Journal of the Institution of Locomotive Engineers
Volume 21 (1931)

A paper by J.W.R. Broadfoot Locomotive progress on the West Australian Rys. was presented in Perth presumably during 1931 as an extensive abridgement appeared in the October issue of the Locomotive Mag., page 340 et seq

Journal No. 99

Smith, F. (Paper No. 270)
Smithy practices of an Argentine locomotive repair shop. 8-61. Disc.: 61-79. 52 illus./diagrs.
September Quarterly Meeting of the 1930 Session was held at the Gorton Workshops, Perez, on Friday 19  September, 1930, Mr. W.P. Deakin presiding.
The subject was treated mainly from the practical side, giving methods and practices adopted at the smithy of which the author was in charge at Perez.
Due to various causes the scope of the blacksmith and the opportunity to demonstrate his skill in hand forging work was gradually lessening. Looking back over a period of a few years, we find that the method of production, as in other trades, has undergone a great change. The steel foundry, drop-stamps, forging machines, hydraulic presses the oxy-acetylene jet cutting machine, electric and oxy-acetylene welding, cutting and building up all combine to reduce the necessity for hand forging.

Journal No. 100

Sanders, T.H. (Paper No. 271)
Locomotive suspension, and its influence on derailments. 133-55. Discussion: 155-63; 205-15; 859-65.
Third Ordinary General Meeting of the North Eastern Centre (Session 1930-31) was held at the Hotel Metropole, Leeds, on Thursday, 20 November, 1930, at 7.15 p.m., the chair being taken by Mr. L.W.R. Robertson.
The laminated spring is almost universal in locomotive suspension and the discrepancies which can arise in this and connecting rigging to affect weight distribution are numerous and may be defined as follows:
(a) Loss of washer bearings, or wear of solid ends, combined with wear of gibs or pins which take the weight thereto.
(b) In overhung springs, hoop dimensions inaccurate at bottom end. In underhung springs, hoop dirnensions inaccurate at top end.
(c) Discrepancy in loading rates of springs of the same design.
(d) Discrepancy in loaded cambers of springs of the same design.
(e) A proportion of broken plates.
(f) Variation of lengths of fixed hangers, or varying adjustment on adjustable hangers.
(g) On engines with equalising rigging, items (a) and (f) additionally.
Any or all of these discrepancies may affect one single spring, so that the possibilities of certain axles being over- loaded and others underloaded is very considerable, as the underloaded axle is obviously the danger point when derail- ments have to be considered, and the overloaded axle when wear and tear of bearings and. track are regarded. In coiled springs, the use of which for locomotive suspension is almost limited to this Country, discrepancies (c), (d), (f)
The first instance is that of a 2-4-2 type tank, travelling chimney first, which, passing at 50 m.p.h. over a somewhat defective crossing, lurched to such an extent that it did not recover, and went off the track after travelling about 300 yards. Certain aspects of super-elevation also arose, and, generally speaking, the broad conditions would all lead to blame being placed on defective track. The engine was not seriously damaged, and after being towed dead to adjacent sheds was weighed, a variation of 3½ tons being present between the left leading coupled wheel and the right leading coupled wheel, the latter being the lighter, and one ton between the trailing coupled wheels, the left hand being the lighter. The reason for this weighing can be best put in the official form after the engine had been inspected at the sheds :-" This inspection, in fact, failed to reveal anything which was likely to have caused or contributed to its derailment on this occasion. On the other hand, unsatisfactory though the track was, it hardly seemed possible that the severe and continuous oscillation described could have been set up—-even at high speed—-particularly as there was no evidence of previous undue movement of engines at this point-unless some unseen defect had suddenly developed in the carriage of the engine. " It should be stated that the class of engine concerned was very numerous, about 330 being in use on this particular railway, and giving always very excellent service with a minimum of trouble.
After the weights had been found as erratic as mentioned, the engine was sent to have the wheels dropped, with a view to a still more detailed overhaul, and during this process it was found that the back plate of the right hand trailing spring was broken inside the hoop. As the result, all the springs had the hoops stripped off, and out of 112 plates, no fewer than 24 were found cracked, or broken, as follows :-

