Edward Thompson: Wartime C.M.E. Discussion

I would suggest that by then (1945) the shortcomings of the swing link pony truck combined with the then permanent way conditions were becoming realised, even more so later with two derailments of V2s near Hatfield soon after. I gather that the L1s had a spring side control system based on the LMS 8F from new, and the V2s were modified to this system not long after.

 

Point of information - but over 5000 posts. Many thanks everyone for your support, interest, information and patience.

 

Thank you: that's one putative reason for the L1s' reputation ruled out.

 

In my upcoming book I go into great detail on the prototype no.9000, which was developed during and then built at the end of the second world war, and subsequently tested on the LNER for nearly a year. The comparative studies of this locomotive done are quite fascinating, and the types of traffic it was intended for are also interesting. The testing was varied and the results influenced heavily the decision to order more.

It is not unfair to say that L1 no.9000 was the most tested locomotive in LNER history, quite frankly. I can't find a comparable prototype that toured the LNER system quite so much, and did so much under testing conditions. Thompson and his design team really, really wanted to get it right first time, and on the conclusion of the tests you'd be forgiven for thinking they'd created something magnificent.

The Thompson L1's reputation has several minor myths at the heart of it. They are:

• Wheel diameter too small - no basis in fact
• Valve gear type not rigid enough (or conversely argued, not flexible!) - same type as on Peppercorn K1s - no basis in fact
• Pony truck - this is a new one on me - it's based on the Stanier 8F design and all of the Thompson designs with a pony truck incorporated side control of some kind by way of this new design

What is definitely true about the Thompson L1s:

• There were issues with detritus from leaking tanks getting into the axleboxes causing overheating - early welded tanks tended to suffer more. This was largely rectified early on in the 50s
• The axleboxes were later fitted with shims which cured some of the overheating issues too
• The cabs were described as draughty - but no.9000's original cab was praised - this appears to be modifications made post-Thompson
• Braking capacity - this has a ring of truth to it, and the L1s were not seen pulling much in the way of freight

Overall it feels more like stubbornness from some quarters that the Thompson L1's reputation has taken such a hit in railway history. The stats don't suggest a poor locomotive class, and at the end of the day a locomotive more of a struggle to a driver - but does the daily mileages and has the availability to match - is going to appeal to the accountants and directors more than a locomotive which is a love in from all operating it, but isn't achieving the same level of reliability or work done for the company.

 

A classic case of Fact v Fiction perchance ?

 

When you say the axleboxes were fitted with shims, do you mean adjustable wedges (as on p19 of Part 9A, later replaced in some instances by manganese steel liners)?

 

I think the issue with the Maunsell "Rivers" was a bit more nuanced. "Track-sensitive design" was one description. In other words, they were dangerous on the poor-quality SE&CR track (still with sections of 45 foot rail on shingle ballast in 1928), but to put it all down to the track is not the conclusion that Herbert Walker at least came to: he required a programme of both track upgrades and loco rebuilds. The rebuilds concentrated on the primary cause in somewhat drastic fashion: surging water in tanks = just remove the tanks (by rebuilding as tender engines). It didn't go further to look at the level of side control in the pony trucks, and perhaps in 1928 there was less theoretical understanding of such issues than was the case 15 years later. My memory of working on our U class (essentially the River class rebuilds) was that they are still somewhat lively at the front end, so I suspect more side control on the pony truck could have been desirable, though for whatever reason it wasn't done and I'm not aware that there were any subsequent derailments of the class even though they worked at times up to about 70-80mph.

FWIW, the Billinton L class 4-6-4T also had problems with surging water and had to have the side tank capacity cut down and replaced with well tanks, which proved a maintenance nightmare; later they were also rebuilt as tender engines.

Finally, on track condition: the GN mainline north of London was used for testing the K class 2-6-4Ts after the Sevenoaks disaster, and the ability to run smoothly at speed was a credit to the condition of the line at that point in time. Whether it was still in such good condition during and immediately after the Second World War is another question.

Tom

 

Yes.

 

On the track issue, the same thing happened with the Hastings DEMUs, after the Hither Green accident in 1967. Their ride was poor on the Southern region, so one was tried out on the GN, as with the River class, and it rode safely at speed there

 

I also wonder if the Rivers made more sense as tender loco's on the expanded Southern system than as tank engines?

 

Possibly. Electrification took over what would have been some of their work. The Brighton 4-6-4 tanks were also converted to 4-6-0 tender engines, presumably for the same reason, Tom will no doubt know more.

 

That's also my understanding. Hopefully someone (hello Tom!) might be in a position to throw some light on why a class, well regarded in it's original guise, seemed so undwhelming once rebuilt. Perhaps it was just a case of their shortcomings being exposed to wider comparison?

