If you register, you can do a lot more. And become an active part of our growing community. You'll have access to hidden forums, and enjoy the ability of replying and starting conversations.

Edward Thompson: Wartime C.M.E. Discussion 2012 - 2019

Discussion in 'Steam Traction' started by S.A.C. Martin, May 2, 2012.

  1. Jimc

    Jimc Well-Known Member

    Joined:
    Sep 8, 2005
    Messages:
    2,878
    Likes Received:
    2,305
    Occupation:
    Once computers, now part time writer I suppose.
    Location:
    SE England
    Heritage Railway Volunteer:
    No I do not currently volunteer
    Trouble is if you want to make a living from your writing there's only so much research you can do, and serious stats bashing takes a particular mind set and skill set as well as a lot of time. Much easier now since the spreadsheet was invented... I don't know about you, but if I calculated the hourly rate from my book it would probably be in pennies.

    I do find myself drawn to some of the numbers stuff, but when I've gone into it I really don't have the expertise, and some of it is so arcane... I found, for instance, a GWR repairs and partial renewals register at Kew, and thought that looked interesting, but when I got it out it was just endless weekly tables of expenses against each factory with no mention of what was being repaired and renewed.
    For 1936, and the GWR, the percentages were:
    Salaries, 2.9%
    Office Expenses 0.3%
    Wages 46.3%
    Materials 32.5%
    Machinery and Plant 5.2%
    Other Expenses 12.75%
    It might be interesting to compare those numbers against a modern enterprise, but there's no book in it...
     
    Last edited: Aug 27, 2019
    S.A.C. Martin likes this.
  2. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    The spreadsheet work has taken up a lot of time, much more than I was anticipating. I keep the folder with all the numbers in it, in my work bag so I can work on it when I have spare moment or need something else to work on when I'm trying to work out a problem in my head. Since going to Kew in August 2018 I have managed to do a huge amount more research and reading. It's the reading and bibliography writing that's the biggest occupier of time, moreso than the actual writing on Thompson.

    Thank you all by the way for your thoughts on the stats. I intend to give the raw data in an updated, modern format for reading (but of course keeping the accuracy of the figures and adding relevant notes to assist reading).
     
  3. jnc

    jnc Member

    Joined:
    Apr 3, 2012
    Messages:
    821
    Likes Received:
    904
    Gender:
    Male
    Location:
    Western Atlantic
    Heritage Railway Volunteer:
    No I do not currently volunteer
    Speaking of 'means' about classes, the same sort of math can be used for fleet-wide numbers there, too. E.g. if you have the mean availability of each class (as opposed to averages), if you want to get an idea of the overall mean, don't just simply add the per-class means and then average them, as that will again over-emphasize the small classes. First, multiply each mean by the size of the class.

    (Not that I ever subjected myself to a statistics class, that kind of math didn't mesh well with my brain, I was more into things like graph theory. So this is all from first principles. But the basic idea is the same - to prevent the small classes from having an undue impact on fleet-wide numbers.)
    “I say, I say, don’t bother me with the facts son, I’ve already made up my mind.” -- Foghorn Leghorn
    :)

    Noel
     
    ross and S.A.C. Martin like this.
  4. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    Looking at the archives in York is going to be interesting on Saturday. It might make or break my thesis!

    But whatever happens (supporting my views or not) I’ll share the results.

    The mad thing is how big this whole Thompson project has become. I never expected to be typing down hundreds of numbers into spreadsheets or making graphs, or looking through archive material like I am doing. It’s now so much bigger than the book.
     
    jnc likes this.
  5. 30567

    30567 Well-Known Member Friend

    Joined:
    Aug 7, 2012
    Messages:
    2,804
    Likes Received:
    1,515
    I think you have gathered the raw material at the most disaggregate level possible. But then, depending on the question you want to investigate, you need various levels of aggregation. For example, supposing you were looking at the 0-6-0 fleet, a question might be whether the GE, GN or NE designs performed better , or whether later designs were any better than earlier. That way, a couple of D17s would be for most practical purposes in with the rest in their category.

    Gathering all this data has been a labour of love, but it should not distract from defining the interesting questions and tailoring the analysis to fit.
     
