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

Lots of interesting posts on this thread - and I'm struggling to keep up!

It's interesting to hear that your views are evolving as your researches proceed. I imagine that this will continue.

One of the issues to consider is how Gresley managed to bring together his department in the 1920s and apparently avoid the factional infighting that characterised the LMS during this period. It nay be that this task was made easier by the fact he did not undertake a widespread "Scrap and build" programme to replace inherited types with new standard models. There was no "Great Northern knows best" approach akin to the "Midland knows best" philosophy reigning to the west.

In contrast to the "small engine policy" that held sway on the LMS during the mid-1920s, Gresley was very much a "big engine" advocate. As well as the Pacifics, he built substantial numbers of K3s, D49s and later V2s, while even the J38/39 were (I think) the heaviest 6-wheelers anywhere outside North America. There were many routes, particularly in East Anglia and Scotland, where none of these high axle-load machines could go. And of course, for a given budget, you could build fewer of these engines than of smaller and lighter types.

Not until 1941 did Gresley introduce a general purpose type with a low axle-load and high route availability - the V4. You and other commentators have suggested that he should have made that move much earlier. There would certainly have been benefits from such a move, but I would enter one note of caution. It would almost certainly have been a 3-cylinder machine with conjugated gear, and that would have exacerbated the difficulties that would face the LNER during WW2.

I don't see that the LNER locomotive fleet was any more diverse than that of the LMS in 1923, but of course far more of the LMS fleet would be replaced by 1939. One point to remember is that Pre-Grouping types mostly stayed in their home area, so that particular Workshops and Loco Depots were mostly dealing only with just two sets of designs - their own Pre-Group types and the Group Standards.

The Pre-Group company loco fleets were mostly fairly well standardised, with plenty of large standard classes and sharing between classes of boilers and other components. There were of course some glaring exceptions, such as the 9 classes of GC 4-6-0 and several dead-end NE developments (A6, B14, D21, etc). But if you think the LNER had a monopoly on diversity, note that the Caledonian provided the LMS with 4-6-0s of 10 different classes using 9 different boiler types (and that's not counting saturated and superheated variants of the same boiler).

 

Ah, thanks for that Peter, I knew I should have checked first...

I certainly remember 847 - a Shire horse of a locomotive (but as horses go, give me a heavy horse over an Arabian stallion any day of the week ...)

Tom

 

It’s probably best summed up by thinking about what designers are inherent in the makeup of the pre-grouping stock of the LNER.

If you take even Robinson as an example, there’s dozens of varieties of boilers across his locomotive types, the only “standard” boiler and locomotive he designed, really, being the impressive 8K (or LNER O4) for which further boiler types were added by Gresley and Thompson, understandably as the original boilers became life expired.

But that then limited these locomotives to pools of those boiler types and thus pushed up timescales for repair and pushed availability down. Double edged sword, upgrading boiler types, IMO, unless they’re done en-masse to justify and - definitely - interchangeable with other members of the same class (e.g as the newer K5 boiler was for the K3s that were fitted with this type later) then you tend to get not entirely unexpected changes to your overall availability.

 

Somebody can do the maths but it wouldn't surprise me if they weren't far off the same in terms of underbridge numbers (purely gut instinct) - you trade off the underbridges of the GC* and in Scotland against the vast number of level crossings in East Anglia and (then, and to an extent now) the ECML.

*and then add into that the GC London Extension, which was hilariously over engineered and only 20 years (ish) old at the grouping, so that knocks out another swathe of underbridges which didn't need upgrading.

Like I say, could be very wrong, but I wonder if there was that much in it.

 

Looking at the LNER Boiler Diagrams list (in RCTS "Locos of LNER Pt1), I was surprised to see that the GER had rather more boiler types than either the GNR or the GCR. The reason seems to be that the GER had a lot of types of small tank engine with unique boiler types.

You mentioned in a previous post how the SECR H-class boiler was fitted to several classes from both the SER & LCDR. Some years later,the GWR adopted a policy of "Swindonizing" many of the locos absorbed at Grouping, including fitting of standard GWR boilers. Many of the newer Taff Vale and Rhymney engines survived well into the 1950s.

That prompts the question of what was the LNER's policy on improvement & rationalization of its absorbed fleets? Its approach seemed to vary between Areas, suggesting that initiatives were coming from the Area Mechanical Engineers rather then being centrally directed. In Scotland, I don't think there were any major changes to absorbed NBR & GNSR types. On the GER Section on the other hand, there was quite a lot of upgrading, involving fitting of Gresley-type boilers with round-top fireboxes. In the case of the B12/3, that involved a substantial size increase. But note that these new boilers were specifically designed for the GER fleet, and were not used elsewhere.

