Proceedings Institution of Mechanical Engineers: 1940s
Steamindex home page  Updated 2013-02-10 Key file The IMechE virtual library is inaccessible (full papers, all diagrams, photographs, extensive tables, etc).at SAGE.

Volume 146 (1941)

Stanier, W.A.
The position of the locomotive in mechanical engineering. 50-61 + 4 plates. 13 illus., diagr., 3 tables. (Presidential Address).
When commencing his training in January 1892, locomotive practice on the Great Western, under the guidance of William Dean, was very much the same as that of other railways of the time. The locomotives were comparatively small, with steam pressures up to 140 psi., but very quickly another phase began; steam pressures were raised to 160 psi and a bogie became necessary in front to provide a lengthened wheelbase on which to carry the larger boilers. About the year 1902, Churchward brought out the first big departure from current practice, when he built six-wheel-coupled express passenger engines with cylinders having 30 in. stroke and fitted with valve gear having an unusually long travel and a greater lap. These characteristics made it possible to work the engine so that greater advantage was obtained from the expansion of the stcam. Churchward continued to adopt these features throughout the whole of his career as chief mechanical engineer of the Great Western. This practice has been adopted and developed gradually on all the other English railways and it is the development that has taken place on the LMS over the last ten years to which he principally refered.
Table 1 lists the approximate thermal efficiencies of various steam locomotives: representative locomotive of c1880; c1912, Coronation, Chapelon superheated 4-8-0 compound locomotive and advanced steam power station practice electric drive (the last based on particulars given in Sir Leonard Pearce’s Thomas Hawksley Lecture, Proc.I Mech.E., 1939, vol. 142, p. 305), Table 1 attempts to set out the relative thermal efficiencies for different stages in the development of steam motive power, showing first of all the basic theoretical efficiency of the cycle, then the actual engine and boiler effciencies, and finally an overall thermal efficiency for the plant as a whole on a basis of indicated horse-power. The first column represents a saturated steam engine as designed in the last century (of which many are still running); col. 2 represents a superheated design of the period 1908-12, still retaining old-fashioned cylinder and valve gear design ; col. 3 the position of representative best present-day design in this country, while col. 4 is illustrative of the work done by Chapelon in France, and represents very nearly the best which can be expected from further refinement in.the normal reciprocating locomotive. The last column gives comparative figures for an “ideal” application of the most advanced power station practice to the locomotive, leaving on one side for the moment the question of how far the various features of power station practice could in fact be applied. The record of the locomotive is not, as is sometimes thought, entirely bad, and Table 1 shows clearly where it has advanced and where it still falls short. Table. 2: particulars of representative locomotive boilers (L .M. & S. Railway and Table3 Dynamometer car test results with various L .M. & S. Railway locomotives: No. 5917 Claughton class Euston to Carlisle and return; No. 6158 Royal Scot class Euston to Carlisle and return (low mileage and high mileage); Princess Royal No. 6210, Turbomotive No. 6202 and Coronation class No. 6225 with light load Euston to Glasgow; Euston to Glasgow and back No. 6220 with Coronation Scot load and timing and No. 6234 with maximum load Crewe to Glasgow and return; and Class 5 St Pancras to Leeds and return with No. 5067 with 14 element superheater and No. 5079 with 21 element superheater.
A review of the efficiency of the steam locomotive, based on LMS testing plus a forecast of future development: makes reference to Goss and thr Altoona test plant. Extensive precis of this Paper in Locomotive Mag., 1941, 47, 245-8.

Halcrow, W.T.
A century of tunnelling (Twenty-eighth Thomas Hawksley Lecture). 100-16. 16 figs.
Advances associated with mechanical engineering progress:

Cox, E.S. Balancing of locomotive reciprocating parts. 148-62. 4 illus., 10 diagrs., 3 tables.
Since the British Bridge Stress Committee’s reportt of 1928, an increase had occurred in the speed capacity of locomotives. After showing how the recommendations of that Committee have been met in modern British locomotive design, the effects of higher rotational speeds are discussed, and tests with the lifting of coupled wheels off the track at maximum speeds and with various proportions of reciprocating balance are described. Since the necessity may arise for reducing hammer blow values still further to meet civil engineers’ requirements, the possibility of reducing or eliminating the amount of reciprocating balance becomes important. The effect of this factor on the locomotive itself is discussed theoretically, expressions being given for the amplitude of the resulting horizontal and lateral oscillations. Available practical evidence in support of the theoretical considerations is described, and conclusions are reached regarding the conditions under which it is still necessary for locomotive reciprocating parts to be balanced.

Table 1 includes hammer blow data on 4-cylinder Coronation class Pacific, Jubilee class 4-6-0, Class 5 4-6-0 and both 2-cylinder and 3-cylinder 2-6-4T LMS types; LNER A4, K3 and V1 classes, GWR King, Hall and 5101 2-6-2T classes and Southern Railway Merchant Navy (nil hammer blow), Lord Nelso, King Arthur and Schools classes.

Colam, Sir Harold Nugent and Watson, John Douglas
Hammer-blow in locomotives: can it not be abolished altogether? 163-6. 2 diagrs., table
Abstract only: full paper Min. Proc. Instn Civ. Engrs., 1941 (Paper 5243)

Volume 148 (1942)

Thompson, S.J.
Boiler development past and present. 132-61.  62 figs. (mainly diagrs.)
Review paper: locomotives only receive slight attention (George Stephenson at Killingworth): most concerns large boilers associated with electricity generation

Volume 149 (1943)

Colbeck, E.W., S.H. Smith, and L. Powell, B.
Caustic embrittlement. 63-73. Disc.: 150, 88-98. 20 illus., 3 diagrs., 7 tables.
Results of experimental work on caustic embrittlement. The apparatus chosen was based on one used in recent American work pertformed by Straub and Bradbury (1938). In this, the solution to be tested is contained in a hollow sealed tubular specimen, heated to any desired temperature in the steam boiler range, and subjected to a tensile load by means of a spring. It had not been possible to reproduce the results claimed by Straub, but a number of examples of intergranular cracking were produced. Some of these have been obtained when using dilute solutions containing amounts of NaOH and silicate such as would be found in boiler waters. There is evidence that embrittlement occurs more readily in poor-quality steel and under conditions of non-uniform stress distribution, but it has not been possible to reproduce results with any degree of A few experiments with an entirely different type of specimen have been carried out; and in The paper also contains a survey of the more important literature on caustic embrittlement Certainty. these, intergranular failures have also been produced. published between 1935 and 1941.

Turner, T. Henry
Corrosion of boiler tubes. 74-88. Disc.: 150, 88-98. 6 diagrs., 5 tables.
A very extensive review which included locomotive boiler tubes. Turner's response to the extensive discussion was given on pp. 97-8 Also published in Journal Institution Locomotive Engineers, 1942, 32 Paper 438...

Volume 150 (1944)

Meyer, Adolf.
The first gas turbine locomotive. 1-10.
Traces history of electrical transmission for locomotives in which steam or Diesel engines have been the prime movers, and refers to the important contributions made by Messrs. Brown Boveri and Company, Ltd., Baden, Switzerland in this. Although there were over 1,000 exhaust and gas turbines in service, the locomotive forming the principal subject of the paper was the first example in which this form of power is applied. After dealing with the design of the turbine in general, an imaginary trip is taken with the driver of the gas turbine locomotive. Explains the principles of operation, and the method by which the system of governing enables the machine to adapt itself to running conditions. The safety devices are described in detail, and some attention is given to the opportunity afforded by a locomotive of this type for the adoption of power braking. Concludes with an analysis of the economic prospects of gas turbine locomotives, which comprises some useful comparisons with other types of motive power, including valuable information in graphical and tabular form.

Volume 151 (January-December 1945)

Puttick, H.W.
Diesel Traction on the North Western Railway (India). 87-98. 8 illustrations, 5 diagrams, 2 tables
In describing chronologically the Diesel traction developments on the broad-gauge and narrow-gauge lines of the North Western Railway, India, the author covers the Beardmore branch-line Diesel-electric locomotives, the Armstrong-Whitworth mountain-line railcar, the 1,300 b.h.p. mainline locomotives, and the eleven Ganz Diesel-mechanical railcars introduced in 1939. Electrical transmission troubles were the main cause of poor performance of the small locomotives, and led in 1940 to a decision to scrap the two locomotives after an aggregate mileage of only 163,000 had been attained in ten years. The small Diesel-electric railcar continues to give good service; apart from teething troubles, the principal defects have been a cracked cylinder block caused by a circlip breakage, three cracks in a bogie frame, and, some years ago, burning of the combustion chamber venturis.
The two main-line Armstrong-Whitworth locomotives were intended for the Lahore-Karachi mail service, but one failed during the preliminary trials and the other only ran about 1000 miles on trial. As the results were unsatisfactory it was decided to recondition both locomotives, and the Diesel engines, generators, and traction motors were returned to England. Shortly after this the makers asked for cancellation of the contract, and this was agreed to. The paper is mainly concerned with the operation and maintenance of the Ganz railcars. They were intended for fast service on branch lines and were therefore based on Jullundur (about 100 miles east of Lahore), where there is an extensive system of branch lines. In addition to the Jullundur service an intensive short-distance suburban service was operated in the Karachi area for some months, due to an emergency arising from the withdrawal of motor bus services because of severe petrol rationing.
Originally these cars were maintained on a mileage basis by the maker. Within the first seven months an aggregate mileage of over half a million was built up, but frequent troubles in traffc then led to the withdrawal of the cars for modification: The failures were connected mainly with the compressed air system, starter motors and battery, Hardy flexible couplings, main clutch, and engine water-cooling system; the gearboxes gave good service.
There were no less than 74 failures during the seven months’ service in 1939, giving the extraordinarily low mileage of just over 7,000 per failure. After reconditioning, a modified four-car service was introduced-the modifications, together with the taking over of the maintenance by the staff of the North Western Railway, greatly improved the service, so that at the end of the threeyear period (1st October 1940 to 30th September 1943) the mileage per failure had been raised to 173,854, as compared to 88,312 miles per failure for steam locomotives during approximately the same period. During this time the cars completed a passenger mileage of 1,043,122. The operating and maintenance costs are discussed, and the author also discusses the relative costs of Diesel-electric as compared to Diesel-mechanical railcars. Brief reference is also made to a two-stroke Diesel engine which has recently been fitted to a narrow-gauge (2 ft. 6 in.) railcar.

Volume 152 (January-December 1945)

Symposium of papers on Some Modern Aids in the Investigation of Materials, Mechanisms, and Structures. 211-43.

Johansen, F.C.
High-speed cinematography. 224-5. + 4 plates (including 1 colour). 36 figures
George Stephenson Address: LMS Reserach Department: described the techniique as such, rather than its applications.

Symposium of papers on Some Modern Aids in the Investigation of Materials, Mechanisms, and Structures.

T. Henry Turner (239-40) said that it was surprising to find testing with a brittle lacquer coating (W.J. Clenshaw: The measurement of strain in components of complicated form by brittle lacquer coating, 221-3) described without reference to the very extensive research which had been published 10 years ago  [Z.V.D.I., 1934, vol. 76, p. 973]. Thirty-three illustrations were given of the process called in German "Das Dehnungslinienverfahren", which he translated as "Stretch-Lines Process". It was used in the development of the Maybach high-speed railcars, and aeroplane and airship engines, making possible a rapid determination of working stresses at any place in a machine part. At that time one of the Transatlantic airships was nearly lost; four of the five engines failed, and the airship only just managed to limp home. That was such a shock to the designers that they undertook research on a big scale. Used as a basis for strength calculations, this German research had taught the designers of the motor for the Maybach high-speed railcar facts which astonished them, giving striking proof that it was wrong to judge the strength of a dynamically loaded part by its size, without considering its shape. One example, which he had seen cited, showed that the removal of material from the threatened section of a machine part, reduced the load on the section by one-third.
The German publication referred to showed the application of this stretch-lines process to the following typical components :-
Top and bottom halves of a crank case.
Steel tube stressed in torsion, the tube having a hole drilled -
A crankshaft. A hollow cylinder in compression.
A rectangular rod when bent.
A twisted shaft.
A shaft in longitudinal tension and simultaneously twisted.
A flat bar supported at either end and depressed in the centre.
The root of the curvature of a part similar to a connecting rod end.
A piston.
It was found necessary to choose different lacquers to suit each type of material—steel, light alloys, etc. Moreover, the temperature and the humidity had to be carefully controlled. With steel, the "stretch-lines" appeared in the coating when the part under investigation elongated to an amount (depending upon the room temperature at which the measurement is taken) equivalent to a stress between 3.2 and 12.7 tons per sq. in.
It was of fundamental importance that these cracks in the lacquer should form when the loads in the steel were far below the elastic limit. A picture was thus obtained of the stress conditions as they actually existed in service where the normal stresses were appreciably below the fatigue limit and yield point. The cracks in the lacquer were found always to lie at right-angles to the greatest tensile stress, as shown in Fig. 20, Plate 2, of Mr. Clenshaw's paper.
He regarded these brittle lacquer coatings as a most valuable tool for designers. It must be remembered that fatigue generally started at the surface, and these stretch-lines proccsses indicated stresses on the surface of the component being tested. through it. The German experimenters found that :-
(1) The lacquer must be suited to the material.
(2) The surface of the material must be given a suitable preparation.
(3) The temperature at which the test was made influenced the formation of the "stretch-lines".
Most lacquers were brittle in the cold and were made plastic by heat, so that this third point was easily understood. There appeared to be a fair amount of agreement, therefore, between the earlier German work and the more recent American work; and he hoped that Mr. Clenshaw's paper would result in this test procedure being better understood and more widely used in this country.
He would like to say a word about temperature-indicating paints. The habit of painting a locomotive crank axle red had for years past proved useful in indicating whether an axle which eventually failed under fatigue had during its life, or since the last shopping, been overheated. When an axle cracked, before blaming the steelmaker, they asked whether there was blackening on the red paint, because if so it was their fault as users for having run a hot axle.
The high-speed cinematography described by Mr. Johansen was a research tool which wrote its own reports, and there was no need of mathematics or microscopes to confuse the issue or camouflage the results; those reports were written so that "he who runs may read".

