JimC

The other option to consider firebox wise would be to shoehorn in a Duchess one. There's no real reason to use a narrow firebox on a Garratt. A King barrel with a duchess firebox is arguably what the Great Bear should have had.

That's not the half of it! It seems they had 4'9 and 4'10in lines, plus 5'4, 5'6 and 6ft as well... https://campus.fsu.edu/bbcswebdav/users/jcalhoun/Economic_Standards/Puffert%20-%20The%20Standardization%20of%20Track%20Gauge.pdf

That's not quite what I'm saying. Over the last 200 years many engineers have contributed to optimising the design of railway track. The vast majority of that work has not been directed at the rather abstruse question of what the laws of physics tell us about what the optimum gauge might be, but at the far more practical problem of how do we get the best possible performance out of 4'8.5 in track. The result is that a large amount of what we know about the laws of physics relating to track is really what the laws of physics are telling us about standard gauge track. Its a problem for the introduction of anything radical. Suppose I have a radical idea for a more efficient engine of some kind. Lets say that the current tech is theoretically capable of delivering 50% efficiency, and the current engineering implementation delivers 90% of that in practice. Total efficiency 45%. Now supposing I come up with a new concept theoretically capable of 60% efficiency. Sounds good. But that means my implementation of the new tech must deliver 75% of theoretical efficiency from day one to be even level with the old tech, even though it took successive engineers decades to get the old tech up to 90%. But this is just kite flying. Out in the world the best gauge is the one everyone else is using.

I think it likely, too, that the numbers game meant the majority of the best engineers round the world were working on standard gauge or something close, which meant the majority of the best innovations and ideas were optimised for that sort of infrastructure. If for some reason broad gauge had won, then every thing would be different. Not being a world class innovative engineer I can't speculate much on what those differences might have been, but I bet trackwork would look very different. With room for 36in cylinders between the frames compound steam locomotives might have worked better too.

On lengths, I reckon there was/is a horrendous cockup in CJ Freezers' 94xx drawing in Railway Modeller and reprinted in his book "Locomotives in Outline GWR" where the back of the locomotive is ~10 inches too short. I tried scratchbuilding one as a teenager and gave up when it didn't look right, but didn't figure out the error until I was making drawings for the below. I make the 2721 365", 6400 373", 8750 374" and 9400 398". (sorry about the inches, my electric sketches are scaled 1mm = 1 inch) The 54/64/74/16 series was a much lighter locomotive with a smaller boiler than the 2721s/57s with greater RA. My reading of the drawings makes the 2721s and contemporaries 9in shorter at the front than the 57s, which gives the pre group large classes a distinctive truncated look. The 54/64/74s were only a couple of inches longer than the 2021s at the front so the effect is less marked.

Ultimately isn't the point of the project to have fun building a model. So is that lost? Taking it to extremes 90% of us could have better model railways if we shelled out megabucks to get the best available pros to build every last thing, but where would be the fun in that?

The broad gauge had the same overhang each side as the standard gauge, which was how mixed gauge worked. Whatever else one may say about the broad gauge, it does seem it saved a good few lives in the derailment and accident prone early days, as the numbers suggest that broad gauge carriages were a lot less prone to tip over and smash in accidents after they had left the track. But I hold no especial candle for the broad gauge. The larger track gauge, with hindsight, was *at best* unnecessary. The larger loading gauge, on the other hand was pretty much dead on. The reason many GWR lines were built to standard gauge is quite simply that the GWR didn't build them. The whole of the Northern division of the GWR was standard gauge acquisitions.

This isn't very well drawn, but here are some reasonable approximations of loading gauges. W6A is, I think the basic gauge on British railways. You can see that fundamentally the GWR gauge wasn't much different. Red is an 1875 Broad gauge loading gauge. These did vary - the Bristol and Exeter and South Devon lines were built to a smaller gauge than the GWR. The thin gray lines are various modern European loading gauges Its obvious that the marginally larger GWR loading gauge was not as popular myth would have it, much to do with the Broad gauge. Hardly surprising as probably the majority of the GWR route mileage never had broad gauge track. The broad gauge loading gauge, as can be seen, is basically the same sort of size as the European gauges. It does have the same problem of low eaves to suit arched bridges.

