Miss Prism

I've also always been sceptical about promoting the idea of the need for short-wheelbase locos working the Paddington ECS duties, considering standard 15'6" wheelbase locos had been doing the job for yonks, and I think your 'weight balance' theory is far more plausible.

What was needed at Paddington was a strong/heavy ECS loco because of the gradient on the 'carriage flyover', and in that respect the 15xx was ideal, although it could be said the 94xx were perfectly adequate for the task, and they existed in plentiful numbers.

Bottlesford is a bit of a one-horse town (very small village, actually) but it has some nice trees.

https://www.rmweb.co.uk/community/index.php?/topic/5436-how-deep-are-pits-in-loco-sheds/

9 minutes ago, Compound2632 said:

Can your list of allocations for 1921 be taken as any sort of guide to allocations in, say, 1902?

That's too much of a time spread to expect any sort of similarity in respect of specific engines. Mikkel's 1919 scenario however is very close to 1921, so one would expect some 1919 inhabitants of a shed to still be there a couple of years later. Low-mileage goods engines weren't serviced anywhere near as often as high-mileage passenger ones. What I think will hold true over a longer time period is the general geographical disposition of a particular class of loco, hence the value of the mapping exercise, which shows a predominance for the 1854 and 2701 classes in Wales.

Even Newbury didn't have its own shed, which is a bit strange considering its status as a significant junction. I guess nothing originated or terminated there.

1 hour ago, Mikkel said:

One issue with the allocation map is that it only shows the home shed. Some classes would obviously have wandered far during their daily work. I'm not sure about these particular classes though. Something to investigate.

For Farthing, I guess Salisbury is the obvious first choice for an 1854 loco. (Presumably Salisbury GWR shed was probably closed when the GWR station closed, but I'm not sure.)

You'll probably want the chimney in the rear position for a pre-1917 unsuperheated loco (but check with Part 5 on dates).

Love the shed mapping exercise, it's very informative.

Outstanding.

This is I think a ex-RR P class loco, not sure what the differences are to the M (frame length?), but I think they were later lumped into the same category after Swindonisation. The pic, at Cardiff in November 1923, shows a standard repaint in Swindon style of a non-Swindonised example. If the repaint was done at Caerphilly, it proves Caerphilly was obeying Swindon orders for local repairs immediately after the grouping.

20 minutes ago, kitpw said:

I can't even identify the diagram number!

L22

Lovely. Gives good depth to the whole scene.

Some Welsh things were too weird for Swindon to consider carrying on with:

(this ex-RR well tank received a GWR number, but I don't think survived much beyond the time of this pic (August 1924))

Thanks Jim. It seems likely that the footplate height of an 1860 engine (especially with 5' wheels) was probably a bit lower than it was in later days. Often the buffer axis setting on the bufferbeam is a telltale clue.

I'm not convinced the running plate needed to rise to clear the cranks.

If they were London Plane they would probably be double that height.

1 hour ago, Mikkel said:

They still do the Medium Metro and MSWJR locos in 4mm:

 https://www.roxeymouldings.co.uk/category/11/4mm-scale-etched-loco-kits/

Thanks, Mikkel. My mistake. Perhaps Roxey took over the Neep artwork.

The Roxey small Metro is available only in 7mm. The Rod Neep small Metro is different artwork IIRC.

I think the first thing the GWR would have done is put a frame extension on the rear with a larger bunker.

The GWR would have fitted a new boiler (I expect they would have had one that fitted that size), so chimney, dome and safety valve would have become standard GWR items.

Post-grouping GWR livery would have been standard unlined green and black. With probably GREAT WESTERN insignia on the tank sides (because, ironically, it was an absorbed engine), but maybe no insignia depending on where the numberplate was placed. It may or may not have received GWR boiler fittings at that stage.

2 hours ago, JimC said:

Do you think, though, the other rivet lines represent tank boundaries or simply fabrication of the bunker assembly? I've taken the viewpoint that only the very closely spaced lines of rivers represent a tank edge, and others are boundaries of plates and/or transverse structures. With my 'no rivets' policy those are best left off. 

I see what you mean, in which case, the tank boundaries are probably better denoted by a line that does not resemble rivets. e.g.:

Brian Daniel's albums always a good source of pics

I suppose what I'm saying is that maybe there should be an extra diagram:

Should there be an extra bunker diagram for 3131-50?

No 1 in post-1882 2-4-0T form. I'm not sure why it was classed as 'experimental', because the Metro tank design had been established by then. As larger versions of the small Metro, I'm surprised these didn't make it into a production run - I don't know what their weights were, but they would have been very useful on the heavier suburban work.

3516, a Stella condenser (built apparently to work the Severn Tunnel). I tend to think of these locos as large outside-framed Metro tanks.

What I find interesting is how these long 0-4-4Ts got round curves. 34/35's total wheelbase is a comparatively modest 20'2", but seemed to cope reasonably on the St Ives branch. The 3521s, at 21'10", disgraced themselves in the Doublebois incident, and although dodgy track was perhaps primarily to blame, it was enough to convince the GWR that 0-4-4s could mean trouble, and they quickly reversed them into 4-4-0s. Not sure exactly what an M7 is offhand, but it is probably approx 23'4" total wheelbase, so there must be significant sideplay allowance in the bogie suspension.