ARCHITECTURE OF MACHINERY 
 
 AN ESSAY 
 
 PROPRIETY OF FORM AND PROPORTION, 
 
 Intended as a Guide to assist the Student in the Drawhu 
 and Designing of Machinery . 
 
 BY SAMUEL CLEGG, JUN., C. E. 
 Mcnstfue, for tyt to of >tutrent antr 
 
 .EIGHTY -FOUR WOOD-CUTS AND PLATES. 
 Price 6s. 
 
\ 
 
 m* 
 
 
 UNIVERSITY OF CALIFORNIA 
 
 ANDREW 
 
 SMITH 
 
 HALLIDIE: 
 
 
- 
 
 7? 
 
ARCHITECTURE OF MACHINERY. 
 
 as 
 
BY THE SAME AUTHOR, 
 
 In one handsomely printed 4to. Volume, with very numerous Plates and Wood-cuts ex- 
 planatory of the whole System of Gas Manufacture, in extra cloth boards, Price 1. 8s. 
 
 A PRACTICAL TREATISE 
 
 MANUFACTURE AND DISTRIBUTION OF COAL GAS; 
 
 CONTAINING 
 
 Explanations of the Chemical Changes which take place during the destructive distillation of 
 Coal ; Working Drawings and Experiments upon different kinds of Retorts, in which the best 
 methods of treating the Coal are considered, with a view to make them practically useful to 
 the Gas Engineer ; Working Drawings and Explanations of Retort Houses, Chimneys, &c., 
 calculated for the reception of any number of Retorts ; Estimates and Examples. 
 
 Working Drawings and Explanations of the different Apparatus used in the Manufacture of 
 Coal Gas ; amongst which are the Condensers, Dry and Wet Lime Purifiers, Wash Vessel, 
 Hydraulic Valves, Station Meter, Gas-holders, Governor, and their Details. 
 
 Also, Rules and Formulae for the calculation of the discharges of Gas from mains of 
 different diameters and lengths, and under different pressures, with Tables of Reference. 
 
 Concluding with Remarks upon the Secondary Products, as Coke, Tar, Ammoniacal 
 Liquor, &c., and the Manufacture of the Carbonate and Muriate of Ammonia. 
 
of 
 
 AN ESSAY 
 
 PROPRIETY OF FORM AND PROPORTION, 
 
 WITH A VIEW TO ASSIST AND IMPROVE DESIGN. 
 
 BY SAMUEL CLEGG, JUN., C. E. 
 
 AUTHOR OF A PRACTICAL TREATISE ON COAL GAS. 
 
 ADMONERE VOLUIMTJS, NON MORDERE ; PRODESSE, NON 
 CONSITLERE MORBIS HOMINUM, NON OFFICERS. 
 
 ARCHITECTURAL LIBRARY, 59, HIGH HOLBORN. 
 
 MDCCCXLII. 
 
< 
 
 PRINTED BY W. HUGHES, 
 KING'S HEAD COURT, GOUGH SQUARE. 
 
TO 
 
 JOSEPH MILLER, ESQ., 
 
 CIVIL ENGINEER, 
 
 THIS ESSAY 
 
 18 
 
 RESPECTFULLY INSCRIBED. 
 
 104009 
 
LIST OF PLATES. 
 
 PLATE I. Mouldings and Pedestals . . ... . to face p. 8 
 
 II. Regular and Irregular Columns contrasted . H 
 
 III. Independent Engine Frame of 14 horses' power . . . 12 
 
 IV. Ditto . . . . ... . . . . ib. 
 
 V. Column from Parthenon, and an Imitation, contrasted . . 14 
 
 VI. ,, Temple at Psestum, do. . " ... . . ib, 
 
 VII. Main Beam for 80-horse and 45 -horse House Engines . . 41 
 
 VIII. Connecting Rods in contrast .:.... 49 
 
 WITH SEVENTY-SIX WOOD ENGRAVINGS. 
 
INTRODUCTION. 
 
 THE object of the following Essay is to inquire into the reasons of error, 
 and to endeavour to point out certain rules for design, by attention to 
 which the many irregularities in construction and/orm so constantly met 
 with in machinery are to be avoided; and to explain the correct prin- 
 ciples of " taste." That they are open to such an inquiry no one can 
 doubt, and it is earnestly hoped that those who peruse this Essay will 
 do so without partiality, or allowing any personal feeling to warp their 
 sounder reason. That the opinions and remarks contained herein are 
 open to criticism and animadversion, the Author cannot deny ; but, at 
 the same time, he cannot help expressing his conviction that the data 
 upon which his reasonings are founded are correct, and might, if 
 treated in a more able manner, be productive of the greatest benefit, 
 not only to the manufacturing engineer, but to all those connected with 
 machinery. 
 
 The daily increasing demand for machinery of every description, the 
 perfection at which workmanship or skill in manufacture has arrived, 
 and the facility with which every part is executed, render it of the 
 utmost importance that the contrivances which constitute the value of 
 those machines, and the workmanship which renders them capable of 
 being made so truly beautiful, should be reduced to such rules of con- 
 struction, that while they retain all their essential forms, they may 
 produce pleasing effects upon the eye of an observer ; and the best means 
 to produce such an effect are now attempted to be set forth. 
 
If 
 
 2 INTRODUCTION. 
 
 It must be remembered that the majority of those for whom machines 
 are constructed cannot enter into the merits of their internal action or 
 of their comparative performance; and therefore, not being willing to 
 yield the right of opinion, judge from " outside show," as they would 
 of a picture or statue, where the only aim is to charm the eye, or excite 
 pleasureable sensations in the mind, by the representation of Nature in 
 her more sublime effects. It is only on those who by their education 
 are qualified to judge of intrinsic merit, that either Nature or Art can 
 be said to produce correct impressions, since the mind is always influenced 
 by custom. It may therefore be inferred, that in many instances the 
 want of elegance in the contour of a machine is not only displeasing to 
 the spectator, but disadvantageous to the manufacturer. 
 
 The numberless contortions that have crept into the "patterns" of 
 some even of our best and most scientific engineers loudly demand that 
 some attempt should be made to point out the causes of the defects, 
 and means of obviating them. Meagre attempts at imitation of the 
 antique should be carefully avoided when the adjacent parts are not 
 capable of being made subservient or of corresponding form; but so 
 seldom can this be done that the propriety of such an arrangement 
 becomes the exception, not the rule. The leading features are often 
 twisted and turned into distortions so unsightly, that it would appear 
 as if the designer wished to show his contempt for regularity of forms 
 " which, by the common consent of ages, are esteemed beautiful," by 
 clothing his ingenuity and talent in ugliness. First impressions are 
 always the most lasting, and it is evidently prudent that they should 
 be pleasing. 
 
 Manufacturers are often bound down to work at low contract prices, 
 and an excuse often occurs, " a handsome thing cannot be made for such 
 a sum ; " but this is entirely a mistake. It does not cost one farthing 
 more to construct an elegant figure, having beauty without an attempt 
 at grandeur, than it does to create a deformity, or to make a machine 
 in such a manner as to raise disgust by its pretensions at display. If 
 
INTRODUCTION. 
 
 it be a casting, the first and principal expense is the " pattern ;" and since 
 this, once done, will serve for many machines, is it not advisable that 
 great care should be taken in its original design ? Surely the perpetuation 
 of a good thing is preferable to the multiplication of ugliness. In 
 wrought iron work this argument will not hold; but within certain limits 
 it is always as cheap, nay, frequently cheaper, to work to a perfect as 
 to an imperfect shape, more especially if it can be reduced, to turning 
 in the lathe or fitting by templates. Nor is it only in the first for- 
 mation that the cost is reduced ; for if the work be set out by scientific 
 rule, there is no necessity for change either in shape or disposition, 
 and the same pattern or template will therefore serve as an universal 
 model. 
 
 The beauty of a piece of mechanism will never depend upon the 
 quantity of "bright work;" a little, to increase the effect of light and 
 shade, and confined to the more prominent parts, is advisable in well 
 " got up " work ; but to overdo it is far worse than leaving it entirely 
 black. A frame, a pedestal, a wrought iron rod or cross head, if out 
 of proportion or possessing superfluous dimension, will look ill, whatever 
 be its polish or material. Deformity decked out in finery is hideous, 
 while that less obtrusive may be passed by. " Pretend to nothing but 
 what you are" will apply to man's handy work as well as to man himself, 
 and to attempt to give beauty by polish, when the form is unsightly, must 
 always be pretence. 
 
 Another very mistaken notion is that of hiding the essential parts of 
 a machine by adding useless ornament. Mr. Pugin, in his work on 
 * The True Principles of Christian Architecture,' complains loudly, and 
 on good grounds, that architects "construct ornaments" instead " of orna- 
 menting construction," and the same holds good with engineers. 
 
 Every part and detail of a machine should have a distinct meaning 
 with reference to the duty they have to perform ; and if those parts are 
 made in such a manner that they have strength in the proper place, 
 stability without unnecessary weight, and simplicity of form without 
 
 
INTRODUCTION. 
 
 meanness, they require no other ornament, because their own proportions 
 will constitute their greatest. 
 
 If we had not the means of making our material in suitable forms 
 without running to great expense, excrescences, distortions, and un- 
 meaning incumbrances would be more excusable; but we have the 
 means. Every facility which the ingenuity of man can devise in the shape 
 of tools is given to the manufacturer; and let him argue as he will, 
 he can bring forward no sound reason why the mechanical contrivances 
 which have gone so far to raise us in the scale of nations should not 
 be rendered as pleasing to the eye by their elegance, as they are useful 
 by the perfection to which they have enabled us to bring our fabrics. 
 
 K 
 
'3/i^'< 
 THE ^ 
 
 VV UNIVERSITY 
 V 
 
 .., 
 
 ARCHITECTURE OF MACHINERY. 
 
 THE first consideration in the design for a steam engine (and to these 
 machines the observations are confined) is the quantity of work each 
 part has to perform : in other words, what weight it has to resist ; in 
 what direction the force or forces are applied with reference to its cross 
 section and points of support ; and the times of impact or velocity. 
 
 The calculations necessary to obtain these must not be confined to 
 theory alone, neither should they be entirely deduced from the "rule 
 of thumb " conclusions, which practice without the guidance of mathe- 
 matical knowledge would arrive, or rather, jump at ; because, by the first, 
 the parts would be liable to insecurity from the sudden jerks or changes 
 in velocity always occurring in machinery, but which cannot be expressed 
 in formulae ; by the second, it is more than probable that weight would 
 be added in unnecessary places from the determination to avoid fracture, 
 and the want of knowledge respecting the true figure. 
 
 The figures best adapted for resistance in any known direction have 
 been fully demonstrated, and the student is referred to the works of 
 Tredgold or Barlow for information upon these heads, books well 
 adapted to their purpose, and possessing the rare qualities of theory and 
 practice combined to a valuable extent. 
 
 As general rules, to which there are but few exceptions, the following 
 may be given : 
 
 For transverse strains the figure must have a parabolic section opposed 
 
ARCHITECTURE OF MACHINERY. 
 
 Fig. 1. 
 
 Fig. 2. 
 
 to the force, or " some simple figure to include the parabolic form." 
 Fig. 1. 
 
 The vertex must be at the point 
 where the force is applied. Fig. 2. 
 
 Where the strain is first in one 
 direction and then in an opposite 
 one, the curve must be on both 
 sides. Fig. 3. 
 
 Against direct strain a straight 
 line must be opposed; and sta- 
 bility and non-vibration insured 
 by the intersection of planes, so 
 that the section will be a cross or a 
 portion of one, or some form that 
 will include that figure. Fig. 4. 
 
 And for tortion the section must 
 
 Fig. 3. 
 
 Fig. 4. 
 
 be a circle. 
 
 All parts of the machine, in 
 whatever situation and under what- 
 ever circumstances they may be placed, may be shaped in these simple 
 figures; and if the calculations for their dimensions are made from 
 correct data, the form of greatest beauty is arrived at directly. The 
 most conspicuous parts should have boldness of outline, and as much 
 relief as possible, consistent with the purposes to which they are ap- 
 plied. 
 
 A bold outline being gained, it will remain to relieve the separate parts, 
 and it may, perhaps, be said that relief should be the term used to 
 express ornament in machinery; but if it be considered essential to use 
 ornament, the stationary portions alone should be thus embellished. 
 
