JimC

Dammit, why didn't I find that? I have a copy of that photo from RCTS, but much worse quality. [later] especially as you'd already posted it in this blog!

Not to be confused with the 0-4-2T 3521s that ended up as 4-4-0s, these were 2-4-0Ts that ended up as 2-4-0s. Built in 1885, the ten double framed 3511 2-4-0Ts were a tank engine version of the Stella class 2-4-0s, a part of a Dean standardisation exercise that also included the 2361 outside frame version of the Dean Goods and the 1661 0-6-0ST. All had major components in common. They were a much larger engine than the Metros with 17in by 26in cylinders and class P (Dean Goods) boilers. Originally (as shown) they carried condensing apparatus for use in the Severn Tunnel. 3501-10 were similar, but built as convertibles for the broad gauge. The 3511 series lost tanks and gained tenders to become additional members of the Stella class in 1894/5. This sketch was quite a challenge. I started it because I was given a couple of blueprints of weight diagrams of the class, but when I came to trace the blueprints the frames just didn't look right to me against photographs. I've decided, on *very* limited evidence, that the running plate on the drawings is a little too low, and so I lifted it and now I think it better matches photographs. Sketch updated 1st Feb

The GWR bought 100 and kept 50, which had the boilers considerably upgraded to GWR standards, including copper inner firebox. The others were run until they dropped, but I imagine they kept enough spare boilers to constitute a reasonable boiler pool. Hence there was no immediate need to put standard boilers on. Fitting Std 1 boilers to the RODs was apparently considered during WW2, but (doubtless cheap) Riddles 2-8-0s were bought after the war, and one may speculate this was partly at least to replace the now aging RODs. The GWR austerities and their changes in (G)WR ownership seem to be very poorly documented in GWR sources.

I'd kinda assumed that the Sodor railway had a sort of Manchester or Birmingham extension or at least running powers so that there was a daily out and back for Gordon's express train, the fish train and maybe a couple of other heavy trains.

Perhaps not the 3150: it has the larger diameter Std 4 boiler whereas the 5101s and the 61s had a Std 2, so the 3111s are, I suggest, the better bet.

For sure, but it also complicates admin, stock holding, distribution etc. Basically its producing a new model, albeit one with no development costs, for a run of 100. I don't know how the relative costs of body and chassis production line up, but it seems to me not utterly impossible that a genuinely profitable price for a chassis might get sufficiently close to that of a complete model that it would generate more bad publicity in complaints about price gouging than the income from sales would be worth. But I'm just speculating.

Trouble is the numbers game. Supposing a manufacturer decides to order some extra running chassis. How many could they sell? A hundred? And how many of those would be taking away sales from complete models?

For sure. And its hard to hack up or chuck that nice detailed body to use the chassis as a donor for something different that (unless you are sufficiently skilled not to have to bother with converting R2R anyway) probably won't look nearly as good. In ten or fifteen year's time of course... But that won't help sales now.

Ah a customer:-) That's what I like to see! A few 2021s were sold off as industrials, which maybe widens their appeal. If anyone here isn't aware, the 2021 is a 'small' type, ancestor to 54/64/74/16 and with ~4'1 wheels. One might also mention their cousins, the 850/1901s with similar variety but shorter wheelbase. There were 158 of those. The 1134s were a 'large' type, distantly ancestral to the 57s with ~4'7 wheels. The 1134s also came with different design/class/type boilers which varied in diameter, length and were higher and lower pitched, which changed their appearance subtly but would also make life 'interesting' for a R2R designer seeking a chassis that would fit as many variants as possible. Mind you a R2R 1134 with a chassis that could fit all its cousins would be a great thing for bashers, makers of 3d print bodies etc because there were almost unlimited numbers of variations, although the widest variety was back in the 19thC.

I've been given some GWR archive material, including a lot of locomotive diagrams, and to my surprise and delight they include weight diagrams of 342, 95/6 and 92. unfortunately the 342 drawing especially has suffered, but I may be able to work up some improvements on my drawings of these little oddities.

To be fair though the war ended a lot sooner than anyone expected which threw all the planning out.