Lomonossoff, G.V. and Lomonossoff, G.
Condensing locomotives. 275-88. Disc.: 289-303. 8 illus., 25 diagrs., 4 tables. Bibliog.
Covers both reciprocating and turbine type of locomotive.
The condenser was introduced by James Watt in 1763 and constituted an important part in his historical patent of 1769. The first attempt to apply partial condensation for heating feed water in locomotives was made by T. Hackworth in 1827. But this idea did not receive a wide application until 1851.
The first attempt to apply full condensation to locomotives was made by Messrs. W. G. Armstrong and Company in 1848, but this engine failed to find a purchaser. Between 1863 and 1586 temporary condensation was employed on the Metropolitan and certain other railways in this country.' The aim of this arrangement was only to improve the hygenic conditions in tunnels, and not to raise the efficiency. The next attempt at condensation was made by the Hunslet Engine Company, of Leeds, when they built five locomotives for the Sudan in 1895. From the information received it seems to have been a case not of full but of partial condensation, because these engines were fitted with a valve which allowed the exhaust steam to go either into the blast-pipe, as in normal locomotives, or into a condenser, at will. Actually these locomotives never ran using the condenser at all.
About 1906 Ramsay renewed the attempts to apply full condensation to locomotives to increase their efficiency considerably. In the same year, and for the same purpose, the first Diesel locomotive was designed by Sulzer.
The Ramsay Turbo-electro Locomotives. The first turbo-electric locomotive was built by the North British Locomotive Company in 1910, by and the second by the Ramsay Condensing Locomotive Company, in co-operation with Messrs. Armstrong Whitworth and Company, in 1920.
The Zolly Turbo-locomotives turbine with a gear transmission, water condensers, and special turbines for reversing. The first of these locomotives was built in .1921 by the Swiss Locomotive Works, Winterthur, in cooperation with Messrs. Escher Wyss in Zurich for the Swiss Federal Railways.
The main turbine, of 1,200 h.p., was situated at the front of the engine and the two surface-water condensers on either side of an ordinary locomotive boiler working at a pressure of 170 lb. per sq. in. The cooling water was recooled on the tender by evaporation. The first arrangement of this sort was of a primitive type, and was unsuccessful. A new tender was therefore ordered from the Krupp Company, Essen. Krupps had also been working on the problem of condensing locomotives since 1920. The first idea, due to Dr. R. Lorenz, who was in charge of this work, was to build a reciprocating express engine with an output of 2,000 h.p. at the rail. To realize all the advantages of a high vacuum, he applied compounding with a high ratio between the volumes of the high-pressure and low-pressure cylinders. But in order to achieve this, the latter had to be so large that it was impossible to fit them into the loading gauge.
Dr. R. Lorenz then decided to utilize the Zolly turbine; and so the second Zolly turbo-locomotive was built. It is better known as the “Krupp turbo-locomotive".Its main characteristic was a special tender with a “locomotive tower”. A similar tender was also then ordered for the Swiss turbo-locomotive. The construction of this “tower” is clear from Fig. 6 in which the shaded areas represent layers of ordinary locomotive.
Besides the main turbine the locomotive had four additional turbines :-
(1) For reversing ;
(2) For the tender fan
(3) For other auxiliaries; and
(4) For the draught fan
Only the last used the exhaust stearn; and the first was co-axial
In 1926 the loconiotive was tested scientifically by Professor Nordmann at Grunewald. During these tests it was discovered that when running forwards turbine (1) absorbed over 400 hp. During 1927 the locomotive was reconstructed, and in 1928 it was tested again. The main results of these later tests will be given in section , p. 285. The authors have had opportunities of seeing this locomotive many times both in service, and in the shops.
The Ljungstrom Turbo-locomotives. The main features were
(1) Application of the Ljungstrom reaction turbine, which is very compac
(2) Application of the Ljungstrom air condensers;

(3) A special gear transmission for reversing; and
(4) A layout having the boiler and turbine on two separate vehicles.
The third Ljungstrom locomotive (Table 1 and Fig. 10) was built by Beyer, Peacock and Company in 1926 and was tried on the London, Midland and Scottish Railway.” By courtesy of the late Sir Henry Fowler, the first of the authors had an opportunity of seeing it at Derby in 1928. Owing to its complexity it was not very popular with engine crews and the vacuum was not always sufficiently high. In this respect tunnels had a very bad influence, as soot from the chimney entered into the condenser with the air and blocked the narrow passages between its tubes.
The Reid-MacLeod Turbo-locomotive. This locomotive was built by the North British Locomotive Company, and was exhibited in 1924 at Wembley. Unlike the Zolly and Ljungstrom locomotives it consisted of only one vehicle with a rigid frame placed on two double bogies. The Reid-Ramsay Iocomotive had a similar arrangement, and the Reid-MacLeod locomotive was possibly a rebuilt version of the earlier locomotive.
Each of the bogies was driven by a separate turbine by means of gears only, without any coupling rods. The turbines, of 500 h.p. each, worked in series. Thus the total horse-power of the Reid-MacLeod locomotive was about 950 at the rail. This was too small for British conditions and explains why it was not sold.????
Its construction is clear from Fig. 11, where, as in Fig. 10, thick lines represent water pipes, thin lines steam pipes, and dotted lines the flow of air. From the superheater chamber a, the steam was directed to the high-pressure turbine TI, then to the low-pressure turbine T2 and thence to condenser X and condensate reservoir C. The co
The Maffei Turbo-locomotive." In 1921 the Maffei the tender and drives two cooling fans as well as a circulating Locomotive Works at Munich began to design a condensing pump for the recooling water; and the third drives all the compound locomotive with a reciprocating high-pressure engine remaining pumps. and a low-pressure turbine,*Z but it was too complex and The authors saw this locomotive in service twice, in 1930 finally the reciprocating part of this design was eliminated. and in 1935. It has worked in parallel with the new 4-6-2 The Maffei design approximates to those of Zolly. It has two engines of the German State Railways; and, with non-stopping main impulse turbines placed at the front end of the locomotive. trains gave considerable economy in coal. But with trains One of these turbines is used for forward motion, and the stopping frequently it works even less economically than other for reverse. Two water condensers of the surface type lie ordinary locomotives. Nevertheless the crews liked it, thanks to along the two sides of the boiler, but are connected in parallel. its reliability and extremely steady running. But the vacuum The cooling tower is placed on the tender.
%%14) The Henschel Condensing Locomotitye for Argentina.23 The Argentine State Railwayi.'cross some districts where water is extremely scarce and usually of very bad quality. The second Ljungstrom locomotive (1923) was purchased mainlv in order to obtain an engine requiring no make-up feed. In this respect the locomotive was quite successful, but the maintenance of aLomon%(16) Soviet 2-10-0 Locomotives of Classes SO and SOk. Until 1931, the 0-10-0 locomotives of Class E were regarded in the Soviet Union as a standard type for goods service. In 1931 the 2-10-2 locomotives of class FD (Felix Dzerjinski) appeared, but owing to their axle load of over 20 tons they cannot operate on many lines. Therefore attempts were made to develop the existing 0-10-0 and 2-10-0 locomotives. It was in this way that the 0-10-0 locomotives of class Em and the 2-10-0 locomotives of class SO (Sergo Ordjanokidze) came into being. The latter has apparently been very successful and has been widely adopted since 1934.
Discussion: Bulleid (289) remarked that the paper treated the subject on an academic level. The steam locomotive to-day was in competition with other forms of traction. It had betn criticized as being inferior to diesel-electric locomotive and electric locomotives, chiefly because of its lack of availability. Lack of availability was fundamentally due to the use of raw water in the boiler, which caused the steam locomotive to be out of service for about 12% of its time; consequently anything which could be done to reduce that loss of service, necessitated by washing out the boiler or repairing damage due to dirt in the boiler, would at once contribute to the greatly increased availability of "what was, after all, the best traction machine in existence." The present paper, therefore, by calling attention to the question of the recovery of the water, was very valuable; for that recovery was much more important in locomotive practice than any saving of, say, 2%, 3%, or 4% in the fuel burned. For a coal consumption of the order of 50 lb. per mile (including lighting-up), the fuel thrown away when the engine was taken out of service, the losses when the engine was standing, and, above all, the fact that at one moment the locomotive would be working at high rates of output, using the whole of the steam available, and at another moment would be running under very low conditions of pressure and very early cut-off, it was not to be hoped that any substantial reduction in the fuel used would be effected by condensing. On the other hand if it was possible to avoid filling the boiler with raw water, that would be a means of very substantially improving the use which could be made of the locomotive. For that reason, Mr. Knudsen's experiments in the Argentine were very interesting, though there again the commercial factor was being ignored. Mr. Knudsen's locomotive cost 1.45 times as much as an ordinary engine, and had a tender which was even more complicated than the locomotive itself; and that would finally prevent any such arrangement becoming normal practice. After the war, when it was possible to devote attention to such matters, it would be appropriate for the Institution, which had been founded by a great locomotive engineer, to carry out a scientific investigation into the means whereby the use of raw water could be avoided, and above all to find what could be done towards recovering the latent heat which was at present discharged into the atmosphere.
Stanier (289)
: In this country, circumstances did not permit the efficient use of turbine power, because the conditions varied foot by foot along the road. When an engine had a demand, as Mr. Bulleid had said, for maximum output over a short section and then for minimum output over the next it did not promote the efficient working of the turbine. The experiment made on the LMS with a non-condensing turbine locomotive had indicated, so far, that the coal consumption per drawbar h.p.-hr. was very much the same as with the normal four-cylinder locomotives on work of the same nature. The turbine locomotive was employed only on straightforward runs between London and Liverpool, and not on trains stopping at many intermediate stations. The coal consumption per drawbar h.p.-hr. was of the order of 2.8 lb. for the reciprocating locomotives and 2.78 lb. for the turbine, while the water consumptions were 24.7 lb. and 24.8 lb. respectively. The corresponding evaporations per pound of coal were 8.3 lb. and 8.4 lb. In regard to condensing apparatus, the difficulties experienced up to the present in this country had, he thought, usually been mechanical; the condenser had given more trouble than the engine and the turbine. In Russia they seemed to have overcome the difficulty with the Henschel arrangement. The height of the condensing gear, however, was about 14 ft. 6 in. above rail level. It might not be necessary to have so great a height, but presumably if the height was reduced a much longer condenser would be required, and there would be difficulty in accommodating it within the normal limits. He was interested in the authors’ statement that the reciprocating engine with condenser could be useful if it was desired to conserve water, but that if a saving in coal consumption was required it was necessary to sacrifice some of the saving in water and have a turbine to supply the power. The turbine locomotive might be very satisfactory on long stretches of line such as were found in Russia or in Australia or in the Argentine, but he had already indicated that it was difficult to find a satisfactory field for it in Britain. ,H. Holcroft (292-4) Anderson system

E.S. Cox, (291-2) said that two of the condensing locomotives mentioned in the paper were tried on the London, Midland and Scottish Railway, and some notes as to their performance would allow certain conclusions to be drawn. The first of these locomotives was the second of the two Ramsay designs mentioned in section (9), p. 278 of the paper. This was a two-unit locomotive of the 2-6-0-+-0-6-2 type. It was designed with four 275 h.p. traction motors, and the intention of its designers was to produce a locomotive of approximately the same power as the London and North Western Railway fourcylinder 4-6-0 “Claughton” locomotive of those days. This engine was delivered at the Horwich works of the London, Midland and Scottish Railway in February 1922, and on weighing it was found to be 35 tons above its designed weight, with individual axle loadings as high as 24 tons—an unheard of figure in those days. At first it was rejected out of hand by the engineer, but afterwards permission was given for it to run as far as Southport, a distance of 25 miles, and throughout its whole life its test runs had to be confined to that short distance.
The results at first were very unsatisfactory. The motor-driven forced-draught system was so defective that proper cornbustion was impossible, and even when running light there was heavy smoking and inability to maintain the pressure. Moreover the condenser, which was of the evaporative type, was not at first able to produce a higher vacuum than 20 inches of mercury, though designed for a vacuum of 27½ inches, and therefore increased steam consumption made impossible demands on the limited boiler capacity.
By the end of June of that year various alterations had been carried out with a view to improving the running, including a new chimney with a blast arrangement taking steam from the condenser ejector; and a larger condenser of the same type was fitted. Between November 1922 and June 1923 several trials were carried out with this engine to Southport with a trailing load of 65 tons, when a maximum speed of 59 m.p.h. was attained. Throughout those trials the condenser vacuum was very variable and did not rise above 25 inches, and the boiler pressure fluctuated between 140 and 170 lb. per sq. in., as against the designed pressure of 200 Ib. Coal consumption worked out at something like 40 Ib. per mile. On other occasions loads up to 170 tons were hauled, but at lower speeds. The performance of this 155-ton engine, therefore, at no time exceeded that possible with the 56-ton 2-4-2 tank engines which carried the traffic of the district. The failure was one not so much of principle as of design and application, in that both the draught and the condenser capacity were inadequate. The engine was cut up and sold for scrap in 1924.
The second condensing locomotive which was run on the London, Midland and Scottish Railway was the Beyer-Peacock Ljungstrom, a direct-drive turbine locomotive, called the third of the type in section (ll), p. 280 of the paper. This was a much more practical design and ran for a while on regular express passenger train work between Manchester and London. In May 1927, on a test run from Derby to Bedford and back, a maximum speed of 76 m.p.h. and 1,200 d.b.h.p. were obtained. The coal consumption, however, was 57 lb. per mile and 5.6 lb. per d.b.h.p.-hr., because at that early stage the same troubles were being experienced as with the previous turbine locomotive; i.e. poor combustion and failure to maintain the vacuum.
During the ensuing year many detailed modifications were made to the engine, with intervening tests, but right up to the end the defects were never entirely overcome. Finally, in April 1928, a series of dynamometer car tests was undertaken on fast vacuum-fitted freight trains between Leeds and London, with the Ljungstrom locomotive running against a standard L.M. & S.R. 2-6-0 mixed traffic engine. The average running speeds were 35 m.p.h., with maxima up to 60 m.p.h., calling for a maximum drawbar-horse-power of about 1,050. The trailing loads were in the region of 500 tons, and the average results over a number of runs in each direction gave the following figures :-
Coal consumed per d.b.h.p.-hour (pounds) :-
Standard 2-6-0 reciprocating engine, 3.15 Ib.
Ljungstrom turbine engine, 3.28 lb., an increase of 4 per cent.
Water used (gallons per mile) :-
Reciprocating engine, 425.
Turbine engine, 6.8, a reduction of 84 per cent.
Generally speaking, boiler pressure and condenser vacuum were well maintained, except that on the 1 in 100 bank starting south from Sheffield there occurred the incident referred to in the paper, which was and, he thought, must be a feature of all air condensers, namely that when the locomotive had to pass through a tunnel very slowly the extremely powerful fans drew off all the soot on the tunnel lining and such smoke as might be coming from the chimney, and that led to trouble. Water consumption, which represented that lost by leakage and by the working of certain auxiliaries such as the vacuum brake, varied widely from trip to trip because of minor defects which developed.
The object-lesson of those experiences was clear. To supplement the theoretical conclusions so ably put forward by the authors, three practical considerations had also to be met. First, an effective substitute must be found for the automatic relationship between the power output and the steam production provided in an ordinary locomotive by the blast-pipe. The difficulty of doing that had always been under-estimated, and was greater in a turbo-condenser engine, since steam demand not only depended on the varying traction load but had also, in past engines of that type, depended on a fluctuating condenser vacuum as well. The authors had suggested a solution in the case of the reciprocating condensing locomotive, namely passing the whole of the exhaust steam from the cylinders through the draught fan turbine, but he would be interested if they would indicate how that could be done in a practical manner in the case of the main turbine exhaust, which had already been brought, for the sake of efficiency, to the lowest attainable point below atmospheric pressure. It seemed clearly to be useless to construct and submit to testing any locomotive for practical traction purposes until that whole problem had been completely solved.
Second, means must be found, possibly by automatic control of the condenser fan output, to ensure that additional steam demands were not made on the boiler due to a periodical drop in vacuum. Alternatively, if such loss was inevitable, the boiler must be made big enough to meet such demands; in other words, for reliable operation turbo-condenser locomotive designers should not assume that a smaller boiler could be provided, sufficient only to give the reduced amount of steam theoretically required. Experience seemed to show that as large a boiler should be put on such a locomotive as on an ordinary non-condensing engine.
Finally, the many auxiliaries inseparable from a turbine condensing installation must be reliable. In the case of the two locomotives which he had mentioned, the turbines and gears and traction motors in the first case were very reliable and gave no trouble, but the same could not be said of the many auxiliaries. No turbine condensing engine, therefore, had yet appeared in this country which sztisfied those three conditions, and the economic possibilities of this form of motive power had therefore never been demonstrated here. Both the engines to which he referred earlier had fundamental defects, and, although each was given a long period-ver a year in each case-of patient and costly trial and alteration, at the end no attention to details was able to overcome the defects, and the engines were not acceptable. There clearly still remained scope for further attempts, and the paper would be of great value in outlining the boundaries between which such attempts could be made.

