Miss Prism

Here's one, and surprise, surprise, it's 3203: 3203 - 1937, tapered chimney, shirtbutton, standard cab windows, parallel loco buffers, fishbelly rods That's about it though - most ran with parallel chimneys: 3205 - 1938, fat parallel chimney, standard cab windows, sandboxes 'on top', parallel loco buffers, fluted rods 3212 'Earl of Eldon', straight from the shops, fat parallel chimney, standard cab windows, tapered loco buffers, fluted rods 3218 - 1938, fat parallel chimney, small top feed, standard cab windows, insignia not visible but certainly shirtbutton, parallel loco buffers, fluted rods 9001 - 1953, thin parallel chimney, standard cab windows, parallel loco buffers, fluted rods, BR black 9004, late, fat parallel chimney, large top feed, standard cab windows, tapered loco buffers, fishbelly rods 9004 - undated, fat parallel chimney, large top feed, standard cab windows, tapered loco buffers, fishbelly rods 9008 - 1951, fat parallel chimney, sandboxes 'on top', tapered loco buffers, fluted rods, BR black 9015 - 1955, fat parallel chimney, 'eyebrow' cab windows, parallel loco buffers, BR black 9015 - undated, but late, fat parallel chimney, 'eyebrow' cab windows, parallel loco buffers, fluted rods 9028 - 1953, fat parallel chimney, standard cab windows, tapered loco buffers, fluted rods Sandbox style was inherited I think from whatever style the donating Bulldog loco frame had. I believe 9026 ran with a small 2500 or 3000g tender in BR days. 9019 ran with a 2500g for a while as noted previously in the thread. Locos Illustrated No 50 (The GWR double-framed 4-4-0s) also has quite a few pics, so you might find something in there. Btw, before Bachmann jump at the shirt-buttoned tapered-chimney 3203, check out the rivets on the bogie frame... Swindon standardisation, huh? Yer just can't beat it!

I'm not convinced by the black livery being suggested for BR. My childhood recollections are very hazy, and to be honest, even if the things had black or grey paint underneath, they all looked uniformly dingy and dirty. I had assumed the BR livery would be standard grey unfitted, or standard bauxite for the few through-piped specimens. If they were black, that would indicate they were considered 'engineering' wagons. Was this so? The yellow lettering doesn't ring true either, but I suppose it depends what one means by 'yellow' - I think the more accurate colour for engineering wagon insignia was a pale cream. Perhaps Paul Bartlett can shed some light on this matter.

With a parallel chimney?

I understand Mike Smith's CSB Pannier article will be in Snooze 182, which should appear in a month or so.

All single-arc roof stock was 8' wide as far as I know. On the Leamington picture, the majority of clerestory non-corridor bogie stock was 8' wide, particularly ones with 6'4" bogies, as that one seems to be.

Point taken about smokeboxes appearing or not appearing above the tanks - there do seem to be variations according to era, and I had momentarily forgotten you were aiming for 1912-ish. (For the 1016, compare a very late example.) I think it is very difficult to draw body-width conclusions from the angle of the pic of the the Buffalo behind the 43xx tender, and I referenced it primarily for the width relationship between the Buffalo's rods and its footplate. (I may be sent a better version of that pic soon, btw.) Here's another example, unfortunately extremely murky, with the cab appearing to be narrower than its adjacent 8' coach body. We've probably bored everyone else to death by now!

That's an interesting possibility, Nick, but I'm inclined not to go along with the idea, since only 50 were convertibles, and I also suspect that BG to sg conversion wasn't as simple as is often portrayed, and would have involved a decent GA drawing to iron out all the knotty frame/splasher/spring etc problems. On the standard 'Churchward' outside-framed 4-4-0s, footplates were 8'3" wide, distance over cranks was approx 7'10" and over the outside of the crankpin was 8'9.5". Distance over lower steps was 8'0.5". If, for the Buffalo, Maskelyne and Templer are saying 8'5 1/4" over pins, then that might mean the overcrank dimension could be down to say 7'6", but I wouldn't take issue with your estimate of 7'8" to 7'10", because previous generation rods were probably a bit thinner than the later Churchward ones. (And 7'10" might have been a standard dimension going back to Armstrong days?) I'm not sure what that would then mean for the dimension over frames - somewhere close to your 6'7" I guess, assuming crank thicknesses of 6" tapering to 4.5" at the pin end. From pictures, I would agree with you the Buffalo footplate width was certainly a bit wider than the overcrank dimension but does not encroach much over the rods themselves. It certainly doesn't encroach as far as the stated 8'5 1/4" Maskelyne/Templer overpin dimension. Have a look at this one. (The hoboking pic is also very useful for this judgement.) Concerning the chimney sitting on the smokebox, I'm now not so sure there would have been a aperture, because the top of the classic Buffalo smokebox is just above the top of the tank level. It's subtle though, and the top might look like a 1854 class. The Great Western in South Devon book (WSP) is a bit tasty for Buffalo pics, btw.