 For the purposes of relief, the eight regular architectural mouldings 
 may be used, which are, the Fillet, fig. 5, introduced principally to 
 separate the members, or relieve a profile : in the latter case, as in fig. 6, 
 
ARCHITECTURE OF MACHINERY. 
 
 7 
 
 the under side should be joined to the face of the die or other surface 
 
 Fig. 5. 
 
 Fig. 10. 
 
 Fig. 6. 
 
 Fig. 11. 
 
 Fig. 7- 
 
 Fig. 12. 
 
 Fig. 8. 
 
 Fig. 9. 
 
 by a hollow : its projection is equal 
 to its height. 1 
 
 The Scotia, fig. 7, is also used to 
 separate members, should always be 
 placed below the eye of an observer, 
 and never upon a cornice or mould- 
 ing : its projection is equal to one- 
 third its height. 
 
 The Ovolo, fig. 8, on the contrary, 
 should never be employed but with 
 upper mouldings : its projection is 
 equal to its height. In this example it is composed of 
 the quadrant of a circle ; in the Grecian character it may 
 be the quadrant of an ellipsis. 
 
 The Torus, fig. 9, forms a portion of base mouldings, 
 is employed on circular figures, and is usually separated 
 from other members by fillets : it projects half its diameter from the 
 fillets. 
 
 The Talon or Ogee, fig. 10, is used in upper or supporting mouldings : 
 its projection is equal to its height. 
 
 The Cyma Recta, fig. 11, according to some, should only be used 
 in " crowning members ; " though in many specimens of the antique it 
 is found in base mouldings : its projection is equal to its height. 
 
 The Cavetto, fig. 12, " must not be seen in circular bases or capitals;" 
 that is to say, it is only valuable where angles occur: its projection 
 is equal to hal its height. 
 
 There are various manners of describing the contour or outline of 
 mouldings; the simplest, however, and the best, is to form them of 
 quadrants of circles, as in the annexed designs, by which means the 
 
 
 1 When it has a less projection, it is sometimes called a ' Plat-band.' 
 
ARCHITECTURE OF MACHINERY. 
 
 different depressions and swellings will be more strongly marked, the 
 transitions be made without any angle, and the projections be agreeable to 
 the doctrine of Vitruvius, and the practice of the ancients. 
 
 In a late work on architecture there is the following note, which agrees 
 with some observations before made. " All sense in the application of 
 appropriate forms in mouldings seems now extinct, and Palladio set at 
 defiance. He who can in the present day produce the newest and most 
 extraordinary moulding in profiling an order is the greatest genius." 
 
 It is not advisable to give the same projection for a cornice, or other 
 system of mouldings in cast iron, as in stone, for many reasons, but 
 there is no cause for deformity and burlesque. If the general style of 
 the engine will not admit of architectural contour, lay it aside entirely, 
 and be content with relieving the mechanical contour. 
 
 In Plate I. will be found some examples of mouldings that have been 
 executed in metal; those which are to be avoided were taken from 
 similar situations to those where the upper combinations were found, 
 and are fairly contrasted : the eye will be the best judge of their respective 
 merits. 
 
 The three pedestals were also taken from similar works, and were all 
 of nearly the same height. No. 3 is meagre, and totally devoid of that 
 appearance of strength which a foundation ought to exhibit, and which 
 is possessed by the other two. 
 
 Pedestals cannot be dwelt upon without notice being taken of the 
 very general error of breaking the lines of the base and cornice used 
 to relieve the tanks of " portable engines," in order to give a projection 
 to each of the six columns, and to form a separate pedestal to each ; 
 they have the appearance of figures upon stilts, requiring effort to retain 
 themselves in a vertical position; but when in addition to this they are 
 elaborately panelled or stuck over with laurel wreaths, they become 
 very miserable failures in the attempt at something elegant. A cornice 
 on a pedestal or tank is necessary for stiffness; a base is also useful to 
 convey an idea of stability. 
 
Jteyular Mouldings . executed in Cast 
 
 Mouldings to be avoided. 
 
 -Regular Pedestals, executed irv Cast Iron. 
 
 TUSCAN. 
 
 DORIC . 
 
 .Win 
 
ARCHITECTURE OF MACHINERY. 
 
 Cast iron being a material of so much greater strength than stone, such 
 high relief is not essential ; but this must not lead to meanness of outline. 
 
 From the best works that have been executed, the proportions may be 
 taken as follow : divide the pedestal into ten parts, fig. 13 ; give one part 
 
 Fig. 13. 
 
 Fig. 14. 
 
 to the cornice, and two to the base ; let the projection of the cornice be 
 equal to its height, and that of the base to one-third thereof: or divide 
 the base into three parts; give one-third to the mouldings, make the 
 projections of the upper members equal to their height, and the plinth 
 will drop from beyond the lower fillet. The pedestal may be divided 
 into twelve parts, without being poor; but the first proportion is con- 
 sidered the best. 
 
 Fig. 14 is from the condenser front of a large marine engine, and is in 
 tolerable proportion ; the base mouldings are dispensed with, because the 
 
10 
 
 ARCHITECTURE OF MACHINERY. 
 
 beam will hide the lower portion: the panel is introduced as a relief 
 against the large flat surface of the die. Fig. 15 is both less correct and 
 
 Fig. 15. 
 
 Fig. 16. 
 
 less pleasing in its effect ; its boldness is destroyed by projections which 
 are useless, and the mouldings are not well relieved ; still it is not so 
 unpleasantly mean as that shown in fig. 16, which is cut up by ill-shaped 
 panels and pilasters, and distorted by the line of the cornice being made 
 to follow that of the framing to which it is bolted. The lower parts 
 being hidden by the beam, much work might have been dispensed with 
 in the pattern. 
 
 When any superincumbent weight has to be supported, it is as usual 
 to introduce an architectural column in machinery as it is in a stone 
 building. In the latter the strictest and most severe attention is paid 
 to the relative form and proportion of the parts, that the same feature 
 may be carried out from the base to the coping stone; whilst in the 
 
CONTRASTED D OHIO COLUMNS. 
 
 -Jiiililiiu 
 
 M I) K H A . 
 
 ^ f 
 
 T . _J 
 
 'v. J 
 
 v- j- 
 
 
 
 
 
 
 
 
 1 
 
 v f 
 
 7 
 
 - 
 
 
 
 
 
 C 3 
 
 1 
 
 IRRE OULAR. 
 
 1 V- 
 
 / 
 
 V J \_ 
 
 
 ijliJlij 
 
 <; j 
 
 \ 
 
 
 \ ( 
 
 \. 
 
 ? 
 
 T r 
 
 r-^ ~~l , ^ 
 
 i . * 
 
 > 
 
 1 1 
 
 Vn 
 
 7 
 > 
 
 i ^ 
 
 c 
 
 
 
 ^ ^=^ 
 
 W.A. B 
 
ARCHITECTURE OF MACHINERY. 11 
 
 former, with few exceptions, a noble disregard is paid to the doctrines of 
 Vitruvius, and the practice of Palladio and Inigo Jones. The shaft of the 
 column is made sufficiently long to reach a desired point; while its 
 diameter remains unaltered, appearing to have no relation whatever to 
 the composition of the order. The Doric of the Greeks and Romans is 
 used indiscriminately, placed often one above the other without inter- 
 vening entablature, bases added or taken away, and the mouldings cut 
 and carved in a manner that to any one but those who propagate them 
 must appear monstrous. These absurdities exist, but there is not any 
 reason why they should remain. 
 
 There are many instances in which architectural columns may be 
 employed in machinery with effect and propriety, and the most appro- 
 priate use to which they can be put, is to support the entablatures upon 
 which the pedestals of any main feature are secured, as the beam of a 
 steam engine for example ; and for the reasons presently to be stated, 
 may be made to deviate to a considerable extent from the proportions of 
 the antique. It is a generally received rule in architecture, " that no 
 support should be burdened with a greater quantity of matter than itself 
 contains ;" or, in other words, that the weight placed on a column should 
 not more than equal its own weight. 
 
 Now, since the quantity of matter in a cast iron entablature is so 
 much less than in one of stone, having the same strength, may it not be 
 allowable for the same reason, to reduce the diameter of the columns 
 supporting that entablature? When the material is known to be metal, 
 and the surrounding objects are kept in subservience, we may deviate 
 from the antique quite as much as that given in Plate II., in which the 
 modern column, executed in cast iron, was employed for the purpose 
 previously named. 
 
 The antique Roman Doric is there divided from the top of the en- 
 tablature to the base into five portions, one of which is given to the 
 entablature. In the modern example the space is divided into six parts, 
 which may be adopted as the standard. 
 # 
 
12 ARCHITECTURE OF MACHINERY. 
 
 The two columns noted as irregular in Plate II. are taken from engines 
 of the same kind as that of which the second example formed part. The 
 one with the triglyphs can hardly be called an imitation of any order, 
 it is so vilely mis-shapen : its fellow is evidently an attempt at Roman 
 Doric ; but its members are so distorted, and its moulding and details so 
 deformed, that instead of adding beauty to the machine, it excites only 
 ludicrous ideas in the mind of the spectator. 
 
 These two columns are examples that may be found scattered very 
 generally throughout the works of our engine -makers who are not 
 engineers ; but that marked modern is the work of a man of judgment, 
 and may be copied with propriety. 
 
 The cost of ill-designed "patterns" in the first instance, and the 
 castings afterwards, differs little, if any, from that of the more perfect 
 model ; it is therefore impossible to assign any reason for the use of such 
 bad proportions, and assuredly there is no excuse. 
 
 Plate III. is an elevation of part of an " independent engine," consisting 
 of a tank and upper frame supported by six slender columns, " the 
 usual bedstead-looking machine," which does not even possess the ad- 
 vantage of cheapness to render it excusable that the same style should be 
 continued : the only convenience of the frame running entirely round 
 the top is, that the end of the radius rods of the parallel motion can be 
 fixed upon it without a support being erected expressly for that purpose. 
 But the " overhead" parallel motion, which is the simplest kind, re- 
 quires only a slight bracket, more easily fitted than the former, even with 
 its advantage of having an entablature. 
 
 Plate IV. also shows an elevation of a portable engine of the same 
 power as the former example (viz. 14 horses). These drawings are both 
 to a scale of ^ an inch to a foot ; the dimensions can therefore be taken, 
 and the relative weights of each be arrived at with tolerable accuracy. 
 The thickness of metal of the small columns is f ths of an inch,, that 
 of the larger fths of an inch, and its entablature f ths, with a piece 
 of oak timber 2 inches thick fitted in at the bottom ; both the tanks 
 
SIX COLUMN INDEPENDENT ENGINE OF 14 HORSE POWEU. 
 
 Plate .J. 
 
 Scale % In. to a, Foot. 
 
 John. Wea2e.Jr<-/iif,-<-tiit;>/ f.i/>r.<i-i ..'>:>, ffiffh Jfofborn. 
 
DOUBLE COLUMN INDEPENDENT ENGINE OF 14 HORSE POWER 
 
 Plate. 4. 
 
 In. to (L foot . 
 
 John 
 
 al Lil<rar\'.5{>.Ifi<i1> H?lk"rn . 
 
ARCHITECTURE OF MACHINERY. 
 
 13 
 
 are |ths thick. The stability of these engines is 
 equal. 
 
 The simplicity of the Tuscan mouldings ren- 
 ders this order more applicable to the uses of 
 the mechanist than either the Grecian or Roman 
 Doric ; but the height of its shaft being less in 
 proportion than that of the latter order, it can- 
 not be made in cast iron, except in rare in- 
 stances, with any degree of accuracy ; or rather, 
 its correct proportions, as given by the ancients, 
 must be deviated from to such an extent that its 
 character would be lost. The Roman Doric is 
 therefore given in Plate II. in preference to the 
 more simple order; although, as before stated, 
 its dimensions must also frequently be made at 
 variance with architectural rules. 
 
 Grecian Doric is frequently met with in the 
 erections of the engineer ; and in no instance, 
 nor under any circumstances, is it introduced 
 with even an approach to propriety. The rules 
 for this order are more severe than those of any 
 other, and they cannot be deviated from without 
 entirely changing its character. 
 