H. Holcroft (292-4) called attention to a novel form of condensing applied to a Southern Railway locomotive some fifteen years ago. No information regarding it was disclosed at the time, and no results of the trial had been published so far, but with the expiry of the patents and for other reasons it was now permissible to give a description of the system and experience obtained from its use.
The apparatus as installed with a stationary plant arranged for demonstration purposes did not lend itself to locomotive conditions, and it became necessary to devise a cooling system more suited to the object in view, and generally to aim for what was termed in locomotive circles “a sound running-shed job.” It had been his task to collaborate with the consulting engineer of the company owning the patent rights in producing a practical scheme, and after a few experiments they hit upon a very simple solution.
Fig. 22 illustrated the layout as adopted on a locomotive. It would be seen that the whole of the exhaust steam from the cylinders entered an oil separator on its way to a multi-tubular cooler, the headers of which were formed to make the steam pass three times through the cooler. On emerging from the bottom set of tubes, the steam was dealt with by a three-throw combined compressor and feed-pump, from which it was discharged as feed to the boiler. The working fluid of the engine therefore operated in a closed circuit. Water carried in the tender was fed to the cooler and was evaporated there at 212 deg. F. under atmospheric pressure, and the vapour resulting was conveniently discharged with the products of combustion in the chimney. The function of this water was to abstract a portion of the heat contained in the exhaust steam and so condition it that the special form of pump could deal with it. In order to create a temperature difference to bring about the necessary heat transmission across the tubular cooling surface, the temperature of the exhaust steam had to be in excess of 212 deg. F., and it was found that the temperature corresponding to a back pressure on the cylinders of from 4 to 7 lb. per sq. in. above atmospheric was sufficient under normal conditions, being no more than that usually found at the base of a blastpipe in an ordinary locomotive. An induced draught fan was provided in place of the blast-pipe. That was the scheme in essence, but some details were omitted from the diagram for the sake of clarity. For instance, the exhaust from the auxiliaries was carried to the grease separator and joined the main stream, and there was a small gear-type pump driven from the main pump to regulate the flow of cooling water.
The following points should be noted :-
(1) Apart from any superheat which might remain in the steam at exhaust, the temperature was constant throughout to delivery in boiler, the feed temperature being in the region of 225 deg. F. Condensation was by pressure rather than temperature, as in the normal manner.
(2) The work obtained from the steam passing through the cylinders was not greater than in an ordinary locomotive. Fuel economy was entirely on the boiler side, due to the return of heat to the boiler; in other words, there was an increased evaporation per pound of fuel.
(3) The work expended on the pumps was small, and little or no greater than upon an ordinary feed-pump, the high-speed steam engine driving the pump being of some 5 i.h.p. capacity.
(4) The amount of water used in cooling was substantially less than that consumed as feed in the ordinary locomotive.
(5) The apparatus was largely automatic in action and required little in the way of control. The pump ran at a governed speed and within wide limits took whatever quantity flowed to it. The amount of water contained in the cooler was only enough to cover the lowest bank of tubes when still. With increase in quantity of exhaust steam passing through the interior of the tubes, the bubbles, and spray reached and wet the middle and then the upper banks, and in that way only the minimum of heat necessary was abstracted from the exhaust steam and the maximum economy was secured automatically. Loss of cooling water by spray carried away with the vapour was reduced as much as possible by the provision of a collector pipe in the cooler.
(6) The apparatus was of a character which could be maintained by shed fitters. The controls were simple to operate and did not add appreciably to the enginemen's duties.
(7) The system was not suitable for waterless regions, but was of value where water supplies contained impurities. It was independent of atmospheric temperature and could therefore be used in hot climates and high altitudes.
(8) It was cheaper to construct than other condensing systems and was no more elaborate than some feed-water heaters.
Apart from any question of fuel saving, advantages to the locomotive arose from reduced boiler maintenance due to absence of deposits associated with raw feed waters, these troubles being transferred to the cooler. Concentration of soluble salts there was avoided by discharging the small quantity of water in the cooler periodically. Any deposits were soft and easily removed by washing out the cooler, which could be done between turns of duty, and so avoiding the loss of availability of the engine to traffic associated with cooling down and washing out the boiler. Reduced cylinder maintenance arose from the avoidance of contamination of interior surfaces by smokebox gases and dust passing down an open exhaust-pipe while the engine was running with steam shut off.
Fig. 23 showed the locomotive as fitted. The apparatus was in duplicate, with a complete set on each side of the locomotive, one or both of which could be operated as necessary. That arrangement could be applied to a standard design of engine without exceeding the permissible axle weights or loadinggauge clearances and without obstructing the driver's view from the cab. No alteration to the tender was required, and therefore there were none of those awkward large-diameter flexible pipe connexions between engine and tender required with other condensing systems. The apparatus was of a somewhat makeshift character, the original intention being to replace it in time by permanent sets designed as the result of the initial experience gained.
In order to give access to the smokebox for ash removal and attention to tubes, the draught fan, of the propeller type, was carried on the smokebox door. It was directly driven by a rotary steam engine of the radial type, having a fixed cut-off of 25 per cent and made self-starting on account of the multiplicity of cylinders. A mechanical lubricator was provided to make the appliance self-contained. Steam and exhaust connexions to the engine had trunnions in line with the door hinges, so that the door could be opened in the normal manner.
After two or three years of desultory running, the condensing apparatus was removed and the engine released to ordinary duties, so that it was natural to ask what difficulties were found. Three-quarters of the trouble experienced was due to the draught fan, which was never adequate for the full capacity of the boiler, and that in itself was a handicap throughout. A reducing valve was originally included to limit steam pressure to the fan engine to 160 lb. per sq. in., and so leave a margin of 40 Ib. below the 200 lb. boiler pressure. This valve was discarded because full pressure was needed for the fan, and so there was no reserve. If boiler pressure fell off, so did the fan speed, and the effect was cumulative; and things went from bad to worse with the lengthening of the locomotive cutoff. It would be appreciated that this draught problem was more acute in the case of a high-pressure engine exhausting to atmosphere than it was for a turbine locomotive of equal output working on a 29-inch vacuum.
Due to the high speed of the propeller type fan and to direct coup!ing, fatigue flaws developed in parts of the rotary engine, and fractures ensued in the course of time. Some components were strengthened, but trouble persisted. A 2-to-I gearing was then interposed between fan and engine, and that necessitated a second bearing for the overhung fan shaft, somewhat cramping the gearing, which was of the sun-and-planet type to preserve the axial alignment. In order to make up for the lower revohtions of the engine and to maintain its horse-power output the fixed cut-off had to be lengthened, and steam consumption was increased. The arrangement was never very satisfactory and eventually, in the later stages, a small turbine of the impulse type was substituted. While this reduced mechanical trouble, it proved to be very extravagant in steam.
The original compression pumps were liable to set up violent water-hammer in the delivery pipes at times. The barrels were redesigned and the trouble eliminated, after which some successful running was experienced with freight and passenger trains. Success was, however, rather fickle, because at other times high back pressures were encountered ; sometimes these were temporary only but at other times they were persistent. In view of the reliability and regularity of working of the stationary plant, this was evidently due to the conditions peculiar to locomotive working. One probable cause of high back pressure was the formation of air-steam mixtures. The admixture of air lowered the condensation point of steam, and conversely the pressvre for a given temperature was higher; thus a higher back pressure was required in the cooler for the necessary rate of heat transmission through the tubes. The presence of air also retarded condensation of steam. Air was drawn into the steam space of a boiler when it cooled down below 212 deg. F. in the shed; it entered dissolved in the make-up feed water and was entrained at the injector, and of course it filled the cylinders and steam-pipes every time the regulator was shut. Rejection of any accumulation of air in the steam was effected by opening a valve while the engine was at work, allowing the steam to exhaust to atmosphere for a short period, thus breaking the circuit.
Another cause suspected was flooding of the oil separators on the engine starting up from cold. Lubricating oil carried over with the water would coat the interior of the tubes with an oil film and interfere with the rate of heat transmission, and so necessitate higher temperatures and thus greater back pressures.
Those “teething troubles” would no doubt have been overcome with further experience, and by the design of apparatus for permanent use. The experiment was terminated because the induced draught was insufficient for the engine to be able to take its place in the link with other engines of its class. When it was realized that only some ten or twelve trial trips with train loads could be arranged per year, due to all sorts of difficulties and setbacks, it would be seen that progress was painfully slow. No doubt progress would have been more speedy if a fully equipped testing plant had been available.

Bulleid, O.V.S.
Locomotives I have known.
341-52 + 6 plates. 18 illus., 12 diagrs., 11 tables.
A select group of locomotives reviewed by the author. An unusual paper in that it selects a number of locomotive types for detailed consideration. Three are obvious (Lord Nelson modifications, Merchant Navy and Q1 classes) in that they were his own designs. The others considered were the Gresley Al, A4, 01, 02 and P2 designs, the Ivatt Atlantics and a French design (Chapelon Paris-Orlians 4-8-0 compound engine of 1935). From the choice of L.N.E.R. types it appear that Bulleid agreed with much of Gresley's design policy. Perhaps the most intriguing are the detailed analyses of the Sturrock 0-8-0T, the Ivatt 0-8-2T (considered with the previous), the Ivatt 0-8-0 and big Atlantics. Between 1902 and 1908 inclusive, 81 of these engines (Big Atlantics) were built and the results obtained in service were very good, both as regards speed and hauling capacity. They readily met Mr. Ivatt’s requirements of a sustained drawbar pull of not less than 2 tons at 60 m.p.h. Mr. Ivatt used to make two remarks about the proportions of these engines. The first was: “It is no good having a large purse and nothing to put in it” (referring to the vogue for large cylinders and small boilers); the second, “The difficult thing is to get the steam out of the cylinders”, related to the attention given to the exhaust passages. The simple form of firebox is noteworthy, the inwardly inclined sides of which, it was claimed, allowed the bubbles of steam to pass through the water instead of creeping up the plate as in the narrow box. The rather flat brick arch-obtained by using large quarries in place of ordinary firebricks-will be noted. The external appearance of the smokebox is deceptive; the smokebox tubeplate is recessed into the barrel and the smokebox volume is thereby increased considerably. This large volume in conjunction with the lower smokebox vacuum found to be adequate when a wide firebox is used, materially reduced the emission of sparks. This large boiler feeds two comparatively small cylinders, 18¾ inches diameter by 24-inch stroke. Balanced slide valves were used. As in this case they were arranged to work vertically, a very direct exhaust was attained. A detail of interest is the tyre section. The coupled wheels are 6 ft. 8 in. diameter on tread, and their centres are 6 ft. 10 in. apart. The clearance between the flanges was only 29/32 inch, the depth of this flange having been made inch, as compared with the standard of 11/8 inches. . .

