754-

A/4

UC-NRLF

INVESTIGATION
Of the Conditions Governing the Choice of a Proper

Quality Standard for Artificial Gas

with Conclusion and Recommendation of the

JOINT COMMITTEE ON CALORIMETRY

of the
PUBLIC SERVICE COMMISSION

and
GAS CORPORATIONS

in the

SECOND PUBLIC SERVICE DISTRICT
NEW YORK STATE

TP 75-+

/V-f

CONTENTS.

Page
Title Page 1

Joint Committee's Letter Submitting Its Report to the Commission.... 5
Report of Joint Committee 7

APPENDIX A 15-19

Origin of Investigation 15

Organization of Committee 15

Classification of Companies Making Tests Table 2 16

Definition of Heating Value of Gas (Footnote) 17

Calorimetric Tests, Number Reported by Each Company, Table II. .. 18

Photometric Tests, Number Reported by Each Company, Table III. 18

20-41

....... 20

npany,

r with

ximum

Errata sheet will be found opposite Page 94 Units

22-39

ie and
41

42-60

nating

Unenriched Coal Gas 42

Recent Development in Coal Gas Manufactured 44

Coke Oven Gas 45

Carburetted "Water Gas 46

Effect of Distribution on the Heating and Illuminating Value 47

Compression and Transmission Tests on Enriched Coke Oven Gas. . 54
Laboratory Experiments, Compression and Freezing Carburetted

Water Gas 56

Heating Value Calculated by Analysis 56

Comparison of Continuous and Intermittant Operation 59

Comparison of Efficiency of Open Flame and Mantle Burners 59

APPENDIX D 61-62

Standards in Other Places 61

APPENDIX E 63-67

Calorimetry 63

Photometry 65

Instruments Used in Investigation and Calibration Work of Public
Service Commission 66

Efficiencies of Calorimeters Determined by Public Service Commission 67

APPENDIX F 69_ 81

Reprint of Pamphlet, "Calorimetric Rules, Regulations and Specifi-
cations ' ' Used During Investigation 69

APPENDIX G : 82-94

Reprint of Pamphlet, "Plan of Calorimetric Investigation and
Explanation of Test and Report Forms" Used During Investigation 82

M259795

/V-f

CONTENTS.

Page

Title Page 1

Joint Committee's Letter Submitting Its Report to the Commission.... 5
Report of Joint Committee 7

APPENDIX A 15-19

Origin of Investigation 15

Organization of Committee 15

Classification of Companies Making Tests Table 2 16

Definition of Heating Value of Gas (Footnote) 17

Calorimetric Tests, Number Reported by Each Company, Table II. .. 18
Photometric Tests, Number Reported by Each Company, Table III. 18

APPENDIX B 20-41

Tabulation of the Results of the Investigation 20

Monthly Averages Heat Units and Candle Power of Each Company,

Table P7 20

Plant Data and Graphical Illustration of Table IV., Together with
Average Annual Heat Units and Candle Power and Maximum
and Minimum Variations Occurring Each Month in Heat Units

and Candle Power 22-39

Chart Showing Indefinite Relation Between Heating Value and
Illuminating Value in Manufactured Gas 41

APPENDIX C 42-60

Manufacturing and Distribution with Reference to Illuminating

Value and Heating Value 42

Unenriched Coal Gas 42

Recent Development in Coal Gas Manufactured 44

Coke Oven Gas 45

Carburetted "Water Gas 46

Effect of Distribution on the Heating and Illuminating Value 47

Compression and Transmission Tests on Enriched Coke Oven Gas . . 54
Laboratory Experiments, Compression and Freezing Carburetted

Water Gas 56

Heating Value Calculated by Analysis 56

Comparison of Continuous and Intermittant Operation 59

Comparison of Efficiency of Open Flame and Mantle Burners 59

APPENDIX D 61-62

Standards in Other Places 61

APPENDIX E 63-67

Calorimetry 63

Photometry 65

Instruments Used in Investigation and Calibration Work of Public

Service Commission 66

Efficiencies of Calorimeters Determined by Public Service Commission 67

APPENDIX F 69-81

Reprint of Pamphlet, "Calorimetric Rules, Regulations and Specifi-
cations" Used During Investigation 69

APPENDIX G : 82-94

Reprint of Pamphlet, "Plan of Calorimetric Investigation and
Explanation of Test and Report Forms ' ' Used During Investigation 82

M259795

March 6, 1913.
Honorable F. W. STEVENS, Chairman,

Public Service Commission,
Second District,

Albany, New York.

Sir:

On December 8, 1909, your Honorable Commission issued a circular to cor-
porations engaged in furnishing or distributing coal gas, water gas and
mixed gas within your jurisdiction, and appointed February 1, 1910, as a
date for conference to interchange views on the necessity for a calorific
standard and all questions necessary and incidental thereto.

On February 1, 1910, the representatives attending appointed a committee
to co-operate with the Commission in the consideration of these questions, and
thereupon your Honorable Commission appointed representatives to meet with
this Committee. After a preliminary meeting on the same date, the represen-
tatives of the companies and of the Commission organized as a "Joint Com-
mittee on Calorimetry. "

Since that date the investigation of this subject has continued and the
history of the work and matters relating thereto will be found in the report
transmitted herewith.

In accordance with our instructions: "If there is anything in yoflr
conclusion that requires the action of this Commission in any way, we shall
expect that it shall be reported to us, and we will take it into consideration
as to whether it is the proper thing for the Commission to do," we would
respectfully direct your attention to paragraphs 12, 26, 35, 37 and 41, of the
report herewith.

We have the honor to be,

Very respectfully yours,

W. R. ADDICKS, Chairman.
T. R. BEAL,
M. J. BRAYTON,
H. H. CROWELL,
J. C. DeLONG,
A. H. ELLIOTT,
J. B. KLUMPP,
C. F. LEONARD,
WM. McCLELLAN,
W. T. MORRIS,
R. M. SEARLE,
C. H. STONE,
C. H. B. CHAPIN, Secretary.

REPORT OF JOINT COMMITTEE ON CALORIMETRY.

1. The first commercial distribution of artificial gas for illumination was
in open luminous flames and quite naturally its quality was stated in terms of
the most convenient unit at hand the candle. With the introduction of the
much more efficient mantle burner, and the increasing use of heating devices,
the heating value of the gas became important. As a result, scientific men in
both Europe and America have recognized that to continue the use of the
candle power (illuminating) standard was, for modern conditions, illogical and
unsatisfactory, and in lieu thereof have advocated the adoption of a heat unit
standard. More than four years ago the Public Service Commission of the
Second District of the State of New York noted the trend of development, and
started an investigation of the actual conditions existing throughout the Second
District. This led to the appointment of a Joint Committee on Calorimetry,
composed of representatives of the Commission and of the Gas Corporations of
the State.

2. This Committee, after three years of continuous research and investiga-
tion, having had the assistance of the laboratories of the Commission and of
tests made at sixteen gas plants in the State, and the results of numerous
experiments conducted elsewhere to aid it in its conclusions, now makes its
report.

3. The object constantly in mind has been the selection of a standard for
artificial gas which will enable the consumer to obtain the most value for the
least money, and will enable the Company to obtain its profit at the smallest
expense to the consumer. The interests of the consumer and the Company are
one. This one interest demands a standard which will fit in with present eco-
nomic conditions, which will permit the most efficient use of modern invention
and which will conserve resources instead of wasting them.

4. The yielding of the open flame burner, the only device requiring the
gas to have an illuminating value, is the first reason for suggesting a standard
based on the heating value. The mantle burner is from four to eight times as
efficient as the open flame burner, and its use reduces the cost of lighting to
the consumer. As is well known, the light is obtained by heating a mantle of
rare earths to incandescence. The gas needs only heating value because the
burner is merely a heater for the mantle.

5. As in all heating devices the burner is adjusted so that the gas is com-
pletely burned and shows a blue or almost colorless flame. Consumers, if
properly informed, would substitute mantle burners for open flames in prac-
tically every case. In addition to the greater economy there is greater safety
in many cases arid more effective illumination always.

6. In addition to modern gas lighting devices which require heating value
only in the gas, there is a rapidly growing demand for gas for cooking and
heating purposes. Artificial gas is being supplied in increased amounts for
melting, tempering, metal finishing, drying, gas engines and hundreds of
other industrial uses. Inventors are actively at work designing apparatus
which will greatly increase this use. Heal storage furnaces for heating
buildings economically with gas are proposed. Indeed it seems to be true
that it only needs the design of proper gas-using apparatus to make gas the
most economic means of transporting the heat content of coal. Under such

circumstances to give artificial gas an expensive and unnecessary illuminat-
ing value is illogical and indefensibly wasteful.

7. The illuminating quality in gas, which, with the disappearance of the
open flame burner becomes unnecessary, may become a costly feature if it must
be added to the gas by a special process of enrichment. This enrichment is
usually made by means of a petroleum oil which for a number of years was
worthless for anything else and consequently was very cheap. But the enor-
mous growth in the demand for gasoline for automobiles and motor boats has
stimulated chemists to invent processes by which the enriching oils hereto-
fore used by gas companies can be turned into light oils suitable for internal
combustion engines.

8. Inventors of oil engines are perfecting their devices rapidly, which
results in much more extended direct use of oil for power generation. Oil
used in this way commands a higher price than when used for gas enrichment.
The United States and other governments are resorting to increased use of oil
fuel for war vessels, and their needs are so paramount that price is not a
critical factor. (Reference: "The Production of Petroleum in 1911," U. S.
Geological Survey, 1912.)

9. This sudden demand for enriching oil products by the people for
pleasure and industrial purposes, and by governments for power purposes, and
the consequent rise in the selling prices, has within a year increased the cost
of manufacturing water gas from 10 to 15 cents per thousand cubic feet.
In addition there is every reason to believe that the present price of
oil is by no means the maximum, so that cost may operate in the future to
require that enrichment be kept to a minimum. It was the presence of a large
and cheap supply of enriching oil that made water gas commercial after the
manufacturing apparatus had been made practical from 1877 to 1882. It is
probable that the high price of enriching oils will make carburetted water gas
useful chiefly for peak demands and as a reserve to retort gas and oven gas,
which need no enrichment if heating value only be required.

10. The present rise in the price of oil would result in a condition seriously
affecting the price of gas to the consumer if it were necessary for artificial gas
to 'continue to have the present high illuminating value. Fortunately the
availability of the mantle burner modifies the seriousness of the situation.
It may be argued that gas oil has risen in price before and afterward dropped.
It must be added, however, that the price never returns to its previous low
figure. Moreover, as shown above, the present rise is due to plainly apparent
and quite natural causes, and it does not appear that these causes will
abate in force.

11. In passing it may be stated that water gas must be enriched to be
practical for community use. Retort gas (so-called "coal gas") has ample
heating value and illuminating value to be distributed without enrichment to
the community. Run of oven gas (by-product from coke ovens) has a large
heating value without enrichment, but in candle power is materially lower than
retort gas.

12. It should also be noted that whatever reasons there may have been
in the past for different standards for coal gas, mixed gas and carburetted
water gas (16, 18 and 20 candle power in New York State, Second District),
they certainly are without force now, and only one standard is necessary or
desirable.

13. Mere increased cost, though important and almost compelling, is not
alone the cause for a change from a candle power to a heat unit standard. As
a matter of fact, when this Committee was appointed this feature could not
have been in any degree a reason for changing the standard. The present
standard actually retards the extension of gas service and as a direct conse-
quence retards the development of communities.

14. Present development in gas distribution falls into two classes first,
distribution in comparatively densely populated large territories such as cities,
with closely attached suburbs; and second, distribution of gas from one large
central plant to a number of more or less distant communities with intervening
territory in which there is little or no demand for gas supply. In either case
the present candle power standard is a burden. This is for the reason that
because of temperature and pressure changes and friction in mains a
part of the enrichment added to a gas to give it illuminating power drops
out during transmission, and the loss becomes more and more serious as the
distance of transmission increases. Higher pressures are necessary if the
gas is to be transmitted economically for a long distance and it is impossible
to avoid some exposure to low temperature. As a result either the gas must
be given excessive candle power at the plant, or it must be enriched after
transmission so that the gas distributed after transmission may be up to
standard. In 'either case it is sometimes difficult to make the operation of the
system satisfactory and the cost increases to an amount which makes such
distribution often commercially impracticable.

15. Within single areas or communities, extension of service is possible
so far as the present law is concerned which requires inspections to be made
about a mile from the works. This does not mean, however, that the same
quality of gas can be supplied economically at the center and on' the outskirts.
The company undertaking to give standard service at all points must of
necessity spend much more on its manufacturing and distribution cost because
the average candle power must be higher in order to make up the loss.

16. In the case of one large central plant distributing gas to a number of
more or less distant and separate communities, the burden is especially heavy,
for but one quality of gas can be ordinarily distributed from the plant. The
long distribution system with its higher pressure and exposure to low tem-
perature entails a very great loss in candle power during transmission. In
addition, the operation is likely to be difficult, because the enriching oils
which condense in the system must be taken care of in larger pipes, traps
and other devices and the labor cost of operation is increased on account of
the maintenance and operation of these extra devices. With the increased
cost of enriching oil it is probable that such extended distributions will not
be possible without a serious increase in the selling price.

17. Too small a community cannot support a gas plant of its own, if
first-class service is to be given, ample financial support secured and
adequate business and engineering superintendence supplied. For a long
time the same conditions obtained in the supply of electricity but the
problem of supplying the smaller community has been solved by the develop-
ment of high-tension, long-distance transmission of power. By this means
any number of small communities and intervening farm territory can be
served from one large central station. High-pressure gas distribution bears
the same relation to the gas industry as high-tension transmission bears to
the electric industry.

18. As shown later, the loss in heat units in transmission due to pressure
or low temperature is very much less than the loss in candle power. A heat
unit standard not very different from the heat unit value of gas at present
supplied would permit gas distribution over long distances under pressure at
a loss which would be in no sense burdensome. Such a result would permit,
as soon as development could take place, gas service to many small villages
and towns which it is quite impossible to supply under present conditions.

19. A further reason why a change from a candle power standard to a
heat unit standard is desirable rests on a broad economic policy. Even though
oil were not increasing in price the present standard spells waste. It is a
waste of resources and it is wasteful of money. Conservation of resources

would demand that there should be no unnecessary resort to the use of oil for
gas enrichment. It is wasteful to maintain a standard beyond what is re-
quired for efficiency and when the standard means an unnecessarily high cost.
The public wants the best gas for the least money and it is to the business
advantage of the Company to supply the demand. The present standards,
under existing conditions, do not assist in attaining this desirable end.

20. Summarizing then, the movement toward a heat unit standard is
based on three important factors :

1. Modern appliances for the use of gas require that it have heat-
ing value only. The open flame burner is rapidly disappearing
on account of its inefficiency and expense.

2. The present candle power standard seriously impedes desirable
distribution in extended communities and for long distances, and
as a consequence retards cpmmunity development. The rising
price of enriching oils adds to the difficulty.

.3. The present standards are wasteful of resources and unduly bur-
densome on the consumer and the Company.

21. In order to obtain accurate information on which to base the choice
of a proper standard, particularly with reference to the needs of New York
State, the Committee turned to a number of gas corporations of the State for
assistance. Laboratories for calorimetrical measurements were established at
sixteen different plants of the State and regular daily tests started. The in-
struments were checked first at the laboratory of the Public Service Commis-
sion at Albany. The cost of the apparatus and the expense of the tests were
all carried as operating expenses of the plants where the tests were made.
Constant attention had to be given by the Company's officers and their
employees to the investigation, and the expenditure of time and money was
not small. Result's of this work make up the most valuable data that the Com-
mittee has in this report. In Appendix B will be found tabulations and curves
showing the results obtained by the various Companies with comments and
discussion in considerable detail (see also Appendix C). The monthly reports
of the Companies summarizing their daily tests when received by the Commit-
tee were scrutinized closely for errors and critical features. Every effort
possible has been made by the Committee to make sure that the work was
being done with uniformity and accuracy. The co-operation of the traveling
gas inspectors of the Commission was of marked assistance in this respect.
As a final test on this point, a demonstration was held at Amsterdam, N. Y.,
at which all the calorimeter operators of the various Companies making tests
were present. This gave an opportunity for a further demonstration in regard
to uniformity and accuracy. The Committee feels confident that the results
are accurate within one per cent.

22. Certain other important facts demonstrated by this experimental
work should be mentioned.

23.. It is known that a calorimetrical laboratory can be established at
comparatively small expense.

24. Calorimetric measurements can be made with great accuracy by men
with no special scientific training except experience in and attention to proper
operating directions.

25. The calorimeter as a practical instrument is more accurate than the
photometer. There is no uncertain feature in connection with its use as there
is with the type of burner and standard unit of light used with the photometer.

26. From the test results no law could be discovered showing a relation
between the candle power and the heat unit value of artificial gas. The Com-

mission's preliminary investigation indicated this, but the results, involving
6,738 calorimetric and 9,167 photometric observations, obtained by the
Committee make it a demonstrated fact.* For this reason it would be
very difficult indeed to state the heat value of artificial gas of a quality equal
to the State standard for candle power inasmuch as the Companies generally
distributed gas above the legal standard, in some cases as much as 17 per cent.
For the information of the Committee, however, two plants were operated
close to the State standard. As the results in Appendix B show, gas meeting
the State standard of candle power would have approximately a monthly
average of 585 B. t. u. ** The question immediately arises as to whether
this value should not be taken for the heating value standard of gas to be
distributed in New York State. The several steps in the reasoning necessary
to properly answer this question are important.

27. It is desirable that a new standard shall not differ greatly from the
heating value of gas of the present legal standard. To have it materially less
would require the distribution and use of a larger volume of gas in order to get
the same useful effect. This in turn would necessitate radical changes in the
selling price annoying to both consumers and Companies without benefit to
either.

28. To meet a heat unit standard of 585 B. t. u. means that most
Companies must enrich the product during a portion of each twelve months.

29. There are a variety of combination methods of making artificial gas
from gas coal, anthracite coal, bituminous coal and; oil, which are discussed in
Appendix C.

30. Any enrichment is expensive and it has been shown above that it is
becoming more and more so with the increasing price of oil. It is safe to pre-,
diet that if the present price of oil continues, carburetted water gas will no
longer occupy the important position that it has for some years past in the
gas industry. Indeed, the idea is now taking firm hold that, owing to the oil
situation, with the practically inexhaustible supply of gas coal now in
sight, the gas industry must depend upon coal gas of some sort for the bulk
of its output and use water gas as a reserve. In any case excessive enrich-
ment is useless and unsatisfactory, especially in connection with modern gas
appliances. Gas unnecessarily enriched interferes with manufacturing pro-
cesses, and when distributed to the consumer deposits carbon in burners and
mantles and, as heretofore stated, the .illuminants drop out in transmission,
especially under pressure and at low temperature. Other things being equal,
it will be to the advantage of consumers and manufacturers if enrichment
is reduced to a minimum.

31. As shown later, with the most modern horizontal retort settings
and machine stoking, coal gas from high-grade gas co;als and with, high
yields of gas per ton of coal, without enrichment, varies in heat units from
approximately 550 to 600 B. t. u. monthly average. If the general use of
carburetted water gas as a staple product becomes impossible on account of
the very high price of enriching oils, and must be replaced by retort or oven
gas, and if the heat unit standard is set at such a point that the manufacturer
will need the highest grades of coal in order to meet this standard or else
be compelled to use high-priced enriching oils, it is obvious that the price
of these higher grade coals will rise so that the very object of the change

* Note Chart Appendix B, pages 40-41.

** B. t. u. is the accepted abbreviation for the British thermal unrt, which is the amount of heat
required to raise the temperature of one pound (avoidupois) of pure water from 39.1 F. to 40.1 F.
The variation in the quantity of heat necessary to raise the temperature of a pound of water one
degree F. is so slight for any temperature between 32 and 212, that in general the B. t. u. may be
safely taken as the amount of heat necessary to raise the temperature of one pound (avoidupois) of
water one degree F.

will be defeated. It is interesting to quote here from Bulletin 6 of the

Bureau of Mines of the United States, published in 1911:

"In a consideration of the various means whereby more eco-
nomical and more efficient use may be made of the fuels in the United
States, the possibility of obtaining for the production of illuminating
gas other and cheaper fuels than the Pennsylvania coals demands at-
tention. For the Government, as well as for private corporations and
the householder, there can be no more economical and efficient way
of using some coals than through the medium of illuminating gas.
In the stove, gas reduces the labor cost of heat production and lessens
the drudgery of the kitchen ; burned in the Welsbach mantle, it is an
excellent and cheap illuminant. In addition, the coke that remains
after the gas has been recovered furnishes a smokeless fuel that has
about the same heating value as anthracite. Hence any investiga-
tions that will indicate how local coals through proper treatment
may be substituted for the higher priced and rapidly vanishing
Pennsylvania gas coals will bring about lower prices for both gas and
coke, and will also aid to conserve for use in metallurgical processes

the coking coals of Pennsylvania and of other States.

*******

"There are few well-developed coal fields in this country that
furnish coal satisfying all the requirements of illuminating-gas manu-
facture. Most of the coal used hitherto has come from Western
Pennsylvania, the quantity supplied by other fields being relatively
small. The introduction of gas-coals from new or little-known dis-
tricts, because of the lack of necessary testing stations and of scien-
tific study of the complex process of gas manufacture, has been dif-
ficult."

32. We must, therefore, think that it would be inadvisable to set the
standard for artificial gas so high that the best coals only could be used. The
standard should be placed so that average coals may be used without enrich-
ment, and thus give the very greatest economic value to the consumer at the
lowest cost.

33. Certain methods of operation are now being discussed that may be
desirable, or even become compulsory under conditions which seem to be ap-
proaching. The disposition of the coke resulting from the manufacturing of
coal gas has been in the past a serious problem to some Companies, and at a
time when coke was used in cooking ranges since discarded for more desirable
gas ranges. For this and other reasons it may be desirable in the future to
manufacture a mixed coal and carburetted water gas, using substantially all
of the coke as fuel in the water gas sets. If this becomes a general practice it
may be desirable to lower the standard. Coke oven gas in which the coal
is carbonized primarily to obtain coke for industrial purposes and the gas a
by-product is also being considered in many places. Run-of-oven gas would
require excessive enrichment if the present standard was in force. It is
quite probable that should this coke oven gas be distributed in larger
quantities it would be desirable to reduce the standard. Present data from
these various processes show that it might be necessary to fix the standard
at 525 B. t. u. or even lower.

34. It is difficult indeed, in view of the uncertainty as to just how fast
certain changes in the conditions governing gas manufacture and distribu-
tion will take place, and as to what the final situation will be, to determine the
proper value at which to set the standard. It has been shown that some time
in the future the standard may have to be 525 units or lower. It has also been
shown that, at present, the monthly average, even with the best coals and
highest grade plants, may be as low as 550 units. All plants, of various sizes
and locations, cannot become highest grade plants, at least immediately, and

the smaller plants never. The best coals are not available to all, and if the
demand is increased the price will rise. Notwithstanding these facts it is
believed that the standard adopted must be close to the heat unit value of the
present standard gas.

