Adrian P. Tester


A defence of the Midland/LMS Class 4 0-6-0 - also why frames cracked and axleboxes ran hot. Aberystwyth: Crimson Lake, 2011. 274 pp.
Post grouping the 4F 0-6-0 tender locomotives were generally regarded by non Midland Railway crew and locomotive cognoscenti as pretty feeble, especially compared with some of the similar LNWR and Scottish versions, which they ran alongside. They also tended to run hot, and their frames crack. The Author offers a considerable defence of what was, by any reckoning, a large class, and of the modifications to them. The book is highly critical of much of what Cox wrote about locomotive development on the LMS and on British Railways. The book covers much more than the title suggests and there are refernces to work on frame cracking, and its causes, on many classes of locomotive, not all of them on the LMS. The book alsoo provides an in-depth examination of frame cracking which includes insights into other work on other railways: Steamindex will be gradually modified to reflect some of these observations. The concluding summary is shown below:..
Whilst the 'Big Goods' was not one of the finest locomotive classes to grace the rails of Britain it was however very far from being the poorest, although to read some commentators' views one could be mistaken for believing it to have been the worst designed locomotive ever and that it should never have been built. Had this really been the case, we must wonder quite how powerless and/or incompetent were the first three Chief Mechanical Engineers of the LMS who allowed the class to be multiplied, unmodified until it represented 10 per cent of the company's locomotive stock – another view is that its performance was not as black as some would have us believe. Indeed it seems far more reasonable to believe those engineers in the Operating Department who requested more Class 4s and those in the Chief Mechanical Engineer's department who agreed to their production made that decision because they thought the locomotives would well serve the new company, after having determined the class' performance from comparison tests.

Other critics of the 'Big Goods' take the view because other railways had introduced 0-8-0s or even 2-8-0s by the time of its appearance, Derby should not have been so 'timorous' in its response. Contrary to the opinions of some, the purpose of engine designers was not to keep up with the 'locomotive Jones' although that certainly occurred on occasions, but rather to produce machines which complied with the restraints of the Engineer while meeting the traffic and route requirements of the Operating Department.

No doubt, had the need been there, or perhaps the LMS been less awash with money for its scrap and build policy, some improvement might have occurred, but instead, it seems later engineers preferred to criticise the machine, thereby providing themselves with the excuse for building a completely new replacement. Perhaps this criticism of the Class 4 et al would have been more fairly justified if the locomotives built to replace them had been shining paragons of the craft of the locomotive engineer – sadly, they were not. Furthermore, the development of the successor to the Class 4 0-6-0 Freight was a long drawn out affair with a gestation period of ten years. From this we might observe that for all its faults, it was not until 1947 that LMS locomotive designers had proved capable of producing a replacement engine which possessed sufficient power, lightness and route availability to satisfy the operating people and the Engineer, to become the nominal successor to a design introduced thirty-six years earlier. The following summarizes the main criticisms of the Midland/'LMS Class 4 0-6-0; most it will be observed individually are comparatively minor, and could have been addressed quite simply, had the will been present:-

Frame cracking: Although the engine had a tendency to crack its frame, so did many, more modern classes, many of which could develop cracks more extensively and at a faster rate. Furthermore, in the case of some, particularly the pure LMS designs and derivatives, this was after considerable study into its causes, yet, applying the lessons had not produced a frame immune from cracking. This was understandable, since steam locomotive engineers did not have access to the necessary design tools to prevent it from occurring. In view of this, it does therefore appear rather unfair for LMS and BR engineers, and others, to criticize a design made two, three or even four decades earlier when the knowledge regarding the causes of frame cracking was even less thorough. This is particularly so when their own efforts were no better and in some cases considerably worse. On the last day of 1947 the LMS handed over to the embryonic British Railways 7,805 steam engines, of these 772 comprised Standard Class 4 0-6-0s, yet the frame of these engines remained the 1911 design. Contrast this with the then 742 extant members of the Class 5, whose frame design had undergone considerable revision and modification. Indeed it was the predilection for frame cracking exhibited by so many of William Stanier's newly introduced classes that initiated the major investigation into the causes and the extent of the problem.

Axleboxes: While these were small by later standards i.e. when compared to those that could be fitted to outside cylinder engines, this was not the real reason for their poorer performance. The evidence is there demonstrating that engines fitted with similarly, or even more heavily loaded bearings, running on other companies, fared better. Cox demonstrated that in 1930, one important non-standard LMS class of inside-cylinder goods engine had returned a hot-box performance – one every ten years – that was identical to that obtained from the Stanier-box fitted to the taper boiler classes. This achievement demonstrates the error in assuming it was bearing size alone that automatically determined a propensity for hot boxes. The statistics also suggest that hot driving wheel axleboxes was a problem that became more troublesome under the LMS but it was one largely of its own making. Changing the white metal, plus poor re-metalling techniques in sheds, together with a less than optimum choice of lubricating oils all served to worsen the situation on the LMS. Later, improved finishing techniques coupled with an improved oil certainly increased their life as well as reducing the incidence of hot boxes. Thus, by the time of British Railways incidences of hot boxes had fallen right away becoming much rarer – yet the bearings themselves were not increased in size nor at the time had the work demanded of the class significantly reduced.

