2703, of the 2701 subclass, received panniers in 1920 and survived until just after WWII. This view is probably late-'20s/early-'30s. Swindon cab (I think) with large spectacle glasses, and 'simple' injector. The small handrail on the tank side is yet to be fitted. It was never superheated. Note the extraneous handrail knob.

They certainly vary. I have a lot of trouble with cab heights, they often don't seem to scale well on the drawings when you have a measured dimension. The dimensions I have for 645/655 are:

K - 10'10 3/4 (equiv of M but 645 class) 

B28 - 11'4 5/8

B65 - 11'4 5/8

The others are scaling from the drawings I have, which are often distorted. I usually have boiler pitch and wheel centres and try and make those fit the measurements. The diag M was particularly difficult, I only have a photo with much distortion and had to manipulate more than am really secure with to try and make parallel lines parallel. 

Regarding diagram A18, I've never seen a pic of a large saddle tank with such a stumpy chimney.

4 hours ago, Miss Prism said:

Regarding diagram A18, I've never seen a pic of a large saddle tank with such a stumpy chimney.

Odd, isn't it. There were cut down locomotives for various highly restricted lines, but you'd think they'd start with an 850 or a 2021 for something like that. 

12 hours ago, JimC said:

Odd, isn't it. There were cut down locomotives for various highly restricted lines, but you'd think they'd start with an 850 or a 2021 for something like that. 

Indeed. There is a list in RCTS of the eight 850 locos fitted with very low chimneys, but it is not clear whether all of them were so fitted at the same time. Here is 863 at Laira on 15 June 1928. I think the purpose of the low chimney is to clear one of the tunnels in the docks.

2 hours ago, Mikkel said:

Except on numerous kit-built locos 🙂

I seem to remember a notorious Cyril Freezer diagram of a large saddle tank with a stumpy chimney appearing in the Railway Modeller. Whether any of the first Wills '1804' kits featured such a chimney I don't know, but certainly in later years the Wills kit was provided with a correct chimney. (As you have on yours.)

As another illustration of cab variety, here is a Wolverhampton tall cab with a small radius roof. This is 1548 of the 1501 class, c 1900. The tank main handrail does not wrap around the front, which seems to be an early Wolverhampton quirk. The handrail also continues onto the side of the cab on this particular loco.

On 16/05/2022 at 04:56, Mikkel said:

Exactly, I thought perhaps it was just me. Drawings of the 2721 and 1854 classes are another example. Thank you for the measurements.

Mmm yes. I drew out the 57xx inside valve gear from a GA drawing yesterday, and I've revisited a couple of other drawings to see if it can add anything, I got to the 2721 drawings I did for the book, and revisiting it with what I have learned and the skills I've acquired since I reckon that in places there's up to 3 inches discrepancy between the GWR drawings for weight Diag A11 and Diag B47. 

For interest here's the two superimposed. I normally align wheel centres, because its the most convenient reference, and that's how I've done the top one. The lower one has the main part of the body aligned, which is a better fit for these two drawings, but is a lot more difficult to do. When I do my drawings I align with guide lines draw on on the main dimensioned points, so the crossing points of the wheel centres are ideal. The overlain drawings also show up the distortions that have occurred in the paper over the years and the transfer to pixels. Look at the various guide lines on the drawings and how they drift from parallel in varying amounts in different places.
From my POV the discrepancies look rather large, but I bet if I were to prepare a scale drawing of a Hornby 2721 and superimpose that the issues with the GW drawings would suddenly look much better!

2721 and 57xx cab heights were the same - 11'4 5/8", at least for the closed cab version of the 2721. I can't find a dimension for the cab height of an open cab version. There was a lot of cab height variation across the saddle tanks, but I think by pannier days there probably should have been a set dimension - the 2721 kit of parts had become fairly standardised by then I think. Probably the driver for standardising cab heights for closed cabs was the need to interface with bunkers.

Boiler pitch for the 2721 and 57xx was the same, at 6'11 3/4" (and standard for most of the large tanks).

3 hours ago, Miss Prism said:

Boiler pitch for the 2721 and 57xx was the same, at 6'11 3/4" (and standard for most of the large tanks).

I think we should qualify that as being standard for P Class boilers. I haven't spotted any exceptions in a quick trawl through my spreadsheet of 0-6-0T dimensions. And, perhaps more surprisingly the same seems to be true of cab heights of the full length roof/rear sheet style as per 57xx (but not 8750). Other boiler types, O, Q, T, Std 10 and the smaller classes are different as one would expect. Chimney heights are more variable. Must admit I hadn't noticed that consistency.

I don't know how consistent heights were on the older cab styles with the various pre group and local bunker styles. The weight diags I have are often not dimensioned - the drawings of sections would be better for that. In those days there appears to have been an element of local variation as if the minor works were cutting the sheet metal on the fly rather than working in detail to Swindon issued drawings. Incidentally one does note that like all smaller than 10 wheel classes the 0-6-0Ts were kept within the 19thC 9'0" x 13'3" loading gauge, presumably for wider route availability.

Here's another overlay, not perfect, but good enough I think, even if it does appear you should look at it with 3D glasses!
 

Green is B65, the last diagram for the 655/1741 Class.

Red is B47, similarly the last diagram for the 2721s.

Blue is B74, the last diagram for the 57xx.

You can see the extent to which originally rather different classes merged, with just the shape of the footplate and the end overhangs being major differences. 

A very interesting post which answers a few questions for me, but leaves some more, to be answered. I have been thinking of drawing and printing a body of the 1854 class, and wondering if 2721 Hornby chassis, is a viable option. 

My understanding is that the Hornby 2721 chassis is too long and has an inaccurate wheelbase.

Yes, it all depends on the compromises you are prepared to accept. I haven't found a single GWR or absorbed class that the Hornby chassis is a really good match for, you are really stuck with LMS types with that chassis, but plenty of folks have accepted the 2721 as being satisfactory for them, and why not. 

Differences between 1854 and 2721 past as built status are not huge, suspension is perhaps one of the bigger ones. On the whole though I think a Bachmann is a better starting point for the pregroup panniers.

I've got a spare 2721 chassis and have mulled over attempting a Rhymney style body, but the RR 0-6-0Ts like S and S1 look notably short ended to me and I don't think you could get the look. 

A thought for a Hornby 2721 conversion might be a fictional absorbed type. Not a few retained their original cabs and bunkers (sometimes extended) , when given GW boilers and pannier tanks,  so a new cab with say a Barry shaped cab entrance might be an interesting little project. 

thanks for the advise, I need to find myself a used Bachmann chassis. Not being familiar with the Bachmann product is the 8750 the model to use.

8 hours ago, 57xx said:

 

A bit late to the conversation here, but... No, the Hornby chassis has the wheelbase wrong dimensions, I think it is the generic Jinty chassis. Either of the Bachmann ones should be ok (57xx or 8750).

thanks for the reply, now I just need to find one on this side of the pond.