 It is placed sometimes singly, to support the 
 " plummer blocks " for the beam of a steam 
 engine, as in fig. 17 ; and although the pro- 
 portions of this example are in accordance with 
 those given by the Athenians, it looks strange, 
 and out of place : there is no authority for 
 single columns of this order having been found, 
 the beauty of which depends upon the general 
 assemblage of many features, rather than the 
 
14 
 
 ARCHITECTURE OF MACHINERY. 
 
 contour of detached portions. It is perhaps 
 hardly fair, whilst looking at machinery, to 
 think of architectural arrangement; but when 
 we are so strongly reminded of it by the erection 
 before us, it is difficult to forget the associations 
 to which the image gives rise. 
 
 The column from which fig. 18 is drawn is a 
 model from those of the temple at Paestum ; is 
 more fanciful than the other ; contains less metal, 
 with the same stability ; and gives a larger base 
 to the " block." For these reasons it is prefer- 
 able to the former, although the same general ob- 
 jections exist. [Both examples are from 30-horse 
 high-pressure engines.] (Scale f inch.) 
 
 Grecian Doric in its more perfect form being 
 objectionable, what must be said of those dis- 
 torted imitations of this order which are some- 
 times to be found ; when not only the pro- 
 portions of the shaft are disregarded, but the 
 contour of the mouldings disfigured under the 
 hand of the draughtsman or pattern-maker, as 
 in Plate V., where, contrasted with a shaft from 
 the Parthenon at Athens, is one from a marine 
 steam engine. The comparison is sufficiently 
 severe, though made without comment. 
 
 In Plate VI. we may again see exhibited the 
 love of imitation and the unsuccessful attempt of 
 the imitator. 
 
 The following rules are deduced from a large 
 collection of examples accumulated during several 
 years. 
 
 Fig. 18. 
 
 mm 111 
 
CONTRASTED GREEK COLUMNS. 
 
 Platt.5. 
 
 Parthenon . Athens. 
 
 JMarin& JSngine,. 
 
 John WkaJ.e, . slrctutectural Libran'. 3ff. ffu/h ffoUorn, . 
 
CONTRASTED GREEK COLUMNS 
 
 Temple at Fees turn. 
 
 Plate. 
 
 House 
 
 John Wfafe. Architectural Library, 5ff.Jfiffh HoUr&r 
 
 It: ./ 
 
ARCHITECTURE OF MACHINERY. 
 
 15 
 
 Fig. 19. 
 
 1 . A column must never be fixed in any position but the vertical ; and 
 unnecessary as this remark may appear, it is not made without 
 cause. 
 
 2. A column should never have a less diameter than ^th the height of 
 its shaft. If essential that the support be of less strength than this 
 rule will give, depart from the regular orders entirely, and employ 
 some pleasing curve in the upper member. Fig. 19. 
 
 3. An entablature should be equal in depth to 
 between Jth and jih of the space occupied by 
 it and the column. 
 
 4. A pedestal is ^rd of this space, or nearly so. 
 
 5. A cornice must have a projection equal to its 
 height. 
 
 6. Bases which include a plinth and mouldings 
 above must be twice the depth of the cornice, 
 and project ^rd that depth. 
 
 7. Avoid disturbing the outline and direction of 
 principal mouldings by small breaks. Use 
 similar mouldings throughout the work. 
 
 8. Triglyphs should be omitted generally, or 
 except in instances where columns are the 
 correct distance apart, which seldom happens 
 in machinery, because they must be placed, 
 
 c 
 
 not with reference to themselves, but to the parts of the machine 
 they carry. 
 
 9. Never make a moveable part to resemble any thing we are accus- 
 tomed to see stationary. 
 
 Beyond these limits we can borrow no rules from Architecture that 
 will not lead us into error and absurdity : all we can do with " framings," 
 beams, and cast iron work, which partake of no uniform figure, is to 
 design them with judgment, good taste, and proper regard to those laws 
 
16 ARCHITECTURE OF MACHINERY. 
 
 which regulate the actions of forces. " Perfect proportion," if considered 
 only with regard to the relations between the different objects in a com- 
 position, and as it merely relates to the pleasure of the sight, seems to 
 consist in this, that those parts which are either principal or essential 
 should be contrived to catch the eye successively from the most con- 
 siderable to the least, according to their degrees of importance in the 
 composition, and impress their images on the mind before it is affected 
 by any of the subservient members ; yet, that these should be so con- 
 ditioned as not to be entirely absorbed, but be capable of raising distinct 
 ideas likewise, and such as may be adequate to the purposes for which 
 these parts are designed. The different figures and situations of the 
 parts may in some degree contribute towards this effect; for simple 
 forms will operate more speedily than those that are complicated, and 
 such as project will be sooner perceived than such as are more retired ; 
 but dimension seems to be the predominant quality, or that which acts 
 most powerfully on the sense ; and this, it is apprehended, can only be 
 discovered by experience, at least to any degree of accuracy. When, 
 therefore, a number of parts, arranged in a particular manner, and under 
 particular dimensions, excite in the generality of judicious spectators 
 pleasing sensations, it will be prudent on every occasion, where the 
 same circumstances subsist, to observe exactly the same arrangement and 
 proportions, notwithstanding they may in themselves appear irregular 
 and unconnected. 
 
 It will perhaps be impossible to give an exact definition of the word 
 Taste, when employed in the construction of machinery ; and it may be 
 judged an useless, if not an absurd undertaking, to lay down rules of 
 caprice, and legislate for whims and fancies. By the word Taste is 
 meant that proper and well-directed exercise of the judgment by which 
 a man acquires facility in detecting impropriety of form, and distin- 
 guishing with readiness the masterpieces of great hands from the per- 
 formances of vulgar artists. 
 
 In considering any complex machine, he will examine every distinct 
 
ARCHITECTURE OF MACHINERY. 17 
 
 portion of the composition one by one, and reduce every thing to the 
 utmost simplicity : he will deliberate upon the uses of each part, on the 
 forces which will influence them, and on the best means of effectually 
 resisting those forces : he will pass nothing by as insignificant, and 
 condemn nothing without previous reasoning upon its excellences or 
 defects. The capability of thus forming a correct judgment is very much 
 accidental, as it depends upon experience, observation, and a knowledge 
 of practical mathematics. 
 
 There may be differences in taste ; for the same causes do not operate 
 with equal effect on the minds of all. But the principles of taste do not 
 differ, because they operate by principles in nature, and which are not 
 derived from any particular habits or advantages. Prejudice derived 
 from a defective education may give a colour to the mind, through which 
 medium deformities may seem beautiful ; from a long and close attention 
 to a certain object the imagination may be cramped and confined in such 
 a manner that nothing pleases it beyond the narrow limits to which 
 custom has bound it. This may influence the degree of taste, but it does 
 not affect the principle. It is upon these principles of taste that the ideas 
 of perfect beauty are founded, whether in nature or in works of art ; and 
 it will be 'found that beauty of proportion in machinery is sequent upon 
 the use of those figures which possess the greatest strength with the least 
 possible matter. A little attention will convince us that this must of 
 necessity be the case. 
 
 It is painful to contemplate a huge mass of matter sustained upon such 
 slender supports that they bend and totter under it; or to see a feeble 
 stop opposed to overwhelming force. These are extreme cases ; the sen- 
 sations they produce are extreme likewise, and are apparent to the most 
 unpractised eye. A different feeling is generated when we see prodigious 
 strength opposed to yielding weakness, or gigantic props beneath a light 
 superstructure ; and it is frequently as painful to be thus excited to 
 ridicule as to fear. As these absurdities vanish and approach more nearly 
 to propriety of proportion, the difficulty of distinguishing the faults 
 
 D 
 
18 ARCHITECTURE OF MACHINERY. 
 
 increases in the same ratio. If the excess or diminution be not glaring, it 
 is only a practised eye can discover inaccuracy, but it is for such eyes 
 that we must work ; and while we remember that beauty may be seen 
 and appreciated by the commonalty, let us not forget that the silent 
 approval of those whom we consider masters is more estimable than the 
 loudest praise of the vulgar. The latter will be guided by the eye alone, 
 but, with the former, judgment and reason will unite in the decision. 
 
 The part of the engine which first presents itself for consideration is 
 the framing, and the laws of statics will be found to combine with those 
 of taste in the production of a good design to so great an extent that they 
 will appear dependent upon each other. 
 
 The effect of forces upon framings, as they are usually applied, is to 
 compress them in the direction of their length. Therefore their parts 
 should be disposed in straight lines, so arranged that, whilst transverse 
 strain is avoided, each separate single force, or the resultant of any 
 number of forces, is opposed by some portion placed in the same direction 
 with it. 
 
 The dimensions necessary successfully to resist any given stress are 
 arrived at theoretically by very simple means, and when this force is 
 constant, and uninfluenced by jerks, the calculated result may be relied 
 upon without fear of failure. In machinery the dynamical effects of 
 forces are constantly varying, and it requires much well-directed thought 
 to enable us correctly to determine the quantity of variation. 
 
 The greatest power the engine is capable of exerting, the momentum 
 the parts put in motion acquire, the leverage at which the opposing 
 forces act, and the impediments or checks to which the machinery is 
 liable, are the chief points on which to found the data of our calculations. 
 But the re-action occasioned by sudden opposition to the impulsive power 
 is often greater than that power, and the parts which are secured from 
 yielding would be liable to fracture were the dimensions of them fur- 
 nished by theory alone. 
 
 Beauty of form and proportion are therefore both evidently dependent 
 
ARCHITECTURE OF MACHINERY. 
 
 19 
 
 Fig. A. 
 
 upon the proper adaptation of those rules which are gained by experience 
 and a well-directed judgment. 
 
 The principal thing to be guarded against in the design of a framing 
 (or any feature possessing considerable extent of surface) is sudden 
 variation in the direction of a member. There is no particular line 
 which is always found the most beautiful ; if there were, it would be easy 
 to give a rule for curves or form ; but we seldom see natural objects 
 entirely angular, and those most pleasing to the eye swell or sink so 
 gradually that it is often difficult to ascertain the point of their beginning 
 or end. A reason for this may be that a sharp angle causes a momentary 
 surprise from the sudden change which is exerted upon the eye, and 
 is therefore displeasing. Fig. A. 
 
 The direction of a varying dimension must be 
 guided by statical laws, and the evidence of its 
 propriety be left to the judgment. 
 
 There are exceptions to this as to most other 
 general rules, for boldness of outline (a very 
 essential requisite in composition) is often pro- 
 duced by sudden projection, and this can 
 only be made with an angle (fig. B) ; still 
 this angle may be formed in a graceful 
 manner. 
 
 The degree of its inclination will be 
 regulated by the direction of some force 
 
 which may be here presumed to operate against it ; and as an exception 
 to the rule, it may be said, that if this direction be evidently arranged 
 to oppose some strain, it will be correct, though perhaps not actually 
 graceful. These circumstances are constantly occurring in practice ; and 
 knowing this, some may smile that any attempt at maxims should be 
 made. But there is a great margin for regulation; and even amongst 
 those necessary deviations, some guide derived from previous observation 
 is certain to present itself. 
 
 Fig. B. 
 
 fc. 
 
20 
 
 ARCHITECTURE OF MACHINERY. 
 
 This may be further illustrated by reference to the drawings, figs. 20 
 and 21, which are framings from 20-horse land engines: they neither 
 
 Fig. 20. 
 
 Fig. 21. 
 
 of them have much claim to beauty, but the most pleasing design is 
 certainly that in fig. 21 ; because, in the first place, its mechanical pro- 
 portions are more correct, the main limbs are made gradually stronger as 
 they recede from the centre of vibration, and become more affected by 
 leverage, and they are arranged so that a straight line is opposed to the 
 
ARCHITECTURE OF MACHINERY. 
 
 21 
 
 force. The other example, on the contrary, exhibits neither of these 
 requisites; the consequence is, that the engine is exceedingly unsteady, 
 often out of repair, and is noisy. This design abounds in sharp angles 
 and crude proportions without gradation, which do not occur to nearly so 
 great an extent in the other framing, the angles of which are rounded off, 
 and the same feature extended throughout. The ladder-like figure pos- 
 sessed by both is not a pleasing one, though so often employed in cast iron 
 work. If made with timber in the same way, the " truss " (if it may be 
 called one) would tend to " rack " under the influence of the vibratory 
 
 motion communicated to 
 it by the beam ; and not- 
 withstanding that the 
 principle has been de- 
 nied, the same rule by 
 which a truss of timber 
 is framed ought to some 
 extent to be considered 
 in the formation of an 
 iron one, as the mathe- 
 matical demonstration is 
 the same in both cases. 
 For the limb opposed to 
 the upward stroke of the 
 beam will tend to be- 
 come vertical, and is re- 
 strained by the cross 
 pieces of the frame, which 
 being of cast iron, ought 
 not to be exposed to a 
 tensile force. In proof 
 of this assertion the an- 
 nexed framing, fig. 22, 
 
 Fig. 22. 
 
