JimC

And depending on whether they had an excess or a shortage of spoil in the immediate area no doubt. We may guess it would be different for additional or extended platforms built after the line was constructed. I think I'm right in saying in my (Southern Electric) area original platforms are typically brick and presumably infill, but there are later extensions on reinforced concrete 'scaffolding'.

I'm minded to get into CAD based cutting of plasticard. After a fair bit of forum browsing there seems little traffic on this recently and a lot on 3d printing. I don't find design for 3d printing especially congenial, and assembling things from CAD cut plasticard has more appeal. Is there a consensus on whether this is still a good way to proceed? And what is current good kit? Laser cutting doesn't seem compatible with materials I like to work with.

The GWRs 28s were very occasionally used for passenger trains if the need was severe enough, so they should have been well capable of faster timings, whilst the 47 is the obvious next step. They'd just have been able to scrap the RODs earlier, which would have led to few tears on the footplate Weight is the issue. The GWR 3150s - the larger boiler Prairies - were getting towards the red route limit. My not-very-well-informed guess is that a Manor boiler on a 4-6-2T chassis is as large a six coupled tank engine as could be managed in the GWR weight limits. A Hall equivalent would need to be 8 coupled. As for need, well an intensive suburban service perhaps. That enthusiastic Swindon trained imaginer of large locomotives, Dusty Durrant, worked up a study of a 2-10-2T to match Southern electric schedules on 10 coach suburban trains, but it was well over the standard red route limit at 21T on the driving wheels.

Fun job. Enjoy. Yes, I wondered about reversing the whole plot too, but decided I wanted light under the boiler more than an empty cab.

Well probably, but to play devil's advocate, the services had to be run, and run as reliably and cheaply as possible. I wonder at what point the new locomotives stopped being cost effective over keeping the old crocks running? As it was there were a lot of locomotives well past their sell by date in 1945. The replacements certainly wouldn't need to run a full life to be cost effective. Unless someone has done a really thorough analysis of cost per mile of different classes I submit its impossible to draw a line in the sand and say "these ones were a waste and those ones weren't". We can be pretty confident that earlier introduction of diesels would have been a bad thing, so the services did need to be run with steam engines of some kind.

It does indeed. I think we are talking about very different things. Take, for example Cox' criticism of the LNW Claughtons in 'Chronicles of Steam'. He doesn't talk about the headline things, but such details as oil pipes that would break in service, return cranks with square corners which damaged the big end bushes, brake pipework which was forever coming loose due to poor layout, sanding gear that was unreliable, warping smokebox doors, and so on. All this was costing money and reducing availability, nothing to do with the headline items. The state of the art in all these sorts of details advanced over the years, and these hidden things made big differences.

I'm not sure that the moderate updating of such types really counts as modernising. The pre group GWR pannier tanks had all been updated with new design boilers, and often top feed and even superheating at the same sort of time, but they were still replaced with new locomotives. I imagine its not the headline parts but all the detail like bearing design, lubrication arrangements and so on. The 57s were superficially very similar to the final form of the pre group types they replaced, but the maintenance costs seem to have been much improved. One may look at the Dukedogs for instance. The 1895 design frames on the Dukes were replaced with the 1900 design frames from Bulldogs to make the Dukedogs, and it was widely considered a worthwhile improvement. How much more might the difference between 1885 and 1925 components be?

But as has been said before the diesel shunters simply couldn't do the work that the pannier tanks did. The 350hp diesels were dedicated shunters and that wasn't a type the GWR built in any numbers. When you do an analysis of the design its evident that the GWR pannier tanks, and especially the 94s and 15s, had very large boilers compared to pure shunting classes like the Austerity and the S100, because they worked traffic as well as shunting. In the event that short haul trip work largely disappeared but hard to predict that in 1947.

