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

French steam with Chapelon thermal improvements did an awfull lot of work from 1945 t0 the utter end.
Gresley high speed non stop steam hauled not so much I think and LNER could as well have bougth some danish designed Lyntog (1932) and offered a faster service for the select few.

 

https://en.wikipedia.org/wiki/Railway_electrification_in_France

 

Certainly the German 2-10-0s were the World's most numerous 3-cylinder locomotives, although I think Britain had a larger variety of 3-cylinder types.

Although many forum members will be familiar with the BR44 2-10-0, the older G12 is much less well-known among British railfans, partly because the locos in West Germany were withdrawn by the Bundesbahn in the early 1950s. I have seen very few references in the literature to the Henschel conjugated valve gear used on the G12 and a few other classes. ES Cox gives a brief description ("World Steeam in 20th Century" - page 115) and indicates that the gear had similar issues to the Gresley gear, due to wear in the pins and linkages building-up and distorting the inside valve timings. But on slow-moving freight locomotives, this was possibly less of an issue than on express types.

So far as countries outside Europe are concerned, I think it is true that 3-cylinder propulsion was only popular in a few countries during the 1920s and then fell from favour. As always, there are exceptions, for example Malaya had some attractive metre-gauge 3-cylinder pacifics built in later times.

German use of 3-cylinders rather ebbed and flowed. The Prussian and Saxon State Railways introduced several 3-cylinder types in the 1915-20 period, some with the Henschel conjugated gear and some with 3 separate valve gears. But after the Reichsbahn unification of 1920, a new central design office was created under Herr Wagner, who generally favoured 2-cylinder simple designs for the new standard classes. An exception was made in the case of the BR44 heavy 2-10-0, where the 3-cylinder design was preferred to a 2-cylinder alternative. From the mid-1930s, 3 cylinders came back into favour for express engines to cope with the higher speeds being introduced.

Elsewhere the French Est Rly had one large class of 3-cylinder 2-10-0s (SNCF 150E) and among the smaller countries, 3 cylinders were popular in Denmark, Czechoslovakia and Bulgaria, but rare or non-existent in most other places.

Overall, I don't think any other railway so consistently championed the 3-cylinder design option as did the LNER under Nigel Gresley.

 

Continued by Bulleid of course! (140 out of 181 steam locos designed for the SR were 3 cylinder which must be something of a record in percentage terms).

Where does Maunsell stand against Gresley in percentage of three cylinder locos? N1, K1, U1, W, Z, V were all three cylinder designs. Though the difference is that Maunsell doesn't seem to have an ideological fixation on any particular format - 2 cylinder, 3 cylinder conjugated, 3 cylinder non-conjugated and 4 cylinder designs as required.

Tom

 

I cannot argue with Tom's thoughts!

Maunsell certainly contributed to Britain's variety of 3-cylinder types. In the case of the Maunsell Moguls, he seemed to be building all 4 types virtually in parallel around 1930. Possibly the building of more N1s and U1s was partly to ensure an adequate stock of locos able to work on the width-restricted Tonbridge to Hastings route?

Bulleid claims the prize of being the most prolific builder of 3-cylinder Pacifics. For the interest of those who like their statistics (but apologies to those who don't), I attach a list of what I think were the 10 most numerous classes of 3-cylinder simple-expansion locos.

 

Had Maunsell built any 3-cylinder locomotives before Holcroft arrived on the SECR?

 

No.

Maunsell arrived at the SECR in 1913 (and Holcroft shortly after). The first things Maunsell did - on the loco front - was to make a few revisions to the L class inside-cylinder 4-4-0s which had been designed, but not built, by Wainwright. Following those revisions, construction was completed by Beyer-Peacock and Borsig (Berlin) (1913).

The next locos were the D1 class, essentially reboilering the Wainwright D with a larger superheated boiler. (Starting 1913 and continuing as locos fell due for overhauls).

He also sorted out the two "missing" H class locos which had been ordered under Wainwright but somehow disappeared into the pool of spares at Ashford without actually being built. (1915)

The last Maunsell "Wainwright" design was the S class shunter, essentially rebuilding a single C class 0-6-0 into an 0-6-0ST after the board had turned down the request for a new design of shunter. (1917)

The first pure "Maunsell" design was the N class two cylinder 2-6-0, which was designed in 1914 but not built until 1917. The prototype K class 2-6-4T dated from the same time, the first appearing in 1917.

The three cylinder designed started in 1922 with the prototype N1; further locos followed after the formation of the SR.

