Edward Thompson: Wartime C.M.E. Discussion

I will need to check this, but my recollection is that the B16s continued to have poor availability throughout the war.

I can't comment on the coal consumption figures - I don't have that data to hand. Would be interesting to see though!

 

The B16s are interesting .
Very long lived .
Quite some rebuilt after Chapelon by Gressley,and even more by Thompson.
Who banged most per buck?
Raven,Gressley or Thompson?
I for one would like to read the Stannier/Cox paper

 

Thank you for such a detailed response, that has certainly enlightened me. But it has thrown up another couple questions.

The P2 certainly looks like it turned out to be a shocker, but it was designed to take the trains over the tricky Aberdeen Edinburgh line, so did the trains get shorter or the timetable longer when locos of less adhesion were drafted in to replace them? Or were the rebuilds truly up to the job, the A4s aswell? Are we saying that the A4s were capable of doing the same work as the P2? This is where my initial misunderstanding lies?

And secondly, "2. The Scottish based locomotive classes all consistently achieve better mileages between general overhauls than their English based counterparts..." what are your thoughts around that? Were the Scottish shops run better/better engineering departments? Were the Scottish drivers less whip handed? Is there a material difference in general maintenance in Scotland? One would have been inclined to think the Scottish locos would have been in the shops more often...gradients/curves/the weather...

 

I will send you a PM shortly so I can send you a copy of the document via email.

Yes - the W1, A2/2 and A4 classes were all fully capable of doing the same work the P2s were doing. The difference was in how far they were pushed to their limits. The lower rate of adhesion on the A2/2 in particular is noted. But then, all of those Pacifics had lower rates of adhesion than the P2s. All proved competent, if not outstanding.

The P2s were masters of their work, make no mistake. Their eight coupled driving wheels and double kylchap exhausts together with the bigger boilers and fireboxes gave us a very potent machine. It's why I bought into the P2 Trust's founder members club. The design has exceptional potential for high performance.

The issue isn't, however, the performance when doing the work: it's how reliably said loco class can be rostered to do the work.

I honestly don't have an answer for this. All I know is that I have the mileages for the loco classes in front of me, and for every year, without fail, where there were multiple groups of the same classes across the LNER, invariably during the war years the Scottish locos did better than their English counterparts.

 

Fair. It really is going to be something else seeing 2007 in full flight!

Any objections to me just chalking that up to world renowned Scottish engineering then ? (Not that I'm biased much!)

 

How much do we know about the issues which were causing the P2s to cover less mileage between overhauls? We know about their propensity to break the A3 type driving axles, problems with the pony truck and hot bearings, but were there other factors? By the time the above tables were produced they would all have been equipped with the A4 front end and Walschaerts valve gear.

 

This is the thing though. If we cross reference that with the 1937 document I have on intended mileages for these classes, the Scottish ones are always lower targets. Presumably they took into account mileages for the actual journeys, maintenance, costs, etc, when deciding on these figures.

But throughout the war the English based classes never achieve very high availability and the Scottish ones always achieve a higher percentage of their own targets (which is almost always a physically higher mileage than their English counterparts too).

I am making some assumptions there based on the raw data and my initial observations, so don't be surprised if, once I've put all the data into the spreadsheet, that view changes. My views have to change as the facts change.

 

Were there other factors? I don't know, but I have a very particular hunch. If you look at the A2/2s, they were the P2s fitted with front bogies and one set of driving wheels removed, with one set of walschaerts gear added. Virtually identical in all other respects in terms of setup.

But the A2/2s suffered in later years with long waits out of use due to their boilers being overhauled. There were no spare boilers for the A2/2s, which in later years (and before the decision was made to fit them with Peppercorn boilers) were always having to wait for one of their pool of six boilers to become available or be swapped from a classmate.

I have no evidence to suggest this - but given the experience after rebuilding as Pacifics, why would it not also be the case before their rebuilding, knowing that they kept to their own small pool of boilers? We are talking about the same components.

 

Better water?
I am looking forward to read the Cox paper.
Thank You.

 

According to Cox Scots based engines on the LMS achieved much better figures due to far better water and thus boiler life. Whether that was a factor on the LNER I have no idea.

 

I don't remember the correct word (no doubt it will spring into my head next week!) but railway workers were exempted from national service .(reserved occupation??) although they could volunteer .

is there any evidence to suggest that the Scottish area retained more skilled men in the workforce during hostilities

 

A relative factor perchance is that the Scottish ECML depots usually allocated a locomotive to specific crews (i.e. one crew one engine) who were thus able to monitor the locomotive performances - including identifying the little things that could be fixed by fitters overnight before they became big things requiring withdrawal from service or even workshops attention. I believe I posted earlier that such attention at Haymarket led to the official maintenance regime being changed at the local level thus improving reliability of the Gresley conjugated valve gear and reducing the casualty rate of Gresley locomotives by a considerable margin.

 

No problem. PM coming in now.

Where did he say that, out of interest? Would be interesting to read.