Steam locomotives were relatively cheap to build and expensive to maintain, so its well worth replacing old stuff if the replacements have a much lower cost of ownership. Keeping old crocks running comes with a cost in increased repairs and lower mileage between overhauls. The old crocks also have higher coal and oil consumption. So execs with a good handle on the numbers may not be keen on patching up. In 'Swindon Steam' Cook, the head of the works, claims Stars were upgraded to Castles because the bigger boilers would be working at a lower evaporation rate, and they calculated that would save a penny a mile on maintenance. There's an interesting contrast between GWR and SR policy. The GWR scrapped 19thC and pre group classes through the 30s, (and late 40s and the 50s), and replaced with new build because the new kit was cheaper to run and the investment worthwhile. By the late 50s there was almost nothing 19thC and little pre group left. The SR used the equivalent money on electrification in the 30s because that also enabled them to scrap 19thC kit *and* make major savings in the workforce, but they were left with some old crocks where electrification hadn't reached. Both policies were rational but the war putting a halt on electrification put a major spanner in the SR works. Still, it means we have more interesting old southern types than GWR.

You've inspired me to play around with the old 2721 chassis I have lying about to see what I could do with that. I have a collection of drawings of Welsh class drawings that might make a fictional absorbed class. Sadly though it seems I have a very old school chassis, and no matter what I try I get the motor in the cab - even a Barry 0-6-4T. Still, maybe if I have closed cab shutters. I won't say watch this space, cause any actual modelling usually takes me months or worse!

An interesting vignette from the GWR Elist. Apparently on ECS workings the 15s didn't ingratiate themselves with the S&T department "because nearly every week it seemed one would take out a shunt signal somewhere between Paddington and OOC - and there was no room to move the signals, so the same ones got clobbered time and time again!"

Thanks for that. It would be interesting to hear a comparison with the USA/S100 and/or the Riddles Austerity from someone who has relevant experience.

