Journal of the Institution of Locomotive Engineers Volume 8 (1918)

Main file

Journal No. 25/26

Rea, J.W.H. (Paper No. 55)
Some notes on the working of superheater engines with special reference to lubrication. 3-10. Disc.: 11-60.

Journal No. 27/28

Jones, A.D. (Paper No. 58)
British railways and the War: Address by the President. 78-87.
A very general presentation of Britain's railways contribution to the War effort during WW1. The most significant item of mechanical engineering was the assembly of ambulance trains by several individual companies. Locomotives were loaned to the Government. The movement of troops and munitions to the departure points for the battle fronts required a major effort. Leave trains for troops also contributed to inferior services for former customers who were forced to meet fare increase. Cross Channel steamers were greatly affected and the murder of Captain Fryatt, master of the GER vessel Brussels is given emphasis. .

Sanderson, R.P.C. (Paper No. 59)
American practice in the draught appliances of locomotives. 90-108. Disc.: 108-30.

Journal No. 29/30

Kelway-Bamber, H. (Paper No. 60)
Coal and mineral traffic on the railways of the United Kingdom. 135-53; Disc.: 154-67.  235-51.
Had 45-ton bogie wagons been used to haul the Admiralty coal traffic during WW1, about 25% less coal would have been consumed by the locomotives hauling the trains, and the reduction in dead weight hauled would have been 33%.

Webster, H.C.  (Paper No. 61)
The arrangement of enginemen's duties. 171-7. Disc.: 177-92.

Journal No. 31/32

Gresley, H.N. [Paper No. 62]
Inaugural address [Chairman of Committee of the Leeds Centre]. 199-214.
From a historical standpoint this is highly interesting as Gresley made observations about Webb's compounds. He noted that the Greater Britain 2-2-2-2 Queen Empress was known by the footplate staff as the Scarlet Runner and that the white Queen Empress aged like a meerschaum pipe. He noted that Webb's experiments with friction wheels led to showers of sparks and wheel flats. He emphasised the need for large boilers noting the size of that to be introduced with the K3 class. He also observed that some of Ivatt's boilers on the large Atlantics had been in service for sixteen years and had still be scrapped. He could not see any advantage in Belpaire fireboxes as it was possible to direct stay round-top fireboxes. He favoured the use of outside cylinders with Walscchaerts valve gear as these were simpler to oil. He also favoured mechanical lubricators and was an advocate of standardisation in principle, and without question the standardisation of wagons. He complained that three quarters of the heat generated in the firebox was wasted up the chimney. He noted the need for feed water heaters: live steam heaters were used in marine applications, and there was a need to develop exhaust steam heaters. There was a need for better grates. He also commented on articulated rolling stock.

Journal No. 33/34

Carlier, S. (Paper No. 63)
Heating of trains and the problem of coal saving. 255-67. Disc.: 267-92.

Journal No. 35/36

Paterson, W. (Paper No. 64)
District supervision of the locomotive department of a British railway. 296-352.

Holcroft, H. (Paper No. 65)
Three-cylinder locomotives. 355-68. Disc.: 368-95; 476-91.
"Probably no bcttcr example of successful design can be taken than the 8 ft. single wheelers formerly running on the Great Northern Railway. The first of these appeared early in the history of that railway, and for nearly forty years, defied all competition in thc running of express trains, the type being perpetuated with scarcely any alterations in detail until recent times. They gave excellent results, ran at the highest speeds, and no locomotives could have been simpler, although the single wheelers on the Midland Railway and other railways ran them closely" (opening page). "They [the Stars] exerted considerable influence, and their lead is due to the fact that they appeared at the proper moment in the development of the locomotive, and also because definite mechanical advantages were cheaply bought with the minimum of added parts." Holcroft then surveyed British three-cylinder designs with three sets of valve gear: the GCR and NER designs where "the mechanical advantanges gained are fully recognised, but the presence of three complete valve gears neutralises them in the eyes of many." He then mentioned the Gresley three-cylinder 2-8-0 and followed this by a detailed description of Holcroft's own conjugated valve gear and the advantages to be gained by three-cylinder propulsion: better starting acceleration, better balancing and elimination of hammer blow, better steaming and less spark throwing, lighter valve gear which is easier to handle.. Argued that a three-cylinder 4-4-4T could take the place of a two-cylinder 4-6-2T or 2-6-4T. This idea was pursued still further in the paper, and in the discussion, where it was considered that three-cylinder 4-4-0s would be able to perform the work normally allocated  to 2-cylinder 4-6-0s: these reamarks may be regarded as the genesis for the Schools class. Discussion: Sanderson (Baldwin Locomotive Works pp. 369-71) argued that only very short locomotives with outside cylinders were liable to roll and sway: longer locomotives did not as the momentum of the mass of the engine will not permit these disturbing effects to produce a waving or sinuous motion., especially if leading and trailing trucks are fitted. He made reference to four-cylinder compound locomotives where "the two-cylinder engines are out on the road earning money when the four-cylinder engines are more frequently in the shop losong money".