35. Taking all these conflicting factors into consideration, it is the
judgment of the Committee that a total heat value not exceeding 570 British
thermal units monthly average measured at the point where the gas leaves
the manufacturing plant, corrected to a temperature of 60 F., and to a
pressure of 30 inches of mercury, as measured by the rules of the Committee
accompanying this report, is the standard which will best serve the interest
of the people of New York State.

36. The standard suggested above is referred to the standard atmo-
spheric cubic foot, i. e., at 30 inches barometer and 60 F. It will be
perceived that the only time a consumer would get the standard number of
heat units would be when his meter was at 60 F. and the barometer was
at 30 inches. Such conditions cannot obtain, however, with localities at
different heights above sea level and with meters located in all kinds of
places giving different and varying temperatures. Therefore, some average
conditions must be chosen. These might be the average annual barometer
and temperature if they could be obtained for each locality and a "local
cubic foot" might be fixed on. these terms. All such "local cubic feet" could
then be required to have the standard number of heat units. This would be
possible for a group of localities not varying too much from a certain
average altitude. It would be very inconvenient however. A certain mass
of coal gives a certain mass of gas at best economic yield, and the volume
of the gas is solely dependent upon pressure and temperature. Therefore, if
a "local cubic foot" is used, operators would operate differently at different
altitudes and temperatures, even though using the same coals, oils and
machinery. A comparison of detail methods of operation, the study of
proper amounts of oil and steam, temperature of various parts of the sets or
benches and other features, are sufficiently complex now without making
them more so by introducing accidental atmospheric conditions. There could
not be even a mere comparison of results by State authorities and others
interested, in order to increase efficiency, until the results were brought
to a common basis. In a State having largely different altitudes several stand-
ards might be required owing to the impossibility of making a uniform
commercial gas in all cases. The operators would still have to observe the
daily barometer and temperature, and make corrections to the "local cubic
foot." The only suggested advantage discernible is that the consumers
everywhere throughout the region or State in question would get the same
number of heat units in the yearly average "local cubic foot." "What they get
from day to day will vary by the same amounts under any system. All features
considered, it will be much more satisfactory to fix the requirement in terms of
the atmospheric standard cubic foot, i. e., at 30 inches barometer and 60 F.
The average "local cubic foot" sold will then contain slightly different num-
bers of heat units according to the height of the locality above the sea and to
the climatic conditions. In New York State these differences are
unimportant.

37. The conditions governing the use of the standard are important.
Gas manufacture is not an exact science but is a complex operation including
a number of distinct processes. Quality of coal, methods of firing, tempera-
ture of retorts, the human factor, the failure or breakdown of parts of the
plant, and other factors not easily controlled, make it impossible for a Gas
Company to deliver an absolutely uniform product. This points to the
necessity of applying the standard as an average for a reasonable length
of time. A month has been adopted elsewhere and is recommended for
New York State. If a Company falls below the standard for a few days it

will then be necessary for it to produce above the standard, at an economic
loss, in order to have its monthly average satisfactory. In order to protect
the public against improper management by which there would be wide
departures from the standard, should a minimum value be set? It is not
necessary that this minimum be set too close to the monthly average, as
there is a financial loss to a Company if it departs too far from it. The
cheapest and best operation for both Company and consumer will obtain by
a close adherence to the standard. A wide departure due to careless operat-
ing means an increase in operating cost which will not be to the Company's
profit. A 5 per cent, deviation for not exceeding three consecutive days
would be adequate protection to the consumer. In extraordinary conditions
due to failure to obtain supplies or to accident in the plant, the Commission
might properly suspend the operation of the standard in its discretion.

38. As a matter of fact even a properly fixed minimum is of little prac-
tical importance. Well-managed companies would never reach it except under
circumstances absolutely beyond their control. The saving and satisfaction in
operating close to the monthly average is very great and induces good manage-
ment. A management continually inefficient and incompetent would be
exposed in so many ways that a change would eventually come through
reorganization or new ownership.

39. Penalties have been used in an attempt to compel good management,
but as a rule, experience has shown them to be ineffective. The difficulties
of placing the blame on the proper persons and conditions, of proper legal
phrasing, of collecting the penalties, of fixing equitable penalties and penalties
that are real, the fact that through carelessness they so frequently fall into
disuse, the opportunity that exists for abuse and persecution all operate
against the effectiveness of a penalty system. Continual and broad publicity
is very much better. The greatest force in the country to-day is public opinion.
No company could ignore or withstand the effect of frequently published state-
ments that its product was not up to a prescribed standard. A weekly publica-
tion of tests, for example, would keep the public informed, would keep the
company active in good management, would prevent careless and irresponsible
complaints, and would prevent abuse and criticism.

40. It is reasonable to ask what disadvantage there will be, if any, to
persons using flat flame burners if a standard is fixed according to heating
value only. It is fair to exclude from consideration all persons who continue
to use flat flame burners through indifference to their own interests.
That a smaller and smaller number of people are doing this is evident from
the results reported by Gas Companies in regard to the reduction in the num-
ber of consumers using open flames. Mantle burners have become so cheap
and the saving is so great that in a short time no one will use open flame
burners except for some peculiar reason. The cases will be remarkably few
where open flame burners will be thought desirable, but for those who feel
that they must use them it may be stated positively that any artificial gas hav-
ing the heating value recommended in the above standard would have sufficient
illuminating power, though at times lower than at present, to make the gas
useful in locations suitable to open flame burners. The use of a very small
percentage of the gas for such a purpose should not prevail against the
general usefulness of the whole product.

41. The Committee recommends, therefore, that no candle-power stand-
ards be considered in connection with the heat unit standard heretofore
recommended.

APPENDIX A
HISTORY OF COMMITTEE AND ITS WORK

1. In August, 1908, an investigation was started by the Public Service
Cqmmission, Second District, N. Y., through its Division of Light, Heat 'and
Power "into the subject of the calorific power and illuminating power of the
coal gas, carburetted water gas, and mixed coal and earburetted water gas
supplied." (Page 21, Third Annual Report.)

2. This examination, as stated in the Third Annual Report of the Com-
mission, was of a preliminary nature, and was completed in October, 1909,
and the data embodied in a report by the Chief of Division of Light, Heat
and Power.

3. On December 8, 1909, notice was sent by the Commission to all the gas
companies operating in the Second Public Service District of a conference to
be held on February 1, 1910, in reference to this subject.

4. In December, 1909, following the receipt of this notice and report,- the
Empire State Gas and Electric Association appointed a Committee to investi-
gate the matter as thoroughly as might be done prior to the hearing of Febru-
ary first. This Committee held a number of meetings, discussed the matter
contained in the report and such other data as was available, but was unable
to arrive at any definite conclusion in the 'very limited time at its disposal.

5. At the hearing on February 1, 1910, after some general discussion,
a vote was taken on the question as to whether or not the investigation started
by the Commission should be continued. The result of the vote being in the
affirmative, the Chairman of the Commission suggested the appointment by the
representatives of the gas companies present, of a Committee to co-operate
with the Commission's representatives. This suggestion having met with the ap-
proval of all those present, a recess was declared, during which the companies
held a meeting and elected as their representatives :

W. R. Addicks,
T. R. Beal,
J. C. DeLong,
W. T. Morris,
M. W. Offutt,
R. M. Searle.

6. Upon the continuation of the conference the Chairman of the Com-
mission named as its representatives :

H. C. Hazzard,
H. H. Crowell,
C. H. Stone:

The persons above named convened after adjournment of the hearing and
voted to hold the first regular meeting in the Capitol, Albany, on Friday,
February 11.

7. On February 11, 1910, the Committee appointed as above outlined,
met and elected H. C. Hazzard, Chairman, and C. H. B. Chapin, Secretary.
It was voted that the Committee should be known as the Joint Committee on
Calorimetry.

8. Since its original appointment, the personnel of the Committee has
undergone some changes. The Commission has appointed William McClellan
and C. F. Leonard as its representatives, H. H. Crowell and C. H. Stone hav-
ing severed their connection with it. H. H. Crowell continued to serve upon
the committee, and C. H. Stone resigned, but by unanimous invitation con-
tinued to sit with the committee and was later re-elected a member. M. "W.
Offutt resigned as a member of the Committee and M. J. Brayton was elected
in his place. Dr. A. H. Elliott and J. B. Klumpp were elected additional
members of the Committee. H. C. Hazzard having resigned from the service
of the Commission, thereupon resigned from the Committee, and W. R. Addicks
was elected Chairman.

9. At the commencement of the investigation, the Committee deemed it
desirable to secure the co-operation of Companies in different parts of the
State and operating under different conditions of manufacture and distribu-
tion of gas. Ten Companies decided to purchase calorimeters and make
such tests as the Committee desired. Before the conclusion of the investiga-
tion additional Companies joined in the work, so that the Committee had
results from sixteen plants located in widely separated parts of the State to
aid it in its conclusions.

10. Statistics are given in Table I showing the kind of gas made by these
Companies, the magnitude of the daily output, and the date of beginning of
tests. Companies are designated by number instead of by name throughout
the report. (For further information regarding the different Companies see
Appendix B.)

TABLE I.
Company

Number. Class. Kind of Gas. Tests Started.

1 A Coal gas, enriched. Oct. 1, 1911.

2 D Coal gas, enriched. Aug. 1, 1911.

3 A Carburetted water gas. Aug. 1, 1911.

4 A Carburetted water gas. Aug. 1, 1911.

5 C Carburetted water gas. Aug. 1, 1911.*

6 B Carburetted water gas. Aug. 1, 1911.

7 B Carburetted water gas. Aug. 1, 1911.

8 C Carburetted water gas. Aug. 1, 1911.

9 A Carburetted water gas. Aug. 1, 1911.

10 B Carburetted water gas. Oct. 1, 1911.

11 A Mixed coal and carb 'd water gas. Aug. 1, 1911.

12 A Mixed coal and carb 'd water gas. Aug. 1, 1911.

13 C Mixed coal and carb 'd water gas. Aug. 1, 1911.

14 C Mixed coal and carb 'd water gas. Apr. 1, 1912.

15 A Carburetted water gas. Oct. 1, 1912.

16 B Carburetted water gas. Feb. 1, 1912.

*Tests discontinued November 30, 1911, and calorimeter moved to another
plant.

Class A Companies having a maximum daily send-out of over 1,000,000

cubic feet.
Class B Companies having a maximum daily send-out from 500,000 to

1,000,000 cubic feet.

Class C Companies having a maximum daily send-out from 100,000 to
500,000 cubic feet.

Class D Companies having a maximum daily send-out of under 100,000
cubic feet.

11. At the meeting of the Committee in February, 1910, it was deemed
advisable to prepare specifications for calorimeter installations and rules for
their operation. This work, which was done by a sub-committee, was com-
pleted and adopted by the full Committee on May 6, 1910, and printed for dis-
tribution under the title ' ' Calorimetric Eules, Regulations and Specifications."
Copies were furnished to all Gas Companies operating in New York State.

This pamphlet is divided into six general sections as follows :

I. Heating Value of Gas (Definition).*

II. Primary Standard To be maintained at the laboratory of
Commission at Albany (Specifications).

III. Secondary Standard To be used in checking Calorimeters of

Gas Companies in situ (Specifications).

IV. General Specifications and Recommendations for Calorimeter

Installations by Gas Companies.
V. Directions for Operating Calorimeter.

VI. Suggestion of Several Types of Calorimeters Suitable to Use
when Checked by the Primary Standard Adopted.

12. Following the adoption of these specifications, the Public Service
Commission, Second District, purchased necessary instruments and equipped
a laboratory where the instruments of the different Companies could be cali-
brated.

13. The delays in delivery of instruments were considerable, so that the
calibration of the companies' calorimeters at the State laboratory was not com-
pleted until early in 1911. It was deemed wise by the Committee to allow a
preliminary period after the instruments were finally installed for the com-
panies' operators to become acquainted with the methods of testing before
asking that the results be submitted to it for inspection.

14. During this preliminary period forms were prepared to be used by
the companies in recording their daily readings and in submitting the results
each month to the Committee.

15. Observations of the results obtained during the first few months of
testing prompted the Committee to prepare a second pamphlet which was
printed under the title "Plan of Calorimetric Investigation and Explanation
of Test and Report Forms." A copy of this pamphlet was furnished each com-
pany engaged in the investigation. Amended forms for recording and report-
ing daily readings and works data were also prepared.

16. This second pamphlet, which was tentatively adopted January 6,
1912, treated in further detail the following subjects:

1. The making of daily Calorimetric tests and the recording daily
of certain works data.

2. The submitting to the Committee monthly the results of the daily
tests and of monthly averages and details of works data.

3. The furnishing to the Committee of information regarding oper-
ating conditions, and apparatus and methods in use.

17. Beginning with August 1, 1911, and ending October 31, 1912, a period
of fifteen months, reports have been regularly received by the Committee and
each month tabulated by the Secretary so that copies could be in the hands of
each member of the Committee for individual study. During this period 6,738
Calorimetric tests and 9,167 photometric tests were reported as shown in
Tables II and III respectively.

* The definition of the heating value of gas adopted by the Committee for the purposes of
this report and the investigations conducted is as follows:

"The heating value of a gas is the total heating effect produced by the complete
combustion of a unit volume of the gas, measured at a temperature of 60 degrees
Fahrenheit, and a pressure of 30 inches of mercury, with air of the same temperature and
pressure, the products of combustion also being brought to this temperature.

"In America the unit of volume is the cubic foot and we recommend that the
heating value be stated in terms of British Thermal Units per cubic foot of gas."

TABLE H.

CALOEIMETEIG TESTS

i-H

Company

rH
4-T

CD
X5

t-4

CD
1

i 1

4-a

-tj
Oi

O
-W

CD
>

CD
t>

3
a

t-t

-IJ

"cS

fr.

1-5

No. 1

1 25

252

No. 2

387

No. 3

354

No. 4

400

No. 5

103

No. 6

433

Works

< i it

267

Outlying Station

No. 7

376

No. 8

383

Inst. A at Works

it 1 1

382

Inst. B at Works

It S t

380

Inst. C at Office

No. 9

397

No. 10

322

No. 11

376

Coal Gas

it tt

376

Water Gas

it tt

378

Mixed Gas

No. 12

314

No. 13

374

No. 14

210

No. 15

Works

it it

Outlying Station

No. 16

201

Total

409

385

464

435

394

414

405

444

470

481

472

487

498

466

514

6,738

TABLE HI.
PHOTOMETEIC TESTS

r-i

Company

)tember

t-,

vember

h
Q

CD
O

j>>

ptember

O
-t-i

^*^

r-i

^ .

No. 1

' 25

278

Works

No. 3

406

Works

11 it

1 25

351

Office

No. 4

' 32

414

Works

No. 5

103

Meter Shop

No. 6

458

Works

tt ti

305

Outlying Station

No. 7

396

Works

tt it

370

Office

No. 8

383

Works

1 1 ft

383

Works

tt it

383

Office

it it

383

Office

No. 9

457

Works

a a

402

Office

it tt

402

Office

No. 10

324

Office

No. 11

376

Coal Gas

it ti

378

Water Gas

it a

3.0

392

Mixed Gas

No. 12

420

Works

it ft

270

Office

No. 13

385

Office

No. 14

213

Works

tt it

211

Office

No. 15

Works

Outlying Station

No. 16

226

Meter Shop

Total

562

538

615

577

548

603

557

598

623

'680

653

629

645

651

688

9,167

18. In May, 1912, a meeting of the Committee was held at Amsterdam
which was attended by the men operating the calorimeters in the several
plants. At this time a general conference was held and a discussion of the
work, with particular reference to uniformity and accuracy, took place.

19. Sub-committees have taken up in detail matters that were considered
of enough importance to require special study. Frequent meetings of the
Committee have been held during the past two years and the work con-
stantly reviewed with an endeavor to consider every phase of the question.
An analysis of the work, the results of the tests and the conclusions drawn
therefrom, will be found elsewhere in the report.

INTRODUCTORY OBSERVATIONS RELATING TO THE STUDY OF

APPENDIX B.

1. Laboratory accuracy cannot be applied in commercial gas produc-
tion. The engineer cannot predict from day to day the quality of gas that
will be produced, not only because of the uncertainties in the character of the
raw material, but also because of climatic conditions. It will be observed
therefore that, owing chiefly to atmospheric changes, an excess candle power
exceeding 10% at the plants is frequently not realized at official testing
station, even though the minimum realized meets the State candle power
requirements. This necessary condition tends to the serving of an irregular
product which the charts clearly disclose.

2. Similarly, when operating under a heat unit standard, the engineer
must continue to make his product in excess of the standard adopted. The
information derived by the test indicates that the consumer will receive a
much more uniform and satisfactory product which should work for greater
efficiency in its use at the point of consumption when compared with operat-
ing under the candle power standard where, even with uniform pressure, the
essential readjustment of air supply is neglected; this is wasteful in use of
gas and through carbonization (a too familiar sight with over-enriched gas)
is destructive of gas mantles. This condition would be eliminated under the
proposed heat unit standard and the present economic losses and annoyance in
the use of gas due to this neglect in readjustment of air supply will be
eliminated.

3. The adoption of the proposed standard will be a conservation of
resources through the elimination of unnecessary wastes in production and
distribution without loss in effectiveness of the product when compared with
all elements of waste resulting from pursuing present methods.

4. It should be noted that a percentage variation from a standard by, for
example, 5% is but 1 unit in the case of 20 as used in candle power, while the
same accuracy when dealing with the larger heat unit figure becomes 29
units (nearly) when dealing with heat unit standards, yet both 1 and 29 are
figures that show equal percentage accuracy.

5. It should be kept in mind that a difference in reading by two observers
of the same gas might reasonably be even .5 of a candle or nearly 3% in
candle power. It is probable that the variation in B. t. u. observation by the
same observers would be less than 1%. The following table may be found
useful. It shows, for example, that a 5% variation from 20 candle power is 1
or 21 candle power; for 18 candle power is .9 or 18.9 candle power; from 16
candle power is .8 or 16.8 ; while from 570 B. t. u. it is 28.50 or 599 (nearly).

TABLE SHOWING RESULTANT ILLUMINATING OR HEATING VALUE

FOR VARIATIONS IN THE QUALITY OF THE GAS, ABOVE

STANDARD, OF FROM 1 TO 13 PER CENT.

Standard 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% 11% 12% 13%

20 C. P 20.2 20.4 20.6 20.8 21.0 21.2 21.4 21.6 21.8 22.0 22.2 22.4 22.6

18 C. P 18.2 18.4 18.5 18.7 18.9 19.1 19.3 19.4 19.6 19.8 20.0 20.2 20.3

16 C. P 16.2 16.3 16.5 16.6 16.8 17.0 17.1 17.3 17.4 17.6 17.8 17.9 18.1

570 B. t. u.,576 581 587 593 599 604 610 616 621 627 633 638 644

APPENDIX B

1. As already noted in Appendix A, page , there have been 6,738
calorimetric tests and 9,167 photometric tests reported to the Committee dur-
ing the period August 1, 1911, to October 31, 1912. It has seemed unnecessary
to include all of these tests in detail in this report, but the monthly averages
are given in Table IV. These averages are in all cases based on the actual
number of tests made during the month. The table also gives the minimum
monthly average of heating value for each Company and the average illuminat-
ing value for the month during which the minimum average heating value
occurred.

TABLE IV.

CAEBUEETTED WATEE GAS

STANDARD CANDLE POWER REQUIRED AT TESTING STATION 20

MONTH

Company No. 3
At Works
Average

Company No. 4
At Works
Average

Company No. 5
At Meter Shop
Average

Company No. 6

Company No. 7
At Works
Average

At Works
Average

Testing Station
Average

B t. u. | C. P.

B t. u.

C. P.

.Blt-fu.

C. P.

B t. u. | C. P.

August, 1911

621

22.8

642

23.3

624

20.7

613

23.8

605

20.7

September

631

22.5

644

23.7

616

21.1

613

24.9

615

20.7

October

636

22.9

632

22.5

635

20.6

612

23.4

617

20.5

November

657

22.5

635

22.1

634

19.6

615

22.9

623

21.1

December

661

22.4

627

21.9

637

23.1

625

22.6

January, 1912

669

21.8

622

20.5

631

22.8

619

19.9

(135

25.1

February

650

23.1

645

22.0

634

23.2

618

20,4

631

25.6

March

643

24.1

631

22.4

624

22.8

620

20.4

630

25.6

April

626

22.1

627

23.3

617

20.2

628

25.5

May

628

23.3

634

23.5

638

23.2

621

20.5

621

22.5

June

622

22.1

630

22.7

643

23.2

636

20.7

610

21.7

July

622

21.3

644

22.4

646

23.1

639

21.1

607

21.1

August

619

20.6

643

22.3

636

23.1

631

20.7

613

21.5

September

624

21.3

644

22.5

630

22.9

616

20.2

619

21.5

October

630

20.7

652

22.3

638

23.4

626

20.4

629

21.6

Min. Ave. B t. u. and
C. P. same month

619

20.6

622

20.5

616

21.1

612

23.4

616

20.2

605

20.7

Company No. 8

Company No. 9

Company No. 10

Company No, 15

At Works

At Office

Testing Station

At Office

At Works

Outlying Station

At Meter Shop

Average

B t. u.

C. P.

B t. u. 1 C. P.

B t. u. I C. P.

B t. u. | C. P.

August, 1911

617

19.1

613

20.0

589

20.8

September

610

18.9

605

19.6

597

20.6

October

594

18.7

584

19.3

619

21.2

625

21.2

November

633

19.7

622

19.9

625

20.4

632

21.2

December

645

22.5

642

20.9

638

20.7

626

20.7

January, 1912

653

22.4

643

21.1

631

20.9

626

20.5

February

650

22.6

631

20.6

621

20.7

634

20.6

625

20.5

March

621

20.3

607

19.8

610

20.7

618

20.4

619

20.8

April

602

18.7

591

20.4

600

20.2

620

20.6

620

20.6

May

592

19.5

590

19.9

589

20.6

629

20.9

621

21.2

June

592

19.3

596

20.7

595

20.6

624

21.2

631

21.0

July

584

17.9

586

20.6

597

20.6

619

21.2

639

20.7

August

592

19.3

591

20.1

588

20.5

633

21.9

626

21.2

September

607

19.7

605

21.0

592

20.2

615

20.8

627

20.9

626

20.7

October

615

20.4

618

20.5

597

20.6

605

20.4

629

22.2

606

17.6

622

21.1

Min, Ave. B t. u. and
C. P. same month

584

17.9

584

19.3

588

20.5

605

20.4

627

20.9

606

17.6

619

20.8

MIXED COAL AND

COAL GAS

COAL
GAS

CARBURETTED WATER GAS

Enriched

Unenriched

Company No. 11

Company No. 12

Company No. 13

CompnnyNo. 14

Company No. 1

Company No. 2

Company No. 11

At Works

At Office

At Works

At Office

At Works

MONTH

Average

Averaje

Average

B t. u. I C. P.