Five of these in the spring of the right hand trailing coupled wheel had broken, four through fatigue cracks, and many of the fatigue cracks in the other springs were sub- stantial, and would have gradually caused failure of the plate in a very short period.
As the result of this discovery considerable attention was concentrated on the spring aspect, and many comments were made. It must be borne in mind that a laminated spring has this disadvantage in comparison with a coiled spring-there is never any doubt regarding the failure of the coil. With a plate spring, however, which is properly fitted with a substantial hoop, it is not an easy matter to detect the failure of plates, as the greatest percentage occurs on the planes of maximum stress, which are within the hoop width. The load testing machine will not necessarily give any information, as plates which are broken completely through, will, if the hoop be tight, flatten out and resist by the action of adjacent plates, with the exception of broken short plates, which will stand off. Some Continental railways have made a standard practice of slots in the middle of the hoop width, with a view to detecting failures at this location, but the slots choke with dirt in service, and, furthermore, the failures will not necessarily take place conveniently opposite the slots-so this device cannot be regarded as specially useful. A method by which the spring is registered to the hoop by a key piece secured to the hoop side, and fastening into a slot cut in the spring side, is good, but still remains limited to the immediate centre of the spring. To be certain of detecting broken plates, therefore, the hoop must be stripped. A method of hammer testing can sometimes detect the fracture, but is not certain, and it will not detect cracks only. However, before a plate completely breaks it has, in nine cases out of ten, been previously cracked—-fatigue cracks due to normal service being the usual life limit of a spring plate. It is, therefore, imperative that, as springs are overhauled the plates should be so examined that these fatigue cracks are discovered, but this is nearly as difficult a matter as finding broken plates in the hooped spring. Fairly obvious cracks are, of course, discoverable, such as some of those which are indicated in the present set of springs, where, in one instance, the fatigue crack, on a plate of  5 in. by 5/8in. section, was 3½in. by ½in., but this had grown from one which, perhaps three months earlier, was  ½in by 1/8in., and in consequence not so readily detected on a greased and dirtied plate. If the camber of the overhauled spring including this plate had been correct, the plate would not have been heat-treated or hammered (a spring fitter's hammering in setting will often detect such flaws) but merely included in the spring as re-assembled. The detection of fatigue cracks with certainty can, it appears to the Author, be performed only in two ways (1 by heating, flattening, and grinding the upper or tension surface of every plate—-as fatigue cracks are invariably on this surface only—-or (2) employing the magnetic method such as is now used for discovering flaws in axles and other important steel details. The former is undoubtedly the more costly method, and it certainly appears that the latter has therefore many advantages, although it is not employed in repair spring practice so far as the Author is aware.
The particular spring which included all these cracked and broken plates is well designed for the load, but the two upper plates, being  5/8in. thick, pass through a range of stress 25 per cent. greater than the following plates, which are tin. thick. The failure of the bottom plates took place across the edge of the hoop-a point to be noted. Calculations for locomotive springs are invariably based on the maximum stress being central, which is normally the case; but if heavy hoops are fitted, which are " panned" solidly on to the short plates on the " as-made" or light spring, obviously the short plate has its bearing lines on the hoop edges, and the effective length is substantially reduced. In the case of springs working with positive cambers, the hoop bottoms should be left flat; and in the case of springs working flat, the hoop bottoms should be radiused. This spring design included "upward nibs" which is the most objectionable form of centre fastening, as usually made. Most of the cracks in the springs now under examination started from these nibs. The railway concerned decided, as the result of this study in fractures, to abandon the upward nib in favour of a hole, which is certainly preferable, though not as good as the downward nib. The upward nib is used where it appears to the draughtsman that a set screw and packing plate cannot be readily introduced at the top of the hoop, and so allow for a downward nib. There are, however, ways and means of providing a downward nib in the most difficult form of hoop. The heat treatment of a plate taken from the broken spring came in for considerable study, but nothing unduly serious could be urged against it. The material was found soft on the surface and· hard towards the middle, as the result of the usual methods of handling water-hardening steels, but nothing abnormal presented itself. The metallurgical laboratory report indicated a possibility of the plate being over-heated (the sample was  ½in. thick) which is possible, but not as probable as the fact that the plate had been under-tempered. The foregoing comments indicate the possibility of reducing spring breakages on the road, with derailment effects such as in this case, by attention to the details of design mentioned and the most rigorous care and attention in overhaul to ensure that no cracked plates are replaced in a spring. Properly organised, it would probably be found that the magnetic method indicated would detect every flaw.
The second instance to be examined is the case of an 0-6-4 type tank, which, after running 440,000 miles in a life of about 20 years (being one of a numerous class), came off the road, chimney first, at a speed of about 50 m.p.h. The engine was running on the straight, but on a low embankment which had the outer rails super-elevated to throw the weight towards the six-foot way, and the leading left hand coupled wheel was the first derailed. The springs throughout were of the laminated type, with the exception of the leading coupled axle, which was fitted with a pair of coiled springs under each axlebox. A detailed examination of the suspension details after the accident resulted in the back plate of one of the driving springs being found broken, with a fresh fracture which was evidently the result of the accident, and two other plates in different springs had fatigue cracks. No question therefore arises as to any connection between the laminated bearing springs and the derailment. The coiled springs under the leading axle were found intact, but on testing a very substantial difference in load carrying was revealed, the left hand deflecting 0.86in. under the specified load, and the right hand  1.21 in. under the load, the desired figure being 1.02in. The track was not good, being out of alignment, and the super-elevation was far from being as regular as was intended, so that rolling was set up, which continued, and resulted in the derailment. The track condition met, of course, with substantial criticisms, but it appears to the Author that the springs on this leading axle could also be severely criticised.
The usual drawings supplied for the manufacture of coiled and laminated springs specify a loaded height or camber with a definite load. As the dimensions of the springs, and sections of the steel, are usually given in full detail, it is assumed that the deflection under the specified load will be reasonably uniform for springs to the same drawing ordered from different makers at different times. In practice, however, such is by no means the case. Regarding laminated springs for locomotives, which are invariably positive length springs, that is, having pin or gib ends to definitely defined centres, as distinct from the plain end of the wagon type, the bearing centres of which may easily vary 2 in , over different makes—-the length is a definite figure and the width is a definite figure. So many plates of a certain thickness are specified, which also appears definite, but actually provides the variable. Assume a steel section 5 in. by ½in., and a spring of twelve plates built up from this. The overall depth may be exactly 6in., or it may be 61/8in., which could not be reasonably objected to, as it is only an excess of 1/ 96in. in thickness per plate. The difference in load carrying is, however, six per cent., as the deflection depends upon the cube of the thickness. Additionally to this, assume the spring which is exact to thickness is also up to the maximum of "concave." It should be known that all steel bars for the manufacture of plate springs are rolled with a concave in each surface, and not dead flat, the object being to facilitate the fitting. The British Standard Specification permits a maximum of 1/64in. per side, or a total of 1/32in., which is much too great if advantage is taken of it. All engine springs should be made from steel which certainly does not exceed in concave 1/100 in. per side, and less than this is preferable. Therefore, if the 6in. deep spring has the steel of maximum concave and the 61/8in. deep spring has the steel flat, the unit deflection due to the concave aspect alone provides a difference of 11 per cent. It can, therefore, be seen that a loading rate of no less than 17 per cent. can exist between two identical springs as regards figured and passable dimensions. If the load be assumed as eight tons, and a deflection of 0.2 in. per ton, the deflection of the weaker spring will be 1.6in. (taking 0.2 in. per ton as representing the weaker) whilst the stronger will only deflect 1.36in., a difference of ¼in. Both springs will pass the inspector, as the maker using the weak steel will produce a spring ¼in. higher in camber free than the maker using the strong steel—or the same maker may produce these two different springs at two different times. With the free camber ¼in. higher in one case than the other, the loaded camber will be identical.
In the case of coiled springs, in which the unit deflection depends upon the fourth power of the section, a similar discrepancy is easily obtained. Assume a section lino round, nominal, supplied by one maker as 1 1/64in., and by another as 63/64in. The strong section will be six per cent. stronger than the nominal, and the weak section seven per cent. weaker than the nominal, the difference between the extremes being' 13 per cent. In the particular case of the leading-axle springs now being examined, on opposite sides of the axle there appear one pair of springs deflecting 19 per cent. greater than the nominal figure, and on the other side the pair deflecting 16 per cent. less than the nominal figure, a total difference of 35 per cent.—or, with the weak spring as a basis, the strong spring was 40 per cent. stronger. The steel section of the springs was flat, in one pair at one axlebox, the separate springs measuring iin. by 19/16in. and 27/ 32in. by 19/16in.; whereas the strong pair on the first derailed wheel measured each iin. by I!in. It should be noted that greater variations are likely to occur in springs of rectangular section than in springs of round section, as the former has to be rolled of special trapezoidal shape in order to properly produce the rectangle section during coiling. The suggestion is that rectangular sections should be avoided, although it is recognised that this is not always possible in underhung locomotive springs, owing to lack of clearance below. In the present instance adjustment was provided on the axle-boxes, but no details of adjustment will alter the rate of spring deflection as such. In order to conform with the drawing, the weak spring would have been higher than the strong spring, as made, and assuming the adjustment symmetrical on each side, theoretically the weak spring would take a part of the load before the strong spring contacted, as the former would have been  3/8in. higher free. It had, therefore, the advantage of this 3/8in. to hold its wheel to the road. It is obvious from this case that more attention should be given to the correct pairing of locomotive springs, and a limit for total deflection under load should he imposed, and included in specifications, in the same way as a limit is given for height or camber under load. Some Continental specifications stipulate this, usually as plus or minus five per cent., so that a spring specified to deflect 1.02 in. under load. as il' this case, would only be permitted to vary on deflection between 0.97in. and I.07in.-whereas in the present instance it has been seen that the variation was 0.86in. to1.21 in,
The coiled springs for bogie coaches are, on some railways, most carefully paired to within 1/32in., as the good riding of the coach is thereby maintained-and if this can be done for coaching stock, it is far more imperative that it is done for Iocomotive coils,' whieh are of very much lower deflection, and therefore more susceptible to'derailment over imperfect track.
The third study involves also coiled springs. An 0-6-2 type tank engine, with all axles on laminated springs except the driving axle, came off the road on the straight, after passing a curve. The engine was running chimney first, and derailed at about 35 m.p.h. The track generally was in fair condition, but a special point of comment was "excessive and incorrect super-elevation on the straight portion between the curve and the point of derailment," in other words the run-out of the super-elevation was unsatisfactory. The engine was practically new, having been put into service only six months before, the mileage being 12,500. This case provided the somewhat mysterious aspect of hroken coiled springs of doubtful manufacture. The engine, as stated, was new, and made by a contract shop, who would, in the ordinary way of purchase, indent for the springs from some approved manufacturer after the order for the engines was obtained. One of the broken springs was, however, found to be dated four years prior to the delivery of the engine, and the other broken spring was doubtful, as the date could not be deciphered. There was a record of the change involving the latter spring, but not of one involving the former. The reason as to why springs on a new engine required such a quick change was not inquired into, but it would seem as if some faulty design or manufacture was responsible. The main point is, however, the fact that two springs, new to the original engine, were introduced in six months, and, presumably as the result, it was found that the weight on the driving axle exceeded the design weight by 2¾ tons. The design weight was undoubtedly correct on the engine as built, so that the change to casual springs obtained from the local stores must be held responsible for the alteration. As will have been consistently noted, adjusting gear, unless handled in conjunction with springs of correct deflection, and engine weighing machines, can be worse than useless. In both the foregoing cases, adjusting gear was fitted, and in the last case it was stated that the spring hanger nuts were adjusted to exactly the same positions as they were in with the original springs, though the explanation as to why these had apparently been dispensed with was not forthcoming.
The fourth instance to be detailed is that of a 2-6-4 type tank engine, running chimney first, which came off the road at about 60 m.p.h. The springs throughout were of the laminated type, and all were intact and in good order after the accident. The track itself was in reasonable condition, but had not been improved by some recent heavy rainfalls. The train was travelling round a curved road, on which the super-elevation should have been 3iin., but which varied on the length previous to the first point of derailment, and, as noted, commencing at a point on the transition, broadly as follows; 25/8iin., 21/16in., 35/16in., 25/16in., 213/16in. At this position occurred the first sign of derail- ment, the leading coupled wheels -being off and running for a considerable distance with the engine rolling over the switchback of the super-elevation, which continued as 3in., 21/8iin., 3¾in., 21/8n., 3½in., and similarly—the left hand wheel on the keys from time to time, with the right hand wheel filling up the spaces of the unmarked keys by being on the chairs, until, at a pair of trailing catch points, complete derailment took place. The permanent way and locomotive both came in for much attention as the result of this accident, and in view of doubts existing in certain quarters as to the stability of the locomotive type, special trials were ###made with similar engines, on first class track, during which trials the engines were run in both directions at speeds up to 83 m.p.h. with entirely satisfactory results, on some trials the tanks and bunkers being full, and on others nearly empty. At a later date, on track not quite as good, similar trials were made up to a maximum of 76 m.p.h. and certain rolling was experienced. Trials made with a 4-6-0 type tender engine, with stiffer springs than those on the tank engines, were reported on the first class track as: "The engine ran with perfect steadiness without any roll or lurch, but the vibration on the footplate (at speeds of 75 to 81 m.p.h.) was so great that it was impossible to obtain any instrument record." On the somewhat less than first class track it is stated: "The vibration was, if anything, worse than previously obtained. (This at 67 to 86 m.p.h.) At the places where rolling occurred with the tank engines there was considerable rolling and lurching with the tender engine, and the riding can be described as very rough and uncomfortable. "
The above trials confirm a statement in the earlier part of this Paper, that, broadly speaking, the engine with stiff springs, approaching in value to a springless carriage, is compelled to follow vertically a track of bad levels, and, accordingly, to roll more dangerously than an engine more softly sprung, which may actually appear to roll somewhat more. In this last instance very specially it appears that some equalisation would have prevented the derailment. When a coupled wheel tends to lift its flange above rail head, the action of the driving rods tends to throw it over the rail head, and in all the instances mentioned no assistance could be obtained from the weight on adjacent axles to hold down the rolling axle. Had this particular 2-6-4 type locomotive —which is an excellently practical type—been provided with a truck of the design in common use on the Continent, and variously known as the Italian, Flamme, and Zara, there would certainly have been some assistance to keep this leading coupled axle in position with the wheel treads pressed to the rail, not only from the equalised weight, but also from the throw-over of the axleboxes. This type of truck appears a better arrangement for high speed engines than the usual American equalisation between the Bissell and the leading coupled wheels via longitudinal and trans- verse beams.
Considerable additional matter could have been introduced into this rather complex subject, which is, it will be admitted, of the highest importance, but it is hoped that sufficient points have been touched upon to impress upon Members the connection between derailments and spring suspension. The general assumption in this Country is that British permanent way is so excellent that no equalisation is necessary, or even desirable—but the four cases which have been examined have sufficiently shown that, whilst the track here may nominally be all first class, it certainly is not on certain sections. On the other hand, in no case was the track involved inferior to quite normal Continental and American track. The fact that in this Country excellent squared and creosoted sleepers are combined with heavy chairs, and well fitted keys, to hold a rail of 100lbs. per yard, not overloaded with 20 tons on an axle—as compared with track of unsquared and untreated sleepers, on which a 90lb. rail is held down with dog spikes, and takes an axle load of 25 tons—must not cause to be lost sight of the fact that it is the levels and alignment of the track which make for good running and lack of derailments—and if the former track is not properly levelled and aligned, it has no serious superiority over the latter. Admittedly, first class British track is amongst the finest in the world, but it is obviously improbable that all our 22,000 miles of route, equalling a track mileage of approximately SS,ooo-is in the same excellent condition as, say, the main line running roads from King's Cross to Doncaster, or similar sections which will occur to the minds of Members, according to the railways in which they have interest. With the standardisation of engine classes, it is now more than ever necessary that good, bad, and indifferent permanent way must be worked over by the same class of engine, and the suspension accordingly should be designed with the vision directed to the most inferior in being. If these few remarks are sufficiently pertinent to provide some little more attention to the general aspect of suspension, combined with useful equalisation; and also to the aspects of spring design and the accuracies of spring deflections, it will be felt that they are not altogether without value, and will perhaps compen- sate to some small degree for the useful time the Members have diverted to the patient hearing of the present Paper.
Discussion. A. Hird (272-3): I have not had much running experience, but I think that a good many of the failures and derailments are due to conflicting ideas in design. We are liable to think of a locomotive as a rigid machine, like a planing machine, and running on a true plane surface. The actuality is far from that. The Author has shown us that springs change their loading very quickly. Springs change camber with every ton variation. It is easily seen that if one puts the wheels on a weighing machine which has been set carefully and is practically on a true plane and one weighs them, setting the springs carefully to their designed load, they won't be the same when the engine has moved twenty yards, owing to inequalities in the road. Railway officials and inspecting engineers have to work to specifications which are too re- stricted in tolerance or clearances, with the result that there is not the flexibility provided in the locomotive to make it able to meet the conditions of the rigid and uneven road it has to run upon.
Take the case of a coupled engme, there is usually no provision for lateral play to allow it to traverse a curve beyond the clearance between flange and rail. Some inspectors will not allow 1/ 32in. play between the faces of wheels and axleboxes. When engines have to run on a bad road the axleboxes should have flanges tapered at top and bottom so as to allow them to rise freely without smashing off the flanges. There are some axleboxes in Continental use that have no flanges so that they may have plenty of freedom to rise. Each side of the box has a pivot formed On it and which fits into a recess of perhaps 4in. or sin. diameter in the slide, so that the slide has freedom to rise and fall up or down the hornblock and the box free to pivot according to the angle of the axle.
With regard to the derailments described, no mention has been made of the position of axle bearings whether they were inside or outside the wheels. Many years ago I had the following experience. It concerned an 0-6-2 engine, aft. gauge, working on a branch line at a water works, which derailed frequently, and they could not account for it. I had to go and see why the trailing two-wheel bogie derailed frequently and yet the coupled wheels did not de-rail.
I had the places shown to me where it derailed, and saw the track was in bad condition. The coupled wheels could be on a comparatively level piece. then there would be a curve with a sudden drop at one side; the wheel went down on the left-hand side and the load-leverage on the outside bearing actually lifted the right-hand wheel off the rail into such a position that the tyre flange mounted the rail. A new axle was put in and the bearings changed from outside to inside, and the wheels and the engine never derailed again. With inside bearings there is no leverage tending to lift the wheel off at either side.
The Author has given a lot of instances which should prove useful to our younger members. The Paper enforces the truth that there should be provision made for play of every kind, or else the shocks will break the springs or damage the bearings and possibly cause derailments. T,H. Sanders was obliged for Blundell's remarks, but think he has misunderstood my reference to " fixed length" springs. I mean thereby a spring. with definite pin or bearing washer centres, of the usual locomotive type, as distinct from a spring with plain ends bearing on shoes or similar fitments, in which the bearing points are rather indeterminate at any time, and, in any case, vary in span (ancl, in consequence, "straight length" of top leaf) due to the deflection from time to time. In wagon springs there is sometimes as much as ain, difference in the bearing point span free, with springs from different makers. We know it is hopeless to take an engine fresh from the shops to the weighing machine, whatever type this may be, and expect to obtain correct weights. When I had to deal with this aspect, my practice was to pour plenty of oil over hornblocks and suspension gear and run the engine for half a day under its own steam, over washers r in, thick, which were staggered on the opposite rails and about six feet apart. This method thoroughly shook up the engine, and the weights obtained were reasonably correct when it was taken to the machine.
Four un-nnamed accidents (2-4-2T, 0-6-4T, 0-6-2T with coil springs and 2-6-4T), but clearly including those at Sevenoaks (Maunsell River class) and Buchlyvie (Gresley N2 class). Suggested that had River class been fitted with the Zara-type of leading pony truck that the Sevenoaks accident would not have occurred. Noted that compensating levers were quite usual on British 4-4-0 and 4-4-2 classes. On page 863 it is noted that equalising gear was invented by R.&W. Hawthorn in 1851, but may been invented earlier in Europe, but not in America. On page 864 in his reply to N.J. Longley he noted that the Bristol & Exeter 9ft singles and the McConnell Bloomers both employed direct rubber suspension.
Fourth Ordinary General Meeting of the Manchester Centre (Session 1930-31) was held in the building of the iManchester Literary and Philosophical Society, 36, George Street, Manchester, at 7 p.m., on Friday, 16 January, 1931, Mr. S. H. Whitelegg taking the Chair.
Second Ordinary General Meeting of the Newcastle Centre (Session 1931-32), was held in the Central Station Hotel, Newcastle-on-T’yne, on Friday, 27 November, 1931, at 7.15 p.m., the chair being taken by Mr. P. Liddell.