Turn the lamps off and pretend it's daytime?

White light on whatever went forward, red light on whatever was in the rear.

Nick, yes, buffer centres at 5'8", and this would give say 7'3" or 7'4" over front buffer beams and say a further 4" overhang of the footplate each side would make it up to 7'11" or 8'. As you say, 8'6" is pushing it - I can't imagine Swindon allowing an unsupported 7" or 8" overhang. Here's newly-panniered 1047 at Snow Hill in 1912 - look how the toolbox overhangs the footplate. This 1016 class is also alleged to have been widened in the mid-1920s madness, as per your diagram refs above. One could speculate the Russell diagrams, at least those that don't belong to the 'larger' classes, e.g. 1701 and 2721, are a series of early experiments under Collett desperately trying to improve his ageing tank fleet before commonsense prevailed with the adoption of the 57xx design, but speculation doesn't really get us anywhere. I don't know any of the answers, Nick. All I know is that I've never seen a pic of a wide Buffalo, and maybe they were always narrow. I can only hope someone out there might see this exchange and might have some better evidence.

Regarding my comments querying the chimney, Simon Kohler assures me "the chimney on the Star model is correct and drawn from the actual GWR drawings", so I'll shut up now.

I looked at this footplate/body width conundrum from a buffers point of view, whose pitch dimension is one of the few we can rely on. Taking a known 8'6" footplate (and 8' body), look at the amount of bufferbeam outboard of the buffer. Compare with a classic Buffalo, where the outboard distance is clearly a lot less. That classic Buffalo pannier footplate can't be 8'6", can it? I assume Templer got his 7'5" body width (and one assumes a say 7'11" or 8' footplate) elevation from somewhere. Having said that, a Buffalo saddle tank is also consistent with the pannier pic, so I find difficulty in corroborating a 8'6" footplate for a saddle version, but then I can't defend the notion that footplate widths would have been intentionally reduced on saddle to pannier conversion. Somebody convince me I'm wrong. (I haven't got my Russell book handy at the moment, so can't compare other saddle pics, and other saddle classes.) Btw, be wary of the differences in widths between some front and rear Buffalo bufferbeams.

Mike - your cunning imperial plan is unlikely to encounter significant dissent - 75% of just about everything (G)WR was in Wales.

Interesting letter in MREMag showing that in the 1950s, 75% of the class were in Wales.

