.0
The boilers were single-flued Cornish boilers, four in
number, each 28 feet in length and 6 feet in diameter, with a
single flue 3 feet 6 inches in diameter for the greater part of
the length. The length of the grate was 4 feet 6 inches.
Hence the grate area of the four boilers was 60 square feet.
Measurement of the Feed.—The feed was supplied at a
nearly constant temperature of 51 degrees, the ordinary feed
-
32*
378
PUMPING MACHINERY.
arrangements which supply the boilers with hot water from
the jackets and hot well being disconnected. The boiler
feed-pump took its steam from the boilers in use and ex-
hausted into the tank, from which it pumped. The whole of
the steam used was therefore recondensed and returned to
the boilers. Of the heat supplied by the boilers to work the
feed-pump nearly all was returned to the boilers. A small
portion-viz., that due to the useful work of pumping and that
lost by radiation from the tank-was no doubt lost, a small
error telling against the main engines.
Measurement of the Air-Pump Discharge.—The air-
pump discharge was led into a wooden tank with stilling
screens. From this it was discharged through a sharp-edged
circular orifice freely into the air. The diameter of the orifice
was carefully tested after the trials, and the coefficient of dis-
charge from similar orifices is known to be 0.599. The tem-
perature and head were noted every 72 minutes. The tem-
peratures were taken by a fixed zero thermometer, with open
scale, and verified at Kew.
Measurement of Length of Stroke.-As the stroke is
variable, an arrangement of indicating-fingers was attached to
each engine, and the length of stroke of each engine was
noted every quarter of an hour.
Indicated Power.-The indicated power was taken by
four Richards's indicators, which were afterwards tested, with
the result that no important error was found in any part of
the scale with any of the springs. Diagrams were taken every
half-hour.
TRIAL OF ENGINES ON NOVEMBER 5 AND 6, 1888.
This was a 24 hours' trial, the coal consumption being
measured as well as the efficiency of the engines. The trial
commenced 10.22 A.M., November 5, and ended 10.22 A.M.,
November 6.
DUTY-TRIALS OF PUMPING ENGINES.
379
The barometer varied a little during the twenty-four hours,
the mean being 29.78 inches (corrected), corresponding to
14.627 pounds per square inch.
The temperature of the injection varied from 48.6° Fahr. to
49.5°, the mean being 49.2° Fahr.
The mean boiler-pressure was 60.29 pounds per square
inch (74.92 pounds absolute).
The mean vacuum shown by a mercury gauge on the en-
gine was 27.76 inches, or 13.63 pounds per square inch.
The total head of water on the pumps was about 55 feet at
starting and 53.5 feet at the end of the trial. It varied little
during the trial, and the mean head was 53.68 feet.
The air-pressure in the compensating air-vessel varied
from 118 pounds to 122 pounds per square inch above the
atmosphere.
Speed and Length of Stroke.-The speed was remark-
ably constant, and averaged 17.282 double strokes per minute.
The length of stroke varied from 42.32 inches to 43.56 inches;
the mean length of stroke was 43.06 inches for engine A and
43.05 inches for engine B.
Indicated Horse-Power.-The reduction of diagrams
taken every half-hour for the first eight hours, and every
hour afterwards, gave the following results. The variation of
the diagrams was very small.
Indicated Horse-Power.
Engine A. High-pressure, back, 31.662
A. Low-pressure, back, 31.145 = 62.807
66
A. High-pressure, front, 34.176
66
45
3
"C
A. Low-pressure, front, 31.685 = 65.861
B. High-pressure, back, 35.856
B. Low-pressure, back, 28.073
B. High-pressure, front, 35.236
66
128.668
=
63.929
126.849
B. Low-pressure, front, 27.684 62.920
Total indicated horse-power of both engines 255.517
The Pumps.-The mean lift was 53.68 feet; mean length
of stroke, 3.5879 feet. Number of strokes per minute, 17.282.
380
PUMPING MACHINERY.
Hence the pumps lifted 13,407 gallons * per minute. The
pump horse-power is 217.06, consequently the mechanical
efficiency of the engines and pumps is 0.8495.
The Feed- and Jacket-Water.-The feed-water had a
mean temperature of 51.07 degrees. The total feed-water
used was 108,537.4 pounds, or 4522.39 pounds per hour.
The amount of drainage from the jackets was 706 pounds
per hour. Consequently, reckoned per indicated horse-power
per hour, the quantities were:
Total feed (at 51.07 degrees) per indicated horse-power per hour
Jacket condensation
Used in the cylinders
17.700
2.763
. 14.937
Air-Pump Discharge. The mean head over the orifice
was 1.7033 feet, and the mean temperature 74.965 degrees.
The total air-pump discharge was 2586 pounds per minute, or
2522.4 pounds of injection-water and 63.6 pounds of con-
densed steam.
Heat rejected by the Engine per Indicated Horse-
Power per Minute.-The heat required to raise the whole
air-pump discharge from 49.2 to 74.965 degrees. We get
for the heat rejected 260.7 thermal units per indicated horse-
power per minute. This is Donkin's coefficient. The more
accurate estimate of the heat rejected is as follows:
Thermal Units.
Heat due to 2522.4 pounds of injection-water per minute raised from 49.2°
Fahr. to 74.965° Fahr.
Heat due to 63.6 pounds of feed-water raised from 51.07° Fahr. to
74.965° Fahr.
Heat due to 11.78 pounds of jacket-water raised 256.3° Fahr.
Heat rejected per indicated horse-power per minute.
Add converted into work.
•
64,990
. 1,519
3,020
69,529
272.1
42.7
Which neglects the loss by radiation.
314.8
* It will be understood that these are English gallons of 277.274 cubic inches.
DUTY-TRIALS OF PUMPING ENGINES.
381
Heat used, reckoned from the Boiler-Pressure.-
The total heat of the steam, considered dry, reckoned from
the feed-temperature at the mean boiler-pressure, is 1156.5
thermal units per pound. Consequently, the heat delivered
from the boiler to the engine was 341.1 thermal units per
indicated horse-power per minute. The difference between
this and the previous estimate of 314.8 represents loss by
radiation, error due to the presence of priming water in the
steam, and errors of observation.
TABULATED RESULTS.
Double strokes per minute
Boiler-pressure
Feed-water per minute
Jacket drains per minute
Temperature of steam
Pressure on pump
Pressure on compensators .
Mean pressure in high-pressure cylinders
Mean pressure in low-pressure cylinders.
Temperature of injection
•
Temperature of air-pump discharge
Head over orifice
•
Air-pump discharge per minute
Injection-water per minute
17.282
60.29 lbs. per sq. in.
75.37 pounds.
11.77
307.36 deg. Fahr.
23.26 lbs. per sq. in.
I20 "
•
32.92"
•
6.905 "
(6
49.2 deg. Fahr.
•
74.965 deg. Fahr.
1.7033 feet.
2586 pounds.
2522.4
Heat passing through engine per indicated horse-power per
minute:
Thermal units from boiler in saturated steam through cylinders
from feed temperature
Latent heat of jacket-steam
Heat rejected in air-pump discharge
Converted into work.
Radiation and error
•
Indicated horse-power
Pump horse-power
Mechanical efficiency
Feed per indicated horse-power per hour through cylinders .
Feed per indicated horse-power through jackets
Piston-speed per minute
•
287.8
41.45
329.25
260.24
42.75
26.26
329.25
. 255.517
217.06
.8495
14.937 pounds.
2.763
124 feet.
382
PUMPING MACHINERY.
L
It should be noted here that the engines worked for twenty-
four hours with the greatest regularity of speed and stroke,
and this although the steam- and expansion-valves remained
untouched after their first adjustment at starting.
THE BOILERS.
Measurement of Coal used.--The ash-pits were cleared
before the trial, and afterwards nothing was removed till the
end of the trial. The fires were cleaned before the trial began,
and again at 4 A.M.
The fires were not touched at the end
of the trial, but the ash-pits were immediately cleaned, and the
whole of the ashes treated thus:
First the clinkers, including those removed from the fires at
4 A.M. (six hours before the end of the trial), were separated
and weighed. The rest of the ashes were sifted through
a sieve with a half-inch mesh. All that passed through the
sieve is treated as incombustible ash, although probably one-
third of it is unburned carbon. What did not pass through
the sieve is treated as unburned fuel. Analysis in similar
cases has shown that the cinders retained by the sieve are
almost entirely carbon.
-1
The coal account then stands thus:
Gross weight of coal brought into boiler-house
Left on the floor at the end of the trial
Cinders sifted out of the ashes.
Total coal used
456.2 pounds per hour.
The residue consisted of clinkers.
•
Incombustible ashes
•
Pounds. Pounds.
11,180
99
132
231
10,949
66
366
432
The clinkers and ashes amount to 3.9 per cent. of the coal used.
The rate of combustion was 7.24 pounds of coal per square
foot of grate, or 0.19 pound per square foot of heating surface
per hour. The coal used per indicated horse-power per hour
was 1.785 pounds,—a very good result, as the feed was sup-
plied at 51° Fahr., and the rejected heat from the jacket-drains
DUTY-TRIALS OF PUMPING ENGINES.
383
was wasted. The evaporation was 9.914 pounds of water from
51.07 degrees at 307.36 degrees per pound of coal, including
clinkers and ashes. This corresponds to an evaporation of
11.867 pounds per pound of coal from and at 212 degrees.
Calorimetric Value of the Coal.-The heating power
of the coal has not been directly determined, but good Welsh
coal is known to contain about 89 per cent. of carbon and 4
per cent. of hydrogen, the rest being oxygen, nitrogen, and
ash. The calorimetric value of such a fuel is 14.500 (0.89 +
4.28 × 0.04)= 15.387 thermal units per pound. But this is
reckoned for a dried sample of coal, and makes no allowance
for the latent heat of the steam produced in combustion.
There would be produced by combustion 0.36 pound of water
per pound of coal, and the latent heat of this would be 348
thermal units, so that the available heat of a pound of dry
coal would be 15,039 thermal units. The coal as taken from
the yard would contain at least I per cent. of moisture, so
that the available heat of one pound of the coal as weighed
and used would be:
Heat due to o 99 pound of coal.
Less latent heat of 0.01 pound of water
Thermal Units.
14,888
ΙΟ
14,878
Available heat, 14,878 thermal units per pound of coal as
weighed and used. Taking this value, the total heat due to
the combustion of the coal is 26,557 thermal units per indi-
cated horse-power per hour, or 442.6 thermal units per minute.
per indicated horse-power. Of this, 341.1 has been shown to
be delivered to the steam. There remains 101.5 thermal
units per indicated horse-power per minute to account for as
losses in the boilers. The efficiency of the boilers is 0.77.
The coal gave to the steam 11,466 thermal units per pound
of coal used.
Anemometer Observations.-Observations at each
boiler every half-hour gave the following volumes of air
384
PUMPING MACHINERY.
entering per minute in cubic feet at the temperature 79.5
degrees of the boiler-house :
Boiler
Quantity of air in cubic feet per minute
•
·
J K
420 438 486
L
M
360
Hence the total quantity of air used was 1704 cubic feet per
minute, or 225 cubic feet per pound of coal. The weight of
the air used was 7489 pounds per hour, or 16.42 pounds per
pound of coal. As the coal requires nearly 12 pounds of air
per pound for perfect combustion, the quantity of air used was
moderate.
The mean temperature of the flue from the pyrometer
observations was 422 degrees.
Tabulating the results stated, we get:
Per Hour.
Pounds.
Per Indicated Horse-
Power per Hour.
Pounds.
Coal used ..
Air used
Less ashes and clinkers.
•
456 2
7489.0
7945.2
18.0
1.785
29.310
Total weight of furnace gases
•
•
7927.2
31.03
Heat Used and Lost in the Boilers.-The thermal
units of heat developed in the furnaces were applied thus:
Total heat due to coal and gas
Thermal Units per
Ind Horse-Power.
Per Hour.
Per Cent.
. 26,557
100
Given to steam
20,466
77.1
Carried off in furnace gases.
2,657
10.0
Probable loss due to opening fire-doors to stoke
265
1.0
Due to carbon in ashes..
284
I.I
Radiation and unaccounted for.
2,885
10.8
This calculation depends on an assumption of the calorific
value of the coal, but this cannot be far wrong. It assumes
that the steam supplied to the engines was dry. If there was
any priming-water, the heat given to the steam would be less.
DUTY-TRIALS OF PUMPING ENGINES.
385
On the other hand, probably, the losses due to moisture in
the coal and to air entering the furnaces during stoking are
underestimated.
Duty of the Engines.-The work done by the engines
during the 24 hours' trial was 106,010,000 foot-pounds per
112 pounds of coal.* During this trial the ordinary con-
ditions of the engine were altered and heat rejected which is
ordinarily used.
Correcting for this, the duty of the engines in normal con-
ditions of work must be 111.5 millions according to the results
of this 24 hours' trial.
To accompany this report drawings were sent, which are
reproduced, as follows:
1. A mean diagram, Fig. 236, drawn from the diagrams
taken on engine A at 12.30 P.M.
On this has been plotted a
FIG. 236.
Steam
pressure
16s
per sq. inch
80
70-
30
2D
10
aturation Curve
10
20
30
40
Volume of Cylinder in cubic feet.
50
60.
saturation curve for the mean speed per stroke during the
trial. Since the indicated power varied so little, this satura-
tion curve must be very approximately the true curve for the
actual diagrams. The re-evaporation during the stroke is
* The difference between the British and the American standard of hundred-
weight will be noted, and the proper credit given.
*