Mmm. Well the ARLE committee was a committee. Hence it took considerable time to reach an agreement - indeed I believe it never did. But the Maunsell wasn't a mile from the consensus, and all the drawings already existed for it. Getting the committee to finalise a design and getting all the drawings prepared would surely have been some months more.

Well, they look pretty fair to me. Enjoy!

Gosh, what a coincidence. Not saying the cracks are anything but real, but its amazing how much more difficult it is to repair something you don't want there anyway...

Probably wrong to think of Swindon Works as a single unit in this context. AIUI the loco works used withdrawn engines with a bit of life left, and supposedly the welsh 0-6-2Ts were popular because nice roomy cabs. Presumably they ran them until something broke and then grabbed another. The 1366s (and presumably earlier equivalents) duties were in the wagon works. I haven't heard what was used in the carriage works.

A process which if taken to its logical extreme leaves us with a single topic of "stuff". I submit that this thread generates quite enough traffic already. Mea culpa, I shouldn't have started this diversion.

Large (4'7) panniers were very rarely chosen for auto fitting in the days before the dedicated autoclasses, and by and large those few that were were of the oldest and weakest classes. As noted 4'0 wheel classes esp 2021 were the preferred choice. Larger wheels were almost all 4 coupled types. The gear seemed to be added and removed with such frequency that I wonder if taken off when a loco was stripped down it was reinstalled on the next candidate going to the desired location. But as to why those apparent policies, I haven't found anything.

According to Holcroft it was in fact usage on Canadian lines that impressed him. Holcroft went on a trip to Eastern Canada/US border. On return he wrote up the trip for his management. Then Churchward gave him the job of outlining replacements for Atbara (large wheel 4-4-0) Bulldog (med wheel 4-4-0s) and Aberdare small wheel 2-6-0s which were to be inside cylinder with 10in piston valves over the cylinders. It perhaps should be noted that the County 4-4-0s were already in existence and a small boiler version tried and abandoned. Apparently this didn't prove practical, so Churchward gave orders to draft a medium wheel 2-6-0, which as we know was a huge success and indeed I think the GW never built another all new 4-4-0.

Cook gives us a pretty good insight into the design of the 48/58/54/64/74/16 family in Swindon Steam. Apparently there were a lot of common parts between them. Its a rare insight into the actual thinking behind design decisions. It's mildly amusing that they started with a clean sheet of paper and ended up with the same configurations as their predecessors 50 years before.

It is strange though. It's as if it all comes to a stop when Collett retires and nothing happens until 1952. No Hawksworth railcar for instance.

Which probably should have been replaced with railcars since having a Riddles Std 2 or for that matter a 4575 towing a coach or two with a dozen passengers in was a pretty wasteful exercise. They aren't things I've devoted much thinking time to but I wonder why there was a ~ 15year pause between the last of the GWR cars and the first BR?

With time and study, though, I'm moving to the opinion that the accountancy rebuilds, on the GWR at least, weren't nearly as dubious bits of financial gamesmanship as I used to think. Yes, it's recorded that the renewal fund was healthy and revenue was not, so putting a locomotive through a renewal saved the cost of the heavy General it would otherwise have required. In the event the end of steam truncated the life of all the 1930s renewals, so it's impossible to do a real analysis, but there's some evidence, looking at the large Prairies in particular, that the renewals would have still been in service long after their unaltered cousins had disappeared. So if a locomotive half way through its normal working life had a renewal costing half what a new build would cost, and the renewed locomotive went on to run a normal full life, and also had various small enhancements making it cheaper to run or more capable, then it seems to me the renewal would be sound business practice. But between the end of steam truncating locomotive lives, and the difficulty of extracting exact costs from surviving records - beyond me I fear - I'm not sure it's possible to make a definitive evaluation.

It wouldn't. Its a nice demonstration of why simply lengthening a 4-4-0 isn't a viable design route. The whole locomotive needs to be rethought.

Another what if is that according to Felix Pole's book the LNWR execs were agitating to have the Rhymney included in the LMS.

Scrapped at the first opportunity I expect!