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

Even worse valve events, restrictive steam passages, there must be an awful lot of ways to stuff up the design of a steam locomotive.

 

True, but an experienced CME and a competent drawing office should have been able to avoid most of them.

 

**cough** Dugald Drummond E10 **cough**

I think the evidence is that locomotive design was still pretty hit and miss in the late pre-grouping and early-grouping period, particularly when designers were faced with a problem that went beyond "just give it another 1/4" on cylinder diameter from the previous batch". That goes equally for boiler design, cylinder / front end design and mechanical design. Just read "Raising Steam on the LMS" to see how an organisation with the resources of the LMS still struggled with some pretty fundamental areas of boiler design.

I'm not sure I'd read much into that comparison between an A1 and an A2 beyond neither appeared especially stellar.

Tom

 

If we measure the heigths of the bars I get 99mm for work done by A2 and 95.5 for A1

and

60mm for coal burnt by A2 and 54 by A1.

A2 gave 1,65 unit work per unit coal and A1 made 1,77 unit work per unit coal or an advantage of 7%.
The specific water consumptions are within 1% identical.
That means that the boiler of the A1 was 7 % more efficient than the A2 boiler and that is not a lot.
Spread within a single class of well maitaned locomotives could be just as great.

 

So the A2 had a 200psi boiler in service? Not 220?
The design was not perfected of course, The famous loco exchanges and his adoption of travel on the Castle shows his design could be modified, and the A2 didn't get a chance. The main issue for these and indeed the B16 was excess wear with all pistons attached to 1 axle iirc. Also not liked by drivers due to poor visibility down the massive parallel boiler. Guess fireman had their work cut out too.
So the Gresley rebuild was regarded as improving the B16, such a shame the A2 wasn't treated similarly and kept on through the war. So York shed was expanded to fit them in, and they didn't want them. Rubbish compared the Gresleys magnificent machines. Such reverence is not easily won over.

 

Even with a loco built as late as 71000 designers didn't always get things quite right, either in the design or during the construction. Or the Manors, yet another seemingly run of the mill GWR 4-6-0, but one that wasn't really sorted out for another 15 years. So even with decades of experience, a couple of things very slightly out could massively affect the ability and usefulness of a loco.

 

And Gresley had the rebuilt the Raven Pacific with Ravens massive boiler now improved and his monobloc engine, perhaps might just have created a frankenmonster more able than his own.imagine ex North Eastern drivers beating his own records in mirth.
As it was, when new boilers were needed, he kept Ravens engine and plonked ona Gresley boiler. It gained none of the grace but lost its vanity.

 

According to the LNER Encyclopedia, the design of the Raven A2 was rushed. It was a greatly stretched version of the Z-class (C7) Atlantic, retaining the valve arrangements and steam passages of the Atlantic. So although both the grate size and tractive effort of the A2 were 50% greater than the Z/C7, it is unlikely that it would be anywhere close to 50% more powerful.

https://www.lner.info/locos/A/a2.php

The Gresley Pacifics were also sub-optimal in their original form but would benefit from 15 years of further development and refinement. There would doubtless have been potential to also improve the Raven Pacifics. If more had been built, then perhaps Gresley and Thompson would have looked to rebuild the front-end on the lines of the B16/2 and B16/3. But the design effort was likely not justifiable for a class of only 5 engines.

There have of course been many examples of larger engines failing to produce increased power proportional to their increased size. French Pacific designs were similarly disappointing until Chapelon arrived to sort them out.

But having raised the topics of the NER and Raven, we should acknowledge that they predated Gresley in the take-up of 3-cylinder propulsion. The NER around 1910 appears to have been the first railway in the world to start building 3-cylinder cylinder simple-expansion locos in significant numbers, including the successful Z-class Atlantic. (There were a handful of prior examples elsewhere, and of course 3-cylinder compounds had been around in a few places since Mr Webb in the 1880s). The Prussian State Railways would follow from 1914, starting with the S10.2 (DRG 17.2) 4-6-0, which had a Raven-like drive of all 3 cylinders onto the front coupled axle. Gresley would follow with his first 3-cylinder 2-8-0 in 1918. It will be interesting to see if Simon finds any material on how Gresley came to favour 3-cylinder drive and whether or not he was influenced by the experience of others.

Raven favoured unified drive onto the first couple axle while Gresley favoured unified drive onto the second coupled axle, this preference being shared by Bulleid whereas Thompson and Peppercorn opted for divided drive. And looking across the 3-cylinder designs of other railways in Britain and Overseas, you again find all these 3 options. I don't think there is any conclusive evidence that any one solution was the best choice - it seems very much designer's preference.

