The LMS's pre-grouping express 4-6-0s - a question

A dotted line from the GWR Castle class to the LMS Jubilee, Rebuilt Royal Scott/Patriot classes to show the GWR boiler/firebox influence in the rebuilding will be good.

 

Interesting...

The Castle exchanges were very influential on the LMS and LNER as regards valve gear as well, and in the other direction the short and wide boiler on the County was LMS influenced. Grange and Manor cylinder castings had ideas from Chapelon. Trouble is before long you end up with a rats nest of arrows everywhere

 

Yes, I started getting carried away and had to stop...

 

I think that the situation just after the grouping on the LMS was rather less straightforward than sometimes caricatured.
In the mid-1920s, heavier trains and faster schedules meant more powerful motive power. All the principal main lines faced this challenge, and some had done better than others in meeting the needed.
So at grouping, the new officers (GM, operating, CME, etc.) faced the choice of going with a known good design from one of the constituents (possibly an improved version) or entirely new design. The various companies acted as follows:
GWR: Castle, improved Star, known quantity - but this was essentially continuity and no new challenges.
SR: Maunsell went initially for improved versions of adsorbed designs - N15, S15, N, etc. His new design (LN) was good but not without issues.
LNER: A new express design had just emerged anyway (A1), although some improvements were needed for it to become superlative. Existing designs replicated (A5, D11, B12, N7, etc.).
LMS: Particular problems in some areas (GSWR) and (modified) existing designs replicated to meet the demand, which they did very adequately. On WCML, Claughtons are struggling with inadequate boilers and valve gear, so improved boilers and valve gear fitted (unfortunately this exposes other problems, because LNWR policy was to design for a twenty year hard life with no residual resilience: "a short life but a merry one"). L&Y 4-6-0 (in their second superheated incarnation) are performing well on the home line and do well in comparative tests, so more built (unfortunately they don't suit the WCML duty, which is quite different). There is also the clash of operating cultures: Derby's many short trains versus Euston's fewer longer trains.
The idea still persists that Hughes and Fowler just sat passively by in a motive power crisis: in fact a new 4-6-2 design was started almost immediately on grouping, tests were conducted on the proposed compounding arrangement, and building actually started in 1926.

 

All absolutely true, Andrew.

George Hughes had a whole series of new designs planned to meet all requirements, but his early retirement largely put paid to them. Henry Fowler did take the Pacific project forward in modified form, so both these CMEs had recognised the power crisis facing the LMS and were moving towards a solution. Even James Anderson by the time of Fowler's promotion realised that Midland Compounds were far from the answer, but it must be assumed that the 'Midland Small Engine Policy' was still extant and a Pacific was a step too far for him. The result was the borrowing of a GWR 'Castle' (in which Fowler took no part) and the abandonment of the Pacific project in favour of the North British-designed 'Royal Scots', introduced in 1927. That this was a mistake is shown by the fact that less than five years later, William Stanier's top priority was the design of a new Pacific!

 

I have always fancied a model of the Hughes pacific, 2-8-2 or other proposed designs, or the Fowler compound pacific. Anyone built one?
Having said all that, I can't regret the tortuous processes of early LMS express motive power design, as it gave us the magnificent Royal Scots.

 

Well they bequeathed the four-ply trailing frame arrangement to the Stanier Pacifics. It would nice to see an example of 10456, the Hughes 4-6-0 converted to a compound, as Hughes poignantly puts it "the most modern but least known compound engine to run in this country" (although I think Gresley's W1 post-dated it). Incidentally, Cox's 1946 paper to the ILocoE is worth a read for anyone studying LMS loco history, and he touches on the Scottish contribution, noting that the G&SWR locos rapidly went to the scrap heap and that the Highland Railway locos were on the whole the best, well probably, and notes the only legacy from Scottish practice was the whistle. There are some useful tables in his paper, such as the attached, one of which shows that the 2P, which I think people were being rude about elsewhere, came out pretty well in comparison to its peers, and the other table compares performance of the Caley 14360 against a Prince of Wales, Midland Compound and Royal Scot.

 

Interesting, but statistics can be manipulated to give any outcome you want. Testing was largely in the hands of the Motive Power people led by James Anderson, ex-Midland. The route was often over ex-Midland metals with ex-Midland crews, and great trouble was taken to ensure that the Midland engine was in first class order.

As to the repair costs, and as I said on the above mentioned other thread, the Midland engines were afflicted with very poor steam passages and old fashioned valve events. It was impossible to work them hard, especially the 2Ps, so they did not suffer the wear and tear of similarly sized but harder working engines. Had they been worked harder, mechanical defects, particularly in the axleboxes, would have manifested themselves and maintenance costs would have risen, as happened with the 4Fs. LNWR engines were lightly built and worked very hard so high maintenance costs were to be expected. One expensive item was frame cracks, particularly emanating from the driving horn gaps. The LNWR had provided an additional centre frame and axlebox to distribute the stresses through a greater length of the frames. The afore-mentioned Running Department decided there was no need for it and had it removed...