Volume 153 (1945)

Newberry, C.W.
A study of the riding and wearing qualities of railway carriage tyres having various profiles. 25-35. Disc.: 35-40.
Results from preliminary experiments to devise a technique to record the transverse motion of a railway coach wheel travelling at speed: this used a cine-camera to produce slow-motion films of the contact region between wheel and rail. The visual impression of the cine-screen image was transcribed, by means of a specially developed apparatus, into graphical form from which the wheel motion could be analysed. The wearing qualities of each tyre profile tested have been assessed by a comparison of the tyres at various stages of wear, the profiles being recorded by a specially accurate method developed for the purpose. Tests have been made with tyres turned to the standard ARLE (Association of Railway Locomotive Engineers of Great Britain and Ireland) profile having a 1 in 20 coned tread, and with experimental profiles having cylindrical treads or 1 in 100 coned treads. The records of wheel motion are discussed, together with direct observations on the riding of the trains concerned; and the various profiles are considered from the points of view of riding, wear, manufacture, and maintenance.
Discussion: C.E. Fairburn (36-7) noted that in 1935 Sir Harold Hartley and the writer, when travelling in the United States, visited the Chicago Rapid Transit Company and were very kindly shown the film records and given a full account of the tests with cylindrical tyres which had been made on vehicles using that company’s lines. In the London, Midland and Scottish Railway tests the question was considered largely in its relation to the riding qualities of the stock, but the Chicago tests arose from considerations of track maintenance. During the spring thaw, after the severe winter frosts experienced in that region, very heavy expenditure was incurred in track maintenance and re-alignment. Experience showed that that expenditure was substantially reduced when cylindrical treads were substituted for the 1 in 20 taper profile previously in use. As the lines ran through the city, there were many sharp curves, and flange wear was rapid; but it was found that there was no appreciable difference between the two profiles in that respect. In any case, the saving in civil engineering expenditure was much more than sufficient to compensate for any possible increase in the cost of tyre re-turning. The question was discussed also with the Illinois Central R.R. and the Chicago, Burlington and Quincy R.R. These companies were interested primarily in the riding of the stock; the former had been using cylindrical treads on multiple-unit electric vehicles with success for two years or more, and the latter had adopted them for some axles on the Burlington Zephyr train, but neither company could make a definite pronouncement about their wider adoption on ordinary steam-hauled stock. With this information about American experience available, it was decided that tests should be made in this country, and the electric stock on the Liverpool-Southport line was very suitable for the purpose.

Walter, A.J.R.
Modern methods of water treatment. 282-7. Disc.: 288-93. 6 ddiagrs., tables.
All natural water supplies contain impurities which often have a profound and far-reaching effect on the products being made and on manufacturing costs. The introduction of new processes for materials and more efficient power generation has in turn made essential the removal of even extremely small amounts of impurities from water that a few years ago would have been tolerated. For example, the increase in boiler pressures alone over the last twenty-five years has called for complete revision of the quality standard of boiler feed water. Fortunately those responsible for research into, and practising the art of, water purification have kept a step ahead of these new requirements and most waters can now be treated in one way or another so as to render them suitable for practically any purpose. Revolutionary discoveries have been made and applied in the field of water treatment. This paper describes the major advances, among them new crystallization and precipitation processes developed for lime-soda softening and particularly the entirely new chemical field discovered in hydrogen ion exchange materials and acid adsorption materials. The combined use of these two new discoveries has resulted in the development of equipment for the production of the equivalent of distilled water without heat or evaporation.

Tuplin. W.A.
Rational specification of surface finish. 340-1

Johansen, F.C.
Locomotive practice. 351-2. Disc.: 352-79.
The Author was employed in the Research Department, London, Midland, and Scottish Railway, Derby. The discussion related to the whole Conference on Surface Finish at the Institution on 9 March 1945.
The influence of the conditions of service maintenance and long life of a locomotive has the consequence that the only surfaces to which exceptionally fine finish was imparted were on parts which might give immediate trouble and never be self-rectifying without such attention. Thus, journals of coupled wheel axles have shown, over a number of years, that progressive improvement of surface finish diminishes the occurrence of overheated bearings, especially in the early stages of use. The tendency, therefore, was for all engine and tender journals to be fine-tuned, fine-ground, and finally lapped. Shrouded leather split collars, enclosing felt pads loaded with oil and emery powder, were strapped around the journals, and were belt-driven by electric motors. Simultaneously the motors were oscillated axially, and the wheel-axle assembly was slowly revolved. Crankpins were usually hand polished after being ground, but crank journals could be machine-lapped, the driving motors being adapted to run up and down inclined rails. As an alternative to the lapping machine, the cylindrical parts of journals and the radiused fillets at their ends were polished (after the fine-turning operation) by felt pad and emery, applied while the assembly was mounted in a lathe.

Volume 154 (January-December 1946)

Bulleid, O.V.S.
Some notes on the "Merchant Navy" class locomotives of the Southern Railway
. 316-33. Disc. 333-43 + 4 plates. 13 illus., 20 diagrs., 2 tables.
Very extensive account of the justification for the design and observations on its performance in service. The abstract states: The operating requirements in respect of weights of trains and speeds determined the main dimensions of these new locomotives, and the paper refers to the limitations on the development of powerful locomotives imposed by the loading gauge and weight restrictions. The Merchant Naqy class locomotives embody many innovations in British locomotive practice, and the reasons for these changes from orthodox practice are fully explained, and the way in which the changes were carried out described. Some difficulties were naturally experienced in service as a result of these departures, and they are fully related, together with the modifications which successfully overcame them. Some of the improved results obtained as a result of the innovations are reported.
: T. Henry Turner (334-) In 1944, and again in 1945, the author very kindly allowed him to visit the Southern Railway works at Eastleigh to see the tubes that were being removed from some of the Merclzant Navy locomotives, and he could certainly confirm from his own observation the statement that they seemed to be in fit condition to be put back in the locomotive and run for a further mileage. There was extremely little pitting or corrosion or damage to them, and that was no doubt due to the care taken in the sheds with their washouts and so on as well as to the fact to which the author drew attention, the omission of copper in the firebox construction, so that electrolytic corrosion was reduced.
The wheels were really most ingenious in the way they met the troubles which had been encountered in the past. When he first saw the photographs of them he thought that they were fabricated, and did not like them, but now that he saw that they were cast steel he thought that they were very fine indeed. They bridged the gaps, the place where nearly all tyres broke was nicely supported, they stopped the rocking, and his only criticism would be that the inside edge of the tyre looked rather too sharp-as though it could do with a larger radius and so match the outside edge.
The enclosed motion must be regarded as one of the finest experiments to which the author had drawn attention. He had followed the “little oil bath” of the bicycle, but had had a most formidable problem with which to deal. It was to be hoped that he would persist in his experiment and make “oil-bath” locomotive motion a reality. It was worth doing, because the consequences of hot boxes and hot big-ends were very far reaching. Personally, he doubted whether any crank axle broke that had not had a previous history of a hot bearing. If one could prevent a bearing running hot, one would probably also reduce the number of broken crank axles and all that they entailed. He wondered, however, whether the author would not go fiuther with regard to the cast iron piston. A steel piston flash-butt welded on to the rod was in use, and would seem to offer still further reduction in weight of reciprocating parts.
There were in this country at the present time few men who could hope to put their name to a class of famous locomotives, and everyone would welcome the fact that the author had joined the select few. His paper was a model of precise and concise description, and it showed that the Merchant Navy locomotives bore features which distinguished them from all their predecessors. His design embraced many features which he no doubt learned in those years when he worked so hard in connexion with international railway conferences and inquiries ; he had probably obtained in that way a wider knowledge of Continental, American, and world-wide practice than most engineers; but he had not only adopted those things which he found to be the best-he had also stepped out into the untrodden paths of real invention and inspiration.
To judge any creation one must take note of the background, recall the preoccupations of his staff and the author’s own heavy load from Dunkirk to D-Day. The Merchant Navy locomotives stood out from their background; they stood out at the head of trains, and they stood out as leaders of thought. To railwaymen, Pacific main-line locomotives were more than engines; they resembled the banner, or the band, at the head of a procession; they gave direction and pride and comradeship to the whole of the railway staff. The author had raised a new banner. There was still novelty in the steam locomotive. When one looked across the lecture theatre to the portrait of the author’s old chief, Sir Nigel Gresley, who also designed and wrote papers about locomotives, one felt that Sir Nigel would have wished to join the members present in congratulating the author on an outstanding paper.
T.R. Cave-Browne-Cave (335) of University College, Southampton, where wind tunnel work had been performed on the shape of the front of the locomotive. It was a good example of the way in which a model could be used in the wind tunnel to reproduce the air flow over a large moving object and to allow the effect of various changes of shape to be studied. It also showed how airflow problems other than those connected with aircraft design could be solved by means of a model in a wind tunnel. A model some 8 feet long had been made by the Company and had been mounted in the wind tunnel at University College, so that a uniform stream of wind could pass over it moving parallel to its axis or at a moderate inclination either way. In this case the object was to study the movement of steam emitted from the chimney. Smoke was emitted from the chimney of the model and small streams of smoke were also used to explore the air flow round the front and down the sides of the model. With the shape of the locomotive as first construtted, it was found that steam was drawn down the sides of the boiler and obscured the driver’s forward view so that sometimes he could not see the signals. It was found that this movement of the smoke, emitted from the chimney of the model, satisfactorily reproduced the movement of the steam over the full-scale locomotive. Various alternative shapes were, therefore, tried on the model
The trouble was due partly to air overflowing from the cowl in front of the smokebox and being thrown so far out from the sides of the boiler casing, that a region of low pressure was formed and the steam from the chimney drawn down into it. The cure was effected partly by leaving a more clear passage between the side plates and the boiler, so as to drive a stream of air close along the sides of the boiler, and partly by giving very fiee outlet over the top of the smokebox, so as to give a stream of really high-speed air over the exit of the chimney. This stream carried the steam upwards and well clear of the windows. It probably had the incidental advantage that the region of very high speed-and therefore low pressur-ver the mouth of the chimney did improve the draught. The shape of the locomotive was altered to conform with the shape determined by the model, and it was then found that the airflow over the locomotive exactly reproduced that which had been seen on the model.
It would be interesting to know whether it was found that the new shape gave greater chimney draught, as was to be anticipated. Mr. Turner had raised the subtle question as to whether the shape of the locomotive was pleasing. That was largely a matter of judgement; but it was quite clear that if altered shapes were contemplated for the purpose of improving appearance, it would be wise to use the model to study both appearance and the effect upon airflow.
He would like to raise one further point, though he did so with some diffidence. Part of the cinematograph film exhibited during the presentation of the paper had shown the flow of oil over the moving mechanism inside the enclosed gearbox, and the author had referred to the difsculty caused by leakage. It seemed probable that the oil flow was unnecessary copious for lubrication, and that this large excess of oil was probably the cause of the the cause of the oil leakage which had given trouble, and also, incidentally, of excessive heating of the oil. He suggested that the oil flow might be progressively reduced until there was some indication of trouble arising through diminution of the flow.

Alan E.L. Chorlton. (339: Past-President) wrote that the paper showed that the author had made the most outstanding changes in locomotive practice that had been witnessed for many years. He had not been afraid to design and try many novel ideas in various parts of the engine. He would like to ask the author why eight-coupled express engines had not been a success in this country, whereas in Canada and the U.S.A. they were standard practice. This question did not apply to the Southern Railway, but to the northern lines; for example, on the London, Midland and Scottish Railway there were heavy gradients and the weight of trains was steadily increasing, and the six-coupled Ppcific locomotives slipped considerably. He also asked why no other British railways besides the Southern used blast pipes with multiple exhaust nozzles. Cast steel frames were general in Canada and the U.S.A., and were also used on large Diesel locomotives with which he had been associated. What did the author think of cast steel frames?

E.S. Cox (339-) wrote that the most interesting feature of the Merchant Navy engines was the introduction of the steel firebox with its thermic syphons. Experience in America seemed to connect good steel firebox performance with fully treated water, and it would be interesting to know how sensitive the Southern Railway boiler was to this factor. He understood that external water treatment was applied individually to the tenders of these engines ; but this method, although standard in France, and easily controllable on a small number of important engines, was difficult to control on a large scale. In this connexion, again, American experience was uncertain in the choice between presence or absence of the copper ferrule at the firebox end of the tubes, but only with rigidly controlled water did it seem possible to rely fully on the connexion without such ferrules. One feature in the firebox design was difficult to understand, namely, the seeming reluctance to use flexible stays in the “breaking” zone; to avoid them the unusual step was taken of employing non-ferrous stays in an otherwise all-steel design. In view of the world-wide acceptance of the flexible staybolt, it would be interesting to learn the reason for this-a feature in which this firebox differed from almost all other large steel fireboxes of which he himself was aware. Turning to the valve gear, after re-reading the paper carefully, the question which still arose was “Why was it necessary to design a new type of link motion and enclose it with the inside big-end in a casing, difficult to design and maintain?” Of all the parts of a locomotive, the normal Walschaert gear gave least trouble and maintenance cost, while a correctly designed and maintained inside big-end could give comparative freedom from failure when contrasted with the performance indicated by the author for his enclosed arrangement.
Taking the last point first, and referring to the well known three-cylinder Royal Scot 4-6-0 locomotive on the London, Midland and Scottish Railway, which had, over a period of 18 years, been engaged in top link work with fast, heavy trains, seventy engines ran 35 million miles in the eight years 1938-1945 inclusive, and sustained eighteen big-end failures in all, an average of practically 1,950,000 miles per failure. This compared very favourably with the author’s figure of 700,000 miles. Turning to maintenance, in three sets of valve gear, at service repairs averaging 65,000 miles, expansion links and valve rods might require new die-blocks. Other rods might require lapping and new pins and the eccentric straps might require re-metalling, at an average departmental cost, including material, of £20. At general repairs, every 165,000 miles, the valve rod and expansion link die paths were lapped, and bushes and oiling rings in the rest of the valve gear were renewed, in addition to what was done at service repairs, at a total cost (on the same basis) of £33. These figures represented a very low proportion of the total repair cost of the locomotive. Finally, the normal unenclosed valve gear could be dismantled, repaired, and returned to the erecting shop for refitting to the engine five days from its entry into the erecting shop for stripping. The foregoing facts seemed to suggest that there was only a very narrow margin of advantage in adopting any more complicated forms of construction.