    69530, jnc and S.A.C. Martin like this.
  6. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    I had to look up "disaggregate" - everyday a school day! Thank you for your thoughts.

    I take your point on the "which 0-6-0s are better" comparisons but I don't think that's the point of my research, nor do I feel inclined to get into that particularly.

    Thompson's decision making in wartime has always been undermined, underplayed, questioned. The general consensus from LNER writers was that his standardisation plans were not necessary, and that any issues he claimed existed, did not exist or were not as bad as described.

    The overall point I hope to make with my research is that the issues were real, both Gresley and Thompson sought to resolve them, and that the LNER's time in the war was in reality incredibly fraught, difficult, and there was not much room for manoeuvre.

    The only comparison in terms of locomotives I am likely to make directly is where the pacifics are concerned, as they are the most controversial of Thompson's work (and, IMO, shouldn't be that controversial if at all, when you look at the LNER's wartime situation in the round).

    That is not to say that someone else couldn't do this - they are welcome to, as I will be releasing the full spreadsheet of data when I have completed it.
     
    jnc likes this.
  7. 30567

    30567 Well-Known Member Friend

    Joined:
    Aug 7, 2012
    Messages:
    2,804
    Likes Received:
    1,515
    OK fair enough. Actually the best comparators for the B1s and L1s are probably on the LMS and not in your dataset anyway. I was thinking more of the O1s and B16/3s where it might be possible to assess whether the rebuilds were an improvement in availability and maintenance terms.

    Isn't it sometimes argued that Thompson's contribution in these rather unglamorous areas is unjustly overshadowed?
     
  8. jnc

    jnc Member

    Joined:
    Apr 3, 2012
    Messages:
    821
    Likes Received:
    904
    Gender:
    Male
    Location:
    Western Atlantic
    Heritage Railway Volunteer:
    No I do not currently volunteer
    Right, but by making all this dis-aggregated data available in electronic form, you'll be making it much easier for those who are interested in other questions (such as the one above) to investigate them using actual data - rather than relying on opinion, or received wisdom - which, as you've discovered, can be quite wrong. A lot of work, to read and enter it all, but incredibly valuable.

    Noel
     
    Forestpines and S.A.C. Martin like this.
  9. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    I would go further and would argue that the vast number of LNER supporters who will only talk about Gresley or Peppercorn as CMEs, but never Thompson, do all three CMEs and their own interests an injustice.

    Edward Thompson IS the LNER as much as Gresley is, and as much as Peppercorn is. He achieved more as CME in five years than many CMEs on other railways did in 25 years. That he also achieved much during a very bloody wartime period says a lot about the man too.

    Thompson hasn't just been overshadowed: at times he has been at worst erased from history or dismissed out of hand at best. On almost every level there has been a failure by LNER writers to ask if what they were writing was accurate.

    What has been most damaging to Thompson's reputation has been written by a select few people. It goes like this:

    Thompson hated Gresley > wanted to rid LNER of Gresley > had a go > all designs bar B1 are failures

    It's always loco-centric, doesn't ever mention the Cox report, or look at wartime availability, or the staff levels, or the materials shortages, or...etc.

    It only ever focuses on a very person specific, very one sided and very personal account which claims Thompson based pretty much all of his decision making from a position of hatred of Gresley.

    The problem is, the more I delve into this, and the more I find, the more I see - and others will too - that the problems Gresley in 1939-40 and Thompson and the LNER faced from 1941 to 1945 were very, very real.

    Placing so much undue emphasis on percieved personal views and not looking at the evidence is the legacy of many LNER writers, in my opinion. Which is why I reply so robustly these days on occasion. It's infuriating to read the same things over and over again when you know they don't come from a position of doing the research. I have.

    Maybe I get too passionate about it sometimes. Maybe I do see it increasingly as an injustice on Thompson's side. But it is never about making Gresley look worse - as has been intimated - to make Thompson look better. Not at all. Both CMEs come out looking better the more I get into it.

    In Gresley's case, looking at the board minutes in particular, I feel increasingly sad that Gresley's last years were so tough for him. He undoubtedly suffered while also trying to do his best for the railway he so clearly loved. Thompson too suffered, in different ways.

