The Locomotive Designer and Design

340 JOURNAL OF THE INST. OF LOCO. ENGINEERS.

THE LOCOMOTIVEDESIGNER AND DESIGN.

Paper read before the Iytitution by Mr. J. RODGERS, Member of Council, Brighton, Afiril 28th, Mlf.

PAPER No. 53. It is an unfortunate circumstance, but one which cannot be avoided, that many men pass through the drawing office an,d others remain therein without getting an opportunity of participating in the preliminary calculations and design of a new locomotive, their work being the detailing of the design after its principal dimensions have been fixed. There- fore it, is to the young members of our Institution who find themselves in this position that this paper is principally addressed. The evolutionary period of the locomotive produced a basis from which subsequent experiments could be carried out and designs evolved, thus enabling the laying down of fundamental principles for future development. The remarkable progress that the locomotive has made is ample proof that this steam prime mover was founded on sound lines, for while we have extended, enlarged, and added to the initial designs, the principle of the locomotive still remains the same as' laid down in those early days, but at the same time one is forced to the conclusion that the engineers of that day had anticipated its capacity being reached long ere this, as it is the limitations created by them that the designet of the present day has to combat, because he has to meet the ever-increasing demands of the traffic department by placing more powerful machines on the same gauge of track, through practically the same loading restrictions, and over the same bridges, which were constructed for the lighter types. To enable this to be done, much valuable information has been arrived at by experimental work, very much is waiting to be accomplished by experiment and research, and-it must be confessed that we in this country are far behind in this matter, in that we do not possess a public testing plant, such as is available at the University of

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODOERS. 24r

Illinois, America, from which valuable information is not only obtained, but is issued for the benefit of all locomotive engineers, and it is mainly due to this cause that the most scientific publications on the locomotive are American. From the great amount ,of knowledge that is obtainable by the present-day designer, it is possible to produce types suitable for all classes of work, and to be certain that they will perform the duty for which they are intended. The aim of this paper is therefore to consider very briefly some of the problems with which the designer finds himself con- fronted, the fundamcntal principles in relation to these problems, abd to some extent the laying down of the first lines of the new design. It will perhaps sccm to some extent outside the scope of this paper to say anything about the designer (and by the designer is meant the person who works out the design to the instructions of the chief mechanical engineer, the consultirig engineer, or the chief draughtsman), but as the design is to a very large extent his production, it bears the mark of his capabilities, and these being very necessary for such a responsible position, a brief outline of the training desirable will be of some value. What should his capabilities be, then? Well, he should be a locomotive engincer, not merely a draughtsman, a man who has had a thoroughly good training in the practical as well as the scicntific, and is able to combine and produce such knowledge in the design. The practical training should be as extensive as possible, and if running- shed cxpericnce can be, obtained, it will prove of inestimable value, giving as it does an insight into wear, tear, and repairs, but as many draughtsmen do not obtain running- shed work, they can and,ought to overcome this want by the most careful consideration to design, embodying efficiency with simplicity, so that examination of all parts may be carried out with the slightest inconvenience, and repairs effected without difliculty and at the minimum of cost. He should have had a thorough training in the designing of details, keeping ever before him the advantages of standardisation, but at the same time never hesitating to propose what can be proved to be a decided advantage by adopting another form of design, for, while it may not be duplicate with existing stock, it may, and often does, pay to embody a new design instead of perpetuating that whlch has proved to be faulty for the sole reason.of duplication.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 242 JOURNAL OF TEE INST. OF LOCO. ~:NGINEERS.

He should also train himself into the most careful consideration of the convenience and accessibility of all parts that come under the control of the driver and fireman, and his imagination should be such that he can obtain from the drawing, quite as good as from the engine itself, a correct realisation of parts in juxtaposition, so that there need not be any alteration due to this cause after construction. Standardisation should not only -apply to the article itself, but also to shop tools, such as templates, jigs, drop- hammer dies, etc., as a small alteration of no importance often causes the making of new tools, and this being an expensive item, due consideration should be given to it in the designing of detail work. Then again, he should be aware of the most up-to-date methods of production, and n‘ot be one who hides himself behind the excuse that it is the business of the works to produce the article according to drawing, without giving consideratibn to the most efficient method of forging, moulding, machining, etc., and who is quite content to tabulate the articles as being made of *‘ steel,” “ brass,” etc., never troubling about the tensile strength of the steel, its suitability for welding or case- hardening, or whether the brass should have a high tensile strength or simply pe suitable for brazing. Such a one has a long road to travel ere he realises the importance of up-to-date designing. There should he no dividing line between the drawing office and shops, there should be entire unison between them, and it is the draughtsman’s duty, however difficult it may be, to keep in touch with the latest shop methods 0f production, thus enabling him to design to suit the ever- increasing improvements that take place in this direction. The drawing, as issued to the shops, should be so complete that it represents the article exactly as finished, with a definite statement of the material from which it is made, SO that there will be no doubt on this point when reference has to be made in cases of failure, and under no circumstances whatever should there be any deviation from the drawing without consultat,ion between the shops and drawing oflioe, as it is only by strict adherence to this rule that the drawing will correctly represent the article, and thus save an enormous amount of expense in duplicating parts. There is another point to which the author would draw attention. All parts should be designed as “ foolproof ” as possible, so as to prevent carelessness in assembling, and thus obviate failures under this head, which are the most annoying of all failures, as they are preventable.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODOEBB. 143

In this very brief resum6 of the knowledge required by the designer perhaps enough has been said to emphasise the responsibility that attaches to him, and will allow us to proceed with the consideration of design as it appeals to a man having the foregoing training and qualifications. In laying down the first lines of a new design it is essential that sufficient information should be supplied by the authorities calling for the design of what may be termed the " controlling factors," and while it may seem, and is, unnecessary in some cases (as, for instance, we know the track in our own country is well laid, yet in some foreign countries this is far from true), it is of the greatest importance that the designer should know the exact conditions for which he is designing. These controlling factors are as follows :- Gauge of Track.--This varies from 2ft. to 5ft. 6in., the narrower gauges placing limita'tions on speeds, due to the greater tendency of the mass to overturn, owing to the centrifugal forces on curves, but it does not control the power of the locomotive that may be placed thereon. Permanent IYay.-This determines the maximum weight that is allowable per axle in tons, and is approxi- Weight of rail in lbs. per yard. mately equal to the - The 5 condition of the permanent way, how ballasted and sleepered, and the super-elevation of the rail on curves, are important when dealing with speeds. Bridges.-These structures limit the weight per foot run over total wheel base, fixed wheel base, or over buffers, which necessarily varies, owing to the strength of the structures, and is therefore of great importance. Construction Gauge.-A complete drawing of this is necessary, and should show distinctly the extreme dimensions which the locomotive may have, the designer taking care that these clearances obtain with the engine on a curve, and also with worn tyres, this affecting the clearances of the lower parts of the machine. Grades.-A profile of the road is of the greatest advantage, so that due consideration can be taken of the maximum grade, its approach, length, curves, etc., in order that the necessary resistances. may be allowed for in dealing with loads and speeds. Loads and Speeds.-The actual loads to be taken under varying conditions of grades and speeds, the time to be

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 244 JOURNAL OF THE INST. OF LOCO. ENQINEERB. taken between stations, and the class of stock it is intended to haul. Curves.-The minimum curve radius fixes the rigid wheel base. Turntables.-The smallest diameter of turntable on the road that the locomotive has to work over, especially if it is intended to limit the wheel base to this structure, care being taken that when the whole machine (engine and tender) is properly balanced an the table, the overhanging parts will have ample clearance from surrohnding buildings, etc. Fuel.-To arrive at a proper grate it is necessary to know the class and quality of fuel to be used, also the distance between fuel stations, thus enabling sufficient bunker capacity to be allowed for. Water.-This is also of the greatest importance, and a chemical analysis is valuable, in order that the most suitable design of boiler may be supplied made of material to withstand the chemical action and fitted with ample and efficient wash-out arrangements. The distance between water stations is required, so that tank capacity may be determined. In considering the design in bulk and detail it is advisable to know something of the shop and labour equip- ment where the locomotives will be stationed, as in many cases railway shops are not equipped with appliances for handling heavy work, or if the labour is sufficiently qualified to deal with, say, the treatment of steel or the repairing of the many intricate fittings, which 'are even in the most up-to- date shops a source of trouble. The capacity of a locomotive' for transporting loads is determined by the tractive force developed at the circumference of the coupled wheels. From this can be deduced the I.H.P. and the H.P. developed at the rail and drawbar, and as there are certain resistances which tend to prevent the movement of the load. these must be known, and allowed' for, so that the hauling capacity is equal to what is specified. The tractive force formula+ given in some cases as being Pdas equal to where P is the boiler pressure, d cylinder -I) diameter, s stroke, D diameter of wheel, all in inches, is only used for the comparison of differat engines, 'and as P is never attained in practice, the formula has to be modified

to read M*E.P' d2s where M.E.P. is the mean effective

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE l)l

sideration of indicator diagrams with cylinders using saturated steam it was found to lose from 1% at low speeds to IO?, at high speeds with full regulator opening. Mr. Fowler, of thc Midland Railway (Past President), in his paper on “ Superheating ” (Proc. Inst. Civil 4En- gineers, 1914),points out that from trials conducted by him on saturated and superheated engines the drop in pressure at the steam chest for the superheated engine was only about half that of the saturated engine, this difference being caused by the greater fluidity of the superheated steam.

Fig. I shows two steam chest diagrams taken from a saturated steam engine, under the conditions as stated thereon. It will be seen that the pressure is continually varying, owing to the opening and closing of the valve, the greatest drop in pressure synchronising with the largest opening of the valve. This shows the necessity for consideration in proportioning the sizes of steam pipes and chest, and in designing steam passages, so that they may be arranged in the most direct manner and have a volume sufficient to counteract as far as possible this drop in

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 246 JOURNAL OF THE INST. OF LOCO. ENGINEERS.

pressure. The other sources of loss are cylinder condensation and leakage through the valve spindle glands. In the first case, care should be taken to keep the steam passages in the cylinder away from the outside walls of the casting, if at all possible, and in the second case, when piston valves are used, inside steam admission overcomes this loss to a very large extent, as the packing is then only exposed to exhaust pressure. The steam, in leaving the steam chest and passing through the ports to the cylinder, suffers another loss through wire-drawing and condensation ; therefore the pressure of the steam is reduced considerably after leaving the boiler. Examination of the indicator diagrams (Fig. 2) show that in full gear the admission line right on to the point of cut off, is approximately equal to the initial

*LCR k.cs II0L.S STC.HC~&L? P~ES165~1s STCAM CCCST RCIiULATOI RCCULATM OPTN 100%

SCCCO 12 I* RN

pressure, but as notching up takes place with an increase of speed the port opening is curtailed, thus not allowing sufficient steam to enter the cj’linder fast enough to fill the volume that is created by the increasing speed of the piston, and therefore the point of cut-off falls, as shown on diagram B., Fig. 2. It will readily be seen that the shorter the port the greater will be the drop, and as every endeavour should be made to raise the steam line and reduce the back pressure line, the port area should be as large as possible, thus allowing the steam to enter the cylinder rapidly and, what is even of greater importance, to leave rapidly. For this reason the piston valve is superior to the ordinary flat valve, in that greater length of port is obtained. From a comparison of 35 British-designed locomotives the average length of port was found to be .065 of the cylinder area, but Henderson, in “ Locomotive Operation,” from careful investigations, demonstrates the considerable advantage

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODOERB. 247 obtained when the port length is increased from .05 to .12. The author has not been able to compare the two lengths from indicator diagrams, so Fig. 3 has been drawn up from “Henderson’s ” particulars to give some idea of how the length of port affects the point of cut-off. The full lines represent the ratio obtained with a port length of .12 and the dotted lines with port length of .05, with a clearance allowed of 8%. In comparing the shorter length of port with indicator diagrams, it was found that the curve, while not actually coinciding, was, as Henderson points out, quite reliable for general purposes.