P.C. Dewhurst (339-), wrote that the author was to be congratulated upon his combining, so successfully, features new to home practice, though well tried overseas, with others which were real innovations in the locomotive world. He himself had long advocated and successfully employed a number of the constructional features-having, in fact, championed “international” locomotive design for many years,* and his remarks were intended for the purpose of eliciting further information rather than as criticisms. He believed that the author was right in considering- that the main cause of electrolytic action in boilers was the use of steel and copper-a point mentioned specifically in his comments on Mr. Turner’s recent paper on “Corrosion of Boiler Tubes”.† Regarding corrosion as such, injector feed-water introduced by a plain top-feed system reduced corrosion very considerably.
Where weight was such an important consideration he could not follow the employment of the Belpaire pattern firebox. A round-topped radial-stayed firebox would have saved appreciable weight and this could have been utilized in increasing the barrel diameter, thus providing a few more (perhaps shorter) tubes and, or alternatively, increased steam space to compensate for that lost by the elimination of the “haunches" of the Belpdre outer shell. The only slight practical disadvantage of the round-topped box with radial staying (namely, that a few of the stays were not exactly radial with respect to the outside plate) had practically disappeared with the possibility of welding-on slight supplements in those places having only one thickness of plate, or of using flexible stays.
The success obtained with steel fireboxes, with the particular manner of “setting” or fixing the steel tubes in the firebox tubeplate, and with the syphons, was no surprise to one who had used the first for thirty, the second for twenty-one, and the third for ten years with eminently satisfactory results. These applications were required to meet conditions more arduous for boilers than even those encountered by the Southern Railway locomotives, conditions indeed which copper fireboxes could not withstand-in particular, the keeping of tubes tight in the firebox tubeplate. The use of 3/8-inch plate all round, except the tube area where a thickness of 5/8 inch was desirable, was good practice.
The ratio of superheating to evaporative heating surface (1 to 2.98) was a welcome figure, the superheater units occupying no less than five horizontal rows. He had tried a ratio of 1 to 3.6 upon engines required to work full-throttle combined with 60-70 per cent actual working cut-off for an hour or two continuously; but overheating of piston-rods in the packing under these extreme conditions persuaded him not to exceed that ratio. It would be of value to know the degree of superheat attained by the author’s engines at maximum output, and how the piston glands behaved.
The author had stated that the tube fixing adopted upon these engines had justified itself, “the tubes remaining completely tight, again something not previously experienced”. On the railway with which he had been connected recently a similar . system was in force, and no case of a tube leaking in service or having to be dealt with between general repairs had been known for a number of years. The author’s method of forming the shoulder on the water side of the tubeplate was not clear; his own system was to form the shoulder on the water side of the plate by means of a “prosser” (a roller expander being used only for fixing the tubes lightly in position, ready for the prosser) which also “bellied-out” the fire side ready for the beading tool. In this connexion it was of great importance to have the correct relation between the prosser profile and the thickness of the tubeplate so as to produce the required “+p” between the shoulder and the bead, but no more, otherwise the diametral tightness in the hole became affected. This tube fixing, as introduced by him on the various national railways of Colombia from 1923, was fully described in a paper* and was also discussed at an informal meeting of The Institution of Mechanical Engineers in the previous year.
The author appeared to be eliminating the copper liner. The liner was used mainly in order to preserve the surface of the tubeplate holes (i.e. the “bore”), and this was a matter of importance for future re-tubing, because even slight scratches communicating from each side of the tubeplate might cause trouble. Even the flat surface on the fire side of the tubeplate was important. His practice had been to cut off the whole bead and “weld-seal” flush when taking out tubes, and to re-face the landing for the bead and weld-seal, at which time any cases needing it were touched up with electric welding and machined to a good surface and correct thickness of tubeplate. No sealing of the tubes by welding was allowed until both hydrostatic and steam tests had been satisfactorily carried out; the sealing was always done with the boiler in its natural position and with water in it.
The use of corrugated steam-pipes was an innovation; some further iniormation on comparative maintenance and ultimate life would be welcome.
He admired the enclosing of the parts in the ingenious valve gear, but considered that there were too many pins. Moreover the two chains might lead to “lag” in the valve events, and this without giving the usual audible indications leading to running-shed attention. The valve gear might therefore require maintenance of a “nursing” character. Although reference had been made particularly to the valve-gear lubrication, nothing had been said about the outside connecting and coupling rods. Much convenience, and considerable saving in lubricants and enginemen’s time, could be obtained by grease lubrication on all the outside rods. Existing rods could be equipped for grease by a very small modification of the original oil-wells.
In the description of the special method of tyre-fixing, emphasis had been laid on the importance of evenly heating the tyres for shrinking. Had the author contemdated using an electric heater of the induction type? He himself had been prevented from adopting it by war-time difficulties, but he had been much impressed by the efficacy of the method as used abroad for the tyres of tramway vehicles.
In the main frames the author had succeeded in bringing the axlebox reactions into line with each frame, whereas they were ordinarily, except in bar frames, 4-5½ inches off-set. It was noticeable, however, that quite a lot of somewhat complicated cross-staying was built into the frame system, and the author referred to “ the place where cracks usually develop” and added that “wedges were not fitted”. These were unsatisfactory features of the plate frame; one day someone in England would try bar frames-and be surprisingly successful !
The reaction brake hangers placed between the coupled wheels were intriguing, but whilst he appreciated the advantage of situating the brake-cylinders away from the region of the firebox, he thought that there would be “chatter” as a result of the toggle-like effect upon the “duplex” hanger system when the engine was running forward. (For running backwards the arrangement appeared to be ideal.) Did not wear and chatter, and hence more wear, develop?
It would be interesting to know whether the author had considered a modernized Midland (Smith-Deeley) threecylinder compound arrangement. It would seem that the author * was so limited by considerations of axle-weight versus maximum starting effort as to be prevented, like himself on various occasions, from utilizing the system. The really high pressures foreshadowed seemed to call for compounding where other conditions permitted.
It had been said that nothing could be done about the loading gauge, but could not platform edges and “between line” girders (upon certain types of underbridges) be set back to allow more room in the region of the cylinders, as he had suggested some years ago in the technical press? A very significant feature of this locomotive design was the manner in w hich the adjustment of the whole optimum to the prevailing conditions had been made by the Chief Mechanical Engineer responsible for results, notwithstanding an ephemeral fashion to disparage this essential to satisfactorv locomotive designing.
* Proc. I.Mech.E., 1922, p. 375, “British and American Locomotive Design”
† Proc. I.Mech.E., 1944, vol. 150, p. 102.
Dewhurst, P.C. 1930 JI. Inst. Locomotive Eng., vol. 20 p. 888, “Locomotive Design for Overseas Service”.

D.W. Peacock (341), wrote that a feature of this engine which appeared to have escaped special notice was the unusually great stiffness of the frame to resist transverse bending. It was noticeable that care had been taken to ensure rigidity over the whole of the coupled wheelbase, in contrast to the frame of a conventional engine, which, while it might be stiff laterally at the front end, was necessarily weak where the firebox came between the frames. It was often claimed that some degree of transverse flexibility was necessary to negotiate sharp curves, and it was presumably incorporated in such designs. It would be interesting to know whether any additional axlebox and axlebox guide clearances had been found necessary for the Merchant Navy engines, or whether difficulties had been experienced in negotiating curves of small radius. What was the sharpest curve which the engine could traverse? The axlebox lubrication, though apparently old-fashioned as compared with mechanical lubrication, had the important advantage that extra oil could readily be given to an axlebox in need of it; and the usual objection to a trimming feed, namely, the possibility of a hot axlebox due to the driver’s forgetting to insert the trimming, hardly had much force provided the underpad was adequate. Would the author explain, however, what was meant by a breken syphon, and why the use of this fitting required the trimmings to be vented, since with a simple trimming feed and pipe to an axlebox the trimming worked excellently without any provision for venting at all.

F.A. Pudney (341) wrote that the author’s neat arrangement of three independent sets of valve gear, with a chain drive, was noteworthy. Would the author agree that at least 20 h.p. would be required to operate all three sets under certain conditions of steaming? The author mentioned an allowance of 3 inches for sag in the chain. Did that mean 1½ inches above and 1½ inches below the horizontal, or 3 inches above and 3 inches below? Some years ago, when he was closely acquainted with chain drives for certain Diesel engine camshafts, chain stretch became serious, for as soon as any “whip’y arose (a condition which could occur almost irrespective of transmitted load) chain inertia, during periods of out-of-phase or out-of-balance running, soon caused quite excessive sagging. . In some instances the only possible cure was the addition of jockey pulleys; in others brass and fibre rubbing (or guide) plates were added. It was of particular interest to read that so little trouble had been experienced in the chain drive, irrespective of the not too ideal conditions existing between the driving and the driven wheels. He asked if the chains could be readily inspected in the gear case, and whether the case was of fabricated construction, also whether any leakage had resulted from distortions in the framing. The housing of such an amount of valve gear in an oiltight bath represented a real step forward, but he asked if one or other of the poppet valve gears, with their many claimed advantages, had been considered when the locomotives were being designed.
Bulleid's reply (343) Pudney had asked about the horse-power necessary to drive the three sets of valve gear. Under certain conditions of steaming, this figure might be so high as 20, but no investigations had been carried out to ascertain the power thus absorbed. Actually, however, the gear was driven in the shops by a motor and the horse-power recorded was 3 at 300 r.p.m. It must be appreciated that there was no steam in the piston valves during these experiments, although all the rings were in place. Whilst the figure given for the sag of the chain was considered a maximum, no chain had as yet shown any appreciable amount of stretch after running up to 100,000 miles. He appreciated that there might be stretching due to chain inertia, but this had not been found to cause any distress, probably because the engine had not exceeded 450 r.p.m. The chains could readily be examined through an inspection door at the rear of the oilbath behind the crank axle, in the case of fabricated construction. There had been no leakage caused by distortion of the framing, as the frame was of very rigid construction. Poppet valves had been considered when the engine was being designed, but this type of gear was not entirely without its difficulties.

H.W. Puttick (341) wrote that the Nicholson syphons appeared to be very expensive, owing to the amount of press-work involved and the large number of stays; and it seemed almost impossible to clean them effectively. The cracking of the throat plate in the vicinity of the syphon neck was probably due to contraction stresses set up from the large mass of metal in the syphon body when the boiler was being cooled off. The syphons improved the circulation in the boiler considerably, but it seemed that this could be done equally effectively by having three 4-inch tubes set in triangular formation as shown in Fig. 34; the tubes would support the firebrick Nicholson Syphons arch in a similar manner to the syphons. It would, of course, be necessary to have two manholes in the wrapper plate so that the tubes could be expanded, beaded, and welded in place, but it would be very easy to clean the tubes with a tube cleaner when the engines went to shed for wash-out, and to renew them if necessary at intermediate or periodic overhauls. Even after allowing for the additional cost of the manholes, there should be a considerable saving in initial expense and weight, with very little reduction of heating surface. He was surprised to see that inverted tooth chains were used for the valve drive. Similar chains were used for the final drive of the Drewry railcars working on the narrow-gauge hill sections of the North Western Railway (India). When new chains of similar construction were indented for in 1941, the chain makers stated that these chains should be considered obsolescent, and it would be necessary to install double-roller chains with new pinions. He would be glad if the author would state what advantage-if any-the inverted tooth had over the roller chain. In this particular case a single-roller chain could have been used, as the horse-power transmitted was quite small. With regard to the illumination of the gauges in the cab, the luminosity obtained from ultra-violet rays on luminous figures was of a low order, and it would appear to be difficult for the fireman to read the gauges after looking at the intense glare emitted from the firebox. In India almost all locomotives were fitted with turbo-generators, which were mainly installed for the engine headlight; originally there were separate lighting fittings for the steam gauges, water gauges, and the lubricators. Owing to the excessively high maintenance, all these fittings, with the exception of the lubricator light, were taken out, and the cab was illuminated with one low candle-power lamp of 15 watts in a locally-made reflector fixed to the roof of the cab. This lamp gave sufficient light to read the gauges without being intense enough to interfere with the driver’s lookout
Bulleid's reply (343): the Nicholson syphons were not expensive when one considered the advantages obtained by fitting them. No difficulties had been encountered in maintaining or cleaning the syphons. Although the suggestion offered for the cracking of the throat plate could not be refuted, he still thought that a considerable amount of the trouble caused was due to the pressing of the syphon necks into the throat plate. The choice between arch tubes and thermic syphons was one which must be mzde by each engineer, and he himself preferred thermic syphons As Mr. Puttick was no doubt aware, special apparatus was generally used to clean out arch tubes, but it was not necessary with thermic syphons. The inverted tooth for the valve gear chain drive had been decided upon as it tended to fit more closely into the tooth, and as the chain stretched the inverted tooth tended to increase its width owing to the wedge-shape pivoting about its fulcrum, which was a rocker bar. He was quite satisfied with the electric lighting which had been fitted, particularly the ultra-violet rays for the gauges ; there was no difficulty in seeing these gauges at all times of the day or night.

H.P. Renwick (Bombay, 342) wrote to congratulate the author, first on his excellent paper, and second on his successful efforts in carrying forward the development of locomotive design on British railways. Most locomotive engineers on overseas railways looked (as he did) to British railways to give a lead, though they had no hesitation in adopting American or European practice where such appeared to offer advantages.
The author stressed the limitations of the British railways loading gauge, and it would seem that, since the maximum height and width had now been reached, the only remaining dirncnsion left for development was length. Whether longer and heavier trains would become the rule to meet the post-war traffic trend remained to be seen. In view of the cost of lengthening station platforms, he thought that the need could only be met by more trains; but the cost of providing longer turntables should not limit the development of more powerful locomotives of increased length.
As the author points o9t, the desirability of providing for power to haul freight trains at passenger speeds would necessitate more powerful general-purpose locomotives.
He could not but admire the very great amount of thought that had gone into the design of the Merchant Navy locomotive, with the aim of giving consideration to the footplate staff who would operate them and the running shed staff who would maintain them. It was the little things that counted in maintenance; and forethought in design meant reduced time in the shed and greater availability.
There were, however, one or two points that seemed to call for mild criticism. The author had mentioned the difficulties in designing a boiler of sufficient power within the weight restrictions. He stated that there had been a number of broken firebox water-space stays, all of which had occurred in the normal breaking zone shown in Fig. 5; yet he stated that flexible stays of the Flannery type were only installed in the throat plate around the necks of the thermic syphons. It had been standard practice on Indian railways to fit flexible stays throughout this zone, with remarkably beneficial results, and he would suggest a similar provision on the Merchant Navy locomotives. The additional weight could be offset by the elimination of the foundation ring which might well be replaced by a U-shaped flanged plate welded to the inner and outer sheets with two seam welds.
The weight of the boiler could be carried by sliding bearer brackets at the front end and by flexible “breather” plates at the rear, as was common practice in the U.S.A.
Similarly, the author referred to the stressing of the throat plate by pressing out the syphon neck connexions. Was there any reason why the curved connexions should not be pressed separately and welded into large-diameter holes cut in the throat plate? One might go a stage further and, placing the neck connexion to the syphon farther to the rear, spin a corrugation in the neck piece to aid flexibility at this point and relieve the throat plate. The speed of water circulation through the syphon should prevent accumulation of scale in the corrugation.
He admired the author’s attempt to produce an enclosed valve gear, continuously lubricated as copiously as was shown in the film exhibited at the meeting; but he could not help feeling that the possibilities of leakage from vibration with age were great, and the maintenance costs to prevent this leakage must be high. He had hoped that the author would have given some comparative figures of big-end and slide block wear between the Merchant Navy and the Schools (4-4-0) classes, particularly the man-hours expended on the two classes to reduce and refit big-end brasses and close slide bars.
Until the stage of the multi-cylinder gear-driven high-pressure locomotive of the Sentinel or Besler type had been reached, he felt that the development of the poppet valve gear wirn enclosed shaft drive would give the same desirable end that the author had endeavoured to attain, with much less weight and complexity. For the benefit of overseas members, he hoped that the Proceedings would contain a more detailed description and an illustration of the motion-pin “circlips”, for he felt that their greater use was a move in the right direction. He was greatly indebted to the author and to Wing Commr. Cave-Browne-Cave for their remarks on the experimentation to decide the best shaping of the front end screening to satisfactorily lift the smoke at short cut-offs. They had explained his own failure to achieve satisfactory results with a similar problem on the Great Indian Peninsula Railway.