 . 
 
22 ARCHITECTURE OF MACHINERY. 
 
 from the hands of one of our greatest engineers, is given, which exhibits 
 a complete "truss;" and it is conceived that the excellence of this me- 
 chanical arrangement will be at once admitted. The good taste shown in 
 this design arises from the use of " propriety of form." The angles are 
 frequent, but similar, and disposed uniformly over the work. The eye 
 may range along the lines without being checked by sudden breaks, and 
 the entire contour is bold and pleasing. 
 
 A better illustration of " boldness of outline " can hardly be given. It 
 is devoid of that gradual undulation generally admitted to be beautiful 
 where properly introduced. Many of the rules laid down by professors 
 of taste have been broken through, and it stands an elegant exception. 
 
 The variety of design found in the framings of marine engines furnishes 
 numerous and diversified examples of form and dimension. An inquiry 
 into the necessary arrangement of the parts will lead us to conclusions 
 from which much useful matter may be gleaned; and the reduction of 
 the several forces which act at certain points into their common resultants, 
 will tend to simplify the analysis so far that the direction of the main 
 features may be determined. 
 
 The " axis of motion " is a straight line drawn through the centre of 
 the paddle-shaft. 
 
 The velocity of the motion is the number of revolutions made by the 
 paddle-wheel. 
 
 The force is the quantity of power required to overcome the resistance 
 of the paddle-floats. 
 
 And the momentum is the quantity of matter multiplied into the 
 velocity, and will vary as the diameter of the wheel. 
 
 When at rest, the force exerted upon the framing is in the direction of 
 a vertical line drawn from the axis of motion, but is changed immediately 
 upon the engine being started. 
 
 Supposing the force P, capable of being exerted by an engine, acts 
 upon the crank-pin with a leverage I, and is resisted by the paddle-float, 
 
ARCHITECTURE OF MACHINERY. 23 
 
 equal to a quantity R, with a leverage L ; then P : R : : L : /. Suppose, 
 again, the crank and the paddle-arm to be in the same straight line, 
 they will act as a lever of the second kind, the float being the fulcrum, the 
 crank-pin the power, and the paddle-shaft the weight ; and the direction 
 of the resultant of the two forces will be common to the angles formed by 
 the crank and the paddle-arm as they change their position ; and as each 
 force is reciprocally proportional to the distance of its direction from the 
 fulcrum, so also will the inclination of the resultant towards the greater 
 force (or the power with greater leverage) be exactly proportional to its 
 excess over the lesser force or same power with less leverage. As the 
 engine starts, the actual direction of the resultant of these two forces 
 will depend upon the situation of that line which is a mean between 
 the two cranks ; and as this may be on either side of the vertical, some 
 feature of the framing must be arranged so that a straight line is opposed 
 to the direction of the force of this resultant. Again, suppose that the 
 base of the vertical framing is immoveable, and the part secured to the 
 cylinder immoveable also, the entire structure would tend to vibrate 
 backwards and forwards, with a leverage equal to the height of the axis 
 of motion from the dead plate. The opposition this motion meets with 
 when it inclines towards the cylinder is always ample, (supposing, of 
 course, the dimensions to be correct,) but in the contrary direction, that 
 portion of the framing being exposed to a tensile force, a foot beyond the 
 base is requisite, to act as a strut ; it being far preferable than allowing 
 the pull to be upon the cylinder. The framings of all our best engines 
 are thus arranged ; and it may therefore be considered essential. When 
 the framings of both engines are secured to each other, and form as it 
 were a single fabric, other facilities for giving stability present themselves, 
 and the " foot " may be reduced ; but considered separately, the framings 
 evidently require to be thus supported, and the above observations will 
 apply to independent structures where opportunities for giving transverse 
 stiffness do not occur. 
 
 The same relative proportion being supposed to exist between the 
 
24 
 
 ARCHITECTURE OF MACHINERY. 
 
 leverage of the crank and paddle-arm, the vibratory motion of the 
 framing will be nearly in the directions of the dotted arrows in fig. 23. 
 The resistance against its inclination towards the cylinder is evidently 
 
 Fig. 23. 
 
 sufficient ; in the contrary direction it is defective, and therefore the 
 deficiency in the quantity of power required is made up by a pull upon 
 the cylinder and condenser, to which parts the braces a and b are bolted. 
 The " mechanical proportions " of this design are consequently faulty. 
 The propriety of its form as regards " taste " cannot be called good, though 
 it has considerable merit. In some parts it is bold and effective, in others 
 it is crude and irregular ; but it may be taken as a framing of more than 
 average excellence, since, where there is one better, there are two worse : 
 perhaps its chief merit may be the absence of architectural ingenuity. 
 
ARCHITECTURE OF MACHINERY. 
 
 25 
 
 Fig. 24 is the framing of a 60-horse engine ; it has precisely the same 
 defect in its mechanical construction as the last example, but without 
 
 Fig. 24. 
 
 so much beauty : the angles in its composition are sudden and dissimilar, 
 the direction of its features change abruptly without curves or gradation, 
 and the truss is not carried out. " Changes of form in any system of 
 framing almost always increase the power of the weight, and often 
 produce cross strains that are attended with the worst consequences, 
 when such changes are not foreseen and provided for accordingly." 
 
 Stability is one of the most essential requisites in a marine engine 
 framing, and this is only thoroughly to be attained by the proper and 
 judicious arrangements of its parts. Dimension is not sufficient alone 
 to effect this : quantity of matter, if misapplied, would add weight, without 
 being of any service whatever. Every thing that tends to increase 
 stability, without increasing the weight of metal, is advantageous in a 
 
 E 
 
26 
 
 ARCHITECTURE OF MACHINERY. 
 
 mechanical point of view ; and it can be demonstrated, that a part so 
 added will, as regards propriety of form, be beautiful also. 
 
 It is not intended to be implied that the entire system of framing 
 should be extended in this direction, but only a single feature, for the 
 former could not be done without much weight : the latter carries no 
 defects with it, as the vertical portion can then be lightened to a great 
 extent. 
 
 In illustration of this, the following examples are given in fig. 25 (the 
 framing of an engine of 110 horses' power) : a great base is gained by the 
 
 Fig. 25. 
 
 addition of a truss opposed to the force acting in the direction of the 
 dotted line drawn from the axis of motion to the "foot " of the frame. 
 
ARCHITECTURE OF MACHINERY. 
 
 27 
 
 Now, since the greatest resistance offered to any weight is when a straight 
 line is opposed to it, the parts on the outside of the dotted line are in- 
 appropriately employed ; therefore the form is not correct, and the eye is 
 unpleasingly affected. Respecting the remaining portion of the design, it 
 is crude ; the angles do not compose, from the sudden change in the 
 direction of their lines ; the truss is not in direct opposition to the thrust, 
 and the boldness which the outline might possess, were the idea arranged 
 with " taste," is completely lost. It is a system dislocated and broken 
 up, a compilation of mechanical arrangements loosely and negligently 
 thrown together. The engine is certainly firm ; but the quantity of 
 metal used is much greater than with a different arrangement the same 
 amount of stability might be secured. 
 
 The framing, fig. 26, is from an engine of the same power us the last 
 
 Fig. 26. 
 
28 
 
 ARCHITECTURE OF MACHINERY. 
 
 Fig. C. 
 
 example. The reason for the disposition of each feature is immediately 
 apparent; the propriety of arrangement and form gained by the well- 
 directed adaptation of scientific rules is here made manifest. The parts 
 through which the power acts are carried as nearly as practicable in right 
 lines from the points of resistance to the centre of motion. The dimensions 
 of the various portions are regulated by experience, and the work is simple 
 and elegant. The stability of the engine amounted to perfect non-vibra- 
 tion ; and the reason undoubtedly is, that the action of the power passes 
 from particle to particle of the metal in the same direction : for it is 
 always the change in the direction of a force, or, which is the same thing, 
 change in direction of the resistance, that causes vibration. 
 
 In the foregoing examples, ornament is produced by relief, not deco- 
 ration. The cross sections of all the members are in the form represented 
 in the annexed diagram (fig. C), or some modification of 
 it. This arrangement may be carried out to almost any 
 extent, the degree of relief heightened or lowered ac- 
 cording to the quantity of strength required, and the 
 angles formed by the intersecting lines made gradual 
 by mouldings: the angles must not, however, be filled 
 up too much, owing to a practical difficulty, which 
 would then exist, of having the density of metal uniform 
 throughout the casting. When a mass of if on is placed in juxtaposition 
 with a thin plate, the cooling, and consequently the contraction, will be 
 unequal, and breakage likely to ensue ; but this will only be in extreme 
 cases. It may not be out of place to mention here, that when an ex- 
 tensive casting is to be made, the engineer would do well to mark the 
 places where the moulder shall make his "gates" and "vents;" for if 
 the air be not allowed free exit, spaces called " air-blows " will invariably 
 occur in the centre of the metal, which are difficult of detection, except 
 when near the surface, and consequently highly dangerous. It is there- 
 fore essential that the design for such a purpose should be made with a 
 due regard to the above observations. Numerous instances might be 
 
ARCHITECTURE OF MACHINERY. 29 
 
 quoted where accidents have occurred solely from these causes ; and here 
 again it is apparent that theory, without her companion, experience, 
 would be liable to error and accident. 
 
 Marine engine framings are by very common consent devoted to the 
 display of architectural ingenuity, and it would be difficult to find any 
 engineering work less appropriate for such arrangement. It must not be 
 thought that these remarks are made invidiously, or that the designs are 
 exhibited with the intent to ridicule : the purpose is far removed from 
 either of these; and difficult as it may be to point out error without 
 giving offence, it is nevertheless trusted the end will justify the means 
 by which it is attained. 
 
 Those men, by whose unwearied exertions these machines have been 
 rendered almost perfect in their performance, while they consider quantity 
 of work done as the "ultimatum" of their labour, look with indifference 
 at the external beauty of their productions. But why should they ? An 
 architect who alone studies convenience and applicability in his struc- 
 tures, who forms his doors simply as a means of admission, and his 
 windows only with a view of giving light, would be looked upon as a 
 mere labourer from the rude school of nature : his edifice might be com- 
 fortable, strongly built, well ventilated; nay, it might possess every 
 requisite that man might require for existence ; but " taste " will put 
 in her claim to notice, education will look for something by which its 
 cultivated ideas and refined habits may be gratified : a rude log, raised 
 to a convenient height from the ground, might serve as a chair or table ; 
 but polished luxury has brought from far and near material and art to 
 minister to its wants ; and whole lives are devoted to the accomplishment 
 of new designs, still further to extend and beautify. 
 
 Why then should externals be disregarded in those things which con- 
 stitute a real good ? It will be said, perhaps, they are not neglected : the 
 answer must be, if the attempt has been made, the result has proved a 
 failure. Why have volumes been written to prove, that, to be beautiful, 
 
30 
 
 ARCHITECTURE OF MACHINERY. 
 
 certain proportions must exist in certain figures ? and why should these 
 same figures be employed with different proportions, while the possibility 
 exists of forming them after the received rule ? There is no reason, argue 
 as we may ; nor excuse either, however logically the argument may be 
 treated. 
 
 Fig. 27 is a framing from an engine of the nominal power of 142 
 horses, (60-inch cylinder, 6-feet stroke). The dimensions of the vertical 
 
 Fig. 27. 
 
 <- 16...'!. 10 
 
 and transverse frames bear no proportion to the relative strains which 
 go through them. The curved line of the latter is contrary to every 
 statical law, for every force acting in the direction of the chord of 
 the arc would tend to break it with a leverage equal to the versed sine ; 
 
ARCHITECTURE OF MACHINERY. 
 
 31 
 
 and hence, to make up for the extra resistance, more metal is necessary, 
 and consequently more weight. The supposed necessity of this form was 
 evidently to enable a large condenser to be erected ; but the ingenuity of 
 the artist must have been at a low ebb indeed if he could find no other 
 means for its introduction than the distortion of his framing. With 
 reference to the vertical portion, it may be designated a very usual imi- 
 tation of architecture as applied to enginery, except the huge mass of 
 metal in the plummer block, which is unusual. 
 
 The three following drawings (figures 28, 29, and 30,) represent a 
 
 Fig. 28. 
 
