Edward Thompson: Wartime C.M.E. Discussion

What I will do is to post the written report in its enteriety without edits this evening and you can make up your own mind accordingly. However I baulked initially at Cox's stats and then on closer examination found myself astonished. I always maintained that I felt the conjugated valve gear problem was blown up out of all proportion too - the report makes me question my own view and that of others more markedly. However, I would encourage you to read it and make your own mind up.

 

Well no. We now know that the problem, at least for the headline failures of the A4s, was the combination [no pun intended!] of the conjugated gear with inadequate big end design, especially in lubrication. And also that neither Gresley nor Thompson sorted it out. In his book Cook says that there were big end failures on Thompson and Peppercorn Pacifics too, so the design fault didn't go away with the third set of valve gear. That may be an interesting observation to put alongside the Cox paper: if the criticism is mainly around big end failures then, according to Cook at least, it seems conjugation was not the primary cause, nor a third set of valve gear a complete solution.

 

And they didn't do so bad in LNER and post war days. Look at 2509's performance on the Silver Jubilee. Virtually ex works and part from problems with the brick arch (rectified over night) yet it ran the service single handed for two weeks before 2510 became available. And here's a good piece of info "The A4s were also mechanically reliable. On only 10 occasions out of a total of 1952 Silver Jubilee services, did an A4 experience a mechanical problem." Hsrdly evidence of a valve gear "not fit for purpose." Would be interesting to compare these figures with those for the LMS front line locos of the same era.

 

On the subject of what the report says here are the bits that I put in my earlier post (#542) - you can read the whole (brief) report on pp152 - 155 of the Oakwood Press biography of Stanier (OL114)

Cox / Stanier conclude (inter alia) that "a good case can be made for not perpetuating it in any future design". Then after commenting on the need for frequent overhaul he says / they say "it is a matter of consideration, therefore, as to whether certain of the classes should not be fitted with an independent inside valve gear".

Now bearing in mind that folks were much more polite in the olden days I interpret that as stop building engines with conjugated valve gear and give serious consideration to replacing it on those engines that already have it. I take the point that the LNER had a great many engines with and they ran for a long time so it was feasible to use the gear but was it economic? Bluntly was the cost of frequent overhaul worth it?

Many organisations develop their own way of doing things and persist with them even where evidence suggests the same level of service can be achieved more cheaply. I've seen this in the businesses that I've worked for myself. I've also worked with businesses that had "distinctive" approaches. After a while even the contemplation of changing something seen as fundamental to the way a business runs becomes almost unthinkable, anyone who suggests that change runs the risk of being ostracised or viewed as simply not understanding what they are talking about.

Gresley had served the LNER (and GN before it) for so long that the railway had his stamp all over it and the 3 Cylinder Pacifics were the flagship of that line. So anyone contemplating changing the way that things were done faced a real challenge. I think Thompson saw this and realised that his best chance was to draw on the views of the only other CME of comparable status to SNG. Stanier's comments were as polite and respectful if SNG as they could be without actually saying junk it.

Incidentally I don't think SNG's reputation would be affected one iota if, say in twenty years time, the Conjugated Valve Gear was seen as inefficient or uneconomic. On any rational analysis his status as a titan of engine design (and much more besides) is assured. Thompson's reputation should also be based on a rational analysis, not the vilification to which he has been subjected.

 

Can you ever eradicate the possibility of failure of a big end? The Peppercorn A1 had the best reliability record of all of the ECML Pacifics in BR days. Three separate sets of valve gear had much to do with this. Isolated incidents of big end failures on Thompson and Peppercorn Pacifics is not the same as the trends recorded with Gresley's engines.

The middle big end was am issue, and it was not just regarding the express passenger locomotives either.

The paper makes it quite clear that both the big end strap design and the choice of conjugated valve gear are the overall problem.

But that is pre-war conditions yet again - nobody here is disputing that the conjugated valve gear when it is able to be looked after in the correct manner, with high levels of attention afforded it, with the specific and expensive oils that were developed for use with the A4 Pacifics - that it is perfectly fine in theory and worked.

That is not in dispute at all.

During the war and after it, conditions of maintenance changed. There were less people on the LNER by nearly half during the war in the cleaning and maintenance divisions - due to men going off to war. Women took up a number of these positions and did their jobs admirably in very difficult conditions. With decreased maintenance and a complete change to traffic demands - look at the length of trains, for instance - the conjugated valve gear problem was appearing more markedly and it was not just on the A4 Pacifics - Cox states this across ALL of the Gresley conjugated valve gear locomotives in his report.