R
33
Z
386
PUMPING MACHINERY.
very marked, as was to be expected from the large jacket
condensation.
2. Mean diagrams, Fig. 237, from all the diagrams of both

lbs.
60000
FIG. 237.
Mean of all diagrams from Engines A&B
taken at 12.30.p.m.
60000
L.P. cyl.
Total Pressuren
Resultant
40000
Pump
30000
H.P.Cyl.
20000
Compensating Cyl.
10000
20
Nov! 5th 1888.
Engine Thrust.
25
30
35
40 inckes
Stroke of Piston
engines taken at 12.30 P.M., are plotted so as to show the effective
thrust of the engines at each point of the stroke. A curve
FIG. 238.
FIG. 239.
of cosines is drawn,
giving the
thrust

of the compensators.
Combining this with
the engine diagram,
the resultant thrust
is obtained. The ef-
fect of the inertia,
however, is neg-
lected. It will be
seen that the result-
ant thrust is remark-
ably uniform, and
probably the effect
of the inertia of the
moving pistons and

plungers is to increase the uniformity of this thrust.
Indicator diagrams of both the high- and low-pressure
cylinders are shown in Figs. 238 and 239.
:
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 387
203
CHAPTER XVIII.
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING.
THE direct-acting engine possesses features which are of
especial value in handling water. As originally constructed
they were wasteful in the extreme; subsequently, by the
addition of compound steam-cylinders, a liberal saving of
steam was secured, but with the best types of compound
condensing direct-acting engines they fell far short of the
duty attained by the crank and fly-wheel engines. The want
of a suitable attachment to direct-acting pumping engines in
order to secure the advantages of steam expansion, and thus
compete with the crank and fly-wheel pumps, were recognized
long ago, but realized only within the past ten years.
Cameron's high-duty engine, the invention of the late
A. S. Cameron, was patented in 1876, but so far as the writer
is aware has never been placed upon the market. It is prob-
ably the first device of its kind for using steam expansively
in direct-acting pumps. The following engravings and de-
scription present Mr. Cameron's views upon the subject.
In Figs. 240 and 241 letter A designates the steam-cylinder
and letter B the pump-cylinder, the pistons of which are con-
nected by a rod, C. On this rod is secured a cross-head, D,
which connects with the compensating-gear by two rods, E,
and with the expansion-gear of the steam-cylinder by a rod,
F. These rods are provided at their ends with hooks, which
catch over pivots formed at the end of the cross-head, so that
said rods can be readily disengaged, and the pumping engine
can be worked in the ordinary manner without expansion.
The compensating-gear which is represented in the drawing
5
-
1.
·
388
PUMPING MACHINERY.
consists of a cylinder, G, which is secured to the top of the
air-vessel H of the pump. From one side of this air-vessel
extends the ascension-pipe I, and from its opposite side ex-
tends a pipe, J, which leads into the top of the cylinder G.
In this cylinder works a piston, K, from which extends a rod,
L, through a stuffing-box in the cylinder-cover, and on the
upper end of this rod is secured a yoke, M, which is guided
between flanges a a, cast with or otherwise secured to the
sides of the cylinder G. With this yoke are combined two
pairs of toggle-levers, each pair consisting of two sections,
NO, which are connected together by
pivots b, while the upper sections are at-
tached to the yoke M by pivots c, and
the lower sections of the sides of the
pump-cylinder by pivots d. The pivots b,

FIG. 240.
B
AD
which form the connection between the sections NO of the
toggle-levers, support the outer ends of the rods E, which
connect the compensating-gear with the cross-head D.
The cut-off valve or expansion-slide P is moved by a lever,
Q, which connects with the cross-head D by the rod F. The
valve-gear represented in the drawing is of that class known
as “steam-moved valves.”
Working without Expansion.-When it is desired to
work the engine without expansion, the rods F and E are dis-
HIGH-DUTY PUMPING ENGINES—DIRECT-ACTING. 389
.
connected from the cross-rod; but when the rods F and E
are connected to the cross-head, as shown in Fig. 241, the
operation is as follows: When the steam-piston moves in the
direction of the arrow, Fig. 241, the water ejected by the
pump is forced up through the ascension-pipe I; but a por-
tion of such water passes through the pipe
J into the upper part of the cylinder
G, while at the same time the piston K
Fig. 240 is moved upwards against the

FIG. 241.
B
珊
​T
pressure of the water acting on it. As soon as the toggle-
levers NO pass their centre of motion, however, the piston K
begins to descend, and as it is continually exposed to the
pressure of the water acting on its upper surface, it moves
down with considerable power; and if steam has been cut off
at any point of the stroke of the steam-piston, the decreasing
power of the expanding steam during the latter part of the
stroke of the steam-piston is compensated for by the pressure
of the water on the piston K during its descent. It will also
be noticed that by the combination of the toggle-levers N O
with the piston K and cross-head D, the power exerted by
the piston K on said cross-head increases during its descent,
while at the same time the pressure of the expanding steam
on the steam-piston decreases.
Employing Weights or Springs. It will be seen from
this description that the power exerted by the compensating-
33*
390
PUMPING MACHINERY.
piston K during the second half of each stroke of the steam-
cylinder changes with the resistance to be overcome; that is
to say, if the height to which the water is forced increases,
the downward pressure of the water on the compensating-
piston, and consequently the power exerted by the compen-
sating-gear, increases in a corresponding ratio, and vice versa ;
and for this reason, whenever applying a compensating-gear
to a steam-pump, Mr. Cameron preferred to use the mech-
anism shown in the drawing; but, if desired, a simple weight
or spring may be substituted for the piston K, said weight
being raised during the first part of each stroke of the steam-
piston, and being made to descend during the second half;
but it is obvious that the compensating power exerted by
such weight will be uniform no matter what may be the
resistance to be overcome.
Davies's High-Duty Engines.-In 1879, Mr. J. D.
Davies sought to accomplish the same result as that of Mr.
Cameron, though in a somewhat different manner, as shown in
Fig. 242, a device which he patented the following year.
Mr. Davies's object was to cause the piston of a steam-
engine or other motor to transmit a constant and equable force
throughout every portion of its stroke, and prevent any undue
strains on the different parts of the engine when operated ex-
pansively, and to regulate the power to compensate for the
variable resistance offered by the mechanism to be actuated.
D
Referring to Fig. 242, A represents the frame, B the steam-
cylinder, and C the pump-cylinder, of a direct-acting steam-
pump, which parts may be of any approved construction.
is the piston-rod, to the centre of which is secured a cross-
head, E. At opposite sides of the piston-rod D are located
the auxiliary or equalizing cylinders F F', each of which is
provided with trunions a a', which are journaled, respectively,
in the lateral bracket G of the engine-frame and cross-bar or
frame G', the latter being supported on the uprights or pillars
HH'.
The piston-rods II of the oscillating cylinders F F' are
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 391
provided with eyes on their outer ends, through which are
inserted the bolts L L', which extend through the opposite

མ
في
阿
​H.
PA
FIG. 242.
ends of the cross-head E, thus pivoting the pistons to the
cross-heads.
The oscillating cylinders F F' are, in fact, single-acting os-
cillating engines, steam, compressed air, or liquid being ad-
392
PUMPING MACHINERY.
mitted through the trunnions in any suitable manner to the rear
portions of the cylinders to exert a constant outward pressure
on the pistons within said cylinders. When the main piston
is at the commencement of its stroke, the auxiliary or equal-
izing engines or cylinders will offer the maximum resistance.
to the outward movement of the main piston, as, of course, the
main and auxiliary piston-rods are located more nearly in
parallelism at the opposite ends of their stroke than at any
other part of the stroke of the engine. As the main piston-
rod moves outward the equalizer is gradually turned on its
trunnions, and its resistance to the main piston gradually di-
minishes until the main and auxiliary piston-rods are at right
angles to each other, at which point the equalizers or auxil-
iary cylinders will have no effect on the action of the engine.
When the centre of the main piston-rod passes the first half
of its stroke the equalizers or auxiliary cylinders begin to act
and assist the movement of the main engine, as the power of the
main steam-cylinder is then re-enforced by the power exerted
upon the pistons in the auxiliary cylinders, and the power of
the latter constantly increases as the cylinders are swung
around and their piston-rods approximate more closely a line
of movement in parallelism to that of the main piston-rod. The
angles of the auxiliary piston-rods are constantly changing
from the commencement to the end of the stroke of the engine
or other motor to which the equalizers are attached, and hence
the relative amount of resistance offered and assistance af-
forded by the equalizers varies throughout every portion of
the stroke. The force of the equalizers may be varied
throughout the different portions of the stroke of the engine
or motor by a cut-off applied to any part of the supply-pipe,
or by varying the cut-off or pressure on the main piston of
the engine or motor, or by varying the pressure on the
pistons of the equalizers during the different parts of the
stroke of the main piston-rod, or by varying the proportions
of the stroke of the equalizers to that of the stroke of the
main piston-rod, or by changing its position during the stroke
to vary the pressure, as desired.
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 393
In single-acting engines, or engines having a greater load in
one direction, the equalizers may be placed nearer one end of
the stroke, thus offering greater resistance in one direction
and giving more assistance in the other than when located
midway the stroke.
The Worthington High-Duty Engine, the invention of
Mr. Davies, was purchased by the firm of Henry R. Worth-
ington. The value of the main features of this design were
fully recognized by them, but its want of completeness in
order to make it valuable in practical use in pumping engines
required much thought, time, and money. Mr. C. C. Worth-
ington, of the above firm, became personally interested in
working out the several devices, which, in combination with
the original invention, have demonstrated fully the practica-
bility of this method of compensation in direct-acting engines.
Nor is this all, the compensating device must be applicable
to duplex engines; therefore this new train of mechanism
must be applied to each engine separately, and both sets of
compensators must be controlled by a common regulating
device yet to be designed. A sectional elevation of this engine
is shown in Fig. 243.
Adaptation to Duplex Engines.--In the case of the
duplex pumping engine, its one peculiar feature is the variable
and adjustable pause which its valve motion permits at the
end of each stroke. The effect of this pause when dealing
with the pumping of fluids is well known to be of great prac-
tical value. The combining with such an engine of a fly-
wheel or any other contrivance in which the momentum of
moving masses is employed for the purpose of enabling it to
run under higher ratios of expansion, and consequent econ-
omy, would rob it entirely of this distinctive and valuable
characteristic, because only a determinate and instantaneous
pause at the end of the stroke would then be possible. This
fact has hitherto prevented the highest rate of economy being
attained by this type of engine. By combining with it the
•
394
PUMPING MACHINERY.