B t. u. 1 C. P.

B t. u. I C. P.

B t. u.

C. P.

B t. u. | C. P.

B t. u.

C. P.

B t. u. | C. P.

August, 1911

626

20.0

590

21.3

611

18.2

6tiO

602

14.9

September

640

20.4

592

21.3

623

18.3

654

616

15.3

October

628

20.6

596

21.5

661

18.7

683

18.7

654

611

13.5

November

643

20.9

611

21.9

624

18.6

674

18.0

651

607

13.8

December

649

20.5

615

21.9

633

18.4

667

17.7

647

621

14.2

January, 1912

672

20.5

635

21.6

632

18.5

653

17.5

630

628

13.7

February

656

21.4

645

21.5

641

18.4

636

<o
*>

614

13.7

March

646

20.4

647

21.1

626

18.3

626

16.5

616

612

14.9

April

636

20.8

647

20.9

626

18.6

605

18.6

641

16.4

637

612

13.5

May

628

21.5

642

20.8

633

18.4

634

19.4

656

16.9

651

601

14.7

June

624

19.7

631

20.6

620

18.4

623

18.5

642

16.6

662

-u

610

15.3

July

610

19.9

617

20.3

635

18.6

618

18.6

638

17.0

646

593

14.3

August

614

19.7

628

20.4

627

18.4

618

18.9

645

16.8

644

591

13.7

September

612

19.8

622

20.5

621

18.4

611

18.7

653

17.0

646

595

14.2

October

623

20.4

633

20.5

625

18.5

603

18.4

650

17.8

618

592

13.1

Min. Ave. B t. u. and
C. P. same month

610

19.9

590

21.3

611

18.2

, 603

18.4

626

16.5

616

591

13.7

2. The results of the tests throughout the entire period are shown graph-
ically in the following pages. Data in regard to the works and operation of
the various Companies are also given.

3. The charts were prepared to show the variations in the quality of gas,
both daily and from month to month. There are separate diagrams for the
heating value and the illuminating value.

4. The zero line represents the average for a complete year, except when
tests did not cover so long a period. In each case the actual figure represented
by the zero line is given, and the months included in the average are stated.
The average for each month is shown by a heavy line indicating the percentage
of variation above or below the yearly average. The cross-sectioning repre-
sents the extreme high and low variation of any daily readings during such
month in percentage of the monthly average.

5. The illuminating values and their variations from the yearly and
monthly average are shown according to the same method in the second
diagram.

6. A careful study of these diagrams indicates that the percentages of
variation in heating values from day to day and from month to month are
considerably less than the percentages of variation in illuminating values, that
the variations in monthly averages for the two measures of quality do not
parallel one another and that there is no definite relation between them.

COMPANY NO. 1

Works Kind of coal % screened Pennsylvania gas coal.

Class A Table I. Page 16.
Duration of charge 4 hours.

Operation Coal gas plant with water gas auxiliary, not in use daily.

One holder housed, six exposed.
Yield per Ib. coal-^.79 to 4.94 cu. ft. (cor.)
Tests Tests made at works.

Coal gas enriched with oil gas.

Type of calorimeter Junkers American Meter Co.
Temperature of atmosphere not reported each month, prob-
able range during period of tests from to 100 F.

Curves Zero lines represent average heating power or illuminating

power for period October 1, 1911, to September 30, 1912,
excepting February, 1912, for which month no tests were
reported.

Average heating power=652 B. t. u.
Average illuminating power=17.2 C. P.

VARIATIONS IN HEATING POWER

OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

OCT. NOV. DEO. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

COMPANY NO. 2

Works Coal gas plant.

Class D Table I. Page 16.

Holders exposed.

Horizontal retorts.

One-half depth furnace.

Operation Enricher cannel coal 8.57 to 9.23 Ibs. 'per 100 Ibs. coal

carbonized.

Kind of coal Pennsylvania.

Duration of charge from 5 hrs. 35 min. to 7 hrs. 26 min.

Yield per Ib. coal December to July 4.78 to 5.25 cu. ft. (cor.)
Tests Tests made at office.

Type of calorimeter Junkers.

No photometric tests reported.

Temperature of atmosphere ranged from 10 to 102 F.
Curves Zero line represents average heating power for 12 months

August 1, 1911, to July 31, 1912.
Average heating power 645 B. t. u.

VARIATIONS IN HEATING POWER

AUQ. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

COMPANY NO. 3

Works Carburetted water gas plant.

Class A Table I. Page 16.'
Holders exposed.
Generators 7' 6" and 12' sets.

Operation Enricher 34 to 35 B. gas oil 3.68 to 4.48 gals, per M. (cor.)

Kind of fuel Anthracite grate coal.
Generator fuel per M. (cor.) 31.55 to 37.44 Ibs.
Hours per day works operation from 10 to 24.
Tests Tests made at works.

Type of calorimeter Junkers.

Temperature of atmosphere ranged from 16 to 98 F.

Curves Zero lines represent average heating power or illuminating

power for period August 1, 1911, to July 31, 1912, except-
ing April, 1912, for which month no tests were reported.
Average heating power=640 B. t. u.
Average illuminating power=22.6 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

COMPANY NO. 4

Works Carburetted water gas plant.

Class A Table I. Page 16.
Holders exposed.
Generators U. G. I. Improved Lowe up and down steam,

7' 6" and 8' 6" sets. Air and steam meters.

Operation Enricher Gas oil 3.80 to 4.19 gals, per M. (cor.)

Kind of fuel Broken anthracite.
Generator fuel per M. (cor.) 29.7 to 35.7 Ibs.
Hours per day works operation from 4.5 to 23.8.
Tests Tests are made at works laboratory.

Samples of gas taken from outlet of street main governor.
Gas has been exposed to atmospheric temperature in storage

holder and relief holder.

Type of calorimeter Junkers, 1910 American Meter Co.
Type of photometer U. G. I. 60" Bar. Edgerton Standard.

No. 7 Bray burner.

Temperature of atmosphere ranged from 4 to 97 P.
Curves Zero lines represent average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.

Average heating power=634 B. t. u.

Average illuminating power 22.4 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10*

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20% T

Works
Tests

Curves

COMPANY NO. 5

Carburetted water gas plant.

Class C Table I. Page 16.

Relief holder housed.

Tests made at a test station one-half mile from works.

Type of calorimeter Sargent.

Temperature of atmosphere ranged from 27 to 97 F. during
period of tests.

Note. This calorimeter was moved during December and Jan-
uary to plant of Company No. 16.

Zero lines represent average neating power or illuminating
power for four months August 1, 1911, to November 30,
1911.

Average heating power=627 B. t. u.
Average illuminating power 20.5 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV.

20%

10%

20%

COMPANY NO. 6

Works Carburetted water gas plant.

Class B Table I. Page 16.

Holders 1 housed, 1 exposed, Works A ; 1 exposed, Station B.
Generators U. G. I.' Standard up and down steam, 6' sets.
Operation Enricher Gas oil 3.59 to 4.11 gals, per M. (cor.)

Kind of fuel Anthracite grate.
Generator fuel per M. (cor.) 27.31 to 30.48 Ibs.
Hours per day works operation from 9.18 to 22.45.
Tests are made at works (A).
Tests Tests are also made at outlying station (B). See next page.

Samples of gas, works A, taken at outlet of station governor.
Type of calorimeter Junkers.

Type of photometer U. G. I. Standard 60" Bar. Edgerton
Standard checked by Pentane lamp. No. 7 lava tip burner.
Temperature of atmosphere ranged from 20 to 108 F.
Curves Zero lines represent the average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.
Average heating power 628 B. t. u.
Average illuminating power=23.3 C. P.

TESTS AT WORKS
VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

Tests

Curves

COMPANY NO. 6. (Continued.)

Works and operating data given on preceding page.

Tests made at outlying testing station (B).

Samples of gas taken from inlet side of the governor on the

outlet of exposed storage holder.
For course of gas from works see map on page 48.
Type of calorimeter Junkers American Meter Co.
Type of photometer U. G. I. Standard 60" Bar. Pentane

lamp standard. No. 7 lava tip burner.
Zero lines represent average heating power or illuminating

power for 10 months January 1 to October 31, 1912.

Average heating power=:624 B. t. u.

Average illuminating power=20.5 C. P.
For further information see Appendix C, page 49.

10%

TESTS AT OUTLYING STATION
VARIATIONS IN HEATING POWER

JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT OCT.

20%

10%

20%

COMPANY NO. 7.

Works Carburetted water gas plant.

Class B Table I. Page 16.
Holders exposed.

Generators Western Gas Const. Co., 7' 6" sets.

Operation Enricher 34 B. gas oil 3.06 to 3.84 gals, per M. (cor.)

Kind of fuel Anthracite coal.
Generator fuel per M. (cor.) 31.7 to 35.4 Ibs.
Hours per day works operation from 7.5 to 20.7.
Tests Tests are made at works.

Samples of gas taken from outlet of station governor.

Gas has been exposed to atmospheric temperature in city

holder.

Type of calorimeter Junkers.
Type of photometer Suggs-Letherby open type. Standard

Hefner lamp burning imported Amylacetate. Burner

Argand F.

Temperature of atmosphere ranged from 20 to 95 F.
Curvei Zero lines represent average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.

Average heating power 621 B. t. u.

Average illuminating power=22.7 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. <JEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

Works

Operation

Tests

Curves

Carburetted water gas plant.

Class C Table I. Page 16.

Holders exposed.

Generators 5' and 7' 6" sets. Pyrometers on sets.

Enrich er 28 B. oil 3.5 to 4.65 gals, per M. (cor.)

Kind of fuel Broken antLracite and from 80 to 89 % anthra-
cite remainder gas house coke during March, April, May
and June.

Generator fuel per M. (cor.) 35.90 to 40.8 Ibs.

Hours per day works operation from 8 to 23.8.

Tests are made at works (A) and also made at testing station
(B). See next page.

Samples of gas at works (A) taken from distribution main gov-
ernor inlet.

Type of calorimeter Junkers and Sargent. (Although re-
sults of tests with Sargent calorimeter were reported, they
have not been used in connection with accompanying
curves to avoid duplicating curves unnecessarily.)

Type of photometer 60" Open Bar. U. G. I. Standard Gen-
uine English Spermaceti candles weighing 6 to the pound.
Burner No. 7 L. P. Slit Union Bray and "New F" Ar-
gand Sugg pattern.

Temperature of atmosphere ranged from 3 to 100 F.

Zero lines represent average heating power or illuminating
power for 12 months August 1, 1911, to July 31, 1912.
Average heating power*=616 B. t. u.
Average illuminating power=:20 C. P.

N. B. Average heating power Sargent same period, 616 B. t. u.

* See also Appendix C, page 56.

TESTS AT WORKS
VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

107o

10%

20%

COMPANY NO. 8. (Continued.)

Tests

Curves

Works and operating data given on preceding page.

Tests made at testing station (B).

Samples of gas are taken direct from service entering build-
ing from street main.

Type of calorimeter Junkers.

Type of photometer 60" open Bar. American Meter Co.
Standard Genuine English Spermaceti candles weighing
6 to the pound. Burner No. 7 L. P. Slit Union Bray and
"New F" Argand Sugg pattern.

Zero lines represent average heating power of illuminating
power for 12 months August 1, 1911, to July 31, 1912.
Average heating power='609 B. t. u.
Average illuminating power=20.2 C. P.

TESTS AT TESTING STATION
VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

COMPANY NO. 9.

Works

Operation

Tests

Curves

Carburetted water gas plant.

Class A Table I. Page 16.

Holders exposed.

Generators 11'xlG' 7" and 12'xl6' 7" Lowe reverse steam sets.

Indicating and Recording Pyrometers.
Enricher 28 B. gas oil 3.41 to 4.34 gals, per M. (cor.)
. Kind of fuel Anthracite.
Generator fuel per M. (cor.) 30.45 to 32.67 Ibs.
Hours per day works operation continuous running.
Tests are made at testing station exceeding 1 mile from works.
Samples of gas taken directly from service entering building

from the street main.
Gas has been exposed to atmospheric temperature in holders

and ground temperatures in mains.
Type of calorimeter Junkers.
Type of photometer 60" open Bar, American Meter Co.

Standard Genuine English Spermaceti candles weighing

6 to the pound. Burners No. 7 L. P. Slit Union Bray and

"New F" Argand Sugg pattern.

Temperature of atmosphere ranged from 2 to 98 F.
Zero lines represent the average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.

Average heating power 609 B. t. u.

Average illuminating power 20.7 C. P.

VARIATIONS IN -HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

COMPANY NO. 10.

Works Carburetted water gas plant.

Class B Table I. Page 16.
Storage holder exposed. Relief holder housed.
Generators 5' Western Gas Const. Co. sets, one with no down-
run valve, one with down-run valve.

Operation Enricher 28 to 34 B. gas oil 3.91 to 4.47 gals, per M. (cor.)

Kind of fuel Anthracite grate.
Generator fuel per M. (cor.) 38.3 to 46.9 Ibs.
Hours per day works operation 3 1/7 to 15%.
Tests Tests are made about % mile from works.

Samples of gas taken direct from regular service into the

building.
Gas has been exposed to atmospheric temperatures in holder

and to ground temperature in mains.
Type of calorimeter Sargent.
Type of photometer 80" closed Bar. American Meter Co.

Standard Candles. Burner Either ' ' New F ' ' Argand or

No. 7 Slit Union Bray.
Temperature of atmosphere not reported.
Curves Zero lines represent the average heating power or illuminating

power for 12 months October 1, 1911, to September 30,

1912.

Average heating power=626 B. t. u.

Average illuminating power=21 C. P.

VARIATIONS IN HEATING POWER

OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

Tests

Works Mixed coal and carburetted water gas plant.

Class A Table I. Page 16.
Holders exposed.

Generators 8' 6", 10' 0", 10' 0" Twin Gen., U. G. 1. sets.
Benches 10 Parker & Russell 9s.

Operation Water gas Enricher 35 to 37 B. gas oil 3.77 to 4.48 gals,

per M. (cor.)

Kind of fuel Anthracite coal, retort coke and oven
coke.

Generator fuel per M. (cor.) 32.27 to 36.19 Ibs.
Hours per day works operation continuous running.
Coal gas Kind of coal Pennsylvania gas coal.

Yield per Ib. coal cor. gas 4.71 to 4.96 cu. ft.
Duration of charge 4 hours.
Mixed gas Mixture from 16.4% coal gas and 83.6% water

gas to 24.7% coal gas and 75.3% water gas.
Tests are made at works.
Tests were made of the earburetted water gas, of the coal gas

and also of the mixed gas.

Type of calorimeter Junkers American Meter Co.
Type of photometer not reported. Burner No. 7 Bray Slit

Union.

Temperature of atmosphere ranged from --4 to 94 F.
The curves given below are for the mixed-coal and carburetted
water gas. (Curves for straight coal gas are on following
page.)

Zero lines represent average heating power or illuminating
power for 12 months August 1. 1911, to July 31, 1912.
Average heating power=638 B. t. u.
Average illuminating power 20.6 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

Curves

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

20%

COMPANY NO. 11. (Continued.)

Unenriched coal gas.

Data regarding works, operation and tests are given on pre-
ceding page.

Curves Zero lines represent average heating power or illuminating

power for 12 months August 1, 1911, to July 31. 1912.
Average heating power=611 B. t. u.
Average illuminating power=14.3 C. P.

UNENRICHED COAL GAS
VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUQ. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUQ. SEPT. OCT.

20%

10%

20%

COMPANY NO. 12

Works Mixed coal and carburetted water gas plant.

Class A Table I. Page 16.
One holder exposed, others housed.
Generators 7' 6" H. & G. and 8' 6" U. G. I.
Benches % depth horizontal.

Operation Water gas Enricher 35 B. gas oil 3.52 to 4.33 gals, per

M. (cor.)

Kind of fuel Retort house coke.
Generator fuel per M. (cor.) 27.96 to 33.45 Ibs.
Hours per day works operation continuous running.
Coal gas Kind of coal Pennsylvania gas coal.

Yield per Ib. coal (cor.) gas 4.55 to 5.22 cu. ft.
Duration of charge 4 hours.
Mixed gas Mixture from 19.21% coal gas and 80.79% water

gas to 26.8% coal gas and 73.2% water gas.
Tests Tests are made at works.

Samples of gas taken at inlet to works governor.
Type of calorimeter Junkers.

Type of photometer 60" Standard U. G. I. Bar. Standard
Edgerton standard checked daily by Pentane lamp
Burner 7' lava tip burner.

Temperature of atmosphere ranged from 8 to 116 F.
Curves Zero lines represent average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.
Average heating power=622 B. t. u.
Average illuminating power 21.2 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

10% -

20%

COMPANY NO. 13

Works Mixed coal and carburetted water gas plant.

Class C Table I. Page 16.
Holders housed.
Generator 6' Lowe set.

Operation Water gas Enricher gas oil 3.4 to 4.3 gals, per M. (cor.)

Kind of fuel Coke.

Generator fuel per M. (cor.) 41.09 to 48 Ibs.
Hours per day works operation from 2 hrs. 43 min.
to 12 hrs. 45 min.
Coal gas Kind of coal Pennsylvania.

Yield per Ib. coal (cor.) 4.75 to 5.5 cu. ft.
Duration of charge from 4 hrs. to 6 hrs. 20 min.
Mixed gas Mixture from 53 % coal gas and 47% water gas

to 68% coal gas and 32% water gas.
Tests Tests are made at office.

Type of calorimeter Junkers.

Temperature of atmosphere not reported regularly from

- 10 to probable 100 F.

Curves Zero lines represent average heating power or illuminating

power for 12 months August 1, 1911, to July 31, 1912.
Average heating power=630 B. t. u.
Average illuminating power=18.5 C. P.

VARIATIONS IN HEATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

AUG. SEPT. OCT. NOV. DEC. JAN. FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

20%

COMPANY NO. 14

Works

Operation

Tests

Curves

Mixed coal and carburetted water gas plant.

Class C Table I. Page 16.

Holders at works, exposed. Outlying, housed.

Generators 5' double superheater, Lowe set, U. G. I. pattern

with down-run connections.

Benches % depth benches of 6s by Improved Equipment Co.
Water gas Enricher gas oil 3.93 to 4.63 gals, per M. (cor.)
Kind of fuel Gas coke.

Generator fuel per M. (cor.) 34.37 to 39.28 Ibs.
Hours per day works operation from 5 hrs. 38 min. to
17 hrs. 15 min.
Coal gas Kind of coal Pennsylvania.

Yield per Ib. coal (cor.) 4.29 to 4.96 cu. ft.
Duration of charge 4 hours.
Mixed gas Mixture from 31.18% coal gas and 68.82% water

gas to 54% coal gas and 46% water gas.
Tests made at office 3^ miles from works.
Samples of gas taken from regular distribution main through

service entering building.

Type of calorimeter Junkers American Meter Co.
Type of photometer 60" open Bar. U. G. I. Standard Double
candles. Burner Sugg F or occasionally Sugg D or Bray
special.

Temperature of atmosphere ranged from 22 to 84 F.
Zero lines represent average heating power or illuminating
power for 7 months April 1 to October 31, 1912.
Average heating power=616 B. t. u.
Average illuminating power 18.7 C. P.

VARIATIONS IN HEATING POWER

APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

APR. MAY JUNE JULY AUG. SEPT. OCT.

20%

10%

2O%

COMPANY NO. 15

Works Carburetted water gas plant.'

Class A Table I. Page 16.

Holders exposed.

Generators 9' and 12' Williamson.
Operation Enricher 34.1 B. gas oil.

Kind of fuel Anthracite coal and coke.

Hours per day works operation 12 to 18 hours.

For information regarding transmission, see map, page 51,
and explanation of tests, page 49.

Tests not having been started until September 1, 1912, no
diagram has been made of the results.

COMPANY NO. 16

Works Carburetted water gas plant.

Class B Table I. Page 16.
Holders exposed.

Operation Enricher Gas oil 3.60 to 4.21 gals, per M. (cor.)

Kind of fuel Anthracite.
Generator fuel per M. (cor.) 33.82 to 41 Ibs.
Hours per day works operation from 6.24 to 14.4.

Tests Tests are made at testing station, iy 2 miles from works.

Type of calorimeter Sargent.

Temperature of atmosphere ranged from 2 to 102 F.

Curves Zero lines represent average heating power or illuminating

power for 9 months February 1 to October 31, 1912.
Average heating power 625 B. t. u.
Average illuminating power 20.9 C. P.

VARIATIONS IN HEATING POWER

FEB. MAR. APR. MAY JUNE JULY AUG. SEPT. OCT.

10%

VARIATIONS IN ILLUMINATING POWER

FEB. MAR. APR. MAY JUNE JULY AUG. SEPT OCT.

20%

10%

20%

7. The accompanying diagram has been prepared for the purpose of de-
monstrating the fallacy of the impression which still exists in some quarters
that there is a definite relation between the illuminating power and the heating
power of the gas. The results of all tests, made at the works, of coal gas, car-
buretted water gas, and mixed coal and carburetted water gas, have been con-
sidered in the preparation of this diagram with the exception of tests of unen-
riched coal gas. Tests showing a heating power below 575 or above 650 have
also been excluded, as they were exceptional and as they were doubtless due
to some extraordinary conditions.

8. With the exceptions noted, all of the results obtained during the fifteen
months covered by the tests when plotted fall within the shaded portion of the
diagram. In other words, gas having a heating power of 575 B. t. u. has been
shown to have an illuminating power anywhere between the limits of 16.6 and
21.4 candle power. Similarly, gas of 620 B. t. u. is shown to have an illuminat-
ing power anywhere between the limits of 16.8 and 25.6 candle power.

9. It might have been expected that the limits would be fairly wide apart
in all cases, but that the tendency of the shaded portion of the diagram would
follow a direction corresponding to higher candle power for higher heating
values. Possibly if all of the tests were plotted and the values weighed a
tendency of this character would be noted. It is, however, a fact that the
minimum candle power, reported at any time 14.9 candle power, was found
when the gas had a heating power of 590 B. t. u. and that the next lowest
candle power, 15.1 candle power, was found when the gas had a heating power
of 638 B. t. u. Similarly, the highest candle power reported, 26.8 candle
power, occurred when the gas had a heating power of 634 B. t. u. ; the next
highest candle power, 26.6 candle power, occurred when the gas had a heat-
ing power of 629 and 632 ; the third highest candle power, 26.5 candle power,
occurred on five occasions, the gas having heating power of 603, 627, 633, 634
and 639.

(See Diagram on opposite page.)

Range in candle power, at works, of gas having heating values
of from 575 to 650 B t. u.