32 
 
 ARCHITECTURE OF MACHINERY. 
 
 favourite style of composition, in imitation of the Grecian Doric ; but the 
 two last stand very far above that framing shown in fig. 28, the design of 
 which admits of no comparative expression, it is " hideously ugly." It 
 has been before observed, that this order is singularly inappropriate to this 
 purpose, even if in the execution of the model the detail were faultless ; 
 but in these, every feature is distorted to render it applicable to a purpose, 
 for which nothing but the strictest attention to form and dimension can 
 produce either beauty or propriety. 
 
 Fig. 29. 
 
 The most insignificant domestic utensil is in the present day fashioned 
 in some way to please the eye : for those which are most frequently in 
 view, some classical design, some graceful figure, is chosen, in which it 
 may be made ; and frequently things are turned out of the shop of the 
 
 
ARCHITECTURE OF MACHINERY. 
 
 33 
 
 simple coppersmith of as beautiful proportion as those from which they 
 were copied. How then can the engineer, who often is, and invariably 
 ought to be, a man of education, suffer such absurdities to exist as we 
 constantly see in the productions of his labour ? He cannot lead where he 
 
 Fig. 30. 
 
 attempts an architectural order, for ages have consented to receive the 
 proportions given by the ancients for their five, nor has a sixth yet been 
 produced ; and if he cannot even follow, let him eschew the fashion of 
 architecturalizing his machinery, and content himself with studying the 
 more efficacious method of framing given in the first part of these 
 observations. 
 
 To turn to another practical point. It may be observed in several 
 examples, that the flanch of the transverse frame is made to bolt on to 
 the top as well as to the sides of the cylinder ; but as a bolt can only 
 
34 
 
 ARCHITECTURE OF MACHINERY. 
 
 be efficaciously employed to resist tension, those which pass through the 
 top are useless; they become affected by a constant alternating thrust, 
 and pull and "work loose;" a slight lip to rest the frame upon while 
 fitting is all that is required. 
 
 Those stationary appendages to all machinery, plummer blocks or 
 pedestals, are apparently thought by many engine-makers too insig- 
 nificant to claim the smallest attention, at least as far as proportion is 
 concerned. A lump of metal, cast of a sufficient size to admit the brass 
 in which the shaft revolves, with another lump placed over it, and 
 secured by bolts, will answer for a general description of these parts. 
 If, however, in some rare instance ornament be attempted, the block 
 becomes diversified, but it still remains a block in every sense of the 
 word. Their dimensions can only be given by experience; therefore 
 their form must be regulated by a practised eye, and their propriety by 
 example. 
 
 Figs. 31, 32, and 33 are blocks of a large size, and show different 
 
 Fig. 31. 
 
 i: 
 
 Fig. 32. 
 
 ffl n 
 
 arrangements. The first is met with in the better class of work, but has 
 fallen considerably into disuse, from the difficulty, or rather the expense, 
 of fitting the brass properly on its bearings. 
 
 Fig. 32 is seen in Millwright's work : it stands far below the former 
 in point of shape, and is both weak and poor at the angles of the base. 
 
 Fig. 33 has superseded or is fast superseding the other kinds of 
 
ARCHITECTURE OF MACHINERY. 
 
 35 
 
 Fig. 33. 
 
 block, and is evidently more appropriate for its purpose. The iron 
 base and cap are fixed in their position, and bored out; the brass is 
 turned and accurately fitted, and then bored out concentric with the 
 outer circle; the bolts are also turned 
 and fitted into bored holes, which take a 
 portion of the brass, and serve as keys : 
 this mode of fitting, which is both cheaper 
 and more accurate than the older plans, 
 with octagonal or square brasses, will no 
 doubt eventually be brought into general 
 use: it has also the recommendation of 
 character, and the capability of being 
 made of very pleasing form. For in- 
 stance, if the pedestal be high, the base 
 
 may be extended by a gradual curve in the moulding to any dimension, 
 without increasing the weight of metal ; and the form of moulding may 
 be struck to coincide with any that may be employed in the other por- 
 tions of the machinery of which it forms a part. The figure is taken 
 from two examples of this kind of block. 
 
 Fig. 34 shows a pedestal from a 50-inch cylinder pumping engine : the 
 curve uniting the vertical part 
 with the base, it will be perceived, 
 is gradual, and more sharp as it 
 joins the horizontal plate; and 
 it may be observed here, that 
 segments of circles should never 
 be used in " rounding off;" they 
 are always formal, and devoid 
 of that elegance which the sweep, 
 if properly arranged, must pos- 
 sess. Workmen are sure to use 
 their compasses to strike out a curve, unless expressly forbidden to do so. 
 
 Fig. 34. 
 
36 
 
 ARCHITECTURE OF MACHINERY. 
 
 The drawing office is furnished with moulded curves of a size suitable to 
 the smaller scale to which the draughtsmen work. Would it not be well 
 to have the same curves increased to full-sized dimensions, and made of 
 sheet iron, for use in the pattern shop ? There would then be a certainty 
 of the intended design being carried out. These curves may, if thought 
 desirable, be numbered to correspond with those in the office. The 
 pedestal which is now referred to is exceedingly good, and may be 
 increased in height 6 or 8 inches, with the same curve, and still retain 
 its character. The semicircular recess is 3 inches deep on each side. 
 
 Fig. 36. 
 
 Fig. 35, which is a pedestal from a 45-horse land engine, represents 
 one of those ludicrous attempts at show that have been referred to. It 
 means nothing, but is still evidently an attempt: its weight is nearly 
 9 cwt. 
 
 Fig. 36 is the pedestal for the centre gudgeon of a beam for an engine 
 of 12-horse power; the usual diameter, if of cast iron, being 3J inches. 
 The arched form still further increases the difficulty, which always to 
 some degree exists, of giving a graceful form to these features, and not 
 the least advantage is gained in a mechanical sense. 
 
ARCHITECTURE OF MACHINERY. 
 
 Fig. 37. 
 
 37 
 
 Fig. 38. 
 
 Fig. 37 exhibits a pedestal from a 25-horse engine, which broke at the 
 part marked, owing to an " air-hole " in the casting. The lines are 
 formal and shapeless ; it was replaced by the block represented in fig. 38, 
 which is of a better figure, but still not perfect. 
 
 All the five last examples are for the bearings of the centre gudgeons 
 of beams. 
 
 On smaller pedestals, which support the bearings of the starting gear of 
 marine engines, for instance, and from which the three following examples, 
 figs. 39, 40, and 41, are taken, it is only necessary to make two remarks, 
 given rise to by occurrences which theory cannot account for. 
 
 Fig. 39. 
 
 Fig. 40. 
 
 Fig. 41. 
 
 In fig. 39 the pedestal is fixed in a sloping position, very properly if 
 the force applied to the starting lever acted always in a contrary direction ; 
 but such not being the case, the pedestal in the case here quoted constantly 
 
38 ARCHITECTURE OF MACHINERY. 
 
 became deranged by the bolts a and b working loose ; so that it was found 
 necessary to substitute the pedestal shown in fig. 40. 
 
 Another circumstance leads to the conclusion that these parts should 
 not be cast upon the framing ; for the pedestal represented in fig. 4 1 broke 
 at the part marked at the time of starting the engine, owing to an air- 
 hole which existed in the centre of the metal. Nor is this a singular 
 instance, for many accidents of the like nature have happened. 
 
 Attention to the effects of different forms upon the eye may make every 
 detail, however insignificant, subservient to the rules of " taste," and 
 pleasing under circumstances which would render others intolerable : 
 practice will enable us to portion out the different degrees of relief or 
 ornament necessary in the various parts of machinery with great facility : 
 a well-directed judgment will point out to us the reasons why the means 
 employed should produce the required end: and these, combined with 
 theoretical knowledge, will render it a task of delightful labour to design 
 and execute. Nothing grows so much upon us as the love of our pro- 
 fession, when once the way to reasoning upon its advantages is opened : 
 each new thought discloses some new path leading to greater imaginings, 
 and still we are led on; emulation, the desire of excellence, and the 
 knowledge we acquire of our own powers, are fascinating impulses which 
 cannot fail to stir up in the minds of the most insensible, those feelings 
 which nature has engrafted on all men ; and what profession affords so 
 much scope for energy of intellect as that of the Civil Engineer ! 
 
ARCHITECTURE OF MACHINERY. 39 
 
 In the practical operations connected with the strength of beams for 
 steam engines, that sectional area which would be found by calculation is 
 never thought sufficient, and every one adds to it as he thinks necessary ; 
 hence results the variety in the strength of beams for engines of the same 
 power, the velocity and length of leverage remaining constant. 
 
 The examples here quoted, with a view to furnish proper dimensions, 
 are from engines that have been at work at least thirteen years, and one 
 twenty-five years. 
 
 It has been remarked, (but upon what data we shall not stop to con- 
 sider,) that it is impossible to tell if the parts of a machine are strong 
 enough until they have been at constant work for a year; the reason 
 given being, that the total amount of deflexion will not be arrived at 
 before. It is needless to say, that this is an absurd dogma; for if a 
 force permanently alters the shape of a beam, it may be considered as 
 broken. 
 
 We are at no loss at the present day for data on which to found cal- 
 culations for the theoretical determination of the load any form of beam 
 will carry, in whatever part and in whatever direction the force acts, or 
 whether it be stationary or moving. Therefore, the only difficulty, (if it 
 can be called one,) seems to be in the determination of the quantity of 
 weight that is actually applied to the end of a beam. 
 
 If a cast iron beam, 1 inch square, and 1 foot long between bearings, is 
 capable of supporting a load in the middle of its length and at rest, equal 
 to 900 flbs., one double that length will only support 2 - a or 450 flbs., or 
 nearly so ; but if it be double the depth, with the same bearing, it will 
 carry 900 X 4 = 3600 fcs. ; if it be double the breadth, with the same 
 bearing, 900 X 2, or 1800, will be its load ; and this is self-evident. 
 
 When the force acts by impulsion, as in the engine beam, the velocity 
 of the motion must be taken into consideration ; for if the effect of the 
 moving force be greater than the elasticity of the material, it would break. 
 Now the velocity of a double-acting or rotatory engine may be taken at a 
 
40 ARCHITECTURE OF MACHINERY. 
 
 constant quantity of 3*6 feet per second ; therefore the number of pounds 
 pressure acting upon the end of the beam, multiplied by this velocity, will 
 give the resistance necessary to be overcome. 
 
 As regards propriety of form, the vertical section or outline of a beam 
 ought to be a figure including the parabolic curve ; that is, two parabolas 
 joined base to base at the centre of motion. 
 
 The following are calculations made for a beam of an 80-horse double- 
 acting engine ; but at the time they were made, the imperfections in the 
 machinery being so much greater than at present, the same results will 
 not be arrived at, but the error will be on what is called the safe side. 
 
 The velocity of the piston was 216 feet per minute, or 3*6 feet per 
 second, and 200 fibs, raised through that space in that time was called a 
 horse power. The pressure in the boiler was 2Jibs. on a square inch 
 above the atmosphere. But the pressure on each square inch of piston 
 surface was reduced by friction, imperfect vacuum, and rarefaction of the 
 steam by the throttle- valve, in its passage from the boiler to the cylinder, 
 to 9 Ibs. per square inch ; and 47f inches was taken as the requisite 
 diameter of the cylinder. Again, since the beam had to overcome the 
 total resistance of the machine, 17*5 Ibs. per square inch was calculated as 
 the force that would act upon it, minus the velocity, or 17 '5 X 1781 '3 
 = 31172-75 Sbs. X 3-6 = 112221-90, or 50 tons, equal the force of im- 
 pulsion. But it was considered that the momentum acquired during the 
 ascending or descending stroke of the engine had to be overcome at the 
 moment of the change of motion, and that it was equal to the friction of 
 the entire machine multiplied into the velocity, 3*6 feet per second. This 
 was equal to nearly 6 Ibs. per square inch of piston surface ; therefore the 
 additional strength required for the beam was 1781*3 X 6 X 3*6 = 
 38476 fos., or 17'13 tons, making a total of 17'13 -f 50 = 67'13 tons. 
 
 The known dimensions of the beam were as follow : length from centre 
 of gudgeon to point where piston rod acts, 12 feet; thickness of each 
 plate, exclusive of the flanches, 2'5 inches. It remained therefore to find 
 the necessary depth. 
 as 
 
45 Horse . Diam, r of Cylinder 36 % *? Stroke 6 feet . N 18 P- Mm 6 . 
 
 - -9.3- - : 
 
 J 
 
 John Weak. Architectural Library. Sffjiigh, ffotborn . 
 