J. Clayton (p. 371-3) took exception to Holcroft's statement: "The locomotive is therefore always somewhat of a compromise between conflicting interests." While [he thought] there is a measure of truth in this remark, [he suggested] that all engineering is compromise and that the successful engineer is he who can so design his locomotive or other machine that the best all-round results are obtained, rather than conflicting interests satisfied. That is a distinction with a difference, for in endeavouring to secure the best compromise all round, so far as ohtaining good mechanical and efficient results, all other interests arc satisfied; so, rather than call them conflicting, we had better say they aid and assist in obtaining the best. He then cited Holcroft's mention of  the 4-4-2 engine, the North Star, on the Great Western Railway and that this engine was fitted with the "Deeley" valve-gear. It was a good gear of the "Walschaert" type, but, unfortunately, differing from it hy having the right-hand valve-gear different to the left-hand, and so added considerahlv to the cost without making the gear any better as such, though it does avoid the use of eccentrics. It also makes a hreakdown on one side of the engine a total disablement, as the motion on either side is dependent on the other.

It is rather singular that the three-cylinder locomotive has not been adopted more by British locomotive engineers, and it may be as Mr. Holcroft suggests, that it. has always been considered necessary to employ a third valve-gear for the third cylinder. This did not, however, deter the Midland Railway from adopting a three-cylinder compound, so eminently satisfactory on that road. Then Sir Vincent L. Raven, of the North-Eastern Railway, is not afraid of thc so-called extra complication of one extra valve-gear, as used on his successful three-cylinder locomotives.

Personally, I am not convinced thut any arrangemcnt of levers and cranks such as suggested so ingeuiously by the [Holcroft] in substitution for a third valve-gear would be quite so good or necessarily simpler. In any case, there would appear little to choose between the two methods of controlling the valve of the third cylinder. May I suggest, however, that the three-cylinder engine has one great weakness, especially over the two-cylinder eilgine, and that is the crank-axle. This axle, having the crank in the centre, is undoubtedly weak, und necessitates a stronger axle and better-designed crank-sweeps than is usually assigned to the ordinary crank-axle. I am, however, entirely with the Author in his advocacy of the three-cylinder locomotive as an improvemcnt ovcr the usual but useful and faithful old two-cylinder type, especially where high powers are required. The turning moment is more uniform and regular than with the two~cylindcr or the four-cylinder types, as with three cranks no crank being opposite to the other the variation between maximum and minimum turning efforts is less marked. The balancing, however, is somewhat better in the four-cylinder engine, and this, taken in conjunction wilh the greater case with which only two valve-gears may be adaptcd or used, the stronger design of crank consequent upon the two instead of one inside-cylinder and the still more equable division of stt'esses, the first place should probably be given to the four-cylindcr locomotive.

With regard to the point raised as to thc improvemcnt of the 4-4-0 engine by the use of three cylinders, I should say that, whilst some improvement would undoubtedly be effected, the chief limit to the type is the lack of adhesion, and recourse must be had to the three axles counted for heavy powers. Persoually, I favour the three or four-cylinder compound of either the 4-4-0 or 4-6-0 type, on the "Smith" system as used by the Midland Railway. This point of view, however, should be reserved until our next paper on the compound .locomotive, to be read shortly by Mr. Rieckie.

I was interested in Mr. Sanderson's remarks, and should like to support what he says about" rolling" and " swaying," and that you must take into account the design of the engine, and in the particular engine to which attention was called it is probably due very largely to the way the mass of the engine is steadied at its extremities.. As Mr.Sanderson very dearly pointed out to us, if the mass is great enough, the inequalities referrcd to are imperceptible.

Mr. H. Kelway.namber (The Leeds Forge Co.): I too would like to congratulate Mr. Holcroft on his excellent Paper, and to ask him how the starting effort of Mr. Gresley's three-cylinder 2-8-0 type Great Northern Ruilway goods and mineral engine compares with that of the two-cylinder engines of similar wheel arrangement and adhcsive weight, which have now been for some years in service on that Railway. Could he also tell us what amount of reduction in coal, oil and watcr consumption has resulted from the use of threc-cylinder engines?