FIG. 243.
HHH
屋
​THHHHE
HH
HHHHH!
HH
HHHHI
HHHHH
H H H H H H I
HH
THAT
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 395
compensating-cylinders, as described, a result heretofore con-
sidered impossible with it is accomplished,-viz., not only
does it realize as high ratios of expansion as are possible with
any fly-wheel type, but the valuable and adjustable pause at
the end of the stroke is retained. The duplex pumping en-
gine thus becomes capable of attaining the highest economic
results without the quality of its motion being affected in the
least or its well-known smooth and quiet action sacrificed.
Fluid used in the Compensating-Cylinders.—The
first suggestion was to use steam in the compensating-cylin-
ders, but, upon further consideration, compressed air was
thought to be better suited for that service. The first engine
built was for experimental purposes only, and this engine was
fitted with an air-compressor and tank with a distributing
system leading to the four compensating-cylinders. The air-
pressure in the tank could be fixed and maintained at any
pressure best suited to the work, or conditions of service.
Notwithstanding the many advantages which compressed air
seemed to possess for this particular service, it was thought
preferable to fill the compensating-pipes, cylinders, etc., with
water, and have the compressed air act upon it, as by this
means a liquid packing is interposed between the compressed
air and the pistons of the compensating-cylinders and con-
nections, and thus acts to prevent the escape of air by leakage,
and also to keep the pistons well lubricated.
Safety Attachment. Up to this point reference is had
only to the compensating-gear and its usefulness as a means
of using steam expansively, but a new feature was now intro-
duced by which the steam-valves were to be operated from
the cross-head of the compensating-gear. In the regular
duplex engines the distribution of steam is effected by suitable
links operated from the cross-head attached to the main
piston-rods; but by connecting the steam-valve gear with the
compensating cross-head, it is apparent that if a breakage of
a rod should occur at any point between the steam-pistons
396
PUMPING MACHINERY,
and the water-plungers, the compensating-gear for that side of
the engine would be thrown out of service, and the motion
of the opposite engine would cease immediately upon the
completion of the then unfinished stroke. This is a feature
of great importance, as, if the valves were operated so as to
allow the steam to act upon the pistons after the rod was
broken, the pistons, being relieved of their load, would be
moved violently in the cylinders, and would be liable to cause
great damage to the engine.
In ordinary engines this can be done by operating the
throttle-valve so as to properly regulate the amount of steam
admitted to the steam-cylinders, and, so far as the available
power which is developed by the main cylinders is concerned,
this can be readily done in the present engine; but as to the
power developed by the compensating-cylinders, the case is
different. The power developed by these cylinders, as will
be readily seen, is always the same, regardless of the quantity
of steam admitted to the main cylinders, and cannot be regu-
lated or controlled by the operation of the throttle-valve.
When, therefore, the engine is to be started, or run with little
or no load, there must of course be sufficient steam admitted
to the main cylinders to overcome the resistance offered by
the pistons of the compensating-cylinders up to the middle of
the stroke. As soon, however, as the middle of the stroke is
reached, the main pistons are not only relieved of this resist-
ance or load, but are assisted by the pistons of the compen-
sating-cylinders, so that, unless means were provided for
preventing it, the last half of the stroke would be made with
great violence. In pumping against heavy pressure there is
also, of course, some danger that the force-main may burst,
which, in case it should happen, would suddenly relieve the
engine of the whole or a large part of its load, thus allowing
the speed of the engine to become suddenly accelerated to
such an extent as to be liable to occasion damage.
The original or experimental engine had the compensating-
cylinders placed between the steam- and water-ends, but at
the conclusion of the experiments, which lasted several
HIGH-DUTY PUMPING ENGINES—DIRECT-ACTINC. 397
months, the compensating-cylinders were changed from be-
tween the steam- and water-end, and placed at the outer end
of the water-cylinder, as shown in Fig. 243. There is also
shown immediately above the compensating-cylinders an
accumulator for securing the necessary pressure in the com-
pensating system.
The accumulator used is of the differential type; it has
below a small cylinder filled with water or oil, within which its
plunger moves, while above it has a larger cylinder filled with
air, and within which there is a piston-head which fits closely
to the cylinder, and is at the same time attached to the top of
the plunger in the lower cylinder.
By this arrangement it will be seen that the pressure per
square inch on the plunger or ram of the accumulator will be
the pressure per square inch on the piston-head in the upper
cylinder, multiplied by the difference between the area of the
piston-head and the lower plunger. This difference of areas
is a matter of calculation based upon the particular service
for which the pump is constructed. The pressure in the air-
cylinder is controlled by the pressure in the main delivery-
pipe of the pump as it is connected to that pipe. This con-
nection with the main has another very important use, as the
power exerted by the compensating-cylinders is a very con-
siderable part of the power used in driving the pump-plunger
at the latter part of its stroke, and it will be seen that if, for
any cause, either by the breaking of the main or otherwise,
the load is entirely thrown off the pump; the plunger cannot
make a disastrous plunge forward, for the reason that the
steam in the steam-cylinder is, by reason of its expansion,
too low in pressure to drive it, while the fall of pressure in
the main has robbed the accumulating-cylinders of their
power.
The practical working of the completed machine
may be illustrated graphically as follows: The indicator dia-
grams, Fig. 244, are from an engine of this type, the ratio of
34
398
PUMPING MACHINERY.
4
A
B
Atmos.
D
Net
Beginning-
result
ony
FIG. 244.
FIG. 245.
zime
1
•Water Plunger.
Line
cort
compensator.
Resistance
Compensator
E
3 stro
|
high- to low-pressure
cylinders being as I
to 4 in area. Re-
taining the high-
pressure lines, and
multiplying the ver-
tical low-pressure
lines by 4 (the differ-
ence in areas), there
is had as a resultant
the line of steam-
effort shown in Fig.
245. Immediately
underneath this lat-
ter illustration is a
sketch showing five
positions of the com-
pensating - cylinders,


corresponding
to
each quarter of the
stroke from the be-
ginning.
Curved
H lines are drawn in
this diagram, Fig.
245, by one of which,
in the lower left-hand
End of Stroke, tod
·
corner, is shown the
resistance of the com-
pensating cylinders
to the forward piston-
movement for the first
half of the stroke, and
in the lower right-
hand corner is an-
other curved line
showing the effort
of the compensating-
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 399
cylinders in assisting the main piston's movement on the
last half of the stroke. By deducting the resistances in the
first half of the stroke, the steam line AB is effective only
in. A D.
In the middle of the stroke the effort of the compensating-
cylinders is nil, so that the resistance A C at the beginning of
the stroke is o at half-stroke, or at the point E in the dia-
gram; but immediately on passing the centre of the stroke
the effort of the compensating-cylinders gradually increases,
as shown in the curved line EH, so that the steam-pressure
FG at the end of the stroke is augmented by the compen-
sator-effort FH, which two combined efforts give the terminal
pressure FI, which is practically the same as at the beginning
of the stroke. The combined effort on the water-plunger
throughout the whole stroke, for the several ordinates in the
diagram, is shown in the line DI.
As a proof of the remarkable manner in which the com-
pensating-cylinders do compensate for the irregular pressure
of the steam as thus used, if a measurement be taken between
the upper or steam curved line, and the lower, or compen-
sating curved line, on any of the ordinates, in any part of the
stroke, it will be found that the distance between these lines,
which is the measure of the available power for driving
the pump-plunger, is exactly the same as is the distance be-
tween the upper and lower lines of the water-card on any or-
dinate, in any part of the stroke of the pump. In fact, it may
be said that the lines of power, as well as the lines of resist-
ance, are practically parallel, and thus it is that there is ex-
erted at all times just power enough and no more to force
the water column along so quietly, so steadily, that on en-
gines of this construction it has been found there is no
possible use for an air-chamber on the delivery-mains; and
thus is the apparently paradoxical problem solved of pro-
ducing a perfectly uniform motion and pressure in a steam-
pump using steam-power variable to the last degree, and with-
out the use of shafts, cranks, fly-wheels, or heavy vibrating
beams.
400
PUMPING MACHINERY.
THE GROSHON HIGH-DUTY ENGINE.
This is a recent invention by Mr. John A. Groshon, of
New York City, and patented some three years ago. The
invention has for its object the economic use of steam in a
direct-acting pumping engine, and for the few machines thus
far constructed it has, under test, demonstrated its fitness to
take rank as a high-duty engine.
A perspective view of a compound duplex pumping engine,
fitted with Mr. Groshon's valve-gear and compensating-device,
is shown in Fig. 246. The high- and low-pressure cylinders are
arranged tandem on a bed-plate. The steam- and exhaust-
valves are modifications of the Corliss type, and are so dis-
posed in the cylinder-casting as to permit the employment of
separate steam- and exhaust-ports, thus retaining the advan-
tages of the five-ported cylinder. The steam-valves have a
positive movement at the beginning of the stroke, and are
released at the proper point of cut-off by an hydraulic regu-
lator controlled by the water-pressure in the force-chamber, or
delivery-main. Dash-pots are provided as in other types of
releasing-gear.
The compensating-device is arranged on a girder frame
of I-beam form between the steam- and water-cylinders; the
compensating-cylinders are located in the pit below the engine-
room floor. These latter cylinders are in communication
with the main discharge-pipe from the pumps, and subject to
the pressure of the same, either with or without multiplying
device. Fig. 247 is a side elevation of this engine, showing
the hydraulic regulator and its connections, together with the
compensating-cylinder.
The levers and their connections being in the position.
illustrated in Fig. 248, it will be observed that the pressure
within the cylinders S will, through the medium of the rods.
P, levers M, rods J, and levers E, operate to oppose the move-
ment of the piston-rod A during the first portions of its stroke,

Truj
SARJETARIETYYNON
1
FIG. 246.
三
​ΠΟΙΟΙ
4 1
A
a
Ammin
401

402
FIG. 247.
C
O
ㅇㅇ
​DISCHARGE
SUCTION
HIGH-DUTY PUMPING ENGINES—DIRECT-ACTING.
403
·
or until the levers E have turned sufficiently to come at right
angles to the piston-rod A, after which the pressure within the
cylinders S, acting on the pistons T, will, through the system
of levers and rods, operate to assist the movement of the said
main piston-rod A.
There are indicated by the dotted lines in the drawings the
movements of the levers M and levers E, and also the extent
FIG. 248.

P
M
K
L
T
D
J
X
2
of the movement of the piston-rod Q while under pressure to
permit the action of the rods under the force applied to the
longer arms of the levers M. It will be understood that during
the first portion of the stroke of the main piston-rod the
\
404
PUMPING MACHINERY.
longer arms of the levers E will, through the rods J, cause
the levers M to rotate and bring their longer arms, with the
rods P, upwards in the direction of the arrow or dotted line
at the upper portion of Fig. 248, the effect being to draw the
pistons Tagainst the pressure within the cylinders S, and
thereby to transmit said pressure to the piston-rod A in a line
opposite to its movement during the first portion of its stroke;
and it will also be noted, upon following the movement of said
levers and rods, that after the levers E have passed their centre
the pressure within the cylinders S will be transmitted to the
main piston-rod A in line with its movement, and thereby
assist the same during the latter portion of its stroke.
Dotted lines in the drawings illustrate arcs of two circles
differing in diameter, one being long and flat, and the other
of a greater curve. These indicate the lines of travel of the
lower end of the rocking-lever E and of the rod J. If the
lever E and rod / were allowed their natural movement from
their points of suspension, they would travel on the arcs
indicated by dotted lines in the drawings, and hence in oper-
ation there must be either a force applied or the overcoming
of a force to compensate for the difference between these two
arcs, and this force is that which is transmitted to the piston-
rod in line with its length.
In the moving of the lower end of the lever E on its line
of travel it is retarded by the rod J, which has a different line
of travel, and to overcome this the piston in the auxiliary
cylinder is compelled to rise and overcome the force from the
accumulator to an extent which would turn the auxiliary lever
M sufficiently to permit the lower end of the lever E and the
lower end of the rod to travel on the same arc, although
hung from different centres, and if left in their natural state
would travel on different arcs.
The indicator diagrams shown in Fig. 249 are from an
engine having 12-inch high-pressure steam-cylinders, 24-inch
low-pressure cylinders, water-plungers 18½ inches diameter,
all of 18 inches stroke, using 75 pounds steam-pressure, and
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 405
pumping against 70 pounds water-pressure. The scale of the
upper diagram before reduction was 30 per inch, and that of
the lower diagram
was 12 per inch.
These cards show a
rate of steam expan-
sion which promises
FIG. 249.

well, other things
being equal, for the
future of this engine..
Atmospheric line.
The
Davey
High - Duty
En-
gine. - Mr. Henry
Davey read a paper
14.7
75 15s.
before the British
Association (Section G) in 1887, entitled "Expansive Working
in Direct-Acting Pumping Engines." The author commenced
by referring to a paper previously read before the British As-
sociation, in which the expansive method of working steam-
pumps can be secured without the aid of a fly-wheel. The
object of the paper under notice was to describe a more recent
invention of his by means of which a far greater degree of
expansion is made possible.
When an engine has heavy reciprocating parts, such as long
pump-rods or loaded plungers, expansive working is possible
because of the inertia at the beginning and the momentum
WV 2
towards the end of the stroke, expressed by the formula.
2g
By this method, for a considerable degree of expansion, a
very high velocity must be given to the mass when the weight
is small. In engines which have not long pump-rods it is not
always convenient to provide weights sufficiently heavy to
enable a high degree of expansion to be employed. The
mechanism which Mr. Davey described equates the engine-
power and pump-resistance, by causing decreasing pressure
of the expanding steam on the piston of the engine to bring
406
PUMPING MACHINERY.
a nearly constant force to bear on the pump throughout the
stroke. By means of the annexed diagrams the action of the
pump is made clear.
The pump resistance is represented in Fig. 250 by the
parallelogram c, a, b, d, and the engine-power diagram by the
figure c, e, f, g, d; and supposing the parts of the engine to
have no weight, then means are required by which the piston
FIG. 250.

3
a
4
9
of the engine may move with varying velocities relative to that
of the pump-piston, exceeding the mean velocity by the ordi-
nates 1, 2, 3, and falling short of that velocity by the ordinates
4, 5, 6.
Decreasing Velocity of Pump-Piston.-In Fig. 251,
A is the engine, B the pump-piston, and C a triangular
frame turning on the fulcrum D. The pump-piston is at-
A
18
FIG. 251.