 

That is a good spot and I will correct this now.

Edit: Table as edited - which shows the Raven machine was always within 1% of its designed for boiler pressure.

 

One thing that occurs to me though. I've heard a suggestion (correct me if this is wrong) that with pop type safety valves best practice is to keep the pressure just below the red line because having the valves blow off wastes a lot of steam, whereas with GWR direct loaded safety valves good practice is to keep the pressure right up to the line. Could it be the case that there's a similar situation here and a difference in best practice is the reason for the lower average pressure rather than shortcomings in design? I'm just throwing this up as a hypothesis, I have little idea what differences existed between the two types.

 

I did say "should".
On a different point: When the Raven A2s were given A1 boilers, did that mean 180 psi? That wouldn't have done wonders for their performance.

 

That (comparison of A1 and A2) is more or less what Locomotives of the LNER Part 2A says. The RCTS has re-issued this series, which is also available as pdfs, and for reading on line for free - I don't think you have to be a member to register for access. https://archive.rcts.org.uk/

 

A very good shout, for which I have registered this morning. I do have a full set of the RCTS books but online access makes it easier to rummage through the books between my two screens.

 

So City of Newcastle had the safety valves on much of the trial then? Curiouser and curiouser.
Wonder if that's why they fell out.
Darling and Donnie are a parody of Crewe and Derby. Both still doing their own thing, Darlington belching out J39, D49, and B1, perhaps. And a 1500volt DC Mainline, somewhere. Really the 4-6-0 was Gresleys achieves heel, without a wide firebox, B1, B12, B16 and dare I say it Some of the GCRs were all more than a match for a B17! However by straying g from Britosh conventions he trumped the lot with a V2

 

say

As you say, there are two schools of thought on the matter. With spring balance valves such a Ramsbottom and GWR ones, they don't suddenly open and as the red line is reached they will start to feather. This allows you to run at full pressure with a slight waste of steam. With a pop valve, once the red line is reached the valve will fully open and a lot of steam is lost until the pressure is reduced sufficiently to cause it to close. Because of this, firemen will generally run with a pressure slightly below the red line to avoid blowing off at all. Both schools of thought have their followers and I don't think there is a right answer. If the boiler was a modern automatically controlled steam plant the controls would be set slightly below the red line and the safety valves would be there just in case and not used as a control valve, which is what those who advocate having the feather showing are using them as. With a pop valve, the firemen is acting as the control mechanism with the valves being the override control. Personally, I think there is no noticeable difference between running a loco at (say) 200psi on the red line and running it at 195psi to keep below the red line so I'm inclined to the pop valve as being the better system.

 

A question occurs. As always, apologies in advance if it's a daft one ....

Does the sudden action of the 'pop' safety valve place any more stress on the boiler barrel (or firebox outer wrapper) than the Ramsbottom variety, or would the design of the mounting po#ition account for any such factor?

 

Britain did have an example of a 4-6-0 with a wide firebox - but it didn't work on the LNER.

https://en.wikipedia.org/wiki/GWR_2602_Class

 

Gresley had more narrow firebox designs to his name than wide firebox ones. Horses for courses.

 

Whilst I agree that locomotive design was “pretty hit and miss”” in terms of efficient design ( Swindon was,
within the UK, ahead of the game until probably the mid twenties ); most designs actually produced
sufficient power to carry out the prescribed design duties , albeit often at the expense of a very high
coal consumption.

The E10s regularly hauled the 17.15 Bournemouth Waterloo in 1914 ( in previous years motive power
had been L12 and D15 4-4-0s). There are ( ex Railway and Travel Monthly and some private papers of
H.Brindley ) some performance logs. The E10s kept time and indeed from the few runs available
performed better than the L12 and D15s.

It would appear that footplate crews had mastered the eccentricities of the 4-2-2-0s, although on one of
the four occasions there was bad slipping restarting from a Clapham Jct. signal stop. However there was
probably a significant cost in fuel. There are figures from 1911, based on MPD information, that suggest
E10s were using 30% more fuel than T9 4-4-0s.

Ref. p.92/3 20th Century Southern Steam Vol 1 ( can be viewed on the Railway Performane Society website)

Michael Rowe

 

Bearing in mind the improvements Ell was able to make to Ivatt, Hawksworth and Collett designs, the fragility of Gresley big ends and the vicissitudes of SR Bulleids I think all of the design schools were still well capable of misses well into the 40s...