 

Simon - If you haven't already got a copy of it, may I suggest that you get your hands on 'Bill Harvey's 60 Years In Steam', written by him, ISBN 0-7153-8712-X, published in1986. Interesting in showing the 'fun' he had in keeping a fleet of L1s running at Neasden around 1950.

I used to live near Hatfield so regularly saw the L1s in action in the 1950s, mostly shedded at Hitchin for the outer suburban and Cambridge trains. For the clattering noises that came from them they equalled the WD 2-8-0s in noise, no wonder they were nicknamed 'Concrete Mixers'! Peter Townsend in his writings wasn't over impressed with them either.

One interesting job they were used on was the 5.52 (?) p.m. Kings Cross to Cambridge and Peterborough, usually double-headed with a B1, although 'Black 5" 44911 ex 6J was seen at times. This train divided at Hitchin, L1 to Cambridge and the B1 to Peterborough.

 

I heard that Neasden preferred the ex GC A5 tanks to the L1s, even though they were much older

 

Bill Harvey mentions this in his book.

 

To help with your appraisals of loco designs, you may want to get hold of Adrian Tester's books "A defence of the Midland/LMS Class 4 0-6-0 - also why frames cracked and axleboxes ran hot" and "An introduction to large-lap valves and their use on the LMS". Tester has an engineering background and much of what he says is well argued albeit uncomfortable reading for those who are rather partial to Stanier's and his team's work. At times, the writing comes across as a bit of a polemic, and Tester is rather eclectic in his use of other authors' work to support a particular argument, but nevertheless highly recommended. You can order them from the ASTT site.

 

I do have a copy, and it is referenced in my book. Note the dates. I have said it a great many times - Neasden's "issues" with the L1s are actually:

• limited in timeframe
• limited to a small number of the 100 strong class

We are talking about locomotives that were 6 months to a year old in 1950, for full context. Does it strike you as odd that only brand new locomotives are apparently so bad? What a coincidence - a Thompson designed locomotive.

Bill Harvey is spot on in many of his writings, but his writings on the Thompson L1s doesn't match up with other records. We have to be clear in this - the biggest underlying issue in the Thompson story is the number of issues that are exaggerated, confabulated, or actually don't exist. The number of times railwaymen have written passionately on their favourite and least favourite locomotive designs is many: and there is so much taken at face value that shouldn't be.

The Thompson L1 is a perfect example of this.

As for the sound of the thing - they received the nickname "concrete mixers" and they probably did clank when run down. There's enough video to prove that. The reliability and mileages of the class suggest they were never so poor as to be withdrawn for maintenance, so where do the issues lie? Is this all just because the designer was Thompson?

Wearying to repeat myself, but we have to be clear and use primary evidence/common sense/objective viewpoints when analysing this. If Bill Harvey's recollections don't match the records: then we have to ask why?

 

That may be so, Simon, but the problems with the L1s persisted, certainly still giving trouble in 1955 with these locos now being at least five years old and no doubt having had major overhauls during that time. This 1955 period is covered by E. S. Beavor in his book'Steam was my calling', ISBN 0 7110 0491 9.

Both being Shedmasters, I just find it difficult to believe that either of these two gentlemen would 'make things up' unnecessarily.

 

Why were there no such similar problems with the LMS and BR standard ones?

 

Yes, that's basically it. Electrification to Brighton took away the work they were designed for, after which they went to Eastbourne and were well-regarded. After Eastbourne electrification in 1935 there was no further Central Section work that was also within range.

At that point they were rebuilt as 4-6-0s and sent to the Western Section.

On the Western section, they were classified N15X and initially put in the same link at Nine Elms as the King Arthurs (and Lord Nelsons). Which they clearly weren't the equivalent of. (Lots of ways you could quantify that, but a simple number - nominally an N15X was what became BR class 4P; a King Arthur was 5P and a Lord Nelson 7P. Things like grate area, heating surface and so on would flesh out the detail). The result was a poor reputation, but it stemmed basically from what was rostered to be a King Arthur equivalent clearly not being so.

Bulleid did a few Bulleidisms sorting out some issues of draughting etc., which made some enhancement, but fundamentally you weren't going to turn what was very largely a late-Edwardian class 4 into an early-modern class 5. They later went to Basingstoke and worked on semi-fasts and van trains in a link that included various Maunsell U class moguls, to which they were more comparable, and acquitted themselves well. Ultimately though, in comparison with later Maunsell locos, they had the Brighton disease of being more expensive both to run and maintain than comparable locos, so they went comparatively early in the 1950s.

(One interesting snippet is that during the war they were loaned for a time to the GWR; that railway classified them as equivalent to a Saint, but below a Hall - which feels about right).

Tom