D
B
tached to the frame at the point E by means of a vibrating
connecting-rod, and the engine-piston to the point F by means
of a similar rod. While the engine is making its stroke in
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 407
the direction of the arrow, the pump-piston is decreasing in
velocity relative to that of the engine-piston, the ratio being
determined by the relative positions of E and F.
Ratio of Expansion.-In applying this mechanism to
pumping engines it is first necessary to determine the ratio of
expansion to be employed, and then to see how nearly the
force and resistance can be equated. In Fig. 252, a, b, c, d, e
are the combined diagrams of a compound engine working
FIG. 252.

f
72
C
19
2
e
with the given ratio of expansion a, f, g, e, the diagram of effects
of the varying velocities of the engine and pump-pistons, and
a, h, i, e, the pump-resistance diagram. Then acceleration of
velocity takes place from h c, and, knowing the weight of the
moving parts, the acceleration may be calculated. It will at
once be seen that the mechanical advantage obtained by this
mechanism greatly reduces the acceleration for a given mass.
Clearance. In answering the discussion which followed
the reading of this paper, and speaking on the variation in
length of stroke, Mr. Davey said that all direct-acting engines
have the stroke controlled by the action of valve-gear, and
depend on the adjustment of the gear for clearance. The
engine described might have 8 feet to 10 feet length of stroke,
and work safely to I inch of the covers, or in special cases
even less; but that was a fair allowance. The Cornish engine
had a higher economy than any engine working with the
same ratio of expansion, but the clearance there would be 2
408
PUMPING MACHINERY.
inches to 3 inches. The economy in such cases had always
been a puzzle to him. With a pressure of 30 pounds to 40
pounds and four expansions, there would be an equal effect to
a rotative engine working with eight expansions and a press-
ure three or four times as great. Rotative engines were made
for pumping under the sentimental notion that they were more
economical on account of less clearance. But it must be
remembered that a rotative engine must be designed to meet
the worst condition of work, and therefore could not, from
the nature of their design, take the full advantage, that might
otherwise be obtained, of more favorable conditions. A rota-
tive engine must depend for its expansion on the number of
revolutions, and mining engineers, to take one instance, must
work sometimes at slow speeds, and the energy of the fly-
wheel was not then enough to work with a large degree of
expansion. For this reason rotative mining engines do not
in practice always work expansively; though direct-acting
pumping engines are always at their maximum expansion,
whatever speed they may be working at.
Fielding's High-Duty Engine.—The publication of Mr.
Davey's paper was the occasion of a letter addressed to
Engineering by Mr. John Fielding, in which he says that
he does not doubt that the arrangement described by Mr.
Davey as a recent invention of the author" was so described
in good faith, and without the knowledge that it had been
previously invented. Nevertheless, a search through the
patent records would have convinced Mr. Davey that Field-
ing's patent, No. 788, of 1874, most clearly anticipated him.
Mr. Fielding enclosed a tracing showing the action of his
compensating-gear as applied in one form of pumping engine,
which is reproduced in Figs. 253 and 254.
Mr. Fielding's object was to enable engines to be worked
expansively (without necessitating heavy masses to be put in
motion), for the purpose of absorbing surplus energy during
the first part and of giving it out during the last part of the
stroke, with the end of obtaining an approximately constant
HIGH-DUTY PUMPING ENGINES-DIRECT-ACTING. 409


T
FIG. 254.
H
FIG. 253.
E
C
S
35
410
PUMPING MACHINERY.
pressure on the pump-piston throughout the whole of the
stroke, while having a varying pressure on the steam-piston,
resulting from the expansion of the steam after the supply
had been cut off.
Fig. 253 shows a longitudinal section and Fig. 254 a side
elevation of one form of Mr. Fielding's invention. Two
pistons, A B, work within the cylinder C, and attached to
these are two pump-rams, D E, working through stuffing-
boxes in the cylinder-covers into pumps F G. A rod, H,
connecting to piston B, extends through centre of ram D,
terminating in a cross-head, I, working within a slot formed
in ram D.
A similar cross-head, J, is formed on ram D. I and J are
connected by side-rods to quadrants K K in such a manner
that piston A drives ram E, while B drives D, and the quad-
rants K K are so proportioned that at the commencement of
a stroke the speed of the piston will be relatively smaller than
that of the ram, say about one-half; at half-way the speeds
will be equal, while at the end of the stroke the speed of the
piston will be double that of the ram.
These relative speeds correspond roughly to the proportions
shown on tracing, but the levers may be proportioned to suit
various degrees of expansion. The drawing shows a single
cylinder only; the compound principle would greatly reduce
the variations in the pressure on the steam-pistons, and so
reduce the difficulty of equalizing them.
Mr. Fielding's patent, not having been kept up, is void, and
the arrangement is therefore public property. Mr. Fielding
takes occasion to say in his communication, that, compared
with the ingenious compensating motion introduced by the
Worthington Pumping Engine Company, his had the advan-
tage of directness and simplicity, seeing that it involves the
use of no extra cylinders, stuffing-boxes, or reservoirs.
HIGH-DUTY PUMPING ENGINES—FLY-WHEEL.
4II
די
CHAPTER XIX.
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
THE high degree of economy attained by some engines of
this type over ordinary direct-acting, and the general run of
crank and fly-wheel pumps, is due to the design and construc-
tion of the steam-engine portion of the machine; the use of
a high initial steam-pressure, together with the advantages
which may be secured by a liberal steam-expansion in two or
more cylinders, may be accepted as the real reason for the
high degree of efficiency which engines of this type have
developed.
A comparative diagram of steam-expansions has already
been given, together with a simple explanation so far as relates
to steam-pressures; the consideration of the steam-engine as a
heat-engine is not entered upon, it being quite foreign to our
present purpose, which is more especially with water-end de-
sign than with that of the steam-end.
The degree of refinement to which steam-engine design has
been carried seems to leave but little more to be accomplished
in that direction.
The enormous loss often referred to in recounting the dif-
ference between the known energy of coal and the net result
of the best type of steam-engine is, in the present state of the
art, wholly outside of the province of the engineer to correct.
The heat rejected by the furnace cannot, except in minor
quantity, be economically recovered, and must, therefore, be
considered as one of the unavoidable losses which accompany
the generation of steam. Mr. J. T. Henthorn contributed,
in Vol. XII., "Transactions of the American Society of
Mechanical Engineers," the results of a test of a triple-
<
412
PUMPING MACHINERY.
expansion engine, showing an indicated horse-power for
each 12.94 pounds of steam per hour, the engine developing
532 H. P. In the discussion which followed the reading of
this paper, Mr. George H. Barrus, the efficient chairman of the
Duty-Trial Committee of the above society, who recommends
that the economy of a pumping engine be referred to heat-
units rather than to coal and water consumption,. worked out
these quantities roughly during the meeting, in which he made
the number of thermal units consumed per I. H. P. per hour
14.038, and the quantity consumed per minute 234. This
may be taken as representing the best modern practice in
steam-engine performance; some tests recording as low as
12.6 pounds of steam per I. H. P. per hour are reported, but
taken altogether, 13 pounds may be said to fairly represent the
best steam economy at this time.
The simple fact that a water-end is operated by an engine
having an automatic or adjustable cut-off does not make it a
high-duty engine. There are pumping engines thus equipped
now in use in water-works pumping stations in which the
average duty is no greater than could be had by the employ-
ment of a simpler and cheaper type of compound direct-
acting engine.
A high-duty engine must be designed for the particular
service required of it, so that the adaptation of the steam to
the water-end shall be one favorable to high economy. The
steam-pressure usually determines the number of cylinders to
be used. For steam-pressures averaging 100 pounds the
engine would be compound, the engine having one high and
one low-pressure cylinder, or one high and two low-pressure
cylinders; the latter is sometimes employed in three-cylinder
punips. For steam-pressures averaging 150 pounds, the
engine would be arranged for triple expansion, the work in
the three cylinders being distributed approximately, thus: the
150 pounds (total pressure) in the first cylinder, 60 pounds
initial pressure in the intermediate cylinder, the third or low-
pressure cylinder having an initial pressure of 20 pounds,
exhausting against a back-pressure of say 3 pounds above
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
413
the vacuum; this would make a fair distribution of tempera-
tures, and would be productive of good results if properly
applied.
Three pumping engines have been selected for illustra-
tion,-viz.:
1. A single, vertical, bucket- and plunger-pump, designed
by Mr. E. D. Leavitt, Jr., and erected at Lynn, Mass.
2. A pair of horizontal, double-acting plunger-pumps, de-
signed by Mr. George H. Corliss, and erected at Pawtucket,
R. I.
3. A vertical, triple expansion, single-acting plunger-pump,
designed by Mr. Edwin Reynolds, and erected at Chicago,
Ill.
The Leavitt Compound Pumping Engine.—A com-
pound rotative pumping engine, designed by Mr. E. D. Leavitt,
Jr., for the city of Lynn, Mass., is shown in sectional elevation
in Fig. 255. In this engine the steam-cylinders are placed
very nearly under the main centre of the beam, and are in-
clined outwardly, at the proper angle to connect with the
dependent projections cast on the beam, the low-pressure
cylinder being connected in the pump-end of the beam, and
the high-pressure to the opposite or crank-end. The pistons
thus have opposite movements; and the connections between
the cylinders, for carrying the steam from the high-pressure to
the low, are very short and direct. The strain on the beam
is, by this construction, reduced to a minimum. The top of
the high-pressure cylinder exhausts into the top of the low-
pressure cylinder, and the bottom of the high-pressure cylin-
der exhausts into the bottom of the low-pressure cylinder. A
single valve controls the connection between the two cylinders
at the bottom, while two valves are used in the connection at
the top, one close to each cylinder, so that the capacity of
this passage is not added to the clearance of the high-pressure
cylinder when being filled with steam, nor is it exhausted when
the low-pressure cylinder is being exhausted. It thus an-
t.
35*
414
PUMPING MACHINERY.
swers, in some measure, to the reservoir of a compound
marine engine. Both cylinders are steam-jacketed at sides
and ends, and, together with the steam-connections, are
clothed with asbestos covering, and lagged with black walnut.
FIG. 255.

In general appearance the high-pressure cylinder approxi-
mates the diameter of the low-pressure cylinder; this is due
to the fact that the lagging has been kept away from the body
of the cylinder, an air-space being thus formed between the
two.
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
415
All the steam-valves are of the gridiron form, and are
moved by revolving cams, which have grooves of the proper
shape cut in their sides, in which run steel rollers on pins
projecting from the sides of rocker-arms. Those for the
high-pressure steam-valves are composed of two disks, fitted
closely together, but not attached. One of these disks, for
the opening movement, is keyed fast to the cam-shaft, and
has its outer edge of the proper shape to open the valve, so
as to give it the required lead; the other, for the closing
movement, has on its inner side a projecting lip, whose in-
terior edge has the proper shape to move the rocker-arm so
as to close the valve. This latter plate is not keyed rigidly to
the cam-shaft, but is driven by a double feather working in a
slot cut in its boss parallel with the bore, and also working in
a slot cut in the shaft at an angle with the axis, so that any
motion given to the feather in the line of shaft changes the
relative position of the cam with its shaft, and consequently
with the crank, thereby causing steam to be cut off earlier or
later. The motion is imparted to the feather by the governor.
The air-pump is double-acting, and worked by a connection
with the beam. The discharge from the hot well is into the
pump-well, and the boiler feed-water is drawn from the hot
well by an independent steam-pump.
The pump is of the Thames Ditton variety, bucket and
plunger type. There is a supplementary pipe and valves to
reduce the friction incident to the passing of the water through
a single bucket-valve in the piston. The valves are of the
usual double-beat form; the bottom; or foot-valves, are seven
in number, the upper valves three,—one in the piston and two
in the supplementary passage; each of the exterior bottom
valves has an independent cover on the chamber for the
readier inspection and removal of the valves. To supply air
to the air-chamber there is a small air-cock on the lower
valve-chamber. The air drawn in through this cock on the
up-stroke of the pump somewhat reduces the quantity of
water discharged by the pump. Careful experiments, pre-
416
PUMPING MACHINERY.
:5
#
vious to and in the early part of the test, established the effect
of this cock by measurements at the weir, and a uniform
opening of this cock was kept during forty-eight hours of the
trial. During the other four hours the average loss of action
was a trifle less than during the forty-eight hours, during one
hour the cock being completely closed.
The water is drawn by the pump from a large well supplied
by a main from the collecting-pond, and discharged through
a 20-inch force-main. This main is of cast iron for a portion
of its length, thence by a cement-lined sheet-iron pipe of the
same diameter to the bottom of a gate-house in the service-
reservoir. During the experiment it was thence conducted
by a vertical cast-iron pipe of sufficient height to discharge
into a wooden trunk, connected with a weir-box, where the
capacity of the pump was determined.
The principal dimensions are shown in the following
table:
Diameter of high-pressure cylinder
low-pressure cylinder .
high-pressure piston-rods
•
17 inches.
36
3
low-pressure piston-rods.
(6
air-pump
•
66
pump-barrel
plunger .
(C
bottom and supplementary valves, out-
side lower seat
bottom and supplementary valves, in-
side upper seat
piston-valve, outside lower seat.
(C
inside upper seat
fly-wheel
Length of stroke of steam- and water-pistons
air-pump
Distance between end-centres of the beam
Steam-lead, high-pressure cylinder, top.
bottom
low-pressure cylinder, top
"
bottom
Exhaust-lead, high-pressure cylinder, top
•
3/21/20
111
<<
2010
18/1/
157
ΙΟ
22
(C
161/1
(C
26 feet.
7
((
44 inches.
II feet.
о
•
O
о
7 inch.
ㅎ
​bottom
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL. 417
Exhaust-lead, low-pressure cylinder, top
66
Closure before end of the stroke:
High-pressure cylinder, top.
•
bottom
Low-pressure cylinder, top
(C
Clearances:
•
bottom
•
High-pressure cylinder, top.
bottom
Low-pressure cylinder, top
•
bottom.
Top connection between cylinders
51 inches.
bottom.
3
97
1411
6
"C
51%
.0231 of vol. of cylinder.
.0360
(C
.0109
"
66
"
•
.0134 of high-press. cyl.
.0104
Weights:
Fly-wheel.
Beams
10.7 tons.
Moving parts connected with beams
Force-main:
Length from pump to top of vertical pipe at
reservoir
•
Height of top of vertical pipe above bottom of
well
4.2
5
1904 feet.
•
163.34
The boilers are two plain horizontal tubular set in brick-
work, fired beneath, flame passing the length of the boiler
and returning through a number of 3-inch tubes to a front
connection, thence passing up into a single flue on the top of
the boilers in brick-work, and to the chimney. The principal
dimensions are:
Diameter of shell.
"L
steam-drum
tubes
Length of shell
5 feet.
3
3 inches.
16 feet.
Height of drum
Length of grate
Width of grate .
6
5
(C
· 5
❝6 inches.
The duration of the trial was fifty-two hours, without a
stop. The engine at the conclusion of the test continued its
work, the fires and water in the boilers being kept as uniform
as possible. The capacity of the pump was established by
weir measurements at the reservoir. The actual delivery was
bb
418
PUMPING MACHINERY.
found to be 96 per cent. of the calculated capacity of the
pump. The capacity of the full pump is 194.37 gallons; the
delivery per revolution, as determined by weir measurement,
was 186.55 gallons; the number of revolutions made in the
52 hours was 57,357; and the total quantity of water deliv-
ered at the reservoir during that time was 10,700,163 gallons,
-or the delivery was at the rate of 205,772 gallons per hour.
The delivery required by the terms of the contract was 200,000
gallons per hour.
The duty was established as follows: The area of
the pump, ascertained by direct measurement, was 534-53
square inches. The load was determined from a gauge on the
main, near the pump, the average reading of which was 64.2
pounds; the difference of level between centre of gauge and
surface of water in pump-well reduced to pounds = 8.21
pounds; allowance for friction and bends of main, between
gauge and pump-well, I pound; total pressure per square
inch, 73.41 pounds. The length of the stroke, as measured,
was 7 feet.
From the above data and the number of revolutions the work
done at the pump was determined,-no deductions having
been made for loss of action. The coal was picked Lacka-
wanna coal, and the contractor was allowed to make use of any
coal from the cinder that he deemed of value as combustible,
but no credit was allowed for screenings or unconsumed coal.
The coal was charged against the machine in gross, and the
total quantity fed into the furnaces during the 52 hours
was 15,160 pounds. The division of the foot-pounds of work
by the weight of coal determined the duty, and the result was
103,923,215 foot-pounds of work for every 100 pounds of
coal fed into the furnaces.
The Corliss compound pumping engine, erected at
Pawtucket, R. I., designed by George H. Corliss, is shown in
sectional elevation in Fig. 256. It presents several novelties
HIGH-DUTY PUMPING ENGINES—FLY-WHEEL.
419