ARCHITECTURE OF MACHINERY. 41 
 
 The datum of the computation was that a beam, supported at the ends, 
 1 inch broad, 12 inches in depth, and 12 feet between supports, would 
 sustain, without permanent deflexion, the weight of 4'90 2 tons at the 
 middle of its length. And this, in the case of a beam, would resolve 
 
 itself into - = length, . = weight ; because it must be considered 
 
 .* /w 
 
 as if it were permanently fixed at one end, and loaded at the other ; and 
 the stress at the end by the weight is the same as the stress upon the 
 middle of a beam of twice the length, with twice the weight laid on its 
 middle ; this beam being supported at both ends. 
 
 The thickness of experimental beam, increased to 5 inches, would bear 
 2'45 X 5 = 12-25 tons. The length increased to 12 feet would reduce 
 
 12-25 ,. 67-13 
 
 this one half, or Lf = 6-125 v ^^ = 10-96 and 10-96 X 12 
 2 6 '125 
 
 = 39'72 inches, the required depth of the beam. With these dimensions 
 it was actually erected, and has now been at constant work rather more 
 than twenty-four years. 
 
 The elevation, plan, and cross section of this beam are represented in 
 fig. 42, Plate VII. The lengths of the ordinates are figured, and the line 
 of the parabola dotted. Unsatisfactory as this mode of proceeding may be 
 to theorists, it was nevertheless that adopted by one whose knowledge of 
 the doctrines of statics and dynamics was perfect. The success attending 
 this example furnishes a dangerous precedent ; for it is much weaker than 
 most other beams for the same power : and although it might not actually 
 break, if subjected to violent concussion, it would be very liable to do so. 
 
 The following are the areas of the cross sections of beams taken at the 
 centre from three different sources : area in above example, 198 '60 inches ; 
 average of seven cases from beams that have been at work three years, 240 
 inches ; by Tredgold's rule (Practical Essay on the Strength of Cast Iron, 
 p. 273, et seq.j 4th edition), 427*4 inches. 
 
 2 Tredgold, 4 '55 tons ; Turnbull, 4 '95 tons. 
 G 
 
42 
 
 ARCHITECTURE OF MACHINERY. 
 
 It has been remarked, that the only difficulty which presents itself in 
 these calculations is the determination of the quantity of impulse given 
 to the end of a beam. In an engine employed for spinning, pumping, 
 blowing, or any regular work, the smaller areas may be used with 
 confidence, and the total pressure that can act on the piston, plus Jrd of 
 this pressure, multiplied by the velocity, will give the force. But for 
 engines used to drive bone-mills, stampers, falling-stocks, or breaking 
 down rollers, the dimensions can only be given by experience. Work- 
 men are very apt to do mischief by bringing " blooms " to all the rollers 
 at once, and what they term " stall " an engine, when either it is brought 
 to a dead stand, or something breaks. These pages are not, however, 
 intended for disquisitions on points of theory. The present digression is 
 merely an attempt to convey some idea of the causes which lead to the 
 error of making prodigiously strong beams, and to show that there is 
 more excuse for the use of unnecessary metal in the moving, than in the 
 stationary, features of engines. 
 
 The friction of any part, cceteris paribus, increases very nearly in the 
 same proportion as the weight ; therefore a beam that is exactly strong 
 enough is advantageous in this respect, as well as being more pleasing 
 to the eye. 
 
 In the outline of a beam, that which approaches nearest to a parabola 
 will be the most perfect, both with regard to mechanical laws, and beauty 
 of proportion. Within this figure it will be weak, if the centre be only of 
 the right dimensions ; and all beyond that curve will be weaker also, by 
 
 Fig. 43. 
 
 39J-inch cylinder 4 feet stroke. 
 
ARCHITECTURE OF MACHINERY. 
 
 43 
 
 the weight of such extra matter. As to propriety of form, then, for an 
 engine beam, there cannot be two different opinions. 
 
 Fig. 43 is an example of the beam of a 50-horse rotary engine, with a 
 velocity of 200 feet per minute. The area of its transverse section is 
 63'75 inches ; the outline is a parabola, having its vertex at the centre of 
 the end gudgeon. 
 
 Fig. 44 is a beam for a 45-horse engine, upon the same construction as 
 the last. Its sectional area is 90 inches ; and supposing it to be no more 
 
 39-inch cylinder 3' 9" stroke. 
 
 than necessary, the neck at a is too weak, because it falls within the line 
 of the curve; the portion at b is too strong, because it projects beyond. 
 These errors could not fail to strike the eye of the most casual observer, 
 and to a man of taste and judgment would appear highly offensive. 
 It is impossible to guess the designer's motive for making his beam in 
 this form, since every deviation from the proper form is disadvantageous, 
 both for beauty and strength. 
 
 Fig. 45, Plate VII., shows a beam for a 45-horse house engine, used 
 for rolling lead. The end gudgeon is fitted upon a turned bearing at 
 right angles with its centre of motion. By this arrangement the contour 
 is perhaps slightly broken ; but as it is one possessing the advantage of 
 allowing the perpendicular links play, and of self-adjustment to some 
 extent, it may counteract the small defect above mentioned. Custom also 
 lessens minor imperfections; for taste is generally thus influenced when 
 
44 
 
 ARCHITECTURE OF MACHINERY. 
 
 there is no axiom for its regulation, or decided law to govern it. Beams 
 for engines of a higher power than this are made double, as in fig. 42, for 
 the convenience of lifting or fixing them in their places. 
 
 The dimensions of cast iron beams for marine engines also vary to a 
 great extent, and rules for the determination of their strengths are ar- 
 bitrary, especially in sea-going ships, where the machinery is exposed to 
 heavy blows from the sudden emersion of the paddle-wheels to a greater 
 depth than that estimated. 
 
 Fig. 46 is a beam for an engine with a 42-inch cylinder, differing in 
 
 Fig. 46. 
 
 42-inch cylinder 4 feet stroke. 
 
 construction from any of the previous examples, and having the centre 
 swelled out to allow a greater quantity of metal to be placed round the 
 gudgeon. 
 
 Fig. 47 represents another beam for a sea-going vessel of the same 
 
 Fig. 47. 
 
 42-inch cylinder 4' 2" stroke. 
 
 power as the last, but with a much greater sectional area. 
 
ARCHITECTURE OF MACHINERY. 
 
 45 
 
 Fig. 48 is a valuable example, because it furnishes us with material 
 from which much that is useful may be gained. The boat was employed 
 
 Fig. 48. 
 
 33-inch cylinder 3 feet stroke. 
 
 for river service in the first instance, and performed its duties without 
 difficulty ; but during its first voyage to sea, the unequal shocks to which 
 the beam was exposed caused a fracture at the place marked in the 
 drawing, having a sectional area of 27 inches. The new beams erected 
 in the place of the old ones had precisely the same dimensions at the 
 centre, but the arms were increased to the size shown by the dotted line, 
 the place of the fracture having now the area of 31 inches; and this has 
 been found ample. The design of this beam is very inferior, from the 
 frequent occurrence of crude angles : the square centre especially is dis- 
 pleasing, besides being more difficult and expensive to fit. 
 
 The next illustration, fig. 49, exhibits another dimension, much at 
 
 32-inch cylinder 3 feet stroke. 
 
 variance with the greater majority of these parts of marine engines. The 
 engine is of less power than that to which the last beam was attached, 
 
46 
 
 ARCHITECTURE OF MACHINERY. 
 
 and its sectional area at the same distance from the centre of motion 
 is 56'37 inches. Now discrepancies to a certain extent are allowable, 
 but there is moderation in all things, and by this enormous addition of 
 weight, the friction of the machine is increased to even a greater degree 
 than the difference in the quantity of matter. Extra weight in any 
 stationary feature is objectionable; but when a considerable velocity is 
 to be conveyed to it, the defect becomes very serious, and an injurious 
 influence is communicated to the entire structure. These two last beams 
 are single, or rather, are not formed of parallel plates in separate 
 castings. 
 
 Fig. 50. Amongst the numerous incidents by which experience is 
 
 Fig. 50. 
 
 31 -inch cylinder 2' 6" stroke. 
 
 so much assisted, and which adds to the useful stock of memoranda to 
 be met with in the note-book of the practical man, the following may be 
 quoted. These beams, made with two plates, each 14*5 inches deep in 
 the centre, by 1 inch thick, were fixed to an engine with a cylinder 
 31 inches diameter, 2 feet 6 inch stroke, and with a velocity of 155 feet 
 per minute ; the pressure of steam in the boiler being 4|- fcs. above that 
 of the atmosphere. The engines had performed their work steadily for 
 six months, when suddenly, during a heavy sea, a plate of one of the 
 beams gave way. The engineer being called upon to replace the beam 
 at his own cost, refused to do so, upon the ground, that as it had 
 performed its ordinary duty with security, it was not just that he should 
 become responsible for accidents to which all machinery, exposed to such 
 risks, was liable : he moreover affirmed, that notwithstanding the above 
 event, the same sectional area was ample for an engine of that power ; 
 nor would he increase the dimensions in any similar instance. The case 
 
fc 
 ARCHITECTURE OF MACHINERY. 47 
 
 was referred to arbitration, and the remaining sound plate of the same 
 beam was subjected to the following experiment: it was fixed upon a 
 turned bearing, and secured by keys ; a fine line was tightly stretched to 
 correspond with the centres of the gudgeons, and each end loaded 
 gradually until it broke, when the weights were found to be 16 tons, 
 3 quarters, and 1 1 fibs, on the side of the fracture, and less by 27 Ibs. on 
 the other ; which made the resistance of the four plates, when at work, 
 equal to 67 tons, 1 quarter, and 16 fts., very nearly three times that 
 given by the supposed necessary calculation. 
 
 As it could not be thought the engine had exerted this enormous 
 force, it was presumed that the fractured plate must necessarily have 
 had a flaw; but every particle of the metal exhibited the lustre of 
 newly broken iron. In fact, such occurrences as these cannot be ex- 
 plained, and as it is utterly impossible to assign a reason why they should 
 happen, it is equally difficult to insure perfect freedom from accident. 
 The means employed for connecting the different moving parts with 
 one another will generally have great influence, as far as regards their 
 safety ; since, if they are not allowed play, the bearings require to be so 
 exactly fitted, that if one brass or gudgeon wears more than another, 
 some deflexion must take place in that which is farthest from the point 
 of impact ; and if that deflexion is greater than the elastic power of the 
 material, it will of course break. 
 
 Wrought iron beams appear for this reason singularly appropriate 
 for marine engines; and it seems a contradiction to employ wrought 
 iron cranks, and then to break the connexion by making cast iron 
 beams. 
 
 The great diminution in the weight of matter is also an advantage 
 which these beams possess over those of cast iron. 
 
 The expense of fitting, the complication of parts, and consequently the 
 increase of friction experienced in beam engines, have led engineers to 
 contrive means by which they may be simplified, and the beam dispensed 
 with. The latest war steamers are fitted with engines of this kind, and 
 
48 
 
 ARCHITECTURE OF MACHINERY. 
 
 it will not be long before a beam in every description of engine will be 
 thought an absurdity. 
 
 With reference to the title of this Essay, the following Table may be 
 considered out of place; but since the dimensions contained in it are 
 practical, and those which have been collected during its compilation, 
 they may perhaps be useful. 
 
 Nominal 
 
 Diameter 
 
 Length 
 
 Velocity 
 
 
 Length of 
 
 
 Thickness 
 
 
 power of 
 
 of 
 
 of 
 
 in feet ^ 
 
 Description of 
 
 beam from 
 
 Depth at 
 
 of 
 
 Sectional 
 
 engine. 
 
 cylinder. 
 
 stroke. 
 
 minute. 
 
 work. 
 
 centre. 
 
 centre. 
 
 metal. 
 
 area. 
 
