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

I'd go along with that breakdown, but do we know how Gresley himself categorized his designs?

Another question - when did the term "mixed traffic" come into general use on Britain's railways? When the Midland Rly renumbered its locos in 1907, it grouped them into just four categories - passenger tender, passenger tank, goods tender & goods tank. The LMS based its 1923 renumbering on the same breakdown, but ended up with some anomalies. For example, the first batch of Caledonian 5ft 9in 4-6-0s (908-class) were numbered in the passenger series and acquired red livery, but the superheated variant of the same type (179-class) were numbered in the goods series and painted black.

Ahrons (British Steam Rly Loco 1825-1925) uses the term "mixed traffic" a couple of times late in his book, but after describing small-wheeled 4-6-0s under the heading "4-6-0 Express Goods Engines" and 2-6-0s (including the GW 43XX) as "2-6-0 Goods Engines". Gresley's first 2-6-0 of 1912 does indeed appear to have been originally conceived for express goods work, but was found suitable for a wider range of duties and initiated a large family of LNER mixed-traffic types - K2, K3, V2.

 

Coming back to the graphing question ...

Suppose instead of Gresley, you'd chosen to concentrate on that doyen of inter-war CMEs, Mr L.H. Crankpin of the famed Much Meddling and Little Snoring Railway.

The MM&LSR's locomotive output from the famed Little Snoring Works in the interwar years was as follows:



As all aficionados of the MM&LSR know, the locomotive classes were as follows:

GA - A goods 0-6-0, the type tracing its ancestry back to Victorian times
GB - Crankpin's famed enlargement of the class GA, with half an inch more cylinder diameter and 10psi more boiler pressure
GC - A more goods modern 2-8-0

MA - A mixed traffic design, designed along Churchwardian lines during World War I but not produced until after hostilities ended
MB - A bigger mixed-traffic 4-6-0

PA - The apogee of MM&LSR types, an express passenger pacific

SA - a suburban 0-4-4T, again tracing its ancestry back to Victorian times
SB - a modern suburban passenger loco, essentially the tank engine counterpart to type MA.

So how could you present those data graphically? It seems there are a couple of ways, depending on the story you wanted to tell.

One way is the stacked line chart (or a stacked bar chart might in some ways be preferable).



What's the story here? Well, three things are really obvious. The first is that total locomotive productions sharply declined, such that by 1938 the famed Little Snoring works was turning out fewer than half the locos they had been only a decade before. Of course, we all know about the Meddling depression and the damage it wrought upon the company's traffic, and in particular its goods traffic. The second is that the company more or less completely stopped building passenger locos - and particularly suburban passenger locos - by the mid 1930s. That has another explanation - the great Little Snoring electrification project, part funded by the Government which made new passenger locos redundant. (Steam was still preferred for goods traffic). The third feature is the recovery of fortunes from 1940 onwards, but for goods and mixed traffic locos only - a consequence of wartime exigencies.

The other way you could present the same data is in a kind of timeline view that just shows the years in which any loco of a given type was built - regardless of numbers:




The notable features from that view is that the two most modern types - the 2-8-0 goods and the 4-6-0 mixed traffic design - were built for a period, but then construction reverted to the older types that they had nominally superseded. That's the Meddling depression again: not only did the company need fewer locos, but it needed smaller, cheaper ones. No point building a highly-capable goods 2-8-0 at £7,527 4/- 6d per engine and tender, painted and delivered, if the traffic on offer could be readily hauled by a simple 0-6-0 buildable at £5,216 3/1 9d.

OK, that's a bit of whimsy - but shows how you could make graphical sense of a large data set of when particular types were constructed to illustrate the relationship between the works output of different types, and the external factors that led to those decisions about construction.

Tom

 

I really like this Tom, but there isn't enough time to develop this fully for the Gresley book unfortunately.

It's an approach I think should be followed for future books and research though. Thanks for sharing your thoughts on this.

 

In "Master Builders of Steam" H.A.V. Bulleid claims that Churchward always referred to it as the de Glehn bogie.

 

We are at the final stages of the book's development. The final stats (give or take the final editing and setting out) are:

• 422 pages
• 320 images, including over 250 photographs together with graphs and line drawings
• A bibliography with over 130 titles, primary evidence sources and more
• A piece on the history of the Gresley Society, part of SNG's legacy

And much more.

I feel that I've learned much from the experience and I am hoping that it ends up better than the Thompson book, which I will revisit in a few years time to apply the lessons learned (once I've finished other books!)

Overall, I am feeling pretty happy with the draft and I am grateful to everyone who has given up their time to review, pass on their ideas and edits to improve the quality of the book.

There is still time to get an editing credit in the book, if you have the time to read all 422 pages at this stage!

 

@Fred Kerr you have a PM.

 

422 pages? That's nearly 2½ times the length of the Thompson book. Which is perhaps to be expected, as there is a great deal to cover in Gresley's 30 years as a CME. But you seem to have written it much quicker than the Thompson book?

I sense that a mighty tome is coming. Potentially a most interesting arrival, but one that I must be careful not to drop on my foot!

 

Well done Simon....will one of the future books (Next?) be about OVSB?