    If my research does one thing, it will be to show that Gresley, Thompson and Peppercorn, whatever their personal views of one another were, were undoubtedly professionals in their work and their only interest was in doing their best for the LNER and its workforce.

    That's very true and I will be happy to make it available. To that end, here's an updated dashboard showing the work I have been doing on the spreadsheet. One year is complete and I am working my way through the rest!

    upload_2019-8-28_11-16-21.png

    Boiler types I have determined by only counting the diagram numbers, and not variations within diagrams (such as dome/safety valve types where you might argue it's a different boiler type). It looks on paper like Thompson reduced the number of boiler types but not by much overall.

    If his standardisation scheme had been done in full, it would have been 19 classes with 25 boiler types comprising around 7000 locomotives. Compare that to 1942 - extraordinary.
     
    60525, Forestpines, ragl and 2 others like this.
  10. jnc

    jnc Member

    Joined:
    Apr 3, 2012
    Messages:
    821
    Likes Received:
    904
    Gender:
    Male
    Location:
    Western Atlantic
    Heritage Railway Volunteer:
    No I do not currently volunteer
    It will be interesting to see how much impact your work has. Don't be too disappointed if the first result are muted, it may just be the Semmelweis effect at work. Rest assured that to the extent you're drawing on demonstrably real data, eventually the tide will turn. (It is not without reason that Planck said 'science advances one funeral at a time'!)
    Is the 'average availability' calculated in the old style (simple average across class availability)?

    Noel
     
  11. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    We will see most certainly.

    Yes - will need modifying. Your ideas superior, for sure.
     
    jnc likes this.
  12. Forestpines

    Forestpines Well-Known Member

    Joined:
    Jun 5, 2009
    Messages:
    1,414
    Likes Received:
    1,791
    Location:
    Bristol
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    Has any railway had the topic considered to this depth before at all?
     
    S.A.C. Martin likes this.
  13. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    I like to think this may be the most in depth discussion on Thompson railway enthusiasts have ever had. Hopefully most constructive too!

    So, on a different note. I took a trip to York yesterday and examined some engine cards.

    So.

    Engine cards in York.

    Availability statistics in Kew.

    Yeadon's Register.

    RCTS Green books.

    All of these sources give us information on locomotives owned by the LNER. The statistics we are interested in are dates to works, how long in works, how long actually being used, how long on shed, and mileage.

    Engine cards for the LNER come in several forms, dependent on loco class, which works was responsible for their repairs, and when they were built. Some locos have more than one engine card - up to seven, in fact - in various forms, dependent on age.

    I was privileged to read Flying Scotsman's, Mallard's, and W.P. Allen's at length. I photographed every single Pacific's engine cards yesterday. I've done some preliminary work on a new spreadsheet, which will give full details of every single Pacific locomotive from the LNER, from its build date to its scrapping based on the engine cards. We will be able to see their annual mileages, time in shops, their availability per year.

    We can then cross reference it with our other sources.

    Again, I intend to make this available to anyone who wants to look at the statistics - I beg some patience mind, it's a big undertaking. I am concentrating on the years 1939-1950 for the time being.

    My initial look at the cards available has confirmed one thing to me. The number of times into works is irrelevant. It's for what kind of work and for how long the locomotive is unavailable.

    And my current line of thinking is that the number of times in works for the Thompson Pacifics has been massively overplayed.
     
    60525, 69530, ragl and 1 other person like this.
  14. Jimc

    Jimc Well-Known Member

    Joined:
    Sep 8, 2005
    Messages:
    2,878
    Likes Received:
    2,305
    Occupation:
    Once computers, now part time writer I suppose.
    Location:
    SE England
    Heritage Railway Volunteer:
    No I do not currently volunteer
    You don't think that's an error in the opposite direction? From the point of view of the shedmaster isn't any time a locomotive is unexpectedly unavailable, even for a couple days, going to make life difficult? Whether a repair is scheduled or unscheduled will also make a lot of difference.
     
    jnc and Jamessquared like this.
  15. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    That is a fair point Jim - but it’s the scheduled versus unscheduled maintenance that is at the heart of this.