As the tables appearing earlier reveal 4F class members were returning annual mileages typically 85-90 per cent of those obtained from Standard Class 8 2-8-0s - the other principal ex-LMS goods locomotive class. This does not seem a bad effort, similarly, this improved axlebox performance also ensured less axlebox wear and deterioration in service. Consequently workshop attention that had been demanded typically at 30,000 miles during the 'twenties and early 'thirties had by BR days been roughly doubled. As figure 3 reveals, although as to be expected the mileages between repairs were higher for 8F 2-8-0s compared to the 4Fs, nevertheless 40 per cent of the eight-coupled engines ran 60,000 miles or less, before entering the works for an Intermediate following a General while 54 per cent of the six-coupled engines ran 50,000-60,000 miles or more. By the 1950s, having corrected the mistakes introduced earlier and combined them with improved workshop techniques ensured that these slightly shorter mileages between workshop visits were simply because the bearings wore out quicker due to their smaller size. They were not due to a reduced technical performance.

Size for size Class 4 axlebox performance was comparable proportionately to that obtained from the 8F. Indeed, we ought really point out that high mileages were obtained from LMS and BR Standard locomotives fitted with Stanier-axleboxes not because the engines were fitted with wonderful bearings but rather it was because the engines concerned were not producing power outputs commensurate with their size. Thus, modern locomotives running elsewhere in the world and shod with designs of plain axlebox similar to, or even nominally better than the LMS Stanier-box, could require attention at mileages as low as 25,000-30,000 miles. On that basis, one could argue that LMS and BR engineers had not obtained economic use of capital, since in their desire to reduce maintenance costs, they had produced engines that were were larger and heavier than they need otherwise have been, thereby increasing their fuel and water costs. Availability and productivity/utilization figures together with the annual mileages obtained infer the 'Big Goods' were reliable and cost effective, but due to their nominal replacement by new classes this meant that a scheme to provide them with roller bearings – the optimum solution – was not deemed worthwhile. Nevertheless, in the mid-1950s a handful of examples were provided with manganese steel liners to their axleboxes, which confirms bearing mileages had improved sufficiently to give the experiment some credence, but this modification would have been better combined with roller bearings.

Steam utilization: Despite the carping, there was nothing wrong with the Class 4 valve events indeed Ivatt subsequently altered the LMS three-cylinder valve events so that they more closely resembled them! Not only did he reduce their lap length while increasing the lead, in order to increase the port opening but also he returned more or less, to the Midland Railway's favoured lap:lead ratio of four. While in the case of his Class 5 4-6-0 built with outside Stephenson's valve gear the valve events were more or less identical with those of the Class 4, differing only in that the port openings to exhaust were proportionately smaller. In all of these outside cylinder classes the port openings to steam were larger for equivalent cut-offs than those in the 'Big Goods' simply because of the longer lap.

The presence of short-lap valves compromised steam economy and performance for high speed traffic, but the class had been designed for goods and mineral traffic so although it was also found suitable for fitted freights, local speed passenger and excursions, unless it spent a sufficiently high proportion of its time on such higher speed traffic, there was probably little benefit to be gained from fitting large-lap valves. Whilst consideration appears to have been given to such a modification, it did not proceed. The ex-LMS locomotive designers whose stars shone post-war opined that the speed of fitted freight trains was such that the 0-6-0 was no longer suitable and that engines in possession of a leading pony truck or even a bogie were more appropriate, although the Southern Railway thought otherwise.

Steam production: The 'Big Goods' was introduced to haul the Midland Railway's immense coal traffic, under which circumstances, it would have generated plenty of hot steam. Essentially it was only the gain in enthalpy obtained from superheating that gave the engine its class 4 rating – the larger diameter cylinders were primarily to accommodate a sufficient mass of the hotter less dense steam. Smaller loads, frequent starts and stops also poorer quality coal would all have affected its performance for the worse. If that was going to be their lot, then Geisel's superheater booster, which comprised a loosely pivoted baffle in front of the small tubes might have been worth investigating. It represented a cheaper alternative to the optimum one of altering flue and small tube diameters and numbers.

The steaming difficulties could have been overcome by attention to the draughting/exhaust system. That such a large class of engines could be left with so reduced a steaming capacity in the face of poorer coal, says a great deal about the abilit!es and misplaced priorities of certain locomotive engineers. The draughting could readily and very cheaply been improved – certainly for far less money than the contemplated changes associated with increasing the lap length even if the latter had been restricted to when new cylinders were required.

Bad as draughting design could be on the LMS, which tended to be rather hit-or-miss witness the fiasco that accompanied the first few 'Doodlebugs', nevertheless, some improvement could have been effected if only perhaps by the simple expedient of adopting the draughting design of the parallel boiler 2-6-4T. Not only would an improved front-end have given it more of a reserve to cope with poorer coal, but also the resulting increased quantity of steam, would have helped offset its poorer steam economy at speed. There is a hint of this in the superior power-to-weight ratio of No. 44030 compared to No. 43094.