Fig. 4 has been made from comparisons of 80 saturated steam indicator diagrams, the lines being plotted to give a Jase average to the slight variations that arise through varying conditions. It shows thk ratio of the initial and mean effective pressures to the boiler pressure for various speeds and points of cut-off. The mean effective pressures, while actually representing the engine plotted from, will probably ’dnly approximately represent another engine, as the actual steam dieribution and condition of each individual engine has much effect on the M.E.P. The line Boiler Capacity is from a report by the Master Mechanics’ Association, and gives the limiting speed at which the boiler will supply steam continuously to the cylinders without drop of gauge pressure.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 248 JOCRNAL OF THE ISYT. OF LOCO. ENGINEERS.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 TIIE LOCOMOTIYE DESIGXER ASD DESIGN-RODGERS. 249 On comparing this diagram with superheated indicator cards very little difference was found in the M.E.P., but the back pressure line was found to be lower, thus giving an advantage in favour of the superheated steam. Mr. Fowler, in his paper on “ Superheating,” referred to previously, carried out investigations on this point, and states: “ In order to show that although the expansion curve of saturated steam, partly owing to re-evaporation, does not drop, as does that of superheated steam, no decrease in the M.E.P. takes place, diagrams taken from two similar engines have been compared. The loss on the expansion curve of the superheated steam is made up by the gain in the lower exhaust and compression curves. Whenever a fair comparison could be made this has been done, with the result that there is no difference in the A1.E.P. An average of 30 diagrams gave the back pressure as 7.41bs. per square inch with saturated steam and 3.71bs. with superheated steam.” Mr. Lawford H. Fry (Member) investigated the steam action in locomotive cylinders from a series of trials conducted by Professor Goss with saturated and superheated steam in the same cylinders (see “ Engineering,” Vol. 95, page 95) and tabulated the results obtained of the M.E.P. for various conditions of pressure, speeds, and cut-off, and the author takes the privilege of reproducing the table here :- Cylinders, I6in. diameter x 24in. stroke. Clearance space, 7.6% of cylinder volume. Superheat averaged about 150 deg. Fahr.’ lloiler Mean effective pmsures in lbs. per sq. inch. pressure. Cut ofc 20 Cut off 40°!,. Cut off 60 OI0. Ibs. per sq. Satura- Super- Satura- Super- Satnra- Super- inch. ted. heated. ted. heated. ted. hented. I20 36.4 33.2 68.6 68.9 87.0 101.0 ,, I80 24.4 22.2 50.0 50.2 66.7 77.6 99 300 12.4 11.3 31.2 31.4 46.5 54.2 I 80 60 61.8 57.9 108.0 109.0 130.0 149.0 9, I 80 42.2 39.5 78.0 79.0 96.8 111.0 9, 300 22.6 21.2 47.5 48.1 64.3 73.9 240 60 90.6 86.0 154.0 155.0 179.0 199.0 ,, I80 62.3 59.2 110.0 111.0 132.0 147.0 99 300 34.2 32.4 66.4 66.7 79.6 88.4

It will be observed that at 20% cut-off the M.E.P. is slightly higher for saturated steam, at 40% they are nearly equal, and for 60% the advantage is in favour of superheated steam. There appears to be a slight difference in the results obtained by Mr. Fry from those of Mr. Fowler, but we shall

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 apparently be erring on the right side if we take the M.E.P. as being the same in both cases, and even that hypothesis proves that the engine economy is increased when using superheated steam, in that, while the volume remains the same, the weight-of steam used is less. To assist in the further study of this important question it will help if we consider an example working out the steam used per stroke and I.H.P. hour, and the means whereby the work done in the cylinders can be increased.

In indicator diagram R., Fig. 2, the particulars are as follows :-Two cylinders, 19in. dia. x 26in. stroke, wheels 6ft. gin. dia., speed 40 M.P.H. = 166 revs. per minute, cut-off 5z%, boiler pressure 16jlbs. gauge. The initial pressure from Fig. 4 is found to be 89% of boiler pressure. The cut-off pressure from Fig. 3, taking port length as .05, is found to be 6776 of initial pressure. 'Therefore Ahsoluta Initial pressure = 147lbs. gauge = 16albs. 67 x Cut-off pressure = '47 = 981bs. gauge = i13lbs. I00 Volume of steam at rut-off = 2.218 cubic feet. Weight of steam at cut-off = 2.218 x .258 = .5721bs. steam per stroke, as accounted for by the indicator. Allowing 20% for condensation, the steam used per stroke = .6a61bs. Steam used per hour = .626 x 166 x 4 x 60 = 2 4,9401bS. Steam used per I.H.P. hour = 3%= 22.glbs. I 166.8 No allowance has been made for clearance space, as it will be seen the compression tills this space to the initial pressure. From this it will be seen that three factors reduce the area of the diagram, and therefore the M.E.P., these being back pressure, condensation, and clearance volume. Back pressure, at starting, is practically non-existent, but with increasing speed and cut-off becomes very evident, due to the restricted ports and passages of the exhaust, a point that requires careful consideration in cylinder design. It may be reduced by giving exhaust clearance to the slide valve and keeping the exhaust nozzle as large as possible, consistent with good steaming of the boiler. The clearance volume includes all space between piston head and valve face when the piston is at end of its stroke,

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIONEP AND DESI&+K--RODGERs. 251 and as a rule averages 8% of the cylinder volum. This is necessary in order that the final compression pressure may approximate the initial pressure an4 also to overcome the inertia of the reciprocating parts. In ordinary valve motions, if this clearance is less than 8%, the terminal pressure becomes higher than the initiaf pressure, but there are valve gear attachments whereby the terminal pressure can be kept to its proper limit and at the same time reduce the clearance volume and cdmpression curve. Condensation is difficult to estimate, and still more difficult to overcome, and while compounding and superheating reduce this item, it is advisable in estimating the steam capacity of the boiler to allow about 20% more steam used in the cylinders than shown by indicator. For this reason it is necessary to clothe all cylinders with an efficient non-conducting material. The I.H.P. can be obtained from the standard formulae x cyl. area x stroke x 4 x revs. per min, I.H.P. = M.E.P. 12 x 33000 or for speed in miles per hour, if V = speed and D = dia. of driving wheel, M.E.P. x cyl.,dia.2 x stroke x V I.H.P. = 375 x D It has been seen that at slow speeds the M.E.P. is about 87% of the boiler pressure, and, as will be mentioned later, the internal friction of the engine is approxi- pately S%, which actually becomes a deduction from the M.E.P. It gives 80% of the boiler pressure available for calculating the tractive effort exerted at the circumference of the coupled wheels, and the T.F. formula would then 80% of boiler pressure x dZx s, mad -______. and this is obtainable D in practice. It must be remembered that rolling friction is not included in this deduction, but will be considered under Train Resistances. In dealing with two-cylinder %ompound engines, the formula must be modified, owing to the exhaust pressure of the high pres%ure cylinder becoming the initial pressute of tb low pressure cylhder, and as the ratio of the cylinders shduld be arranged so that the work done in each cylinder is equal, we can write If pe is the exhaust pressure of the H.P. cyl., P the

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 252 JOPRNAL OF THE ISST. OF LOCO. ENGINEERS.

.boiler pressure, d the'diameter of the H.P. cyl., and D the diameter of the L.P. cyl. rDZ ada Pe - = P-PJ - 4 4 and as the ratio of cylinder areas is in the proportion D2 dc = r, the expression ultimately becomes P Pe = - Tf I The above is however correct only if there is no drop in the receiver and if the piston strokes are the same. If the strokes are different, the volumes and not the areas of the cylinders should be taken. This expression gi\ 2s the initial pressure in the low pressure cylinders, but as has been already stated the maximum available pressure is 80% of boiler pressure, the tractive force for equal work in each cylinder becomes .8 x boiler press. x dia. low press. cyL2 x stroke

(Ratio of cyl. + I) x dia. of coupled wheel and for four-cylinder compounds the factor of boiler pressure becomes 1.6 instead of .8. Compound engines are arranged to start as simple engines, and the steam from the boiler to the low pressure cylinder is reduced in pressure so as to give equal total pressures in each cylinder, so that when operating thus the tractive force is 80% boiler press. x dia. H.P. cy1.2 x stroke dia. of coupled wheel It must be borne in mind that up to the present we have considered the available tractive effort at slow speeds, and have assumed that the boiler is of ample capacity to fill the cylinders with steam at every stroke, but as the speed increases, the greater is the demand on the boiler, and very soon the valve must be notched up so that the steam pressure may be maintained and generated sufficiently rapidly to keep the cylinders supplied at increasing speeds. There- fore it is advisable to know the greatest speed at which the boiler will supply the cylinders with steam at any point of cut off.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THIS LOCOMOTIVE DESIGSER AND DESIGK-RODGERS. 253

An example will make this clear. Cyls., 18in. x 26in. ; boiler pressure, rhlbs. ; cut off, 80% ; heating surface, I ,627 sq. ft. ; grate area, 23.75 sq. ft. ; volume of I cylinder, 3.44 cub. ft. The ratio of heating surface and grate area is

~- = 68.5, and with ordinary bituminous coal, the boiler 23.75 will produce about rolbs. of stoam per square foot of heating surface at boiler pressure, so that the total weight of steam generated per hour = 1,627 x 10 = 16,270lbs. The initial pressur,e=gj% of boiler pressure at 50 revs. per min. = 171lbs. The cut off pressure=gs% of initial pressure at 50 revs. per min. = 162.4lbs. Weight of steam at cut off pressure=.3941b. per cubic foot. 16,270 Therefore the supply amounts to = 688 cubic -394 x 60 feet per minute. The boiler has to supply a cylinder volume of 3.44~4 688 per revolution, so - = 51 revolutions per minut6, 3-44 x 4 which is the greatest speed at which the boiler will supply the cylinders under the conditions stated.

ADHESION. In dealing with this question two kinds of friction present themselves, viz. static and dynamic. Static friction holds good as long as the wheel revolves without slipping, but immediately the wheels begin to slide or slip, that is, revolve without advancing, the friction then becomes dynamic, thus proving that the friction which holds the wheel to the rail has been materially reduced, and is therefore greatly less than static friction, giving as a result very severe tyre wear. The rotative force at the rail should then bear some relation to the weight on the coupled wheels to ensure that this slipping will be reduced to the least possible minimum, and as the co-efficient of friction varies with the condition of the rails, being greatest when the rails are very wet or dry,. and least when frosty or greasy, the adhesion should be arranged to cover the worst conditions of rail friction.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 254 JOURNAL OF THE INST. OF LOCO. ENGINEERS.

Molesworth gives the adhesion per ton of load on coupled wheels as follows:- &-efficient of friction. When rails are very dry ... 6001bs. 0.268

9, 9, wet ... 55olbs. 0.245 Ordinary English weather.. . 4501bs. 0.2 Misty weather, rails greasy 300lbs. 0.13 Frosty or snowy weather ... 2001bs. 0.09 Forward movement then can only take plaoe if the adhesive weight and friction on the coupled wheels are greater than the maximum tractive effort. The adhesive factor is simply a comparative figure and is not to be taken as representing the co-efficient of friction between wheels and rail, this being according to conditions given. The rotative force is not uniform throughout the stroke, the maximum being about 20% greater than the tractive force formula; and as the greatest maximum effort. is at slow speeds, and assuming a co-efficient of friction of .3 I or - , slipping will not take place under ordinary 3: 33 conditions with dry rails so long as 3.33 x 1.2 tractive effort is less than adhesive weight, or if the factor of adhesion = adhesive weight is greater than 4. In high speeds, owing tractive effort to the shorter cut off, causing a greater difference than 20% over tractive effort, and also owing to the wheel pressure on the rail varying considerably in one revolution, caused by the counterbalance on the top centre reducing the weight on the rails, with a corresponding reduction of friction, there is an extra liability to slip, so to avoid this the factor for high speed engines is increased to 4.25. The reciprocating weights should be kept as light as possible consistently with strength to give a valuable reduction of counterbalance. In two-cylinder compounds, when starting simple, there is an excess piston pressure of approximately so%, so that it becomes necessary to use a lower factor for compound working to prevent slipping when working simple. One must also remember in dealing with adhesion that in deciding cylinder diameter, boiler power must be taken into account, as the desire to use all the adhesive weight by increased cylinder power generally means a very short cut off at high speeds due to insufficient boiler capacity. Table No. I gives particulars of British locomotives, and the following is a summary giving the adhesive factors

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGSER AND DESIGN-RODOERB. 255 for the various classes and types. From a comparison of American and Continental designs the adhesive factor, whilst not actually agreeing with British design, is a close approximation to it.