The phrase isn't used in the book that was produced from those drawings. Its just captioned "The Biggest 2-8-0"

Quite a few extra inches. The GWR broad gauge was 11ft 6 across the vehicle, 11ft 8in at the eaves and 15ft at the centre The SDR and B&E were 11ft 0 across the vehicle, 10ft 10 at the eaves and 14ft 9 at the centre I've seen a drawing showing an average Scots loading gauge pre grouping as being 9ft 0 across the vehicle, 11ft at the eaves and 13ft 6in at the centre I think a contemporary UK loading gauge is about 9ft 3 across the vehicle, 10ft 9 at the eaves and 13ft in the centre Obviously as things have worked out its height at the eaves that has turned out to be critical, but the extra width of the broad gauge would have given appreciably more room for containers, which are of course relatively narrow, as would the extra height of the Scots lines.

Track gauge apart though, Brunel was certainly correct in his belief that the Stephenson's loading gauge was far too small, and the british railways have been paying the penalty ever since and still are. One of the bad effects of the grouping was the adoption of lowest common denominator loading gauges: I believe the Scots lines at least had been somewhat more sensible with appreciably large loading gauges than the GWR, whose slightly larger loading gauge can have had very little to do with the broad gauge. I've always been bemused by the notorious engravings of transfers- I've never understood why changing trains between two narrow gauge passenger trains should be any worse than broad gauge and narrow gauge, and I've heard it said the disruption was at least partially staged. Freight, on the other hand, was quite another matter.

If you want some inspiration for very large British outline locomotives see if you can find a copy of A E Durrant's "Swindon Apprentice". Dusty Durrant was a Swindon apprentice in the late GWR/early BR days, and a draughtsman at Swindon when the standards were being drawn up. He filled his sketchbooks with drawings of the enormous locomotives that he felt BR should have been building. Among his sketches are an express 4-8-4 with a matching 2-12-2 freight locomotive, and the comment that a mixed traffic 2-10-4 could also have been schemed out from the same components. As Durrant, flights of fancy apart, was a trained locomotive designer, one may assume his sketches are not utterly ridiculous and might well be a useful pattern. You can be assured that they have nothing that is obviously GWR about them.

That's interesting. When did the last new set go on? I was wondering whether the new sets of frames a very few Castles (4037, 4090 I believe at least) got in the 50s were associated with a need to fit new inside cylinders, and they didn't have the tooling for joggled frame cylinders any more, but clearly if 4079 received new cylinders fairly late on that can't have been the case. BTW can you add removing spring compensation to your list?

For a what if of my own I imagined a running shed very unwisely sited on the flood plain at Peasmarsh, between the unused triangle from Shalford heading southwards and the Cranleigh line junction...

There may always be contractual restrictions. Its common enough nowadays, I don't know about the 19thC.

The trouble is that the fairy tale is told first hand by executives who were directly involved. Its hard to ignore that. Beyer Peacock had no need for the LMS to specify (inadequate) axleboxes for the LMS Beyer-Garratts, but that doesn't mean it didn't happen.

2708? http://www.rmweb.co.uk/community/index.php?/topic/14790-imaginary-locomotives/?p=3029554

The Great Bears tender bogies are indeed assumed to be the locomotive bogie design shortened.

There are drawings in Russell's Pictorial Record of Great Western Absorbed engines, not in full detail though. changes seem to include smokebox chimney top feed safety valves bunker extension That looks to me like a GW standard toolbox too.