APPENDIX C

Manufacture and Distribution with Reference to Illuminating Value

and Heating Value

UNENEICHED COAL GAS

1. Only one Company (No. 11) participating in the investigation reported
the illuminating value and heating value of straight unenriched coal gas. From
the reports submitted it has been found that the average illuminating value
of this coal gas, for a full year, as read at the manufacturing plant, is 14.3
candle power, with a maximum individual daily reading of 18.1 and a mini-
mum of 11.4 on a No. 7 Bray special burner. This gas is reported to be made
from the best gas coal obtainable, and generated under good average condi-
tions in horizontal retorts, with yields of only 4.71 to 4.96 cu. ft.

2. The present State requirements demand at the testing station a 16
candle power coal gas. It will be seen that under these requirements, the
manufacture and sale of straight coal gas, without enrichment, is not per-
missable. To meet the requirements a coal gas must be enriched the greater
part of the year with either gas made from eannel coal or gas made from oil.

3. Should a coal gas be enriched by mixing with a carburetted water gas,
the gas would then be classified as a mixed gas and the present requirements
would demand 18 candle power. To obtain this result would mean a large
percentage of a high candle power water gas from 22 to 25 candle power to
bring the mixture up to the required quality, an increasingly difficult process
with the deterioration in the quality of oil obtainable.

4. The enrichment of coal gas with a gas made from eannel coal is not
generally practiced. The coke made from this coal is of little value as a
by-product, and if mixed with the coke obtained from gas coal, reduces the
value of the entire product, thus increasing the cost of manufacture without
compensating results in the quality of gas thus obtained. The supply of a
good grade of eannel coal is limited, and if generally used, it is doubtful
whether an adequate supply could be obtained. This method of enrichment
is used only under peculiar and unusual conditions.

5. Enrichment with heavy oil by generating oil gas in coal gas retorts
has been considered generally inefficient and expensive; and light oils are no
longer available. It is a practice that, where possible, has been abandoned by
nearly all coal gas companies; but some method of enrichment is compulsory
under the present requirements of this State.

6. Benzol enrichment, so-called, consists of adding light oil vapors in
the shape of benzol and toluol to the gas in the form of a spray, to increase its
illuminating value. This increase is only effective under certain favorable
conditions and is lost, to some extent, when the gas is further subjected to low
temperatures or high pressures. (See pages 45 and 46, paragraphs 25, 26
and 27.)

7. In enriching coal gas by making a mixed gas, the present standards
demand an illuminating value two candles higher than required for coal gas.
-This method of manufacture was employed by Companies 11, 12, 13 and 14,
but only one of these Companies (11) reported the illuminating value and heat-
ing value of the two gases separately, and it will be seen from the following

table, which is a summary of its results, that a corresponding increase in the
heating value was not obtained in the effort to bring the illuminating value of
the mixed gas up to the required standard. It is seen that while the illuminat-
ing value of the coal gas is increased about 45 per cent., the heating value is
only increased 5 per cent. It should also be noted that the coal gas is mixed
with three to four times its own volume of carburetted water gas of high
illuminating value.

Coal Gas. Carburetted Mixed Gas.

Water Gas.

Illuminating value 14.3 C. P. 22.5 C. P. 20.6 C. P.

Calorific value 609 B. t. u. 647 B. t. u. 639 B. t. u.

8. The results of the entire fifteen months' readings by this Company,
as reported to the Committee, on the illuminating value and the heating value
of the straight coal gas, are shown below.

9. This gas was made from Pennsylvania gas coal in stop end retorts and
is straight coal gas without enrichment; the illuminating value being deter-
mined on a No. 7 Bray special burner. The minimum day's reading for each
month and the average for the month, are shown, as well as the' average for
the entire fifteen months. (See also Appendix B, pages 34 and 35.)

Company No. 11 Straight Unenriched Coal Gas:

Month 1911. Candle Power. B. t. u.

Min. Avg. Min. Avg.

August 13.3 14.9 546 602

September . 13.3 15.3 587 616

October 11.7 13.5 584 611

November 11.5 13.8 586 607

December 12.7 14.2 573 621

1912.

January 12.4 13.7 580 628

February 12.4 13.7 579 614

March 12.2 14.9 550 612

April 11.4 13.5 578 612

May 12.6 14.7 570 601

June 13.9 15.3 561 610

July 12.5 14.3 565 593

August 12.3 13.7 569 591

September 12.2 14.2 557 595

October . .. 11.7 13.1 562 592

Minimum 13.1 591

10. The following figures (Par. 11) show the result from a coal gas plant
not participating in the investigation using coal with a volatile
constituent of 36 per cent., by weight, and producing a straight unenriched
coal gas.

11. The generators and settings are designed to give a maximum yield
of gas per pound of coal. The retorts are set horizontally and machine
stoked. All results are corrected as to temperature and pressure. The burner
is the new Sugg F, Argand. The heat unit averages are the result of ten calor-
imetric tests per month. A year's results are as follows:

Average Heating Average Illuminating Yield Per Lb.

Month. Value of Gas. . Value of Gas. Coal.

B. t. u. Candle Power. Cu. Ft.

January 556.1 14.54 5.36

February 567.3 14.79 5.17

March 573.7 14.79 5.26

April 568.3 14.32 5.21

May 594.2 14.82 5.25

June 600.5 14.74 5.24

July 600.9 14.16 5.19

August 602.3 14.32 5.28

September 596.6 14.34 5.21

October 599.6 14.30 5.19

November 589.1 14.14 5.20

December . 583.0 14.64 5.17

Minimum . 556.1 14.14 5.17

12. For the purpose of confirming the above results, we submit the aver-
age candle power of straight unenriched coal gas manufactured by two large
gas plants, for each month covering a period of two years.

13. These values, which are the averages of hourly readings, are indi-
cative of the illuminating value that may be obtained in an efficiently man-
aged plant, using the best West Virginia gas coal, in horizontal retorts. Candle
power readings were made each hour of the twenty-four, on a No. 7 flat flame
burner against a pentane lamp. Candle power variations are due largely to
changes from freshly mined coal to stored coal. Yield slightly in excess of
5 feet per pound.

Plant 1. Plant 2.

Candle Power Candle Power

1910. 1911. 1910. 1911.

January 12.61 11.44 12.65 11.97

February 12.40 12.20 13.34 13.28

March 12.00 13.01 13.50 12.63

April 12.69 12.46 12.91 12.11

May 11.65 13.00 12.03 13.08

June 11.96 12.68 11.84 13.22

July 11.44 13.11 11.40 13.11

August 11.25 12.20 12.07 11.82

September 11.39 13.30 12.27 13.10

October 11.11 13.62 13.38 12.83

November 12.60 13.34 11.96 12.37

December 12.59 12.80 12.54 13.99

Average 11.97 12.76 12.49 32.79

RECENT DEVELOPMENTS IN COAL GAS MANUFACTURE

14. Recent developments in connection with the manufacture of coal gas
have appeared in the form of vertical gas retorts and chamber ovens. Among
the features of these types of installation have been improvements in the meth-
ods of handling materials and by-products. The character of construction of
these retorts has permitted the adoption of devices for charging coal and dis-
charging coke that have eliminated much of the heavy work of stoking labor,
which is particularly arduous during the hot summer months.

15. The installation of these types of retorts has been of benefit in the
saving of ground space occupied and in some instances in overcoming the
ventilation retort house problems during the time of charging and discharging.

16. Gasifying coal in bulk in relatively large units has allowed a longer
time for carbonization of the charge, and has been somewhat effectual in in-
creasing the total yield of gas from an equal quantity of coal, with, however,
a slight reduction in the heating value and illuminating value per thousand.

17. Lengthening the carbonizing time has been helpful in increasing the
total heat in resultant gas, per ton of coal carbonized, thereby producing a
general economic saving.

18. Developments along this line are showing rapid improvements at the
present time and it seems advisable that the quality requirements of a gas
should be so placed as to allow the adoption and use of these more recent meth-
ods in the manufacture of coal gas throughout this State.

COKE OVEN GAS

19. During the past decade a large number of coke oven plants have been
erected throughout the United States. Many of these plants were installed in
connection with industrial undertakings, requiring coke for their operation,
and where they have been compelled to seek a market for the by-product gas.

20. In many instances the coals used by these coke ovens were selected
for their ability to produce a good metallurgical coke and often these ovens
were not operated to handle by-product gas most efficiently. But in later years
the methods of operating coke ovens have been somewhat modified, with the
result that the quality of the gas has been much improved. "With a selection of
coals better suited for gas-making purposes, a better quality of gas may be
expected from the coke ovens. But where these ovens are operated for the
purpose of producing a high-grade furnace coke, and not primarily for the
production of gas, the so-called best gas coals cannot be used to the best
advantage, and, therefore, we probably will not see the highest quality of
coal gas made in coke oven plants.

21. A number of instances have already arisen where such by-product
coke oven gas could have been utilized to advantage. The heating value ap-
proached that of unenriched retort coal gas, and could have been made to
equal it by enrichment, or mixture with carburetted water gas. The illuminat-
ing value, however, could not be brought up to the present requirements,
except by enriching costs that were commercially prohibitive, or by using
too large a percentage of carburetted water gas, and therefore the supply
could not be availed of. To permit the utilization of this gas by proper
standards would prevent this useless waste, and would constitute a great eco-
nomic saving.

22. It seems probable that such situations will occur more frequently in
the future than in the past and it would appear to be good public policy to
permit the use of such gas for general distribution.

23. Coke oven gas has been used by many gas companies in other States,
in whole or in part, for distribution to general consumers.

24. The treatment of this gas, in a number of instances, by manufacturing
companies, however, is quite different from that of ordinary coal gas produced
by gas companies, in that the operators of the coke ovens scrub the gas and
remove the light oil vapors, consisting of benzol, toluol and xylol, reducing to
a considerable extent its illuminating value, at the same time reducing in a
slight degree only its heating value.

25. In a number of cases where coke oven gas has been purchased by the
gas companies, its illuminating value has been restored by the addition of re-
fined benzol to a value even greater than that of the gas before it was first
scrubbed.

26. This method of increasing the low illuminating value of this gas has
been the result of a requirement for a higher illuminating value, which has
been obtained without a corresponding increase in heating value. And the
addition of this increase in illuminating value may not be considered at all

times to be permanent. Under certain conditions where the unfixed hydro
carbon vapors have been removed in a previous washing and scrubbing of the
gas, the imparted enrichment from benzol is more effective, as the gas is in a
condition to absorb these vaporous hydro-carbons having such a relatively
low percentage of saturation. Under these conditions the increased illuminat-
ing value will stand with rather severe changes in temperature and pressure,
and the gas may be considered suitable for general consumption where the
illuminating requirements are low, but where a normal heating value is
demanded.

27. But, as a general rule, the enrichment or increase in illuminating
value of any gas, by the addition of benzol, may not be considered permanent
and its addition has a tendency to create a gas variable in quality when deliv-
ered to the consumer and objectionable because of carbon deposit in burners
and mantles.

28. Carburetted water gas has been used as an enricher for coke oven gas,
and its use may be successfully employed where the requirements do not
demand an illuminating value above that of unenriched retort coal gas, but
otherwise the quantity of carburetted water gas necessary for this enrichment
results in too great a proportion of the total amount of gas supplied.

29. Carburetted water gas may be used to advantage to provide any
deficiency in supply and to take care of the peak loads where the coke oven
gas is produced at a uniform rate from day to day, and provide reserve in
times of a depressed market for furnace coke.

CARBURETTED WATER GAS

30. An analysis of the records of the Companies manufacturing car-
buretted water gas, and reporting their results, shows a great variation in both
illuminating value and heating value, and proves conclusively that there is no
positive relation between illuminating value and heating value.

31. It has been proven, however, that in general the carburetted water
gas of the highest average illuminating value has had the highest heating
value, as the illumination imparted to the open flame is a function of the
quantity as well as the quality of oil used, considering the manufacturing
apparatus is operated with equal efficiency. It has also been demonstrated
that the heating value is a factor of the quantity of oil used, but there are
limits to the quantity of oil that can be efficiently handled and turned into a
constituent gas.

32. Therefore gases of equal illuminating value do not necessarily have
equal heating value and vice versa. This is true of the gas when manufac-
tured, without introducing any of the uncertain elements of distribution. Such
a condition is due more particularly to the variation in oil efficiencies obtained
in producing illuminating value, and which are occasioned by different types
of generating apparatus and different methods and constituents of operation,
as well as the quality of the oil available. These economies or efficiencies are
greatly influenced by climatic and temperature conditions, which cause extreme
variations between the warm summer months and the cold winter months.
And the economic results obtained by scientifically operated plants cannot be
expected of all plants throughout an entire State.

33. In the plants of the reporting Companies, the conditions of gas manu-
facture vary greatly and some variation in the results obtained was, therefore,
to be expected. At the same time, the location of the testing station, whether
at the works or at some remote point in the distribution system, introduced
factors that had to be considered in making an analysis of the results obtained.

We believe, however, that the reports indicate what may be considered as good
practice, representing fair and average conditions of operation, and that the
conclusions drawn are based on representative data.

34. As regulatory requirements have frequently been based on monthly
averages, an analysis has been made of the results obtained by months, and
from these results have been calculated the heating value imparted to the car-
buretted water gas per gallon of oil employed in manufacture.

35. From the tests made by six of the Companies the heating value of
the gas, as read at the works, for a whole year, averaged 162 heat units per
cubic foot, per gallon of oil used per thousand cubic feet; while the average
obtained in the summer months was 168 and in the winter months 157. The
minimum monthly average for all plants was 151 heat units. These same tests
show that in one plant the average for an entire month was only 141 heat
units per gallon of oil used. The indications are that a yearly average of 150
heat units per gallon, per thousand cubic feet, can be obtained with good
operating conditions.

36. It is evident from the reports made by four of the Companies, testing
at a point some distance from the works, that the heating value delivered
would be somewhat below this figure. These plants show a yearly average of
154 heat units per gallon of oil used, with an average of 167 during the sum-
mer months, and an average of 146 heat units during the winter months. The
minimum for all plants for any month occurred during January, when 142
heat units per gallon of oil was delivered, with an individual plant average in
any month of 139 heat units per gallon.

37. The above figures indicate that, on the basis of the average quantity
of oil used in these ten plants during twelve months, at least 150 heat units
per gallon of oil during the winter months may be expected when measure-
ments are taken .at the works, or 140 heat units per gallon of oil when measured
at a point some distance from the works.

38. The quantity of oil used as an average by all the Companies was
about 3.9 gallons, and if this is considered as a normal quantity to be used,
then the average heating value for any one month, of a gas, as read at the
works, would be 3.9x150 or 585 heat units.

39. The average quantity of oil used in the above deduction represents
the conditions when manufacturing gas to meet the present illuminating stand-
ard. To meet such a standard requires the manufacture of a gas having an
illuminating value of from 6 to 8 per cent, higher than that specified in the
standard, on account of the impossibility of manufacturing gas of an exact
illuminating value and the necessity for an excess in quality to allow for vari-
able losses due to handling, to transmission and to changes in temperature.

40. In addition to determining the heating value and the illuminating
value of gas as generated at and delivered from various gas plants, it was
deemed essential by the Committee to discover as far as the present operat-
ing conditions would permit the effect on the quality of , the gas of trans-
mission through distribution systems. It was known that losses in the quality
of a gas occur during transmission, caused by the scrubbing action on the gas
in passing through the mains, which is aggravated by a reduction in tempera-
ture and by any increase in pressure. Readings, therefore, were taken, where
opportunity permitted, to determine the extent of losses due to these factors.

CA6 WORKS

AT THIS POINT
/a ABOUT, / rr

UflSOEft OO Of Jr/fA/H

c/ry

MAP

COMPANY WO.

MAM MOT XfaseO

/VOr XPOSED

/JESTS ON ran

Of BH/OG ARCH

"Ot-OEfl

41. At the beginning of the investigation there was little authoritative
data available relating to this question. At the suggestion of the Committee,
however, some experiments were undertaken to determine the effect of dis-
tribution under varying conditions.

42. During a period of eleven months, beginning January, 1912, read-
ings were taken by one of the participating Companies (No. 6) of the illumi-
nating value and heating value of carburetted water gas as manufactured, and
again as delivered in an outlying district three miles distant, at the end of a
pumping main; the initial pressures being from eight to fifteen inches water
pressure. These readings were made daily with some few exceptions and
include some 294 individual tests for both heating and illuminating value. (See
tables II. and III., Appendix A, page 18.)

The location of the works and the testing stations are shown on the map,
on opposite page, and it will be seen that at some few points the main is ex-
posed to temperatures of the streams that are crossed, and at other points to
atmospheric temperature. The gas was further subjected to atmospheric tem-
perature in an outlying storage holder. Eesults of these tests indicate the losses
in heating value and illuminating value that may be expected at various seasons
of the year for gas delivered under similar conditions. The average loss in
heating value is 1.4 per cent, and the average loss in illuminating value is 12.8
per cent.

Loss in Loss in

Illuminating Value Heating Value

Month. No. Tests Candles. Per Cent. B. t. u. Per Cent.

January 20 3.2 14.0 11 1.75

February 16 3.2 13.8 13 2.06

March 26 2.7 11.7 4 .64

April 30 3.4 14.4 4 .64

May 31 2.7 11.6 17 2.66

June 30 2.5 10.8 7 1.09

July 31 2.3 9.8 8 1.24

August 29 2.6 11.2 5 .79

September 27 3.0 12.9 14 2.22

October 26 3.0 12.8 12 1.88

November . 28 4.6 18.3 5 .77

Average 12.8 1.4

43. The daily variations in heating value and illuminating value between
Stations A and B are shown by diagrams (page 50). The zero line represents
the heating value or illuminating value at A. The solid line curve shows the
heating value at B in percentage of the heating values at A. The broken line
curve similarly shows the illuminating value at B.

44. Another Company (No. 15) made observations of the loss in heating
value and illuminating value of gas when pumped under the relative high
pressure of fifteen pounds, for a distance of about twenty-three miles. These
results cover daily observations for the months of October and November,
1912, consisting of about sixty individual tests. They indicate what may be
expected under similar conditions. The readings were made when the ground
temperature was about 50 F., and, therefore, do not show the extreme condi-
tions that will be met during the winter months when the temperature of the
ground is 32 F. Plate III shows how the main is exposed both to atmospheric
and stream temperatures. (See page 51.)

45. The readings made on this high pressure line are summarized below,
and it will be seen that the loss due to transmission of gas under these condi-

SCALE

PIPE fXfOSfO on

PIPE HUNS UMQEH

P/PE ftt/AfS t/MDEH WAT Ef{

f>if> fXPQseo on arno&e

II I

] STAT/OA/

tions is much more in illuminating value (20.7%) than it is in heating
value (6.05%).

Heating Illuminating

October 1912. Value, Value, 0. P.

B. t. u.

Uncompressed gas 629 22.2

Compressed gas 619

Delivered gas 23y 2 miles 607 17.6

Loss per cent 3.5 20.7

November.

Uncompressed gas 31 21.3

Compressed gas 627

Delivered gas 23% mi ls 577 16.9

Loss per cent 8.6 20.7

Average loss per cent 6.05 20.7

46. This is in accordance with the expectations, as it is apparent that the
decrease in quality is due to the deposition of the unfixed or light oil vapors
that are removed from the gas by pressure, and by the low temperature and
scrubbing action of the jnains.

47. To further confirm these results, at the request of the Committee,
investigations were also made under the direct supervision of one of its mem-
bers, on carburetted water gas, manufactured in a large city, in another State,
compressed pressure varying from ten to twenty pounds and delivered to
several outlying districts. These districts consisted of two smaller cities, sev-
eral small boroughs and small settlements on both sides of a large river. A
large portion of this territory is so situated that the inhabitants would have
been unable to obtain gas under any other conditions. The high pressure is
reduced and the gas supplied to the two cities through low pressure distribu-
tion systems, but the boroughs and small settlements are directly supplied by
high pressure which is reduced on the consumers' premises. An outline map
page 53 shows the general plan of the system.

48. The gas as manufactured at A averaged about 22 candle power. At
the time of the investigation the gas was compressed to about 16 to 20 pounds
pressure, as shown in the tables, and then delivered through a high pressure
system, consisting of 9,700 feet of 6" and 1,875 feet of 4" main, to a second
testing station, B. From this point the 4" high pressure main is continued
across the river for a distance of 16,625 feet, to a third testing station, C.

49. In crossing the river the main is exposed for a considerable distance,
and acquires the temperature of the river water, which at the time of the test
approximated 50 P.

50. At Station A, where the gas was compressed, tests were made before
and after compression. At Station B tests were made on the gas direct from
the high pressure lines before entering the low pressure system. At Station C
the gas was tested directly from the high pressure lines.

51. In order to obtain samples of the same gas, the capacity of the mains
and the estimated rate of consumption was determined, and the lag was
calculated.

52. To determine the illuminating value of the gas, tests were made on
standard 60" bar photometers at Stations A and B, and on a 60" portable
photometer at Station C, ten candle power pentane lamps being used as
standards.

53. The heating value of the gas was read on new calorimeters of the
American Meter Company, which were compared with each other just previous
to testing. All instruments and accessories were tested and calibrated before

Di5~rr?iBL/r/or</

being used. The humidity of the atmosphere was determined and correspond-
ing corrections made in the calorific values.

54. The temperature of the atmosphere, the temperature of the ground
and the dew point of the gas were determined. The temperature of the ground
ran from 56 to 58 F., while the dew point gave indications as low as 30 F.,
indicating combined effect due to the compression and cooling.

55. These readings were made over the period from October 31 to No-
vember 14, 1912, and the results obtained from day to day, as shown by the
following table, substantiate the conclusions presented above that the illum-
inating value cannot be maintained if the gas is to be transmitted to outlying
districts at relatively high pressures; but that under the same conditions it is
possible to maintain the heating value with only slight loss.

56. If these tests were duplicated in the extreme cold weather months of
the winter, we might expect even greater losses in illuminating value at Sta-
tions B and C.

Loss in illuminating and heating value of carburetted water gas, com-
pressed to 16 pounds and transmitted distances of 2.2 and 5.1 miles :

Illuminating Per Cent. Heating

Value, Loss. Value, Per Cent.

0. P. B. t. u. Loss.

Initial gas 21.69 ... 630

Loss due to compression 2.66 12.3 8 1.27

Loss due to transmission of 2.2

miles . .74 3.4 2 .32

Total loss 3.40 15.7 10 1.57

Initial gas 21.69 . . . 630

Loss in compression and trans-
missionof 5.1 miles. . 6.68 30.8 11 1.75

COMPRESSION AND TRANSMISSION TESTS ON ENRICHED COKE
OVEN GAS, NOVEMBER-DECEMBER, 1912.

57. The Committee also made some investigation of the effects of com-
pression and transmission on the illuminating and heating values of enriched
coke oven gas. As there was no such installation in the State, permission was
obtained elsewhere to make such a test on a plant consisting of forty ovens
and having a capacity for carbonizing three hundred tons of coal daily.
A good grade of Pennsylvania gas coal was used, with an average coking
period of twenty-two hours. Run-of-oven gas, enriched to about 16 candle
power, flat flame, was obtained, using a 90 per cent, benzol enricher. During
a part of the test a small amount of high candle power carburetted water gas
was made to help out the deficiency in the make from the coke ovens. In, both
cases, after enrichment at ''A," the gas was compressed at "A-l" to between
30 and 40 pounds per square inch, and then delivered to "B," a distance of
nine miles. The terminal pressure at "B" was between 26 and 36 pounds.
This high terminal pressure was necessary because of distribution to surround-
ing territory. After compression the gas was cooled to about atmospheric
temperature in a condenser, the condensate being removed.