Adhesion Factor = Weight on Coupled Wheels +Tractive effort. Summary from Table No.. I. Tender Passenger Engines. Tender Goods Engines. Adhesion Adhesion Type. Factor. Type. Factor. 4-44 ...... 4.4 to 4.8 0-6-0 ...... 5.2 to5.4 4-4-2 ...... 4.0 to 5.2 2-69 ...... 5.0 to5.8 4-6-2 ...... 5.1 to6.5 0-8-0 ...... 4.0t05.6 2-8-0 ...... 3.7 to 5.0 Tank Passenger Tank Goods Engines ..: 4.2 to 6.7 Engines ... 4.7 to 7.0

RESISTANCES. It should be noted that, in dealing with tractive force, 80% of the boiler pressure was taken as being the maximum obtainable from the work done in the cylinders, the re- maining 20% being accounted for in the loss of steam pressure, and the internal or machine friction of the ehgine itself. This machine friction is made up of the frictional resistances caused by the piston pressure on axle journals, connecting and coupling rod bearings, crosshead slippers, piston rings and rods, slide or piston valves and spindles, and valve motion. Most of these resistances are very difficult to calculate, and the authorities differ in the way that it should be computed, some considering that speed does not enter into the sum, and look upon it as a constant quantity, while others consider that it should be taken in relation to the cylinder power. Prof. Goss experimented on the subject and from a large number of tests produced this formula, which is regardless of cut off or speed:- dia. of cyL2 x stroke = internal resistance in lbs. 3'8 dia. of wheels Henderson assumes the percentage of indicated power consumed in friction = -15 velocity in miles per hour+C,

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 256 JOUHNAL OF THE INST. OF LOCO. EXCflSEERS.

where C is a constant varying from 2 to 8, the higher figure for heavy and slow work. Wellington gives 5 to 8 per cent. of indicated tractive power as being the internal friction. These formulae will give differences and yet under certain conditions will be much the same, so that it is difficult to say which is more nearly correct, but if we take 8 per cent. of the indicated tractive power at full stroke as being equal to the internal friction we will be on the safe side for calculating the maximum tractive effort at slow speeds. The diversity of opinion regarding the exact amount to be deducted for internal friction does not alter the fact that it is there and has to be taken into consideration by giving special attention to design so that it may be reduced to a minimum. This is proved in the case of slide valves, from a series of experiments carried out by the Master Mechanics’ Association of America at Purdue University. There were four different types of valves used, all adaptable for the one engine, but it will be sufficient for this paper to give the comparisons between the ordinary unbalanced D valve and Richardson’s balanced valve.

MEAN PULL AT 100 LBS. STEAM CHES.T PRESSURE. 22in. cut off. r8hiii. cut off. 94in. cut off. Miles per hour. 10 20 40 10 20 4a 10 20 40 Unbalanced D Valve Ibs. 1118 1062 1207 1187 140 924 1322 1240 1180 Richardson Balanced Valve Ibr. 382 396 772 362 370 694 361 442 468 The analysis of these tests proved that the average friction of the unbalanced valve was about twice as great as the balanced valve and the area that should be balanced was recommended to be equal to area of exhaust port + area of two bridges+area of steam port. In British practice it is customary to balance from 50% to 60% of valve area. In tests between piston valves and balanced va,lves it was found that the frictional resistance of the former wa8 about half that of the latter, so from this pint of view as well as from that of the length of steam port the piston valve deserves very serious consideration. Axle journal friction depends on the intensity, of pressure per unit of surface of projected area (length x dia.), the material of the surfaces in contact, end to a very large extent the system of lubrication.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODGERS. 257

This unit pressure, if excessive, prevents the oil from getting between the journal and the box, especially when the oil way is on the crown of the box, with the inevitable result of “ hot boxes.” It is generally calculated as Weight carried by journal in lbs.

Projected area of journal and from investigations the following figures seem to be general practice :- Unit pressure in lbs. per sq. in. of projected area. Coupled wheels ... from 145 to 218 Average 185 Trailing ,, ... 9, 153to250 9, 181 Truck ,, ... 9; 140to ‘99 9, ‘73 Tender ,, ... ,, 2ootogro ,, 236 The desire to reduce the unit pressure by extending the bearing unequally from the point of application of load, thus causing it to be eccentrically loaded, should be avoided, for although the,average pressure may be less, the unit pressure at the shortest arm from the load will be greater than at the longest arm, and the concentration of the load is greater than wben the journal is equally loaded, thus causing heating and uneven wearing. In dealing with coupling and connecting rod, and crosshead pins, the unit pressures are much higher than those taken for axles, for the reason that the maximum pressures are not maintained for a long time at high speeds and are continually reversing in direction. Moreover, in the case of crosshead pins, the motion is very slight. The load on the connecting rod and crosshead pins is usually taken as the total piston load, and for coupling rod pins the same load divided in proportion to the number of wheels to be rotated. Unit pressure in lbs. per sq. in. of projected area. Connecting rod pins ...... 1,500t02,ooo Coupling rod pins ...’ ... 1,5ooto2,000 Crosshead gudgeon pins .. . 4,000 to 4,500 Crosshead slipper unit pressure averages about 8olbs. per sq. inch, but this figure can be increased for the bottom bars on engines that do very little backward running, since in forward running the pressure is upwards during both forward and backward stroke, thus accounting for the wear on the top bars. The low unit pressure is necessary because of the exposed position of the bars to grit.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 258 JOURNAL OF THE INST. OF LOCO. ENOIX’EERS

The piston, rod, and link motion give a frictional resistance of approximately 24% of the piston load with metallic packing, assuming that the lubrication is efficient. The weight of the piston head is relieved from the bottom of the cylinder by using a tail .rod, but the bushes carrying the tail rod wear very rapidly on the underside owing to the lubricant, when delivered from the top, being carried away by the steam. To overcome this, lubrication should take place on the bottom, which is the actual working surface. Motion pins should have ample bearipg surface and be fixed so that friction will only take place in one or the other of the connected parts. From these considerations it will be acknowledged that the system of lubrication should be given very careful attention. All moving parts should have the lubricant delivered at and distributed by means of channels from the most effective point of application, by oil boxes, cups,. or pumps, placed in a really accessible position and suitable for the class of oil used on the railway for which the engines are to work. These then constitute the internal or .machine friction of the locomotive, and as has been already stated is about 8 per cent. of the indicated tractive power. So far the resistances that have been considered concern only the machine friction, but when a lmd is moved through space, work is performed equal to the power developed, and as the load constitutes the resistance of the engine, tender, and train, the various factors constituting it must be allowed for. These may be very briefly stated as follows:- Journal Resistance.-Mr. Aspinall, in his paper on I‘ Train Resistances ” (“ Proceedings of Civil Engineers,” Vol. CXLVII.), states that axle friction per ton does not vary with the length of train, but with the weight on the boxes, and gives the following table compiled from his experiments :- Speed of Speed of Co-efficient axle. train. of Axle Feet per Miles per Friction. Friction. min. hour. P lbs. per ton. ‘57 20.0 0.01 r .65 209 26.6 0.0087 1-43 262 33.4 0.0085 1.4 3’4 40.0 0.0078 1.29 366 46.6 0.0085 ‘-4 4’9 53.3 0.01 1.65

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESION---BODOEBS. 259

For speeds up to 5 miles. it will be safe to take .I and from 5 miles to 20 miles per hour .02 as the *efficient of friction, then the axle friction in pounds per ton drawn equals p x dia. of axle x load on axle x 2,240 dia. of wheel x total weight of train drawn taking the diameter of wheel and axle in inches, the load and total weight of train in tons. 8tatting Reeietanae of these tests worked ouf at about 17lbs. per ton, end this figure is pretty generally acapted as being a good average. Wind Reeistanos.-Professor GOSS,from a series of elaborate experiments, arrived at the Mowing :- Locomotive ...... IIP First coach ...... 001 V' Second coach ...... dY' Intermediate coaches, each ... .oO01 V' Last coach ...... 00026V' V = Velocity in miles per hour. Mr. Aspinall estimates the pressure in lbs. per square foot at P=.003V'. FORMULAS FOR TRAIN RESISTANCES. R=Tractive resistance in lbs. per ton (2,240lbs.). V=Velocity of train in miles per hour. L=Length of train in feet. Authority. Value of R. Conditions. Reference. V' Clark .... 8+- Whole train ..." Railway Machin- 171 ery," 1855. V'

$9 *a. 6+- Carriages only. .. Ditto 240 V2 Deeley ... 3+- Bogie carriageswolff, " M o d e r n 290 Loco. Practice.'' do. do. Ditto

Henderson 3.36 + - American cars.. . Kent's " Mechanical 193.5 Pocket Book."

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 JOTHSAL OF THb: ISST. OF LOCO. ENGINEERS.

9 + 0.007 V‘ Carriages only ... Pettigrew, t‘ Manual v+ of Loco. Engi- 2.5+- 5 bogie coaches neering. ” 58.7 and dyna. car v+ 2.5 +- 10 do. do. 65.82 vg Aspinall, I‘ Proc. 2.5+- 15 do. do. . Ins. C.E.,” VO~. 73-05 cx LV I I. V$ 2.5+- 20 do. do. 80 v+ 2.5 + General formula 50.8 + 0.0278 L

Grade Resistance.-This is simply an inclined plane and therefore becomes equal to Total weight of engine, tender and train in Ibs. Rate of grade Curve Resistance .-The reduction of speed which generally takes place on curves generally compensates for resistances on curves, but where curves are frequent they must be allowed for. This can be obtained from the following:- Resistance in Ibs. per ton equals Gauge of rails xrigid wheel base - 2,240 x xi z x radius of curve where f=co-efficient of friction varying from .I to .27. The Master Mechanics’ Association of America recom- mends .71bs. per ton per degree of curvature for cars and 1.4lbs. for locomotives, taking ton weight as 2,oooIbs. one degree of curvature = 5,730 radius in feet Acceleration, or the force which the train offers to a gradual increase in velocity, or retardation from such velocity to a dead stop.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-EODOERS. 261

If P = the accelerating or retarding force in lbs. per ton of engine and tender, including rotary acceleration of wheels and axles, (without rotary acceleration the oo+fficients would be 75 and 102). V=velocity in miles per hour. &=distance in feet in which the acceleration or retardation takes plaqe. Then to produce a velocity V in a distance 8 VZ P=h- 8 To determine force P to produce velocity V V P=Iog- t In the variations of velocity, both for acakration, which is a positive quantity, and retardation, which is a negative quantity, the difference of the squares of the velocities must be taken. If V=smalIer velocity. v =greater ,, $hen v1 - va P=&- is or V1- 01 P=zOg- t If it is desired that the speed at the top of a grade must be = V, with an a proach speed at the foot of the grade s u, then the assP stance in lbs. per ton of train which inertia will give to essist the engine up the grade +--V =p=&)- 8 BOILER. The ever-increasing demands for greater loads and biglwr speeds is not altogether solved by cylinder capacity and wheel diameter, but in conjunction with theso must be considered perhaps what is more impprtant, the steam capacity and generating power of the boiler; more impot-

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 262 JOURNAL OF THE ISST. OF LOCO. EXGINEERS. tant because it is useless to look for high speeds with a large cylinder volume, unless the enerating unit is pro- portioned in such a way that it will %e able to maintain the steam supply necessary for the speed, and at the same time have a margin which can be drawn upon for the various accessories, such as carriage heating, etc. The boiler then is the important item of the locomotive, and while the various proportions of heating surfaces and grate vary considerably, there are sufficient data obtainable to be sure that these proportions are good. Nevertheless, the proportions that are actually the best have not been definitely decided. In deciding upon the size of boiler it is better to calculate the steam used to make certain that the boiler capacity is ample, and not merely increase the halting surfaces and grate, because the cylinder volume has been made larger. The main features in boiler design are grate, firebox heating surface and volume, tube proportions and heating surface, and one of the biggest problems in boiler design is so to arrange these that the maximum amount of heat may be obtained from the fuel burned on the grate, and given out by the heating surfaces in such an efficient manner that the necessary amount of steam may be produced from the minimum amount of coal. The area of the fire grate is determined by the amount of fuel required to be burned on each square foot of grate per hour to give the necessary evaporation, and as it is well known that the number of heat units contained in the various classes of fuel vary considerably, it is necessary to have information regarding this important item in deciding the size of grate. The various classes of fuel used on locomotives are coal, coke, wood and oil. The following table gives the calorific values and theoretical evaporative power of fuels :- CALORIFIC POWER AND THEORETICAL EVAPORATI\'E POWER OF FUELS. F.r.porat1ve Power in Ibs. of water Fuel. B.T.U. evaporated to swm from and at 212. F. Welsh anthracite ...... 15,800 16.3 American anthracite ... '3,s" '3.9 Welsh coal, good ordinary ... 15,m '5.5 9, medium quality '3,428 13.9

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN~BODGEBS. 263

Newcastle coal ...... 14.4 Lancashire coal ... .c. 14.4 Yorkshire coal ...... 13.8 to 14.9 Derbyshire coal ...... '4.3 Scotch coal ...... 13.9 t0 14.4 Coke, good quality ...... 14.0 Lignite ...... 10.0 Wood, dry ...... 8.2 ,, air dried ...... 6. I American Royal Daylight Oil 20.4 Refined Russian Petroleum 21.0 American Crude Petroleum 20.2 Coal is divided into two classes, anthracite and bitu- mbus, the heating value of the latter varying considerably. Anthracite is composed of nearly almost free carbon and gives out an intense heat, but as it burns sluggishly and forms a close mass, it requires a large grate, a thin fire, and a well distributed supply of air through the fire to produce the necessary amount of steam. Lignite also requires a large grate, since its evaporative power is small, as it is low in carbon and high in oxygen. Bituminous coals differ considerably in their compasi- tion, oalorific values, and therefore evaporative capacity, but being of good heating capacity, and as a rule burning freely, do not require such a llarge grate as the two previous classes. Coke is composkd practically of pure carbon, gives off very little smoke, burns slowly, and has a greater bulk for a given weight than coal. Wood is not a good fuel, and whun air dried it contains a large quantity of moisture, which has to be evaporated by the heat of combustion. It contains very little available hydrogen and its calorific value is low, being when dry about 8,000 B.T.U. and air dried 5,960 B.T.U.; it therefore requires a large and deep grate. oil does not require a grate as it is burned in the firebox, being injected in the form of a spray, after being atomised by a jet of steam, the burners being placed below the fire door and directed in such a manner that the spray strikes below the brick arch. As it contains a large quantity of hydrogen it has a higher calorific value than coal and gives out an intense heat and is therefore suitable for rapid