The final tally of Maunsell 3 cylinder locos was:

• N1 2-6-0, conjugated: 1 (Built by SECR, later rebuilt with three sets of independent valve gear) (1922)
• N1 2-6-0, three sets of independent valve gear: 5 (Second batch, built by the SR) (1930)
• K1 2-6-4T, conjugated: 1 (Later rebuilt as U1, later rebuilt with independent valve gear) (1925, rebuilt 1928)
• U1 2-6-0, three sets of independent valve gear: 20 (plus one K1 rebuild) (1931)
• W 2-6-4T, independent valve gear: 15 (1930 onwards)
• Z 0-8-0T, independent valve gear: 8 (1929)
• V 4-4-0 ("Schools"), independent valve gear: 40 (1930 onwards)

So 90 locos in total, of which only 2 had conjugated valve gear - the N1 prototype and the K1 / U1 prototype - both of which were rebuilt into conventional form.

Withe the exception of the Schools (which was a very different design lineage), all the three-cylinder locos of the "Mogul" family (N1, K1, U1, W, Z) had the valves offset inwards from the cylinders, giving a very "sloped in" appearance to the outside cylinders. That was to provide space for the conjugating levers, but in practice only a couple of locos were thus fitted.



Source: https://commons.wikimedia.org/wiki/File:SECR_N1_class.jpg



Source: https://commons.wikimedia.org/wiki/File:K1_class.jpg

Tom

 

Maybe its just my imagination, but is there just a hint that the conjugated Southern locomotives were a sop to Holcroft to stop him defecting to Gresley?

 

I think those photographs you've posted Tom confirm to me why Gresley's patent was accepted as being different to Holcroft's by the Patent Office.

His version of the 2-1 lever, by being at the valve spindle ends, or behind the cylinders (as per the B17) meant you could still use the same patterns and dimensions of the outside cylinders without having to put the linkage on the outside of the cylinder.

It's also why I am loathe to continue the use of "Gresley-Holcroft" as seems to have been the practice by LNER writers to date.

Gresley's conjugated valve gear does the same thing, but does it very differently and mechanically is different. It is perhaps telling that Gresley's conjugated gear patent was never amended other than to extend the patent.

I don't believe the two gears are that similar, other than the basic principle of them both to drive the central cylinder by means of links and pins and not a separate set of valve gear. It's that outside linkage that does it for me, Gresley's does away with that entirely. On the original O2, no.461, there were additional rocking shafts which were then deleted by the production O2s and the relative simplicity of the Gresley gear is best shown by this photograph and drawing:





Source for both: https://www.lner.info/article/tech/valvegear/gresley.php

I think you could argue quite cogently that Gresley's gear isn't really Holcroft's or vice versa. They're achieving the same basic principle but quite differently.

 

The question will be about the dates of construction, IMO. I don't know enough about the SECR locomotives. However Gresley and Holcroft did have a meeting after the introduction of Gresley's prototype O2 and his patent being approved.

Holcroft's patent ran out in 1913 and Gresley's was approved in October 1916.

 

I think Maunsell was curious enough to want to see a trial; there were numerous examples of such trials on the SR where a loco was built to either try a design direction, or validate a previous one. (For example, the Lord Nelson's notably had 135° crank settings with the aim of getting more even torque - but one was built with conventional 90° cranks to validate whether there was actually an advantage).

One of Maunsell's great strengths though as a CME was that he was decisive without being especially dogmatic in design terms. So the conjugated locos were built, evaluated - and then the decision was made that the disadvantages outweighed the advantages, such that no more were built and those that existed were converted.

Tom

 

That's an interesting comparison, are there any reports or data on it? Just out of interest. I assume the 90 degree locomotive was rebuilt to match the others after a trial?

Contrast this with the treatment of Thompson post 1941. It's really interesting to see how the two different schools of locomotive design are assessed by their enthusiasts.

 

I think there is more similarity than you give credit for.

Holcroft's original conception had been for the conjugating gear to be set behind the cylinders. When Maunsell charged him with designing a three cylinder 2-6-0 (the N1, built in 1922 but with design starting in 1919) Holcroft found that there was insufficient space since the leading coupled wheels were in the way. (Notably, prior to that he had pitched a three cylinder conjugated 4-4-0 to Maunsell, where there may well have been space to mount the valve gear where intended).

At that point, I suspect Holcroft had discovered what Gresley had already found out on his three cylinder mogul: you can't fit conjugated valve gear behind the cylinders. It is at that point that the designs diverge. Gresley took the obvious answer of taking the drive for the inside cylinder from the front of the cylinders. That simplified things mechanically, but introduced the timing error caused when the valve spindles expanded as they got hotter. Holcroft, by contrast, maintained the drive from behind the cylinders, avoiding the timing error - but introducing the long drive rods running past the cylinders which in practice turned out to be insufficiently stiff.