58. At the terminus of the high pressure line the gas was expanded into
a holder and distributed at low pressure.

59. The illuminating value and the heating value of the gas at "A,"
"A-l" and "B" is shown in the tables. These tables also give the losses both
actual and in per cent, of original quality of the gas.

60. The ground temperature at "A" was about 47 F. and at "B" it
was 49 F. The dew point at "A-l" was 31 F. and at "B" it was 12 F.
The average outside air temperature was 39 F.

61. The tests were made with improved American Meter Company calori-
meters of the Junker type, and standard 60" bar photometers. All candle
power readings are flat flame value against a standard 10 candle power pen-
tane lamp. All apparatus was tested and checked for accuracy.

Loss in illuminating and heating value of enriched run-of-oven coke oven
gas, due to a compression of 30 pounds and a transmission of 'nine miles:

Illuminating Per Cent. Heating Per Cent.

Value, Loss. Value, Loss.

C. P. B. t. u.

Initial gas 16.2 ... 598

Loss due to compression 30 Ibs. 1.8 11.1 20 3.34

Loss due to transmission 9 miles 4.0 24.7 21 3.51

Total loss 5.8 35.8 41 6.85

Run-of-oven coke oven gas with an addition of 15 per cent, carburetted
water gas :

Initial gas
Loss due to
Loss due to

lumSnating
Value,
C. P.

16.9
2.7
5.1

Per Cent.
Loss.

16.6
30.2

Heating
Value,
B. t. u.

597
18

Per Cent.
. Loss.

3.02
3.85

compression 40 Ibs.
transmission 9 miles

Total loss 7.8 46.2 41 6.87

62. A previous test had been made on the same system, by one of the
Committee, in March and April, 1907 ; and while at that time no determina-
tions in loss of heating value were made, the loss in illuminating value was
determined.

63. In addition to the effect of compression and transmission, a test was
made on the effect of low temperature by passing the gas through a freezing
coil.

64. The following tables give a summary of the results obtained on these
tests.

65. The first table shows the effect of freezing on the candle power of
low pressure coke oven gas, carburefted water gas and mixed coke oven gas
and carburetted water gas.

Enriched Coke Carburetted Enriched Coke
Oven Gas. Water Gas. Oven Gas and

20% Carburetted
Water Gas.

Initial candle power 15.92 19.15 16.32

Loss due to compression and freez-
ing to 32 F 2.89 2.63 2.43

Loss in per cent, of initial candle

power 18.2 13.7 17.5

66. The second table shows the effect of compressing to pressures approxi-
mating 30 pounds and transmitting through a distributing system 11.5 miles.
The first test was on run-of-oven coke oven gas, benzol enriched, and the sec-
ond test was made on this same gas with a mixture of carburetted water gas,
equalling 27 per cent, of the total volume.

67. Hygrometer readings indicated a dew point of 18 F. on the com-

pressed and delivered gas.

Enriched Coke
Oven Gas.

Initial candle power 17.45

Loss due to compression of 30 pounds and

delivered 11.5 miles at high pressure.. 6.02

Loss in per cent, of initial candle power. . . . 24.5

Enriched Coke

Oven Gas Mixed

With, 27% Car-

buretted Water

Gas.

17.0

4.82
28.4

LABORATORY EXPERIMENTS TO DETERMINE THE EFFECT OF COM-
PRESSION AND FREEZING ON CARBURETTED
WATER GAS.

68. To determine further the effect upon the illuminating value and heat-
ing value of the gas, certain laboratory experiments were carried on under the
supervision of the Committee, with the following results :

69. Carburetted water gas of 23.5 candle power and 654 B. t. u. heating
value, was compressed first to low pressures of three and ten inches of water
and then in stages of five pounds each to a total of thirty pounds gauge
pressure, and while under compression reduced to a temperature of 35 F.
The condensation formed, due to compression and cooling, was removed, and
the gas, when expanded again, showed a dew point of 15 F. No serious losses
in heating value were found to take place in pressures up to ten inches of
water. Such losses, however, became evident as soon as the pressure had
reached five pounds and over. At thirty pounds' compression the illuminating
value had dropped 32.4 per cent., while the heating value dropped 6.8 per cent.
It was noted that the gravity of the gas changed but very little, as shown by
a test on an effusion apparatus. The illuminating value seemed to drop
uniformly with the compression, while the heating value dropped 30 heat
units or the first ten pounds and only twelve additional heat units for the
next twenty pounds. The results are shown in detail below:

B. t. u.

Pressure. Candle Power Gross.

3 inches, before cooling. . 23.5 654

3 inches 20.2 656

10 inches 20.3 654

5 pounds 18.1 631

10 pounds 16.3 621

15 pounds 15.6 624

20 pounds 14.95 615

25 pounds 14.2 617

30 pounds 13.7 609

Note. Original gas passed through 70 F. coil,
through 35 F. coil.

Loss Per Cent.

Candle Power B. t. u.

10.6
19.5
23.0
26.2
29.9
32.4

3.5
5.0
4.6
6.0
5.6
6.8

Compressed gas passed

HEATING VALUE CALCULATED BY ANALYSIS

70. One Company, No. 8, ran its plant (Class C, Table I.) to determine
as nearly as may be the heat unit equivalent to the present standard of candle
power. This plant was equipped with a Junker automatic continuous regis-
tering calorimeter, which was destroyed by fire before the completion of the
full year's test. In addition tests were made daily on both a Sargent
and on a Junker calorimeter. Samples of gas were taken and analysis of gas

made and recorded. The candle power was likewise determined on a 60"
U. G. I. open bar, using No. 7 L. P. Slit Union Bray and new F Argand Sugg
pattern gas burners against candles in accordance with New York legal
requirement.

71. Not until the completion of the year's test was any use made of
the gas analysis so that the calculated heat units therefrom have special
interest as a check.

72. The daily heat unit readings therefore were checked by an automatic
instrument self recording and finally by the calculated B. t. u. from the gas
analysis. The gas manufactured was carburetted water gas. The gas generat-
ing apparatus, 5' and T 6" sets, have up and down steam connections and
automatic continuously registering pyrometers on superheater. As the plant
is a small one the running was intermittent, varying from 8 hours to 23.8
hours. Gas oil was used, "28 B. " which had been contracted for prior to
beginning of test. Anthracite broken coal was used, with some gas-house
coke at times for experimental purposes. Oil corrected varied from 3.5 to
4.65 gallons per thousand. Fuel per 1,000, corrected, varied from 37.62 to
40.76.

73. The equivalents used in calculating the B. t. u. from gas analysis were :

Illuminants 2350.0 B. t. u. per cu. ft

Co 323.5

H2 326.2

CH4 1009.0

The results were as follows :

COMPANY NO. 8.
Average Monthly Results at Works B. t. u.

B t. u. B. t. u. Calculated

Candle Power. Junkers Sargent. from Analysis.

1911.

August 19.1 617 613 606

September 18.9 610 607 605

October 18.7 594 591 589

November 19.7 633 628 626

December 22.5 645 651 651

1912.

January 22.4 653* 656* 658*

February 22.6* 650 653 636

March .*....' 20.3 621 625 614

April 18.7 602 601 585

May 19.5 592 593 569f

June 19.3 592 591 582'

July 17.9f 584f 584f 580

* Maximum, f Minimum.

74. The official testing station of this Company was equipped with a
Junker calorimeter and 60" American Meter Company open bar photometer
using similar burners and candles. Gas was analyzed as at plant and B. t. u.
calculated with the following results:

Monthly Averages.

B. t. u.

B. t. u. Calculated

Candle Power. Junker. from Analysis.

1911.

August 20.0 613 608

September 19.6 605 602

October 19.3f 584f 581t

November 19.9 622 618

December 20.9 642 645

1912.

January 21.1* 643* 655*

February 20.6 631 634

March 19.8 607 610

April 20.4 591 582

May 19.9 590 568

June 20.7 596 586

July 20.6 586 577

* Maximum, f Minimum.

75. The Gas Inspector of the State tested the candle power for the State
record twelve times during the period of testing, finding candle power
averaging 20.1. The Company's test for the same days averaged 20.0 candle
power.

76. It will be observed that the B. t. u. calculated from analysis varies
from the actual readings made by calorimeters in different seasons. The effect
of temperature affects the character of the illuminants remaining in the gas
at different seasons. The following calculations were made as a matter of
interest. The values of B. t. u. per cubic foot of C O of H 2 and C H 4 were
taken as stated heretofore and the actual B. t. u. observed in the Junker
calorimeter was assumed in each case as the B. t. u. value of the gas. From
these figures the B. t. u. value of the illuminants was calculated with the
following results :

Monthly Averages.
Plant.

-Plant -Testing Station-

Date. B. t. u. Cal. Val. of B. t. u. Cal. Val. of

Junkers. Illuminants. Junkers. Illuminants.

1911.

August 617 2458 613 2399

September 610 2400 605 2377

October 594 2404 584f 2378

November 633 2410 622 2385

December 645 2304f 642 2321

1912.

January 653 2311 643* 2251f

February 650* 2465 631 2320

March 621 2408 607 2325

April 602 2450 591 2429

May 592 2582* 590 2576*

June 592 2449 596 2454

July 584f 2396 586 - 2336

* Maximum, f Minimum.

The above results clearly indicate the uncertainty of chemical analysis

as a means of accuracy determining the calorific value of manufactured gas.

Comparison of Continuous and Intermittant Operations.

77. The manufacture of gas by Company No. 9 was under the same
engineering superintendence as Company No. 8. Tests in this case were made
only at the testing station over a mile from the manufacturing plant. This
plant (Class A, Table 1.) runs continually night and day manufacturing
carburetted water gas; generators 11' and 12' reverse steam Lowe type with
indicating and recording pyrometers. Gas oil 28 B. using (corrected) 3.41
to 4.34 gallons per 1,000 cu. ft. Anthracite fuel (corrected) 30.45 to 32.67
pounds per 1,000. Calorimeter used Junkers. Candle power determined by
American Meter Company, 60" open bar photometer; burners No. 7 L. P.
Slit Union Bray and New F Argand. Analysis of gas was made only occa-
sionally. The results follow:

Monthly Averages.

B. t. u.

Candle Power Junker,

1911.

August 20.8 589

September 20.6 597

October 21.2 619

November 20.4 625

December 20.7 638

1912.

January * 20.9 631

February 20.7 621

March 20.7 610

April 20.2 600

May 20.6 589

June 20.6 595

July 20.6 597

78. A comparison of the Company's candle power reading taken at the
time B. t. u. results were taken on thirty days when the State Inspector tested
the gas of the Company shows 20.4 candle power average by State test and
20.8 candle power by Company's observer.

Comparison of Efficiency of Open Flame and Mantle Burners.

79. In replacing the flat flame burner with the mantle burner, it is a
matter of importance to determine the relative efficiency of these burners when
using the same quality of manufactured gas. It is aJso important to inquire
if similar results are obtained when using mantle burners with unenriched
coal gas and the present 20 candle power water gas. The State Commission,
Second District, requires that five cubic feet of carburetted water gas shall
give 20 candle power when burned in the burner best adapted to it, and
applicable for general use, when compared with the light of two standard
English sperm candles. The candle power obtained with the normal mantle
burner is uniformly greater than obtained in the flame burner even when
using only three cubic feet of the same gas. It is customary to state the
efficiency of the gas mantle in terms of candle power per cubic foot of gas used.

80. Under the supervision of a member of the Joint Committee mantles
costing at retail 10 cents, 15 cents, and 35 cents, mantle burners costing 10
cents and 50 cents, and chimneys at a uniform cost of 10 cents were purchased
at shops patronized by the general public.

81. These mantles and mantle burners were tested using carburetted
water gas from 19.4 candle power to 21.8 candle power at burner pressures,
varying from one to three inches water. The B. t. u. varied from 603 to 622.
Sixty-five separate illuminating power tests were made with the mantles

when using this gas. The average candle power per cubic foot of gas when
used in the flame burner was 4.15 candles, while the same gas developed 13
to 21.7 candles per cubic foot when used in the mantle burner. The 13 candle
power test was obtained with gas giving 4.12 candles per cu. ft. in the flame
burner, yet the same type mantles gave 16.4, 19.1 and 19.9 candles per cu. ft.
with gas of lower candle power than 4.12 candles.

82. Seventeen tests, with 2.5 inch burner pressure, were made with these
mantle burners and the same type mantles, but using a 14.38 candle power,
unenriched coal gas, equivalent to an efficiency of 2.87 candles per cubic foot
of gas when used in the flame burner. The heat units of this coal gas varied
from 601 to 607 B. t. u. When this gas was used in the mantle burners from
13.7 to 21.4 candles per cubic foot of gas was obtained.

83. It will be observed that the efficiency of the mantle burner was
equally good with either 20 candle power carburetted water gas or with
14.38 candle power in enriched coal gas. It will also be noted that the mantle
burner is many times more efficient than the flame burner. In some caess the
flame burner shows only 13.4 per cent, relative efficiency when compared with
the mantle burner using the same gas or as one is to eight nearly.

84. A mantle burner ordinarily consumes three cubic feet of gas per
hour and with 16 candles per foot gives 48 candle power. To obtain this
candle power with a twenty candle gas in an open flame burner would require
twelve cubic feet of gas. This is an hourly saving in gas bills of nine cubic
feet. Assuming one thousand hours' burning per annum there is a saving of
9,000 cubic feet of gas, which represents a cash saving of from $8 to $18,
depending upon the price of gas. Four mantles per annum would be a fair
average for this number of hours' use at an annual cost of from 40 cents to
$1.40; as stated, a burner which should give good service for many years
costs from lOc. to 50c., and the chimneys lOc. each. Smaller mantles consum-
ing but one foot to one and one-half feet per hour, where less illumination is
required, would show a corresponding saving.

APPENDIX D
STANDARDS IN OTHER PLACES

1. Up to 1906 practically all standards for gas, both in this country and
abroad, prescribed a certain illuminating value but contained no requirements
as to heating value. As early as 1894, however, gas engineers had begun to
recognize the inadequacy of a photometric standard and the necessity for
determining a calorific value.

2. The earliest date of a calorific standard being established was 1906.
Since that time this measure of the quality of gas has been adopted quite gen-
erally abroad and in a considerable number of instances in this country.

3. Generally speaking we find no scientific reasons for the figures which
have been adopted, and in few cases does there appear to have been any pre-
liminary investigation made along the lines of practical operating experience
under varying conditions.

4. Wisconsin, which was the pioneer in the establishment of a state
standard of heating value, did, through the Wisconsin Railroad Commission,
conduct a series of tests before the adoption of the requirement now in force.
These tests, however, did not cover a sufficient period of time or a wide enough
variety of conditions to be conclusive. It should also be remembered that long
strides have been made in the science of calorimetry during the last year
or two, and manufacturing conditions have undergone important changes.

5. At the present time heating value requirements are in force in five
states and in some thirty-one cities. A list of the states and cities with their
respective requirements follows:

States.

Wisconsin Monthly average 600 gross B. t. u. Minimum 550

New Jersey Monthly average 600 gross B. t. u. Minimum 550

Nevada Monthly average 550 gross B. t. u. Minimum 500

Washington .Monthly average 600 total. Minimum 550

Indiana 600 B. t. u.

Cities.

Aurora, HI. 600 gross Elkhart. Ind. 600

Birmingham, Ala. 575 gross Elyria, 0. 600 "heat units"

Cedar Rapids, la. 600 Freeport, HI. Monthly ave . 600

Chicago, 111. 600 gross Freeport, HI. Minimum 550

Dallas, Tex. 650 Helena, Mont. 500

Detroit, Mich. 600 gross Kankakee, 111. 600 "heat units"

Elgin, 111. 600 net Ottawa, HI. 600

Ft. Wayne, Ind. 550 Sault St. Marie, Mich. 500

Indianapolis, Ind. 600 Port Huron, Mich. 600 gross

Jackson, Mich. 600 Stockton, Cal. 600 gross Mo. ave.

Joliet, 111. 600 "heat units" Minneapolis, Minn. 600

Kalamazoo, Mich. 600 gross Minneapolis, Minn. Daily min . 550

Lansing, Mich. 600 (low value) Oakland, Cal. 600

Lincoln, Neb. 625 Omaha, Neb. 600 net

Los Angeles, Cal. 600 gross San Francisco, Cal. 600

Milwaukee, Wis. 635 gross Springfield, HI. 650
Adrian, Mich. Average of 600

6. In 1906 the calorific standard was adopted for Tottenham, the first
place in England to have such a standard. This standard was fixed by agree-
ment, but the Gas Light and Coke Company of London was the first to have
the heat unit test statutorily applied to town gas. This was accomplished
through the Parliamentary Committee of the London County Council advised
by Dr. Frankland, Charles Hunt and Dr. Clowes, the figures being set at ap-
proximately 500 net B. t. u., with a minimum of 450 B. t. u.

7. In Europe Paris has adopted a calorific standard of 528 net B. t. u. and
abandoned photometric requirements. The same standard holds in Rheimes.
Marseilles has adopted 551 B. t. u. and Milan, Italy, 573 B. t. u.

8. In 1909 German chemists concluded that they ought to have a calorific
test, and that the figures should be set at 543 gross B. t. u. with a minimum of
522, these results to apply to tests made at the works. There is even yet no
generally accepted standard of calorific value in Germany, although tests have
been regularly made in Berlin, Magdeburg, Bonn and Breslau for at least
seven years. Zurich was also testing for calorific value at least five years ago.

9. A member of the Committee from personal observation ascertained
that many continental cities do not even maintain photometric apparatus for
testing gas. Included in this number are Paris, Brussels, Ghent, Bruges,
Nuremberg, Stettin, Berlin, Charlottenburg, Warsaw and Zurich. No attention
is paid to candle power. Another authority from personal inquiry and obser-
vation confirms this tendency by information derived within six months that
the same conditions may be found- in Austria and Austria-Hungary as well
as Germany and other continental countries.

10. It should be noted that in Continental countries, it is customary to
correct the gas volume to 0C (32 F.) and 760 M. M. (30" Barometer). The
correction to 32 F. instead of 60 used in the United States and by this
Committee, means that the standard abroad is actually over 5% lower than
the figures quoted above would indicate. For example, a German gas con-
sumer receiving 543. gross B. t. u. per cubic foot is actually receiving not
more than 51.6 when United States methods of measurements are followed.

11. In South America Colombo has had a heat unit standard of 400
B. t. u. for about five years, and in Buenos Ayres it is required that the net
B. t. u. shall not be less than 539.

12. It will be seen from the preceding figures that the calorific power
requirements in Great Britain, on the Continent and in South America are
generally lower than those in the United States, due to a recognition of the
economic advantages of permitting the use of unenriched coal gas manu-
factured from such grades of coal as were available.

13. The standards which have been adopted in this country in the past
are in most cases in excess of what should be required. They were usually
set arbitrarily and without thorough investigation. It is inconceivable that
they would have been so fixed with present available raw materials. Refer-
ence to Table IV., page 20, shows monthly averages of heating power of gas,
enriched to meet the present illuminating power standards as low as 584
B. t. u. and the table in Appendix C, page 43, shows monthly averages of
heating power of unenriched coal gas, manufactured, however, from the
highest grades of coal, as low as 556 B. t. u. These are not individual readings,
but are averages for entire months.

APPENDIX E
CALORIMETRY AND PHOTOMETRY

Calorimetry

1. Assuming that a suitable calorimeter is employed, that the operator
has had reasonable experience and proper instruction, and that the accepted
rules of procedure are followed, it is proper to inquire as to the consistency
of the readings and the accuracy of the results that will be obtained.

2. One of the smaller Companies participating in the investigations has
had check tests made on practically every working day during the entire
period and with three observers, of whom two were no more than high school
graduates, only 5 per cent, of the time was there a difference of 3 B. t. u.
between the tests, while on 88 per cent, of the days the difference. was 2 B. t. u.
or less. The determinations came out exactly the same, decimals being of
course excluded during 32 per cent, of the time.

3. Such results can, however, only be secured by an absolute conformity
to operating instructions. The great danger with calorimetric determinations,
as in photometric work, is that the operator will drop into the habit of regard-
ing this or that minor detail as unimportant and consequently will finally dis-
regard it altogether. To illustrate, some operators only read their outlet water
thermometers to tenths of a degree F., yet an error of 0.1 F. means an error
of about 4 B. t. u. in the result.

4. The pressure at which the gas is metered is an important factor, al-
though by some it is considered only a negligible quantity and is left out of
consideration. If this is not corrected for, and the gas is delivered to the
meter at 3 inches water pressure, the final result will be about 4 B. t. u. too
high.

5. The rules as laid down by this Committee in its two pamphlets are not
difficult to follow and every point emphasized has a practical value and is nec-
essary to successful determinations. If this is borne in mind there should be
no difficulty, under ordinary conditions, in obtaining consistency of results.

6. The accuracy of results, however, assuming that the test has been care-
fully and correctly carried out in every particular, is a different question and
demands separate treatment. We shall assume that the operator is endeavor-
ing to obtain the total heat value, or as near thereto as is practically possible,
and that he has made all of the usual corrections, i. e., for thermometer error
and stem exposure, temperature and pressure under which the gas is measured,
and efficiency of instrument. If the inlet water were of the room temperature,
and the products of combustion left the instrument also at room temperature,
there is still a correction to be applied for the humidity of the air entering
the calorimeter. This can be obtained by the use of a psychrometer, the cor-
rection then being found in the table furnished by the National Bureau of
Standards and published in the Proceedings of the American Gas Institute for
1912. This table, however, consists of two parts: (a) Where seven volumes
of air are used per volume of gas, and (b) where nine volumes of air are used.
Since there is at present no practical method of measuring the air passing
through the calorimeter, it would seem as if there were an insuperable diffi-
culty at the outset. But if we examine the two tables we find that' for ordinary

working conditions the differences are not large, as will be seen from the fol-
lowing which have been compiled by taking the difference in B. t. u. between
the corrections in the two tables for the same temperature and humidity:

Humidity in P.O. Boom Temp. 65 70 75 80 85 90

10 1.9 2.2 2.5 2.9 3.3 4.0

20 1.7 1.9 2.3 2.5 3.0 3.5

30 1.5 1.6 2.1 2.2 2.6 3.1

40 ....: 1.2 1.5 1.8 1.9 2.2 2.6

50 1.0 1.2 1.5 1.5 1.9 2.2

60 0.7 0.9 1.2 1.2 1.5 1.7

70 0.6 0.6 1.0 1.0 1.1 1.3

80 0.3 0.4 0.6 0.6 0.7 0.8

90 0.1 0.2 0.4 0.4 0.4 0.5

100 0.0 0.0 0.0 0.0 0.0 0.0

In every case the corrections in Table B are algebraically the greater, but
it must be remembered that the low ranges of humidity, as well as the ex-
cessive room temperatures, are comparatively rare, at least in this part of the
country, and moreover, these extreme conditions do not, as a rule, occur at the
same time. If, therefore, the ordinary working conditions are considered,
such as a room temperature between 65 and 80, and a humidity over 30 per
cent., it will be seen that the maximum error likely to occur from use of the
wrong table is only 1.5 B. t. u. or about one-fourth of 1 per cent. But there is
another factor which still further tends to diminish the chances of error from
this source. With the exhaust damper set properly and with the air mixer on
the burner so adjusted as to give the most perfect combustion (a condition
easily judged by the color and general appearance of the flame) the excess of
air admitted to the calorimeter is a constant within reasonably close limits,
and the seven volumes of air per volume of gas will ordinarily be the mixture
employed. Thus it will be seen that, with the aid of the correction table and
a humidity determination, the error due to humidity will never be over
4 B. t. u., and, under working conditions, will probably average about 1.5
B. t. u. The application of surface combustion may well be studied with the
object of minimizing the quantity of air used in the calorimeter.