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 264 JOURNAL OF THE ISST. OF LOCO. ESGINEERS. generation of steam ; the evaporation of water from and at 212' F. is 12 to 13lbs., when burned at the rate of Ialbs. per square foot of heating surface per hour. This rate of consumption, when compared with an average quality of coal burned at the maximum rate and using the same boiler, gives 25% more steam, so therefore weight for weight less oil need be carried. The early experience in using oil showed that it proved very severe on the firebox, which had to be renewed about every two years, and as the cost of repairs and renewal would probably be more than the difference in price between oil and coal per ton mile, the only advantage to be gained is in oil producing districts where oil is much cheaper than coal, the absence of smoke and sparks, and also the easier firing of the locomotive. Experiments on oil burning are being carried out at the present time, so probably in the near future many difficulties will be overcome and oil be used much more than has been the case hitherto. The supply of air to support combustion must be delivered above the fire to effect the combustion of the fuel gases, and below the fire to effect cornbustion of the solid carbonaceous portions on the grate. The draught axid the distance between the grate bars both largely affect the quantity of air admitted, and care should be taken that the amount furnished is neither deficient nor excessive, and that it should be introduced in the right place. The weight of air required to effect complete combustion is at least 50% in excess of that theoretically required. This forms a guide when fixing firebar spaces and ashpan openings. The human element must be taken into consideration when deciding the length of grate, as grates over 8ft. oh. long are very difficult to keep properly covered, and with a large area, such as is adopted on some of the large American engines, it is beyond a fireman's endurance to keep the fire in proper trim, so that mechanical stoking is being adopted for grates of abnormal size. The following figures show the wide variation in ratios .of grate to evaporative heating surface, compiled from T,able No. I of proportions of British locomotives. The author found it impossible to make comparisons of this ratio with American and Continental designs, as the quality of fuel was very seldom given. Whether the ratio was adopted proportionally to the heating surface or to suit the quality of fuel used, is not apparent.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGSEE AND DESIGX-R0J)GERY. 265

As bituminous fuel is in universal use in this country the ratios given in Table No. I form a good guide in design. Total evaporative heating surface + grate area= for passenger engines 45.3 to 79.1 ; for goods engines 51.1 to 108. For anthracite and low grade coals the ratio will require to be much less to give the necessary evaporation. To obtain the maximum delivery *from the boiler the fuel must be burned at the greatest possible rate, measured as the amount of coal fired per square foot of grate, and giving an evaporation in lbs. of water per square foot of heating surface or per lb. of coal. Professor Goss, in his Purdue experiments, burned I8olbs. of coal per square foot of grate, the evaporation being 12lbs. of water per square foot of heating surface, or 144lbs. from and at ziz0 F., with a grate ratio of I/,,,; and Mr. Pettigrew, of the Furness Railway, with Welsh coal, obtained an evaporation of 9.231bS. of water per square foot of heating surface from and at 212' F., or 11.6lbs. of water per lb. of coal, with a rate of firing of 50.1lbs., and I a grate ratio of __ 55.3 The rate of firing thus varies from 50 to 2oolbs. for bituminous coals, the latter amount being fired only for very short periods, and when the engine is working on heavy grades or at high speeds, and certainly with little economy. A rate of aoolbs. may therefore be taken as the maximum firing, and in average service the rate may probably not be over Ioolbs. Fig. 5 gives an approximate diagram of boiler performance of ordinary design and is self-explanatory. Heating Surfaces.-It is essential to have sufficient heating surface so that the maximum heat units will be con- veyed to the water from the gases as they pass on their way to the atmosphere. Firebox heating surface gives the highest evaporative efficiency, and therefore the depth over the grate should be ample, so that there is sufficient room for the development of radiant heat and for the combustion of the gases. before they enter the tubes. Bituminous coal gives off a very much larger quantity of combustible gases than anthracite, so that the firebox depth should be greater. The volume should be approximately 6 cubic feet per square foot of grate, with ample air inlets for a rate of combustion of about 140lbs. per square foot of grate per hour.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 FfC. 5.

APPROXIMATE DIAGRAM OF BOILER ERFORMANCd FOR LOCO.BO/L€RS OF ORDINARY DES/GN

CALORlFfC VALUE OF COAL = /4500 6.XU. a /5L8S OF WATER FRON & AT 2/zoE

0 POUNDS OF &YATER FROM $I AT %/2 F

PER LB. Of COAL.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 TEE LOCOMOTIVE DESIGNER AND DESION-BODQEBS. 267 Tube Heating 8u7face.-The ratio of tube heating surface to firebox heating surface varies very largely, as will be seen from Table No. I, and therefore it is di&ult to say what is the best proportion. The vacuum in the smokebox, which averages from 3 to 7 inches of water, is employed in overcoming the reGist- ince of air in passing through the fuel bed and in drawing the gases through the tubes, therefore the length of the tubes will depend very largely on the draught, and sbld be suWient to absorb the heat of the fuel gases. When tbe tubes are short in length the gases pass through them before the heat has become absorbed by the available surfaces, thus causing a loss of fuel, and as long as the temperature of the gases at the smokebox is greater than the temperature of the steam in the boiler there will be a transference of heat. The spaces between the tubes should be sufficient to allow of free circulation of water and the rising of steam from their surfaces. The tubes should be placed as far as possibk above the grate so that solid particles may not be drawn through them. This is also prevented by the assistance of, the brick arch, which at the same time deflects the gases, causing fhem to go forward instead of directly into the tubes, thus giving more time for their combustion in the chamber formed by the arch. The ratid of firebox to evaporative tube heating surface will be found on Table No. I: The variations are very marked, being as follows:- Total evaporative tube heating surface ifirebox heating surface=for passenger engines 8.8 to 14.1 ; and for goodo engines 8.3 to 17.9. The heating surface is measured on the water side ot both firebox and tubes. The estimated I.H.P. of a simple engine is frequently worked out from particulars arrived at by Prof. hss, who averaged the consumption of steam as z81bs. per horse-power per hour, and an evaporation of 12lbs. per square foot of bating surface per hour, in which case the heating surface required for one I.H.P. = #%=2+ square feet. It is difficult to estimate the saving from compounding as SO much depends on the condition of the enghe, but Hendierson estimates that one I.H.P. requires 2 square feet of heating surface. Approximately for one 1.H.P.-

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 268 JOURNAL OF THE INST. OF WCO. ENGINEERS.

Compound locomotives ...... 2 sq. ft. heating surface. Simple locomotives, cut off under half stroke ...... 2* 9, 9, 9, Simple locomotives, cut off + to

2 stroke ...... 1.- 2% 9, 9, ,, Simple locomotives, cut off full stroke ...... 3 9, 9, 9, A superheater gives a saving in water 2nd fue!, but it must be remembered that if placed in the boiler, it takes away a certain percentage of tube heating surface, so that in making comparisons in Table No. I this point must be borne in mind, and also that the ratios are calculated on evaporative heating surface only, that is, neglecting superheater surface. Feed water heating gives a saving in fuel and is a great factor in increasing boiler efficiency. Therefore, wherever possible, some form of feed heating should be seriously considered when working out a design. One has only to mention the blowing off at the safety valves when standing at stations to realise the loss in energy that occurs and which could quite well be used in heating the feed water. TYPE. Space does not permit of consideration of the various reasons, apparent and otherwise, for the adoption of the many types now in existence, so this question.wil1 be dealt with in a general manner, but at the same time it must be borne in mind that when proposing a design, the designer ,must take into consideration all the factors as stated in the opening paragraphs of this paper, and on these the ultimate decision of type should based. Any tendency to follow .the lead of a new type just introduced should be carefully held in check, as it is well known that while a certain type does good work on one railway its performance is inferior on another. The engineer’s restrictions regarding loads are generally :- Maximum load on rail per axle.

9, per foot run over fixed wheel base. 11 9, 9, total ,, 9, 9, ,, buffers. The type then to a very large extent is decided by the bad restrictions, adhesion weight which is determined from

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIQNER AKD DESIGN-BODOEBS. 269 the tractive effort, the minimum curve to be traversed, and the coal and water capacity. This will be best explained by referring to the GlasgQw and South Western goods engine, 0-6-0 type, Table No. I. The adhesion factor is 5.4 and the maximum weight per axle is ao tons weight per foot run over wheel base, mini- tpum curve 5 chains. It will be observed that this engine is built up to 'the maximum limit of weight, the total weight is available for adhesion, and that the wheel base is sufficiently flexible to allow it to pass round the minimum curve easily. In an engine of this type the disposition of the wheels should be such that the loads per axle are nearly equal. If the total weight of engine is more than is allowable by load restrictions or hecessary for adhesion, then carrying wheels require to be added, and as the total wheel base will be longer, the fixed wheel base will require to be reduced and redisposed, so that the minimum curve will be traversable with the minimum side play of carrying wheels, Sufficient weight must also be allooated thereon. The greater the number of coupled wheels on a fixed wheel base the closer must they be pitched so as to allow tr-aversing on curves ; and to give the necessary freedom for such, thin or flangeless tyres may have to be adopted on the middle pairs of wheels. This closer grouping of the wheels gives an increased weight per fwt run. Ten and twelve coupled engines have been built to over- take the larger loads, but in adopting these types one has to give particular attention to the lateral play in wheels, axles, and coupling rods necessary for such inflexible wheel bases, and it is for this reason as well as that ofethe maintenance costs of these parts that the twelve coupled engine has the maximum number of coupled wheels that can be adopted on a fixed wheel base. Fig. g gives a good idea of what happens on a five chain curve with an eight-coupled engine. The position of the engine on the curve is taking full advantage of the lateral play between tyres and rails, axleboxes and guides and wheels, with rails laid exaFt to gauge (neglecting the practice sometimes adopted of laying rails wide on curves), thus ahwing for any errors that occur in laying curves to. the specified radius. The sketch also takes into account the check rail and crossing, the reason being to show the necessity of spacing

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 270 JOC'RSAL OF THE INST. OF LOCO. ENGINEERS. the middle coupled wheels wider and giving them thin flanges, or making certain of them flangeless.

The questions of " spacing," " thin or flangeless tyres," are all debateable points and depend greatly on the curvature of the roads over which the engine works. In the case of a truck being used the centre of oscillation should be such that the truck wheels will do their proper work in guiding, thus saving flange wear on the coupled wheels. The location of the truck centre, therefore, is of importance, and is generally arrived at by the following con- structions and formula, all three of which give the same result :-

FIG, 6.

On railways where the rail loading is low, curves frequent, and a high powered engine is required, where it is

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODGERS. 271 not possible to place a rigid wheel base engine, then the articulated _type deserves serious consideration. There arc four types at present in vogue,. namely,

" Fairlie," '' Mallet," " Kitson Meyer " and " Garratt," and while they all differ materially in design, yet the principle of a flexible wheel base is common to all, giving a double engine as one unit, obtaining a long wheel base with tb required low rail loads and weight per foot run over wheel base. These consid*erationsof type have dealt with goods and mixed traffic engines, where tractive effort was the main consideration and speed only secondary, but in dealing with fast passenger stock speed is of vital importance and therefore has to be carefully ansidered. The following are a few of the types in use at the present time :- Tender engines. Tank engines. 0-4-2 0-4-4 2-4-0 4-4-2 4-44 4-4-4 4-44 46-2 4-6-2 4-6-4 4-6-4 It wilt be noted that the 0-4-2 type tender engine is the only one without guiding wheels, the 2-4-0 the only one with a single pair of guiding wheels, and the 0-4-4 tank engine has no guiding wheels in front, but a four-wheeled bogie at the rear. Mr. Stroudley, of the L.B. and S.C. Railway, designed the 0-4-2 tender engine for fast passenger traftic, and claimed that by placing the coupled wheels in fmt, where the greater weight was, the centre 05 gravity is kept well forward, thus giving steadiness and stability not only on the straight but also in negotiating cufves. Moreover, in having a small carrying wheel behind, a larger grate was obtainable and the cast iron weight at the back of the engine could be dispensed with. These engines have proved themselves equal to the work for which they were designed, and do guod work at the present time without any abnormal flange wear on any of the wheels. The 2-4-0 type has done good work for many years on the L. and N.W. Railway, so the question arises as to when it is necessary to adopt a truck.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 272 JOURXAL OF THE ISST. OF LOCO. I:SGISEEIIS.