Windle, E. (Paper No. 272)
Some notes relating to cylinder performance. 178-97. Disc. :197-204. 7 diagrs., table.
Fourth Ordinary General Meeting of the North- Eastern Centre (Session 1930-31) held at the Hotel Metropole, Leeds, on Friday, the 19 December, 1930, at 7.15 p.m., Mr. T.H. Sanders occupying the chair.
This paper outlines the development of long lap valves on the LNER.
Discussion: J. Blundell (200): The Author has compared the locomotive trials from King’s Cross to Doncaster with others from Carlisle to Crewe. Surely the gradients on those two lines are not sufficiently similar to allow of fair comparison. With regard to the decrease of power available behind the engine when running at high speeds, is not this due to the very large increase in internal friction and energy required to work the engine itself.
A short time ago I was running a four-cylinder locomotive, the gear of which had been changed from the Stephenson to the Caprotti poppet valve type, and over a trial of just over 1,000 miles the coal consumption compared with a similar engine with Stephenson gear and piston valves showed a saving in the neighbourhood of 71 per cent. and a water saving of about 18 per cent. Taken over some months’ actual running the coal saving was about 27 per cent. With regard to the much debated point as to the best at-off to work with, I think most drivers will notch an engine up until it begins to knock. One engine under my control had been converted to Caprotti gear, and it was found that out of several drivers who worked it there was one who could not pull the engine up within 20 per cent. Normally it could be run inside 10 per cent. Why that man could not work the engine in the same way as the other drivers, I cou!d never discover.
I should like to ask whether superheat is more when running well notched up and whether there is a greater efficiency for this reason on account of the smaller quantity of steam passing through the elements than when running in full gear. Referring to the back pressure diagrams, we had one rather peculiar case with piston valves, after changing from the broad ring type with compression release discs to the solid head type with narrow rings. It was found that with the latter type the steam chest air valves were continually breaking, and when drifting in full gear and when well notched up there was considerable chatter of these valves. It was found that if the gear was placed at an intermediate position the trouble did not occur. Compression had no doubt something to do with it.
One also gets high exhaust pressures after having run an engine some time, owing to excessive carbonisation in the steam ports, partly from faulty lubrication and partly from ashes drawn down the blast pipe, and that chokes an engine and leads to the loss of efficiency. The engine gets sluggish and fails to keep time.
The Author says he prefers broad piston valve rings. My experience is that the broad piston ring wears valve liners away very much more quicklv than the narrow, and maintenance in the shed bears no comparison with the narrow ring valve, and drivers book the valves blowing through frequently with the broad ring time and very seldom with the narrow. With regard to the lap of valves, we had a case with a mixed traffic engine, where we increased the lap by gin. After some weeks we got out the coal consumption, which was 49.41bs. per mile. Without the increased lap the coal consumption on the same work was 52.31bs., which gave practically 5 per cent. saving, and there was less choking and much freer running.
I think with Caprotti gears one gets increased climbing power, economy, and freer running, and a converted engine can certainly handle a much heavier train than previously and still show savings in fuel and water. Perhaps Mr. Windle could give us some comparisons between Caprotti, Walschaert and Stephenson gears.
C.S. Cocks (203): The old problem of high speeds, cut-off and regulator throttling has again come into discussion. By obtaining cut-offs of 10 to 15 per cent., the steam becomes throttled or wiredrawn through the valve opening, and, in consequence, the initial pressure is considerably reduced in the cylinder. Moreover, the exhaust passages also become restricted. Why not, therefore, throttle at the regulator, leaving the cut-off about 30 per cent., thereby giving a better steam distribution. With regard to cylinder clearance volume, I am aware that some of the modern engines have a clearance of 6 per cent., but I think a better figure is between 8 and 10 per cent.

