Edward Thompson: Wartime C.M.E. Discussion

Cox's book "World Steam in the Twentieth Century" gives a further statement of his general views on conjugated valve gears. He discusses the Prussian G12 (later Reichsbahn BR58) 3-cylinder 2-10-0, which had the Henschel version of conjugated gear:

"The inside valve on all of these engines was driven by a system of cross-shafting, deriving its initial motion from the two outside Walschaert gears, and arranged with vertical arms disposed in two to one proportion, which was a different way of arriving at the same effect that Gresley (or rather Holcroft) achieved subsequently in England by a system of levers working horizontally. It can be said at this point that this conjugated gear did not prove ultimately satisfactory, the curse of all such motions being that small wear or distortion at individual pins was capable of producing an altogether disproportionate error in events at the inside valve itself. Its use was not therefore continued for any new design after 1918."

I interpret the term "curse of all such motions" to mean that Cox saw all types of conjugated valve gear to be defective in principle. That does not of course mean that a loco design with conjugated gear will be unsuccessful, since all designs are compromises with both good and bad features. The G12 itself continued in production until 1924, with 1478 engines built. Most served for normal life-spans and retained conjugated gear until withdrawal, although some were rebuilt in East Germany in the 1950s with 3 independent valve gears, the inside gear being driven from an extra outside crank.

 

The thing that I have noticed about this topic is that following a burst of activity it goes into hibernation for a few weeks before another eruption begins.....!

 

At the end of January, the WSR thread went silent for over a week. There has since been a super-eruption at the wayside halt of Washford!

 

To give time for correspondents to find more support for their pet theories perchance ?

 

You're suggesting that the contributors *research*? Have they no idea how the internet works? The idea is to blast away as loudly as possible, preferably in an echo chamber of the like minded, but failing that to aggressively flame with personal attacks for seasoning. You don't do research...

 

Although of course the vast majority of the locos with conjugated valve gear ran quite happily until the 60's. So some exploration of the phrase 'incapable of being made in to a sound mechanical job' is called for, to say the least. All mechanisms wear, the question is a financial as well as technical one - do the costs of a mechanism's maintenance exceed it's benefits? We have extensively explored that question on here as regards works costs - but how much time was saved every day by not having to lubricate and inspect an inside mechanism?

We will never know the complete answer.

 

But can infer a broad answer from things like availability statistics. If I've read the thread right - and I'm sure @S.A.C. Martin will be on to clarify - then it seems that conjugation was effective (and cost effective?) in "normal" times, but relied on a level of maintenance that couldn't be sustained in "abnormal" times. That seems less a matter of financial cost, and more a matter of value - could the locomotive be relied upon to be available when needed?

 

Isn't the valid comparison between purely inside sets of Walschaerts grar and conjugated motions? I could well believe Walschaerts was more resiliant, but hardly completely immune to the reduced maintenence issues of the wartime era.

 

I'd suggest that the valid comparison is between different 3 cylinder designs, doing approximately equivalent work. And similarly, it's not whether alternatives were immune to wartime conditions, but how well they coped in those conditions.

 

That's simply not the case.

We could split conjugated valve gear into several eras on the LNER.

1914-1928 - original design, flawed and outperformed (demonstrably so) with the largest engines by GWR in exchange trials.
1929-1939 - long lap travel valves, non-stop services, high speed record
1939-1949 - wartime, low overall availability plus lower mileages
1950-1968 - Indian summer

I have posted the whole report before several times in this thread, but the cogent points I took from it were:

• wear in the pins
• overrun on the middle cylinder
• inadequate middle end bearing and associated big end

But what might appear to be a saving on time in the sheds doesn’t translate into an overall saving when you consider that swapping out conjugated valve gear in the LNER required replacement of both the inside set and the outside pair, as they were all treated as ONE set, unlike two cylinder and three cylinder engines with separate valve gears.

If we accept the LNER statistics as an accurate representation of their rolling stock availability then we have to also accept what is being shown.

And by and large, it’s showing that Gresley conjugated classes were less available for work and spending more time in works than all other classes.

Again, I refute this. The statistics I have collated and put into an incredibly extensive spreadsheet show the Gresley classes in WW2 that were fitted with conjugated valve gear, were less available for work, ran lower mileages, and were in works for longer than the equivalent two cylinder or three cylinder locomotives with two or three sets of valve gear.