The main considerations appear to be speed, curvature of road, and weight that truck requires to carry. When a fixed yheel base engine enters a curve the leading wheels, be they truck or coupled, must take up the guiding of the mass, which owing to natural laws will tend to run off at a tangent to the curve. This guiding is done by the wheel base making a series of chords round the curve, and it will be at once apparent that if the leading wheel is of the fixed wheel base, it will be driven hard on to the outer rail, thus causing excessive rail friction and tyre wear, and the longer the wheel base the greater will be the friction. The wheels of the leading two-wheeled truck must be placed in front of the cylinders and sufficient weight allocated to them to give steady running. From the construction already given the radiating centre is found to be behind the wheels, thus giving the effect of being pushed from a point well behind the contact of wheels with rail. At the same time it is free to swing on this centre, which allows the axle to radiate to the centre of the curve. As there is only one pair of wheels the pushing action from behind has a tendency to form a tangent to the curve instead of a chord, and therefore cause the leading coupled wheels to do a good deal of the guiding. A four-wheeled truck has a definite wheel base and can be located under the cylinders, thus carrying a greater load which gives it stability. When running forward it is also being pushed through the pivot, which is located in the middle or within a few inches of the middle of its wheel base, and being able to swing on this pivot the guiding action is in the form of a chord, which relieves the coupled wheels and gives an easier movement of the whole mass round the curve. Referring to Table No. I, several of the 4-4-0 type engines have on their coupled wheels practically the maximum weight allowable per axle, this adhesion being ample for cylinders 20 dia. x 26 stroke with boiler capacity sufficient for high speeds, and having ample weight on the bogie to ensure steady running. In engines of this type the driving wheel is arranged as close to the firebox as possible so as to obtain connecting rod length, and as the fixed wheel base is determined by the coupling rod length, the maximum now being Ioft., it decides the length and depth of firebox. In a 10 feet wheek base it is possible to drop the firebox between the axles

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIQN~BODGEBS. 273

and obtain a grate area of 23 sq. feet, with a deep box, but if an increased grate is necessary, the firebox has to be sloped over the trailing axle to obtain the necessary length, and the depth is reduced thereby.

NG. 7. -7-

In fixing the boiler diameter, due consideration must be given to wheel diameter and the height of the centre of gravity. Fig. 7 shows the relative position of boiler and wbeels, and if it is necessary to increase the boiler diameter, it must be raised, or the wheel diameter must be reduced. The centre of gravity should be at such a height as to ensure stability against the centrifugal forces tending to overturn the machine normal to the curve. If the resultant of the cenJrifuga1 force and the weight falls even on the rail the engine will overturn, and therefore the resultant should fall inside the rail. The 4-4-2 ,type is an extension of the four-coupled type, which allows of larger grate a- and boiler capacity. Since the coupled wheels are in front of the firebox the cylinders should be outside the frames to ensure a long connecting rod. The firebox can be arranged to extend over the main frames, and owing to the weight carrying wheel being mall in diameter a good depth of firebox is obtained. This type, however, whilst having the advantage of

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 274 JOCHSAL OF THE ISST. OF LOCO. KNGISEERS. larger grate and boiler, has also the disadvantage of not being able to utilise any more weight for adhesion, so that if the cylinders were made large enough to take full advantage of the boiler oapacity there would be trouble with slipping at starting. It therefore seems reasonable, when designing a high powered engine, to make use of the extra weight which the trailing wheels are carrying, by making a 4-6-0 type, for although the placing of the trailing wheel under the firebox curtails grate and heating surface, it is probably compen- sated for by the increased tractive effort obtainable. In countries where the fuel used is of poor quality the grate must be larger to obtain the necessary evaporation, and this increase of weight on the trailing end very often means the addition of eitheyz or 4 carrying wheels, thus giving the 4-6-2 or 4-6-4 types. The foregoing remarks deal with tender engines and are based on the assumption that sufficient water and coal capacity would not be available if placed on the engine itself, owing to the maximum weight limits being exceeded, but in a great many instances tank engines can be designed to cover the distances between water and coal stations, and if such can be arranged, a tank engine should be preferred, owing to the lower initial cost and maintenance. This might be instanced by reference to the 4-6-4 tank engines of the London, Brighton and South Coast Railway, designed and constructed by L. Billinton, Esq., at the Brighton Works, for working the non-stop trains, London to Brighton and London to Porhmouth. These engines take 300 tons without any wavering, and maintain a fairly constant speed throughout the journey, without having to rush on down grades to make up for dropping speeds on up grades. They are also remarkable for their steadiness in running at all speeds. To apply these considefitions, it is now necessary that a general arrangement of the locomotive should be pro- ceeded with, in order that the design may be worked out to have the specified capacity, also to prove that the type which has been chosen IS suitable as far as weight is concerned, by calculating the total weight and disposing the wheels so that the distribution on the axles may be what is r.equired and not exceed the restrictions laid down by .the engineers. The question of weight is of great importance not only for axle loading, but also in that details may be made as

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOOOYOTIVE DESIGNER AND DESIGp-RODGERS. 275 light as possible consistent with strength, thus curtailing as far as possible all weight that utilises engine power, but which could with advantage be used in the more vital parts of the engine. Much scope lies open to the designer in this field of the work, and one has only to instance the saving in counterbalance weight that accrues from scientifically designed reciprocating and revolving weights, combined with the use of the highest grade quality of alloy steels, to realise that every pound in weight saved here means that one is enabled to place on the rails a more powerful engine for the same weight. Of course dead weight cannot be avoided in some cases, such as the 4-4-0 tender engine, which generally requires a cast iron balance behind the firebox to obtain sufficient' load on the trailhg axle, but if the type allows of it this should be avoided as far as possible, and the weight adjustment on the axles arrived at by the re-disposition of the wheels, boiler, tanks, etc. There are two methods of arriving at the %eight distribution, one being by calculating the weight and centre of gravity of each individual part, taking their moment from a fixed position, thus arrivihg finally at a common centre of gravity, when the total live weights can then be distributed. This is a long process, but it is far better to spend a month arriving at a correct result than to have the engine built by guess work and then find out, when it is too late to alter them, that the axle loadings are not what are desired. The second method is by comparison, with the nearest existing type, of which there is a correct record of axle loadings, when it is then only necessary to calculate the weights of parts which are different on the two types, or parts that have different centres of gravity from the common line. If records of details are kept a very great deal of labour is saved. Fig. 8 explains the method. The whole subject has of necessity been treated very briefly, but the author's aim has simply been to give the main outlines of the subject. Appended are nine examples of modern locomotive design, Figs. 10-18.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 -/,-o JOCHSAL OF THE ISST. OF LOCO. KNGISEERS.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIQN-WDOEBS. 277

a n 1

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 278 JOCHSAL OF THE ISST. OF LOCO. KNGISEERS.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN--BODGEES. 279

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 I -T~PE C-6-2 JUPERHEATED L 6 LJC R - FIG.14.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 -NPE. 2-64, SUPERHEATED. c.c.a - I FIG.15.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 284 JOCHSAL OF THE ISST. OF LOCO. KNGISEERS.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODQEBB. 28 j

DISCUSS 10N. The Chairman: Gentlemen,-I am very interested in seeing SO many junior members here this afternoon. I am sure I should like to have had the benefit of hearing a paper such as we have had this afternoon, when I was going through the works, particularly before I got into the drawing office. The author has put before us in this paper some very valuable information, and the paper I think acts as a guide to all junior engineers, who want to have an idea of the directions in which they should study, particularly if they are thinking of getting into the drawing office, and going on seriously either in the designing of locomotives or of at- taining to any position in which they will have charge of locomotives, such as district superintendent or an officer of a railway company. The author has practically treated of most of the subjects with which we have to deal in the locomotive profession, and I will not occupy your time at present. We should like to hear anyope who has any remarks to make. Mr. J. Clayton (S.E. and C. Ry., Ashford): I shouid like to have heard some of the junior members, for.whom I think the paper was chiefly intended, rather than those who have had riper experience, open the discussion ; at th6 same time, I would like to make a few observations on one or two points. I must say that the paper is full of good things; too full, I think, for us to do justice to it in the short time we can give this afternoon. I should think it is equal to two or three of the average papers rolled into one, and I am sorry it was not divided into three. I am sure Mr. Rodgers has been to an enormous amount of labour. Most of the points Mr. Rodgers has dealt with were points usually kept quite secret by the man whose business was the design of the locomotive “ in the little office over there in the corner,” and we used to wonder what calculations were made by him there, and how we should like to get hold of his calculation book, and just take a few notes from it.. That has been my experience and I expect it is not unknown to others. I think it is very thoughtful indeed of Mr. Rodgers to put this subject before us as fie has done. I do think though that every youth in the drawing office and in the shops, while he jmay not have the chance to enter into the actual design of the locomotive, may in his own interest

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 286 JOURXAL OF THE INST. OF LOCO. ENGINEERS. find out from those, with whom he comes in contact in the works and in the classes, and from the books which are now available how to design the locomotive himself. I was just going to remark that one of the best places to get experience is the running shed. Unfortunately all cannot have experience in the running shed, which is so valuable, and one wishes it could be made compulsory for all drawing office juniors to be given six months in the running shed. Failing that, however, go to the scrap heap. It is the most instructive place in any works, and when you go there, look for something. Pick up a bit here and a bit there and see how it is worn, and where the failures occur. Mr. Rodgers refers to standards. .It is good that we should not make a fetish of standards. They help expedi- tious work in the drawing office and shops and are quite good up to a point, but it is necessary that we should not be tied down to them. I worked under a well-known locomotive engineer in this country, who used to say, “No standards for him !”, for if there are no standards everyone will always endeavour to find out something better. We require some standards of fundamental parts which have been tried and found to be right; such things as pins, Cotters, bolts, etc. Those are standards which can really be applied, but the most useful standards you can have in the designing department are the records of locomotives existing. Just to give you one example. When designing the cab or footplate, the best thing to do first of all is to turn up the record made in Jhe office of the relation of all the principal handles, heights of the footplate, seats, hand- rails, firedoors, tender coal plate, etc. As you design other engines put down the fresh particulars as a guide, and such records of locomotives existing on that particular line or of good practice on other lines, will be found extremely handy for reference and great time savers. With regard to the points which the designer requires to know in the first instance about the road Over which the engine will run, the most important, I think myself, is the question of the bridges. It is nearly always bridges that decide its weight and distribution. You can generally strengthen up the road, but you cannot strengthen bridges readily. They are nearly always the limiting factor of the size of locomotive you can put on your road. One particular line in this country, which I know well, is entirely bound up with this question of bridges. For years and years it ran a good large type of locomotive, which could not be employed on an important section because there was on

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCCNfOTIVE DESIGNEIt ASD I>ESIC;S-RODGERY; 287 that particular portion one bridge which it would have cost an enormous amount of money to strengthen. The engl- Deer should keep the locomotive department posted with the particulars regarding his bridge curve, which is a diagram of the span and safe loads per span on that line, SO that the locomotive engineer may know to what extent an increase in the loads per axle is possible. Strangely enough, it is not often the big bridges that are the trouble, but the short span bridges. Another point is that the engineer usually, when he is considering bridge strength, assumes two locomotives put together, so that you may get a big concentration of load on a short span. Notwithstanding the author's remarks on the subject, I think it is generally conceded that slight wiredrawing is a good thing in order to get a drop between the boiler and the cylinders, and so assist the steam ,to find its way readily through the superheater elements to the cylinders. The authof remarked that engine economy is increased when using superheated steam. I would like to ask if he did not mean boiler economy. Perhaps, however, I do not .quite grasp his point. It occurred to me that it was boiler economy rather than engine economy. I notice the author bases the steam production on heating surface. I think the better way is to base it on grate area. Heating surface is such a will-o'-the-wisp. You may have plenty of it on paper, but a great deal may not be of much use; but the grate decides largely your ability to consume coal in order that you may get heat units out of it. The next thing is to consider the heating surface in its relation to the grate, and then the number of pounds of coal burned per square foot of grate area, say 80 to 100, and the calculation is quite simple. To base steam production on heating surface alone is, I submit, a questionable procedure. With regard to the adhesive factor, you notice that this varies from something like 3.7 up to 7. Well, a reliable figure used to be 4 to I for passenger work and 5 to 6 for tank engines. The author did not state it, but it is interesting to note the reason why tank engines have a higher factor is that the water is not a constant quantity. With regard to eccentric loading of bearings, that is a very good point indeed, but is rarely seen.in good practice, but out of centre loading of crosshead shoes is not uncom- mon. That is, the crosshead pin is not in the centre of the length of the slide block. I have one railway in mind which has m'any engines with this feature. It would be