 
 
 
 
 
 / // 
 
 
 
 
 80-horse 
 
 47-f 
 
 7 
 
 216 
 
 Blowing 
 
 12 
 
 39-72 
 
 2-5X2 
 
 198-6 
 
 
 471 
 
 8 
 
 210 
 
 Rolling 
 
 12 4 
 
 48 
 
 2-5X2 
 
 240 
 
 65 
 
 421 
 
 7 
 
 216 
 
 Grinding corn 
 
 10 9 
 
 42 
 
 2-5x2 
 
 210 
 
 56 
 
 40J 
 
 7 
 
 180 
 
 Pumping 
 
 10 4 
 
 36 
 
 2-25x2 
 
 162 
 
 55 
 
 39 
 
 6 9 
 
 220 
 
 Blowing 
 
 9 6 
 
 38-5 
 
 2-5 
 
 96-25 
 
 50 
 
 39i 
 
 4 
 
 200 
 
 Blowing 
 
 6 9 
 
 25-5 
 
 2-5 
 
 63-75 
 
 45 
 
 39 
 
 3 9 
 
 210 
 
 Mill-work 
 
 6 6 
 
 30 
 
 3 
 
 90 
 
 45 
 
 36| 
 
 6 3 
 
 216 
 
 Rolling 
 
 9 3 
 
 30 
 
 2 
 
 60 
 
 36 
 
 311 
 
 6 
 
 216 
 
 Blowing 
 
 9 
 
 30 
 
 2 
 
 60 
 
 20 
 
 24 
 
 5 
 
 190 
 
 Mill-work 
 
 8O 1 
 * 
 
 24 
 
 2-25 
 
 54 
 
 20 
 
 231 
 
 5 
 
 216 
 
 Mill-work 
 
 8 
 
 25 
 
 2 
 
 50 
 
 14 
 
 20 
 
 4 6 
 
 200 
 
 Rolling 
 
 7 8 
 
 21 
 
 2 
 
 42 
 
 12 
 
 18* 
 
 4 
 
 216 
 
 Mill-work 
 
 6 10 
 
 22-25 
 
 2-25 
 
 50 
 
 8 
 
 15 
 
 4 
 
 216 
 
 Oil-mill 
 
 6 8 
 
 18 
 
 1-75 
 
 31-5 
 
 
 42 
 
 4 2 
 
 180 
 
 Marine 
 
 6 6 
 
 27 
 
 2x4 
 
 216 
 
 
 42 
 
 4 
 
 200 
 
 Marine 
 
 6 3 
 
 23 
 
 1-5x4 
 
 138 
 
 
 33 
 
 3 
 
 180 
 
 Marine 
 
 5 
 
 17-25 
 
 2-25x2 
 
 77-6 
 
 
 32 
 
 3 
 
 180 
 
 Marine 
 
 5 
 
 22 
 
 3x2 
 
 132 
 
 
 31 
 
 2 6 
 
 200 
 
 Marine 
 
 4 3* 
 
 14-5 
 
 1x4 
 
 58 
 
-9 ' fZ- 
 
 VfUOJf Off 
 
 o 
 
 c 
 
 -~\ 
 
 
 .jr< 
 
 ~ff-n~ 
 
ARCHITECTURE OF MACHINERY. 
 
 The transmission of power through the connecting rod, the alternating 
 thrust and pull which influences it, the constant variation in the direction 
 of the motion, and consequently the change in the quantity of strength 
 required at each different portion of its path, renders the calculations for 
 its dimensions exceedingly complicated ; nor are they at all satisfactory, 
 since in no single instance has the theoretical resistance been found 
 correct when adopted in practice. The quantity of deflexion or " swag " 
 that affects the direction of the downwards thrust is perhaps the most 
 essential point to be considered in the analysis, and the spring from 
 the elastic properties of the metal, which must be sensibly felt through 
 the rod immediately and on the instant of the change in the motion, 
 exerts a serious power, which experience alone can correctly appreciate. 
 And since this is felt more in long than in short rods, the dimensions will 
 not be in the theoretical proportion to their respective lengths, although 
 it has been so stated. 
 
 The examples of connecting rods given in Plate VIII. will show by 
 mere inspection how very much their proportions differ from one another, 
 and how seriously mistaken some have been in their ideas of necessary 
 strength. Those for 8-horse power have a difference in their sectional 
 area at a of 16'75 inches ; the greater section belonging to the shorter 
 rod, which is rendered clumsy in appearance, and is an injurious drag 
 upon the engine. The rods for 30-horse engines also exhibit discre- 
 pancies which would not exist if the theory of their action were properly 
 considered. As in the case of beams, the different kinds of work for 
 which they are employed would make some difference in their requisite 
 dimensions, but not nearly so much as the drawings show. Wrought 
 iron is very appropriately used in instances when the rod is of incon- 
 siderable length, since it gives an idea of even greater security, with 
 much more elegance. 
 
 Propriety of form and an exact adjustment of material are equally 
 essential in these features. It is perhaps impossible to convey to the 
 
 H 
 
50 ARCHITECTURE OF MACHINERY. 
 
 mind any just reason for the cause, or to make it sensible to the merits 
 or demerits of a design, without mathematical demonstration, and even 
 then well-directed judgment must be exercised in the arrangement of 
 the parts; for a mere theoretical rod would look strangely deficient to 
 the practised eye, and would surely be equally disadvantageous when 
 comparison had to be made with quantity of work performed. . The 
 jingling contrivances for transmitting power which we constantly see 
 exhibited, arise solely from the want of practice or of care in the artist 
 who planned them ; and the theorist would be equally lost if he should 
 venture beyond his construction upon paper : his neutral axis, so essential 
 to the nice solution of his equation, would be found any where but in the 
 expected place : his vis viva, estimated to the uttermost fraction, would 
 contradict him, and the modulus of elasticity of his material would display 
 itself in a different light under every varying gradation of quality, from 
 Old Park iron downwards. Notwithstanding all this, however, it is 
 doubtless a gratifying exercise to draw from assumed data the ever- 
 lastingly changing results which arise from mathematical investigations, 
 and to find them lead step by step to the anticipated solution; but it 
 would be just as unpleasant to find that the magical Q. E. D. had lost its 
 charm when put to the test of practice, and which it assuredly would in 
 the cases just referred to. 
 
 Propriety of form in the detail of machinery depends upon two circum- 
 stances. The first is, that the parts subject to wear and tear, and in- 
 fluenced by strains, should be capable of motion or adjustment : the 
 second, that every portion should be equally strong, and present to the 
 eye an uniform figure, or one that is consistent with its degree of action : 
 theory, practice, and taste, all must combine to produce such an one. 
 A great extent of beauty is attainable in all the details, but mathematical 
 reasons cannot be given why a certain arrangement of lines should be 
 preferable to another, provided they are equally strong. Truth does not 
 strike us without the assistance of custom; but so great is the force of 
 custom, that unassisted by truth it has worked the greatest miracles ; and 
 
r 
 
 ARCHITECTURE OF MACHINERY. 
 
 51 
 
 it certainly must be this universal Mentor which gives us the power to 
 choose between forms. It has been already frequently observed in 
 speaking of general contour, that sharp angles and irregular curves 
 disturb the eye : these must be also considered in detail. Owing to 
 the above circumstance, (the absence of a decided law,) opinions alone 
 can be given upon the various Fig. 51. 
 
 parts of machinery which fol- 
 low : it is not possible to do 
 this without disagreement, but 
 it will produce at least one good 
 effect, if no other ; it will sug- 
 gest thought, so often absent in 
 these designs. 
 
 Figs. 51 and 52 represent two 
 different methods of forming 
 the crank ends of cast iron con- 
 necting rods for land engines. 
 Concerning the first there can- 
 not be a dissent : it combines 
 nearly all the essential properties 
 for strength and beauty ; it is 
 capable of high " finish," with- 
 out being rendered " meanly 
 showy;" and the expense of 
 planing the flat face would be 
 trifling. The swell on each side 
 is to allow a broad bearing for 
 the key, and therefore has a 
 meaning. 
 
 The second figure is in the 
 opposite extreme: it is heavy, 
 
 devoid of SVHimptrv and Histinrt Connectin g rod ends 80-horse land engines. Cast iron 
 [ byn letry, anc gcale ^ inch ._ Length between centres 25 feet. 
 
52 
 
 ARCHITECTURE OF MACHINERY. 
 
 Fig. 53. 
 
 Crank end. 
 
 means are employed for resistance, by which the determination of the 
 strength is rendered uncertain, and the liability to error increased. 
 
 To gain any end, no means can be too simple, and, if possible, the 
 same material should be used throughout ; for if wrought and cast iron, 
 for instance, be brought in close contact, performing the same extent of 
 duty, their dimensions 
 will certainly cause an 
 unpleasing effect. There 
 are of course exceptions, 
 but this may be taken as 
 a general rule. 
 
 The circumstances con- 
 nected with the safety of 
 moving parts (mentioned 
 at page 47) may here be 
 more fully explained. 
 
 The most approved 
 system of attaching the 
 ends of rods to their cross 
 heads is by a joint, move- 
 able in the direction of 
 their strain, as in fig. 53, 
 thus allowing some play 
 and wear in the brasses, 
 without giving rise to 
 bad consequences. The 
 expense of fitting is not 
 at all increased, nor is 
 the appearance less per- 
 fect. 
 
 The arrangement to 
 
 be avoided IS represented Connecting rod en ds 80-horse marine engine, wrought iron. Scale J i 
 
ARCHITECTURE OF MACHINERY. 
 
 53 
 
 Fig. 54. 
 
 Crank end. 
 
 in fig. 54, and it must be obvious, that unless every bearing be exactly 
 true and square, the strain will be unequally distributed ; and if the 
 quantity of error be greater than 
 the elastic power of the material, 
 a permanent deflexion will be 
 established, which may be con- 
 sidered as the first stage of frac- 
 ture. 
 
 The form of ends made with a 
 strap and cotter does not admit of 
 much variation : those shown in 
 the annexed diagrams are the 
 most approved. The octagonal 
 neck below the square head should 
 always be introduced ; it is a gra- 
 duation towards the round part of 
 the rod, and gives a character to 
 the work. The shaft of the rod 
 should swell out from 4f inches 
 at the neck to 5|- inches at the 
 middle; and mouldings at this 
 point should be studiously avoid- 
 ed, since it is decidedly " con- 
 structing an ornament" when it 
 can be of no use whatever: it 
 also breaks the line, and prevents 
 the eye from following it. 
 
 It may also be mentioned here, that the pins and gudgeons about 
 which there is any constant motion should be larger in diameter than is 
 actually required for strength ; for the rubbing produces heat, slight ex- 
 pansion consequently follows, and thus the friction of a small pin is found 
 in practice to be greater than that of one moderately larger in diameter. 
 
 Connecting rod ends 80-horse marine engine, wrought 
 iron. Scale in. 
 
54 
 
 ARCHITECTURE OF MACHINERY. 
 
 The description of work put upon wrought iron 
 rods should be plain and square, perfectly free 
 from any unnecessary projection, and their shapes 
 should, if practicable, be such that they may be 
 finished in the lathe and planing machine with- 
 out much vice-work, except at the octagonal 
 necks. 
 
 The same remarks apply equally to all the dif- 
 ferent rods connected with the steam engine and 
 to machinery in general. 
 
 Side rods are constructed in two different ways, 
 according to the custom of the engineer, either with 
 one " dumb end" and strap, or with both ends 
 having straps. The selection of these is a matter of 
 practice or convenience, and may therefore be left 
 open to choice. The strap end has, or rather 
 ought to have, the same form in every instance; 
 consequently what has been previously said will 
 not be repeated. The proper form for the eye is 
 readily explained. Fig. 55 exhibits that which 
 is correct; the oval shape being necessary to give 
 equal strength to the metal surrounding the brass, 
 and to allow of its adjustment and wear. The 
 head metal is increased in thickness from the sides, 
 because the strain is not altogether tensile, and 
 the particles become more influenced by stress, 
 requiring a greater area for equal resistance. The 
 external figure is often varied, as in fig. 56, but 
 not with good effect, as far as figure itself is con- 
 cerned. 
 
 The circular eye, fig. 57, is not calculated for 
 equal strength, nor is the brass bush practically 
 
 Fig. 55. 
 
 Fig. 56. 
 
 Side rod 50-horse marine 
 engine. Scale H in. 
 
ARCHITECTURE OF MACHINERY. 
 
 55 
 
 Fig. 57. 
 
 correct; for there is no provision 
 for wear, and the pin that holds 
 it presents an unpleasant excres- 
 cence. 
 
 Fig. 58 is a sketch from the air- 
 pump side rod of a 50-horse ma- 
 rine engine. In actual construction 
 
 Fig. 58. 
 
 there is not a material objection to 
 the use of such a form when hidden 
 from the sight, as this would be; 
 but it would be better to avoid the 
 frequent use of this figure, since it 
 does not present, in common with 
 the other ends, an equal idea of 
 security and finish. 
 