 

Holcroft certainly laid out a 4cyl conjugated arrangement. Whether it was ever used in anger I'm not sure, but certainly some locos utilised 4 sets of Walschaerts, so (historic record keeping permitting) there's scope for dieect comparison twixt inside and outside motion doing precisely the same job.

 

Not only that, but the answers are complicated by other matters. On the GWR, for instance, preparing locomotives was a piece work task with a fixed rate (two rates I believe, large locomotives and small locomotives IIRC), so the financial saving from a reduced preparation time was zero. There is also a question as to whether, especially in wartime conditions, locomotive preparation would tend to be skimped under stress of work. I haven't analysed the supplied data, it would be interesting to see whether there was a statistically significant difference between the availability of inside and outside gear locomotives on the LNER.

 

Jim, I’m more than happy to share my data. It’s still being collated in the excel spreadsheet but you are welcome to look at my scans of the source document with my notes if you so wish.

 

Pretty much this.

 

The crucial point in the Cox report is that conjugation makes the effects of wear considerably worse. Therefore, presuming comparable maintenance and comparable rates of wear, conjugated valve gear will need parts to be replaced sooner, and the longer that is delayed the worse the effects become. That was a good enough reason for Thompson to abandon conjugation for future designs and for Peppercorn to follow the same policy, but not for wholesale rebuilding of the existing locomotive stock when resources were severely limited by the war. After the war, when wholesale rebuilding would have been possible, it was unnecessary because the maintenance regime had improved again. As with most engineering, there are pros and cons, and different engineers will choose different compromises. Gresley wasn't wrong, nor were Thompson and Peppercorn.

 

An interesting extension of this line of thought is that designers would accept - hopefully - that machines such as locomotives are individual hence the depots where they are based would look to individualise maintenance accordingly. Earlier I drew attention to Haymarket's amended maintenance regime for its Gresley locomotives with conjugated valve gear which begs the question of how much leeway the CME would grant to depots to undertake local variations to the recommended maintenance regime. Presuming that the CME was willing to allow such flexibility it is then interesting to note how much use depots made of that freedom and the consequence it had on locomotive availability.

Not having analysed Martin's tabulated availability figures I wonder if such assessments could be made for individual depots and - more relevant perhaps - how much flexibility did Gresley, Thompson and Peppercorn as CMEs give to depots to vary the recommended maintenance regimes. In this context IIRC I am also minded to note Townsend's adoption of the Kylchap double blast pipe and chimney on an unofficial basis until he had proof of its efficacy thus BR's acceptance of the wider fitment to the LNER Pacifics.

 

But what about staff costs saved by reduced preparation time? That's my point. Exchanging a 10 minute saving every day for a more frequent works visit is a valid equation - of course depending on the actual figures.

I have no doubt that Thompson and Cox reached the generally correct conclusion in wartime. I'm just objecting to the phrase I quoted below -

"So some exploration of the phrase 'incapable of being made in to a sound mechanical job' is called for, to say the least."

Which is accurate. In non-wartime conditions the trade-offs were clearly acceptable. Or BR would have ordered a mass withdrawal as they did with the J39's. It was mainly the A4's that were sent north for further work in Scotland when the ECML was 'Deltic'd' , not the A2/3's or A1's.

 

That’s where you need to look at the point made by @Jimc about piece work though. If there is a formal agreement that, say, 90 minutes is allowed to prepare a locomotive, then saving ten minutes doesn’t actually save the company any money at all, it just gives the prep crew time to a have a quick fag. Whereas improved availability by dint of less time in the works or under the attention of shed fitters has a direct impact on the bottom line by virtue of having less of the capital tied up non-productively out of use.

Tom

 

I don't know whether it could be called piecework as such, but on BR midland ,going from memory, it was 45 min to dispose or prepare any thing under a class 5 ,and 75 mins for anything above. As far as disposing was concerned if the driver had a reliable fireman with him and loco's came in thick and fast ,they would do one loco' each and go home when the required amount equalling 8 hours was done. Six hours were nothing unusual and away home.

 

A Top Shed driver I knew reckoned that the 2.00pm goods from KX to York was always given a clear road by signalmen. This was because that trip took nearly eight hours and on arrival at York, the crew had just enough time to dispose of the loco and get to the bar for a pint before closing time. If they were held up, the crew knew they would not be able to get a drink, so would take their time and slow down all North bound traffic on the ECML.
Did the pubs in York shut at 10.30 or 11.00 in the 1950's?