I was mulling over the design of the (to me at least) strangely appealing 1948 15xx. It was a pure GWR design, and it appears from the NRM drawings list that it was actually on the drawing boards as early as 1944. As Cook tells us it was designed as a "24 hour shunter", not needing to be serviced over a pit: a worthy aim, but rendered largely obsolete by the early 350HP diesel shunters that were being introduced at the same time. I've seen an interesting comparison made between the GWR 0-6-0PTs and their theoretical equivalents on other lines, the Riddles Austerity/J94, the LMS Jinty and the USA tanks used on the Southern. The numbers indicate that the GWR locomotives have considerably greater boiler capacity than the others, but a similar tractive effort. A pure shunter doesn't really need much boiler capacity, since there is plenty of time for boiler pressure to recover, whereas a locomotive used for traffic work does need continuous steam, and 57s and 94s were regularly used on branch and even short trip main line services. A flaw/feature in the 15xx design is commonly held to be the relatively short wheelbase, which is reported as rendering them somewhat unstable at speed, and it seems they rarely if ever undertook the traffic roles of other pannier tanks, although the survivor with the Severn Valley seems to do well enough at preserved line speeds. The actual wheelbase is 6ft 4in + 6ft 6in, 12ft 10in. Its interesting to compare this with dedicated short wheelbase dock shunters, such as the GWR 1361 and 1366 classes , 6ft + 5ft - 11ft, , the USATC S100 at 5ft + 5ft for 10ft and the Riddles Austerity 5ft 9in + 5ft 3in for 11ft. The shortest wheelbase regular 0-6-0T on the GWR was the 850 class, 7ft 4in + 6ft 4in - 13ft 8in. The short wheelbase on the 15s is commonly held to be intended to improve their ability to traverse curves, and their work in Newport and on the Paddington ECS workings stated to support this. Its interesting that the wheelbase on the 15s is intermediate between the pure shunting types listed above and traffic locomotives such as the 850, and even more the other large pannier tanks, 94xx, 57xx and their 7ft 3in + 8ft 3in 15ft 6in wheelbase. As such it has occurred to me that the 15xx wheelbase might be for other reasons than curves. My theory is this: the 57 and 94 cylinders are set partially between the wheels, as is possible with inside cylinders. The big outside cylinders on the 15 can't be, so the leading wheels have to be set back relative to the smokebox in order to clear the cylinders. In addition, whereas on the inside cylinder locomotives the cylinders themselves brace the frame, on the 15xx there's a large structure between the cylinders to perform the same function. All this makes the locomotive heavy, and in particular front heavy. This in turn means that the trailing wheels have to be set forward for the locomotive to balance. The 15 is heaviest on the leading wheels and lightest on the 3rd pair, whereas the 94 is opposite. It would be interesting to know what someone better versed than I on the subtleties of steam locomotive designs makes of that idea. Its often claimed that the 15xx was inspired by the S100/USATC 0-6-0T. The locomotives are indeed superficially similar, with prominent outside cylinders, outside walschaerts valve gear, external steam pipes, water tanks that do not flank the smokebox and no footplate. However this claim doesn't appear in any of the memoirs of contemporary GWR staff that I am familiar with. Given the design aim declared by Cook, a 24 hour shunting locomotive that did not need to go over a pit for servicing, then when examined in detail the comparison is less certain. The design aim forces outside valve gear and outside cylinders. GWR practice was to use walschaerts gear on (their few) outside valve gear locomotives - notably the railmotors and the VOR 2-6-2Ts. By this time external steam pipes were standard on GWR outside cylinder classes. The S100, with its very short (10ft) wheelbase drives to and has the eccentric on the rearmost driving wheel. Apart from anything else there would be no room for the valve gear driving on the middle wheel. The 15xx, on the other hand, has 12ft 10in wheelbase, and a connecting rod driving the trailing wheels would be some 13ft 6in long. The longest connecting rod on any of the Churchward standards was 10ft 8½ inches. I wonder if 13ft 6in would be practical. Here's a list of similarities and differences. Similarities No footplate Outside cylinders with prominent steam pipes Outside Walschaerts gear Wheel size 4'6 v 4'7.5 Differences coal capacity (1 ton , 3.25 ton) parallel/taper boiler driven wheel wheelbase (s100 as short as possible, 15xx longer) boiler proportions (much bigger boiler on the 15xx) All in all, I submit that there's a strong case to describe it as convergent evolution, rather than consider the 15xx to be a direct descendant of the S100. Arguably the only feature of the 15xx which may not be extrapolated from previous GWR practice is the absence of footplate. This could well be a weight saving feature, and in that respect we might also look at Bulleid's 1942 Q1 as an inspiration. On the other hand the GWR drawing office must have had drawings for the S100 available, since the first weight diagram for the type at Swindon is dated July 1943. They may have been in service at WD sites adjacent to the GWR as early as 1942, but RCTS states they were not used on GWR metals until June 1944. The first drawings at the NRM for what became the 15xx are dated February 1944.

According to RCTS no 15 received new boiler barrels in both the 1866 and 1887 rebuilds, but kept the original domed firebox for her entire life. Apparently new frames were fitted in 1890, which seems a surprising repair on such an oddball locomotive. I did have trouble getting a dome shape that I was happy with.

Yes, the scratch built plasticard bridge is simply glued on top of the Peco one with the power conductor wires extended. If appearances are of no concern then a slightly longer deck ought to be straightforward.

If a scale 37'6" turntable was acceptable, this work in progress is a 4mm deck built over a Peco N gauge unit.

One of the side effects of the grouping was that almost all of the small lines who didn't build their own locomotives disappeared. We can note the GWR 1101 class, which was really a lightly modified Avonside design and built by Avonside. The GWR had a curious reluctance to build 0-4-0Ts.