Mr. Lelean: Always keeping in mind the point to which Mr. Sanderson referred-viz., the mass of the engine and its effect in smothering small disturbances and vibrations and so steadying the locomotive-the question arises whether it is really desirable to place the outside cranks opposite the inside cranks in the case of the four-cylinder locomotive. Mr. Clayton's diagrams showed the effects of two series of four impulses per revolution superimposed on one another in the case of the four-cylinder engine, and six impulses in the case of the three-cylinder c:ngine, but by setting the insidc cranks at 135 degrees to the outside cranks eight impulses per revolution would be obtained, and it has to be considered whcther the balancing troubles and the upkeep of thc extra motions would not be more than compensated for. I would like to ask the Author if he could give us any information as to whether there are any four-cylinder engines in which the cranks are thus arranged so as to give eight independent impulses in the rcvolution.

Mr. Clayton: I ought to just remark that, as I explained with the diagram just now, it is not so much that the four-cylinder engine gives eight impulses in the revolution; the point was that, occurring as they do, the difference between the maximum and minimum efforts was so very marked.

Mr. Lelean: But I understand that it is due to the inside and outside cranks being opposite each other. Why could they not be set at 135 degrees? "

Gairns: (p. 374): took it that when referring to three cylinders, and the equable turning moment realised thereby, his point is more particularly that the three cylinders are using high-pressure steam. When it comes to compounding there are two low-pressure and two high-pressure cylinders, and two of them would he comparable with ordinary two-cylinder praetice, the cranks heing generally at 90 degrees; hence there are two cylinders of the same character and one to a degree variable. Presumably, in running, as Sanderson and others have suggested, there would be compensation for this, and a certain steadiness is obtained from what may be somewhat unsteady units at starting, but at any rate there is not the equality that three-cylinder simple working gives, and it seems to me that most of our three-cylinder compounds—the Midland and the Great Central particularly, for they are the most successful —are really two-cylinder engines adapted and brought into the system.

I should think that the relatively variahle working value of the third cylinder must in some degree affect the balancing and equalising of turning moment as compared with the use of three simple cylinders.

I.E. Mercer (LNWR p. 374): The Paper deals more with means of working the centre valve by means of rocking-levers from outside valve-gears than with the three-cylinder locomotive as compared with other arrangements of cylinders. The Author scouts the idea of using more than two sets of valve-gear, but fails to advance arguments in support of such a basis for his Paper. He states that recent practice has been very largely in favour of such designs, but does not follow it up with the comparative results obtained. The first and maintenance costs of a locomotive are very important things, but are certainly secondary to successful operation with economy. Recently a leading article appeared in an important journal which strongly favoured rocking-levers and pin-joints, with small movement to a complete valve-gear. The point of view advanced was, that it is not the number of joints that count, but the type as well. This is all perfectly sound theory, but is it as perfectly sound practice? To commence with, very great care and size are needed in connection with all rocking-shaft hrackets.

His experience of several designs on more than one railway is that such brackets are generally the first parts to give trouble in the valve-gear; therefore such experience would, suggest that a double set of valve-gear is less likely to cause anxiety than the use of, rocking-shafts and brackets. Massive construction, however, overcomes this weakness. Another point is that when any valve-motion has become worn, the locomotive suffers at a much more rapidly increasing rate than the wear of the motion. With a valve indirectly driven there is the loss of travel and poor operation due to the worn valve-gear, plus the loss by wear at each, pin-joint transferring the motion to the indirectly-driven valve. This serious defect can be largely overcome by again adopting massive joints, with every provision for lubrication, but all this addition in weight of parts means further increase in the dimensions of the valve-gear parts proper, and greater chance of enhanced wear and possibility of running hot as compared with a motion that only drives one valve.

With inside cylinders the bracing of the main frames is a more difficult matter, and the use of inside valve-gear as well will not greatly affect this problem.

If possible, the rocking-shaft drive should be taken direct from the valve-gear, and not from the front end of the valve-spindle, as although an approximately correct valve-setting for the second valve can be ohtained by experiment, it is quite surprising what the effect of the steam-chest temperature will be on the setting of the indirectly-driven valve, owing to the varying expansion of both valve-spindles. In some instances it is found necessary to set one valve without lead at one end, as the expansion of the spindles will alter the setting of the valve when running to this extent. The steam-chest temperature varies, so the setting of the second valve must also vary under different running conditions. In any case the valves nre necessarily badly set when such engines are getting under way.

It is not suggested on these accounts that the vogue of indirectly-drivcn valves is bad practice, but rather that the Author is mistaken in suggesting that the alternative practice is not worth consideration. The method the Author advocates can, without doubt, be made very successful if proper precautions are taken and the design of the engine in general, is suitable, but, even then, it is a question whether the result would be any better than that obtainable by Sir Vincent Raven's plan of, using three complete gears. So, far as he knew the NER. Class " Z" give no trouble, speaking particularly of the valve-gears, and most certainly ware giving excellent results in the haulage of traffic. The use of three inside link-motions instead of one and two outside gears, as adopted by Mr. Robinson on the GCR is, of course, nothing to do with the subject.