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 288 JOVHSAL OF THE ISST. OF LOCO. ESGINEERY. instructive to look at the slide blocks. They must wear more at one end than the other. With regard to lubrioation, one of the saving features of the locomotive which is always a wonder to every other class of engineer is that it runs so well with such small bearing surfaces. It is due, as the author points out, to the reversal of thrust. This change of direction of pressure at each end of the stroke enables the oil to work in and save the situation. There is iust one other point about six-coupled engines. hZr. Rodgers mentioned that in a 0-6-0engine it was better to have the loads equal on each axle. My experience has shown that it is far better to have the weight on the trailing axle ,mnsiderably less than on the other axles by as much as two tons, for the simple reason that when the engine is pulling at the drawbar the locomotive is pulled down there by the reaction, and it will generally result, if you do not have the weight on the trailing axle less &an that on ,the other axles, in what the shed foreman calls “ the engine being down on its springs,” and the remedy is to keep the trailing axle distinctly lighter by 2 to 24 tons. I am very much obliged to you for listening so patiently to my remarks. Mr. J. P. Maitland (L.B. and S.C. Ry., Brighton): I am sure that all of us have been very pleased with the very able paper which Mr. Rodgers has presented to us this afternoon. One thing that arrested my attention was that Mr. Rodgers laid before us so many of the fundamental formulae for locomotive design without burdening US with details which are useful only to those responsible for special parts of locomotive construction, and I am sure that, to many, his paper will be very instructive for reference. Mr. Rodgers spoke about the need of a thorough understanding existing between the shops and the drawing office. I am sure that it is impossible to emphasise this too fully, but I wish to go a little further. I am of the opinion that there is great need in many cases for the existence of a closer and more cordial understanding between the drawing office and the running department. The latter departmerrt is one which has come to the fore more, perhaps, during the last ten years than was previously the case, and has seemed to have materialised as a buffer between the locomotive and traffic departments. The running department is brought into contact with all the essential problems of traffic mani- pulation, and, in addition, those in authority in it have had

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODGERS. 2% practical experience of actual construction and maintenance of locomotives, and are thoroughly in touch with the -working of the line, not only as an engineering concern, but also as a railway in its primary sense, viz., a means of transportation of passengers and freight. I know that in a good many cases mistakes have been avoided, and many great improvements have been initiated and elaborated by a conference between members of the running and drawing offices, whilst the design of a new type of engin4 was in preparation. Some time ago there was much controversy in railway circles over the topic of “ watertight compartments ” ; we certainly do not want watertight compartments in one department, and for that reason I greatly weloome Mr. Rodgers’ suggestion.

Mr. Rodgers mentioned outside ” conditions which Qccur in the actual working of the engine, with special reference to the necessity of bearing in mind, when designing the engine, the class of shed to which the engines would be allocated. It has been my experience that it is preferable to station engines of similar classes, fittings, or details, as much together as possible. I am convinced that even at the risk of losing a little in other directions, it will be found better to arrange thus, and only in the most exceptional cases to distribute the engines in isolated units. Otherwise, it will be found necessary to maintain at many depots a large stock of special spare parts to wver any possible failure. With regard to the tests and researches conducted in America into the science of the locomotive, we must admit that they have been of very great value to the locomotive world, e.g., the tests of locomotive boilers to destruction, which have given us some of the best data which it was possible to obtain. It would however appear that despite all these experiments, the Americans do not seem to have produced a more efficient type of engine than has been constructed over here, although we have simply prooeeded On data based almost entirely on actual practice. I believe that it is an accepted fact that the British built engine is more efficient than any foreign production, which has shown that our system can be justified, and is due, without doubt, to the very variable conditions under which a locomotive operates, which it is a practical impossibility to ascertain accurately by any purely theoretical method. Mr. G. Harben (London): I should like to add to the last speaker’s remarks, and thank Mr. Rodgers for a very interesting paper. Mr. Rodgers sugests that running shed

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 290 JOUHSAL OF THE ISST. OF LOCO. I

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESI6N-RODGEBS. 291 the running shed doing that work which he will afterwards have to design. Mr. A. R. Bennett (London): I have very little to say, because the paper is too important to discuss offhand, but I should like to remark that I think Mr. Rodgers’ contribu- tion will form one of the numbers of the Proceedings which will be most in demand hereafter, for it is a real compeo- dium of information in respect of the design of locomotives. In view of a recent discussion, I should like to ask Mr. Rodgers-he refers to the case of oil burning destroying furnaces in two years-were these fireboxes of copper or steel? Mr. Rodgers: Steel. Mr. Bennett: In this country? Mr. Rodgers : American boxes. Mr. Bennett: Then perhaps we shall have the pleasure of hearing Mr. Sanderson say something on the subject. Mr. Rodgers remarks that it is well known that a certain type of locomotive may do good work on one railway and yet perform in a very inferior manner on another. Is that absolutely ascertained to be the fact? Because, in the United States, the custom is, I believe, for the large locomotive builders to design an engine which will perform a specified amount of work under specified circumstances and supply such engines to roads of corresponding character all over the country, with, I understand, very acceptable result,. Now, if that is thecase in America, and it is the reasonable case, surely engines of similar weight, cylinder power, and heating surface ought to do practically the same work on one line in this country as on another, supposing the conditions are not very different indeed. On our main lines curves are not very different or difficult and the effect of gravity is the same on all. Why, therefore, it should be alleged, as it is, that 0 separate design is necessary for each line I cannot understand. When, a few years ago, companies interchanged engines experimentally, I don’t think such a result was brought out, and I know that in one or two cases engines on a foreign line did better than the home locomotives. And now the war has brought about such a general interchange of engines the alleged effect ought to be very apparent indeed, but is it? All companies have heavy sections, of course, but my view is that the engine designed for such ought to &a1 as effectively with similar grades or curves elsewhere.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 292 JOCRSAL OF THE; IXST. OF LOCO. F:SGINEEIIY.

There is such a lot in-the paper that it is impossible to follow the author from point to point; and, in that respect, Mr. Rodgers differs from the author of a little book that I read many years ago. It was published in America in the ’~o’s,and the subject was, “The Designing of Locomo- tives.” The writer went to the point, for he said, supposing an engine has cylinders of a certain diameter, then ekery other portion of the engine may be ascertained directly by computation from that diameter, and he gave a list of factors. If you wanted to ascertain the area of the grate or the diameter of the chimney, of the exhaust nozzle, or anything, you took a figure which he gave and used it as a multiplier or divisor on the cylinder diameter, and there you were. Very much simpler than Mr. Rodgers; but still, on the whole, I think Mr. Rodgers’ paper will attain a greater reputation than that very practical author’s. I do not know whether Mr. Sanderson will remember the book? It consisted of only 30 or 40 pages, but with it and a ready reckoner a consulting engineer could get out the specification of a locomotive in a quarter of an hour. 1 remember that the heating surface provided was very much less than was usual with British locomotives of the period. Mr. Smith Mannering (L.B. and S.C. Ry., Brighton): I shall not attempt to offer any criticism on the paper, as it is quite beyond the scope of my own particular present calling, and I feel any such attempt on my part would be altogether out of place. I should like, however, to make an allusion to the term “ standardisation,” of which Mr. Rodgers speaks, and which should always be kept in view by the locomotive engineer and his staff in preparing speci- fications and drawings for new engines and renewals to existing ones. I consider one of the greatest drawbacks to a system of standardisation on a railway concern is the unavoidable and periodical change of personnel. Mr. Rodgers says that to perpetuate any particular design Qf p-art or parts for the sake of duplication, when such parts are out of date and troublesome in various ways, is bad practice. With this reasoning we can all agree, but there is the other side, viz., pulling things to pieces, and intro- ducing new methods and designs where the existing ones have proved quite satisfactory and trustworthy. A systeni and a certain type of bogie truck, for example, is in vogue for a time, which gives every satisfaction to its designers and originators, but a change of personnel alters all this, and introduces a different design altogether, and for what purpose? The same thing applies to all parts of the locomotive from the boiler to the cylinders, valve gear, and other

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 ‘rm LOCOMOTIVE DESIGKER AND DESIGN-ROIMERS. 293 parts. One could go on indefinitely in this line of argument and make “ standardisation ” a by word. I am gladof the opportunity of saying a few words in support of the principles set forth in Mr. Rodgers’ most instructive paper, which should prove of very real value to those who are aspiring to become (like Mr. Rodgers and other members) leading lights in the designing of locomotive engines. Mr. Bennett made an inquiry just now about oil fuel. We tried oil fuel on the Brighton Railway some years ago in the copper fireboxes of some of the locomotives, and it gave a certain measure of success, but it was discarded on account of the damage done to the copper plates. Mr. Rodgers states there is no grate required in a firebox burning oil. This may be true in his experience, but .ell the engines on the Brighton Railway so fitted retained their grates, which were earticularly useful for raising steam 5n the sheds. I was engaged on one occasion on some trials with a special kind of burner, in which a copper spiral was introduced in the body, of the burner. This had tho effect of gasifying the oil instead of atomising it. The trials were satisfactory so far, but nothing further came from them. I was assisted at the time by Mr. Lovick Johnson, now in India, who read a paper at this Institution on oil fuel, and based, I believe, on these particular trials.” Mr. Bennett: When I made that inquiry, I had in mind Mr. Urquhart’s experience in Russia with oil burning, and I do not think he found it severe on the fireboxes there, and that is why I asked whether they were of copper or steel. Mr. Rodg8rs: Mr. Urquhart’s experience was not the same as that in America. Mr. V. T. Barn- (Graduate, L. and S.W. Ry., East- leigh): I should like to mention the question of apprentices going through the running sheds as part of their training. Frequently when an.apprentice starts in the locomotive works he goes through several of the departments, but it is not ;very often that the opportunity is given him to go into the running shed. When he gets into the erecting shop, it Seems to be the chief object to keep him there in the form of cheap labour. I know that in some works it is quite 3 usual thing for apprentices to remain in the erecting shop

‘I Liquid Fuel ” by F. S. Lovick Johnson. Proceedings, 191I.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 294 JOL‘HSAL OF THE ISST. OF LOCO. KSGIXEERS. for two or three years. 1 think that the fault lies more with the locomotive superintendent than with the apprentice in connection with this matter. It is mostaessential that the draughtsman should have running shed experience, and I think that if the earlier draughtsmen had had part of their training there the British locomotive would not be such an awkward machine to work on as it is to-day. One of the worst features are the splashers fitted by same railway companies to their engines. I do not think that it is possible for a fitter to work in a more awkward or dirty position than between a splasher and the wheel, and yet it is quite a common feature to have feed-pipes and brake pipes running round inside the splasher. Other companies, on engines with outside cylinders, more especially on four- cylinder locomotives, place the greater part of a set of motion up inside the splasher, where a good blow cannot be got upon any of the motion pins. The latest engines of the G.W.R., L.S.W.R., and Caledonian Railway, with raised footplates, show that the splashers are being done away with, and it will not be much loss when they arc entirely eliminated. I should like to hear further reference made by the author to a valve gear attachment to prevent excessive compression. The usual method adopted in this country to prevent this is not by an attachment to the valve gear, but by means of valves fitted to the back and front ends of the cylinders, with ;L spring loaded to about 51bs. per sq. in. above the working pressure of the boiler. Mr. Clayton: There was one question I should just like to ask, viz. : The length of the steam ports. Do I understand correctly that the author means the length of the steam ports or the cross sectional area? Mr. Rodgers: The length of the steam port. Mr. Clayton: Well, I have recently had one interesting experience in this respect with regard to back pressure. FVe altered an engine recently which had not been thought as good as it should be. It was taken in hand and altered as far as possible without much change in the valve gear, and just by increasing the length of the valve travel about an inch, we made the engine what the driver said was “ Simply all right.” It made a very great improvement in the power developed. I should say the increased horst- power developed on average working was something like IOO in 800, just by increasing the length of the valve travel about one inch. Simply by varying the length of the rocker