7. Another factor which is intentionally disregarded in practical work is
the specific heat of the water. Since the custom of weighing the water is now
almost universally adopted, any error from this source has usually been con-
sidered so small as to be unworthy of attention. In a study of this matter by
Leo Loeb, in March, 1911, he states that, in gas calorimetry, the possible varia-
tions due to this cause are from 0.13 to 0.25 per cent. This would mean a
maximum error of about 1.4 B. t. u.

8. An error which cannot be readily computed is introduced through the
fact that the gas is not saturated with moisture as it enters the burner. It has
always been assumed that gas after passing a wet meter would be very nearly
saturated with water vapor, one author stating as the result of his experiments
that the saturation was over 98 per cent. Recent investigations seem to render
so high a result questionable, but it is the consensus of opinion that the error
due to this cause is so small as to be negligible in practical work ; it will prob-
ably be less than 0.5 B. t. u.

9. One other factor affecting the accuracy of results remains to be
considered, and this is loss from radiation. This matter has received con-
siderable attention from the National Bureau of Standards, which has found
that if proper baffle plates are placed on the burner, radiation losses may be
reduced to about 0.3 per cent, or about 1.75 B. t. u.

10. If we now combine all of these errors, and assume them to be all in
the same direction (which is not true) we should have a total maximum error
of about 5.5 B. t. u., or less than 1 per cent. The error due to radiation, how-

ever, tends to make the observed result too low, while the fact that the gas is
not saturated makes the observed result higher than the true figure. The error
introduced by the specific heat of water makes the observed result too high,
while that due to the unknown volume of excess air admitted may be either a
positive or a negative correction. Combining these algebraically, we get a
possible error of -f-1.5 B. t. u., or about 0.25 per cent.

Photometry

11. Even with the most complete photometric outfit it is not possible to
secure results which are accurate within less than 1 or 2 per cent. This is
due to a number of factors which are variable and cannot be corrected for :
the personal equation, atmospheric conditions, quality of pentane, deteriora-
tion of standards lamps and of discs, varying water line in the meter and in-
accuracy of the latter, etc. If this be true of the best type of apparatus, how
much more true it is of photometers as generally employed, with candles as
standards and burners which do not begin to develop the full illuminating
power of the gas.

12. In New York State candles have been made the official standard, prin-
cipally for the reason that it did not seem practical to employ either a pentane
or a Hefner lamp in connection with the portable photometer which the Com-
mission's inspectors are obliged to use in a large number of places.

13. It appears to have been conclusively established that the candle as a
standard is unreliable. Instead of attempting to take up this subject our-
selves, we refer to the following as being among the most notable discussions
to be found in the voluminous literature dealing with this question Dutch
Photometric Committee, 1894, C. 0. Bond's paper on Working Standards of
Light, American Gas Institute Proceedings, 1907 ; Dr. Love 's paper on Stand-
ard Candles, and the recent work by C. E. Crittenden on the variations in
illuminating value of candles.

14. To cite the last authority only, Mr. Crittenden publishes curves show-
ing fluctuations from minute to minute of over 10 per cent, while the average
approached closely 2.1 candle power per pair. If to this be added the errors in-
herent in a portable photometer, the inaccuracies of a dry meter for photo-
metric work, the troubles furnished by the candle-balance and the use of im-
proper burners, the result obtained is most unsatisfactory. The last item, how-
ever, is one of the most important and should receive special consideration. '

15. In this country the two states which have given the most thorough
study to photometric work have adopted the practice of employing the burner
best adapted to the gas to be tested ; the burners selected would be either a
slit union Bray, a Suggs table top or one of the various styles of Argand
burners. The result of this is that, in order to make a satisfactory test, several
burners must be tried.

16. While, as a rule, the new style F Argand will give the best results
from a water gas of between 17 and 21 candle power, it very frequently hap-
pens that, even under these conditions, the Bray burner is superior, and there-
fore both burners must be tried in every case if a correct result is to be
assured.

17. With a coal gas the matter is still worse, for there are at least five
styles of Argand burners which may be tried, each adapted to a little different
quality of gas and each liable to give the best result on a gas of unknown
candle power.

18. After all of these have been used, and the maximum result secured,
is this the candle power of the gas ? No ; for there is at least one other burner
which might be employed, the Metropolitan No. 2, which would give a much
higher candle power than any of the above.

19. This latter burner has been adopted as official by the London Referees
and is recognized by Parliament. It has not met with favor in this country for
two reasons : First, its initial cost, which places it beyond the reach of most
consumers; and, second, the fact that it gives so much greater candle power
than the burners now used. Neither of these is a scientific reason, for science
is not concerned with cost and is concerned with securing as nearly a theoret-
ically correct result as possible.

20. It has been argued that the illuminating value of gas as measured
in candle power is not a definite scientific quantity and in this respect differs
from the calorific value. This does not nullify the argument that the manu-
facturer is entitled to a judgment of his product based upon the best that can
be attained therefrom using the most accurate scientific instruments.

21. It is also sometimes urged that illuminating value should be measured
with an open flame burner, because the results thus obtained would more
nearly represent the value of the gas to the consumer. Such a method would
not measure the illuminating value of the gas, but rather the efficiency of the
burner used. There are a number of types of open flame burners in general
use whose efficiency varies from 50 to 90 per cent.

22. If to test the gas the Metropolitan No. 2 burner is used, the consumer
will be getting the gas of a certain fixed illuminating value and the measure of
light which he will obtain from a given quantity of gas will depend entirely
upon the efficiency of his burner.

23. From the above discussion it is evident that a calorimetric test meas-
ures much more accurately the heating value of the gas than a photometric
test measures the illuminating value. In the test itself, even eliminating the
question of burners, the percentage of probable error is much less in a calori-
metric test than in a photometric test. It will also be seen that calorimetric
tests may be made with sufficient accuracy by other than highly technical men
as long as the prescribed rules of procedure are followed.

INSTRUMENTS USED IN INVESTIGATION AND CALIBRATION
WORK OF PUBLIC SERVICE COMMISSION.

24. The selection of those calorimeters that assure accuracy in determin-
ing the heating value of the gas, and yet are simple in operation, received con-
siderable attention from the Committee. For this purpose the work that had
been done on the subject of gas calorimetry was reviewed. A study was made
of the reports of the Calorimetry Committee of the American Gas Institute
made in the years 1908, 1909 and 1912, and in addition consultations were held
with the members of the staff of the National Bureau of Standards, Washing-
ton, who have had this subject under investigation.

25. It was found that at the present time there are a number of instru-
ments in use and on the market designed to measure the heating value of gas,
but employing different underlying principles in their operation. After giving
the matter much consideration, it was found advisable to employ calorimeters
for this investigation of the water heater type, and only those that expressed
directly the heating value of the gas, when burning a known quantity of gas
and imparting the heat developed to a known quantity of water.

26. Of the calorimeters approved by the Committee, only three have been
used by the reporting Companies during this investigation. They are the
Junkers, the American Meter Company and the Sargent. These instruments
have all been calibrated and checked for accuracy at the Commission's labora-
tory at Albany and in their operation have proved satisfactory.

27. It is interesting to note in this connection the variation in efficiency
of instruments after a period of continuous operation for one or two years.
We have received figures from the laboratory of the Commission for one instru-
ment tested on November 23, 1910, and again on October 15, 1912. The effi-
ciency in the first case was 99.6 per cent., and in the second case 99.5 per cent.
Another instrument was tested on January 27, 1911, and again on November
25, 1912, the efficiency at the first test showing 99.8 per cent, efficiency and in
the second case 99.4 per cent, efficiency. Another instrument tested November
18, 1910, and having an efficiency at that time of 99.5 per cent., was tested
again on December 16, 1912, about two years, and showed an efficiency of 99
per cent.

28. The photometrical measurements were made in the usual way and in
accordance with the State requirements.

29. The Primary Standard was tested by the Bureau of Standards at
Washington by means of an electrical burner, with the following results :

October 22, 1910 Efficiency 99.5%

April 22, 1912 Efficiency 99.8%

30. All wet meters were calibrated prior to the investigation. The
calibration of calorimeter thermometers showed that those used in the inves-
tigation were of high class, the corrections applicable being very small.

31. In the table following the efficiencies of the calorimeters of the
participating companies are given. Column I. shows the efficiencies as
determined against the Primary Standard in the laboratory of the Com-
mission at Albany, before the companies began to report to the Committee.
Column IV. gives the results against the Primary Standard after the com-
panies ceased reporting. There is thus an interval of at least fifteen months
between the results in Column I. and IV. in nearly every instance, and in
some cases over two years. The efficiencies given in Columns II. and III.
were determined by the use of the Secondary Standard at the plants of the
companies by traveling gas inspectors of the Commission and at intervals
of several months.

EFFICIENCIES OF CALORIMETERS DETERMINED BY
PUBLIC SERVICE COMMISSION

Company
Number

Column
I.

Column
II.

Primary

Secondary

Standard

99.8

99.3

98.8

99.8

98.5

99.5

99.7

99.3

99.2

97.7

99.7

99.5

99.6

99.4

( i

99.5

98.8

c <

99.1

99.4

i I

99.6

99.3

Column
III.

Secondary
Standard

99.6
99.2
99.6
99.0

99.6
98.9

99.4
99.5
98.7

Column

rv.

Primary
Standard

99.3
99.4
99.4
99.0

98.4
99.1
99.3
99.6
99.2
99.3
99.5
99.6

Company

Number

Column
I.

Primary

Standard

99.3

99.6

99.3

98.2

99.3

99.8

16*

* Same instrument.

Column
II.

Secondary
Standard

99.1
99.9
99.0
99.2

99.0

Column

III.

IV.

Secondary

Primary

Standard

99.8

99.5

98.7

97.6

99.0

99.2

99.3

32. These tests not only indicate that the variation in efficiency of a
calorimeter is slight, but also that a more satisfactory result is obtained
when this calibration is performed in the laboratory of the Commission,
which has been especially equipped for just such work.

APPENDIX F.

REPRINT OF CALORIMETRIC RULES, REGULATIONS AND
SPECIFICATIONS.

Adopted May 6, 1910, by the Joint Committee on Calorimetry, representing
the Public Service Commission and G-as Corporations in the Second Public
Service District, New York State.

INTRODUCTORY NOTES.

(1) A preliminary inquiry into the heat units of gas supplied in New
York State was made in 1908 and 1909 by the Division of Light, Heat and
Power of the Public Service Commission, Second District, by direction of the
Commission. The inquiry was conducted under the immediate supervision of
Mr. Charles H. Stone, the Chief Inspector of Gas.

(2) The results of the determinations were submitted to the Commission
in a report by the Chief of the Division, Mr. Henry C. Hazzard, under date of
October 29, 1909.

(3) Under date of December 8, 1909, the Commission addressed the
following communication to each gas corporation:

"Albany, December 8, 1909.

"To Corporations Engaged in Furnishing or Distributing Coal Gas, Water
"Gas, or Mixed Gas:

"By resolution duly adopted, this Commission has appointed February 1,
1910, as the date for a conference with representatives of gas companies on
the subject of standards for the measurement of the value of gas.

"The particular object of the conference is to obtain an interchange of
views on the necessity for a calorific standard, and on all questions necessarily
incidental thereto.

"A preliminary inquiry into the subject has been completed, the results
of which are embodied in a report by the Chief of Division of Light, Heat
and Power. For your information therein we are sending you under separate
cover a printed copy of this report.

"The conference will begin at 2 p. m., in the hearing room of this Com-
mission, at the Capitol, Albany, on the above mentioned date, and if necessary
will be continued the following day. You are respectfully requested to have a
representative present.

"Very truly yours,

"J. S. KENNEDY,

"Secretary."

(4) The conference on February 1, 1910, decided upon the desirability
of establishing a joint committee under whose immediate charge and direction,
subject to the approval of the Commission, the inquiry should be concluded.

(5) The Commission appointed to serve on such committee :

Messrs. Henry C. Hazzard, Chief of Division of Light, Heat and Power.

Howard H. Crowell, Engineer of Division of Light, Heat and
Power.

Charles H. Stone, Chief Inspector of Gas, Division of Light, Heat
and Power.

The gas corporations represented at the conference appointed to serve
on such committee:

Messrs. R. M. Searle, Rochester Railway and Light Company.
W. R. Addicks, Westchester Lighting Company.

T. R. Beal, Poughkeepsie and Newburgh Light, Heat and Power
Companies.

J. C. DeLong, Syracuse Lighting Company.

"W. T. Morris, United States Gas and Electric Company.

M. W. Offutt, Mohawk Gas Company.

(6) This Joint Committee, of which Mr. Henry C. Hazzard was elected
Chairman on February 11, 1910, appointed a sub-committee of three, consisting
of Messrs. W. R. Addicks, James C. DeLong and Charles H. Stone. The sub-
committee in the discharge of its duties submitted the following report, which
was duly adopted by the Joint Committee at a meeting held March 11, 1910,
and revised May 6, 1910, and ordered printed for the guidance of those
participating in the inquiry:

REPORT OF A SUB-COMMITTEE OF THE JOINT COMMITTEE OF NINE

REPRESENTING THE NEW YORK PUBLIC SERVICE COMMISSION,

SECOND DISTRICT, AND REPRESENTATIVES ELECTED AT A

MEETING OF GAS COMPANIES IN ALBANY,

FEBRUARY 1, 1910.

At a meeting of the Joint Committee held in Albany February 11, 1910,
resolutions were adopted providing for the appointment of a sub-committee
consisting of Mr. Addicks, Mr. DeLong and Mr. Stone to prepare and submit
to the Joint Committee :

(1) Specifications for a primary standard calorimeter;

(2) Rules and regulations for the installation and operation of calori-
meters at plants of gas corporations ;

(3) Suggestions as to suitable types of calorimeters for use, when
checked against the primary standard, at gas plants.

In accordance with the above, your sub-committee begs leave to report
as follows :

HEATING VALUE OF GAS.

(1) The definition of the heating value of gas adopted by your sub-
committee for the purposes of this report and the investigations to be here-
after conducted is that given by the American Gas Institute, Vol. III., 1908,
page 383, as follows :

"The heating value of a gas is the total heating effect produced by
the complete combustion of a unit volume of the gas, measured at a
temperature of 60 degrees Fahrenheit, and a pressure of 30 inches of
mercury, with air of the same temperature and pressure, the products of
combustion also being brought to this temperature.

' ' In America the unit of volume is the cubic foot and we recommend
that the heating value be stated in terms of British Thermal Units per
cubic foot of gas."

PRIMARY STANDARD TO BE MAINTAINED AT LABORATORY OF
THE COMMISSION AT ALBANY.

1. The Primary Standard shall be a new instrument, and shall consist
of the calorimeter proper, as manufactured by Junkers & Company, Dessau,
Germany, and as illustrated by Instrument No. 1221, now in possession of the
Public Service Commission of the Second District, State of New York.

2. The meter for measuring the gas shall be a wet meter, having a drum
capacity of 1/10 of a cubic foot for each revolution, and with an outside gauge
glass for indicating the water level. The dial shall read in tenths, hundredths
and thousandths of a cubic foot. This meter shall be of the size and pattern
supplied by the American Meter Company, fulfilling the above requirement.

3. The thermometers for use in determining the temperature of the
water entering and leaving the calorimeter shall be of the design recom-
mended by the Calorimetry Committee of the American Gas Institute. They

shall have a range of from 60 to 110 degrees F., shall be subdivided to read
1/10 of one degree, and shall have an auxiliary division at 32 degrees F. for
checking the ice point. They shall be calibrated throughout their entire range.

4. The thermometers for reading the temperatures of the gas, the atmo-
sphere and the exhaust products shall be graduated in degrees F., shall be
accurate to within one-half (Va) of one degree, and shall be calibrated through-
out their entire range.

5. Gas used for standardizing purposes shall be stored in a holder of
not less than 50 cu. ft. capacity.

6. The gas pressure at the inlet of the meter shall approximate existing
normal distribution pressure supplied to consumers of artificial gas, and this
pressure shall be added to the barometric pressure and taken into account
when making the barometric corrections as indicated hereafter.

7. The gas governor placed between the meter and the burner shall be
of the float type now supplied with the Junkers Calorimeter.

8. Arrangements shall be made for accurately weighing the water pass-
ing through the calorimeter, and the balance employed shall have a capacity
of ten (10) pounds avoirdupois, and give the correct weight at that capacity
to within 0.001 of one pound.

9. The calorimeter, balance, weights and thermometers shall be carried
to Washington and there standardized and calibrated by the National Bureau
of Standards. The gas meter shall be tested against a cubic foot bottle
bearing the seal of the said National Bureau of Standards.

10. The water supply to the calorimeter shall be filtered, and so arranged
that the water entering the calorimeter shall be of a uniform pressure and
temperature, and that temperature shall be within two (2) degrees of the
temperature of the atmosphere surrounding the calorimeter, and of the exhaust
products leaving the calorimeter.

11. The gas as metered and entering the calorimeter shall have approxi-
mately the room temperature, and shall be corrected to 60 F. and 30" baro-
metric pressure, the latter to be read from U. S. Signal Service type of
barometer.

12. The calorimeters shall be operated with the minimum quantity of
air to effect complete combustion of the gas, which shall be burned at a rate
giving the maximum calorific efficiency.

13. Corrections shall be made for atmospheric humidity.

14. The entire apparatus shall be installed in a proper room of the
Laboratory of the Public Service Commission, Second District, in Albany.

SECONDARY STANDARD TO BE USED IN CHECKING THE
CALORIMETERS OF THE GAS COMPANIES IN SITU.

1. The Secondary Standards used by the Public Service Commission,
Second District, New York, shall consist of calorimeters and accessories which
with the operating methods employed shall give, within 2%, the heating value
of the gas, as determined by the Primary Standard heretofore recommended.

2. The Secondary Standards shall be checked against the Primary Stand-
ard at such intervals as will maintain the calorimeters and accessory apparatus
in condition to fulfill the said 2% requirement, and the results of such tests
shall be recorded and filed.

3. The Secondary Standards shall be used for checking the calorimeters
of the companies engaged in experimental work relating to the heating value
of artificial gas ; said checking should be made not oftener than once in thirty
(30) days, and shall be made at least once in ninety (90) days.

IV.

GENERAL SPECIFICATIONS AND RECOMMENDATIONS FOR
CALORIMETER INSTALLATIONS BY GAS COMPANIES

1. We recommend the adoption of a calorimeter of the water heater type
(see ft 2), which when new shall be tested against the State's Primary Stand-
ard, and we feel that an instrument should be required to have an efficiency
within 2% of the Primary Standard (see page 71).

In determining the calorific value of the gas we recommend :

(a) The measuring of the gas in cubic feet (see ft 9).

(b) Taking all temperatures of air, gas and water with Fahrenheit
thermometers (see fl 22).

(d) Correction of the volume of the gas to standard volume, as expressed
when measured at a temperature of sixty (60) degrees Fahrenheit, and baro-
metric pressure of thirty (30) inches of mercury (see fl 29).

(e) Expressing the result of all calorific determinations in British
Thermal Units (B. t. u. 's) [see page 71].

(f) That at this time, with the information before us, we believe that a
calorimeter in commercial use may be expected to give results with an effi-
ciency within 3% of the Primary Standard, in which case it should be held
to be commercially correct. A record should be kept of the periodic tests
made by the State's Inspector with the Secondary Standard.

Calorimeter Proper.

2. The calorimeter proper shall be an instrument that transmits directly
the heat evolved by the burning gas to a quantity of water: it shall at this
writing be of a design operating on the principle as illustrated by that of the
Junkers Gas Calorimeter. This calorimeter shall be accompanied by acces-
sories that shall measure definitely the gas burned; the water heated and the
temperatures of the gas, water, air and exhaust products.

3. The apparatus should be designed to give a constant head of water
on the Calorimeter. This head should be maintained by having a weir overflow
on the inlet at some distance above the top of the calorimeter, and a weir
overflow at the outlet. The rate of flow through the calorimeter should be
regulated at the inlet by means of a cock with graduated scale.

4. The calorimeter should be so built that the water will circulate freely,
and will be equally distributed throughout the apparatus. Baffle plates should
be so arranged that the water will be thoroughly mixed before coming in con-
tact with the bulb of the outlet thermometer, insuring a correct average
reading. The design should be such that air pockets cannot form in the water
space of the calorimeter.

5. The calorimeter should be made of bright polished metal, air jacketed
in all its parts.

6. There should be a damper in the exhaust gas flue which can be easily
adjusted, and which cannot be moved by a slight jar.

7. The calorimeter should be mounted at a height sufficient to make it
easy to put the burner in place, and on legs with a spread great enough to
insure a firm base.

8. It may prove desirable in practice to have .water .thermometers. on
the same level, to facilitate readings, as recommended by the Calorimetry
Committee of the American Gas Institute. The openings for thermometers
should be large enough to take a No. 4 rubber stopper.

Meters.

9. For a meter, we recommend a wet meter, and one registering 1/10
cubic foot per revolution.

10. The large dial should be divided into 100 equal parts, with every

tenth part distinctly marked to facilitate reading. In addition to the large
dial, there should be a smaller dial to register the number of revolutions of
the large hand ; this dial should register tens, units and tenths of a cubic foot.

11. The face of the meter should be enameled and no glass used on the
front, thereby preventing error due to parallax. The face of the meter should
be easily removable, in order to get at the shaft and the stuffing box on the
shaft. This stuffing box should be of a size large enough to be easily packed.

12. The large hand of the meter should be well pointed, and not extend
to the outer end of graduations of the meter dial. The meter should have
leveling screws.