I can only think of one case where LNER standard boilers were fitted to a Pre-Group design, that being the ex-GCR/ROD Class O4. A couple of these were given the O2-type boiler, an expensive move that required a frame extension. Subsequent rebuilds used shortened O2 boilers and subsequently the Thompson B1 (Diagram 100A) boiler. But note that the GWR, who liked to fit their own standard boilers, retained the GCR boilers on their ex-ROD engines.

Should the LNER have made efforts to design standard boilers that would fit a range of its absorbed engines, across company boundaries, analogous to what the SECR and GWR had done? Perhaps Gresley did look at that possibility, but decided against it?

 

They did and they didn't. The RODs that the GWR kept long term were given top feed, new copper inner fireboxes and Swindon design superheaters. 50 locomotives was enough for a sensible boiler pool although there was a WW2 proposal to fit Std 1 boilers which went nowhere. Might one speculate they realised there were going to be a new selection of cheap ex government freight engines after the war, so upgrading the old ones would be pointless?

There's an interesting passage in Cox' 'Chronicles of steam' about Stanier proposals to fit standard taper boilers to existing LMs and pre group classes. Maybe its just me, but I read into it that the LMS design offices were extremely reluctant about the idea and did all they could to block it - and of course succeeded.

 

That sounds like two threadsworth of debate in it's own right!

Perhaps there's a distinction to be drawn between standardising fleets of locos, designed and constructed in-house by larger railways and those of absorbed smaller lines, which were more likely the product of outside builders?

I'm aware of no company which took standardisation to the extremes of Mr Churchward's meccano set. Even Stroudley's standardisation at Brighton had nothing like the lasting effects on the Brighton school of design .... never mind it's Southern successor. Ditto that of Wainwright at Ashford.

 

James l’Anson Cudworth’s SER? 110 of the 2-4-0s and 53 of the 0-6-0 goods: those are big numbers on a railway with fewer than 400 locos at the time.

There’s also the question of replacement rate. I’d be interested to know what it is for the LBER (i.e. what percentage of the fleet was replaced each year). If you take 3% as a typical figure (probably slightly higher in Victorian times; slightly lower in the twentieth century when locos tended to last longer, at least before the BR modernisation plan) then it takes about 30+ years to work right through a fleet. Gresley was CME of the LNER for 18 years so allowing for a couple of years of war and probably a year or so on formation when each workshop would likely have continued with orders placed in 1922, then realistically upon his death, about half the fleet should have been of his own design; about half older pre-grouping locos. (Including his own GNR designs in the latter, of course). It would be unrealistic for their not to still be considerable numbers of pre-grouping designs in 1941.

A worked example: Wainwright built 287 locos of his own design: of those, 252 were of just three basic types: 109 C class goods; 66 H class passenger tanks and 77 4-4-0 express locos (the D and E, closely related except for two types of boiler). In addition to that, the H class boiler was widely used across older Stirling and Kirtley locos. The combined SECR stock on handover to the SR was 726; I think somewhere around 700 on formation. So Wainwright was CME for about 14 years: at 3% replacement per year he should havebuilt about 300 locos (he built 287) and of those 287, 252 were of three/four basic designs.

Tom

 

I'll always bow to your knowledge of the 19th century stuff trundling around Kent Tom ... but I'm standing by my observation on Swindon's 20th century 'mix'n'match' school of design, which you'll have to admit, took it all to another level!

 

I think the GWR was a special case though, for a number of reasons.

Firstly, on amalgamation, the "new" GWR was essentially the "old" GWR with a fairly small addition of minor companies. By contrast, the other three members of the Big 4 were all formed from at least three companies of roughly comparable size, such that no single loco design school represented even half the stock.

To put it in LNER terms, it was as if Gresley moved from GNR to "new" GNR, with the NER, GCR, GER etc being minor in terms of their bequest to the new group.

Secondly, the GWR followed a largely singular design philosophy for 50 years, from 1902 to the early BR(W) period. That's long enough to get through the entire fleet once, and be well on the way to a second renewal. By contrast, the other three companies had at best 25 years which doesn't get you right through the fleet, and in each case whenever the CME changed, loco philosophy seemed to change direction sharply. Maunsell --> Bulleid; Gresley --> Thompson or Chaos --> Stanier all represented a far more significant change of philosophy than Churchward --> Collett --> Hawksworth.

Even so, I suspect that there is more engineering development (= non-standardisation!) than meets the eye. A 1950 Castle is not the same as a 1922 example. The strength of the Swindon design school was that because Churchward basically got the big calls right (or at least, agreed ...) about boiler design, front end design and general layout, it freed Collett to concentrate on making incremental developments in the field of both improving availability, and making it more consistent. @S.A.C. Martin talks about asset management: my reading is that it is in that field in particular that the GWR were ahead of the field through the 1930s, but largely unappreciated by the "top trumps" school of locomotive history.