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 TIIE LOCOMOTIVE DESIGNEN ASD DI;SIGS--RODGEHS. 295 arms, using the old rocker shaft, in order to give an increased valve travel, at the same time increasing the lap of the valve in order to keep the steam opening all right; it gave us the advantage chiefly on the exhaust side and gave the full port opening for a longer period. That is one way in which the back pressure can be very much reduced. With regard to clearance volume, it is not easy now- adays to get this down to 8 per cent. with piston valves. I think you will find it averages about 10 per cent. Get the port as short and as straight as you will, it is not easy to obtain the 8 per cent. clearance. I should like to know what are these valve gear attachments, to which the author refers, for altering the compression space, Le., the clearance volume at the end of the stroke, by the valve gear. Mr. Rodgers: Purely American and unknown in this country so far as I am aware. I think it is the Peabody valve gear attachment to which I was particularly referring, which gives a less compression curve and fattens up the whole of the indicator diagram itself. Mr. W. Vaughan (L. and S.W. Ky., Strawberry Hill): Mr. Chairman,-One can hardly allow the expressed aspira- tions of the younger members present of obtaining running shed practice to pass unnoticed, and we quite enter into their feelings that a past privilege is not now apparently open, for reasons which I am unable to say, but as the desire exists, one would recommend an appeal, to obtain what appears essential experience for the future generation of locomotive engineers. Speaking upon the subject simi- larly, a young man said to me, “ We cannot gain in theory as in practice,” which emphasises somewhat the desire mentioned. With regard to the author’s most interesting paper, one appreciates the broadness of its scope for assisting those who are delving into the problems of locomotive engineering, and it speaks volumes for the care taken in furnishing the data as presented, and we readily agree with the advantage that must necessarily be gained from an efficient co-ordination between the shops and the running departments to enable the traffic problems to be met. Mr. R. P. C. Sanderson (Vice-president, London): There are just one or two points to which I should like to refer, if I .may be allowed. The author has spoken of the desirability of running

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 296 30URSAL OF THE INST. OF LOCO. ENGISEERS. ghed experience. Other speakers have mentioned it rather feelingly. I would like to add my testimony most stronglv with regard to that. Every man that has to do with the designing of engines should have running shed experience, even if he never works in any other part of the shop. When it comes to a question of other shop knowledge, I believe that, to make a man a thoroughly equipped designer, he should have worked in every department from the foundry up. We do not find that to be practical, however, as it takes too much of a man’s lifetime to educate himself to do any useful work in the drawing office, so a general practice on the other side on the larger railways and in the locomotive manufacturing shops is to have certain of the draughtsmen, besides their running shed experience, trained in the different departments, and the work is specialised. The leading designer takes hold of the general arrangement. The castings are put in charge of a man who has a thorough knowledge of foundry work and core-making. The forgings are worked out by a man who has a thorough knowledge of dies, die forging and work of that kind, and the most economical way to design so as to utilise stock. The boiler work is put in the charge of a draughtsman who has spent a great deal of his time working in and supervising work in the boiler shop and knows how that work should be designed and laid out. The general draughtsman, who is to have charge of the whole design, embodies these in his general arrangement and details. In that way the special knowledge required by the draughtsmen to make a really first-class detail design is acquired without having one man to know it all. Experience in the running shed is of even more importance than anything else, because if the mechanical railway official is held responsible for his cost and for his maximum earnings per engine per hour, the question of the engine being able to keep in service the longest number of hours ahd minutes in the twenty-four hours becomes vital, and every minute lost in the running shed taking out a wheel cover to get at a wash+ut plug and putting it back is time lost in train service. Every minute lost in removing an injector to remove a delivery tube and to put a new tubt in again is money lost from the train service. It is that particular experience which is of such great importance in the earnings of the railroad as compared with the total amount of money invested, and I entirely endorse the idea that every man who has to work in a drawing office afterwards ought to spend a certain amount of his time in the round-house, barking his knuckles, trying to get at inacassi- ble parts.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AXD DEYIOhT-RODC+I.:nY. 297 With regard to the fireboxes mentioned in oil burning engines lasting only two years, I have seen some cases, woqse than that, but in each case the boilers were working with hard artesian water coming from some thousand or so feet under the surface, from very deep wells in the oil districts of California and Arizona ; the water was terribly hard and the sheets burned out on account of the excessive scale formation. I do not think, if the firebox is properly designed with the minimum amount of double metal in the seams and with the braces and stays made of the smallest safe proportions, that the oil will be very much more severe than good coal. The heat is more concentrated and the radiant heat forms a greater proportion of the total. As. mentioned once before, I have seen the heads of crown bolts melt with the oil heat (when the engines were transferred from the divisions where they had been successfully running on coal), until the volume of the head was so reduced that the water could keep the balance of the metal at the proper temperature, and then the injury stopped. In the old drawing office of Mr. William Mason, of Taunton, Mass. (the Mason engines at one time were cele- brated in the United States in the running sheds, because, if the superintendent called for an engine in a hurry, the round-house force always ran to Mason engines first; they were get-at-able in every point; you never had to take anything down to get at something else; you could always reach it with a monkey wrench; never had to look for a socket wrench or special tool to get out anything) on his office wall, over the windows, he had a large sign in large letters, and that was the only rule he kid down, and it was this, " MAKE EVERYTHING GET-AT-ABLE. " That was his principal instruction to his draughtsmen. The Chairman : In Mr. Mannering's remarks he referred to the lighting up of the oil burners in an oil burning locomotive and suggested that it was necessary to have a coal fire going the whole time in order to be able to start the oil fire. I think it will be found that this is not necessary. At the present time, e.g., thgre are running at the Deptford Loading Wharf, locomotives made by the Hunslet Engine Co., of Leeds, which doubtless members in London interested in oil burning could get permission to see. In these the fireboxes and ashpans are lined with firebrick, and after; the engines have once been started in the morning all that is done to re-start the oil burner after a stop is to throw a piece of oily waste into the firebox and then turn on the oil injector when the oil fire will start at once. There

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 298 JOCRS.4L OF T€lE ISST. OF LOCO. ESGINEERS. is thus no need to keep a cod fire going all the time, and to do so appears unnecessary. For starting up from the cold, special arrangements haxe to be made, and connections can be provided for taking steam for warming the oil and working the blower, etc., from a neighbouring engine in steam or ;I stationary boiler. Mr. Mannering: R'e had to keep a very thin fire. The Chairman: I am not sure of the exact details as to the starting of the fire from cold in the Deptford engines referred to, but I thought the matter might be interesting to anyone thinking of using oil fires, and no doubt The Hunslet Engine Co. would furnish information with reference to the Deptford Engines. Several speakers have given great encouragement to the younger men to profit in every way possible by experience during their apprenticeship, and suggested in particular the value to their practical training of experience in the running sheds. I think every one of us will thoroughly endorse that, and would add the strong recommendation to them to get some actual running experience on the footplate if it is at all possible. We quite know the difficulties there are in the way of obtaining this experience as part of an apprenticeship or pupilage. To a certain extent this difficulty has further been created by young men in the past not having appreciated the great value of the experience to be gained by working in the running sheds, and shirking the dirty work there when they have been given opportunity of running shed experiencel but it appears that it is to the detriment of the future of railways themselves that locomotive supefintendents or chief mechanical enginers should not afford an opportunity for running shed experience to those of their apprentices who show themselves promising, and from whom they will be likely to recruit their future officers. If one might venture a suggestion to those locomotive superintendents or chief mechanical engineers who have done such a lot to encourage their apprentices or pupils, it would be to offer as a prize to those showing themselves proficient in their technical classes and in the works, the opportunity for say at least three months in the running sheds and at least one month on the footplate (even if only in shunting yards). That I am sure would be, very much more valuable to the apprentices in their future professional life than any monetary award, or than any offer of facilities for further study at technical schools. It cannot I think be too strongly impressed on the rising engineers that these opportunities, however, will only be available to

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 those who show themselves likely to appreciate and profit by them. Where a hard and fast programme is laid down for all pupils and apprentices one so often sees the effect where a young fellow gets as part af his training (without any effort on his part) a certain time in the drawing office ; he simply becomes a tracer through never having been com- pelled to fit him2elf for anything better before he was allowed to enter the drawing office. With reference to the many points covered by the author in his paper, they are so many that it is impossible to deal with them in detail. There are, however, one or two matters upon which one might touch. With reference to the thinning of the tyres on centre coupled wheels the usual practice I think in this country is to keep the inside faces of all tyres the same distance apart and to thin the flange of the centre coupled wheels where necessary by turning off as much as is required from the outside face of the flange. If, as the author pointed out, an end view is made of the engine when standing on the curve showing the leading and trailing coupled tyres coinciding, and the centre thinned tyre thrown over as much as the clearance of the axles and axleboxes will allow, then by showing the rail and guard rails in juxtaposition on the same drawing it will show the clearances very clearly. Some experiences lately have led one to doubt the correct- ness of the usual practice, and a consideration of such drawing would seem to suggest that it is doubtful whether the thinning- of the centre tyre flange should not be made by turning OK as much as is necessary from both the inside and outside of the flanges, or alternatively to push out the centre coupled wheels so as to increase the distance between the tyres, at the same time turning off the outside face of the flange so as to make the flange thinner. I do nut think the author has had time, through covering so much in his paper, to do justice to many such useful hints. Mr. Clayton referred to the value of study of the scrap heap, and one can quite endorse the value of such study ; in particular this applies specially to the value of the study of scrapped boilers and their fireboxes, etc. Mr. Sandeison referred to the practice larely followed in Amerioa. which seems a very sound one, of making draughtsmen specialise in one department and selecting them from the most promising men in this special department. I cannot recall any case in my knowledge where anyone from the ranks of a boiler shop from which a foreman or assistant foreman boilermaker would be selected was ever made a

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 300 JOUBNAL OF THE ISHT. OF LOCO. EP;GISI.;ERY draughtsman. Were such opportunities of advancement open to men in the boiler department, it would induce promising apprentices to kpecialise in boiler making, and thus tend generally to improve the work in this, the most important, department of locomotive construction. Through lack of the necessary capable young man to choose from in the boiler department, the drawing office has often suffered from lack of actual practical boilermaker’s experience. Mr. Clayton also referred to standardisation ; he rather seemed to deprecate the idea of too rigid standards, though at the same time he appeared to support the principle by referring to the advantage of the practice which he followed of tabulating the heights of all cab fittings and such similar dimensions and adhering to them when getting out a new design. Mr. Bennett dealt with the question of the detail dimensions of the locomotive being determined when you have once settled on the tractive effort required. I think it will be in your recollection tbat our late Vice-president, Mr. Burnett, at one of his last attendances at the Institution here gave us a number of factors for determining such dimensions, starting with the cylinder size. These he said were for the benefit of the younger members, and he suggested that they should compare the figures which he gave with the practice on the lines on which they were engaged to see if the figures tallied with those which he had quoted. These figures, Mr. Burnett explained, he head not originated by calculation, but had simply compiled by averaging the corresponding dimensions of numbers of locomotives of good design, which had passed through his hands in the course of his professional experiedce. There seems no doubt that such a tabulated statement is a wonderful help in the design of a locomotive, and any national attempt to produce a standard locomotive would start with the consideration of such tabulated statement of existing designs. One does not want to adhere slavishly to a standard to the exclusion of any progress, but from experience with the benefits of having adopted standard engines for India, it appears that you cannot too highly value standardisation. Thus you can standardise a boiler without slavishly insisting on the injector or mountings, etc., on that boiler being exactly the same as on the previous boiler of the same class, but you can retain the flanged plate of that boiler the Same as the last class, and make the next class again with the same flanged plates. Unless it can be shown that there is something seriously wrong with the boiler, the alteration of a small amount in the radius of a flange or the width of a firebox or similar

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODGERS. 301 dimensions, which would mean new fanging blocks and another stock of spare flanged plates, should be strongly resisted. Alteration for alteration's sake has been a costly thing for railway companies. Unfortunately in the past locomotive engineering suffered too much from an entirely false tradition, viz., that instead of its being an ideal to adhere to existing designs, if at all possible, it was regarded on the contrary as proof of a locomotive superintendent's ability, if in designing a new engine he originated entirely new patterns so that the engine appeared to have individuality of its own, and it became quite the practice to refer to it as his engine. The 'effect of this tradition was for many years disastrous for railways, since each successive locomotive superintendent considered that it showed his individuality, and was a proof to the directors and the public of his ability if he promptly altered every special detail on the locomotives designed by his predecessors. Unfor- tunately many directors, being without intimate knowledge as locomotive engineers, grew up in this tradition and actually encouraged these introductions of new designs as proof that their locomotive superintendent thereby showed knowledge in advance of previous or contemporary locomotive superintendents. Had it been established as the tradition that no alteration from existing types should be. introduced without very definite proof of the advantages to be gained therefrom, many thousands of pounds would have been saved to the railways. It appars impossible now, unless a very broad and far-seeing view is taken of the matter by all directors, to introduce standard locomotives on all the British lines, as has been done in India, unless the railways are nationalised. The advantage of such adoption of standards for the whole country would be incalculable in the saving of duplication of spare parts and in facilitating and cheapening construction, particularly for the smaller lines. On the larger systems much can be done to standardise, and there is no doubt this is being done at the present day to an extent unthought of say 20 years ago. [t is found possible to get a number of locomotives with aarts which can duplicate. Connecting and coupling rod brasses. axleboxes and brasses, etc., etc., and even complete boilers and their mountings can be made to duplicate for all engines of the same power that have to run on the line. Mr. Maitland, speaking as a running man, was instancing that very very clearly ; he very strongly supported the idea of standardisation when he pointed out the advantage derived through having the details all duplicated, even where there were only two or three locomotives of the same