Volume 155 (War Emergency Issues, 13-24, 1946)

Plummer, G.A.
The Development of the La Mont Boiler in Great Britain. 333-45. 9 illus., 20 diagrs.
The La Mont forced-circulation boiler was introduced into this country in 1936, and although considerable experience was already available on the Continent (where some 150 installations then existed) the need for further development was immediately apparent, not only to bring this type of steam generator into line with British practice, but also because the installations then existing were comparatively small units operated for the most part at moderate steam pressures and temperatures. Author was Director in charge of development and research, Messrs. John Thompson Water Tube Boilers, Ltd., Wolverhampton.

Begg, G.A.J., Hebblethwaite, W.M., and Cooke, G.
Operating experience with La Mont boilers, with special reference to feed water problems. 346-57. Disc. 357-75.
Experience at the power plant of I.C.I., Dyestuffs Division and I.C.I., Alkali Division. The discussion (all in the form of written communications) do not appear to show any direct railway interest, although it would seem that Stanier was well aware of this work going on "not a thousand miles from Crewe".

Volume 156 (1947)

Bulleid, O.V.S.
Presidential Address.
1-5 + 10 plates. 42 illus., diagr., 3 tables.
Included the reason for introducing the Leader design as follows:
1 To be able to run over the majority of the Company's lines.
2 To be capable of working all classes of trains up to a speed of 90 miles per hour.
3 To have its whole weight available for braking and the highest possible percentage thereof for adhesion.
4 To be equally suitable for running in both directions without turning, with unobstructed look-out.
5 To be ready for service at short notice.
6 To be almost continuously available.
7 To be suitable for 'common use.'
8 To run not less than 100,000 miles between general overhauls with little or no attention at the running sheds.
9 To cause minimum wear and tear to the track.
10 To use substantially less fuel and water per drawbar horse-power developed.

Diamond, E.L.
Development of locomotive power at speed. 404-16. Disc.: 417-43.
This was a very important Paper and led to much discussion, only some of which has been extracted.
The power developed per unit of cylinder volume of locomotives in the speed range 250 to 400 rpm had doubled. In the 1900s mean effective pressure tended to diminish with increase of speed, rather than with boiler evaporative capacity to limit engine power. The most advanced designs maintain the mean pressure at a high percentage of the calculated value to the highest speeds, and the effect of valve events and clearance volume on the calculated mean pressure thus becomes of practical importance, especially as designers were endeavouring to use considerably higher steam pressures in single-expansion cylinders.
The effect of these factors over a wide range of steam pressures in the form of basic data graphs of mean pressure, relative efficiency, and steam consumption. It is shown that without the decrease of clearance volume and increase of expansion ratio which compound expansion affords, the improved thermal efficiency which higher steam pressure offers cannot be fully realized above about 250 psi boiler pressure, though greater power can be obtained.
Then an examination was made of the actual deviation of mean pressure with speed for a wide range of locomotives, including some of the most recent designs, and the reasons for the radical improvement are indicated. In 1905 Dalby suggested a simple proportional relationship between mean pressure and speed. Such a relationship did not hold for modern locomotives, and a simple exponential law is proposed with a single coefficient characteristic of the locomotive. This law may be used to estimate the power of a future design at any speed, or as a criterion for assessing the performance of an existing locomotive in respect of power developed, which is generally of more importance to railway companies than thermal efficiency.
Steam locomotive design may appear to have remained static for the previous fifty years as compared with the technical developments of stationary steam plants and other forms of prime mover. It is true that there have been no fundamental changes in the construction or the working principle of the steam locomotive apart from the use of higher steam temperatures. Yet the power capacity of the average locomotive of 1900 was a mere fraction of that of the then best contemporary designs, for the same unit cylinder volume at the speed of running prevailing in 1939. If the performance of locomotives over the range of speed 250-400 r.p.m. was examined, it may justly be claimed that progress has been commensurate with that of other established types of engine. If it had not been so, it is almost certain that the non-condensing reciprocating engine would not have maintained its pre-eminence for railway service.
It has to be admitted that the spectacular developments of locomotive performance in recent years are not entirely the result of devices or principles unknown at the beginning of the century. The performance of a locomotive in practice is, however, so dependent on the steam-producing capacity of the boiler that there grew up a habit of basing any study of its power on the evaporative capacity of the boiler. Some years ago the author made a survey of the various methods that have been used for calculating and measuring locomotive horse-power, and it will be found that almost all the early formulae were based on the boiler, notably those of Frank (1887) and Goss (1901) which expressed the power of a locomotive as a function of the heating surface. It has generally been a fundamental assumption that the power of a locomotive is limited by its evaporative capacity, but whilst this is obviously true in a broad theoretical sense, it has not always been true in practice of large-wheeled locomotives.
D.K. Clark, whose paper on “Expansive Working of Steam in Locomotives” read before this Institution in 1852 can still be read with interest. Clark laid down that the volume of the high-pressure steam chests should be equal to that of one cylinder. This rule had long been forgotten till the Nord Company proved once more its value in some tests in 1897.
Mean Effective Pressure Curves for Various Locomotives on Truly Comparative Basis (The values of K in the characteristic power equation for each locomotive are given alongside the corresponding curve):.
1 Midland 4-4-0 superheater locomotive (1917).
2 Original P.O. compound superheater Pacific locomotive (1909).
3 Pennsylvania K4S locomotive in original form.
4 Pennsylvania K2S locomotive.
5 Pennsylvania E6S locomotive.
6 L.M.S. rebuilt Royal Scor locomotive.
7 Pennsylvania K4S locomotive rebuilt with Franklin poppet valve
8 Rebuilt P.O. compound locomotive No. 3705 (Chapelon).
9 Pennsylvania TI locomotive with Franklin poppet valve gear
Alternative abstract see Loco. Rly Carr. Wagon Rev., 1947, 53, 66-7.
Discussion: See also Tuplin's comments on low pressure locomotive boilers.
E.W. Marten (419 London) said that sustained power at speed in modem locomotive design was not, as formerly, so much a question of boiler capacity as of cylinder performance, the yardstick of which was mean effective pressure, and to maintain this at a maximum figure over the speed range the steam had to flow with the minimum of restriction through passages and ports of adequate proportions, and further, correct relationship and accurate timing of the valve events was imperative. The author pointed out that his main assumption in this respect related to the point of compression as shown by the inset curve in Fig. 1, which, he stated, was desirable for any type of gear. That, however, was open to question, since from high compression, limiting therefore short cut-off working, the curve fell to zero at something over 60 per cent cut-off.
In actual practice some compression was, of course, essential throughout the range, but otherwise the characteristic of the author’s curve well illustrated that, even with a modern Walschaerts gear, it was all a matter of compromise since, with this design, there must be a rapid rise to excessive compression at the shorter cut-offs.
On the other hand, with valve events independently controlled (as with cambox-actuated poppet valves) a closer approach to the ideal was possible, the degree of compression, for a given clearance volume, being made to vary and correspond to the particular cut-off. Since compression aided fuel economy, and as the greater percentage of compression was (as the author remarked) equivalent to a lower percentage clearance, the importance of clearance as stressed in the paper appeared to omit the practical consideration that, in insisting on clearance being an absolute minimum, a loss by wire drawing might be involved where the development of maximum power required large valves. It would be of interest to know how the author would suggest overcoming the difficulty of achieving in cylinder design what in effect he advocated, namely maximum steam-flow area and, at the same time, minimum clearance volume-two conAicting requirements that called for a compromise—for the lowest clearance volume consistent with the large valve and passage areas.
The paper emphasized that the exact point of compression was all-important to the accuracy of the results to be determined by the author’s method, and although the foregoing remarks affected somewhat the value of the coefficient k, his equation curves appeared generally to agree well with the selected tests, and the same might be said in regard to some test records in the writer’s experience. But the more modern designs cited were confined to locomotives having piston valves or poppet valves operated by Walschaerts gear, or with gear working on a similar principle. To these examples might be added cambox-operated poppet valves working on the rotary-cam principle, the latest application of which was soon to be seen on main-line locomotives in this country. Compared with the areas of Fig. 13, the noticeable increase possible with this system, in the maintenance of port openings down to the shortest cut-offs, was apparent from Fig. 25. Coupled with the fact that the lead also increased on notching up, adequate head of steam was therefore assured.
The best showing in the paper was with the poppet valve; this demonstrated again what was invariably found in comparative trials : that for any given cut-off and speed the poppet-valve gear produced a higher mean effective pressure, and hence a higher horse-power in the cylinders, than did Walschaerts valve gear.
The results with the T.l class locomotive in particular served as a classic example of improvement in power output and economy as a result of design of cylinders and valve gear, with correct proportioning of areas throughout the circuit from regulator head to blast-pipe. This design gave the lowest water rate consumption in the forty years of test experience at Altoona.
E.S. Cox (p. 420-) noted dynamometer car tests run in 1945 on rebuilt Royal Scot class showed a 5.5% saving in fuel on 450-ton trains between Crewe and Carlisle.
T. Robson (Darlington: p. 439) agreed with the author that the usual clearance volumes and ratio of expansion at present possible in one cylinder made the use of increasing boiler pressures uneconomical without compound expansion : they had not yet succeeded in makmg the best possible use, in a singleexpansion locomotive, of steam pressures and temperatures that had been in use for twenty years.
Many engineers had drawn attention to this; he himself had submitted, ten years ago, original experimental data which he hoped would prove convincing. There was a long time-lag before new ideas were generally accepted — e.g. the long-lap valve, the most important and most recent economic improvement in the single-expansion locombtive since superheating, was brought out by Churchward about forty years ago, but twenty years had to elapse before it was generally adopted.
There had been great advances made in the development of power at speed in France and America. There had been nothing comparable in Great Britain, because the large engines that were built shortly after the main line railways were amalgamated four large groups were found adequate to deal with the loads and speeds required. The economy in coal consumption, due to superheating and the eventual adoption of the long-lap valve, was so great when compared with the very poor results from the older engines that it was considered satisfactory, having regard to the many more pressing administrative problems. The lower diagram in Fig. 12 was not representative of early cut-off working with Walschaert’s gear, because the lead of inch was about twice as much as usual, resulting in a much better steam-port opening and therefore a higher value for k than was possible with a normal amount of lead. Although the diagram was as good as could be expected from a piston valve, the large clearance-volume considerably limited the economical use of the high steam pressure.
He wished to stress that, in using the author’s method of finding k, it was necessary to compare the actual cut-off with the graduations on the reversing gear, particularly in early cut-off positions, and to correct these: a very slight movement of the reversing gear in the early cut-off positions changed the mean effective pressure, and the drawbar pull, very considerably. Removing all other variables, by making constant-speed tests on the level using a counter-pressure brake-engine, had shown that the drawbar pull of a modem type locomotive in 15% cut-off was more than doubled at ordinary running speeds by moving the reversing wheel half a turn towards full gear. The writer had been present during tests with the latest Chapelon compounds and was impressed with the large drawbarpull at high speed and early cut-off. Expressed as a percentage of the rated tractive effort, it was at least twice as great as that possible with conventional single-expansion locomotives, so it was surprising that the value of k shown in Fig. 15 was not higher.
The low value of k shown in Fig. 14 for the original locomotive might be due to condensation in the low-pressure cylinders owing to insufficient superheat. In the later engines this was increased; he had noted inlet-steam temperatures as high as 800°F. Probably these compounds were the first that had had dry steam at the low-pressure exhaust ; this, in his opinion, was a very important reason for their success. Moreover, as a result of two-stage expansion, there was less port restriction, when working in an economical cut-off at high speed, than with single expansion.
For the provision, with a single-expansion locomotive, of greater power at high speed, the obvious devices of building larger locomotives, or working existing ones in longer and therefore less economical cut-off positions, or using larger piston valves, were presumably considered too wasteful; the aim was to obtain a much higher proportion of the rated tractive effort than at present, when working in an early cut-off. It could not be done by a piston valve: natural laws governing steam flow precluded such a possibility. It could only be done by using independent steam and exhaust valves having wide and straight port openings and passages.
The slope of the curves for the poppet-valve engines on Figs. 18 and 19 was very good, but the values of k for the T-1 locomotive varied a good deal — perhaps as a result of the inaccuracy in graduation already referred to. The flatness of these curves for the different cut-off positions could be attributed in part to the large steam openings, and in part to the valve springs. Above a certain speed the cut-off started to lengthen without alteration in the reversing gear, owing to the increasing kinetic energy in the valve and the increasing force necessary to close it in the decreasing time. There could also be peculiar surge effects in the springs
At about 75 m.p.h., with a usual size of driving wheel, the time interval for steam admission in 15% cut-off was the cut-off to 25%. The delayed closing of the exhaust valve increased the amount of live steam required to fill the clearance space. The combined effect gave a higher mean effective pressure than could be obtained if the valve events were true as with a positively closed piston valve, and a false value of k at high speeds, making the valve appear better than it really was.
The writer had made many tests with poppet-valve locomotives, including comparative tests between the rotary-cam type and long-lap piston valves, modern passenger engines These tests were made at a series of constant speeds in different reversing-geir positions using a counter-pressure brake engine.
The results from two poppet-valve engines were practically identical, but were disappointing when compared with the piston valve, which showed much greater economy over a wide range of power at ordinary running speeds. These engines had four valves per cylinder; the effect explained would be less with eight valves per cylinder but could not be eliminated. In addition to this, the sudden changes in direction through the poppet valve and the steam passages were just as bad as with the piston valve, and hindered the free steam-flow during the short time interval; also the clearance volume was too large. In his opinion, the performance of the poppet valve at high speeds was so far from the ideal, and eight valves per cylinder were such a complication, that it would be better to start developing a valve of new type free from these defects.
As the time interval of admission was 1/45 second or less, to get the maximum possible weight of steam into the cylinder and the minimum pressure drop (from boiler pressure) at cut-off, a type of valve giving a wide opening to a straight steam-port of large area, through a passage as short as possible to keep the clearance volume small, and leading directly into the cylinder without change of direction, was required. There was no other way of getting greater power at high speed with high thermal efficiency. In his opinion, direct openings of adequate size could be provided only by the use of rotating valves for admission and exhaust. The author compared the relative usefulness of a locomotive testing plant and a single-cylinder unit, for making improvements in the steam circuit. The single-cylinder unit was more useful, because alterations could be made more cheaply, and it was, in his own opinion, the key feature in design. If economic development was required, such a unit must be provided, and engineers with the necessary aptitude must be concentrated on the problem.
A unit built up in sections, to enable alteration to be made cheaply, was suggested in this country thirteen years ago. It was a mistake to think that locomotive development in this country had been delayed owing to the lack of a testing plant : it was due to the concentration at the grouping of so much responsibility on so few chief engineers. Whilst great improvements were made in workshop methods, administrative work and correspondence left insufficient time for the problems met with in development work. A lot of experimental work had been done before the war, and this showed where improvements were desirable; it remained only to make a start.
What a difference it would make if those directing large undertakings planned a less concentrated organization, in which experienced engineers with inventive ability were free to give of their best ; he feared nationalization would put an end to such a hope, and this failure to harness the individuals creative instinct was the primary handicap to development.

