Edward Thompson: Wartime C.M.E. Discussion

Absolutely - I have been very lucky to have had Nat Pres here and for the last six years to have the debate and analysis. Once things are better financially, I intend to make a donation to its running costs as my way of saying thank you.

I will also make available an e-copy of the book free to anyone who has taken part in this thread on here - that's a promise.

Draw your own conclusions. On average, Thane of Fife as an A2/2 was in use 100 days more than the P2s were in the same year. 20,000 more miles too.

The reason I am spending a lot of time looking at the locomotive side of things is that that the locomotive side of things has been the stick with which Thompson's reputation has been beaten, repeatedly. A number of questions thus emerge:

• Are his designs as bad as they are made out? If not - why have they been reported thusly?
• If they are - why are they?
• Was he really trying to "destroy all that was Gresley?"

Etc etc. You get the point.

For me, if he was trying to "destroy all that gresley" that would imply there were no problems with the Gresley designs during the second world war. That is demonstrably not the case. To what extent they have been reported thusly has until recently not really been examined.

Mainly because most LNER historians and enthusiasts take the line that Thompson was lying in order to make his mark on the LNER and because he had a grudge against Gresley.

For some time I've been beaten with a stick for speaking up and asking the questions. The more I delve through this source material, the more I am convinced we as railway writers and enthusiasts have got it wrong - and badly - over Thompson.

But we have also misunderstood the circumstances of Gresley prior to his death and I think it would not be unfair to say that Gresley and Thompson, and their relationship with one another, is much misunderstood too.

 

I seem to recall even Norman McKillop '(Enginemen Eite' etc.) had little time for the P2's before or after rebuilding, and he was a Gresley fan who might have been expected to defend the originals if there was much to be said for them.

 

Hi Tom,

I have given this a go, but my figures are different. Here's the table:



"Total engine days available" is (total number of engines) x (365 days) which we will call T.

Number of days off for repair total we will call R.

The "% Availability" we will call A.

So, if I am reading you right, R / T = A

Where we are using one column divided by another and then the % button in excel to give the values.

Which gives us a % figure of a locomotive classes' availability for 365 days in a year based on those figures. In my table, this has been calculated by area, so the figure we are actually seeing is the availability of that class for that area of the LNER.

So this is for 1942 - see what you think.

 

Interestingly - it looks like there are some missing days somewhere in your figures.

For example, to take the first row - 21 locos @ 365 days per year = 7665 days.

Of those:
- Repairs in the shops: 721
- Repairs on shed: 1277
=====
1998

- Available in traffic: 4517
- Available but not in use: 16
=====
4533

That gives only 6531 days,

Essentially, you have:
- Under repair: 1998 (26%)
- Available: 4533 (59%)
- "Missing": 1134 (15%)

I'm wondering where those days have gone (other locos seem to show something similar). I suppose it is possible that certain shed activity could be not counted as a repair but nonetheless make a loco unavailable - the main thing I can think of would be washouts. 1134 days for 21 locos is 53 days per loco, which would imply one day per week per loco has gone missing - anyone know what the washout schedule was?

As has been hinted above, the more you look at the figures, the more questions it raises! Given the stated aim of the GWR and LMS to have 85% availability, it would be useful to know how they calculated that figure, so you can at least compare like with like. Failing that, accounting for the discrepancy would be interesting - I wonder if there are regional or class patterns?

Tom

 

Ah ha! I think I know where the missing days are. Here's how the LNER broke down their figures:



So a "year" of work for England would be 311 and it would be 312 for Scotland. So I guess the question is - should we be multiplying by 365 or 311/312?

Interesting stuff this!

 

So what I observe is

All the numbers seem to be based on a 312 day year ( ie a six day week). This does not affect the relativities between loco types but should be noted. Edit, sorry, wrote this before seeing the two posts just above.

In 1942 the miles per available day for the P2s is 10-15% below the A1s and A3s. Maybe this could be put down to differences in rostering, work available.

The big difference in 1942 for the P2s is shop days per loco--double any other class. Then in 1943 the days in shops and stopped at sheds (especially the latter) look dreadful.

So you are dealing with something real. Thinking of Hardy's books, he often stresses the need for locos to do, say, 50,000 between shop visits (I'm thinking of the B17s here), comparable with your planned mileages between services column. The P2s were not achieving anywhere near their target in 1942/3. Why?

 

Okay so I have made a quick change to the above table - now the stats for the English areas are multiplied by 311 and the Scottish ones by 312. This makes the availability percentages rise a little.