 

Definitely not quicker! It took longer to research Thompson's work and unpick everything. Writing the Thompson book notionally took nearly ten years but the bulk of it actually came together in about a year. The Gresley book is based on the research the Thompson book developed and has been in writing for three years this month.

I hope so, it includes some good material from The Gresley Society (find them here: https://gresley.org/ ) and good amount of primary evidence and information, I hope. Definitely don't drop it on the foot - my first editing copy is pretty hefty!

I am taking my time with Bulleid, there's a to learn and a hell of a lot of books to read. The next one after this will be my book on the full wartime performances of the LNER's locomotive stock, and then I might do a bit of a segway into the GNR for a bit. The Bulleid one is coming, but it all takes time to get it done properly.

 

Dammit @Jamessquared



Watch this space!

 

Getting there. Definite patterns emerging. I am using the figures in RCTS and cross referencing with LNER internal reports and the use of engine power documents to confirm years. I have had to in some cases round up to a single year builds over two years as we simply don't know the spread of releasing to traffic accurately enough. (e.g. for a build that says 25 engines 1933-34 I have put all 25 in 1934).

So this isn't going to be an exact science, but it's as close as I can get with the data available, I think.

 

Never let it be said I don't suffer for my craft...!

Express Passenger
A1, A3, A4, B17, P2, W1

Secondary Passenger
N2, D49, V1, V3

Mixed Traffic
J6, K2, K3, J39, V2, V4, K4, EM1

Freight
O1, O2, J38, P1

Shunting
J50

Specialised
U1

So the tables were then organised as according to those arbitrary groupings, once I had worked out all of the build years for all of the batches (and crposs referenced the class totals by way of a simple sum formula).



This then produces a graph not unlike Tom's (but I am still playing around with it), and a line graph I produced after.





I think these graphs shows us a few trends. Thoughts?

 

Build is interesting of course, but its the (more time consuming to calculate) size of the fleet that really matters. Also age. One might expect the age of the fleet to increase in wartime, and possibly also in the depression. One might also see changes in the proportions of freight to passenger...

An interesting one might be to explore the size/weight of the LNER locomotive fleet to highlight the effects of Gresley's large locomotive policy. There must be some things that could be shown. Larger locomotives potentially reduce the size of the required fleet due to elimination of double heading and fewer/longer trains, improved availability also reduces the size of the required fleet. On the other hand increasing passenger and freight mileage would require a larger fleet.

 

So on that, the overall LNER fleet decreased by around 1000 locomotives in Gresley's time, and the number of mixed traffic engines in proportion to everything else went up. The mixed traffic locos varied from 0-6-0 tender to 2-6-0 and then to 2-6-2 tender locomotives. These were also the three largest classes built under Gresley (J39, K3, V2).

 

I think the proportion of mixed-traffic types increased on all of the Big 4 companies during the Grouping period - and indeed beyond into BR and the early diesel eras. A difference is that Gresley went for a larger and more powerful engine in the V2 when compared with the LMS & GWR mixed-traffic classes.

As you say, the overall LNER loco fleet size dropped significantly, mainly in the 1930-39 period. Looking more closely at some of the stats listed by RCTS, the LNER's stock of 0-6-0s dropped by about a third and the number of 4-4-0s by almost half during Gresley's period of office. I am unsure whether it is possible to quantify between the several reasons that this will have taken place:
(1) Replaced by new engines, albeit not necessarily on a 1:1 basis.
(2) Improved utilisation reduced the number of engines needed.
(3) Withdrawn because there was no longer work for them, due to the 1930s Depression or other changes in trade patterns.

In looking at the overall LNER fleet, it is necessary to include engines that were acquired but not always built to LNER standard designs. Most notably, the LNER built relatively few heavy freight locos, but that was because of the purchase of 273 ex-ROD 2-8-0s and the post-WW2 purchase of 200 ex-WD 2-8-0s. Gresley built quite a few engines to designs other than his own, notably an extra 112 N7 0-6-2Ts. You might consider some kind of graph (or stacked histogram?) that shows the overall acquisition pattern, split between Gresley standards and other acquisitions.

 

Might also be interesting to compare the mileages of different route restrictions on the different lines. From what I gathered from Simon's book the LNER had a much greater proportion of RA9 route than the GWR had of double red. A V2 equivalent on the GWR would have been restricted to the same routes as the Kings.

 

The thing that jumps out is how “spiky” it is. I wonder if that is real or an artefact of your earlier comment about e.g. grouping all 1933-34 locos in one year. (Maybe splitting equally across two years would be more realistic where you can’t definitively say?)

Every graph leads to further questions . On the face of it the workshop output looks up and down but - pace @Jimc - maybe there is a tonnage consideration (presumably it is more work making a V2 than a J-something). And maybe when they were building fewer locos the works were instead rebuilding older locos.

Tom

 

That paper of Wilson's mentioned the Government's New Works Programme 1935-40. I wonder how much difference that made to the case for new build in the late 30s, especially of course the V2s.

 

I did think about that but when I looked back at my figures, it was clear that it wouldn't change any of the most obvious spikes in the data.

 

Weren't some of the A4s built under a Government finance scheme to reduce unemployment?