Sanford, D.W. (Paper No. 273)
Development of the piston valve to improve steam distribution. 217-28. Discussion.: 228- 310.
Fifth Ordinary General Meeting of the 1930-31 Session held at Denison House, Vauxhall Bridge Road, Westminster, on Thursday, 29 January 1931, at 6 pm., H. Kelway-Bamber, M.V.O., President of the Institution, occupying the chair..
After considering the reasons which led to the introduction of piston valves the author directed attention to three features of interest connected with the flat slide valve which the piston valve had replaced: (1) the desirability of keeping the travel short, to avoid friction and reduce wear, notwithstanding the fact that this gave less satisfactory steam distribution; (2) to obviate the disadvantage of having large valves on which the steam pressure acted it was customary to bring the ports close together and make them long, thus increasing the clearance volume and the surfaces on which interchanges of heat between the metal of the cylinder casting and steam took place; (3) the advantage that the flat valve took up its own wear and, provided lubrication was satisfactory, the fact that it remained steam tight throughout its service.
With piston valves the first two defects mentioned were overcome, although full advantage was not always taken, but as regards the third leakage quite a different problem presented itself. The author showed the effect of leakage by diagrams on the screen and then proceeded to give illustrations of the best arrangements of packing rings to prevent such losses. The employment of a number of narrow rings, in place of one wide one he considered advisable.
Dealing with steam distribution, Mr. Sanford contested the assertions often made by advocates of poppet valve gears that the older types of gear, Walschaerts, Stephenson, etc., cannot give a satisfactory distribution because the cut-off cannot be made to occur sufficiently early without incurring release and compression early in the stroke. Diagrams were shown taken from an engine having Walschaerts gear and long travel valves, to illustrate these points and show that satisfactory distribution can be obtained with piston valves having long travel gear. The author suggested the improvement of existing engines fitted with short travel valves, by alteration, and then illustrated the adaptation of a piston valve, normal as far as admission was concerned, but which had double sets of ports for the exhaust. Indicator cards taken from an engine fitted with such valves showed marked improvement. In service the engines of this class, before being equipped, worked as a rule with not higher than about 35% to 40% cut-off; they can now run with valves in mid-gear without undue compression. An economy of 11 % in fuel on the basis of coal consumed per drawbar horse-power has been recorded.
The author concluded his paper by summarising his experience with the wear of piston valves in service, remarking this can be minimised by careful design and accurate workmanship. Further, whilst in no way disparaging poppet valve gear, he contended very satisfactory indicator cards can be . obtained with long travel valve gear or with double exhaust valves. Precis from Loco. Rly Carr. Wagon Rev., 1931, 37, 44-5..
Discussion H. Holcrott (228-9): We have to congratulate the Author very much on his Paper, which is brief, but to the point. As one who has had a long association with the long travel valve, I heartily endorse everything he has said about it. I believe that long travel valves originally came from America, but the first man to take them up in this Country was Mr. Churchward, on the Great Western Railway, somewhere about 1903. After experimenting with them for a short time, he adopted them on all his standard engines, and for a period of twelve years at least he was quite alone in this respect, none of the other railways making any real attempt to follow in his footsteps. I believe the next railway to take them up was the old South Eastern and Chatham, about 1917, when Mr. MaunscII started to design new engines. He not only introduced them into new designs, but re-built some old 4-4-0 type engines, and made an excellent job of them, mainly through adopting the long travel piston valves. These eng ines successfully ran the heavy boat train traffic between Victoria and Dover for a number of years, until the road bed was strengthened sufficiently to take heavier engines, after the formation of the Southern Railway in 1923. As soon as Mr. Maunsell started designing engines for the Southern Railway he adopted the long travel valve as a matter of course, and not until it became evident that another railway was working parallel with the Great Western, did the other railway groups begin to take notice of this piston valve. Although the L.M.S.R. is a late-corner in this respect, we welcome it to the fold at last!
There is a great deal of difference in the running of engines with the long travel valve as opposed to the short travel valve. I remember very well being struck by the difference in the working of heavy trains in this respect. In climbing up a long gradient the engine with a short travel valve would be worked with a full regulator and cut-off about 35 per cent. On coming to the top of the bank, and beginning to run faster, where the next section called for a speed of 60 to 65 miles an hour, it was curious to note that the driver actually had to let out the engine into 40 or 45 per cent. cut-off and ease the regulator. The reason for this was that with the faster running, there was insuffidient exhaust port opening, and therefore he had to increase the travel in order to get rid of the low-pressure steam, and he had to cut down the rate of steam entering the cylinders by throttling on the regulator. The driver of the engine with the long piston valve, on the other hand, coming up the bank with a 30 t9 33 per cent. cut-off would, on reaching the top, notch back to 25 per cent. or less, and so be able to run with full regulator the whole time.
These valves are termed "long travel" valves, but actually that is not a good description, because when the engines are notched up in their runqing condition, the travel of the long travel valve, or what should be called the" long lap" valve, is very little more than that of the short travel valve, perhaps only tin. or tin. more. That is simply because of the necessity of getting rid of the exhaust steam with the short travel valve.
It is quite true, as the Author has told us, that the condition of the piston valves is very much better with the long travel valve. I have often seen piston valves, particularly in three-cylinder engines, appear, after a long mileage, just as if they had been wiped over with an oily rag, and the steam ports quite free of carbon, right up to the exhaust tip. It is also possible with these engines, especially the three-cylinder engines, to run almost in mid- gear. I remember travelling on the footplate from Ton- bridge to Ashford on a fast train, where we had only 28 minutes to do 26½ miles from start to stop, and drawing a heavy train, and as soon as the speed got up to 60 miles an hour the three-cylinder engine was run very nearly in mid-gear for the whole way, with the regulator full open.
W.A. Lelean (229): The Author refers to the effect of long travel valves in giving freedom for exhaust. This feature, copied from the G.W.R. is now fitted to all standard engines in India. Comparing slide and piston valves, the former tended to wear in grooves transversely, but to keep tight by the wear being in fairly straight lines longitudinaIly. Piston valves as first used, through lack of satisfactory lubrication, wore badly, and it was not infrequent to find liners worn into a round-ended depression of iin. or even iin. deep, beginning at the end of the travel of the valve when in normal gear and continuing for the length of the valve travel. To get over this " restrained " piston valve rings were used; the latest form was a broad ring with a solid steel ring about ½in. broad by ¼in. thick let into each side of the ring, the recess being turned in such a way that the steel ring prevented the main ring expanding more than about .006in. In this way the ring having worn itself to a surface against the liner it was restrained from pressing against the liner and wearing it further. Restrained rings have run frqm 200,000 to 400,000 miles in service with practically no wear where the oil has been efficiently introduced in the steam by means of displacement lubricators. We have no figures of the leakage past piston valves and rings to check the actual loss of efficiency to which the Author has drawn attention, and. this possibly could be done on the actual engine by putting the piston valves under steam on the middle of the stroke and measuring the condensed water which could be obtained in a given time from the cylinder cocks.
A novel form of valve (the T.A.B. piston valve) is shown in attached Figs. (a) and (b). When steam is shut off the two valve heads AA are forced towards each other so giving free access from each end of the cylinder to the exhaust and also between the two ends of the cylinder through the valve. One may ask what will happen when the steam comes on between these two heads. There is no flying outwards and hitting the diaphragms BB, because the bushes DD on the valve spindle practically fit in the recesses of the sleeves of the valve heads, and at the same time the valve diaphragms practically fit into the inside of the heads. Thus there is a perfect trapping, giving a. steam cushion. We are told by the Metropolitan Railway, who have some experience of these valves, that the valves sink gently on their seats and that there is no hammer. The Author drew attention to the freedom given to the exhaust by the double-potted valve, i.e., to the free passage between the cylinder and the blast pipe as the exhaust opens, but he did not find this caused carbonisation, since on the return exhaust stroke it gave equal freedom for the gases to ,get out of the cylinder before compression began. I understand from conversation with the Author that he found by fitting thermometers to the steam ports of a cylinder not fitted with anti-vacuum valve that there was a rise of 150° to 190° of temperature when coasting as compared with running with steam on, showing that the hot gases from the smokebox, when their temperature is still further raised by compression at the end of the stroke, might well cause carbonisation in the ports. The T.A.B. valve, by giving the freedom between the ends of the cylinders through the collapsing valves, takes off this tendency to draw the smokebox gases into the cylinder ports.
S.J. Symes (231): I should like to add a word of appreciation of this excellent Paper. I think it is a very valuable contribution to' the Institution's Proceedings. I have been in touch with the Author a good deal with regard to the double exhaust port valve, which is, I think, the outstanding feature of his subject, but the Paper also brings out one or two points very prominently, and one of these is that nobody yet has got a piston valve with which they are satisfied. When I was in a drawing office 30 years ago, I remember our C.M.E. wishing to try piston valves; we got all the information we could from other railway companies, and I think there was such a divergence that we did not copy anybody. It is a curious thing that to-clay if one walks into any drawing office and discusses piston valves with the chief draughtsman, one will find that he has not yet made up his mind as to what is the best. We are all trying different valves and satisfied with more or less successful results until somebody else brings along something which is considered better. This is one of the main reasons why we are looking to the poppet valve. A point arising from Mr. Holcroft's remarks, although the Author himself did not labour it, is that experience with the long travel valve shows that one certainly does get more wear on the motion than with the short travel; it is not always working in the notched up position Mr. Holcroft has mentioned, but the travel is considerable, even at fairly high speeds. One of the advantages which we look forward to obtaining from the double port valve is that the motion will keep in better condition. It does not matter how good a distribution one gets when an engine comes out of the shops if the condition is not maintained over the period the engine is in service. It will be appreciated that piston valves weigh quite a lot, and one point against the double-ported valve is that the weight is somewhat increased. I have advocated for a long time that the possibilities should be explored of 'finding some materials which would be suitable for the valve spindle and cross head and piston valve heads, and still be yery much lighter than the material used to-day. So far, however" not much success has been attained in these directions.
T. Hornbuckle (232): There is no doubt amongst locomotive engineers to-day that a most important point in locomotive operation is the maintenance of a high standard of efficiency throughout the whole period that an engine is in service, and that period needs to be as long as possible. This will depend very largely on the efficiency of important parts, such as the slide valve, and I question sometimes whether these, which are often regarded as minor details, really receive the attention that they deserve either in manufacture in the workshops or in maintenance in the sheds. I think that one thing which the Author has made very clear is that there is a considerable field for improvement in working of the existing locomotive, without any very radical changes in design. -
J. Clayton (232-3): I have been extremely interested in the Paper, because it takes that practical view of things which our Institution so much values. The Author raised a point with regard to the slide valve, and said that such valves were in common use up to a certain time when pressures began to increase, and particularly when superheating came into vogue. I would refer in this connection to the case of the Midland Compound, which is probably one of the most successful engines for its size and power ever produced. The h. p: cylinders have piston valves. and the I. p. cylinders have slide valves. Thus the way for the steam to the chimney is tightly guarded by the flat slide valve, and I suggest that if it had not been for this fact the engines would not have been the success that they are.
With regard to clearances between the valve head and the liner, I am afraid that on the Southern Railway we could not run with anything like I/64in. clearance. I agree with the Author "that ithis- clearance does matter, but, on the other hand, if the valve head is too near the liner, it is very liable to score it when the guide bush wears. We run successfully the Schmidt type of valve. I admit that it is an expensive type at first cost; its workmanship has to be of the highest, but once the men have got trained into it, it does well, and to-day welcok for not less than 100,000 miles from. a piston valve ring, and have a set now which have actually gone through their third shop,'- representing a life of well over 180,000 miles.
We are also trying narrow valve rings; and these seem to be doing well up to the .preserrt. Such rings are the standard form of piston Valve ring in Germany, and appear very successful.
Holcroft referred to the long travel valve, or " long lap" valve) and mentioned its advantages. Another point is that such valves enable the drivers to notch up their engines close and to work with full regulator. With regard to extra -wear in motion parts, which Syrnes referred to, such has not been our experience. Mr. Churchward adopted on the Great Western Railway the " semi-plug" valve which has a restrained ring, in principIe not unlike the " no wear" valve referred to, but in which the ring springs out and adjusts itself to the side of the liner, in which position it is held by the steam pressure. The Author did not refer to the fact that most piston valves do not orovide for relief of water in the cylinder and special release valves have in such cases to befitted.
Mr. Lelean called attention to the Trofimoff valve recently described in these pages, and mentioned its success on the locomotives of the Metropolitan Ry. Other speakers included Messrs. Holcroft, Clayton, Symes, Rogers, Williams, and Poultney, the last-named emphasising the merits of poppet valves in comparison with either slide or piston valves.
A.H. Whitaker (237): The Paper has been a most interesting one, and will be a valuable adtdition to the Proceedings of this Institution. I have been trying to solve the problem of the lessened amount of carbon deposit with a double exhaust valve, but I have not managed it yet. I happen to have two such engines under my charge, and I intend to watch them very carefully. Perhaps I shall find that what has been said on this subject is not true. Possibly it has something to do with drawing carbon down the blast pipe and running without steam. There is one question I should like to put to Mr. Clayton. He spoke about the Midland Compounds, and said that they would not have been so successful if they had had piston valves for the low-pressure cylinders. I should like to know his reason for saying that. J. Clayton (237-8) replied: At the time when the Compound was introduced on the old Midland Railway the best piston valve known was the Smith type. While it was a good valve up to a point, it was not a very tight valve, and if such piston valves had been applied to the low-pressure cylinders of the Compound, those engines could hardly have been so successful because of the leakage.
J. Mitchell: (245-8) I regret that I was unavoidably absent from the meeting, but I take this opportunity of thanking the Author for his interesting and instructive Paper. I understand that during the discussion which followed -the reading of the Paper, attention was called to the N.W. (no wear) piston valve, and that a request was made by the Author for illustrations of these valves, so that they could be published in the Journal.
I have much pleasure in enclosing the illustrations, and it will be seen that the N. W. valve is a very simple one. In my early experience of locomotive valves, nearly forty years ago, the flat slide valves, in the majority of engines, operated in a vertical plane, and they were made of different kinds of materials.
It was very noticeable that the valves and valve faces of engines working on certain sections of the line appeared to wear fairly slowly and uniformly, but on engines of the same type working on other sections of the line, the valves and valve faces wore very quickly and unevenly. The causes of wear at that time could be classified under four headings, as follows
(a) Coasting.
(b) Priming.
(c) Varying quality of oils used.
(d) Inefficient lubrication.
I think we can say that coasting was and still is the principal cause of wear of valves and liners. Owing to the introduction of larger boilers and other improvements, wear due to priming has been reduced, but where bad water must be used, it is still a great menace to the life of valves and liners.
Oils are now obtainable which are practically perfect, but, unfortunately, they are not always used. Lubrication troubles have been considerably reduced owing to the adop- tion of suitable and efficient lubricators. If the distance traversed in coasting, or drifting, is not of any great length, very little harm is done, but on some railways the coasting distances are considerable, and if some method is not provided for the prevention of ashes, etc., falling into the valve chests and cylinders from. the srnokebox, then, undoubtedly, valve and liner wear will be severe.
The action of some valves increases this tendency to bring down material from the smokebox, but in well- designed valves this tendency can be very greatly reduced. Good drivers, who follow carefully the instruction:—- "Never completely shut the regulator until the engine comes to a stop"—-can and do very greatly assist in maintaining the valves, valve liners and main pistons and cylinders in good condition.
A slightly open regulator is particularly necessary when coasting.
I have heard of a simple and effectual method of overcoming the drawing down of ashes, etc., into the valve chest. It consists of a small pipe, carrying live steam, and which is led into the blast pipe, and acts upwards. The mouth of the pipe was not more than one-eighth of an inch in diameter, and the supply of live steam was so arranged that it was automatically turned on when the regulator was closed. It was found that the steam supplied by this small pipe' was sufficient to counteract the down draught in the blast pipe.
I have known of serious valve failures that have occurred, due to incorrect alignment of the engine valve gear, to bent valve spindles, bent during the fitting together of the valves, and to sudden stoppage of lubrication. It has been my experience that wear is a very indeterminate factor, and no real standard is laid down anywhere between a completely worn valve and a valve considered fit for service.
In this connection I would suggest that providing an engine does not suddenly blow very badly, owing to a fracture, for example, the valves can be considered good for service until an opportunity arises, such as washing-out day, to examine them.
One difference between N.W. valves, which are of the restrained ring type, and valves of the expanding ring type, for example, Schmidt broad ring, as well as the narrow ring types, etc., is that a restrained valve is better tested when under full steam pressure, whereas valves with rings having unlimited expansion (Schmidt or narrow ring, etc.), must be withdrawn and the rings and liners inspected. The provision of broad rings permits the use of liners with long, rectangular ports, which can be cast neatly to size, and are easily machined, whereas with narrow rings it is essential to use liners with short, diamond-shaped ports, which are expensive to machine in the ordinary way, or require expensive machinery to produce. With diamond ports the bridges between the ports are necessarily thin, and are very liable to fracture when being handled, inserted into the cylinders, and when under temperature variations in service.
In a recent engineering publication, the chief engineer of a railway published an illustration of an apparatus which he had perfected for drawing piston valves from his engines. In that article it was stated that whereas an hour or more was necessary to draw a piston valve, now, by the use of the simple apparatus, the time of withdrawing each valve had been reduced to ten minutes or thereabouts. This evident trouble of withdrawing valves (which it is obvious does occur, or mechanism such as referred to would not be designed to overcome it), has led the shed fitter to expect a valve to be tight to withdraw, and -when valves such as the N.W. are used, he cannot understand why they lie loose in the liner, and are capable of being instantly withdrawn by hand. Instead of investigating whether this condition is normal, he jumps to the conclusion that there is something wrong with the valve, and immediately commences tarnpermg with the parts, or scrapping perfectly good rings and making replacements which are unnecessary. There is a function of a piston valve which is little understood, and that is, valve rings must be capable of collapsing when, owing to priming or condensation, water has accumulated in the cylinder. A comparatively recent case came to my notice, in which a, locomotive engineer argued that he could not see that such a provision was necessary in a modern locomotive, and accordingly made a set of solid piston valve heads which were a fit in the liners. After two months' service a steam chest cover fractured. He is now convinced that a collapsing valve ring is an absolute necessity. In going round a number of shops, I have been impressed by the number of machines turning out valve rings and liners, and I find it difficult to reconcile the statements made to me that little or no trouble with valves and liners is experienced on their particular railway, and that very few new valve rings and liners were required for replacements.
I feel sure that if a thorough investigation were made, and a count taken of the liner castings and ring castings issued for machining in one year, and the quantities of rings and liners produced from these castings were divided amongst the engines in service, interesting and surprising figures would be obtained.
During the last few years over two thousand engines have been fitted with N.W. valves, and it might interest members of the Institution to know that in not a single instance has a valve ring broken in service. Given normal conditions, the N.W. rings run hundreds of thousands of miles, and the liners give equally excellent rnileage,
Sixth Ordinary General Meeting of the North Eastern Centre (Session 193031) was held at the Hotel Metropole, Leeds, on Friday, the 27 February, 1931, at 7.15 p.m., Mr. E. de H. Rowntree taking the chair.
Fifth Ordinary General Meeting of the 1930-31 Session of the Manchester Centre was held in the Literary and Philosophical Society’s Room, George Street, Manchester, on Friday 13 February, 1931, the chair being taken by Mr. J.N. Gresham.