 

Of course, one could interpret it at face value - or even as "well there's nothing wrong with it that a good grease gun won't fix, but Thompson's in the mire if we just say that"

As to cost of frequent overhaul - whilst the grease guns were applied as required, the need for frequent overhauls of the gear or due to the gear vanished. It was simply war time conditions that tripped the gear up. I don't know if there are figures available for the 50s and 60s detailing costs of repairs of or due to the conjugated gear as opposed to say the A1 separate gears - the A1 had other details which led to their getting better mileage between trips to works than the A3/A4

 

Not just pre war. The A3 and A4 Pacifics were in front line service right up to the end of steam. Peter Townend and his counterparts elsewhere would not have turned them out of the ECML's prestige expresses if there was a fair chance of them sitting down on the job. To judge the valve gear solely by wartime conditions, and a war that the designer couldn't have foreseen, is disingenuous.

 

I'm not being disingenuous and I resent that accusation. You are doing an awful job of twisting what I am saying - to be clear I am speaking, specifically, about the conditions found at the time the report was made. Putting into context the report and why it was written at the time. Cox could not have forseen things post war and neither could Thompson. There are other things in the report which add to the argument to drop the conjugated valve gear, mind - not just the wartime conditions but other considerations as well.

Peter Townend does make it abundantly clear in East Coast Pacifics at Work that it was only with certain other conditions being met - including the use of the Zeiss Optical equipment - that led to increased reliability of the valve gear, and several changes to the strap design too. The A3 and A4 did admirably in the latter days of steam - that is not in dispute either. Decreased maintenance was still the norm post war and all of the LNER locomotives - A4s in particular - never had the same levels of service applied post war as they had pre-war. That surely cannot be in doubt?

 

Can you back that up with numbers? Cook does not much separate the three types. He only states that if worn the Gresley gear would put a greater percentage of the load on centre bearings. I don't think any of us would dispute that the Gresley gear was more prone to find out a marginal or inadequate big end bearing design, but surely the best fix for that is to sort out the big end if possible.

In order to justify putting the primary blame on the conjugated gear rather than big end design, and Cook implies that inside big end design was sub optimal on all LNER classes, what we need to see is :-

Firstly comparitive figures for all the classes in the post War but pre Cook period. Its a tricky comparison because there's a fairly narrow window where you can make a true comparison. No point in comparing the reliability of brand new Peppercorns with twenty year old Gresleys, and no point in including the war time period when all maintenance was sub standard.

Secondly comparitive figures for all the classes in the post Cook era once they had a satisfactory big end design, and excluding all the failures where locos were yet to be fitted with the improved big end.

I think only by doing that sort of comparison can we tell whether conjugation was fundamentally unsound, or whether the real problem was big end design. I cannot see how Cox and Stanier could have distinguished between the two possible root causes, so their data is only of limited value for you - and of course for Thompson. It sounds like a huge job to try and gather that sort of data, if indeed it still exists, but I think its what would be needed to make a fair evaluation of the concept.

Its the age old engineering problem that you don't really know what the problem is until you've fixed it. Hindsight does seem to tell us that effort would be - as indeed under Cook it was - better spent on improving the big ends than rebuilding the Gresley locomotives with an inside set of valve gear. I *think* I'm right in saying that Thompson never did instigate a mass program of rebuilding Gresley locos with inside valve gear (as opposed to 3cl-> 2cyl). If so does that suggest that he came to the same conclusion after his prototype rebuild?

It would be pretty surprising if the Peppercorns were not the most reliable of the Pacifics: they should, after all have included everything learned from their predecessors. However establishing how much of that increased reliability was down to the basic valve gear design as opposed to the detailed engineering implementation is another matter. And we shouldn't forget that when new the conjugation might even have delivered better valve timing than the 3 sets, because the inside set must be different in detail dimensions. Have you read Don Ashton's paper on the Peppercorn valve gear? http://www.donashton.co.uk/html/peppercorn_a1.html He notes that even the A1 put an increased loading on the middle cylinder at speed, though I'm sure far far less than worn conjugation gear.

I think that's an exaggeration. I believe that *every* time an A4 topped 110 before Cook it ran the middle big end. That does indicate a consistent problem, no matter what the root cause.