Volume 157 (1947)

Bulleid, O.V.S., Peppercorn, A.H., Hawksworth, F.W. and Ivatt, H.G.
Railway power plant in Great Britain. 235-9 + 4 plates. 12 illus., diagr. (s. & f. els.) 2 tables. (Centenary Lectures).
The Institution celebrated its Centenary by presenting surveys of the state of the art in many activities. Thus, the surveys were very concise, but in the case of two of the contributors (Ivatt and Bulleid) very concentrated. Peppercorn's and Hawksworth's contributions (especially the latter) were lesser in impact, but still give an excellent survey of locomotive development at that key time. With the exception of Bulleid who was a prolific contributor, the other contributions were virtually their sole public utterances. Greater detail is presented under each of the authors. Loco. Rly Carr. Wagon Rev., 1947, 53, 106-7 carried a long precis, but this is not reproduced (see references to each contributor).

Armand, L. Motive power trends on European railways. 239-45. 9 diagrams, illustration. (Centenary Lectures). 
Following WW2 devastation the prime task for the railways was the rehabilitation of existing machines, but the magnitude of the railways’ requirements permits consideration to be given also to new equipment and the launching of new building programmes, and consideration was given to directions in which the railways might develop. The survey covered steam, diesel, and electric locomotives from the mechanical engineering aspect.
Special versus Conventional Locomotives. Between the two world wars a great effort was made on the continent to carry out experiments with new types, in which a departure was made from the conventional Stephenson locomotive. It was considered that the overall thermal efficiency at the wheel rim was poor, and an early series of researches, carried out particularly in Germany between 1925 and 1935 was aimed at the realization of a higher efficiency by the use of higher pressure (as in the Schmidt, Loffler, and other similar locomotives), or by the use of turbines and condensers (as in the Krupp, Maffei, and such types). These attempts came to an end, as is well known, owing to the conclusion that the savings which they gave were not justified in view of the additional complication and increased maintenance costs.
In France, there was a notable increase in efficiency — a rise from 8% to 11% — had been obtained with conventional locomotives of the Pacific (4-6-2) type on the Paris-Orleans Railway in 1930, trials which as yet are scarcely completed, have been undertaken with special types of locomotives with a view to obtaining : (1) a higher efficiency by the use of high-pressure water-tube boilers; (2) the employment of novel driving mechanism to take advantage of high-pressures and to eliminate coupling rods.
These trials are outlined in the following three subsections, in which the main conclusions arrived at on the French Railwavs are also given.
(a) Velox Boiler.
A Velox boiler was fitted in 1939 to an ordinary recimocating 4-6-0 locomotive; the pressure is 16 kg/cm2, but this boiler was equally capable of producing steam at pressures of 50-80 kg/cm2. This type of boiler was costly and too complicated for railways, and in present circumstances a diesel locomotive might be preferable, as it would appear to offer the same advantage in high-speed service whilst showing also a better efficiency.
(b) Winterthur High-pressure Locomotive with Individual Axle Drive.
The Winterthur 4-6-4 locomotive placed in service in 1939 provided experience with two new departures from conventional practice: a high-pressure boiler working at 60 kg/cm2 and the use of small steam engines — two per axle — each having three cylinders. The crankshaft was driven at 1,000 r.p.m., and each unit developed 500 h.p., giving a total of 3,000 h.p. for the locomotive. Trials of this machine were suspended during practically the whole of WW2, but had resumed. The driving units gave satisfaction, but the boiler was shown by its behaviour to be somewhat delicate owing to its construction in two portions, one boiler carrying water at 20 kg/cm2 and feeding the high-pressure boiler working at 60 kg/cm2. The saving in fuel, which probably did not exceed 20 % did not justify the difficulties of maintaining the boiler, which had the further drawback that its components are inaccessible. Nevertheless this experiment should eventually make some extremely interesting information available for consideration when it is desired to build locomotives with driving bogies, e.g. a machine mounted on two six-wheeled bogies to eliminate entirely the “hammer blow” of conventional locomotives. However, the scope for such applications is limited.
(c) Schneider Locomotive with Individual turbine drive.
In the Schneider 4-6-4 turbine locomotive, the boiler, of the conventional type, works at 25 kg/cm2, but on each axle was mounted a turbine running at 10,000 r.p.m., and developing about 1,000 h.p., i.e. a total of 3,000 h.p. Placed in service in 1938, this engine underwent trials at the Vitry testing station where it developed an actual horse-power of 2,600 at the wheel rim. When running at 100 km/h (62 m.p.h.) the fuel and water consumption was that of a good locomotive of normal type. The experiment, however, was not directed towards the attainment of a high efficiency, but was rather a research into a new type of drive. The engine was kept in service for some time, but unfortunately was severely damaged by the Germans in 1944 and could no longer be run. Nevertheless the experiment enabled the conclusion to be drawn that it is possible to apply individual turbine drive to the axles of a steam locomotive. Thus it should now be possible, as mentioned in connexion with the Wimterthur locomotive, to produce a steam locomotive with driving bogies, or an extra high-speed machine perfectly balanced in regard to hammer blow; but these may well be expensive types of construction, limited to special cases where they are justifiable. The diesel-electric locomotive has established itself since the idea of a locomotive with individual turbine drive has been put forward, and may afford a more practical solution.
Conclusions regarding Special Types of Locomotives.
The conclusion to be drawn was that European experience, first in Germany and more recently in France, had demonstrated that complexities of special locomotive types nullify the advantages to be gained from them, and that simplicity is a great virtue in locomotives. It is this simplicity of construction and of operation which is the chief defence of the normal locomotive against competition from other forms of motive power. It can therefore be concluded that the steam locomotive will probably continue to follow, and indeed to complete, its development on the lines of the conventional type.
It may be stated in general terms that, after having undergone a development in thermodynamic details between 1930 and 1940, steam locomotive design was dominated by an effort to achieve a robust construction in order to reduce maintenance cost and to increase the annual mileage. The increased high level of wages in maintenance work, and the need for amortizing the prime cost whilst materials were continually rising in price, make it of paramount importance to reduce the time during which engines are stopped: the principal means to attain this object, both from the constructional and the operating viewpoints were reviewed.
Increased Axle Loading. The robustness of the locomotive, in regard to its various components, is bound up with the permissible axle load. The drawbar horse-power of the French “Pacific” locomotives, weighing 105 tonnes had by 1930 risen from 1,600 to 2,300 as a result of thermodynamic improvements, without any increase in engine weight. A higher adhesive weight was, however, shown to be desirable, whilst further improvements which suggested themselves were the fitting of a more powerful boiler, the use of roller bearings, and further strengthening of the frames, which were by no means sufficiently rigid. On most French lines the maximum axle load was 18.5 tonnes. Certain sections, however, notably the Nord, could take 21 tonnes. It was decided to raise the axle loading on French main lines to 23 tonnes, and to 20 tonnes on most of the other French lines. In Belgium, an axle loading of 24 tonnes has long been permitted. Spain has railways with a 22-tonne maximum axle loading. This increase is a general tendency in Europe. An axle load of 22.8 tonnes already exists on many British railways, notably the London, Midland, and Scottish Railway and the Great Western Railway.
Concurrently with the use of a higher axle load there has also been a tendency, in order to allow of a general increase in dimensions, to introduce an additional carrying axle at the trailing end; this led to the 4-6-4 or 4-8-4 types for passenger work, and undoubtedly we shall see 2-8-4 or 2-10-4 types for freight.
Boiler Construction. The boiler, as is well known, is the most expensive part of a locomotive from the viewpoint of maintenance. To ensure that the boiler is worked to a reasonable degree, there is a tendency to use the largest possible grates. The 2-8-2 French locomotives recently built for mixed traffic have a grate area of 5 m2 and a grate area of 6 m2 can be envisaged for more powerful locomotives.
The thermic syphon, which is intended to increase heating surface and to improve circulation, has been adopted on all recent French locomotives of the 2-8-2P 2-10-0P, and 2-8-2R types. As regards construction, both American and Belpaire types of firebox are in use. French preferences finally favoured the Belpaire type.
A great saving in maintenance is obtained by using a welded steel firebox in conjunction with complete system of feed-water treatment. The welded steel firebox has the advantages of (1) a lower first cost than a copper firebox; (2) a higher resistance to deformation of the firebox plates, since copper plates are liable to deformation in engines carrying pressures in excess of 16 kg/cm2. and (3) complete freedom from leakages both in the tubeplates and also in the seams. Whilst steel fireboxes have been widely adopted in France, their use is not yet general in Europe. The Belgian railways did not use steel fireboxes until 1945, when they ordered some 280 locomotives, so equipped, from Canada and the U.S.A. Steel fireboxes are, however, in use in Spain; and it is possible that Germany may extend the use of the steel fireboxes which were introduced there during the war.
Water Treatment. Treatment of feed water is one of the essential means for reducing both maintenance costs and the time during which normal locomotives are stopped. The procedure employed in France since 1939 consists of the introduction of a compound which will prevent sludge from adhering to the boiler, whilst at the same time protecting the steel plates against corrosion. The improvement thus obtained has been considerable. Instead of having to replace about 400 firebox stays at each overhaul, in bad-water districts it has been only necessary to replace a few dozen; and the maintenance costs have been reduced in the ratio of 10/1. In one extreme case at Joncherolles depot (Nord), the mileage between consecutive general repairs doubled, rising from 210,000 km to 420,000 km. The treatment is carried out in the tender itself, but can equally be effected at fixed stations. Good results can only be obtained by obedience to strict rules which oblige the water-treatment staff to carry out the blowdown routine in a regular and methodical manner. The amount of water blown down may be about 200 litres every 50 km.
Mechanical Stokers. A mechanical stoker assists in solving the problem of operating long runs because it obviates fatigue on the part of firemen and allows the fire to be built up again, if need be, en route. It also solves the problem of firing powerful engines which may burn more than 2,000 kg. of coal an hour. Finally, it fulfils a need that is very evident from the humanistic viewpoint. In France it has the added advantage of allowing the use of non-coking coals, since that country is not well supplied with coal of the coking variety. Mechanical stokers were first applied in France in 1932. The total number of French locomotives with mechanical stokers, including machines now under construction, will shortly number about 1,700. It is also noteworthy that experiments in their use were begun in Spain in 1935.
Frames. The benefit of a rigid framework has already been mentioned. Until recent years the framework consisted of longitudinal steel plates from 28 to 35 mm. in thickness, which were but feebly supported by cross-stays. A first step towards improvement was made when bar frames were adopted, in which the longitudinal members were 100 mm. thick and which possessed sufficient rigidity by themselves in the transverse direction. This type of bar frame has been employed to some extent in Belgium, France, and Spain. A better solution can, however, be obtained with cast steel frames of the “monobloc” type used in America.
Cylinders. An important step was taken in France, in the direction of achieving a more robust design, by abandoning the four-cylinder machine and adopting a disposition with three cylinders. The axle with two cranks, necessary for four-cylinder locomotives, is too fragile for modern machines.
The use of three cylinders provides a compromise between the quest for a strong crank axle (in which web thicknesses have increased from 110 to 250 mm., and a good stress distribution. The tendency to use three cylinders is general in continental Europe. The Reichsbahn in Germany has been led to adopt this arrangement for its locomotives, e.g. the 2-10-0’s, type 44; the 2-10-2’s, type 52; and so on, as well as for its Pacific (4-6-2) type 03, in preference to the two-cylinder arrangement. This was done for considerations relating to the maintenance of the mechanism and to the distribution of the drive. Locomotives of the 4-8-2 and 2-10-2 types, which have been built in Spain since 1942 — powerful machines with grate areas of 5 m2 also utilize three cylinders. Naturally the preference accorded to three cylinders only applies to powerful machines. For medium-powered locomotives in France and elsewhere, two-cylinder simple-expansion locomotives are now favoured.