Tom

 

'Swindonisation' is one of those odd terms that's broadly understood, though beyond a change of chimney, a lick of paint and adding a Ramsbottom safety valve cover, I suspect most of us would have little idea what it entailed. The only locos for which I've seen any detailed description of rebuilds were the W&LLR's pair .... and I'm guessing they're not exactly typical!

Where it gets interesting to me is a rebuilt absorbed loco which seems to (and I'll put it no stronger than this) inform a subsequent GW design. The "Valleys" 0-6-2T classes and the 56xx, the later Cambrian 0-6-0 and the 2251, or the real oddball, the 1366 class which although supposedly an update of the 1361, actually looked an awful lot like the rebuilds of the CMDPR's Manning Wardle 0-6-0ST pair.

On component commonality, use of parts of dismantled 43xx Moguls in nominally new 68xx and 78xx is a case in point. For that matter, so are the 'Barry 10' parts going into 4709.

 

This ignores the fact that Gresley also made "incremental developments" during the course of his employment. Many critics see his obsessive adherence to the 3-cylinder conjugated valve gear as indicative of a CME unwilling to be forward looking but surely his willingness to use wind tunnels, to accept the erroneous thumb print in the design of the A4, his involvement with the EM1 in the 1930s and his willingness to experiment with different boilers for his V4 design before making a final commitment suggest that Gresley wasn't the unimaginative designer that many portray him to be. Given that Simon has access to more records from the time one hopes that a fuller picture of his stewardship of the CME role can be shown.

 

You are misreading what I wrote. I'm certainly not saying Gresley was unimaginative; nor am I saying that the LNER made no incremental improvements under his tenure. What I am saying is that because the core loco policy on the GWR was very settled over a long period of time, it gave the CME (particularly Collett) a lot of scope to concentrate on detail design, particularly with the objective of improving availability, and making shopping intervals more consistent. Read KJ Cook's book on the subject; I've not seen evidence that any of the other three of the Big 4 went to anything like the same degree of consistent mileage-based overhauls.

Tom

 

I don’t think anyone sees Gresley as unimaginative. What interests me about this discussion is the question of focus, and understanding how & why someone as forward looking as Gresley could be in some matters could also have done so relatively little in others.


Sent from my iPad using Tapatalk

 

A minor point, but there was also a background on the LNER: North Eastern and Hull and Barnsley, and earlier the Great Central and Lancashire, Derbyshire, and East Coast.

 

To be fair, though, neither of those were anything like as material as the LNWR/L&Y merger.

 

But nobody is suggesting that Gresley wasn't innovative. He was. He led the way in so many different aspects of locomotive design, and the design process. Nobody is questioning that.

The point being made is that he also needs to have been a good administrator, recognising the issues of the working railway and how to deal with them. There's a question mark here as the issues suffered by the LNER in WW2 were a coming together of many different and complicated factors: but one suspects had there been a go anywhere 4-6-0 in high numbers replacing the vast quantities of pre grouping stock earlier, the LNER might not have found itself in the position it was in by 1941.

This isn't being critical, incidentally. The question we are trying to unearth is "why". Gresley's place in history is cemented.

 

Interesting word, innovative. I'd say you could apply that to Bulleid, no problem.

With Gresley, was it more a case of brilliant open minded student of the art, knowing a good thing when he saw it, willingness to learn from Churchward, Chapelon and indeed the experiences of his earlier designs? Was there any one single innovation in, say the A1/A3 which was down to Gresley and team? Or was it more a matter of finding a winning combination based on careful study of best practice? Was he one of the best at putting it all together?

 

@30567 speaks for me in part, but there's another point that nags at me - the comparison with what was happening at Swindon. Something came to a head in 1941, but I'm curious not just about the administration, but also where the innovation was applied, and to what ends. We can all wax lyrical about the A4s, and the streamlined steam passages, and Bugatti, etc. - and that was superlative. But where was that detailed asset engineering focus, looking for the minor changes that would deliver LNER better value for money? In an era before the term was coined, where was the kaizen approach in LNER? How did LNER deliver incremental improvements in the way that Collett and Hawksworth did in Swindon?

 

Indeed...

I think also, the GW got it very right with timing. If they'd stuck with Dean's MO, it wouldn't have stuck. If Churchward had just been a Dean+, then it wouldn't have happened. Or, if the Welsh lines had been stronger (either economically or in terms of being a unified bloc, again, you would have a different outcome.

Also, AIUI, he made his call at a point where locomotive design stablised into that general paradigm. Hawksworth's pacific dipped into the LMS design school. Would the Churchward kit of parts cater for the new class 8 locos. Had the GWR and steam run for another few decades, would we see more stanier/riddles-shaped locomotives?

And if Churchward hadn't been Churchward, would Stanier have come up with the same answers? A butterfly moment