 

Hmm, the trials of Silver Link on 27/09/1935 involved 112.5mph twice and 108.7 for 10.6 miles. I don't believe the big end failed then. True, Silver Fox failed after 113 mph down Stoke Bank, just making it to KingsX sounding like a steam hammer, though I recall reading it was more likely the driver shutting off sharply for curves which caused problems by removing the steam cushion at the end of the piston strokes. That whole record attempt was mess, the driver not realising he was making an attempt untill the top of Stoke bank and then thrashing the loco far harder than Mallard had to be thrashed and going too close to the curves. Mallard also had a hot big end after 126mph, though it didn't fail catastrophically. I think the only recorded figure over 110 after cook was SNG at 112? mph down Stoke? which didn't cause any failure at all. Since the conjugated gear was unchanged from Silver Link's debut until the end of steam it actually suggests the biggest problem was the big end. (somewhere I recall reading the conjugated gear bearings were changed to plain oiled bearing late in the war so the drivers could oil them each day, but that might have just been a suggestion. I think they all went back to greased rollers soon after.)

edit for date error

 

i also look forward to reading the full report this evening.

i would also like to endorse Jimc's comments about the middle big end. this wasnt solved until K J Cook replaced the Gresley design with the GWR design. which begs the question why Thompson didnt replace the Gresley design of middle big end when he had ample evidence of big end failures in service plus the Stanier/Cox report spelling it all out!

we know for a fact that pre-war A4s with conjugated gear in excellent condition suffered from middle big end failure.

cheers,
julian

 

Presumably the LNER drawing office thought that the existing design was the best that could be done as it seems there were still problems on the Thompson and Peppercorn Pacifics.

Cook tells how Collett instituted a big reworking of big end design and lubrication after signs of heating (but no major failures) were observed on the Kings after a couple of years service, which resulted in some significant changes in bearing design and the introduction of felt pad lubrication. This seems to have been the sort of thing Collett was very good at: unspectacular, invisible and vital. Cook says (Swindon |Steam p102-3) that he didn't replace the Gresley design with a GWR one (apart from a few experimental ones) but adapted the existing marine style design to include the desired features from GWR practice.

 

The big end strap design was changed in 1947 with a mod in 1950; it was the bearing itself which Cook changed from whitemetal pockets in the brasses to a thin continuous whitemetal shell inside the brass (East CoastPacifics at Work). I take it that was the GWR standard you mean.

Just looking at my copy, it wasn't the valve gear that had increased reliability due to the use of Zeiss Optical alignment and strap changes, it was the locos themselves. The valve gear wasn't changed at all. I quote "The effect of lack of maintenance of the conjugated valve gear and consequent overtravel of the middle valve has been given as a popular reason for the middle end bearing overheating, but it was apparent that certain weaknesses lay in the middle big end itself and this is where the changes were made which basically overcame the problem not only on the Gresley engines but also on the other Pacifics which incurred similar failures. The only alteration made to the conjugated gear in the postwar years was to give more protection to prevent smokebox ashes being shovelled through the nearest hole onto the centre bearing. "

As an aside on P131 PT states that the introduction of the zeiss gear actually increased the hot axleboxes for a while.

 

Further to the reply from Lplus, the A4s were never designed to meet a 110+ requirement. The max for the streamliners was 90. The fact that they exceeded this on a number of occasions is a matter or record of course but it was not the requirement for which they were designed.

 

Incorrect, as explained elsewhere by Lplus. As for the A4s themselves, remember that they weren't all about the high speed streamliners, they also worked the day to day "ordinary" expresses too. After the introduction in 1958 of the EE Type 4s on intensive diagrams, A4s were often called upon to work these and managed to do so successfully, thus being used more intensively than anything pre war.

 

Happy enough with Lplus' explanation and I thank him for it. Does it really need to be a points scoring exercise with you, however?

 

I was working from memory but am happy to be corrected - always best to be corrected. Thank you.

 

Physician heal thyself.

As for the A4s themselves, remember that they weren't all about the high speed streamliners, they also worked the day to day "ordinary" expresses too. After the introduction in 1958 of the EE Type 4s on intensive diagrams, A4s were often called upon to work these and managed to do so successfully, thus being used more intensively than anything pre war.

 

That thought occurred to me, having been a passenger on the December 5th run. A seventy plus year old loco breezing 90+mph, and seemingly well capable of three figures if permission had been granted. NEWSFLASH... machines tend to work better when they are properly maintained.

I can understand that with manpower shortages and punishing wartime schedules, it is likely that maintenance would suffer. But with more time and a bit of care and attention, a run around with the grease gun and occasional change of bushes does not sound that challenging? Also, how easy is maintaining a full inside set of valve gear in comparison? Those are genuine question, by the way, for those with knowledge and experience of these things.

Dave

 

But - and this is the point I have been desperately trying to make for about 32 pages now - people condemn Thompson for making the decision to change to three sets of independent valve gear on the basis that the valve gear worked in the specific situations of having the maintenance it required.

He had no idea how long hostilities were going to go on, and if the report (which I am putting up shortly - just some typing left to correct) is put into the full context of the time, then it is clear the current situation the LNER was in when Stanier and Cox published the report was starting to become a grave one indeed, particularly where maintenance and manpower were concerned. The statistics for failures of the middle big end are astonishing.

Read what the report says - it is incoming - and think about the time period it was published at and the experiences the LNER and Thompson had had, and were to come.