Best of luck with this one, Nick. I've got a more ancient (M&L version) unbuilt specimen, with everything in whitemetal (footplate, frames, cab). It's not pretty. No wonder I kept bending Martin Finney's ear for a decent etched Buffalo, but alas, in vain. He'd obviously thought about it, the Buffalo also being a favourite of his, but he knew the many pitfalls as well. The central problem with the kit concerns the width. As can be seen below, I came to the conclusion that the whole thing was about 2.5mm too wide. Here are the notes I made at the time in correspondence with someone else tackling one of these engines. In my view, the Buffalos are by far the most complicated of all GWR classes. _______________ In the pannier tank era, there are roughly 4 appearance categories of Buffalo engines as they ended their days: 12 locos with high-pitched boilers, large (1200 gallon) tanks, and where the boiler extends above the tops of the tanks; this type is the subject of the upper Templer drawing in Russell's book. Let's call these the 'high humped' locos. According to the RCTS records, not all of these had enclosed cabs, but I find this impossible to believe for 1925-30 construction. I feel these 12 locos do not really qualify to be called Buffalos, as the entire body was 7" wider than all the rest of the class, i.e. the particular type of boiler being fitted required new tanks, new cab, new bunker and new footplate - the whole of the upper part of the loco was therefore completely reconstituted! Quite why such a subclass came about is a mystery, unless Collett wanted a guinea pig for that new boiler type before launching into the 5700 pannier dynasty, and other inside-framed types like the 2721 and 1854 classes were surely better types to experiment on. I have never seen a picture of one of these high humped Buffalos. 19 engines with high-pitched but otherwise conventional Buffalo boilers and tanks (approx 1000 gallons). I call these the 'high flat-tops'. Some had enclosed cabs whilst most did not - see fig 242 of Russell for the latter. Impossible to tell if figs 246b and 247 of Russell are this type in the absence of dates for those photos, but 1282 (with its 'busby' in fig 247) is known to have been stationed at the Didcot ordinance depot c.1907, but both pictures are obviously a lot later than this, and could be between 1916 and 1924, when that loco was in 'classic' Buffalo style (see next bullet point). The bulk of the rest of the class, which look more or less like the lower Templer drawing (cabs and bunkers differ, primarily), and which I call the 'low flat-tops'. This is perhaps the 'classic' Buffalo look. See figs 236 and 257 (open cab) and fig 238 (enclosed cab) of Russell. Boiler pitch is 6'6 3/4", barrel length 11', and length over firebox casing is 5'4". The 'extended' engines, used exclusively (?) for auto work. These are as per the classic Buffalo (see third bullet above), but have extended smokeboxes and tanks. See figs 237 and 241 of Russell. But we don't have a drawing! We know however that the extension is in respect of the smokebox and tanks, so boiler length is unchanged from 11'. Locos known to be in the extended category are 738 (nice front 3/4 view on page 225 of the South Devon book), 1167, 1234/5/52/65/9/71/84, 1567/70/1600. I would add 1168 to this list, if the caption on page 185 of the South Devon book is correct. An immediate modelling snag in the generality of the Buffalo class concerns the overall length of any particular loco: the rear overhang varies according the lot number (see figs 239 and 245 etc of Russell). You need to translate this info for your chosen loco using the lot number info on page 94 of Russell. Having said that, some of the earlier, shorter, locos might have been lengthened up to the standard 6'9" rear overhang towards the ends of their lives, and this seems most probable if/when receiving new cabs and Collett bunkers. For the extended locos, only 738 comes into the 'shorter original' category, and this appears to have gained the full monty in respect of rear overhang judging from fig 241 of Russell. Most details are easily deduced for the extended engines, others less so, and one particular aspect is difficult. At one time, I thought there might be a questionmark over the body width of the extended locos, i.e. 7'5" (classic Buffalo) or 8'0" (as per the high humped locos): I am now strongly inclined to the former, so the end views of the lower Templer drawing are o.k. But I can't be absolutely certain, and Swindon's standardised bits and pieces policy meant that a loco could go into the shops one day and come out a week later, looking all in the same proportion, but actually 7" wider down the whole length of the loco. No drawings have surfaced for any of this critical width information. All extended engines were autofitted, e.g. 1271 being fitted in 1915 (full details in RCTS for the other engines). All are assumed to have been fitted with screw reverse, probably at the time the new tanks/smokeboxes were put on, and also probably coinciding with the fitting of new cab plates and bunkers. All extended engines were fitted with ATC c.1930-31, but were the shoes at the front or the back? 1600's picture in Russell appears to have its ATC shoe at the back, and there was certainly more room at the back than at the front. The dates of fitting the extended smokeboxes/tanks are not known, but they would have coincided with the other associated mods, i.e. enclosed cabs (but see below), rectangular window front cab plates, increased bunker size, and is therefore generally c.1924-27. Rear cab spectacle plates remained circular. All extended engines have enclosed cabs, with two known exceptions: 1234 probably never got one, and in anycase is a bit of a freak, having an extended smokebox but unextended tanks; 1252, extended sometime between September 1920 (page 186, South Devon book) and July 1924 (page 223, South Devon book) has not got an enclosed cab by the latter date, but could have had one fitted after that. All extended engines have rectangular windows in the cab front plates. I always thought the converse applied, but, stranger than strange, I now discover it does not. There is an interesting twist over 1570. Dr Ian C Allen's (a photographer not known for getting his loco numbers wrong) picture of 1570 at Yelverton (page 196, South Devon book) c. 1930 (known to be between 1928 and 1932, and I suspect towards the later end of that period) clearly shows an extended loco. His well known picture of the same loco at Tavistock in 1931 (see Great Western Album, R C Riley, Ian Allan) clearly shows the rectangular-windowed loco with an un-extended smokebox and tank. Obviously the loco nipped into the shops in between photos, and came out looking distinctly different! Both these pictures appear in Great Western Autotrailers Part 1 by John Lewis (WSP). Bunker shape: no problem for enclosed cab locos - as per lower Templer drawing. Concerning the exact length of the smokebox/tank extension that you have to add on to the front of what's shown on the lower Templer drawing, we can only take an educated guess. Scaling off the 1600 picture in Russell, I would estimate that you would need to add on about 9" to 10", i.e. the extended smokebox length is something like figures 262 and 264 of Russell and the Templer drawing and official weight diagram for the 1661 class (page 97, Russell book). Chimney length will be 3'3", and dome as per the lower Templer drawing (not the upper drawing, whose dome is far bigger). The big uncertainty concerns the positioning of the chimney for the extended locos. I get the feeling from the extended Buffalo pictures that the chimney is in a 'forward' position, i.e. like the unextended lower Templer drawing, where the chimney is approx 1'3" back from the front of the smokebox. But we know that very few of the Buffalos (and none of the extended ones) were superheated. So why the extended smokebox - surely not to gain just a few more gallons in the tanks? Or were the extended ones superheated (which would befit their use on autotrain work) and nobody ever made a record? I don't know. Logically, we have to accept that the chimney is in a rear position, like figs 258/262/264, which gives the chimney at approx 2'3" back from the front of the smokebox. Modelling-wise, it's unfortunately not something that can be left until last thing, like on a conventional visible-top-of-the-smokebox-type of engine, where you can move the chimney about a bit until the whole loco looks like the pictures; for a flat-top pannier, the chimney aperture has to be made in the top of the tank wrapper (see, e.g. the pictures of 2112, 856 and 5402 in Guy Williams WSP book), so it has to be right at the pannier-making and fitting stage.