Vallantin, R.G.E. (Paper No. 274)
Compound locomotives of the P.L.M. Rly. 252-79. Discussion: 280-303; 311-13. 9 illus.
Sixth Ordinary General Meeting of the Institution (Session 1930-31) was held in the Hall of the Institution of Mechanical Engineers, Storey’s Gate, Westminster, on Thursday, 26 February 1931, at 6 p.m., Mr. H. Kelway-Bamber, M.V.O., President of the Institution, occupying the Chair. M. Vallantin, Engineer-in-Chief of Materials and Traction of the PLM. Translated from French.
When the Author joined the staff of the P.L.M. Rly. Company in 1907, eighteen years had elapsed since its first compound locomotives had been designed and built. Simple expansion working had been entirely given up, and of the 845 locomotives ordered between 1 January 1890, and 1 January 1907, no less than 835 were compounds. In addition 140 simple expansion goods locomotives had been altered to compound working. It seemed, therefore, as if the Company had definitely made up its mind on the matter.
However, a little before 1907 superheaters made their appearance, and those who introduced this great improvement claimed that expansion of the superheated steam took place in the cylinders without condensation occurring and that there was thus no longer any need to have recourse to compounding, with the accompanying high steam pressures, in order to reduce, if not to do away with, such condensation, the cause of the high steam consumption of simple expansion locomotives.
Discussion: H.N. Gresley: (280) stated that Vallantin had given us the results, he has not told us very much about how he obtained these results. He had not informed us how these engines were worked, what percentage of cut-cff was used in the high-pressure cylinders and what percentage of cut-off was used in the low-pressure when the locomotives were running under very heavy load and on a fairly level road, and on the other haild when they were working very heavily uphill, perhaps at slower speeds, and when :hey were running downhill at fast speeds. I should like to know the percentage of cut-off both in high and low pressure which is used.
The next point is what degree of superheat was obtained in the high-pressure cylinders, and particularly what amount of residual superheat there is in the low-pressure steam chest. I am particularly interested in this, because in the case of the compound engine No. 10000, which is running in this Country, I have not yet got any satisfactory data as to the amount of residual superheat we are getting in the low-pressure steam chest.
Then there is another question I want to ask, and that is the volume of the receiver between the high and low pressure cylinders. Will M. Vallantin tell us, moreover, what is the effect of increasing the volume above the normal that he employs, and on the other hand, what is the effect of decreasing it? What does he regard as the ideal ratio of the volume of the high-pressure cylinder, the receiver and the low-pressure for, say, boiler pressures of 227lbs. per square inch.
Lastly, this Paper is entitled Compound Locomotives of the P.L.M. I know that Vallantin has not had very much time, but it would be of interest to know what sort of results are being obtained with the latest compound on that line, the Schmidt-Henschell engine, which I understand has been running for a few months on the P.L.M. That would be of very great interest to us here, particularly on account of the interest now being taken in high-pressure locomotives.
Sir Henry Fowler (282) This Paper is a very valuable one, in that so much detailed information has been given. With regard to the point which Mr. Gresley has raised as to the degree of superheat in the high and low-pressure cylinders, if I remember rightly, with our own compounds there is about 80°F of superheat in the receiver between the high and low pressure, and therefore from what M. Vallantin says I take it his is a good deai higher.
There is one interesting point in the Paper which has nothing to do with compounding or superheating. M. Vallantin says it was thought that the use of four cylinders would ensure greater steadiness of the locomotives at high speed, and I should like to know whether that was actually confirmed.
One is struck when looking at Table I. by the very high coal consumption per drawbar horse-power hour, and if M. Vallantin has some more recent figures in this connection I am sure we should he very glad to have them. Later on in the Paper it is stated that the compounds are not so good in starting. We have always found—and it has been placed on record in one of the previous papers read on this subject—that the three-cylinder compounds of the L.M.S. are good in starting.
I come now to a financial question—and the financial question is, when all is said and done, one of the most important. Figures are quoted in francs by the Author, and I should like to know at what value the franc has been taken at those times. I judge from what is said in the Paper that the francs thcre are taken at their present-day value, but I think it would be of interest and would lend value to the Paper if that was stated.
Compounding is an intensely interesting subject. I was associated with it as Chief Mechanical Engineeer of the Midland, and afterwards of the L.M.S., for just twenty-one years, and of course during that time I had considerable experience with the three-cylinder compounds. One of the first questions which may be asked is why, if we found them so successful (and that we did is evidenced by the considerable number of compound engines built by the L.M.S. after the amalgamation), we have not gone on building them. It must be remembered that M. Vallantin has a great advantage over us with regard to the loading gauge, and whereas with smaller engines these difficulties are not felt so much, there are certain difficulties which arise in the case of a larger engine and when one is tempted to try to get bigger cylinders outside.
Early in 1914 I had the honour of reading a paper on superheated steam in locomotives before the Institution of Civil Engineers. Naturally I have looked that paper up, and I find that M. Nadal, of the French State Railway at that time, said that all his new engines were compound engines, and that he had found by superheating he got an advantage of from 10 to 15 per cent. When I became Chief Mechanical Engineeer of the Midland Railway in 1910 we turned our attention to the question of superheating 0u.r compound engines. I referred to this in a discussion on a paper which Mr. Hughes read in this hall in 1910. We found then that generally we were getting an advantage of about 7 per cent. in coal consumption with our compounds as compared with simple engines doing similar work. We had not at that time very many results, but that was speaking generally and I believe the figures were substantially correct. In 1914, as I say, I read a paper on superheating in locomotives, and there we found, as a result of a series of tests over the 197 miles from St. Pancras to Leeds, taking wholeday runs, that we got an advantage of about 26 per cent. in coal and about the same in water by applying superheating to our compound engines.
One thing we did, however, was to take advantage of superheating and drop our pressure 20lbs. The original compounds had a presure of 2201bs., and we had very considerable trouble for reasons which in new boilers have disappeared, and we found a considerable advantage in boiler maintenance by dropping the boiler pressure from 2201bs. to 200lbs.
W.A. Stanier (283): I should like to express my thanks to you for inviting me to hear this most interesting Paper by one of the most eminent locomotive cngineers on the Continent. When reading the Paper I could not help being struck by the fact that M. Vallantin has apparently come to the conclusion not only that compounding is desirable, but that pressures rather higher than are normally used for locomotives are also very desirable. In this Country, where we are confined to a width of not more than nine feet across the cylinders, we have to devote a great deal of thought to the design and construction and-what is more important -the maintenance of higher pressure boilers. It is interesting that the French engines which the Great Western purchased in, I think, 1904, had 227lbs. pressure and were not superheated, and as a consequence a great deal af trouble was experienced with condensation in the receivers. They were afterwards fitted with Great Western boilers and superheaters and to a great extent the condensation difficulty disappeared ; but they were still limited, because the nine feet width restricted the size of the cylinders which could be fitted, and as a consequence the power of the engine was much too small for the increased requirements on the Great Western Railway. The Great Western had therefore to develop four-cylinder simple engines.
As a matter of interest and not as a matter of comparison— because in making comparisons one must have similar conditions— I have turned up the records of some trials of the Castle class, and I find they give a figure of 10lbs. of coal per 100 ton-miles. The coal per i.h.p. is 2.1 lbs. We must remember in considering these figures that that is with South Wales coal which has not been damaged by transport by sea and by tranference from coal tip to ship and out of the ship again, and so on. There is one question I should like to put to M. Vallantin. He has not told us, I think, whether the latest Pacifics he is using are fitted with independent valve gears for each cylinder or whether he has adopted an arrangement for operating the inside valves from the outside valve gear.
It would be interesting to know if M. Vallantin has tried out his new “ Pacific ” engines on a train with a dynamometer car, and if so, what is the maximum maintained drawbar pull he gets at 70 miles an hour.
H. Holcroft (284) We have listened to a very interesting Paper on compounding, which describes what are perhaps the most complete tests that have ever been made between compound and simple engines, and I think it is of very great value from that point of view There is just one question I should like to ask M. Vallantin, and that is the reason for employing four-cylinder simple engines as freight engines. It seems to me that in those conditions the fourcylinder engine is operating under the very worst conditions ; the four-cylinder locomotive is essentially an express engine, because of the improved balancing and other factors ; to get the very best out of a simple high-pressure engine three cylinders are necessary, for then one does get some very definite advantages from the multicylinder arrangement which one does not get from the four-cylinder simple when it is used as a slow speed engine.
As far as the results are concerned, it seems quite definite that compounding has been a great success on the P.L.M., but, of course, we are most interested in this subject from the point of view of the application of c m - pounding to this Country, and to bcgin with we are up against the gauge question. Another factor to be considered is that in this Country the traffic conditions are not always favourable to compounding, especially from the freight point of view. Where an engine is confined in sidings for a good deal of its time or is running without steam for long distances downhill, the opportunity for saving in cad consumption which may result from compounding, is very much reduced, as a lot of stand-by losses occur.
Mr. J. Clayton (285) This Paper comes to us at a very opportune time; in the first place because we have recently had read before our Institution another interesting paper by Mr. Selby (* See Journal Vol. XX., No. 95, page 287) on the same subject. These two papers together will constitute a book of reference which I sincerely hope may give to the question of compounding a new lease of life in England, because I myself feel that we in this Country have tended to relegate it too much to the dim past; we have forgotten, in our anxiety to meet the ever-increasing demands made upon us by the trafic department, that there is still another side of the question, namely that of the more economical use of the steam. Now that we have the advantage of superheating and all the modern improvements in materials, we should try to give by the compound principle a new lease of life to the steam locomotive. It is said that a country always gets the government it deserves. Can it be said with equal force that a country always gets the form of traction best suited to its needs?
I am glad Mr. Stanier has told us something about the Great Western experience, because many of us have wondered why Mr. Churchward, one of the most shrewd locomotive engineers of his time, did not adopt the compound principle. The same thing applies to the L.M.S. experience; why, when the Royal Scots were produced, some consideration was not given to compounding, after the wonderful experience the old Midland and the L.M.S. have had with their compounds.
We are, of course, aware that conditions- vary, and those in France are not quite comparable to the British. When we go to the Continent we generally travel by selected trains from Calais to Paris or from Boulogne to Paris, so that we see largely their best work. When you are comparing the locomotives of different countries, however, you must take the locomotive practice as a whole into consideration. The best English practice is, in my opinion, as good as the best French practice, but conditions vary and we in England have not, as has already been pointed out, the added advantage of the more ample loading gauge which exists in France. Notwithstanding this, however, locomotive engineers in this Country would do well to turn their attention again to compounding.
H. Chambers (286): The point that stood out very prominently in the particulars quoted by the Author in Table I. is the very high coal consumption per drawbar horse-power hour at the tender drawbar. In the case of the compound engine, 5.931bs. of coal is quoted and for the simple engine 4.95lbs. of coal per d.b.h.p. hour. It would be interesting to, know what class of coal was used on these tests as compared with coal on British railways. Probably in this Country the average calorific value of coal for express locomotives is in the region of 13,500 to 14,000 B.T.U.
Again, the water used per d.b.h.p. hour for the compound engine is given as 43.41bs. and with a simple engine 35.2lbs. This again is very high as compared with standard three-cylinder compound figures obtained by the LMS. dynamometer car tests, viz., coal per d.b.h.p. hour, 3.491bs., and water per d.b.h.p. hour, 29.6lbs. The Author undoubtedly makes a strong case in favour of the four-cylinder compound engine, and I would like to hear to which of the three following important factors he chiefly attributes the superiority of the compound
(1) Increased ratio of expansion in compound cylinders.
(2) Smaller temperature range in each cylinder, thus reducing heat losses.
(3) Reduced leakage on pistons and valves due to lower range of pressures on both sides, particularly bearing in mind the wear that inevitably takes place on the piston valve rings, liners, and main pistons, while in service.
This means, of course, that in the compound engine leakage past the piston valves or main pistons passes into the low-pressure receiver and is not entirely lost, as in the case of simple engines.
Another interesting point is that although in the earlier stages a three-cylinder engine was considered, this was not proceeded with as it was stated that a four-cylinder engine would result in a better balance and therefore smoother running engine at high speeds. As a matter of fact, a three-cylinder engine with the cranks at 120° has practically as good a turning moment as a four-cylinder engine, and in this Country three-cylinder simple engines are running with great success. Assuming a three-cylinder engine can be designed to meet the requirements, the following points are strongly in its favour :-
(1) Decreased number of cylinders resulting in less heat losses.
(2) Decreased number of sets of motion parts resulting in lower first cost to produce, and also in less maintenance charges for examination at the sheds.
Mr. E. L. Diamond (287): Direct comparisons between simple and compound locomotives, such as those detailed in the Paper, must be of the utmost value to the administration conducting them, since they take into account almost every circumstance of their operation. But for that very reason they cannot be made the sole basis for any generalisation. The first principle of all scientific investigation is to separate variable factors and evaluate them one at a time. In the case of locomotive practice this may seem to be a very academic procedure, but until it is adopted the century-old controversy on this subject may well last another century. Mr. Chambers emphasised this point when he asked what proportions of the economy obtained by the Author were attributable to three factors which he named. There may be half a dozen other factorsas well, and moreover they will vary in importance according to the design of the locomotive and the conditions under which it is run. For example, at high speeds condensation is, so far as I have been able to ascertain, practically eliminated in superheater simple-expansion locomotives owing to the drastic throttling at admission. Under such conditions the reduced temperature range in compound cylinders is an advantage that becomes discounted. Again, with modern designs of valve gear full expansions of the steam can be secured in simple-expansion locomotives, and I am not surprised that the Great Western Railway found little advantage with their French compounds since their own engines and their running conditions were such that full expansion of the steam could be obtained in simple-expansion cylinders.