    Hence why ultimately the number of times in works is irrelevant, if already planned, if overall time spent in works is minimised.

    The goal is high availability for work and to actually do work.

    Overall productivity is the key stat for the bean counters.

    The interesting thing looking at the cards so far is that there’s an argument we are looking at two kinds of maintenance approach with Gresley and Thompson Pacifics.

    Run to failure, long periods in traffic, few overhauls but long overhaul times - Gresley.

    Preventative maintenance, more works visits, overall shorter times in works - Thompson.

    Peppercorn - we seem to have sublime and ridiculous (A1 compared to the first batch of A2s). Roller bearing A1s undoubtedly the best Pacifics on LNER if the aim is to reduce maintenance time and increase availability.

    It’s all fascinating stuff and I will share pictures of my spreadsheet work for discussion soon.

    One thing is clear: what we think we know and what is true are different - how different will depend on the figures.

    Obviously there will always be the boys own stories of one off high speed runs to point to whenever the debate about “what was best” comes up, but I feel strongly that the incremental development of the LNER Pacifics to their ultimate conclusion would not happened without Thompson.

    Obviously it’s not just about the Pacifics but they do form most of the controversy surrounding him.
     
  16. Jimc

    Jimc Well-Known Member

    Joined:
    Sep 8, 2005
    Messages:
    2,878
    Likes Received:
    2,305
    Occupation:
    Once computers, now part time writer I suppose.
    Location:
    SE England
    Heritage Railway Volunteer:
    No I do not currently volunteer
    Yes, there's a lot in that. Another approach, lauded by Cox in his book Chronicles of Steam, is how the LMS works at St Rollox under HG Ivatt managed locomotive condition so that minor shed repairs kept them running as long as possible between factory visits, and factory work was controlled so that everything was fixed up with just enough life to make it to the next scheduled repair. Cook on the GWR, on the other hand as he documents in his book Swindon Steam, seems to have been much keener on scheduling repairs, but also on intermediate repairs which would extend the distance to the next full repair.
     
    S.A.C. Martin likes this.
  17. Miff

    Miff Well-Known Member

    Joined:
    Jun 17, 2008
    Messages:
    2,050
    Likes Received:
    1,455
    Is there anything in the records which would enable a cost comparison? E.g. overall cost per mile (including construction & maintenance costs, not just fuel) for each comparable class e.g. all the pacifics. Since the LNER Co. was a business I wonder if this kind of exercise would have played a part in investment decisions. It was surely part of Thompson’s thinking in trying to reduce the number of classes and boiler types. I’m aware projected costs don’t always match reality (e.g. HS2 :) ) but you have to start with something.
     
    jnc likes this.
  18. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    Yes, I have construction (and conversion) costs for all of the Pacifics. If we agreed on costs adjusted for inflation, that could be of use. How much, I don't know, but it would make interesting debate.

    Thompson's desire to reduce the number of classes and boiler types was definitely a simplification exercise. Less types of loco, less overall different types of spares, more wide ranging availability and compatibility. GWR did it; I think Thompson thought the LNER could too - but his retirement and then British Railways formation mitigated against his long term plan.
     
    60525 and Miff like this.
  19. huochemi

    huochemi Member Friend

    Joined:
    May 6, 2008
    Messages:
    1,935
    Likes Received:
    844
    Gender:
    Male
    Location:
    UK
    I think you mean "militated against", "mitigated" has a different meaning (and does not need "against").;)

    To what extent was Peppercorn involved in fitting roller bearings to the last few A1s?
     
    S.A.C. Martin likes this.
  20. S.A.C. Martin

    S.A.C. Martin Part of the furniture

    Joined:
    Aug 31, 2010
    Messages:
    3,301
    Likes Received:
    4,051
    Gender:
    Male
    Occupation:
    Apprentice Rail Engineer, MNLPS Treasurer, Author
    Location:
    Sidcup, Kent
    Heritage Railway Volunteer:
    Yes I am an active volunteer
    Autocorrect! Will modify.

    Not sure, will check.
     

Share This Page