Boiler
D
EPOCA
Feed-Water
B
Meter..
FIG. 256.
IN
Exhaust
Meter
SECTION
SIDE ELEVATION, SHOWING LONGITUDINAL
ON CENTER LINC OF PLUNGER.
or
PUMP
1
F
420
PUMPING MACHINERY.
in pumping engine design that a less fearless man than Mr.
Corliss would have hesitated to employ, but which fourteen
years of daily use show to have been judiciously chosen
for an engine of the size selected for their water-supply,
which was the delivery of 3,000,000 gallons of water in 24
hours against a head of about 270 feet.
This engine consists of one high- and one low-pressure
cylinder, placed side by side horizontally, with piston-rods
connected directly to double-acting water-plungers. The
plunger-rods are keyed to cross-heads, from which links run
to vertical beams; the link connection being made in the
centre of the beam and the fulcrum at the lower end. From
the upper end connecting-rods are coupled to cranks, which
are placed at right angles, and have a throw equal to twice
the stroke of the steam-piston and plungers. The bearings
for the crank-shaft are carried on top of the pump air-cham-
bers, and braces run from these air-chambers to the main
centre pedestals of the beams.
The steam-cylinders, valve-boxes, and cylinder-heads are all
thoroughly steam-jacketed; steam of full boiler-pressure is
used in each. The induction-valves of both cylinders are con-
nected to one lever for hand adjustment, and are placed under
the control of a governor, which limits the velocity of the
engine to 52 revolutions per minute.
The steam-valves and valve mechanism are in all their sa-
lient features similar to Mr. Corliss's previous practice, but
with certain modifications which were first used in the engine
which furnished the motive-power of the machinery-hall at
the Centennial Exhibition.
Receiver.-There is a receiver between the steam-cylin-
ders approximately equal in volume to that of the low-pressure
cylinder; the condensed water from the steam-jacket is re-
turned to the boiler-feed, and that from the receiver, after
being re-evaporated into superheated steam, is discharged into
this same receiver with the effect, not infrequently, that more
steam can be accounted for on the low-pressure indicator-card
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
421
than came out of the high-pressure cylinder. The re-evapora-
tion of the condensed water is performed in a coil, placed in
the boiler-flue.
A jet-condenser is employed, the water of condensation
being supplied automatically from the pump-well.
The engine is adapted for the use of steam of 125 pounds
pressure, which is first admitted to the high-pressure cylinder,
and from which it is exhausted into a receiver, and from thence
passes into the low-pressure cylinder, from which it is ex-
hausted into a jet-condenser; the steam being expanded about
twenty times its original volume at the moment of final release
from the low-pressure cylinder.
The principal dimensions of the engines are:
Diameter of high-pressure cylinder .
low-pressure cylinder
Length of stroke, both cylinders
66
•
•
cut-off, high-pressure cylinder
low-pressure cylinder.
Clearance, high-pressure cylinder
66
low-pressure cylinder
•
Port area in per cent. of piston area:
Admission high-pressure cylinder
Exhaust high-pressure cylinder
Admission low-pressure cylinder.
Exhaust low-pressure cylinder .
Diameter of all pump-plungers
all rods.
15 inches.
30%
30
66
66
0.22 per cent.
0.335
66
66
4
3.7
64
•
9.04
14.8
5.04
9.08
(6
•
10.52
inches.
234
The pump-valves are similar to one shown in the full-size
illustration, Fig. 257. The disk of copper forming the valve
is shown lifted off of the valve-seat inch, about the extent
which the valves are believed to lift during the working of
the pump. The area exposed to pressure tending to open the
valve is 1.8 square inches. There are 70 valves in each section
of the pump. A lift of of an inch, therefore, offers an area
of outlet equal to half the cross-section of the pump-plunger.
The weight of the moving parts of each valve is 0.8 ounce.
36
422
PUMPING MACHINERY.
To lift a valve of an inch, and maintain it against the com-
pressive force of the spring, requires a water-pressure of 14
FIG. 257.
pound per square
inch. To overcome
the inertia of the
valve would require
less than ½ pound
of water - pressure,
supposing the valve
to open during the
first fifth of an inch
of piston-travel, or
during of a sec-
ond of time.