So I read it that there is a common genesis; and then varying solutions to a common problem of how you package all the machinery. Gresley found one solution which introduced one problem; Holcroft found a different solution which introduced a different problem.

The difference was then that Maunsell called a halt and reverted to three independent sets of valve gear, whereas Gresley persisted with conjugation.

Were I being mischievous, I might suggest that Maunsell - as CME - had less to lose in stopping work on an interesting design idea promoted by his subordinate; whereas Gresley had more directly staked his personal reputation on the design and therefore would have had more explaining to do had the LNER engaged on a programme of rebuilding his own locos with three sets of independent valve gear - even had it been mechanically possible.

Tom

 

I know a bit (but not a lot) about patents, having worked at the Patent Office for about a year and a half.
I don't understand the statement in that article about the "expiration" of the Holcroft patent. If that was applied for in 1909 (and therefore granted in that year or later) it should still have been in force when the Gresley arrangement was first used*. Possibly the Gresley arrangement was outside the scope of the Holcroft patent. It is certainly sufficiently different to have been entitled to a patent in its own right, but usually the scope of protection which is claimed is rather broader than the specific arrangements described. Possibly whoever drafted the application for the Holcroft arrangement drafted the scope too narrowly, or possibly a scope broad enough to cover both arrangements would also have covered some prior arrangement and would therefore not have been eligible for a patent.
*Edit: I've now seen Simon's statement that Holcroft's patent ran out in 1913. I don't understand such a short lifetime, unless no-one could be bothered to pay to keep it in force.

 

Patents are tricky things. AIUI enforcement tends to be about the detail, not the concepts. Holcroft's original design was symmetrical with the valve rod on the centreline, and seems to have been a clean sheet of paper job, although when applying for the patent some prior art was turned up for a conjugated gear by David Joy for 3 cylinder compound marine engines. Gresley's first design on the O2 was undesirably complicated due to the centre valve not being on the same plane as the others, and it was Holcroft who pointed out to him that just because the piston rod was on a different plane the valve rod didn't need to be. But surely Gresley's original concept wouldn't have had all the complication that the implementation on the O2 had. Its unclear to me whether Gresley came up with his concept completely independantly, or was influenced by Joy or Holcroft's patents. What is certain is that the final form with all valve rods on the same plane and the slight adjustment of the crank angle to suit was Holcroft's suggestion.

 

IIRC B17 and D49 had their derived motion behind the cylinders.

 

I think we need to see a drawing of Holcroft's to compare Tom. You may well be right but having read Gresley's patents and looking at what I have by way of photographs/drawings so far, I am not convinced. I think it's overplayed.

(I am open to having my mind changed, of course!)

But Gresley's B17 and D49 classes are examples of locomotives where conjugated valve gear IS fitted behind the cylinders. In the case of the B17 it is done through divided drive, onto the front and middle driving wheels.

Don't disagree with any of that Tom, but where Gresley is concerned, he had great success with his prototype O2, no.461, and ultimately his conjugated valve gear locomotives did good work up until 1939, and when conditions changed post war, worked to the end of steam, so fundamentally we're not talking about locomotives or valve gear that are so awful they are unusable, quite the opposite in fact.

So his reputation may have been staked on three cylinders and conjugated gear but I personally don't think he had much to worry about. No.461 proved the theory, the production Gresley O2s had changes made, and his prototype K3, no.1000, had this later arrangement, was very successful and from there pretty much all of Gresley's conjugated classes would be more or less like this prototype.

 

In 'Locomotive Adventure' Holcroft says that Gresley found a loophole in Holcroft's patent that a bit more thought might have closed.

 

Thank you. That is making sense and would also explain why no-one bothered to pay renewal fees on the Holcroft patent.

 

Much to agree with here, but ultimately it doesn't matter if Gresley came up with it independently or not. The fact remains is that his interpretation was different enough to Holcroft's to be accepted without argument.

If I may though Jim, Gresley's original patent appears to cater for the valve rods being in the same plane or not by way of its wording. The drawings attached show these two versions of the Gresley arrangement.

The patent was not modified post 1916, and was in place before his meeting with Holcroft, so it looks as if Gresley was already thinking along those lines and Holcroft confirmed his thoughts. That's one interpretation: others are available of course.

Happy to send you a copy of the patent by email if you drop me a PM for your perusal, if you're interested.