Kiefer, P.W. Railway power plant from the United States point of view. 245-51. 5 diagrams. (Centenary Lectures).
Summary of motive-power evaluations. The relative evaluations given below are predicated upon:
(a) Locomotives of equivalent power and representing the latest state of the design art.
(b) Equivalent through-line passenger schedules and freight operations and efficient use of potential availability.
(c) Equivalent maintenance and servicing attention at all times.
(d) Accumulated knowledge and experience.
(e) The exclusion of fixed charges and maintenance expense, for motive-power operating, servicing, and repair facilities. Where the use of steam may gradually decrease, some reduction in the facilities required should take place, but as a partial offset to the resulting savings, moderately increasing investment is required in suitable facilities for other forms of motive power as their number becomes greater.
It is emphasized that the following ratings for the gas- and steam-turbine locomotives are speculative and at the present stage are based on design characteristics, and possibilities only.
For the straight electric, Diesel, and steam, the order of placing are founded on substantial experience, but for the gas- turbine and steam-turbine the placings are estimated:
(1) Availability: straight electric, gas-turbine, Diesel-electric, steam-turbine, reciprocating steam.
(2) Over-all Costs of Ownership and Usage: Diesel-electric or reciprocating steam, straight electric (gas-turbine and steam-turbine not placed as there are insufficient data for estimation).
(3) Capacity for Work: straight electric, gas-turbine, Diesel-electric or steam-turbine or reciprocating steam.
(4) Performance Efficiency: Over-all performance: straight electric, gas-turbine, Diesel-electric, steam-turbine, reciprocating steam.
Thermal Efficiency at drawbar: Diesel-electric, straight electric, gas-turbine, steam-turbine, reciprocating steam.
Finally, to maintain the supremacy of rail transportation-so far as this may be accomplished through the selection and introduction of modern locomotives and cars-properly balanced quantities and kinds of rolling stock, both passenger and freight, must be currently acquired. As conditions permit, this is being done on the railroads of the United States.
For passenger service alone, the New York Central System is now taking deliveries of a total of 720 new day-service and sleeping cars of most modern construction and of thirty-one different types, fitted with the latest auxiliary equipment and furnishings, which are to be made up into fifty-two additional streamlined passenger trains for operation on improved schedules throughout the system.

Bulleid, O.V.S. Closing remarks. (Centenary Lectures). 251-2
The Southern Railway operates the largest electric service in the world—a suburban service, almost entirely consisting of multiple-unit stock.
For main-line work the Southern concentrates on mixed traffic locomotives, and all their requirements are expected to be met by two classes of tender locomotives and two classes of tank locomotives. All recent engines have totally enclosed, continuously lubricated valve gear, steel fireboxes with thermic syphons, improved wheels, and clasp brakes.
The Southern Railway have built two Co+Co mixed-traffic electric locomotives, which have given excellent results in service, and are considering high-speed electric locomotives which will compare with the steam mixed-traffic locomotives. They are 19 fitted with boosters and can run over long gaps in the conductor rails without difficulty. The bogies are of unusual design without bolster.
The Southern also has three 350-h.p. Diesel-electric shunting engines, and have in hand three Diesel-electric locomotives fitted with 1,600-h.p. engines for main-line work at speeds up to 90 m.p.h. Experiments are also in hand with a six-coupled Diesel-mechanical shunting engine which is intended for hump shunting, pick-up goods service, and branch-line working at speeds up to 45 m.p.h.
The Great Western Railway has continued the policy instituted by the late George Jackson Churchward in 1902, with improvement in detail.
This policy has resulted in exceptional standardization of parts. with a booked stock of 3,860 engines, 3,575 of G.W.R. design. There are 3,119 standard locomotives divided into twenty classes, the whole of which are equipped with boilers drawn from a range of seven standard boilers.
The company is at present equipping a number of engines for burning oil and have made some progress in trials of a Diesel-electric shunting engine of 350 h.p. New developments are in the direction of the utilization of gas turbines, and two 2,500 h.p. gas-turbine electric locomotives have been ordered and are expected in 1949.
The London, Midland and Scottish Railway has been able to reduce the standard types to ten classes, and has incorporated a number of improvements to assist availability. The L.M.S.R. are carrying out experiments in roller bearings, poppet valve gears and steel fireboxes, in the hope that there will be a further improvement in the time between shoppings, and in the maintenance in efficiency.
An extensive trial of Diesel-electric locomotives is being made, a most satisfactory shunting engine of 350 h.p. having been developed. An 800-h.p. unit for cross-country and branch-line services and two 1,600-h.p. units for main-line express trains (which can be used singly or as a pair) are under construction.
L.M.S. authorities hope to explore the whole traffic field and settle where Diesel-electric traction is most likely to justify itself, and to ascertain how far this type of locomotive can be considered as an alternative to main-line electric locomotives.
The London and North Eastern Railway has eight new standard designs, which are expected to meet all requirements. Experiments are being made on controlled water treatment, with the possibility of using welded steel fireboxes, and research is being carried out in valve gears and poppet valves.
Diesel-mechanical railcars to cover a wide range of their secondary services, and Diesel-electric units of 350 h.p. for shunting purposes, have been proposed. A mixed-trafiic electric locomotive of 1,868 b.h.p. has been built for the Manchester-Sheffield electrification scheme.
The Chief Engineer of Motive Power of the New York Central Railroad has presented a most valuable review of the locomotive position in the United States as it exists to-day, and he quite rightly emphasizes the fundamental requirements of availability and utilization. Improved rights of way and operating practices demand improved locomotive design to support the more intensive utilization made possible.
A motive power unit must possess a reasonable margin of capacity over that necessary to perform the appointed task. Rapid acceleration from rest, or after slacks, is a characteristic much desired, in addition to capacity for running. In fact, a good big engine is better than a good little one.
In the United States the trend is towards higher standards without over complication ; the characteristic curves, drawbar horse-power and drawbar pull, for various United States engines, illustrate the improvement in power. The Locomotive Development Committee, in association with Bituminous Coal Research and the General Electric Company, is producing a number of new types using pulverized coal as the fuel for a steam-turbine locomotive, and two gas-turbine locomotives with approximately 3,750 s.h.p.
The table giving the principal characteristics of reciprocating Steam locomotives is a valuable contribution to the lecture. The lecture concludes with a review of what is taking place in the Diesel-electric field, and is illustrated with some interesting curves comparing steam, electric, and Diesel locomotive performance: tables of annual costs, and of the acceleration characteristics of steam, Diesel-electric, and electric locomotives complete this very valuable contribution.
The Deputy Director General of the French National Railways reports that between the two world wars numerous experiments were carried out to obtain increased efficiency from the locomotive : Schmidt and Löffler high-pressure boilers, Krupp and Maffei turbine and condenser locomotives were tried. It was generally found that the savings did not justify the additional complications and increased maintenance costs.
In France the experiments made by M. Chapelon opened up a new field of improvement for conventional locomotives. These results were hardly established before experiments were started with the Velox boiler, Wintherthur high-pressure locomotives with individual axle drive, and Schneider locomotives with individual turbine drive, but in general the increased cost of maintenance and the complications required to obtain the theoretical efficiency make the innovations uneconomical.
The trend of modern steam locomotive design in Europe is towards reduction of maintenance costs and towards longer daily hauls, simplicity, increased robustness (leading to an increase in axle load), and to design for general-purpose service. Steel fireboxes are coming more and more into favour with the introduction of controlled water treatment. For high power, a preference is being shown for three-cylinder arrangement instead of four, and a new three-cylinder compound designed by M. Chapelon has just gone into service. The French Railways prefer piston valves to poppet valves—they require less maintenance.
Diesel-electric locomotives of 600 h.p. are being used in increasing numbers for shunting purposes, but very little has been done to develop a main-line Diesel-electric locomotive. Extensive trials have already been made with the gas-turbine locomotive built by Brown Boveri.
Electrification of main lines has resulted in many experiments introducing various forms of electric locomotives, apparently with the object of finding a good total adhesion machine, and tests are being made of the influence on the track of various arrangements of drive and wheels, by means of blades, or with piezo-electric quartz as used by M. Mauzin in France.
Each of these three valuable lectures deserves very careful study, bearing in mind the different conditions, especially geographical, of the three territories considered, and we are indebted to the authors for their valuable contributions.
In all three lectures we find the importance of robustness, efficiency, and reliability stressed, in order that the availability may be as high as possible.
The advantage of the two-cylinder design of steam locomotive (for locomotives within the two-cylinder power range) as regards simplicity has been advocated.
The multi-cylinder locomotive is used when greater power is required, and when it is desired to increase axle load by the elimination of hammer blow.
Whilst the four-cylinder design was largely a French development, there is in France a tendency towards the three-cylinder type so widely used in British practice.
The disadvantages of special types of locomotives are emphasized, the ideal being the engine which can be used most widely (i.e. on most of the lines and on most classes of trains) and which requires the minimum of preparation and attention in service.
A marked reduction in the types of locomotives required to cover all classes of train working is envisaged, with an expected reduction in operating costs.
The use of Diesel locomotives for shunting is extending, and may be expected to become general. Main-line Diesel locomotives are on order for trial purposes to obtain operating results under British conditions, with a view to determining the most suitable field in which they can be used.
Electric locomotives have demonstrated their reliability and great availability. They appear to provide the best alternative to steam power in coal-producing countries, certainly until such time as coal can be made available as a fuel for use in the cylinders of compression-ignition engines or in gas turbines.

Volume 158 (1948)

Parker, R.C. and P.R. Marshall.
The Measurement of the temperature of sliding surfaces, with particular reference to railway brake blocks. 209-20. Disc.: 220-9 + 4 plates. 30 illus., 8 diagrs..
Authors employed by Ferodo Ltd. A pyrometer, which covered the temperature range 200 to 900°C. (390 to 1,650°F.) with a response time of 10-3 seconds, had been developed for studying the temperature of stationary and moving surfaces. The pyrometer had been used to determine the temperature of the surface of a railway tyre, during brake applications under various conditions, just as it emerged from beneath the brake block. Evidence is adduced to prove that the temperatures measured were not very different from those actually under the block. It is shown that the possible change in emissivity of the tyre steel is insufficient to invalidate the results. The variation and extent of the area of contact between brake block and tyre were investigated simultaneously with the temperature measurement. It was shown that high tyre surface temperatures (over 800°C) are an inevitable result of “strip braking”. The preferred formation of heat spots between the spokes of the wheel, and the unequal growth of heat spots across and along the tread due to tyre distortion, were examined. An attempt to eliminate strip braking by reducing the length of the standard brake block to three-quarters, one-half, and one-quarter of its original length was made. It was found that reducing the brake block to one-quarter of its original length completely inhibited heat spot formation, reduced by a factor of 2 the standard deviation of the stopping times under standard conditions, and reduced the maximum temperature attained during brake application from over 800°C. to under 400°C.. The deleterious formation of martensite was thus eliminated. It was found that the wear of the half-length brake blocks, expressed on a thickness basis, was not twice but maybe equd to that of a full-length block. Since the cost of a brake block is somewhat proportional to its size it follows that a shorter block may be more economical than a full-length block. It is concluded that the optimum size of a railway brake block must be determined separately for each material and set of operating conditions

Davies, R.D. and Cook, A.F. The motion of a railway axle. 426-34. Disc.: 434-6. 2 illus., 14 diagrs.
Investigation of the motion of a single axle and pair of wheels by means of a model, paying particular attention to the action of the flanges and to their influence on the development of “hunting”. In justification of the use of a single axle, evidence was produced that, owing to the various clearances, the axles of a bogie are in fact largely free to follow their individual paths. The effects of unsprung mass, wear of tyres and rails, flange clearance, lateral stiffness of the track, and other variables were studied. It is shown from considerations of dynamical similarity that the speed at which oscillation of a given violence occurs varies approximately inversely as the square root of the unsprung mass, and that a reduction of this mass would therefore be beneficial. Possible changes in tyre profile were discussed

Andrews, H.I.
The mobile testing plant of the London, Midland and Scottish Railway. 450-63. Disc. 463-76 + 6 plates. 22 illus., 14 diagrs., 2 plans, table. Bibliog.
A dynamometer car plus a set of coaches which were equipped with electric generators. These generators acted as a variable 'load" for the locomotive being tested. Discussion: D.R. Carling (466) considered that the importance of properly conducted tests, carried out under controlled conditions, could not be over-emphasized, if full and accurate data were to be obtained, on which to base changes in the design and utilization of locomotives. Tests carried out under normal running conditions revealed what a locomotive could do and how locomotives compared with respect to a particular piece of work, but it was a laborious and often fruitless task to interpret the results thus obtained in a manner of real use to the designer, as it was usually impossible to isolate the many factors involved. The maximum of information would be obtained by a combination of tests at a testing station and on the line. Commented on counter pressure testing conducted on LNER. Stanier (466-7) noted that he had been privileged to help test locomotives with Gooch’s dynamometer car, and that had interested him in the testing of engines. Mr. Diamond had referred to the leader in The Engineer in 1898. At that time one felt that the results were rather unsatisfactory. One never knew quite what one was measuring and one never knew that the answers were quite what one wanted. However, those results had given an indication as to what improvement, if any, had been made, and certainly they had helped the Great Western Railway to realize that to get the steam into the cylinder was quite as important as to get it out ! In that coxmexion, Churchward, before designing the long-stroke valve gear, had made a single-cylinder unit and had it carefully indicated as a stationary engine. That was why the Great Western had adopted, in 1900, the long-stroke valve gear, and led the country at that time.

Volume 160 (1949)

Bulleid, H.A.V.
Cinematography in engineering. 185-90. Disc.: 191-5.
Although this paper is related to the production of nylon textile fibres it is of interest (1) because the author was the son of the Bulleid and a significant author of railway books, and (2) it cites two key papers which link high speed cinematography with studies of the ride performance of rolling stock.

Ljungström, Fredrik
The development of the Ljungström steam turbine and air preheater. 211-23.
Inheritance and family surroundings favoured the growth of inventive ability in my late brother Birger, the next to the youngest of the eight children in the family. My father was a land surveyor, whose professional activities included the design and improvement of the various geodetic instruments he used. Amongst his innovations was a field-tube for measuring distances, which by this instrument were automatically projected down to the horizontal surface, when hilly country was surveyed. For these and other instruments, he was awarded a gold medal at the World‘s Fair in Philadelphia in 1876. The instruments were manufactured by the family in a small home workshop, so that his sons, from early childhood, had the advantage of a thorough practical training in the art of instrument Thus, when my brother, who was serving as an engineer in Aktiebolaget Separator, Stockholm, began in 1906 the work which later resulted in the design of the Ljungstrom steam turbine, he had the advantage of being an instrument-making artisan. Some of the details in the Ljungstrom turbine bear wimess to this early training. But the most important impulse was given to him by the inspiring example of the Parsons turbine, which he was able to study at the Durham College of Science. When the Ljungstrom turbine 1910 appeared in a unit, combined with two brakes for testing purposes, and gave evidence of high efficiency, various experts explained that this type of turbine was unsuitable for practical purposes: it had a frail structure, and was more like a piece of clock-work than a turbine. The inventor explained, however, that he would be very glad if the turbine ran continuously like a clock. Significant for this phase in the development