225
The boilers are of the upright tubular variety, three in
number, also designed by Mr. Corliss, each boiler being 4
feet in diameter and 14 feet high, being of the precise form
and arrangement of those which furnished steam for the
Centennial engine. The aggregate quantities of the three
boilers are as follows:
Water-heating surface
Superheating surface
Grate surface.
Ratio of heating to grate surface
1231 square feet.
66
66
66
508
45
38.6 to I.
The original contract contained a clause that when the
engine was completed and ready for service two tests should
be made: one to continue 24 hours incessantly, and the other
to occupy two weeks, omitting the intervening Sunday, run-
ning the fires ten hours per day as nearly as may be, with the
engine at such speed as shall guarantee the delivery of
3,000,000 gallons in 24 hours. In this every day running test,
all the fuel shall be taken into account, the wood used for
kindling being reckoned at 45 per cent. of its weight in coal.
No allowance to be made for ashes or clinkers. The coal
used in these tests to be first-class run of the mine. The
quantity of water to be delivered is to be ascertained by weir
HIGH-DUTY PUMPING ENGINES—FLY-WHEEL.
423
measurement. A duty of 100,000,000 foot-pounds was guar-
anteed on the 24 hours' trial, and 80,000,000 foot-pounds on
the two weeks' trial.
The two weeks' every day running test was begun
August 6, 1878, and the computation of the duty for the
average of the twelve days is as follows:
Number of hours run
revolutions
Average reading of water-gauge, in pounds.
draft-indicator, being distance
120
373,124
109.164
from centre of water-gauge
down to surface of water in
pump-well, in feet
•
16.3169
66
66
of total head against pump, in feet.
temperature of water, degrees Fahrenheit.
269.0185
78.194
Weight of a cubic foot in pounds
62.20625
Displacement of one revolution in cubic feet
5.6952
956.925 coal.
6521
(6
26,598.5
· 34,076.425
Wood used to start fire, in pounds, 2126.5
Coal used to start fire, in pounds
while pumping, in pounds.
Total pounds of fuel used
DUTY.
•
5.6952 cubic feet X 373,124 revolutions X 62.20625 pounds × 269.0185 feet × 100
34,076.425 pounds of coal
104,357,654 foot-pounds per 100 pounds of coal.
The weir measurement was made for four hours on August
28, and showed the capacity of the engine at the average speed
of the two weeks' test to be 3,060,000 gallons in 24 hours.
The twenty-four hours' test was begun on October 3
following; the details and computations of the duty are as
follows:
Number of revolutions
Average reading of water-gauge, in pounds
74,463
108.06
draft-indicator, in feet.
16.6706
of total head against pump, in feet
266.4852
" temperature of water
68.458°
Weight of a cubic foot, in pounds
62.2882
•
Displacement of one revolution, in cubic feet
5.6952
Coal used while pumping, in pounds
5272
424
PUMPING MACHINERY.
DUTY.
5.6952 cubic feet × 74,463 revolutions × 62.2882 pounds × 266.4852 feet × 100
5272 pounds of coal
= 133,522,060 foot-pounds per 100 pounds of coal.
The results show that the engine exceeded the guarantee on
the two weeks' every day running test by more than 30 per cent.,
and on that of the 24 hours' test by more than 33 per cent.
Annual Duty.-During the fiscal year ending November
30, 1888, the water-works report claimed an annual duty of
123,656,000 foot-pounds, largely exceeding its previous rec-
ords. As no determination of the steam-consumption and
boiler-evaporation had ever been published, the determinations
of duty depending on measurements of coal and water-press-
ure only, Professor J. E. Denton requested permission to make
these measurements and to verify the extraordinary duty con-
tained in the above report. This permission was cordially
granted, the management co-operating with Professor Denton
in every possible way to make the investigation complete.
Professor J. E. Denton's experiments had for their
special object the measurement of the steam consumed by the
engine per horse-power under various conditions calculated to
afford scientific data regarding the application of thermo-
dynamic principles in engines of this type. This required the
separation of the performance of the boiler from that of the
engine. The engine had been several times previously tested
without making this separation, the performance being stated
in foot-pounds of "duty," which expresses the combined
economy of the engine and boiler.
PROGRAMME OF EXPERIMENTS EXECUTED.
1. From May 30 to June 1, 1889, inclusive, the engine was
run continuously 72 hours under its usual working conditions,
and measurements made of
a. The coal consumed, anthracite, stove-size, being used.
b. Horse-power performed both for pumping the water and to overcome the fric-
tion of the engine itself.
*
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
425
c. The pounds of steam generated in the boilers and used to operate the engine.
d. The pounds of water used to condense the steam as it left the engine.
e. All temperatures and detailed measurements necessary to determine the heat
rejected by the engine, the steam condensed in the jackets, the economy due
to passing the condensed steam from the receiver through heater in chimney,
and the dryness of the steam.
ƒ. The duty of the engine, the steam per hour per horse-power, and the evapora-
tion of the boilers per pound of coal, were the principal quantities sought.
2. During June 1 and 2 the engine was run 24.13 hours
under the same conditions as before, but with selected
George's Creek bituminous coal for fuel.
The same determinations were made, but the special object
was to check the previous test and at the same time ascertain
how much gain in duty could be expected from choice bitumi-
nous coal as compared to the stove anthracite used in the first
test, and in the usual operation of the plant.
3. During June 3, 4, and 5, short tests of from six to
nine hours were made, the engine being run with and with-
out steam in its steam-jackets. No measurement of fuel was
made, the object sought being to determine the steam con-
sumed by the engine per horse-power with and without
the jackets in action.
General Arrangement of Apparatus.-In Fig. 256 is
shown a sectional view of the engine with the various ther-
mometers and meters attached. Thermometer G measures
the temperature of the feed-water as it enters the boiler.
Thermometer M was a special one 36 inches long, graduated
by H. J. Green. Twenty inches of length were devoted to the
100 degrees between 300° and 400° Fahr. Each degree was
therefore represented by a fifth of an inch, thus making the
temperature of the steam measurable with great accuracy.
This thermometer remained at about 356.1 degrees during
most of the experiments, thereby showing the steam to be
superheated about 3 degrees.
For a short time the chimney temperature rose from 428
degrees, its normal amount, to 500 degrees, and the super-
36*
426
PUMPING MACHINERY.
heating then became equal to 18 degrees, but the average was
3 degrees, and the minimum 21 degrees.
The quality of the steam was therefore determined to be
perfectly dry on entering the engine without the application
of the more or less vexatious devices known as calorimeters,
for the use of one of which, the Barrus superheating form,
arrangements had been made.
Thermometer / determined the temperature of the water
from the suction-well on its way into the condenser.
Thermometers H and I measured the temperature of the
water leaving the condenser, H extended just through the
pipe and I to nearly the opposite side of the pipe; the former
showed constantly about 104 degrees when the latter read
about 92 degrees. This apparent anomaly was finally ex-
plained by the conclusion that the thermometer H showed
the temperature of saturation of the steam or vapor at the
pressure given by the vacuum-gauge, 28½ inches, while I
gave the temperature of the mixture of condensing water and
condensed steam.
Thermometer N showed the temperature of the hot-well
water. Its readings were practically identical with those of I,
and practically with those of G when the latter were not in-
fluenced by the entrance of the condensed steam from the
jackets into the feed-pipe. Such steam was pumped into the
feed-pipe at a point just below M.
I. Principal Results-Anthracite Test.-Date, May 30
to June 1, 1889. Duration, 72 hours.
Revolutions per minute
Boilers in use
Pressures:
Average boiler-pressure.
receiver-pressure
•
•
back-pressure in low-pressure cylinder
water-pressure against pumps
vacuum.
66
66
suction against pumps
barometer-pressure
49.03
3
127 pounds.
8/1/
281 inches.
1.25 pounds.
115
(6
5.78
30
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
427
*
Temperatures:
Temperature of saturated steam at average boiler-pressure
Average temperature of steam in steam-pipe two feet from
steam-chest
Degrees of superheating
Temperature of feed-water entering boilers
leaving hot well .
Increase of temperature due to admission of jacket-water
Temperature of jacket-water entering feed-pipe
66
66
66
condensing water leaving suction-well
gases at exit from boiler-flues
353 deg. F.
356
3
104
92.1
CC
•
11.9
310
63
428
(C
gases after passing heater in flue through
which condensation from receiver was
circulated
354
cr
66
water blown to waste from receiver
the atmosphere
•
225
(C
70
Total quantities :
Anthracite coal.
15,710 pounds.
Equivalent of wood at 40 per cent., 263 pounds
105
66
Total.
15,815
(C
Ashes, clinkers, etc. .
Water fed to boilers, including condensation from jackets 140,969
Condensing water
·
Fuel
•
Hourly quantities:
4,201,337
2,238
(C
"C
Feed-water
Condensing water
Condensed in jackets
through receiver
Blown to waste from receiver.
Average indicated horse-power, steam-cylinders
66
<<
66
of pumps
•
horse-power to overcome friction of engine, in-
cluding air-pump
horse-power to operate air-pump .
Rate of combustion, coal per square foot of grate per
219.5
1,958
66
58,351
107
156
118
143.89
138.36
5.13
2.35
(C
4.9
hour
Economy of engine:
Steam per hour per
inders
indicated horse-power of steam-cyl-
This figure should be regarded subject to a possible error
of 3 per cent. due to this limit of error affecting the
determination of horse-power from indicator cards.
Efficiency of mechanism of engine and pumps
13.64
96.4 per cent.
428
PUMPING MACHINERY.
Economy of boilers:
•
Water evaporated per pound of fuel from 102
degrees into steam at 125 pounds pressure.
Equivalent evaporation from and at 212 degrees per
pound of combustible
Economy of engine and boilers combined:
Duty per 100 pounds of coal at actual efficiency of
boilers
Coal per hour per indicated horse-power of steam-
cylinders
If boilers had evaporated 10 pounds of water into
steam from actual temperature of feed-water, the
duty would be per 100 pounds of coal
Coal per hour per indicated horse-power of steam-
cylinders
•
8.88 pounds.
12.12
66
124,720,000 foot-pounds.
1.54 pounds.
140,430,000 foot-pounds.
1.37 pounds.
CHECK UPON MEASUREMENT OF FEED-WATER BY MEASUREMENT
OF HEAT EXHAUSTED OR REJECTED AT AIR-PUMP OF ENGINE.
The heat which entered the engine per hour was as follows:
1958 pounds of feed-water, to each pound of which there was
supplied 1221 heat-units, the total heat of evaporation, hence:
Heat given to steam 1958 x 1221 =
The chimney-heater evaporates 38 pounds of water at
225 degrees in steam at 22 pounds pressure, making
38 x 960 =
Total heat received by engine
This heat is distributed as follows:
58,351 pounds of condensing water is raised in temperature
from 63° to 92.1° Fahr., making.
2,390,718 heat-units.
36,480
2,427,198
•
1,698,014
118 pounds of the feed-water is blown to waste from the
receiver, representing 118 × 225=
26,550
107 pounds of the feed-water is drained from the jackets,
representing 107 × 310 =
33,170
The remainder of the feed-water, or 1958-118. 107=
1733 pounds, is found in the hot well, representing 1733
× 92.1 =
159,609
The indicated work performed, or 143.49 horse-power, rep-
143.49 × 1,980,000
resents
368,769
772
The estimated radiation is 20 pounds of steam per hour.
38,320
2,324,432
Balance unaccounted for.
102,766
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
429
..
This discrepancy represents 4 per cent. of the whole
amount of heat. Part of this is ascribable to the leakage of
steam from the safety-valves and stuffing-boxes. The re-
mainder is due to undefined causes. The error is, however,
on the right side to make the feed-water measurements
worthy of confidence as being slightly too large and unfavor-
able to an exaggeration of the economy of the engine.
II. Principal Results-Bituminous Test.-All data for
this test are the same, except the following:
Revolutions per minute.
Average boiler-pressure
water-pressure against pumps
horse-power indicated, steam-cylinders
49.44
128 pounds.
66
116
CC
แ
cut-off high-pressure cylinders .
0.235
147.95
pumps
•
140.74
devoted to friction
7.21
Total bituminous coal.
5200
Total wood
40 per cent. X 200
80
Total
5280
Total ashes
510
(6
Duration of test, 24 hours, 13 minutes.
Fuel per hour.
•
Feed-water per hour
Economy of engine :
Steam per hour per indicated horse-power of steam-
218.6
(C
2046
cylinders.
·
Efficiency of mechanism
13.82
95 per ct.
Economy of boilers:
Water evaporated per pound of fuel from 102 degrees
into steam at 126 pounds. .
Water evaporated per pound of combustible, and at 212
degrees
•
Economy of engine and boilers combined:
Duty per 100 pounds of coal at actual efficiency of
boiler
9.35 pounds.
12.II
•
•
127,350,000 ft. lbs.
Coal per hour per indicated horse-power of steam-
cylinders.
Economy of engine and boilers combined if boilers had
evaporated 10 pounds of water per pound of coal from
temperature of 102 degrees:
The duty per 100 pounds of coal
Coal per hour per indicated horse-power of cylinders
1.48 pounds.
.. 145,041,000 ft. lbs.
1.30 pounds.
430
PUMPING MACHINERY.
By reference to the theoretical heating capacity of the two
kinds of coal it will be seen that the increase of duty with the
bituminous coal was not as great as chemical analysis indi-
cates should have been the case.
Tests to determine Influence of the Steam-Jackets.
-The results of these tests indicate that the saving by the
use of the steam in the jackets is not greater than three per
cent. of the fuel used without steam in the jackets.
Economic Value of the Flue Heater.-Assuming the
effect of the heater to be the addition of about 40 pounds of
steam to the volume of the latter, which remains vaporous at
the end of admission to the low-pressure cylinder, the useful
effect would be that due to increasing the admission-pressure
of the low-pressure card. In round numbers, the steam re-
maining vaporous at low cut-off represents 1600 pounds per
hour. The addition of 40 pounds would increase the pressure
1400 = 10. The absolute pressure averages 22 pounds.
Hence the increase of average pressure during admission
would be less than 1 pound per square inch. The gain in
mean effective pressure due to such an increment would rep-
resent about one per cent. of the total power of the engine.
6
10
Reynolds's Triple - Expansion Pumping Engine.-
The engine illustrated in the folding-plate, Fig. 258, is one of
five built by the Edward P. Allis Company, of Milwaukee,
Wis., for the Chicago Water-Works, and is of the same type
and general design as engines built for the water-works of
Milwaukee, Omaha, Denver, Detroit, New Orleans, and several
other cities.
The original engine of this type was built in 1886 for the
high service of the Milwaukee Water-Works, and was the first
triple-expansion engine built for water-works purposes. This
original engine, although operated with only 80 pounds steam-
pressure, has given a duty as high as 128,000,000 foot-pounds
יהוי
}
1800
#
I
C
HAMM
NOODBURY ST.WORCESTER,
I
BUTANAWATSONEN UNDAN LUZ DNE SIG DAN STENT.
FIG. 258.

imm{{ གm[]r«f T/
!!
!
:
!
་
:
•

HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
431
for each 100 pounds of anthracite coal burned, and has de-
veloped an indicated horse-power on 1.35 pounds of coal per
hour.
The following description of the Chicago engine covers the
main features of all the triple-expansion pumping engines
built by this firm. The general arrangement of the engine
above the floor is somewhat similar to the marine type of
engine. The high-pressure cylinder is 27 inches diameter by
60 inches stroke, the intermediate cylinder 46 inches diameter
by 60 inches stroke, and the low-pressure cylinder 70 inches
diameter by 60 inches stroke. The steam-cylinders are steam-
jacketed throughout and are fitted with Reynolds-Corliss
valve-gear, the valves being placed in the cylinder-heads;
this arrangement reducing the clearances to the very smallest
amount, in some cases the total clearance being only three-
fourths of one per cent. of the cylinder volume.
The valve-gear of each cylinder has adjustable trip cut-offs,
the high-pressure valve-gear being also under the control of
a speed-governor. Each cylinder has two piston-rods con-
nected to a forged steel cross-head, from which four rods ex-
tend directly to the heads of the pump-plungers, which are
situated below and directly in line with the steam-cylinders.
This arrangement makes the machine direct-acting in the
best sense of the word, as the power is transmitted directly
from the steam-pistons to the pump-plungers, without passing
through links, beams, or swinging journals of any kind, the
only power passing through the crank-shaft being that which
is transmitted to the fly-wheels at the beginning of the stroke,
and returned from them near the end of the stroke.
There are three cranks set at angles of 120 degrees apart,
and each crank being connected to its particular plunger in-
sures a practically uniform flow of water from the pumps, and
reduces the shock and jar on the pumps and pipes to a
minimum.
The main shaft is 16 inches diameter in the journals and
18 inches in the wheels, the main journals being 20 inches
long. The fly-wheels are two in number, and weigh about 25
J
432
PUMPING MACHINERY.

FIG. 259.
In
.58
« dif' Mungin a
07.7
n
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
433
·
⚫ tons each. The pumps are three in number, being placed
directly under the steam-cylinders, and arranged as shown in
Fig. 259. They are of the single-acting, outside-packed type,
SK
FIG 260.