 Side rod 50-horse marine engine. Scale H in. While Upon the Subject, the eyes 
 
 and arrangements of perpendicular links, &c., may be considered. Fig. 59 
 shows a perpendicular link from an 80-horse land engine. The fitting-in 
 piece between the brasses is somewhat old-fashioned, and unnecessarily 
 strong, but simple, and to the purpose; indeed, it is far superior to 
 
56 
 
 ARCHITECTURE OF MACHINERY. 
 
 many we see at the present day. These dumb pieces used to be made of 
 oak or mahogany ; but this material should never be preferred to cast 
 iron. 
 
 Another form is represented in fig. 60 : it may be difficult to choose 
 
 Fig. 59. Fig. 60. 
 
 A 
 
 \ 
 
 O 
 
 Perpendicular and back links 80-horse land engines. Scale f inch. 
 
 between these, as the propriety will depend so much on the general 
 character of the engine. If the parts require strength for severe work, 
 the first is perhaps the most appropriate; but if the engine is highly 
 finished, the latter would be usually preferred. This circumstance must 
 always have weight in the designs of detail : it would be absurd to place 
 a light and highly finished portion in juxtaposition with a mass of black 
 metal, where the recommendation was merely weight and stability. It 
 would be equally wrong to introduce an unfinished and crude piece of 
 
ARCHITECTURE OF MACHINERY. 
 
 57 
 
 detail amongst those possessing elegant contour and workmanship. A 
 little thought on the duties which each separate part has to perform is all 
 that is necessary to the production of correct design. The figures ad- 
 joining are the " back links " of the same parallel motion, and to which 
 the previous observations have also reference. 
 
 Drag and connecting links, from the different methods of forming them, 
 present subjects for comment, and it is only 
 by gleaning from the works of engineers 
 that diversity of shape can be imagined. 
 So simple a part it would be thought could 
 only be bungled by great want of attention, 
 and yet that represented in fig. 61 is from 
 the hand of a justly esteemed maker. For 
 such a construction it is light, but for a 
 drag link it is heavy and inappropriate. 
 The most efficient means that can be adopted 
 for this purpose is explained by figures 62 
 and 63 ; and this arrangement may be used 
 in every position where a connexion has to 
 be effected, either between two pins, or between a cast iron rod and its 
 
 Fig. 62. 
 
 Fig. 63. 
 
 Drag links 50-horse marine engines. Scale 1 in. 
 
58 
 
 ARCHITECTURE OF MACHINERY. 
 
 gudgeon. They are of course made of wrought iron, being the material 
 best suited to the description of force which goes through them. The 
 finish at their heads may be either square or circular, according to the 
 taste of the engineer, or the general character of the machine. The strain 
 will certainly go through the If -bolts in fig. 61; but the remaining 
 parts being of cast metal, the link becomes necessarily clumsy. 
 
 The next class of detail that may be noticed as presenting scope for 
 good taste are the caps for rods, from whence power has to be commu- 
 nicated to other parts of the machine. These are made both in cast and 
 wrought iron : when of the first material, they must look heavy from its 
 inferior strength ; but notwithstanding this fact, they are by far the most 
 generally used. Why? is a question that would be found difficult of 
 solution; certainly not because they are cheaper, since the variation in 
 cost could only be a few shillings, an insignificant difference where beauty 
 is a desirable attainment. 
 
 Money is so earnestly coveted by some, that every other consideration 
 is swallowed up in that single thought ; propriety of form, security, and 
 even the science of their profession, is neglected in the study of how to get 
 
 the most. With these no argument can have 
 force ; no reason, however just, can convince 
 them that by due attention to the first 
 principles, not only is cost diminished in a 
 single machine, but the constant alteration 
 which is sequent on a slovenly design is 
 rendered unnecessary. But it is needless to 
 repeat these facts, since it is not for those 
 persons the present pages are written. 
 
 Fig. 64 is a cast iron cap for the piston 
 rod of an 80-horse land engine. The prin- 
 ciple of its construction is erroneous, as 
 well as its form. It is too light in one 
 place, too strong in another, and has an 
 
 Fig. 64. 
 
ARCHITECTURE OF MACHINERY. 
 
 59 
 
 Fig. 65. 
 
 appearance of neglect quite 
 unpardonable. 
 
 The cap represented by 
 fig. 65 is likewise of cast 
 iron, and for that metal ex- 
 hibits judicious proportions ; 
 the mouldings at the bottom, 
 however, should be dispensed 
 with, for they are not es- 
 sential, and look frivolous. 
 These things ought not to be 
 treated like the appendages 
 to a fire-grate or a household 
 utensil, but should possess a simple and severe outline, with nothing 
 either superfluous or merely ornamental. Besides, the step that is un- 
 avoidable between the diameters of the two adjacent parts should be 
 decreased, rather than be made more conspicuous by an attractive object. 
 
 Fig. 66 is a wrought iron cap for the same purpose, and from an 
 engine of the same power. It is not handsome, but it is light and 
 
 Fig. 66. 
 
 Fig. 67. 
 
 IK. 
 
60 
 
 ARCHITECTURE OF MACHINERY. 
 
 simple : it might be improved by the lines of the curves being kept more 
 parallel, and the angle at the base of the head being omitted. 
 
 A very appropriate contrivance for communicating a small amount of 
 power from gudgeons of considerable diameter is shown in fig. 67. The 
 cap is cast in brass, which decreases the expense of workmanship, besides 
 making a " better job." 
 
 The next subject that presents itself for consideration consists of cranks 
 and their shafts. The different amounts of tortion felt at the several 
 portions of the revolution, the length of the "journals," and the work 
 they have to perform, have nothing to do with the present investigation. 
 The only point requiring notice is, that the area of one crank shaft being 
 determined and acted upon with a known leverage, all others can be 
 determined by the rule of proportion ; the areas depending directly on 
 the quantity of power and the length of the crank. Attention to this 
 simple rule enables us to retain the proportions of a crank ; for the shorter 
 it is, the less will be the diameter of its shaft. The area of a shaft with a 
 
 Fig. 68. 
 
 Fig. 69. 
 
 80-horsc land engines. 
 
ARCHITECTURE OF MACHINERY. 
 
 61 
 
 leverage of 4 feet being 153*9, that with a leverage of 3 feet will be 
 115*4, which will be observed by reference to figs. 68 and 69. The first 
 is very correct in its form : the diameter of the shaft is Jrd greater than 
 is found by calculation, which appears essential to allow for wear and 
 sudden impulsion; the sides are straight lines joined into the circles by 
 gradual curves, and all the faces are perfectly square. The area of the 
 section at the greatest width is equal to the area of the shaft, minus the 
 difference of leverage; that of the neck is one-half. The breadth of 
 bearing upon the journal is 14 inches, which is ample for a crank of this 
 size; and many of our leading authorities consider the breadth to be 
 always sufficient, when made the same as the diameter of the shaft, for 
 engines of any power whatever. Fig. 69 is from an engine of the same 
 power as the last, but with 2 feet less stroke. The crank is precisely 
 similar in its construction. 
 
 Fig. 70 is a crank for a 45-horse land engine. The correct diameter of 
 the shaft is 9 inches ; but in this example it being made 1 1|-, the figure 
 
 Fig. 70. 
 
 Fig. 71- 
 
 Fig. 72. 
 
 45-horse land engines. 
 
 is rendered dumpy and unmechanical. The others, figs. 71 and 72, are 
 appropriate. The remaining sketches of cranks show the errors that have 
 
62 
 
 ARCHITECTURE OF MACHINERY. 
 
 crept into the calculations of engineers to a still greater extent. Figs. 73, 
 74, and 75 are of proper dimensions ; the others are faulty. 
 
 Fig. 73. 
 
 7 
 
 Fig. 74. 
 
 20-horse land engines. 
 
 Fig. 75. 
 
 Fig. 76. 
 
 10-horse. 
 
 14-horse. 
 
 Fig. 76 represents a crank from a 14-horse land engine, which broke 
 at the place marked. It was replaced by one having the sides lined up 
 to the dotted figure, which has been at constant work since 1825. 
 
 Little can be said about the form of cranks that would not be a 
 repetition of many previous observations ; but there are few, perhaps no 
 parts, that depend more upon proper dimension for their beauty. The 
 same remarks hold good with wrought iron cranks: their shape is 
 necessarily different, owing to the difficulty, if not impossibility, of forging 
 them with back ribs. The main limb is therefore a simple straight lever, 
 
ARCHITECTURE OF MACHINERY. 63 
 
 which should be kept square at the external angles, and planed bright 
 upon the face. 
 
 The principal features and details of the steam engine, as separate 
 members, have now been discussed. A few words may be said con- 
 cerning their positions, having reference to their duties performed when 
 brought into a system. 
 
 Uniform direction, compactness, and simplicity are the three necessary 
 principles to which the strictest attention must be paid before a perfect 
 engine can be designed. The directions of members ought to be straight 
 lines, running as nearly parallel to each other as the nature of their 
 movements will permit. They must go directly to their object, so that 
 their distinct use is not hidden. Cranked or bent levers, made so to 
 avoid other parts, are unsightly and vulgar, giving the appearance of 
 unexpected additions, of which a man possessing true taste would never 
 be guilty. Pipes should have no bends but which are at once seen to 
 be needed, and they should be at right angles to each other on each side 
 of the bend, not slope generally between two points, unless in rare cases, 
 when the variation will not admit of a quadrant bend ; but there is seldom 
 such a contingency that cannot be avoided by attention. 
 
 Compactness consists in avoiding all unnecessary space, or placing 
 any part at a distance from the main body, except such part be a distinct 
 machine, or worked to serve some purpose unconnected with the engine. 
 Collateral members should be fixed quite away, and with a decided space 
 intervening between them and the main body; they then become 
 separate machines, depending on their own form and proportion. It 
 is, however, quite as wrong to crowd parts together as to make them 
 straggle. Indeed, the exercise of the nicest judgment is required in their 
 arrangement. 
 
 Simplicity is by far the most pleasing quality that any machine can 
 possess. The merit which a perfect engine possesses can require no 
 superfluous decoration ; the ornaments of construction are sufficient for 
 
64 ARCHITECTURE OF MACHINERY. 
 
 beauty when designed with good taste ; and proper regard to the two 
 first principles, viz., uniform direction and compactness, are the surest 
 means by which simplicity is to be attained. 
 
 To those who aim at every reduction of cost in the formation of an 
 engine, attention to the rules of taste cannot fail to contribute to their 
 system of economy, since each feature will possess the requisite form. 
 The weight of metal will almost always be reduced ; in many instances 
 workmanship will also be less necessary; and alterations (tiresome and 
 unsatisfactory at all times) will never be found requisite. Upon these 
 grounds it is asserted that expenditure is reduced by a careful investi- 
 gation of those rules which guide the propriety of form and dimension. 
 
 To those more enlightened men whose science leads them to work 
 more with a view to improve the machine, and raise their name amongst 
 philosophers, rather than amongst money dealers, to those it is quite 
 unnecessary to enlarge upon the benefits arising from the study of these 
 laws : it will only be required to point out the true means by which the 
 end can be gained ; and some of these, it is hoped, will be found in 
 foregoing pages. 
 
 If the power of the Author had been equal to his will, no one point 
 would have been omitted, no argument neglected, that could in any way 
 have served to show his views upon the subject of proportion and taste. 
 The love of analytical inquiry into such matters goes far to render a man 
 capable of giving opinions and laying down rules : but there are other 
 requisites to make his opinions and rules valuable which are beyond 
 the reach of every one, and the chief of these are opportunity and ex- 
 perience ; the last following as a consequent upon the first, when thrown 
 in the way of those who strive for excellence. 
 
 THE END. 
 
UNIVERSITY OF CALIFORNIA LIBRARY 
 BERKELEY 
 
 THIS BOOK IS DUE ON THE LAST DATE 
 STAMPED BELOW 
 
 Books not returned on time are subject to a fine of 
 50c per volume after the third day overdue, increasing 
 to $1.00 per volume after the sixth day. Books not in 
 demand may be renewed if application is made before 
 expiration of loan period. 
 
 AUG 4 
 
 MOV 
 
 UL 10 '67 -9 AM 
 
 LOAN DEPT, 
 
 50m-7,'16 
 
01 14