13. Two leveling tubes, placed at right angles to each other, should be
securely fastened to the top of the meter.

14. The meter should have an outside gauge glass showing the water
level. This glass should not be less than %-inch, nor more than %-inch, inside
diameter, as it is necessary to have the glass large enough t6 be readily
cleaned, and small enough that the meniscus formed by the water can be
accurately read. The openings from the gauge to the meter should be unob-
structed, and of a size to correspond with the size of the gauge glass. A fixed
point to show the correct water level, reading to the bottom of the meniscus,
should be put on the outside of all water level gauge glasses.

15. For convenience, a standard 3-light meter union should be used on
all meters, and hose nipples for %-inch hose should be furnished with the
unions.

*
74

16. The meter should be provided with an opening for the addition of
water when needed. This can be done by using a pet cock, with a small
covered funnel mounted on top, connected to the top of the gauge glass
support.

17. An opening must be left for a thermometer in or near the gas outlet.
This thermometer should have a metal case and read to one degree Fahrenheit,
with a range of from about 50 to 100 degrees, and accurate to within % degree.

18. An opening with a plug connection should be left on the bottom
of the meter to drain it when so desired.

19. The number of joints liable to cause leakage should be reduced to a
minimum.

Gas Pressure Regulator.

20. The pressure of the gas when burning in the calorimeter should be
absolutely uniform to obtain correct results, and any small regulator that will
maintain this uniform pressure will be satisfactory. We recommend the use
of a small wet governor, similar to the one supplied with the Junkers Calori-
meter. This will give excellent regulation, and will operate without chatter-
ing. N Such a regulator should be constructed so as to be readily weighted for
altering the delivered pressure.

Burners.

21. The burner should be a long tube Bunsen, having a spreader on top,
and adjustable air mixer which can be easily reached when burner is in
position in the calorimeter. The burner should be provided with a stop-cock.
The burner should be attached to the calorimeter in such a way that the gas
flame cannot impinge on the interior body of the calorimeter, and when the
burner is set at its correct position it should be so fastened that it cannot be
accidentally shifted. The condition of the flame should be observable by the
operator, either directly or by means of a reflecting mirror.

Thermometers.

22. Accurate thermometers are the most important accessories to correct
calorimetry.

23. The thermometers for reading water temperatures should be of high-
grade quality, and should read accurately within 1/10 of a degree Fahrenheit.

24. The thermometers should be graduated from 60 to 110 degrees
Fahrenheit, each degree to be divided into tenths, .with short, distinct gradua-
tions. The thermometers should be so accurately made that in ordinary
commercial work corrections may be neglected. With each thermometer
should be provided a calibration curve, which should enable very accurate
results to be obtained whenever it was deemed necessary to make these
corrections.

25. This matter of high-grade thermometers for calorimetry work has
been taken up with several thermometer makers by the American Gas Insti-
tute 's Calorimetry Committee, which reported that Messrs. Hohmann &
Maurer, of Rochester, N. Y., are now delivering a thermometer that has been
built according to its recommendations. The thermometers have a range of
from 60 to 110 F., and graduated to 1/10 degree, having an auxiliary divi-
sion at 32 F., which is convenient for carefully checking the ice point. These
thermometers are carefully made and have a bore that is exceedingly uniform
and accurate. This Committee hopes that other makers will place on the
market similar instruments.

26. The error of 1/10 above mentioned may seem to be a small matter,
and it is in most measures of temperature, but when the calorific value of an
artificial gas is determined with a rise in the water temperature of 15 F
a difference of 1/10 means an error of 1/150 of the total heat of the gas or
about four (4) B. t. u.'s.

27. When doubt arises as to correctness of thermometers, we recommend
their calibration by the National Bureau of Standards at Washington.

28. Telescopic sights for reading thermometers should be provided, as
much more accurate readings can be obtained in this way.

Barometer.

29. Corrections for variation in barometric pressure should be made in
measuring the volume of the gas. This pressure should either be obtained by
means of a mercury column barometer or by a recently calibrated aneroid
barometer. Where it is possible barometer readings should be checked occa-
sionally with readings of the Government Weather Bureau of the city in
which the readings are made. Where no barometer is available, it may be
possible to get fairly accurate figures on pressure by obtaining from the local
Weather Bureau the barometer readings for the day, and correcting for
variations in elevation.

Water Supply and Measurement.

30. The control of the temperature of the water supply is very important
in calorimetry, and this temperature should be approximately that of the
room in which the observations are being made. Water obtained from an
ordinary house piping system is apt to be variable in pressure and temperature,
due to the uneven consumption in other parts of the building, and possible
exposure of the water main to the extreme temperatures of the ground or
atmosphere. This control of temperature or pressure may be readily obtained
by providing a permanent water supply tank in the upper part of the calori-
meter room, that will contain enough water to enable the readings for the
day to be made. A flat tank of large horizontal area is preferable to a deep
vertical tank. The exposed surface allows the water to come to the tem-
perature of the room more readily, while the shallow depth has less effect on
the head as the water is being used.

31. Should a number of continuous readings be made that will require
more water than is contained in the overhead tank, a simple coil gas water
heater may be employed to raise the temperature of the water supply to the
overhead tank, so that it will enter this tank at approximately the temperature
of the room. The tank will then act as an equalizer and assist in maintaining a
uniform temperature and pressure of water entering the calorimeter.

32. Water may be collected and weighed in thin sheet metal containers,
holding about nine (9) pounds of water. This size container will hold all the
water required in burning 0.2 of a cubic foot of ordinary illuminating gas,
with a range of about fifteen (15) degrees Fahrenheit in temperature between
the inlet and outlet water. The scales, or balance, employed should have a
capacity of at least ten pounds, should read to 1/100 of a pound, and should
be calibrated and certified to as being correct by proper authorities.

33. Should it be desired to measure the water volumetrically, instead
of weighing it, graduated vessels may be employed that will read accurately
the water passed through the calorimeter to within 1/100 of a pound. Such
vessels, however, shall be approved by the Commission and calibrated (by
State or National authority) at 60 F. and accompanied by a curve to correct
for other temperatures.

Gas Piping and Tubing.

34. Gas connections for a calorimeter should consist of metallic piping
or tubing where possible ; rubber tubing is not advisable, but when necessary,
the lengths used in conducting the gas should be as short as possible, and
they should be thoroughly saturated with gas before a test is made.

Humidity.

35. It may be desirable to have the state of the humidity of the atmo-
sphere during the test, in which case percentage readings may be made from
wet and dry bulb llu'rmometers. For accurate work these wet and dry bulb
thermometers should be arranged so that the average humidity of the room
may be obtained. This may be done by having a whirling wet thermometer, or
having a constant current of air impinging upon the wet bulb from an electric
fan ; or, a more perfect instrument in the form of an Assman Psychrometer
may be obtained. These humidity readings of the atmosphere will not be
found ordinarily necessary in commercial calorimetry, but may be 'useful if
it is desired to make corrections for heat absorbed in saturating the products
of combustion.

Calorimeter Cabinet.

36. To facilitate the operation of the calorimeters at the various gas
plants, the calorimeters should preferably be installed in a cabinet, similar to
that recommended in the Report of the Calorimetry Committee of the American
Gas Institute, as contained in the American Gas Institute Proceedings, Vol. IV.,
1909, pages 205 and 206. This sketch represents a typical cabinet, suitable for
use in some convenient building, either at the gas works or gas office, and of
such a design that when the calorimeter is once placed and connected up, it
may be kept clean, protected and ready for use at all times.

37. In construction, the cabinet should be made as dust tight as prac-
ticable. "Where there is not enough head room for a vertical sliding door, hori-
zontal sliding or folding doors may be substituted. This cabinet should
provide for an overhead water tank, and may be most conveniently located
adjacent to a sink and water supply.

38. The gas supply line to .the calorimeter should have a purging con-
nection. All cocks controlling the gas and water supply should be inside of
the cabinet, and the cabinet should be kept closed and locked when not in
service.

39. This cabinet shall not be near any gas flame, register or other object
radiating heat; direct sunlight shall not be allowed to strike upon it, but the
thermometers and meter shall receive sufficient reflected artificial light to
enable them to be easily read. Since drafts must be rigorously excluded, it
is better, wherever possible, to set aside a room solely for the use of the
calorimetric outfit.

40. The adoption of such an installation will enable a calorific reading
of the gas to be made in a very short time, and will warrant the best of care
being taken of the calorimeter and of its accessories.

41. After installation and before undertaking investigations involving
experimental data, the above equipment should be inspected by the Chief
Inspector of Gas of the State Commission and have the approval of the
Commission.

DIRECTIONS FOR OPERATING A CALORIMETER.

1. On unpacking the Calorimeter, see that it is cleaned inside and out,
and free of packing material.

2. Study carefully the erecting directions and cuts and see that all parts
are included.

3. Handle the thermometers with the greatest of care.

4. The Calorimeter should be set up in a quiet, light and well-ventilated
room or cabinet, which is free from draughts and in which the temperature
can be maintained constant at not less than sixty degrees Fahrenheit. The
room should be provided with a sink and with a good supply of running
water. It is advisable to have a large shallow overhead covered tank, from
which the water supply can be taken. Should the tank capacity be small and
not hofd enough water for a prolonged series of readings, a small gas water
heater may be employed as already noted to bring the water to approximately
the room temperature. It is desirable to use water in the Calorimeter that is
clear and free from suspended matter, therefore, a filter should be installed
in the water supply line before it enters the overhead tank.

5. If only a single test is desired, gas may be taken from the house
piping, but if an average value is required, a small gas holder, or averaging
tank, should be used, and the gas flowing into the holder adjusted to a rate
of flow to just fill it in the time during which the sample is to be taken.
Care should be taken to have a short service to this holder in order that an
average sample of gas may be obtained, and if the sample be taken from a
line on which there is no considerable consumption, see that this line is
thoroughly purged before sampling. It is recommended that the gas be
metered at a pressure not to exceed 2 inches of water; if this is not obtain-
able, it is advisable to insert a holder or diaphragm governor in the supply
line to reduce the pressure to within this limit.

6. Set up the calorimeter so that the overflow and outlet water can be
easily led to the sink. Make water connections with rubber tubing, being
careful not to cramp the tubing. To avoid air currents caused by the move-
ment of the observer's body, set up the calorimeter so that the water supply
and waste may be easily adjusted and that all temperatures may be readily
observed. Lead the outlet water to a waste funnel supported a little above
the top of the copper or glass container used in collecting the water, so that
the water can be shifted from the funnel to the container and back without
spilling.

7. Set up the gas meter facing the observer and level it carefully. Then
adjust the water level of the meter, both inlet and outlet being open to the
air. To do this, remove the plug from the dry well, open the funnel cock
and disconnect the tubing on the outlet of the meter. With one finger over
the dry well turn on the gas a little and by removing and replacing the finger
see that there is no water in the dry well. If water be found therein it must
be blown out by gas pressure. Notice whether the water in the gauge glass
moves freely, as, if it does not, the meter is out of order. Now remove the
finger from the dry well and add or remove water (through the funnel or by
the cock under the gauge glass) until the lowest edge of the meniscus just
touches the scratch on the gauge glass, or is even with the fixed pointer.
Replace the plug in the dry well, close the funnel valv.e and connect the
governor. If the meter has been filled with fresh water the gas must be
allowed to burn at least two hours before making a test. When the water
in the meter is saturated with gas, twenty minutes should be sufficient.

8. Fill pressure regulator with water, then connect it to the calorimeter
burner. Metallic tubing is preferable, but when rubber tubing is used to
connect meter, pressure regulator and burner, connections should be as short
as possible, and should be saturated with the gas.

9. Turn on gas and allow it to burn for 5 or 10 minutes with the burner
on the table. Shut off gas at burner and watch hand on meter for leakage.
Be sure that all leaks are stopped before attempting to make a test. Start
: water running through the calorimeter at a rate of about three pounds per
minute. Then regulate the gas to flow at the rate of 4 to 7 feet an hour, as
may be found by experiment to give the highest result with the gas to be
tested, admitting enough air through the burner so that the flame shows a
faint luminous tip, then insert the burner at the proper height in the calori-
meter and observe again the condition of the flame to see that it is all right,
using a mirror.

10. The excess of air passing through the calorimeter is controlled some-
what by the position of the damper in the exhaust port, and the best results

'are obtained by having the excess air as low as possible and still maintaining
complete combustion of the gas. Such position has heretofore been found to
be about bnfe-fourth open with those calorimeters already investigated; care
must be exercised to determine this for each calorimeter.

11. Water should be regulated so that there is a difference. between the
inlet and outlet temperatures of about 15 degrees Fahrenheit. The temperature
of the inlet water should vary but little when an overhead tank is used and
the water maintained at room temperature. Be sure that both overflows are
running.

12. Before making the test the barometer, temperature of the gas at the
meter, temperature of room and temperature of exhaust products should be
recorded. It is desirable to have the temperature of the inlet water and
temperature of exhaust products as nearly as possible at room temperature,
in order to establish more nearly a thermal balance; the difference in these
temperatures should never exceed five degrees.

13. Next allow the gas to burn in the calorimeter until ,a thermal balance
is established, or until there is the least change in the inlet and outlet waters.

14. The test may now 'be started by shifting the outlet water from the
funnel to the container just as the large hand on the meter passes the zero
point. Readings are then made of inlet and outlet thermometers, making the
readings as rapidly as the observer is able to record them during the consump-
tion, preferably of 2/10 of a cubic foot of gas. At least ten readings should be
made of both inlet and outlet water temperatures. Water is again shifted
from the container to the waste funnel as the hand passes the zero point the
second time. Water is then weighed, or measured. The uncorrected heating
value per cubic foot is obtained by multiplying the difference of the averages
of inlet and outlet temperatures, by the number of pounds of water and
dividing by two-tenths. This quantity is divided by the correction factor for
barometer and temperature, obtainable from tables, to give the heating value
at 30 inches pressure and 60 degrees Fahrenheit. The weight or contents of
container should be obtained while the inside is wet. This may be done by
filling it with water, emptying and shaking for about five seconds in an inverted

. position. This will do away with any correction where several consecutive
tests are required with same container.

15. A second, and perhaps a third test is advisable, and these should
be made without disturbing the existing conditions, provided all readings are
within the above prescribed limits. In practice the operator should get con-
secutive results on the same holder of gas within ten (10) B. t. u's. Under
such conditions an average of the results may safely be taken,

Results as Obtained by Calculation.

16. The method of calculating the calorific value of the gas from the
observations indicated is very simple when all readings are made in English
units, as recommended, and entered in some form conveniently arranged. A
simple record sheet is illustrated in the American Gas Institute Proceedings,
Vol. III., 1908, page 320.

17. The averages of the inlet and outlet water temperatures are made
and any correction for thermometer error allowed for. The difference in
these averages should give the rise in temperature of the water. This rise in
temperature of the water is then multiplied by the number of pounds of water
passed through the calorimeter during the test. The product of these two is
then divided by the quantity of gas burned, either 0.1 or 0.2 of a cubic foot
as may be. This quotient will give the heating value of one cubic foot of gas
in B. t. u's. at the indicated temperature and barometric pressure. To correct
this to 60 F. and 30" pressure, divide by the "Correction Factor" for the
indicated temperature and pressure as obtained from some standard table, a
copy of which may be found opposite page 373 of the Proceedings of the
American Gas Institute, Vol. III., 1908. The final result will be corrected
heating value of the gas tested, in B. t. u's.

18. Expressing the above in a formula we have :

B. t. u's. per cubic foot =

WxT

W = Weight, in pounds, of water passed.
T the average difference in temperature, in degrees Fahrenheit,

between inlet and outlet water.
G = corrected volume of gas burned, in cubic feet.

Use of Computer.

19. The labor of making the calculations for determining the heating
value from observations of a calorimeter may be lessened by the use of a
heating value computer. The computer consists of a circular slide rule, with
divisions corresponding to the readings made on the calorimeter. This com-
puter gives the corrected heating value of a cubic foot of gas in B. t. u's,
having the barometer and temperature of the metered gas, and the difference
in temperature between the inlet and outlet water and the pounds of water
passed. This computer is designed to operate within the limits of from 300
to 800 B. t. u's. Should a gas of a lower or higher heating value be measured,
the computer can still be used by dividing or multiplying one or the other
of the factors in its computation. A cut of this computer may be found on
page 373, Vol. III., Proceedings of the American Gas Institute.

Care of Instruments .

20. The calorimeter, being a delicate and sensitive instrument, should
be very carefully cared for when not in use. If the instrument is set up
permanently, provision should be made that it be not disturbed by anybody
except the operator. If the instrument is not erected permanently, when
dismantled it should be carefully cleaned inside and out and the thermometers
removed and carefully packed in cotton.

21. It seems hardly necessary that instruction should be given for the
care of such an instrument, but certain precautions should be noted.

Precautions ' ' Don 'ts " .

22. Don't place lighted burner in calorimeter when water is not running
through the calorimeter.

Don't shut off water while gas is burning, but if water is accidentally
shut off, then shut off the gas quickly, to avoid breaking thermometers.

Don't move suddenly near instrument during test. Slight drafts thus
caused will vary outlet readings and vitiate test.

Don't fail to check daily the water level in the gas meter.

Don't forget to test meter and all connections daily for leakages.

Don't erect the calorimeter too close to any heating or lighting appliances,
where radiant heat might affect the readings.

Don't make the test with the inlet water temperature over 5 degrees above
or below the temperature of the room.

Don't fail to fill the overhead tank with water when through testing so
that it will be ready for the next test.

Note:

23. That an error of 1/10 F. in water temperature means an error of
about four B. t. u's. in the gas.

That an error of 1/100 of a pound of water when burning .2 of a cubic
foot of gas in the test means an error of about .9 B. t. u's. in the gas.

That an error of one degree in the temperature of the gas means an error
of about 1.8 B. t. u's.

That an error of 1/10 of an inch in Barometer reading means an error
of about 2 B. t. u's.

That when metering the gas, each additional inch of water pressure to
which the gas is subjected means an error of about 1.5 B. t. u's.

VI.

SUGGESTION OF SEVERAL TYPES OF CALORIMETERS SUITABLE

TO USE WHEN CHECKED BY THE PRIMARY

STANDARD ADOPTED.

The Committee believes that any calorimeter of the water heater type,
when fitted with the accessories as provided in the recommendations of the
Committee, that, when new, will test with the Primary Standard within two
per cent, would be suitable for commercial use by any company.

From the information available, the Junkers, the Improved Sargent, or
American Meter Company calorimeters are types of instruments which seem
to be available for immediate use by the Companies, but they must in each
case be equipped with the accessories as provided in the recommendations of
the Committee. Any instrument of the above mentioned types must pass the
prescribed test against the Primary Standard.

We believe that all makers of instruments of the water heater type
prescribed should be encouraged to place their instruments in use.

Dr. Arthur H. Elliott, Ph. D., of New York, and J. B. Klumpp, M. B., of
Philadelphia, met with the Committee; they entered into its discussions, aided
in the determinations and join in the conclusions of the Committee.

W. R. ADDICKS, Chairman,
JAMES C. DELONG,
CHAS. H. STONE.
February 25, 1910.

APPENDIX G.

REPRINT OF PLAN OF CALORIMETRIC INVESTIGATION AND
EXPLANATION OF TEST AND REPORT FORMS.

Tentatively adopted January 26, 1912, by the Joint Committee on Calori-
metry, representing the Public Service Commission and Gas Corporations in
the Second Public Service District, New York State.

INTRODUCTORY.

On May 6, 1910, this Committee adopted certain Calorimetric Rules,
Regulations and Specifications which were printed and a copy sent to each
gas company operating in the second public service district in New York State.
References made hereafter to "Calorimetric Rules, Regulations and Specifica-
tions" refer to this pamphlet. (See Appendix F, page 69.)

A number of companies at once purchased and installed instruments in
accordance with these specifications and started daily tests to determine the
calorific value of their gas for the assistance of the Committee in its
investigation.

As other companies are becoming interested in the investigation and are
deciding to participate, it has become necessary to devise a definite plan for
the investigation in order that the results obtained in different localities and
under different conditions may be analyzed intelligently, and correct con-
clusions drawn therefrom.

The Committee is making a very comprehensive study of this entire
subject and the plan formulated is therefore more elaborate than would be
the case if merely the calorific values, without reference to operating condi-
tions, were desired.

PLAN OF INVESTIGATION.

The plan formulated comprises:

1. The making of daily calorimetric tests and the recording daily of
certain works data. Form A is to be used for this purpose. (See page 93.)

2. The submitting to the Committee monthly of the results of the daily
tests and of monthly averages and totals of works data. Form B is to be used
for this purpose. (See page 94.)

3. The furnishing to the Committee of information regarding operating
conditions, and apparatus and methods in use. A map or sketch with an
accompanying letter of explanation and description is to be used for this
purpose. (See page 88.)

EXPLANATION OF TEST AND REPORT FORMS.
FORM A.

Apparatus in Use

1. Each piece of apparatus will be given a designating letter or number
(see page 89, paragraph 6f), and this letter or number may be used in noting
the apparatus in use each day.

Send Out

2. The maximum, minimum and average daily send out will be reported
monthly and the figures will be obtained from these daily entries.

Works Started (First Blast On) at
Works Shut Down (Last Run Off) ak
Duration Intermediate Shut Down
Total Works Operation

3. The maximum, minimum and average hours per day of works opera-
tion will be reported monthly and the figures will be obtained from these daily
entries.

Yield Per Lb. Coal-
Oil Per M.
Generator Fuel Per M.

4. In many instances it would be extremely difficult to determine these
figures with any degree of accuracy on a daily run and in such cases it need
not be attempted. On the other hand, if it is the practice to make these
calculations, the figures should be entered for comparison with the monthly
measurements. (See page 87, paragraph 5.)

Enricher Per 100 Lbs. Coal Carbonized

5. The unit of "100 Ibs. coal carbonized" has been adopted as a fair
basis for comparison.

The calculation should be made and the figures entered daily.

Kind of Enricher

6. If cannel coal is used, the grade of this coal should be given, OT if
oil, the kind of oil. The practice in regard to this subject should be explained
in considerable detail in the letter. (See page 89, paragraph 61.)

Duration of Charge

7. The duration of charge each day should be noted* so that the average
daily duration of charge for the month can be obtained.

NOTE The term "Corrected Gas" means that the quantity of gas, as measured by a
meter, has been corrected to a standard of volume as represented when measured at a
temperature of 60 Fahrenheit and a barometric pressure of 30 inches. The figure for
corrected gas is obtained by multiplying the volume of uncorrected gas by a factor
corresponding to the temperature and pressure at which the gas has been measured. A
table of "Correction Factors" is given in Appendix A. This is copied with the permission
of the author, from the table given in "Practical Testing of Gas and Gas Meters," by
C. H. Stone.

Mixed Gas
Coal Gas %, Water Gas %

8. This is self-explanatory.

Time of Test

9. A calorimetric test consists of one or more sets of readings taken
continuously. On the form, three columns are provided for three sets of
readings. For convenience they are headed Test 1, Test 2, Test 3. It should
be understood that the three sets of readings or "tests" taken together and
the results checked or averaged, constitute one test. The time of starting
this test should be given. (See Calorimetric Rules, Regulations and Specifica-
tions, page 74, paragraph 15.) Should two or more tests be made at different
hours of the same day, a separate sheet should be used for each test.