DISCHARCE
SUCTION
as shown in Fig. 260, having plungers 3234 inches diameter
by 60 inches stroke, and the plungers are so counterbalanced
T
т сс
37
434
PUMPING MACHINERY.
i
that the work done in the steam-cylinders is the same on the
up-stroke that it is on the down-stroke.
The pump-barrel and valve-chambers, shown in sectional
elevation in Fig. 260, are extremely simple in construction.
FIG. 261.
ー​ト
​=
The valves used are of rubber, of small diameter, working on
brass seats; a section through a valve and seat is shown in
Fig. 262. The valves are arranged in groups mounted on
cages; a plan of one of these cages in shown in Fig. 261,
and in sectional elevation in Fig. 262, each cage being re-
A

HIGH-DUTY PUMPING ENGINES-FLY-WHEEL. 435
movable independent of the others, or permitting the removal
of a single valve without disturbing a cage. These cages are
held in place by a single bronze tap-bolt extending through
FIG. 262.

the centre of the cage and screwing into a bridge cast in the
valve-chamber, as shown in Figs. 260 and 262.
The air- and feed-pumps are driven by an arm attached to
the low-pressure plunger, thus giving them the same stroke
as the main pumps. The height of the engine above the
main floor is about 35 feet, and the depth of pump-pit is 35
feet, making a total height of 70 feet.
The weight of one engine complete with its pumps is 450
tons. At Chicago two of these engines are in operation at
the West Harrison Street Station and three in the new station
at Fourteenth Street. The Harrison Street engines have been
in operation about two years and a half; the Fourteenth Street
436
PUMPING MACHINERY.
engines have recently been put in operation on the comple-
tion of the new four-and-a-half-mile intake tunnel. These
engines have a nominal capacity of 15,000,000 gallons each,
against a head of 125 feet, when running at a speed of about
16 revolutions per minute, and have a guaranteed capacity
of 18,000,000 gallons in 24 hours under the same head.
Duty-Trial.—A test of the Harrison Street engines made
by Mr. Bernard Fiend, assistant engineer of Chicago, in Octo-
ber, 1891, showed the following results:
Duration of test, from 8.45 A.M. to 4.45 P.M.
Total revolutions of No. I engine
(C
No. 2
'
both engines
Average
66
No. I engine
No. 2
8 hours.
7,972
8,042
16,014
16.608 per min.
16.754
both engines
(6
steam-pressure
CC
(C
(C
in first receiver
in second receiver, atmosphere
vacuum in inches
•
Total head of water pumped against
•
•
Water fed to boilers, 197 barrels, at 400 pounds each
plus 163 pounds
Average temperature of feed-water
Total coal burned.
·
16.681
124 86 pounds.
25.83
Zero.
27.0 inches.
126.65 feet.
78,963 pounds.
115 deg. F.
10,975 pounds.
Total ashes
Per cent. of ash
Actual evaporation from 115°
Evaporation from 212°
"6
(6
per pound of combustible.
Total number of U. S. gallons pumped by both engines
pounds of water. pumped by both en-
gines at 8.34 pounds per gallon .
Total net foot-pounds of work done by both engines
Duty per 1000 pounds of feed-water on plunger dis-
placement
•
Net duty after allowing 2 per cent. for slip and leakage
Average indicated horse-power of No. I engine
66
No. 2 engine
total indicated horse-power of both engines
per
indicated horse-power.
Water
per hour
1,408 66
12.83 per cent.
7.195 pounds.
8.238
66
9.45
10,511,348 gallons.
87,664,640
11,087,822,390 ft. lbs.
140,416,679
137,608,346
388.5 H. P.
390.2
778.7
<<
12.675 pounds.
HIGH-DUTY PUMPING ENGINES-FLY-WHEEL.
437
Reference to the indicator diagrams taken during the test,
average samples of which are reproduced on smaller scale in
FIG. 263.


HP


INT.P


L.P
Fig. 263, together with the combined diagram Fig. 264,
will show how close the actual expansion of steam conformed
FIG. 264.