Also, to argue that it was Mr. Churchward's adoption of only two valve-gears that brought about the one or two other designs now extant is perhaps saying too much for one point of design. Had Mr. Churchward used four gears and obtained the very fine results he has done, it is quite likely that four-cylinder locomotives would have bcen very nearly as popular. Mr. Churchward, and Mr. Dugald Drummond too, first adopted four cylinders with four valve-gears for their other merits, although there were several hundred four-cylinder engines running on the LNWR with only two gears many years prior to the GWR and LSWR experiments. I am inclined to think the use of more than two cylinders had its fillip from another factor, which arose shortly after Mr. Churchward's Star class was introduced, namely, superheating; this led to increased cylinder diameters, with increased maximum pressure on all bearings, which in so many cases caused bearing troubles; these have by no means been overcome to this day. Yet to-day, in spite of the success of previous designs of three and four-cylinder locomotives, there is a certain amount of return to two-cylinder design, though no doubt two cylinders will in time have to be abandoned for heavy work.

He preferred the three-cylinder engine to any other for several reasons. With three cylinders there is only one inside crank to give trouble, and, as it is only driven by one-third of the power of the engine, it should give a far better account of itself than the inside-cylinder engine, while reducing the objections to outside-cylinders. Also, owing to only one crank in the axle, the driving-bearings can he of almost indefinite size, which is a great benefit. Cylinders are liable, with their fittings, to be more troublesome than valve-gear, so that I would rather have one more valve-gear than one more cylinder; this was before it was even suggested the third gear might be dispensed with if thought necessary. Although there is a somewhat wider variation in full gear between maximum and minimum torque with three cylinders than four, the actual maximum is not so high for a given mean, so that, even taking into account reduction of wheel-pressure on rail caused by the movement of the counter-balance for reciprocating parts, the three-cylinder engine is approximately equal to the four-cylinder–to be precise, it is about two tons to the good–as regards necessary adhesive weight for a given horse-power.

In spite of these advantages, the three-cylinder simple locomotive seems ruled out as the possible future standard high-speed or even heavy mineral type, because he had always strongly advocated compounding. The MR three-cylinder compounds are probably amongst the best running in England, along with their GCR confreres. But as an ultimate type they have certain defects. To obtain the excellent torque of the three-cylinder quoted previously, each cylindcr must be of equal power. In the compound engines mentioned this feature is lacking, as the HP horse-power should equal the LP, which means the inside HP crank transmits as much power as the sum of the other two. Using two HP cylinders and one LP, there is no benefit in this respect. If one designs the horsepower of the HP cylinder to be only half that of the total LP, then one gets defective valve-operation and loss of the best economy possible. At the same time, as designed :at present the load on the inside crank is higher than with a simple engine owing to the above.

Mr. F. W. Brewer: There is one sentence in the Author's Paper which goes straight to - the point, and if it were invariably kept in view would, I believe, prevent a great deal of misunderstanding. That sentence is: "The locomotive is always somewhat of a compromise between -conflicting interests." That is true. In locomotive practice it is virtually an axiom that you cannot very weIl combine every good feature at one and the same time in the one unit. Something has to be sacrificed. Consequently the locomotive engine represents a compromise between what is theoretically desirable and what is necessary on practical grounds. Between the two extremes there is plenty of scope for investigation and trial. In this connection I am glad to find that the fuller development of the 4-4-0 engine is advocated in the Paper. Twelve years ago I wrote an article recommending the enlargement of that type. The Author's suggestion that three cylinders should be used struck me as bcing un excellent one, as the cranks could then be placed 120 degrees apart, thus giving an even turning moment and doing away with the dead centre. In addition, the effect is to increase the adhesion-a very important point in the case of a four-coupled engine, in which the weight on the two coupled axles could not much exceed 40 tons. Probably the Author is right in putting the highest I. H. P. of most 4-4-0

Journal Number 36

Riekie, John (Paper No. 66)
Compound locomotives. 405-29. Disc.: 430-75.
Based on work in India. Discussion: J. Clayton (433-5) noted that the Smith/Deeley system was "highly successful" and achieved 1200-1300 hp over 15 miles. Fuel consumption was very low: 0.0916 lbs coal/ton mile and water consumption was also low: .797 lbs water/ton mile. Also cited the excellent PLM compounds and noted their flexibility. Mercer (LNWR 438-40) spoke in favour of compounding. H.G. King (441-4) refered to the L&YR 4-cylinder compound 0-8-0s. H.W. Dearberg (445-52) commented on the Webb 4-cylinder compounds noting that the proportions of the cylinder volumes were incorrect and that the receivers induced drag in the low pressure engine. He also noted that the French compounds had good cylinder proportions. H. Holcroft (458-60) was critical of the compound concept as there was less friction in simple types. Sith Mannering (460-2) refered to the excellence of the copper tubeplates on the K class 2-6-0s.