There's a good GA of diagram M1 (initially fitted with lever brakes, but soon retrofitted with Dean-Churchward) and an M4 underframe in the 'Bible', plus plenty of pictures. Buffer types, brake details and handrails varied over time, but I would suggest an M2, M3 or M4, with DCIII brakes, would be a good long-timespread choice. Hornby's new specimen is an M4 or M5 with an M1 number. (Credit in a way to Hornby, though, because M1's 41886 really was at Gloucester Old Yard in 1944.)

There are some detail differences between the various shunters truck M diagrams, but nothing a little book and diagram research can't solve.

Plenty of time to fix that chimney, then...

It's undeniably pretty, but I too strongly doubt whether any Dukedogs would have appeared looking like that after WWII. Having been outshopped only in 1936, my guess is that most of them continued on in an increasingly grimy state throughout the war, and the first proper repaint would have been into BR(W) black.

The beam pitch in the Markits WD short austerity knobs is very close to 1.25mm, and these seem to align well with the carrier holes on their HL blocks.

Thanks. I think we are into the realm of a combination of tolerances on the round bearings and the jig etching, so I'm not surprised a little fettling was required. In design terms, Chris Gibbon is quite right to err on the tight side. Thanks also for the beam pitch estimates. I must revisit some etchings on their blocks. Unless Chris has tweaked the carrier artworks (again!) Btw, did you use the HL thinner 'spacesaver' blocks on the front axle to clear the inside motion bars?

Nick - what is the pitch of your beam from the inside face of the frame?

Part 5, 6-coupled tanks.

Nick, when using the HL CSB jig, using the round bushes in the rigid axle holes (your 57xx_03.jpg above), did you find the round bushes needed small flats on them to fit in the jig slots or did you wipe the jig slots with a file?

Reading's moniker was "Bulbs, beer and biscuits". I think David Hyde's book has some details of the night freight nicknamed 'The Biscuit'.