Sir Nigel Gresley - The L.N.E.R.’s First C.M.E.

To this non-engineer the question of unified v divided drive is irrelevant in the discussion of a locomotive engineer's record. In simple terms the CME is charged with providing a locomotive to undertake a specific range of services whether it be high speed expresses or heavy slow freights hence his success is directly linked as to how his designs meet the required objectives. Obviously there will the inclusion of running costs (including maintenance costs, failure rates, and time(s) in workshops) but at the end of the day the man comparison between locomotive engineers is how capable they proved in handling the tasks they were built for.
The Board will set the requirements and these may include such things as boiler pressure, firebox design or they may leave these factors to the individual engineer to adopt; it is accepted that Gresley was committed to 3 cylinders with conjugated valve gear whilst the GWR was committed to its inside motion for all its locomotive until Hawksworth designed the 1500 class with outside Walschaerts valve gear. Even with his LNER background J.F. Harrison committed to Caprotti valve gear for his Pacific design although it has taken half a century to prove his foresight in providing a capable and powerful locomotive.
Given that locomotive designers have their foibles - or identification markers - it seems crass to consider one minor (design) detail against another when the main comparison should be how one design compares with another in meeting the job. In terms of the LNER it is accepted that Gresley led the way in providing high speed locomotives at a cost that was acceptable at the time. As the situation changed - including the operating environment - new ideas were considered and adopted including double chimneys, Kylchap blast pipes and the move from unified to divided drive but this did not impair the ability of earlier designs to meet the needs of the operating department. Thompson may have been tainted by being the first to question - and change - Gresley's 3-cylinder designs but such is the price of progress which Peppercorn benefitted from when producing both his A1 and A2 Pacifics. The current questioning of the unified v divided drive is surely made by those blinkered to a fixed idea and unwilling to accept new ideas of the new operating environment in the post-WWII years hence its uselessness in determining how successful a locomotive designer was in providing locomotives to meet the current operating environment.

 

The roller bearing A1s only numbered 5 examples, so really we should concentrate on the 44 other Peppercorn A1s, which were as per all the other LNER Pacifics with the standard axlebox types. Even then, those locomotives had substantially higher annual mileages and better availability. The roller bearing examples were exceptional, the production models are what we should concentrate on.

Could a three cylinder locomotive with divided drive been designed on the LNER in 1936? Yes, the technology was well known. Raven's class X had this in the 1910s: https://en.wikipedia.org/wiki/NER_Class_X

These had Stephenson's valve gear. Raven preferred unified drive and leading onto the first axle, and this was more prevalent on his other designs for the NER.

Which is an entirely reasonable conclusion, based on the primary evidence we have available.

 

Point of order: conjugated valve gear was superseded by a third set of Walschaerts. That happened to be combined with a change to divided drive, and the third set might have been hard or even impossible to fit in if unified drive had been retained. But they are separate design features.

 

Thank you. That is a very clever arrangement and seems intuitively to be less prone to degraded performance through wear than Gresley's levers.

 

A fair comment, however BR managed it on the rebuilt Bulleid Pacifics (three sets of walschaerts and unified onto centre axle). So again, unified or divided is a bit of a red herring. You are correct of course in that providing a third set of valve gear superseded conjugated for new designs, but also two sets rather than three was the change for all other locomotives.

 

Is it? If the argument is divided v unified then as many variables as possible have to be taken out of the equation surely. Figures may show that separate gear + divided drive gives better/different results than derived motion + unified drive but they won't tell you if it's the separate gear or the divided drive - or indeed a combination of both - that's having the effect.
It's all rather a moot point as no such trial took place so we'll probably never know.

 

But again, the argument isn't divided versus unified. End thereof.

 

So the goalposts have moved?

 

I'm losing track of what is the argument. We know that the conjugated gear had some flaws, which were most significant in wartime conditions. What else made the A1s better than the A4s?

 

Only in your head.

My position hasn’t changed.

 

Townend is your friend to answer that, but for one thing they were Kylchap fitted.

 

You’ve been on about divided drive + separate sets of valve gear versus unified drive + derived motion. Your contention is that the first is better than the second and The conversation moved on to why this may, or may not, be. Was it the divided drive, the separate valve gear or a combination of both? Since none of Gresley’s Pacifics had divided drive and none of Thompson’s/Peppercorn’s had derived motion, it’s impossible to say exactly which feature(s) caused the differences you note. And that’s before we consider material spec and manufacturing standards. It’s entirely possible that divided drive had nothing to do with it but without testing the various permutations, we’ll never know.

 

Initially but eventually all the Gresley Pacifics were so fitted.

 

It's possible of course that the answer is "lots of incremental improvements" rather than one big bang change.

Cook's book "Swindon Steam" is instructive in that regard. In the same period that the GNR / LNER / BR(E) went from a Gresley A1 to a Peppercorn A1, the GWR went from a Castle to ... a Castle. But actually the 1950 Castle is fundamentally improved on the 1923 version, in myriad small ways. The big picture is they are the same loco - at least at the level of detail being discussed (no fundamental change such as conjugated or separate valve gear; or switching between unified and divided drive). But better alignment of frames; incremental increase in superheat area; lots of changes in lubrication and so on make the later versions more efficient and more consistent in mileage run.