Barometer

10. Refer to Calorimetric Rules, Regulations and Specifications, page 76,
paragraph 29.

The mercury column barometer specified in the reference should have
an adjustable zero and a vernier for reading. If the participating company
is relying on barometric readings taken by the local weather bureau, the
reading taken at a time nearest to the time of test should be used. Otherwise
a reading should be taken as a part of the test.

Room Temperature

11. The temperature of the room at the time of the test should be stated
in degrees Fahrenheit.

Candle Power

12. Two spaces are provided for candle power so that in case two types
of burners are used the results obtained with each can be entered separately.
Spaces are also provided in which the type of burner should be noted directly
above the candle power obtained with it. The result of this one photometric
test only is to be noted on this form. The results of other photometric tests
made during the day will be disregarded so far as this investigation is
concerned. (See page 88, paragraph 14.)

Pressure
Meter Inlet Burner Inlet

13. These readings should be taken at the time of starting the test. (See
Calorimetric Rules, Regulations and Specifications, page 78, paragraph 5.)

Minimum Temperature to Which Gas Has
Been Subjected Before Test

14. This temperature may be obtained by the use of an hygrometer.
(See Proceedings American Gas Institute, Vol I., 1906, pages 601 and 602.)

Rate of Combustion Per Hour

15. See Calorimetric Rules, Regulations and Specifications, page 78,
paragraph 9,

Exhaust Temperature
Gas Temperature

16. See Calorimetric Rules, Regulations and Specifications, page 79,
paragraph 12.

Total Pressure Correction

17. This means that the water pressure, at meter outlet, in inches, is
calculated to inches of mercury and added to the barometric reading. It is
desired to correct the pressure of the gas to 30 inches of mercury, from the
combination of the barometric pressure and the inches of mercury calculated
from the water pressure at which the gas is burned.

Correction Factor

18. See Table, pages 90 and 91.

Uncorrected Gas Used in Test
Corrected Gas Used in Test

19. (See footnote, page 83.)

Weight, Water and Pail
Weight, Pail Empty
Weight, Water

20. See Calorimetric Rules, Regulations and Specifications, page 76,
paragraph 32 ; also page 79, paragraph 14.

Temperature of Water

21. Readings 1-20. See Calorimetric Rules, Regulations and Specifica-
tions, page 79, paragraph 14.

Average Temperature

22. The average temperatures will, of course, be obtained by adding all
the temperatures taken and dividing by the number of readings. Space is
provided for this calculation.

Thermometer Correction

23. This figure will be obtained from the calibration curve. See Calori-
metric Rules, Regulations and Specifications, page 75, paragraph 24; also
paragraphs 22, 23, 24 and 26.

Stem Correction

24. See page 92.

Corrected Average Temperature

25. This means the average temperature after the two corrections, ther-
mometer and stem, have been applied.

Rise in Temperature

26. The rise in temperature equals the corrected average outlet tempera-
ture minus the corrected average inlet temperature.

Calculation

27. The general formula is:

WxT

B. t. u. per cu. ft. =

Gxe

"W" Weight of water.
T Rise in temperature of water.
G Corrected gas used in test.

e = Efficiency of instruments in tenths of 1%. This figure is ob-
tained from the most recent comparison of the instrument
with the State standard.

In the blank formula as stated on the form, the numerator contains the
figure 1,000, so that the efficiency can be stated in whole numbers and decimals
thus avoided.

A computer may be used in making the calculation after the blank
formula has been filled out, but if this is done the correction for efficiency will
have to be made separately. (See Calorimetric Rules, Regulations and Specifi-
cations, page 80, paragraph 19.)

Average = B. t. u.

28. The results obtained with the different sets of readings should check
within 10 B. t. u's. Under such conditions the average of these results should
be obtained and this figure will be the one transferred to the Monthly
Summary, Form B. (See Calorimetric Rules, Regulations and Specifications,
page 79, paragraph 15.)

FORM B

Coal Gas Made

Carburetted Water Gas Made

Mixed Gas Made

1. This refers to the gas made during the calendar month. Whether
the figures are for uncorrected or corrected gas should always be indicated.

Daily Send Out, Maximum
Daily Send Out, Minimum
Daily Send Out, Average

2. These figures will be obtained from the entries for "Send Out" on
Form A.

Gas Enriched (Yes or No)
Gas Enriched (How)

3. This subject will be reported on fully in the letter. (See page 89,
paragraph 61) but should also be reported on briefly opposite these headings.
(See also "Average Enricher" and "Kind of Enricher," page 87, paragraphs
7 and 8.)

Hours Per Day Works Operation, Maximum
Hours Per Day Works Operation, Minimum
Hours Per Day Works Operation, Average

4. These figures will be obtained from the entries on Form A for "Total
Works Operation."

Average Yield Per Lb. Coal

Average Oil Per M.

Average Generator Fuel Per M.

5. These figures should be based on measurements of the coal, oil or
fuel on hand at the beginning and end of the month and not on the averages
of the entries on Form A. (See page 83, paragraph 4.)

Kind of Coal

6. The information desired is the commercial name of the coal used and
the mine from which it comes, if this is known.

Average Enricher Per 100 Lbs. Coal Carbonized

7. This figure will be the average of the daily entries on Form A.

Kind of Enricher

8. (See page 83, paragraph 6.)

Average Duration of Charge

9. This figure will be the average of the daily entries on Form A.

Kind of OH-
IO. This entry should give the "kind of oil," the district where the
oil is produced, if definitely known, and the specific gravity in degrees
Beaume, if this figure is available. If the companies have any distillation test
figures, they should be given.

Kind of Fuel

11. This entry should state whether coal or coke is used, and if the
latter, whether retort or oven.

Mixed Gas.
Coal Gas %, Water Gas %.

12. These figures will be the average of the corresponding entries on
Form A.

Calorific Values

13. As explained, page 84, paragraph 9, a test consists of one, two or
three sets of readings. This form provides space for only one test per day at
works and one at office, or some other location. If more tests are made, addi-
tional sheets should be used.

The figure to be entered will be taken from Form A, "Average B. t. u.,"
but the nearest whole number should be given and decimals eliminated.

Candle Power

14. The candle power figures to be entered here should be taken from
Form A, "Candle Power." The entry should not represent the average of all
photometric tests made during the day, but should be the candle power at
the time the calorimetric test is made. The candle power should be stated
with only one decimal.

Minimum Temperature Gas

15. As explained in the note on the form, this refers to the minimum
temperature to which the gas has been subjected before test. The figures
should be taken from the corresponding entries on Form A.

Temperature of Atmosphere

16. The maximum and minimum temperature of atmosphere should be
stated in degrees Fahrenheit. As no space is left for them on Form A they
may be entered daily on Form B.

Maximum
Minimum

17. Refers to figures in columns above.

Note

18. When calorimetric or photometric tests are made at both works and
office or some other location, the tests at the two places should be made
simultaneously.

MAP AND LETTER

1. A detailed map, or if this is not possible, a sketch, on paper 8^/2" x 14"
should be submitted.

2. This map or sketch should show the relative location of the works
and holders and should indicate the points at which the tests are made.

3. If these tests are made at a distance from the works, this distance
following the course taken by the gas should be accurately shown. Also if
any exposed bridges have been crossed, or if the line runs under water, these
points should be made clear.

4. Such map or sketch will be asked for but once unless changes are
made, and it should therefore contain information regarding all matters which
are liable to affect the results obtained in the tests.

5. The map or sketch should be accompanied by a letter, also on paper
' x 14", containing a general description of the apparatus and methods

employed.

6. Such letter should state :

(a) Kind of gas made.

(b) Manufacturing capacity of plant, giving figures for coal gas and
water gas separately.

(d) Gas holder capacity outlying.

(e) Holders housed or exposed.

(f ) List of generator apparatus with type and capacity of each piece
of apparatus. The different pieces of apparatus may be desig-
nated by a letter or number for future reference.

(g) Type and make of calorimeter,
(h) Type and make of photometer.

(i) Type of standard and burner used in photometer test.

(j) General description of the methods used and conditions under
which the tests are made. For example, such a description
might be that the calorific and candle power values are taken
at office located at works, that the gas is taken from inlet of
the street governor and has been in the storage holder, and that,
as this holder is exposed, the probabilities are that the gas has
been subjected to the extreme temperatures of the atmosphere.
Or, for another example, that the tests are made at the com-
pany's office, located a mile from the works, that the gas is
taken from the house piping or that it is taken from an indi-
vidual service; if this service is any way exposed to the tem-
perature of the atmosphere, it should be mentioned; that the
gas has passed over an exposed bridge as shown on- the map, etc.

(k) Concise statement of how the gas is stored and exposed before
it is delivered to the street mains.

(1) Concise statement of methods employed in enriching.

WM. McCLELLAN, Chairman.
A. H. ELLIOTT,
J. B. KLUMPP,
C. F. LEONARD,
C. H. STONE.
January 26, 1912.

13
02
02

P tt

P
H

O
H

o
o

t-t-OOOOCOC55O>OiOOO' li i

09 i

^ OOOOOOOOOOOOOOOOCOOOOOOia5OiO>OiOiO5O5O5OiOiC:C3OlG5O5O5OiOlO5

M i

ti !(M(M(M<MCOCOCOTt<TtiT^TtiinlO

OOOOQOOOQOOOGOOOOOOOOOOOCX30OOSOlO5OlOJO5OiO5O5C5O:O5a3OiO5OiOi

li 1 i 1 i I(M(M<MCOCOCOCC-*

Tj(Tt<lOlOkOiCO5t~t~t^OOOOOOOVi35OiOOO' li (rHi ICvlCgO-lOtCOCOfO

li li li-l(MC<lC<)COfOeO
OOOOCX3OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOSOiO5OO5O5OiOiO>O5OiO5CS

I OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO5O5O5O)OiOiOiOiC5O5O5C5

u5oooginoico?oO5cocoocooojeoioO5(Miocoi iTfiCDOsociTjHi^oco to-oo

COCO^T^-^lOlOkOCO^Ot^l^-t^t^OOOOOOOiOOSOOOOi-HiHrHCvltMCvlCvl
OOOOOOOOOOCXDOOOOOOOOOOOOOOOOOOOOQOOOOOOOO5O5O3O>O5O5O5OiOlO5O5

; OOOOCOOOOOOOOOOOaDOOOOOOCOOOOOOOOOOOOOOOCXDC5O5O3O5CiO5C5O5O5O5

; OOaDQOOOOOOOOOOOOOOOOOOOOOODOOOOOOOOCOQOODOOO5C5O5OJOiOSO5C5C5

loaoicOM^-^^-^iOlOlOWtOtOb-t-^b-tXDOOOOOJOSOSOSOOOr-lTHr- li i
~p OOOOOOOOOOOOOOOOOOOOOOCXDCOOOODOOOOOOOOQOOOOOOOOOO>O5O>ClOiO5O5

n iooacoooooocooDoooooooooooooooooocooooocoooooccoooocioiOJOiOiOi

OOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOOOCOOODOOODOOOOOOOOOOOiOJOlOiOS

OOOOOOOO05050505050505050505GIDXCOGIDCeGIOCCXGID(X)b>b-b-

C5 ^^ 'O ^^ ^5 C> ^2 ^^ O< ^^ G> ^5 C5 O 1 O C) ^5 C 5

Oi Oi Oi Oi Oi Oi O5 OS O^ OS OS ^5 ^5 O> O ^5 (O ^5 ^O ^5 ^^ ^^ ^2 ^^ ^5 ^^ ^^ O <O C> ^5 ^^ O 1 ^^ C* O O ^5

GOrHcOOOarH^I^OSC<J^t^OeOU^OOOCOCOOOi^CO^OOi^CC^OOi^C^<DOOOCOLOCO^
^t^t*b^GpQpOOGOOSOaO5OOOOrHrHrHrHC<]CvlC^C^COCOCCCO^r^TtHT^lO^

OCOlCOOi-H^COOSr^'^i^DOS'-H^COOSi (Tj^CDOSr-lTrtOOST (COCDCO'-^CO^O
OSOSOSOSOiOSOSOSOSOSOSOSOSOiOSOOOOOOOOOOOOOOOOOOOOOOO

COIOOOOCQIOCOC>COIQODC^CQLOOOOC<I
i^i^i^CvJC^C^C^COcOCCCO^^^-^lplO

rHCOlpOppCO^pQOi iCO^DOO
O5 OS

t- O

ososososososososososososososososososososo

ososososososososososososososos

ososososo^osososososososos

IO b* O CO

os o> os os
ososososososososososososososososososOi

OSOSOSOSOsOSOSOi

ososososo-ososds
osososososososos
osososososososos
os os os os os os os

OS OS OS OS OS

STEM CORRECTION

In general, all corrections are determined for total immersion, i. e., for
the condition where both bulb and stem of the thermometer are at the same
temperature. If, however, the stem is emergent into space, either hotter or
colder than the temperature of the bulb, a stem correction must be applied
to the observed reading.

This so-called stem correction may be considerable if the number of
degrees emergent and the difference of temperature between the bath and
the space above it are large. It may amount to more than 68 F. for measure-
ments made with a mercury thermometer at 752 F.

For the glass of which this thermometer is made the stem correction may
be computed from the following formula :

Stem correction=0.000088 x n (T 1).
n=number of degrees emergent from the bath.
T=temperature of bath,
t mean temperature of the emergent stem.

The mean temperature, t, may be approximately measured by means of a
small auxiliary thermometer suspended near the emergent stem, or by sur-
rounding the latter with a small water jacket and taking the temperature of
the water with the auxiliary thermometer, or, more accurately, in the way
suggested by Guillaume, by exposing an exactly similar stem and capillary
mercury thread beside the emergent stem, and thus measuring its mean
temperature.

This is also conveniently carried out with the "thread Thermometer"
(Fadenthermometer) of Mahlke, in which the expansion of the mercury in the
capillary tube (bulb) is measured on a still finer capillary stem.

Example

Suppose that the observed temperature was 85 and the thermometer was
immersed to the 32 mark on the scale, so that 53 of the mercury column
projected out into the air, and the mean temperature of the emergent column
was found to be 70 F., then

Stem correction^ .000088 x 53 (8570).

=r0.07

As the stem was at a lower temperature than the bulb, the thermometer
read too low, so that this correction must be added to the observed
reading to find the reading corresponding to total immersion, i. e.,
85.00 + 0.07 = 85.07 F.

This correction must be considered in addition to any correction shown
by the certificate accompanying the thermometer.

For further information in regard to this subject see "The Correction for
Emergent Stem of the Mercurial Thermometer," published by the U. S. Bureau
of Standards as Reprint No. 170.

RE JOINT COMMITTEE ON CALORIMETRY 2ND P. S. C. DIST. N. Y.

(NKI or COMPANY)

GAS MAKING RECORDS

Apparatus in use..
Send oul

. cu. h. uncorrected
.. M " . corrected

Work, itarted (Irt Wart on) at ; * m ~

p.m.

Worb ihul down (la* run of) at m '

p. m.

Duration intermediate shutdowns. .'..'..... hri.

Total worb operation :..;.......... hn.

COAL GAS

Yeild per Ib. coal uncorrected gas T cu. ft. Enricher per 100 Ibi. coal carbonized.,

corrected " " Kind of eoncner

Duration of charge -

WATER CAS
',,

Oil per M uncorrected gas gal. Generator fuel per M uncorrected gai..

" corrected gas " corrected gai

MIXED GAS

CALORIMETER TESTS

TIME OF TEST

BAROMETER ROOM TEMP.

TEMPERATURE OF WATER

' I'm

1
2
3
4
5
6
7
8
9
10
II
12
13
14
15
16

TEST 1

TEST 2

TEST 3

CANDLE POWER PRESSURE

Inlet Water

Outlet Water

Inlet Water

Outlet Water

Inlet Water

Outlet Watet

Meter Inlet Burner Inlet

C. P.

C. P. in. in.

Minimum Temperature to which CM hat been lubiected before Te*

TEST 1

TEST 2

TEST J '

Exhaust temperatu
Gas
Total piessuse corr
Correction factor...

cu h

cu. ft.

cu ft

. --

clion

Corrected (G)

cu. It,

cu. ft.

cu. h,

Weigh! water and
" pail empty

Ibs.

..Ibs.

(W)

TEST 1

x..lOOO= B. t.u.

17
18
19

TEST 3

x ...1000= B. l.u.

.-. AVERAGE = .-..B. t.u.

. Average
NOTE. For explanation of use of this fotm.see j
"Plan of Calorimetric Investigation Thermom
and Explanation of Test and Report j -
Forms," Joint Committee on Calori- ij
rnetry. Jan. 26. 1912. ! l o,,,,,^

i Rise in le

average tempera
mperarure (T)..

Sip*!..

..Approved..

The actual size of this Form is B l / 2 " x 11"

RE JOINT COMMITTEE ON CALORIMETRY 2ND P. S. C. DIST. N. Y.

SUMMARY MONTH OF

GAS MAKING RECORDS

Coal Gas Made

' , t Uncorrected
CU. ILJ Corrected

Daily Send Out, Maximum ..

, I Uncorrected
<=" n j Corrected

Carburetted Water Gas Made

, ( Uncorrected
(Corrected

Minimum . .

.. I Uncorrected

Mixed Gas Made

.. ( Unconected
(Corrected

4 Average

., ! Uncorrected
(Corrected

Gas enriched (yes or no)

Hours per day works operation, maximum

hours

'" " (how)

.- ,

ii .. ,,.' .;

average ,

Ave. yield per Ib. coal uncorrecteJ gas.

corrected gas

Kind of coal....

COAL GAS

cu. ft. Ave. enricher per 100 Ibs. coal cardonized Ibs.... gals.

Kind of enricher....

Ave. duration of charge

Ave. oil per M uncorrected gas.
corrected gas...
Kind of oil

WATER GAS
..gals. Ave. generator fuel per M uncorrected gas.

.. " corrected gas

Kind of fuel ...

RESULTS OF TESTS

DAY

AT WORKS

TEMP. OF ATMOSPHERE

Mm. Tit" Q*

B. T. . *

Readings

C .P.

ludhifs

M,. T.... G..'

B. T. u.

.,-lnp

C. P.

I,.**.

Mu.MUM

r 2

3 : '

'" 4

""*

'"'

"'.'.IZ

'" ': a

.pro

"""'12

V-

"""'16

. ie

'~"" a

" 17
18

:::::::

" 2O

- ; *

"""7"

" 22

* 23

*"..^^

"""'26

"""26

'. 31

"WSSSST

Avtrate

HZ"

(MOTE: For eiplanmion of UK of thi. form Ke "Plan ol Calorimetric Invertigatioi. and Explanation ol Tert and Report Form.," Joint Committee on Calorimetry. Jan. 26. 1912

Remarks.... " !

Sgned.. ...Approved

The actual size of this Form is 8 ft" x 11"

ERRATA

CONTENTS "Classification of Companies Making Tests" under Appendix A is given in Table I

not Table 2 16

Insert "Introductory Observations Relation to the Study of Appendix B" 19

PAGE 19 paragraph 2, second sentence, reading "The information derived by the test * '" should
read "The information derived from the test."

PAGE 19 paragraph 4, reading "It should be noted that a percentage variation from a standard by,
for example," etc. should read "It should be noted that a percentage variation from a
standard of, for example," etc.

PAGE 20 paragraph 1, reference in first line is to page 18.

PAGE 21 In connection with table, note that the present standard for mixed coal and carburetted water
gas is 18 candle power and for enriched coal gas 16 candle power.

PAGE 22 Diagram showing th variations in heating power, symbol at left of centre line of diagram
should be "O" instead of/'fc".

PACK 38 In diagram "Variations in heating power" same correction.

PAGE 40 paragraph 9, second sentence, reading "Possibly if all the tests were plotted and the values
weighed," etc., should read, "Possibly if all the tests were plotted and the values weighted."

PAGE 49 paragraph 44, reading "Plate III shows how the main is exposed," should read, "Drawing
on page 51 shows how main is exposed," etc. The words "(See Page 51)" at the end
of this paragraph should be omitted.

PAGE 52 The page opposite Page 52 with drawing should be numbered 53.

PAGE 56 paragraph 69, third sentence reading "No serious losses in heating value were found to take
place in pressures up to ten inches of water" should read, "No serious losses in heating
value were found to take place with pressures up to ten inches of water."

PAGE 57 paragraph 73, the symbol for carbon monoxide should be "CO" not "Co". H2 should
be H 2 and CH4 should be CH 4

PAGE 58 paragraph 76, last line of page reading "as a means of accuracy determining," etc., should
read, "as a means of accurately determining."

PAGE 59 paragraph 81, second line, the comma at the end of the line should be stricken out and a
comma inserted before the word "at".

PAGE 60 paragraph 83, first sentence reading "It will be observed that the efficiency of the mantle
burner was equally good with either 20 candle power carburetted water gas or with 14.38
candle power in enriched coal gas," should read, "It will be observed that the efficiency
of the mantle burner was equally good with either 20 candle power carburetted water gas
or with 14.38 candle power unenriched coal gas."

PAGE 61 The standard in Dallas, Texas, is 633 at 32 F, not 650.

The standard in Milwaukee, Wis. is 600 gross and not 635 gross.

PAGE 62 paragraph 10, last line, figure "51.6" should be "516."

PAGE 63 paragraph 2, in the fourth line there should be a semi-colon instead of a comma after the
word "graduates" and in the last line there should be a comma after the word "excluded".

PAGE 64 paragraph 6, the first line after the table, reading "In every case the corrections in Table
B," etc., should read, "In every case the corrections in part b of the table."

PAGE 67 paragraph 27, the word "later" should be inserted after the words "two year," in next to
the last line.

PAGE 67 Table The thre<e tests opposite Company No. 8 showing accuracies 99. 4, 99.4, 99.4, were
made September, 1912, February, 1912 and July, 1912, respectively.

PAGE 82 references to Calorimetric Rules, Regulations and Specifications througout Appendix G
should all be accompanied by a reference to Appendix F, the reprint of this pamphlet.

PAGE 83 title, opposite the words "Form A" should be a reference to page 93, where there is a
cut of this form.

PAGE 83 Foot-note, third from the last line, the reference to Appendix A should be to pages 90 and
91, making this sentence read "A table of correction factors is given on pages 90 and 91."

PAGE 86 Opposite words "Form B" should be a reference to page 94 where there is a cut of this
form.

THIS BOOK IS DUE ON THE LAST DATE
STAMPED BELOW

AN INITIAL FINE OF 25 CENTS

WILL BE ASSESSED FOR FAILURE TO RETURN
THIS BOOK ON THE DATE DUE. THE PENALTY
WILL INCREASE TO SO CENTS ON THE FOURTH
DAY AND TO $I.OO ON THE SEVENTH DAY
OVERDUE.

4)EC 7

LD 21-100m-12, '43 (8796s)

c; -7 "7 :

u I i -~>

THE UNIVERSITY OF CALIFORNIA LIBRARY