Journal No. 101

Andrews, H. Ivan (Paper No. 275)
The possibility of condensing on locomotives. 336-66.  Disc.: 366-78; 537-9. 3 illus., 15 diagrs.
Fourth Ordinary General Meeting of the 1930-31 Session was held at Denison House, Vauxhall Bridge Road, London, on Wednesday, 17 December 1930, at 6 p.m., Mr. J. R. Bazin, Past-President, occupying the chair. Brief abstract in Locomotive Mag., 1931, 37, 24
Such a condenser must possess the following characteristics :-
( I ) Light weight.
(2) Small size.
(3) Low initial cost.
(4) Low maintenance.
(5) Reliability.
(6)Constant performance under varying conditions.
(7) Low power demand for draught.
(8) Adequate capacity.
(9) Reserve capacity for short periods
The question of draught (7) is a matter of primary importance, since in some designs so much power has been allowed for blowing, both for condenser and boiler, that the advantages claimed for condensing have almost been nullified. Above all, however, comes the question of capacity. It must first be agreed what standard of capacity is to be taken; the "Cole" evaporation of the boiler, as used in America, affording a reasonable basis of calculation.
The normal condenser capacity must be something between
The air cooled condenser can be built in many forms, is comparatively cheap, light, and reliable, and is easy to maintain. Its chief disadvantages are its size and its small overload capacity, which necessitates an unduly large condenser to cope with the overload period.
The evaporation condenser, though excellent in design and performance, has the great disadvantage of requiring a supply of water almost as great as the boiler feed, hence it is unable to justify itself as regards weight and space. At z7ins. of vacuum .851bs. of water is required to condense Ilb. of steam. This water may, however, be of poor quality, hence this type is favoured in districts where the water is cheap but troublesome.
The jet condenser is an excellent method of condensing the steam, but difficulty is experienced in discharging the surplus heat. Since this heat is lost in two stages it is obvious that the second temperature drop cannot be so great as the drops in the previous cases, hence this apparatus is more bulky and requires more cooling surface than the others. Its chief advantage is its reserve capacity, but this is offset somewhat by its considerable weight.
In this Paper it is proposed to consider the direct air cooled condenser for normal operation, since this appears to hold out the greatest possibilities of development. This condenser will be designed for the normal rating, and
It remains to compare this form of condenser with some more orthodox design, for which purpose reference is made to the design of a turbine locomotive prepared by the Standing Committee on the design of Turbine Locomotives of the International Railway Fuel Association (see Fig. 17) and presented at the conference at Chicago in 1929~. In this engine two condensers are provided, one at the front and the other at the rear, and at least, as far as size is concerned, form the most prominent feature of the design. The leading condenser is of the honeycomb or " radiator " type, through which air is drawn by three large bladed fans, and expelled through the roof, somewhat after the manner of the Ljungstrom condenser. The rear condenser is of the jet type, acting partially as feed water heater, but has also a large radiator to cool the excess of condensing water. Even allowing for the air temperature of 100°F, which has been assumed, it must be admitted that this apparatus is extremely bulky, and possesses considerable complications liable to require very careful handling to ensure satisfactory operation in service. As 1 million ft³. of air per minute are necessary to condense the exhaust of the 3,000 h.p. turbine fitted, it is obvious that an enormous quantity of heat has to be dissipated. This is found to be too great for a single bank of air-cooled tubes as in the preceding examples, but it can be handled with the assistance of a second bank operating with only about a 40 m.p.h. draught when 60 m.p.h. is supplied to the leading section. The remaining illustration (Fig. 18) shows the same locomotive rearranged, so as to accommodate the alternative form of condenser, and it is at once evident how much simpler the arrangement becomes. Owing to the electric drive, the sudden overloads obtained with the normal locomotive do not occur, so that in this case the condenser can be rated higher than usual, and very little water need be carried for additional cooling. In the rearrangement the turbine and generator, etc., have been moved to the rear, where they are more accessible, allowing the whole of the condenser to be mounted at the front of the boiler, but this is by no means essential, as the two sections of the condenser may be mounted at opposite ends if it is desired to operate the locomotive continually in either direction. The air preheater has been left in the same position as before.
Finally, the Author’s thanks are due to the many people, both in this Country and America, who have contributed their experience to the compilation of this Paper, and to Professor W. E. Dalby for permission to include the experimental results.
Mention must also be made of the other members of the Committee on the design of Turbine Locomotives of the International Railway Fuel Association, also the members of the Committee on Front Ends, Grates and Ashpans, to whom the Author is indebted for a great deal of valuable material which has been borrowed from the various reports.
The criteria were: light weight, small size, low initial cost, low maintenance, reliability, constant performance under varying conditions, low power demand for draught, adequate capacity/reserve capacity for short periods. The Cole evaporation rate should form the basis for the last-named.
Discussion: Sir Henry Fowler (367) I think the way in which the Author has dealt with his subject, and the experiments he has carried out, presumably when he was at South Kensington, are very delightful. I should like to ask him—I ought to know, because I dealt a good deal with wind channels when I had to do with aircraft—whether there is any similarity between the effect with air and the effect with water. I had the opportunity, many years ago, of seeing the very interesting experiments which Dr. Hele-Shaw carried out with regard to the streamlining of ships, and if there is any similarity as between air and water it would be very interesting, and should make it easy to apply the knowledge gained in one field to the other.
Another point which I think is of great interest—and this again goes back to my aircraft work—is that Dr. A. H. Gibson found that by painting metal surfaces—in this particular case it was a, question of an air-cooled cylinder he was able to get a very great increase in the heat transfer; as high, I believe, as 15 per cent. with an ordinary japan paint. That, of course, would add very materially to anything which could be done with a condenser of the type in question.
The question of the use of a condenser on a locomotive will probably be dealt with by some of the subsequent speakers, but as our Chairman has said, and as the Author himself mentioned at the beginning, there are two things which trouble us; one is the question of weight, and the other is the question of cost. Some of us who have looked into this question of condensation have been faced more particularly, perhaps, with the question of cost. I should be glad to know if the last locomotive the Author showed us on the screen has actually been built. I take it it has not, but it would be of great interest to know if the International Railway Fuel Association have actually built such an engine. and, if so, whether it has been operated. One of the points which struck me about it was its apparent great length, which would in this Country at all events offer considerable difficulties. Even now we occasionally have to build a 1ocomotive somewhat longer than we like, because its weight concentration involves certain difficulties. I feel the Author has given us information and-in the early part of the Paper-figures which should be of the greatest use to> those of us who have to consider from time to time in the drawing office how we can continue to improve our locomotives
H. Chambers (368): I have been very interested in this Paper, particularly in connection with the possibilities of economies in locomotis e operation by the application of a condenser. The Author states that with the condensing type he would expect to get a 42 per cent. decrease in fuel consumption. This figure is based on the theoretical Rankine cycle. I would draw the attention of the meeting to an interesting lecture given before the Institute of RIechanical Engineers by Lawford Fry, who had prepared what was called a modified Rankine cycle, and which was more in accordance with present practical working conditions of a steam locomotive.
For the purpose of thic, discussion I propose to consider two engines, one a non-condensing and the other a condensing type, bath working at 200lbs. boiler pressure. Applying the modified Rankine cycle I find that the condensing engine would give a fuel economy of round about 30 per cent., which is considerably less than the figure of 42 per cent. mentioned by the Author. Again, by introducing a condenser, other fittings have to be provided on the locomotive, such as a blower to supply air to the condenser, and draught fan in the smokebox, and to cover these I estimate that 10 per cent. would be taken off the economy, thus leaving only 20 per cent. saving in fuel for the condensing engine.
Now, let us suppose a locomotive runs 50,000 miles 1)er annum and burns 451bs. of coal per mile. Assuming coal to cost 18s. per ton on the tender, the cost of fueI would be £900 and 20 per cent. of £900. is £180, So that by fitting a condenser we shall save £180 per year. The Author may not agree with me, but I suggest that to apply the condenser he mentions, with its auxiliaries, will cost something in the region of £2,000. If we take the interest, dqreciation and maintenance charges on that figure, we shall find they amount to about A300 to A350 per year. We have thus a credit of A180 as compared with a debit of A.350, and the conclusion at which I arrive, therefore, is that the heavy cost of the condenser nullifies the benefit of the saving in fuel it would secure. I do agree with the Author, however, that there are geographical conditions, particularly in the tropics, where water is more scarce, and therefore more valuable, than in this Country, and there the position will be more favourable so far as the condensing locomotive is concerned. I would like to say that it gives me great pleasure to be able to take part in the discussion following such an interesting Paper, which I think we shall all agree is of particular interest, especially to the younger members present.
W. A. Lelean (369): The most striking portion of this Paper to his mind is that in which the Author describes the experiments he conducted and the valuable experimental data obtained showing the flow of air around varying arrangements of tubes and the resulting effects on the transfer of heat. If, as Sir Henry Fowler suggested, there is a corresponding difference in the flow of water as of air, it opens a useful field for investigation as to the effect of arranging the tubes in the boiler in differing ways and with differing pitches. It might be possible to reduce the number of tubes without any loss of efficiency.
H. Hokroft (369-70): I have enjoyed this Paper very much. The Author has given us some valuable data and he has put forward his Paper rather as theoretical work for the consideration of practical men, who must decide whether and how they can apply it in their every-day work. We are often faced with problems of this sort, and the first thing we have to do is to draw up a sort of balance sheet, to determine whether we are going to be better off as a result. We may have an economy of, let us say, 20 to 30 per cent. in fuel, and the boiler may he kept in better condition and not scale up so much, and probably the boiler repairs will be reduced. On the other hand, we have a very much higher capital cost on which interest will have to be paid, and we have very much more in the way otf apparatus to maintain. We have to keep a condenser which is actually on wheels in a bottle-tight condition, so that there shall not be the slightest leakage of air ; otherwise the economy will disappear. Then we have numerous flexible joints and other things to maintain. Whatever is done there must be a great deal more maintenance, and so one tries to construct a balance sheet, and put on the one side the savings and on the other side the increased charges involved; one will find the position is not very attractive, though, as has been said, there are places where the water conditions or the fuel conditions would make the position of the condensing engine more favourable.
A condensing engine will either take the form of an engine and a tender with joints between, or else it will be wholly on one frame with power bogies. Flexible steam pipe connections are required in each case, and precautions must be talieii to prevent leakages, either of steam under pressure or of air getting into the pipes below atmospheric pressure.
The Author has pointed out the advantages of the supply of air under pressure to the ashpan, as opposed to suction by means of a fan in the smokebox. I agree that is the best method of applying it, because one can do it with very much less horse-power, and a smaller fan is required on account of the cold air having very much less volume than the heated gases in the smokebox. But that method is all very well where one has a mechanical stoker or pulverised fuel or oil fuel, anything not requiring the opening of the fire door, but with a hand-fired engine, with a coal consumption of 45 to 50lbs. a mile, it means seven or eight shovefuls a minute, and while this was being fired one would have to stop the flow of air, as otherwise the flames would come out of the firebox door. This means that during a large percentage of the time there will be no air passing through the fuel bed, that is to say, one will get conditions every few seconds in a deep fuel bed where there will be formation of smoke or carbon monoxide, therefore one will not have good combustion, so that I think it is very much better to have a suction fan in the smokebox, although it may not be quite so efficient.
J.R. Gould (370-1): The tendency of the locomotive engineer has always been to try and get the same efficiency out of a locomotive that is got out of a marine engine, and to a large extent he has been successful. The condenser has not proved a successful addition to the locomotive for the simple reason that it lacks the necessary supply of cooling water, which, when replaced by air, does not give the same results due to the change in temperature of the air, caused by climatic conditions. In hot countries, and even in England when we are attacked with a heat wave, an air-cooled condenser would be liable to give trouble. Also the additional bulk and very heavy first costs, to say nothing of the extra complication and attention required, would not warrant the adoption of a condenser. Again the exhaust steam injector or Weir feed-water heater, which give such excellent results, would be done away with. These claim a saving of 10 per cent. One would also have to do away with the superheater which has become standard practice, and is, inversely, a condenser itse!f. the marine practice of compounding really gives excellent results on a locomotive, actually showing a saving of 25 per cent. in fuel, a fact which should not be ignored. Without any shadow of doubt every locomotive should be compounded, either with two or three cylinders, with the exception of perhaps the shunting engine. In addition to this, a feed-water heater, such as the Weir, which absorbs the waste heat from the exhaust steam, acts very well as a condenser, and should give, under all circumstances, as much efficiency as a condenser, as no on~~*heater, which give such excellent results, would be done away with. These claim a saving of 10 per cent. One would also have to do away with the superheater which has become standard practice, and is, inversely, a condenser itse!f. the marine practice of compounding really gives excellent results on a locomotive, actually showing a saving of 25 per cent. in fuel, a fact which should not be ignored. Without any shadow of doubt every locomotive should be compounded, either with two or three cylinders, with the exception of perhaps the shunting engine. In addition to this, a feed-water heater, such as the Weir, which absorbs the waste heat from the exhaust steam, acts very well as a condenser, and should give, under all circumstances, as much efficiency as a condenser, as no vacuum pump or blower is required. blast to be found by experiment.
D.R. Carling (371-3): I wonder if any advantage could be obtained by making the tubes more or less of aerofoil section, in which czse it might be possible to obtain a much greater percentage of really effective cooling- surface. However, If the tubes were wanted in any real quantity, they should be no more expensive than circular tubes, though they are not quite so strong and might have to be a little thicker.
The case for the condenser-fitted locomotive depends, I think, entirely on local conditions. If fuel is cheap and water good, as is generally the case in England, I do not think the chances are favourable for the employment of a condenser. There is a saving in boiler maintenance by using distilled water, but that is probably offset by condenser maintenance. On the other hand, it is also possible that the increased capacity of the locomotive may be worth while in itself. An increased capacity in the locomotive of 10 per cent. may make all the difference between its being useless or useful; it may make it capable of working a given service satisfactorily instead of unsatisfactorily. One has only to be on the footplate of a locomotive which is overloaded or in bad condition-which generally means the boiler is in bad condition-to know what a dreadful business it is to try to keep time. But if the conditions are really bad, with water at the price mentioned of $2.50 a thousand gallons-and I have heard of it reaching the price of $10 a thousand gallons, by the way-one really does have an obvious case for condensing; but on the other hand one may also have a case for the Diesel engine, and that is an alternative nhich has to be considered. In each case -condenser-fitted loconiotive or Diesel-the trouble is the capital cost, and it remains to be seen which is really the better, according to local circumstances.
I beliexe on the Trans-Australian Railway water is not only scarce, but alqo very bad. A condensing locomotive, therefore, would probably be a godsend to the operating department there. But in the majority of cases we have no such conditions to deal with, so that it seems the application of the condenser is definitely limited and must be decided on economic grounds. It must, of course, always be remembered that if a locomotive can be kept in service by putting a condenser on it, a great saving may be effected in that may; foi instance, one may be able to get double the mileage out of it in a district where the water is bad, so that even if it were no more economical than an ordinary locomotive, there would be a considerable saving there. Fuel economy does not seem to lead to such very great saving. As Mr. Chambers says, with 50,000 miles a year in Great Britain one gets a saving of  £180, but one might save ten times that amount by having the engine more reliable instead of more economical, or perhaps by keeping it in service longer. An appliance which effects a 10 per cent. saving in fuel may go' wrong and keep the engine to which it is fitted out of service for 10 per cent. of its time. I do not know what the earning capacity of a locomotive is said to be ; it may be £100 a day, and if the locomotive is out of service for two days, all the fuel saving is gone ! Reliability is the great thing to aim at. In some cases a condenser may make the locomotive more reliable, but in most cases I am afraid it will not.
Fourth General Meeting of the 1930-31 Session of the Scottish Centre was held in the Societies' Room of the Royal Technical College, Glasgow, at 7.30 p.m. on Thursday, 22 January 1931, Mr. G. W. Phillips, the Chairman of the Centre, presiding.