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 302 JOURNAL OF THE INST. OF LOCO. ENGINEERS. type at the same station. If, therefore, on our main railways, all the locomothes on the whole line could be reduced to about five or six types for which standard parts could be retained at eTery running shed, the gain in manufacturing costs land upkeep would be very great. In addition to the direct gain in reduced costs standardisation carries with it many great advantages that are only appreciated after experience with standards-reduction of storage, smaller number of dnawings, templates, patterns, and much less clerical work. Lastly-and this is a considerable item on a busy railway-every man gets familiar with the details of a few standard engines in a way which was impossible when every engine differed, consequently there is a considerable saving of time to all concerned in explaining commands to or receiving reports from subordinates. Mr. Harben referred to a railway, in the running shed of which he has known, when an engine has come in, that they have had to cut them to pieces in order to get the lifting chains on them. There seems to me something wrong on that railway ! The locomotive superintendent should be advised by the running department, and it is they to whom he should look for recommendations as to what is required in any new locomotive to be designed. They know what has failed, they know what is required, and it is they who should say where failures occurred in previous engines of the class and where consequently altera- tions are required. If, therefore the running department on a railway gets a second lot of engines in succession, in which they cannot apply the lifting appliances, then I should certainly blame that running department. It seems that any railway should be so organised that the running department should at once record any such difficulty as that, and that any item having once been noted, e.g., “ lifting facilities,” that item should never be dropped out of the list of details to receixe the consideration of the running department when adlising on the specification for a new design of engine. It seems that in this respect there was some lack of system on thc rtailway system referred to. Mr. Rodgers gave some figures for the pressures on the bearings. Any such list of figures as this given by the author is useful as being a means to enable one to check the designs and practice with which they have to deal from time to time. It is astonishing how anyone in the drawing office, who goes through the different designs of engines, even of standarci engines, which have become accepted and are doing quite good work and appear generally satisfactory, suddenly comes across curious anomalieq, in that the ,engine

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGXER AND DESIGN-RODQEBS. 303 in one particular design does not conform at all to the practice in the re* of the engines of which one has the figures before him. The value of these tables is that it at once calls attention to that detail, and when a new engine is to be designed, that item can be reconsidered and if thought desirable it can be modified. Mr. G. F. Burtt : I should like to express m apprecia- tion of the paper presented by Mr. Rodgers. think the Institution is much indebted to him, for it has Ientailed an enormous amount of work, and will prove of exceptional value to many students of locomotive engineering. Covering as it does such a large field of points in locomotive design a discussion might well be prolonged over several meetings. With regard to locomotive design and running shed experience, when I commenced to serve my apprenticeship nearly 30 years ago, I started in the fitting shops and running sheds at New Cross, living at that time in London, and was there for nearly three years, and a place like that was an ideal one for a young man to gain experience. An apprentice, after he had been there for about twelve months, got a lot of very important work. I remember on one occasion a fellow apprentice and I had to stop one night to take out the pistons of one of the little " A " class engines. Those engines at that time were kept in a round house, and the pits were consequently very short. It so happened that, on one side, the piston was right at the end of the stroke inside the cylinder, and we could not draw it out. The engine being pinched forward, after it had gone so far, brought the buffers up against the shed wall, so we could not get it' out that way. We pinched the engine back. It was then off the pit, and we could not remove the piston then. The only thing to do was to'take down the connecting rod and force the piston right out. When we had taken the connecting rod down the little end brass was broken. In this particular design of connecting rod a strap was fitted with the brass in two pieces, one half being circular, and the half square on the outside, but the semi-circular piece had broken in halves. The job was done with regard to putting in new piston rings, b'ut the trouble came in putting this rod back, and we spent the whole of the night in getting up this connecting rod and trying to get in three pieces of brass where there shd have been two. I came to the conclusion then that if I had had at any time to design a locomotive it would most decidedly have outsidel cylinders. Regarding locomotive design, I have a firm opinion that

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 304 JOURNAL OF THE INST. OF LOCO. ENGINEERS. simplicity and “ get-at-ableness ” should be one of the first considerations of a locomotive engineer. A deal may be said in regard to inside and outside cylinders, motion, etc., but from the running shed view, in the saving of labour, an engine with outside gear has much in its favour. It must be admitted that a lot of time can be saved by taking out valves, pistons, and motion down from such an engine. The idea of symmetry of outline and “ personal ” appearance is very nice, but I think it should be a secondary consideration. For instance, many modern tank engines have the cab side sheets and tanks all in one piece riveted up. What a lot of labour could be saved if they were separate, as in the case of a tank having to come off to get at a few firebox stays. There are also reversing rods, etc., neatly tucked away behind firebox clothing. The Author, in reply to the discussion, said :-Running shed experience has entered very largely into the discussion of the designer’s qualifications, and rightly so, as there is no doubt that very much indeed-can be done in the designing room to facilitate examination and repairs, by arranging that all parts shall be accessible and removable with the minimum of inconvenience, and it is with the full knowledge of such an important necessity that I thought fit to draw attention to such points which have been very greatly emphasised by the various speakers. Mr. Sanderson’s information in reference to the system of specialisation in the American drawing office is extremely interesting, but if a draughtsman is to be a fully qualified engineer, he should be able to design a cylinder quite as well as a boiler, or lay out a motion as correctly as arranging a grate, and so have an intimate knowledge of the design and construction of the whole machine. Life may be too short to obtain complete mastery of each individual branch, but with that close -ordination between drawing bffice and shops, which should exist, there seems to be no reason why the draughtsman should not be a good all-round man and therefore able to produce an efficient design at a minimum of cost. In fact this is what occurs in offices with a small staff, but even in the largest offices there is the difficulty of keeping the specialist fully employed on that branch in which he is an adept, and therefore when transferred to another branch he would not be earning his full salary. Mr. Maitland remarks that unison should not only exist between drawing office and shops, but should also extend to the running department; this is quite right, and in fact it should apply to all departments, in order that failures,

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DEBIGNER AND DESIGN-RODGEBS. 305 inconveniences, and suggested improvements may be notified to th.e drawing office, to enable the design to be corrected and brought up to date. Standardisation has received much attention this afternoon, and whilst I deprecate standardising for its own sake, since this, as Mr. Clayton points out, has a tendency to impede progress, yet when parts have been well tried and found efficient, it is of the greatest advantage to standardise such, not only for the sake of interchangeability, which it. such an important item in the running shed and repair shop, but also with benefit to stores stock; therefore I consider the designer should give this point very careful consideration. In my remarks on this matter, I dealt simply with thz designer standardising for a particular railway, the very much larger question of adopting standards for all railways, such as has been done by the Standards Committee! for India, opens up a wide field of discussion, and so is worthy of a paper on its own account. Mr. Clayton asks whether, when dealing with superheated steam and M.E.P., it should not be boiler economy instead of engine economy that is increased. In superheating a greater saving takes place in water than in fuel, so therefore more heat units are utilised per pound of steam, and as the evaporative rate of the boiler is reduced when the superheater is of the flue tube type, and has the effect of reducing the total heating surface, therefore boiler efficiency is de- creased, and superheating gives an increased engine economy. In reference to Mr. Clayton’s remarks on steam production, I mentioned that the rate of combustion is measured as the amount of coal fired per square foot of grate, and the evaporation is measured in Ibs. of water per square foot of heating surface or per Ib. of coal. Heating surface is to some extent a quantity that cannot be guaranteed, but it seems reasonable to measure water evaporation by it, seeing that it is the means whereby the heat units are transferred to the water in the boiler. Here again the designer should see that the very best design of heating surface is obtained, so that the maximum of efficiency is approached. In dealing with fuel combustion alone it is perhaps bettlr to measure pounds of fuel burned per square foot of grate per hour, because this is directly concerned with the burning of the fuel without relation to evaporative efficiency- Mr. Clayton draws attention to the adhesive factor for

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 306 JOURNAL OF THE INST. OF LOCO. ENGINEERS. tank engines, and his explanation is the reason why F separated tank from tender engines in summary of Table No. I. In reference to the weights of the 0-6-0 engine, it certainly is a slight advantage to have the load on the trailing axle somewhat lighter than on leading and driving axles, owing to the downward tendency of pull through the drawbar, but to avoid serious overloading special attention should be given to height and general design of draw gear. Mr. Mannering raises the question of grates for oil fuel, and it seems from his remarks, that in the experiments in which he assisted, oil fuel was used in conjunction with coal, in which case a thin layer of coal is spread over the grate, land the ashpan dampers are left sufficiently open to ensure the necessary draught. Oil cax ae used alone by covering the grate with firebrick, and when starting up from cold it is necessary to connect up the oil burner, when using steam, to a stationary boiler which is already in steam, until there is sufficient steam generated by the engine itself to keep the burners going. Engines, which have to stand idle for some time whilst in steam, may have the fuel supply cut down or even cut off altogether, as there is generally sufficient steam in the boiler to start the buiners when required. Mr. Barnes and Mr. Clayton refer to cylinder volume and cornpression. The cylinder clearance, which has been stated is approximately 8 per cent., can be reduced to about 3 per cent. by using a valve gear attachment, such as the Allfree gear. This has an eccentric movement, controlled by the crosshead, thus giving a variable speed of travel to the valve, so that the opening and closing of the exhaust port is delayed at all points of cut off, thus giving a late compression with a reduced volume at compression due to the delayed closing of the exhaust port. This then fattens up the indicator diagram because the expansion is increased and the negative work of compression reduced. The spring loaded valve, to which Mr. Barnes refers, is merely for compression and hydraulic relief, in case the terminal pressure becomes greater than a pre-determined pressure of about 51bs. over the boiler pressure. In reference to Mr. Bennett's remarks on designing from tabulated data in relation to cylinder diameter, such a method is only useful for comparative purposes, such as the ratio of length of cylinder ports to diameter of cylinder,

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 THE LOCOMOTIVE DESIGNER AND DESIGN-RODGERS. 307 ar heating surface to grate, but when you come to deal with say the strength or bearing surface of a crank pin, then such a method is only an approximation, as obviously maw points arise, such as the material and its tensile strength, so that it is b>ter to work out such items from actual conditions than to depend on such text book rules. Mr. Bennett also mentions the question of engines of the same type not being suitable on different railways. I agree entirely with Mr. Bennett, in that engines of the same type and power will perform the same work, but always providing that the conditions of road and service are the same, and it is this point that I had in mind when reminding the designer to consider all the factors as laid down before deciding upon a type. Take for instance a case of two railways with a part of their road of equal lengths, and the loads to be hauled the same, but in the first case the road is fairly straight and abounding in grades, and in the second case the road has severe curves, but is fairly level. Now for the hilly and straight road, an engine with moderate wheel diameter, a high factor of adhesion, with the elimination of all unpro- ductive weight bv adopting the smallest number of guiding wheels possible would be necessary, but for the level road abounding in curves the guiding truck is indispnsable, and since the severe grades are absent a large wheel diameter can be employed with a low facfor of adhesion. In reference to the spacing and thinning of tyres, Mr. Lelean is quite correct in saying that the general practice in this country is to keep the distance between all tyres the same, there being, I beIieve, only about four exceptions to this in which the distance between the thin flanges is increased by tin. to *in. Which is tKe best method is con- troversial, as much depends on the frequency and curve radius of the road, and whether the middle wheels should share the guiding, thus relieving flange wear on the truck, and leading coupled wheels. Fig. g was made purposely to draw attention to the necessity of considering these points on curves and crossings, taking into account the guard nail. Mr. Lelean offers a suggestion to chief mechanical engineers, that apprentices showing themselves efficient at work and classes should be granted the experience of the running department. This is being carried out by the London, Brighton and South Coast Railway Co., wherein an appren- tioe who is capable in the workshop and attains the highest position in the third year course of mechanical or elect.rica1 engineering obtains at the consent of the chief mechanical

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016 308 JOURNAL OF THE INST. OF LOCO. ENGINEERS. engineer the privileges of a pupil, which give experience in all departments, including the running department and drawing office. I thank you all very much for your kind remarks about the paper and the criticisms which you hqve given. The paper is purely elementary, it makes no pretence of being advanced, but the aim has been to let the younger men in our Institution know something of the knowledge that a designer requires if he desires to become a locomotive engineer. Mr. A. R. Bennett: I think we should place on record a vote of thanks to Mr. Rodgers for his very excellent paper. Mr. Maitland: I have pleasure in seconding that proposal. The Chairman put the vote of thanks to the meeting, and it was duly carried.

Downloaded from jil.sagepub.com at The University of Melbourne Libraries on June 5, 2016