Modifications: The class was, post-1928, provided with four-ring piston valves in place of the original Schmidt single broad ring, which reduced the internal steam leakage. Exhaust steam injectors were tried but later abandoned – feed water heaters demonstrate their greatest saving in fuel when the locomotive is being worked consistently near its maximum output. In this respect following the introduction of newer, larger engines, it is likely exhaust steam injectors made less impact on Class 4 economy following Sir William's arrival, but by the same token they were probably no more effective on for example the Class 8 2-8-0. For regular use on high speed work a modest improvement could have been effected in the steam circuit, which it will be recalled was not noticeably smaller, in proportion to its cylinder diameter, .than that of two contemporary express locomotives. Nevertheless its weakest features could have been cheaply overcome – increasing the diameter of the superheated steam pipes, revising the porting in the piston valve liners, and possibly streamlining the valve heads. A useful further halfway house might simply have been a modest increase in valve lap by say 1/8ins or 3/16ins, which could have been accommodated by shortening the eccentric rods and substituting launch type expansion links. To increase the lap further would probably have necessitated a revised cylinder design.

Tragically, for the steam locomotive, particularly in the land of its birth, fitting large-lap valves and superheaters represented more or less the pinnacle of its technical development. Unfortunately, the poor understanding of locomotive performance demonstrated by some designers, or perhaps simply the determination to incorporate certain pet 'hobby-horses' into the design, was such that they were able to destroy much of benefit, which even these modest improvements could bestow. For example, ignoring whatever benefit fitting the replacements built in the 'thirties and later with carrying wheels and outside cylinders was meant to achieve through easier maintenance and better riding, one downside was they increased engine weight, while the locomotives themselves produced no more power.

Whilst a Class 4 0-6-0 needed more steam to produce each horsepower, at low speeds this was primarily because of its lower steam pressure and superheat conditions with very little due to its smaller steam circuit. However, it was a considerably lighter engine, thus once allowance is made for the horsepower required to drive the heavier Class 4MT 2-6-0 is considered, the advantage the latter's superior steam circuit presented was less than one might at first suppose. Along with this we must remember that some of the published test data for the 'Doodlebug' is suspect appearing to shew the locomotive in a better light. A conclusion that is confirmed to some extent by the comments made by several footplatemen to the effect the Mogul was no better than a good Standard Freight. Incidentally there was a precedence for this data manipulation in the case of at least one other BR Standard class. Michael Rutherford advised that the 'best' test results were used as the plotting points for the charts recording the Class 7 4-6-2 thereby giving the locomotive in its Performance and Efficiency Bulletin a most optimistic performance. At the time Mr Cox was both in charge of locomotive testing and responsible for the design ofthese classes. When we consider the poor return on capital investment that so many more modem classes actually represented, is it any surprise that 'despised' locomotive classes, such as the Midland/LMS Class 2 or the 'Big Goods' lasted so long? However challenging they might have been to their crews in comparison to more modem locomotives, from the viewpoint of the bean counters, there was no difference. The new modem replacement locomotives, due to inherent design defects, and/or their unsuitable power-to-weight ratio in respect of their duties, meant that frequently they were unable to demonstrate any real economical or financial advantage over these pre-grouping designs.

As Cuthbert Pounder', a director and Chief Technical Engineer of Harland & Wolff Ltd observed, albeit in connexion with marine engine building:-
"Engineering is, first and last, a commercial business. The similarity between building and selling an engine, and making and selling a pair of shoes, may be disguised and overlaid, but, at root, the two things are the same. The engineering works, with its complicated plant, its many trades and its technical officers has, in the last analysis, kinship with the shoemaker, who, taking his raw material, fashions it into a pair of shoes and sells them over the counter. Shoes are sold on suitability, durability, price and so on; so is a power plant. The end is precisely the same; it is only the means to that end which are different. All this may be obvious, but nevertheless it is hard to get technical men to appreciate - in their bones, so to speak - the commercial aspect of their work. In design offices, men – especially those whose cast of mind tends to the academical – are apt to become so absorbed in their problems as to mistake the means for the end. In the works, men with the super-organizing complex commit the same error. " Perhaps in time, once the steam locomotive has receded further into history, there will be more objective and considered appraisals made of its development under the auspices of the LMS derived from the Midland Railway's legacy - as opposed to the current simple acceptance by enthusiasts of the criticism prompted by others, possibly with personal axes to grind. The outcome of these, the author ventures to suggest, may come as a surprise to a great many people.

C C Pounder Human Problems in Marine Engineering Trans I Mar E March 1960.

An introduction to large-lap valves & their use on the LMS. Aberystwyth : Crimson Lake, 2008. 102 pp.
In the complex field of valves and valve gears, one of the less discussed areas is that of short or long-lap valves. In Britain, the Midland Railway’s engineers had used short-lap valves, and were regarded as incompetent by E.S. Cox for adhering to them. But Cox and others within the LMS/Crewe establishment were long-lap valve men and used them.

He also contributes to Backtrack including a series on locomotive testing

Steamindex home pagge