Clayton, T. (Paper No. 276)
Wagon repairing by the Central Argentine Rly. 379-432.
Third Quarterly Meeting of the South American Centre (1930 Session) was held in Buenos Aires on Friday, the 19 December, 1930, Mr. R.E. Kimberley occupying the chair.

Chambers, H. (Paper No. 277).
Improvements in water pick-up gear for locomotives. 450-64. Disc. : 464-72; 787-93 + 3 folding plates. 7 illus., 9 diagrs.
Seventh Ordinary General Meeting of the 1930- 31 Session, being- also the Twentieth Annual General Meeting of the Institution, was held at the Institution of Mechanical Engineers, Westminster, on Tuesday, the 24 March, 1931, at 6 p.m., Mr. H. Kelway-Bamber, M.V.O., President of the Institution, occupying the Chair.
After narrating the early adoption of such to enable locomotives to undertake long runs by Mr. John Ramsbottorn, the locomotive engineer of the L. & N.W. Ry., Chambers proceeded to describe in detail the most approved arrangements used on the LMS Ry., devoting particular attention to the many improvements which had been incorporated in the latest forms used on that line.
A verv interesting account was given of the results obtained in experiments made with a special tender constructed at Derby Shops, in which an observation chamber was provided. Careful notes made in actual working had enabled great improvements to be made in the provision of anti-splash devices to enable the maximum amount of water to be taken whilst an engine is passing over the trough with a minimum amount of waste.
Provision of a deflector plate about 1 ft. 4 in. in front of the scoop has enabled the following records to be made: (1) Average increase of water picked up at all speeds was about 200 gallons, representing about 17% improvement; (2) The amount of water picked up is constant for all speeds between the minimum or critical speed and 60 m.p.h., but as speed increases the amount of water splash causing waste rapidly increases; (3) The reduction of water lost by splash is constant for all speeds and was about 400 gallons, representing a reduction of about 50% at a speed of 40 m.p.h. The water splash as a percentage of the total water taken from trough, for the two cases is as follows

The author proceeded to show the huge amount saved by the total number of engines equipped with this pick-up gear on the L.M. & S. Ry., some 4,000  locomotives, assuming each locomotive fitted made an average of two pick-ups per day, not an excessive figure considering that on some through runs there were as many as nine pick-ups per turn. Daily there was a saving of pumping 7,000,000 gallons of water.
Improvements in the design to save water: invented by H. Chambers. Henry Fowler (464-5) commented upon the MR design; J. Clayton (465) noted the problem of water wastage; H. Holcroft (465-7) commented on th problems of getting rid of excess air from the tanks on tank engines and not that a new design was fitted to The Great Bear. A.M. Bell (467-8) noted that the installation of water troughs was important for operating the Norfolk Coast Express..
Sixth Ordiiiary General Meeting of the Birmingham Centre (Session 1930-31) was held at the Birmingham Chamber of Commerce, New Street, Birmingham, on the 30th March, 1931, at 7.0 p.m., the chair being taken by Mr. H.P.M. Beames.