CHICAGO PUMPING ENGINE.
DIAGRAMS
to the theoretical, and probably no better results have ever
been attained under similar conditions.
As an illustration of the economy of this type of engine in
every-day operation under ordinary conditions of service, it is
37*
438
PUMPING MACHINERY.
1
reported that the 18,000,000-gallon engine of this type in the
Milwaukee Works, already referred to, attains a monthly duty
of about 120,000,000 foot-pounds for each 100 pounds of coal
burned for all purposes without deductions of any kind, the
coal burned being only a fair quality of anthracite, and the
boilers being of the ordinary horizontal tubular type, evapo-
rating about 834 pounds of water per pound of coal.
INDEX.
Accumulator, 162.
A.
friction in, 165.
Acid water, effect of, 290.
Air, clearing pumps of, 112.
quantity to resist shock, 118.
Air- and vacuum-chambers, 90.
Air-chamber, fire-pump, 282.
lined with wood, 310.
mine-pump, 295.
pressure-pump, 164.
Air-pump, bucket and piston, 26.
for air-chambers, 93.
for condenser, 27.
Blake valve motion, 204.
Boiler performance, 363.
test by Prof. Unwin, 382.
tests and duty-trials, 362.
form recommended, 363.
Boilers, Pawtucket Water-Works, 422.
heat used and lost in, 384.
Lynn Water-Works, 417.
Brass-covered piston-rods, 277.
Brewery-pump, 145.
Brick-earth, pumping, 189.
Bucket- and piston-pump, 33.
Bucket- and plunger-pump, 32.
Bucket-piston for air-pump, 26.
American Society of Mechanical Engi- | Bucket-pump, ball-valve, 25.
neers, 71, 321, 334, 363.
Andrews's centrifugal pump, 357.
Annular valve, 64.
Appold's centrifugal pump, 335.
Area of steam-ports, 237.
for valve-seats, 84.
Artesian-well pump, 223.
Assoc. Fac. Mut. Ins. Co., 271-288.
Atmospheric pressure, 9.
Atmospheric pump, 10.
Automatic relief for pressure-pumps,
169.
sprinklers, 274.
Auxiliary throttle-valves, 316.
Ball-valve, 55.
B.
with metal core, 56.
Barrus, George H., 346, 412.
Bell-valve, 55.
Berrenberg rotary pump, 226.
Bessemer works, pumps for, 167.
Birkenbine, H. P. M., quoted, 175.
Björling, P. R., quoted, 53.
mitre-valve, 25.
Bucket-pumps, 23.
Buffalo Stm.-Pump Co., Mine P'ps, 304.
Built-up wing-valve, 67.
Bushings for valve-seats, 82.
Bushing stuffing-boxes, 45.
for plunger, example, 309.
Butterfly-valve, 51.
By-pass connection, 145.
C.
Calorimetric value of coal, 383.
Cameron, A. S., 387.
Cameron's high-duty engine, 387.
sinking-pump, 291.
valve motion, 203.
valve-seats, 61.
Cataract for steam-pumps, 220.
Centrally-packed plunger-pump, 31, 142.
Centrifugal force in pumps, 338.
pumps, 334.
and engine, 354.
blades for, 336.
casing for, 337.
439
440
INDEX.
:
Centrifugal pumps, concave arm wing, | Compound mine-pumps, Buffalo, 304.
351.
design, 337.
double suction, 253.
dredging, 354.
effect of high rotation, 340.
hollow arm piston, 351.
in series, 360.
piston for sand, 352, 354.
pistons, forms of, 351.
speed, Morris Machine Works,
356.
whirling in, 336.
Charging-pipe, 122.
Check-valve, 122.
Chicago Water - Works, Reynolds's
pumping-engine, 430.
arrangement of pumps, 432.
duty-trial, 436.
section of pump, 433.
valves and cages, 434.
Circular steam-valves, 253.
Clack-valves, 47.
and supplemental valve, 49.
experience with, 48.
fulcrum for, 52.
hinged, 51.
lift of, 52.
Clarkson valve motion, 207.
Classification of pumps, 11.
Clearance, steam-cylinder, 280, 407.
Coal, calorimetric value of, 383.
Cold-rolled steel, 37.
Cold-water valves, 56.
Coles, H. J., quoted, 119.
Collars for plunger-rods, 43.
Combined foot-valve and strainer, 104.
Compensators.
tachment.)
(See High-Duty At-
Compound cylinders, ordinary sizes, 250.
ratios, 249.
stuffing-box, 245.
engine piston-rods, 240.
power of, 256.
isochronal pump, 262.
mine-pumps, 314.
T
Jeanesville, 313.
Knowles, 305.
pressure-pump, 168.
pump, Tangyes's, 259.
pumps, duty-trial, duplex, 370.
fly-wheel, 372.
duplex performance, 251.
high-service attachment, 237.
intermediate head, 238.
single, direct-acting pump, 224.
steam-pumps, 235.
action of steam in, 245.
arrangement of, 240.
tank-engine, 254.
Concentric ring-valve, 87.
Conical mitre-valve, 53.
springs, 60.
Corliss, George H., 269, 413.
pumping-engine, Pawtucket, 252,
418.
Cornish double-beat valves, 70-77.
pumping-engine, 310.
clearance in, 407.
Cost of dredging with centrifugal pump,
359.
of fire-pumps, 287.
of pumping earth, 190.
Cranes, hydraulic power for, 168.
pressures in, 152.
Crank, advantages claimed for, 175.
Crank-pumps, 173.
details, 182.
diagram of effort, 176.
geared, 197.
Goodbrand & Co., 185.
Guild & Garrison, 184.
single, 190.
slotted cross-head, 184.
strength for, 174.
valve areas for, 175.
without fly-wheel, 178.
Woodward's, 183.
Cross-head for compound engine, 243.
Cup-leathers, mould for making, 17.
packed piston, 16.
INDEX.
44I
Cushion-valves, 280.
Cylinders, arrangement of comp., 240.
lagging of, 253.
D.
Dash relief-valve, 228, 237, 280.
Davey, Henry, 405.
high-duty engine, 405.
Davidson triple-expansion engines, 264.
valve-motion, 212.
valve-seats, 61.
Davies, J. D., 390.
high-duty engine, 390.
Davis, E. F. C., quoted, 321.
Dean Brothers' valve motion, 209.
Deane's sinking-pump, 291.
Deep-well pump, 223.
Delivery-pipes, size of, 107.
Denton, Prof. J. E., 39.
on steam-jackets, 253.
Differential plunger-pump, 143.
Dimensions, Corliss engine, Pawtucket,
421.
Leavitt engine, Lynn, 416.
Reynolds's engine, Chicago, 431.
Direct-acting mine pump, 295.
steam-pumps, 200.
Disk-valves, 56.
Double-beat valves, Cornish, 70.
India-rubber valves, 70.
Double-plunger pressure-pump, 169.
Drainage of steam-jackets, 251.
Drainage-pipe, 151.
Dredging by centrifugal pumps, 355.
General Gillmore on, 357-
Drip-cocks, 284.
Duplex pumps, 226.
fire-pumps, 272.
mine-pumps, 297.
sinking-pump, 294.
steam-valves, 229.
tanking, 254.
valve-adjustment, 230.
·
valve-gear, 227.
valve-movement, 233.
Duty-trials of pumping engines, 361.
Duty-trials of pumping engines, com-
pound duplex, 370.
fly-wheel, 372.
duration of test, 375.
engine performance, 367.
English method, 376.
guaranteed performance, 374.
Lynn pumping engine, 418.
measuring work done, 366.
Milwaukee pumping engine, 438.
Pawtucket pumping engine, 423-
429.
plunger displacement, 366.
quantity of work done, 366.
standard proposed, 363.
Worthington engine, 385.
E.
Easy-seating water-valve, 63.
Eccentric, Woodward pump, 184.
Eccentrics for pumps, 195.
Efficiency, Appold's pump, 340.
centrifugal pump, 337.
curves, Webber, 343.
Gwinne pump, 340.
large centrifugal pumps, 349.
loss of, in pumps, 149.
rotary pumps, 331.
24-inch centrifugal pump, 346.
Electrical pump, 198.
Ellis, G. A., quoted, 108.
Engine, efficiency, 412.
Engineering News, quoted, 256, 264.
English method duty-trials, 376.
Erste Brünner Masch. Fab. Pump, 136.
Evaporation assumed, duty-trials, 362.
Expansion diagrams, 179.
Expansive use of steam, 177.
F.
Factor of safety, 277.
piston-rods, 38.
pressure-pumps, 163.
Farcot's centrifugal pumps, 350.
Fastening piston-rods, 40.
Fielding, John, high-duty engine, 408.
Fiend, Bernard, duty-test by, 436.
442
INDEX.
--
Final results of boiler-trial, 363.
Fire-pumps, 271.
air-chamber, 282.
brass fitting of, 277.
capacity of, 274.
commercial sizes, 275.
cost, 287.
duplex, 272.
force-valves, 278.
H. P. required for, 273.
hose connections, 285.
necessary fittings, 287.
pipe sizes, 281.
piston pattern, 276.
speed, 99.
plunger pattern, 276.
pressure-gauge, 283.
priming-pipes, 284.
ratio of steam to water, 276.
safety-valve, 283.
size-plate, 276.
sizes recommended, 273.
speed of, 274.
steam-ports, 279.
strength of parts, 277.
suction-valves, 278.
test for acceptance, 285.
vacuum-chamber, 281.
Fire-stream, size and capacity, 274.
Five ports in duplex cylinders, 226.
Flanging machines, pressures in, 152.
Flow of water into a pump, 96, 114.
Floyd and Morton's steam-end, 259.
Fly-wheel, function of, 178.
pumping engines, 411.
Foot-valve and strainer, 103.
multiple, 105.
relief-valve, 104.
single disk, 105.
Force of impact in pumps, 339.
Form of air-chamber, 91.
Foundry-lifts, pressures in, 152.
Four-seated valve, 80.
Freeman, John R., 287.
Friction in hydraulic machines, 165.
loss, table of, 108.
Frictional resistance of water, 150.
Fulcrum for clack-valves, 52.
G.
Gain by compounding, 235.
Geared crank-pump, 197.
pressure-pump, 167.
pump with eccentrics, 195.
Gillmore, General Q. A., quoted, 357.
Goodbrand & Co. crank-pump, 185.
Gordon isochronal compound pump,
262.
valve motion, 218.
Green, Howell, quoted, 310.
Greindl rotary pump, 329.
Grooved water-piston, 19.
Groshon, John A, 400.
high-duty engine, 400.
Guaranteed performance of pumping
engine, 374.
Guild and Garrison crank-pump, 184.
valve motion, 215.
Gun-metal-lined mine-pump, 308.
Gwinne pump, efficiency of, 340.
tests, 350.
H.
Hale, R. A., quoted, 331.
Heat-units, a base for duty-trials, 363.
Heat used and lost in boilers, 384.
Hemispherical valve-guard, 60.
Henthorn, J. I., quoted, 411.
High-duty attachment, Cameron's de-
vice, 387.
Davey's device, 405.
Davies's device, 390.
Fielding's device, 408.
Groshon's device, 400.
Worthington's device, 393.
High-service attachment, 237.
Hill, John W., quoted, 256.
Hinged clack-valve, 51.
Hippopotamus leather, 48.
Holly, A. L., quoted, 167.
Horizontal strainer, 102.
Hot water, pumping, 119.
INDEX.
443
Hot-water valves, 56.
Hydraulic elevator service, 13.
engines, 165.
pressure-pumps, 152.
pressures, notes on, 153.
pump, water-end, 155.
transmission of power, 164.
Hyperbolic logarithms, 180.
I.
India-rubber disk-valves, 56.
Indicator diagram, pump, 50.
Inside-packed plunger-pump, 29, 138.
Intermediate cylinder-head, 238.
head with stuffing-box, 245.
Irrigating machinery, Pacific coast, 360.
Isochronal pumping engine, 262.
Jamb-nuts, 44.
J.
Jeanesville mine-pump, 313.
Joints for mine-pipes, 324.
K.
Knowles compound mine pump, 305.
piston mine-pump, 296.
valve motion, 201.
L.
Lagging steam-cylinders, 253.
Lawrence, Mass., pumping engine, 269.
Mach. Co. centrifugal pump, 344.
Leaking suction-pipes, 100.
Leather-faced wing-valve, 53.
Leather for packings, 18.
Leathers for clack-valves, 47.
Leavitt, E. D., Jr., 269, 413.
compound pumping engine, 413.
Length of pump-stroke, 126.
Lift of butterfly-valves, 52.
of clack-valves, 52.
Limit of suction, 10.
Lining piston-pumps, 20.
mine-pumps, gun-metal, 298, 308.
wood, 308.
removable, 22.
Loss of efficiency in pumps, 149.
Lost motion in duplex pumps, 229.
Lost-motion links, 230.
Lubricating compound mine-pumps,
296.
Lynn, Mass., pumping engine, 413.
M.
McCarty's centrifugal pump, 335.
McCreath, James, quoted, 110, et seq.
Machines for raising water, 7.
Mair, John G., quoted, 176.
Marsh valve motion, 213.
water-valve, 63.
Mean pressure of steam, 181.
Mechanically-operated valves, 88.
Metal disk-valve, 58.
Metallic packings for pump-rods, 46.
Metal rings for water-pistons, 18.
used for mine-pumps, 289.
Milwaukee pumping-engine duty, 438.
Mine-pumps, 289.
arrangement of, 307.
Buffalo S. P. Co., 304.
centrally packed, 300, 302.
compound, 314.
duplex, 297.
electrical, 199.
end-plungers, 303.
lined, gun-metal, 298, 308.
wood, 308.
lubricating piston of, 296.
multiple valves for, 302.
piston-pattern, 299.
plungers, 318.
pot-valve chambers, 306.
rods, 300.
rotative, 408.
single, 295.
valve-area, 302.
valve-chambers, 309, 321.
valve-seats, 62, 300.
valves, 300, 319.
water-ends, 311.
Mine-water, action on metals, 319.
effects of, 290.
pipes, 324.
444
INDEX.
Mitre-valve, conical, 53.
Montreal sewage-pumps, 346.
Morris Machine-Works, 351.
Multiple-beat valve, 68.
N.
Nagle, A. F., quoted, 71.
Noisy pumps, 87.
Non-conductors for lagging, 253.
Nozzle, size for fire-pump, 285.
O.
Outside-packed plunger-pump, 30, 139.
P.
Packed plunger-pump, 135.
Packing a water-piston, 16.
Parallel spiral springs, 60.
Parsons, R. C., quoted, 334, et seq.
Passages through centrifugal pumps,
337.
Pawtucket pumping engine, 269.
annual duty, 423.
Denton's experiments, 424.
Alue-heater, 430.
general arrangement, 425.
performance of boilers, 363.
of engine, 251.
results, anthracite, 426.
bituminous, 429.
steam-jacket, 430.
Perreaux's valve, 81.
Pipe sizes, fire-pumps, 281.
Piping a pump, 121.
Piston water-end, removable lining, 130.
Piston- and plunger-pump, 34.
rods, 37.
Piston-pump, ball-valves, 129.
Erste Brünner, 136.
for mine, 298.
for thick stuff, 145.
linings, 20.
with wing-valves, 128, 145.
Piston-rods, brass-covered, 277.
compound engines, 240.
fastenings, 40.
Piston-rods, protected, 245.
Piston-speed for pumps, 126.
underwriter pump, 99.
Pistons, cup-leathers, 16.
fibrous packing, 15.
fitted with wood, 19.
lubrication of, 296.
metal rings, 18.
movement with crank, 174.
with grooves, 19.
Plunger and ring water-end, 137.
and solid ring, 28.
bushing, example of, 309.
for mine-pumps, 300, 321.
Plunger-pump, 27.
centrally-packed, 31, 142.
central diaphragm, 30.
differential, 143.
inside-packed, 29, 138.
outside-packed, 139.
outside tie-rods, 30.
vertical, 144.
water-end, 129.
Worthington, 132.
Plunger-rod fastenings, 40.
Poillon, M., quoted, 328.
Pot-valve chamber, 158.
mine-pump, 305.
number of, 161.
Power for hydraulic engines, 165.
hydraulic transmission of, 164.
of compound pumping - engine,
256.
pressure-pumps, 169.
required for pumping earth, 189.
Power-pump, 195.
Pressure affected by accumulator, 163.
caused by shocks, 116.
Pressure-gauge, fire-pump, 283.
Pressure-pumps, 152.
air-chambers for, 164.
automatic relief, 169-171.
compound, 168.
design, 158.
double plunger, 169.
duplex, 167.
INDEX.
445
Pressure-pumps, factors of safety in, 163. | Rotary pumps, failures, causes, 328.
geared, 167.
large water-ends, 162.
materials for, 154.
power, 169.
valve and seat, 160.
Pressures in hydraulic machines, 152.
of water, table of, 124.
Priming-pipes, 284.
Priming-tank, fire-pump, 284.
Protected piston-rod, 245.
Pump, electrically driven, 198.
speed of, 86.
Pump-bucket for high pressure, 24.
Pump-diagram, vacuum, 50.
Pump-rod collars, 43.
details, 40.
Pump-valve, Corliss's, 421.
Woodward's, 149.
Pumps, clearing of air in, 112.
pressure caused by shocks, 116.
varieties of, 8.
Pumping brick earth, 189.
engines, duty of, 361.
hot water, 119.
R.
Radial force in centrifugal pumps, 339.
Raising water above suction limit, 10.
Ratios of H. P and L. P. cylinders, 249.
Rectangular-valve, double-seating, 67.
Relief-valve and pipe, 122.
-
for foot-valve, 104.
Removable lining, 22.
Revolutions, duplex, defined, 274.
Reynolds, Edwin, 413.
Reynolds's triple-expansion engine, 430.
Rhinoceros leather, 48.
Richards, John, 360.
Riedler, A., 87.
Riveting, increase of pressure in, 163.
Rood, Vernon H., 313.
Rotary pumps, 325.
classification of, 326.
design, 327.
efficiency, 328-331.
objections to, 326.
Rotative engines for pumping, 408.
Rumsey & Co. rotary pump, 326.
S.
Sand, proportion pumped by centrifugal
pumps, 358.
Sand-pumps, 352.
Sewage-pump, centrifugal, 346.
Shand, Mason & Co. pump, 190.
Shears, hydraulic, pressures in, 152.
Shock in pumps, 116.
Side-pipe and strainer, 102.
Simpson, James & Co., 376.
Simpson's lubricating piston, 296.
Single-crank engine, without fly-wheel,
190.
Single mine pumps, 295.
Sinking-pumps, 290.
Cameron's, 291.
Deane's, 291.
Worthington's. 294.
Size of air-chamber, 92.
for steel piston-rods, 38.
of vacuum-chamber, 96.
of valves, 84.
Slide-valve, duplex, adjustments, 280.
Slide-valves, circular, 253.
Slotted cross head, 184.
Slurry-pump, 189.
Specifications for fire-pump, 271.
Speed and capacity of pumps, 127.
of pumps, 86.
Spindle-valve, 54.
Springs, 58, 278.
Steam, expansion of, 177.
in compound cylinders, 245.
mean pressure of, 181.
per horse-power, 412.
Steam-ends for trade-pumps, 13.
Steam-jackets, 251.
mine-pumps, 316.
value of, 252.
Steam-pipe, 122.
Steam-pipes for collieries, 321.
38
446
INDEX.
Steam-ports, area of, 237.
in duplex pumps, 226.
pressure for compounding, 235.
pumps, direct-acting, 200.
underwriter pump, 279.
Steam-valves, circular, 253.
Steel rods, strength of, 37.
Sterk, A. Elink, quoted, 350.
Sticking of valves, 279.
Strainer and vacuum-chamber, 102.
Strainers, 101.
Stroke-gauge, duplex pump, 281.
Stroke-table for pumps, 128.
Stuffing-box bushings, 45.
Stuffing-boxes, 44.
for fire-pumps, 277.
Submerged pump, mine, 316.
Suction, 9.
Suction-pipe, 122.
enlargements in, 99.
fire-pump, 281.
loss of energy in, 112.
Suction- and delivery-pipe, 98.
practical limitations, 110.
valves, fire-pump, 278.
Syringe, 9.
T.
Tandem compound cylinders, 236.
Tangyes's compound pump, 259.
Tank-engine, compound, 254.
economy of, 256.
Taper-taps for guard-nuts, 57.
Teague, Henry, quoted, 48.
Tensile strength, steel rods, 38.
Test of underwriter pump, 285.
Thermal units per horse-power, 412.
Throttle-valves, 3 for large pumps, 316.
Thurston, R. H., tests by, 39.
Tie-rod collars, mine-pump, 317.
flanges, mine-pump, 318.
Tobin bronze, 39, 277.
Top-guided valve, 66.
Torricelli's experiments, 10.
Triple-expansion engine, Davidson's,
264.
Reynolds's, 430.
Troy valve, 65.
Tuck's packing, 16.
Tweddell, R. H., quoted, 153.
U.
Underwriter pump, 271.
Unwin, W. C., quoted, 376, et seq.
V.
Vacuum-chamber, 96, 122, 281.
and strainer, 102.
in suction-pipe, 97.
Vacuum diagram, pump, 50.
Valve, annular, 64.
area, ordinary, 84.
crank-pumps, 175.
mine-pumps, 302.
ball and cage, 55.
bell, 55.
butterfly, 51.
cap, 58.
chamber lined with wood, 309.
mine-pump, 321.
clack, 47.
concentric ring, 87.
conical mitre, 53.
dash-relief, 228.
double-beat, Cornish, 70.
rubber, 70.
easy-seating, 63.
four-seated, 80.
gear, duplex, 227.
guard, hemispherical, 60.
guides, 65.
india-rubber disk, 56.
mechanically operated, 88.
metal disk, 58.
mine-pump, 319.
motions, Blake's, 204.
Cameron's, 203.
Clarkson, 207.
Davidson, 212.
Dean Brothers, 209.
Guild and Garrison, 215.
isochronal, 218.
Knowles, 201.
INDEX.
447
:
Valve motions, Marsh, 213.
multiple-beat, 68.
Perreaux's, 81.
size of, 84.
top-guided, 66.
Troy, the, 65.
weighted, 83.
wing, 52.
curved, 53.
with cataract, 220.
Valve-plates, 58.
Valve-seat bushings, 82.
and spindle, 57.
dovetailed, 83.
fire-pumps, 279.
inserted guard, 59.
securing in place, 61.
Valve-springs, 58.
Valves, sticking of, 279.
Velocity of flow in suction-pipe, 98.
Vertical plunger-pump, 144.
W.
Water following pump-bucket, 113.
frictional resistance, 150.
loss of energy in flowing, 112
Water, raising of, by suction, III.
Water-elevators, 8.
Water-end design, 126.
for crank-pump, 174.
for mine-pump, 298, 317.
for trade-pumps, 13.
Water-pipe for mines, 324.
Watson and Stillman, 170.
Webber, W. O., quoted, 334, et seq.
Weisbach, P. J., quoted, 112.
Wing-valve, 52.
built-up, 67.
leather-face, 53.
Wolff, Ernest, tank-engine, 255.
Wood-lined air-chamber, 310.
mine-pump, 308.
Wood-packing for pistons, 20.
Woodward pump, 183.
valve, 149.
Worthington, Henry R., 226.
Charles C., 393.
high-duty engine, 393.
duty of (Unwin), 376-385.
plunger-pump, 132.
sinking-pump, 294.
tank-engine, 255.
THE END,
1

;
+
甲
​


A
UNIVERSITY OF MICHIGAN
3 9015 06729 5439