No - as it was only certain disciplines that were a reserved occupation on the railways. The LNER lost great swathes of its maintenance staff to the war effort.

My apologies Fred, you have said this before and I forgot.

I have come across documents regarding changes of maintenance regime on the LNER and will type them up when I can for your perusal.

 

Very interesting. So, other than the known design weaknesses that were never developed out, perhaps the P2s were better locomotives than is generally thought. The excessive downtime may have been due to simply a shortage of spare parts, something which could affect any small class that was perceived as a design dead-end.

In a round about way, Thompson and A1SLT seem to be addressing the same weaknesses of the P2 but with different solutions. Thompson solved the problem of the over-stressed crank axle by removing a drive axle, thus reducing the torque on the crank; A1SLT designed a stronger axle. Thompson replaced the problematic pony truck with a bogie; A1SLT redesigned the pony truck. Perhaps the difference in approach is partly due to the restrictions of wartime conditions.

 

On the contrary, I think the mileages show that the P2s were somewhat worse than is otherwise described, particularly under wartime constraints.

Many critics of Thompson have claimed that he exaggerated the problems of the P2s so that they could be rebuilt. Here are the facts:

1. Thompson was aware of the availability issues across the LNER and investigated this
2. Thompson obtained an independent report on the Gresley valve gear, which was proving troublesome in wartime
3. He carried out a limited rebuilding with a single prototype, as per the report's recommendation (rebuild a locomotive with one additional set of valve gear)
4. This prototype (Thane of Fife) was on test for a year, its mileage substantially improved, the P2s annual mileage fell again
5. The LNER emergency board granted permission to rebuild the rest

But what "shortage of spares" are these? Almost everything on the P2s is standard with another LNER class in some form, besides the unique boilers. Lets say we accept the small pool of boilers is a potential factor: the W1 had the exact same problem and it was achieving double or better mileage than the P2s.

Precisely. And that I have said this for a number of years now.

Welcome to the "dark side" (!!!)

 

I think that is certainly a factor. The quality of water up here is better suited to steam locos.
We have never once had to de-scale a kettle. It just doesn't happen up here in granite country.

I wonder if Haymarket/Ferryhill et all had more autonomy to conduct intermittent maintenance, being so far from Darlington/Doncaster. Then again, Inverurie wasn't far away...

 

Simon - not sure if I am mis-interpreting your data, but mileage between overhauls seems more significant to me than %age of intended mileage run per year.

As an example, to take the first row of your table - class A1 in the SAW area: 21 locos ran a cumulative 951,865 miles, which is 45,327 per loco for that year. You've expressed that as running 57% of the target 80,000 miles between overhauls. But assuming the locos actually were able to manage 80,000 each, it's just another way of saying that locos should last 21 months between overhauls. (80,000/45,327 = 1.75 years).

In other words, you have compared the annual mileage run not as a percentage of the target annual mileage, but as a percentage of the target between overhauls at unspecified intervals.

Yearly mileage will always vary depending on the traffic available. The traffic department presumably had a target for each loco, from which they determine how many locos of each type they needed available.

So the more germane figures are:

• %age availability (because low availability means you need more locos to cover the duties, i.e. they are spending more time in the shops and less in traffic, and more locos required is an excess capital cost)
• actual mileage between overhauls (because broadly one overhaul will cost the same as another, so low mileage means high per mile cost, i.e. excess revenue cost).

I think you can get the %age availability from your data; for example, taking the first row of the table: 21 locos * 365 days = 7665 engine days available; under repair 1998 days, so availability = 74%.

To see if there were real problems, I'd be looking at mileage between overhauls and availability, across classes or across geographic areas. It might then be interesting to compare (if the data are available) with, say, GWR King, SR MN and LMS Duchess - which might flag up whether the LNER in particular had issues with availability. (I seem to recall figures nearer 85% for GWR availability as an example - @Jimc might confirm - but don't know if that is comparable with regard e.g. wartime conditions).

Tom

 

Tom - thank you. A potential big error in interpreting the data avoided!

I’m travelling home right now but when I get a chance I’ll fire up the spreadsheet and see if I can put together some examples like that you suggest above.

 

for example Chronicles of Steam pp161-163, talking about water treatment "The objective was roughly to raise English water to the quality of Scottish water... it had been estimated there would be a net saving of £134,000 per annum" and p 108 "engines... in Scotland ran considerably further between boiler changes... The constantly good quality of the available water was the ruling factor here."

85% availability was a target on both GWR and LMS.

 

Something else that would also need factoring in is that during the war there was a general speed limit of, I believe, 60 mph. That slowing down of trains, plus their being generally heavier, would particularly affect the Pacifics and V2s in that any train that they hauled would take longer to cover a given distance compared with pre-war times and speeds. This inevitably would result in a loco covering less distance per day/month/year than hitherto. This situation I think, would affect ECML trains rather more than those between Edinburgh and Aberdeen which were hardly 'high speed' ones given conditions on that route and could well account for at least some of the lesser mileage of Southern Area Locos.