Journal 102

Holmes, V.W. (Paper No. 278)
A new infinitely variable poppet valve gear. 481-90. Disc.: 490-524.
Eighth Ordinary General Meeting of the 1930-31 Session was held at the Institution of Mechanical Engineers, Westminster, on Thursday, 30 April 1931, at 6 p.m., the chair being occupied by the President,. Mr. H. Kelway Bamber, M.V.O.
One of the very few lady members of the Institution: she asserted that the use of poppet valves prevents wire drawing at admission and back pressure and exhaust. To obtain the full advantage from poppet valves, the gear should fulfil the following criteria: (1) the cut-off should be infinitely variable, not limited to a series of steps; (2) the lead should vary slightly, being greatest with early cut-offs, and least in full gear, in order to facilitate starting; (3) again to facilitate starting, the cut-off should be high in full gear; (4) the valves should open and close rapidly, and should give a good area of opening even with early cut-offs; (5) the exhaust valve timing should not be fixed, but should vary slightly with the cut-off, both releases and compression being delayed with late cut-offs; (6) the cam box should be as compact as possible, with the cam shaft not too high above the cylinder centre line; (7) the cam box should be a standard unit, capable of rapid removal and replacement by a spare box. Left and right-hand boxes should be interchangeable, and left and right-hand cylinders also should be identical; (8) the valves should be carried in cages, which should be capable of rapid removal, and (9) the control gear should be as simple as possible and minimize effort. The discussion included comment from Holcroft, Gray, Twinberrow, Beaumont and Maitland
E.C. Poultney (492): I should like to make a passing reference to one or two points mentioned by the Author in the Paper. ' One is the question of the cage construction for carrying the valve. The Author gives it as a sine qua non that that should be part of the arrangement, and I would like her to explain her reasons for that, because, as most members know, I am associated with poppet valve gears, and while we have used a cage on occasion, usually we have not done so, and we have not found any ill effects. In point of fact, we are always rather inclined to think that the cage construction will more or less cut down the free area through the portways and passages, and if that is so, given a cage, we would have to use a larger valve than would otherwise be necessary. Perhaps the Author would like to comment on that.
Then there is the old question of steps versus infinitely variable cut-offs. Theoretically one would be inclined to say that an infinitely variable gear would be preferable to one which gave a certain fixed cut-off. In practice I submit that it does not make any difference which you employ. We have had quite a lot of experience with the step type of poppet valve gear, and actually know how that arrangement compares in practice on the road with an infinitely variable gear, and I can assure you that in actual practice there is nothing between the two so far as fuel and steam consumptions are concerned. At any rate, the figures I have at my disposal seem to show that that is so, up to the present time. With regard to standardisation, it is a good thing to have standard units, but it is just a question as to how far that standardisation can be carried out. In our experience we have always found that different types of engines require somewhat different treatment. Therefore, while it is pleasant to talk about standardisation, as a matter of fact, it does not work out in practice. At any rate, that is our experience up to date. Of course, certain parts such as valves and spindles and many other details can be very largely standardised, and we have actually done so, but to standardise a unit and to say that that is what one would like to put on-well, we have not quite got away with it yet.
Precis in Loco. Rly Carr. Wagon Rev., 1931, 37, 157.

Kay, Walter (Paper No. 279)
Mineral oils and lubrication. 540-61. Disc.: 561-6.
Third Ordinary General Meeting of the Birmingham Centre (Session 1929-30) was held at the Birmingham Chamber of Commerce, New Street, Birmingham, on Wednesday, the 19th day of March, 1930, at 7.15 p.m., the Chair being taken by Mr. R.G. McLaughlin
Generally speaking, this Paper consists of an outline of the production of petroleum and refining, followed by the application of lubricating oil to machinery, and particularly to railway vehicles. To endeavour to treat the subject in other than a general way would be impossible in the time at our disposal.
There is considerable divergence of opinion with regard to blending mineral and fatty oils, in connection with steam cylinder oils. The problem of the lubrication of steam cylinders is extremely difficult, for although it is possible so to lubricate the cylinders and. valves of the steam engine that there shall not be any excessive wear, it is impossible to obtain anything like the results so far as friction is concerned, given by a well-lubricated journal. The pistons and valves move to and fro in straight lines, and do not tend to place themselves automatically in such positions as to trap the oil properly, and keep the surfaces from touching. Neither can the large extent of surface exposed be kept flooded with the oil, the passage of live steam through the valve chest and the steamports to the cylinders not admitting the presence of large quantities of lubricant. The engineer must, therefore, as a compromise, be content with the presence of a lubricating film of no great thickness and endeavour to keep the loads on the bearing surfaces as small as possible.
When the steam is wet it has a tendency to wash away the oil film on the internal surfaces. In compound or triple expansion engines, even if the steam is dry on entering the high pressure cylinder, the fall in pressure and expansion taking place eventually produces condensation, so that the steam arriving at the intermediate and low pressure cylinders may be wet. In order to lubricate cylinders satisfactorily under wet steam conditions, the cylinder oil must readily combine with the moisture and cling to the cylinder walls. It should, therefore, be a compounded oil, as the fatty oil tends to emulsify with the moisture present, and so resist the washing action on the bearing surface of the cylinder.

Ridge, C.W. (Paper No. 280)
The behaviour of railway material in the Argentine Republic. 528-619; 657-96; 765.
The second Quarterly Meeting of the above Centre was held at Perez on the 19th Jrne, 1931. Through the kindness of the General Manager of the Central Argentine Railway three sleeping coaches and restaurant car were placed at the disposal of the Centre for conveying the members from Buenos Aires. Fifty-one members travelled from the latter point, and on arrival at Perez the following morning were joined by 42 other members who had arrived from other parts direct. At 9.30 a.m. the members proceeded to the Meeting Hall, where Mr. J. G. Mayne, the Chairman, presided over a total attendance of 93 members and visitors.//Second Ordinary General Meeting of the 1931-32 Session w-as held in the Hall of the Institution of Mechanical Engineers, Storey’s Gate, Westminster, on Thursday, October 29th. 1931, at 6 p.m. Major C.E. Williams, O.B.E., in the Chair. The Minutes of the
Certain conditions prevail in the Argentine which are peculiar to the country, and make it difficult to apply rules which are found to be satisfactory in Europe, for the selection of its railway material. The class of workman called upon to handle the material, both in the shops and on the road, is a factor which often determines the success or failure of the products. Many of the Argentine workers are unskilled, or only semi-skilled. Only a few possess any degree of artisanship, which they have picked up in the small factories in Europe. The majority have scarcely any feeling of pride in their work which is so conducive to getting the most out of the material. It is very noticeable that many men take a delight in purposely destroying rolling stock by mishandling it on every possible occasion.
For the most part the railways pass over huge tracts of dried mud, and it is this which blows up in dust storms, which have a great effect upon mechanical appliances. This dust finds its way into axle boxes and causes heavy scoring of journals. It also falls upon motion parts of locomotives which have been oiled or greased and produces abrasion. Even the varnish and paint on the outside of the rolling stock is so bombarded by the fine particles of dust, that within a few weeks both have lost their good surface condition.
Tracks are difiicult to maintain for they pass over stoneless regions. This makes heavy ballasting a necessity, but after heavy rain the mud is so washed away beneath the ballast that rail joints are a constant source of trouble

Journal 103

Agnew, W.A.
Presidential Address: railway electrification. 636-56. Disc.: 1932, 22, 83-8.
First Ordinary General Meeting of the 1931-32 Session was held at the Institute of Mechanical Engineers on Thursday, the 24 of September, at 6 p.m., the chair being occupied by the newly-elected President, Mr. W.A, Agnew.
Notes that consideration had been given to electrifying certain British main lines, and in March, 1931, a comprehensive report was issued giving the views of a committee appointed bv the Minister of Transport. This was formed of Lord Weir of Eastwood (Chairman). Sir RaIph Wedgwood, and Sir Wm. McClintock, with Col. A.C. Trench, as Secretary.
The Third Ordinary General Meeting of the Newcastle Centre (Session 1931-2) held at Central Station Hotel, Newcastle-on-Tyne, on Tuesday 8 December 1931, at 7.15 p.m., the chair taken by the Chairman, Mr. P. Liddell who announced that due to the illness of the President, Mr. W.A. Agnew, his Address on "Railway Electrification" would be read by Colonel Mawby: this led to the discussion.
First Ordinary General Meeting of the 1931-32 Session of the North Eastern Centre was held at the Rletropole Hotel, Leeds, on Friday the 9th day of October, 1931, at 7.15 p.m., the chair being taken by Mr. T.H. Saunders
Opening General Meeting of the 1931-1932 Session was held on Thursday, aand October, 1931, at 7.30 p.m., in the Societies' Room, Royal Technical College, Glasgow, Mr. G. W. Phillips, Chairman of the Centre, presiding.
P. Liddell (22: 83-5) noted the difficulties involved in electrifying from Newcastle to King's Cross and considered that the lines to Sunderland and South Shields should be electrified. A.H.T. Head (87) considered that the cost of repairs to electric locomotives would be high.
The First Ordinary General Meeting of the Birmingham Centre, Session 1931-32, was held in the Queen’s Hotel, Birmingham, at 6.45 p.m., on Wednesday, 14 October 1931, the chair being taken by Mr. R.G. McLaughlin. The Chairman then introduced the President (Mr. W. A. Agnew), who repeated his Presidential Address, which was much appreciated by the members. (Published in Vol. XXI, Journal 103, p. 636.)

Powell-Brett, B. (Paper No. 281)
Modern drop-forging equipment and its services to the railway engineer. 697-730.
Second Ordinary General Meeting of the Birmingham Centre (Session 1930-31) was held in the Assembly Hall of the Birmingham Chamber of Commerce, New Street, on Thursday, 20 November 1930, at 7.15 p.m., the chair being taken by Mr. R.G. McLaughlin.

Gillvray, H.G. (Paper No. 282)
The design and equipment of a modern railway dynamometer car. 731-53. Disc.: 753-60: 1932, 22, 249-55.
18 December, 1930, in Birmingham//Third Ordinary General Meeting of the Birmingham Centre (Session 1930-31) was held in the Assembly Hall of the Birmingham Chamber of Commerce, New Street, on Thursday, the 18th day of December, 1930, at 7.15 p.m., the chair being taken by Mr. R.G. McLaughlin.
Hydraulic type of dynamometer car for service in India], similar to those in use in the United States of America and South Africa where very high drawbar forces were encountered with Mallet articulated locomotives in the former, and multiple electic locomotives in the latter country. Discussion: E.W. Selby (250-2)

Richie, E.G.
Steam storage in relation to the locomotive. 780.
Sixth Ordinary General Meeting of the Manchester Centre (Session 1930-31) held in the Building of the Manchester Literary and Philosophical Society, 36, George Street, Manchester, on Friday, 13 March 1931, at 7 p.m., the chair being taken by J.N. Gresham.
Showed the losses due to time lag in the production of steam to meet the variation in steam demanded in industrial plants which can be overcome by the installation of a steam storage system and advocated application of this system to locomotives. However, he left it to the locomotive engineers to find the necessary ways and means of applying it.
The full text of this Paper was available at Headquarters to members wishing to peruse it.

(North Eastern Centre–Leeds), Visit to Kirkstall Power House. 782.
On 10 and 17 October.

Journal No. 104

Hudd, A.E. (Paper No. 283)
A new system of automatic train control. 825-42. Discussion.: 842-54.
Third Ordinary General Meeting of the 1931-32 Session held in the Hall of the Institution of Mechanical Engineers, Storey’s Gate, Westminster, on Thursday 26 November 1931, at 6 p.m. In the absence of the President, owing to illness, the Chair was taken by J. Clayton.
A system of intermittent inductive automatic train control with three types of track elements: Permanent Magnet, Electro-Magnet and Combined Permanent and Electro-Magnet. A control valve is connected in the train pipe of the locomotile brake system, operated by a magnetic Pilot Vahe. The Pilot Valve is in turn controlled by four iron plates contained in a Receiver mounted on the locomotive. The lowest part of the Receiver is set at five inches above the top of the running rail, and the highest part of the track elements or Inductors at one inch above rail lelel, providing a total clearance of four inches.
Magnetic flux picked up by the Receiver opens or closes the Pilot Vahe in accordance with the position of the signals.
A momentary opening and closing of the Pilot Valve gives a short audible indication on the locomotive when a distant signal is at Clear. Continued opening of Pilot Valve indicates distant signal at “ Danger ” by means of continuous audible indication and partial brake application.
No effect is given on passing a Stop signal at Clear, but a continuous warning and brake application is given on passing a Stop signal at Danger. A manual release is provided to cancel Danger effect in each case.
Locomotive apparatus is operated by vacuum with magnetic control. No electricity used on the locomotive