I wonder whether the fact that LNER pacifics had major design variations might be blinding us to the cumulative impact of many minor changes in trying to identify the factor that makes a 1950s loco better than a 1920s one. It would be a bit odd if a Peppercorn pacific wasn't better than a Gresley one given the years of development between them, but maybe the critical factor is just that: the years of development. Even in the case of the A4, arguably their Achilles heel - the middle engine big end - was transformed by a small design change to one component, not a root-and-branch redevelopment of the whole locomotive design. Many locomotive classes, if they have reasonable longevity, have those kinds of incremental enhancements and they rarely cause much comment.

Tom

 

And when you think about it it would be rather disappointing if the various drawing offices hadn't advanced the state of the art between 1930 and 1950 - compare a 1930 and 1950 aeroplane (I admit that's not an entirely fair comparison, but it makes the point).
I recall reading somewhere that in the early days of the preservation movement, when 'originality' had the same sort of emphasis that 'authenticity' has today, hearing that professional railwaymen were bemused that the preservationists cherished locomotives without the various fixes that had been put in to make the damn things work better...

 

If I were to make an obvious point, why was Gresley pushing ahead with the V4 when what was desperately needed was the B1?

 

And yet again - not, it is not. You would like it to be specific to the Pacifics, but the fact is we do have divided drive on the B17s with conjugated valve gear, and that is the next closest thing as the B17 was drawn up by using the A1 Pacific drawings by NBL, and the arrangement, bar divided drive, is virtually identical.

We have statistics, we have engine record cards, we have reports, we have secondary sources, and all of them show that the conjugated gear, together with the middle big end design, was, with poor standards of maintenance at sheds and at works, left wanting in a way that separate sets of valve gear wasn't. That's not in dispute - I am not writing anything controversial by saying it - we have clear evidence for it.

Divided drive or unified drive is not the issue. The middle big end, slack and wear in the pins and 2:1 gear is. Separate sets of valve gear were more robust and better maintained with better tolerance to poor maintenance. This is clear from the primary sources.

With respect, I think your line of questioning and reasoning is at odds with the evidence - but I'm not particularly surprised.

You want it to be about unified versus divided, I am satisfied that unified versus divided drive isn't the issue. I am able to make that informed decision because I've done the research and looked across the LNER's fleet as a whole.

Where's your research and data?

 

This is the big question mark, certainly, and also begs the question why Freddie Harrison in his address made the claim that the Peppercorn A1 was "the kind of locomotive Sir Nigel Gresley would have designed had he been alive, to meet the changing conditions of maintenance" when all the evidence points to Gresley retaining the conjugated valve gear well into the 1940s, the V4 being an example of that. You do not get a Peppercorn A1 without Edward Thompson changing the design ethos of the LNER. Fact.

As far as it goes, in my research and in my writing my firm, personal view is that conjugated valve gear was the right answer for the questions asked of Gresley in the 1910s, 20s and 30s. By world war two and the 1940s the world had changed, particularly at the shop floor and in the sheds by way of available staff for maintenance, and those locomotives across the LNER fleet with two cylinders or separate sets of valve gear across three cylinders were shown quite conclusively to be better available for work and doing better mileages. (By the by, but repeatedly making this about just the express locomotives and ignoring everything else is remarkable!)

ES Cox did not make this up when writing his report, Thompson didn't make up the statistics the LNER collected across its regions, most LNER writers, whatever their preference on the gear, accept that during world war two, and post war, the conjugated gear was not the answer. Around the world, the gear fell out of use and the only large hold out were the Gresley designed locomotives of the LNER in Great Britain. All new locomotives for British Railways bar Duke of Gloucester used two cylinders, side stepping this debate altogether, and no other railway in Britain used the conjugated valve gear (EDIT - outside of the known, minimal prototypes by the SECR and Holcroft).

It is not denigrating, or wrong, to say that by 1941, the conjugated gear was superseded by three sets of valve gear. Whether that was divided drive or unified is immaterial, and only becomes material if you choose to blindly ignore the well known and publicised issues of the conjugated gear.

That the older Pacific locomotives with conjugated valve gear were made to work better and have better availability by way of a new design of middle big end and to have better setting up of frames, and more structured maintenance at works and in sheds, post war, does not override the inherent issues of the conjugated valve gear, which have still been seen post war (overheating middle big end, slack in pins and wear/tear in the 2:1 gear) - and actually, if everyone here was being honest - into preservation too.

I think getting into a cyclical debate about unified/divided drive is something of a nonsense, quite frankly. I cannot think of a single other engineer where the defence for the type of valve gear is quite so dogmatic in their lack of bigger picture thinking. It is frankly infuriating. The answers are obvious and the primary evidence makes this easy for us.

By admitting the conjugated valve gear had probably had its day for new locomotives by 1941 mean we are denigrating Gresley? No, of course not. We are accepting the world had changed and the needs of the railway had changed by the time Thompson took over.

Perhaps most damning is that neither Peppercorn nor Harrison reintroduced conjugated valve gear, despite having ample opportunity to do so.

 

It was you who implied that divided or unified drive was significant when you said "divided drive locomotives had better availability than the conjugated valve gear locomotives" (post 1033). Perhaps what you meant was "locomotives with three sets of valve gear (which by the way also happened to have divided drive) had better availability than the conjugated valve gear locomotives".

 

But that statement is still correct!