1942 and 1943 for comparison, one below the other.

Notable that Thane of Fife has gone from being in the worst class for availability to almost the best.

 

Sneaky! I wonder why they did that, and also how universal a method of accounting that was? Presumably the LNER ran some services on Sundays?

Tom

 

So in one of the national archive files, Thompson had this to say:

(on a side note, it is incredibly humbling to read Thompson's own letters and first hand accounts on this and other matters).

 

Found this in Enginemen Elite, the only book of his I have.

(discussing being sent from Haymarket to Doncaster to pick up Cock o' the North)

'...…...This Mikado would be so unique that it would attain standards never before achieved. So I thought.

But as everyone knows, those hopes were never fulfilled, and I took a hand in condemning this engine, for I not only brought her to Scotland, but tested her out in the initial stages on the first stretch of the Edinburgh -Aberdeen run. The Mikados were one of Gresley's 'near misses' and I suppose there was no more disappointed engineman in Britain than myself that this was so...…..I learned to dislike the breed through much experience of them.'

But he does say in the next chapter

'The fact was that the P2s, while not popular with enginemen for various reasons, were capable of hauling enormous trains between Aberdeen and Edinburgh...….'

So just looking at that data and Thompson's letter, I wonder if any of that 126 days stopped on shed in 1943 was a slight bending of the truth. No-one wanted to admit they had modern locos available but not in use. There's a war on.....

I would love to see data like that for MNs, WCs, LNs and Arthurs in the early 50s.

 

I don't think the issue has ever been their ability to do a job. We need to stop conflating the two issues. It is perfectly reasonable to state that the Gresley P2s were:

1) Fully capable of taking incredibly heavy trains
2) Were also mechanically unreliable for a variety of reasons

The two issues above can both be true. In fact I would wager that both statements are entirely true. These mileage figures go some way to explaining the steps Thompson and his team took with the P2s, I feel.

And for the record, I don't believe there is a bending of the figures. If there was, why would you make the figures for the P2s worse if you wanted to keep them in service? That doesn't make any sense. You'd massage them the other way!

 

Agree completely on the first point, no intention to do other than quote Toram Beg (not my favourite author but he was there at 64B at the time).

Not intending to imply any bias in the comparison between the P2 and other classes. But actually the extremely low numbers in the days available but not used column for all classes plus the letter comparing LMS availability with LNER availability data ( also referred to in McKillop's book by the way) and referring specifically to how undesirable it was to say something was not available when actually it was might be an indicator that that particular column should be read with a pinch of salt.

 

Just thought of one point re the 1942/3 P2 data. When was Thane of Fife rebuilt and how long did that take? I'm wondering if that accounts for some of the high shop time--- for example if the rebuild took six months.

 

The discrepancy between England and Scotland intrigues me - presumably in England Christmas was also being discounted. But why the discrepancy?

 

Simon, that's a very kind offer. In return, I will make a donation to a charity/loco project of your choice.

Mark

 

That is a very kind offer Mark, thank you.

I will start to put together a list of people from this thread who I will happily PM with a PDF of the book when complete (we're nearly there I think - aiming for end of September), and if anyone else does feel like chucking a few pennies this way, here are my chosen charities:

The Railway Children Charity
The Bluebell Railway
The Brighton Atlantic Fund
A1 Trust (60163 Tornado)
P2 Trust (2007 Prince of Wales)
M&GNJR Society (loco 8572)

Absolutely no obligation to donate - I am happy to furnish people with PDFs of the book who have taken part. I feel that is fair.

 

My apologies ladies and gents, I fear I have spammed this thread recently, but I beg your forgiveness one more time, for I have completed the tables for 1942, 43, 44, 45 and 1946 as promised.

For the % availability column, I have used the formula suggested by Tom.

"Total engine days available" is (total number of engines) x (No. of days in an "LNER Year") which we will call T.

Number of days off for repair total we will call R.

The "% Availability" we will call A.

Therefore R / T = A

In this case, we are using one column divided by another in the excel spreadsheet and then the % button to give the values in %.

I have used 311 for all of the English regions and 312 for the Scottish regions where we require "T". This seems fairest.

I think it makes for interesting reading. See what you think.

 

Nice work Simon - I'll sit down and have a look through all this over the next few evenings!

Richard.

 

That seems to confound a lot of what we believed. A4 Scottish figures for example.....

 

I would also just like to point out that the single locomotive in the North Eastern region in 1946 that achieved 93% availability...

...was 505 Thane of Fife whilst working out of Neville Hill.