The running tests are interesting but there was, as already indicated, some bias towards the Midland locos. But that the Compound achieved god coal and water consumptions is not too surprising, these are normal attributes of any compound engine. But these were made over the S&C with the engines working quite long distances on full regulator, exactly what the Compound was designed to do. But once they were displaced from front line duties to semi-fasts and locals, they were out of their comfort zone, had to be worked for greater distances on part regulator, and the basic economy - and performance - of the engines disappeared.

This is not saying that the engines from the other constituents were wonderful. By the time of the Grouping they were generally old and had survived very heavy use during World War I. But Anderson had a plan to run the LMS on Midland lines with Midland-derived locos, so it was important that these showed up well against the opposition. As history shows, he partly got his way, and then these engines failed to live up to expectations.

 

But am I correct in saying that the rebuilding of the Royal Scots turned them into pretty useful engines? Several videos of 46115 and 46100 in preservation posted on this forum have given me the impression that in the rebuilt form, they were/are highly competent performers. A main line run behind one of them is definitely on my wish list. So far my only encounter with a rebuilt "Scot" has been 46100's visit to the Mid Hant' 2016 autumn gala. The climb from Butts Junction to Medstead in particular whetted my appetite for more.

 

They were very useful engines before, but barely big enough on introduction in 1927 and were themselves outclassed by train weights by 1932. Rebuilding improved what was already a good machine, but by then one step down from the top. They were probably the best 4-6-0s to run in Britain (sits back and awaits the GWR contingent's response!).

Having said that, the LMS / LMR never had more than 51 class 8P (BR system) Pacifics, and these included the Turbo and 71000. Non-availability of a Pacific was a perennial problem, and a Scot would have to substitute, sometimes with 16 and even 17 bogies in tow*. This in turn led to a shortage of Class 7Ps, a gap filled by Class 6Ps (Baby Scots and Jubilees), which themselves were not over-plentiful (compare numbers with the impoverished LNER!). The Class 6 diagrams then had to be filled by Black Fives, and it was either John Powell or J.W. Rowledge who stated that these engines were the main cause of late running on the LMR. Excellent engines though they were, they were a mixed traffic loco and constant express work was asking too much of them.

*For this reason most daytime Pacific jobs fell within the Class 7 load limits, albeit right at the top end of them. Thus when the Stanier Pacifics were withdrawn and Britannias took over their remaining work, it was put about that these BR types could perform as well as a Big Lizzie. But only during daylight, they were never given the enormous overnight loads which the Stanier engines worked throughout their lives.

 

David Wardale suggests that a Standard 5 was more than capable of deputising for a (Bulleid?) Pacific and was easier on the fireman than the Bulleid when it was deputising

 

The L&YS handily have the Cox paper on their website: http://www.lyrs.org.uk/Papers-to-Institutions (scroll to bottom).

 

Cox' paper includes a useful figure showing how Compounds operated efficiently while on "top link" work, but how their performance deteriorated as they moved down into more secondary work. Cox believed this was due to needing trained and dedicated crews.

To be fair, I think it was less manipulative that that. Anderson was an operating officer. He had operated the MR at the peak of efficiency. He did not adapt well to a larger system having different types of operation, and stuck with what he knew - and it wasn't always appropriate for the larger system.

The 'mystery' of my the L&Y 4-6-0s weren't better - given their high degree of superheat, ample boilers, better-than-most valve events, etc. is given (in Cox' opinion - and he was a Horwich man) in the Author's answers to questions in the paper in my previous post. Essentially Cox blamed the piston rings, ball-releief valves (which leaked when they wore) and leakage into the smokebox vaccum. Which goes to show that "little things" can be the undoing of great projects...

 

Thanks, that is useful, as inter alia it also has Hughes' papers.

 

But really the Midland Compound was the zenith of the Victorian 4-4-0 (along with the GWR City) and should not have been in the trials or this thread. And the POfW 'Tishy' is the ancestor of Col.Beames' Black 5 maybe...

 

Tch, I expected netter than that!

 

Don't forget that Anderson had been Chief Draftsman for the Midland, a position with more power and responsibility than the name suggests. He had been Acting Chief Mechanical Engineer during Henry Fowler's secondment to Government service during World War I. He introduced much that was to be Midland 'Standard' practice, and was loathe to let these go.

The smokebox air leak problem was not easy of solution, as it wasn't on other classes, including the original Royal Scots. The problem with valve leakage, again paralleled with the Scots, was easy and was done, but too late.

 

The Midland compound was a very fine Class 4 passenger loco indeed. Given lack of long lap, long travel valve gear (which applied to everything except the most modern GWR locos) they were thermally and mechanically efficient, robust, efficient on coal and water, cheap to repair (even with 3 cylinders) - more than a match for anything in their class. It's just that Class 4 wasn't what was needed in the front line any more, it was (good) Class 5 and up. Hindsight-in-advance isn't available to real engineers...
Fowler had a planned Class 5 compound 4-6-0, which never got off the drawing board.
There was nothing wrong with the basic Smith-Johnson-Deeley design, it could have been modernised and been the basis of multiple cylinder locos to the end of steam, as elsewhere.

 

Don't forget that awful paget loco or the unfortunate Fury..