Well it does, or nearly so, ("Well Tanks with High Sides") but in handwriting and so on that makes me confident it was added at least ten years later and possibly a lot more. I have a gut feeling that some of the annotations in the book are a lot later, and maybe not even official GWR/WR additions. The register entry for the A112s has several scribbled additions that are not there for any other lots, and in handwriting that obviously doesn't match contemporary entries. Another is a list "Originally fitted to following engs in numerical order:- 4088, 4091 4092, 4008, 4083, 4087, 4016, 4032, 4019, 4056". Its difficult to see what possible use that information was to the GWR, and it appears for no other lots. I don't think we need despair too much about the lack of drawings if we are not seeking to manufacture new components to exact specification. The general arrangement drawings tell us an awful lot about the individual components. I've kinda got into this (can you tell!) and I think I'm going to try and write up the variations of the tenders in some detail, maybe for a book(let), maybe for web pages, it depends how it works out. But I'm pretty confident that I'll be able to draw recognisable sketches for the major components and be able to produce a timeline of what was introduced when so it will be possible to identify what one is looking at and produce a mix and match chart. I've found, for instance, that the A112s had a new spring design - unsurprising - but that the NRM has a drawing for it which they've published in large enough detail to redraw, and that that spring was used for the next two or three lots until they were redesigned again. So here's a 71534 spring. There was a new design for lot A117 and again for A120. The A120 spring was used until the end, including on the Hawksworth tenders, so that will be the very familiar flat design, and I'm happy I'll be able to get a good go at that from GA drawings. The A117 spring I haven't looked at yet, but hopefully the A118 GA will give some detail. The downside is that I see myself spending a fortune on NRM drawings if I'm not careful. [Later - and here's 79936 for A117 from the A118 GA. You can see the new spring is different in about every way. [Later still ] 89792 for Lot A120 on doesn't appear to be superficially different from 79936. The spring hangers are different though, the later design has short spring hangers. I've yet to research when that change was made.

The story I heard was that they were completely overloaded in the machining shops, even though there was capacity in the rest of the works, so reusing all the motion components from the Dean Goods was a big advantage. No doubt they also had more 0-6-0 tender engines than they really had work for too, with the 28s coming into service.

According to RCTS the 1392s in their final guise actually had 1361 boilers. The only trouble with picking the 1392, which was Holcroft's model for the 1361, is that they were so similar at the end, that one would have trouble telling them apart. They did have a rather different cab and bunker though.

Tender variations... I'm just having a session scanning works drawings in Russell, and happened to look a bit more closely at Fig 615 in Vol2, which I had assumed was just an extract from a weights diagram. Looking closer its a bit more detailed than that with some useful measurements. Anyway, its labelled Standard 3,500 gallon tender, Swindon, October 1936. What is drawn from the footplate up is apparently a standard well tank tender tank and gear as per all the 3500 gal tenders built up to the mid 1920s, but the frames are 1936 and later type, with wide hornguides, short hangers and straight springs. So it appears to be representative of what a lot A79-A111 tender might look like being turned out by the works after a major repair and new frames, but not apparently anything that was built new! If I've found the right entry in the NRM list then its a drawing that was prepared for modellers. So the moral is that you even need to be very careful about GWR drawings! If you look at the tenders drawn on weight diagrams (as in Russell again) you'll see a similar assortment of combinations which weren't built new, but may have come to exist after components were mixed and matched in the factory.

Mmm, I think you're right, which is ironic since no GWR 2-4-0 had that feature either. Well, I may have a play... Probably the right thing to do with the tender would be to use something completely different if a suitable donor were available. Not many options of course. I wondered about filling in the tender frames - if the frames were continuous below the axleboxes with ovoid cutouts above I think they'd look a lot less GWR. But that really needed doing before assembling the tender. If one did the same with the locomotive frames I think it would make a big difference. Perhaps something like this? (top original, bottom 2-4-0)

Well, here's something literally thrown together. Many bits are just resting on each other which is why it looks so very crooked... I hope... Has a Dapol City died in vain, or does this look vaguely like a generic 2-4-0 and its worth finishing it off and writing London & Surrey on the side? I do think the rolled up plasticard stovepipe "chimney" and the terrier dome go a way towards making it look less GWR. The smaller dome especially. Just need some ramsbottom safety valves. [and a lot more skill and concentration!]