@Dana Ashdown found a photograph of number 15 in her 1887 (final) incarnation, which inspired me to see if I could have a go at producing a reasonable sketch from the photo. As I was going to be colouring the new one I thought I would also colour the previous 1866 rebuild sketch. 1866 Very unusual brake setup for the 1866 rebuild, but I'm confident its a reasonable interpretation of what Ahrons drew in the line drawing I worked this up from. 1887 Looking at the photo carefully I decided it was evident that rather than build a new saddle tank in 1887 Wolverhampton had simply extended the original, which considerably simplified the task. Brakes are more conjectural than I would like, but a similar setup was used by Wolverhampton on early 517s. Colour wise, well, its intended to give the impression of Wolverhampton green. Who knows. As ever I've left out the lining and anything else difficult! GWW seems to be silent on the colours of the painted numbers, but white edged in black seemed feasible to me. I also added these revised images to the original post after the RMweb site image problems in 2022.

Some good spots there. That's an interesting photo of no 15. I sketched it up working from an EL Ahrons photo in a blog post as shown below, but that photograph is obviously from the 1887 iteration. I shall have to try and work something up for that. I also sketched no 92, which was a close relative of 45. There was a sort of not a class of 5 somewhat similar 0-4-0s, which I also wrote up https://www.rmweb.co.uk/community/index.php?/blogs/entry/24886-gwr-no-92/ I'm kinda keen to try and find a ready to run chassis that I could modify into one of those, don't know if any of you have any suggestions. The wheels were 4ftish and the wheelbase 7'2 or 7' 4.

Do you think that's stretching it a bit? No disputing the essential flakiness, but the need to pin down brakes on inclines was part of railway operation from the earliest days when realistically there was no choice. At least some early passenger trains had multiple brake vans to provide the same ability to brake on multiple vehicles. I don't think it can be regarded as a kludge because several generations of wagon were built like that. From what I can make out the primary motive for either side brakes was accidents to shunters crossing the track in front of moving wagons, esp fly shunting, to apply brakes. I imagine the assumption was that there would always be enough wagons with brakes on the desired side to brake a train. I think the bigger challenge was not so much persuading railway executives as persuading private owner wagon execs that continuous brakes were needed. The vast fleet of minimally maintained private owner wagons was the real problem, and even if vacuum fitting had been mandated, wouldn't it have resulted in trains of leaking pipes and cylinders and constant delays with vacuum brake problems?

Complicated, because the pre group GW tanks were upgraded over the course of their lives, with higher boiler pressure and sometimes larger cylinders, but according to RCTS the figures when they reached BR were: 655, 1813 - A 2F. 1854, 2721 - A 3F. (some came in the range for GW power class B, but none seem to have been classified as such) 5700, 9400 - C 4F

Surely not insoluble if the will was there though. Yes, I imagine pretty much impractical on 9ft wheelbase minerals, but the GWR had 20T 13ft wheelbase vacuum fitted coal hopper wagons for their internal power station and gas works traffic.

I suspect a lot is down to ownership. All the various different parties involved had to be prepared to make the investment at the same time for any progress to be made.

Stephenson’s built the first three locomotives of the R class in 1907. Although based on the design of the Ms, they had quite different boilers with a slightly larger barrel and a Belpaire firebox, and cylinders with the valves between the cylinders rather than above as in the Ms. Two more Rs followed in 1909. The last batch of Rs, known semi officially as the AR class, were delivered in 1921 and will follow in another entry. The Rs are commonly held to be the major influence on the GWR 56xx, although to be honest the family style was so strong between all the Rhymney inside frame types that making a distinction seems futile. The most obvious difference between the Rhymney locomotives and the 56s is that the GWR design had piston valves above the cylinders, whilst the Rs had slide valves mounted between them. This is not, of course, evident in the sketches. This sketch is based on a 1908 Stephenson's drawing of the 1909 batch. Later in their Rhymney lives there were safety valve and front suspension changes. The GWR changed safety valves again, and in several cases installed Std 2 boilers. I have a drawing in preparation for the Std 2 boiler, but I am uncertain of several details. That may have to wait until the AR article. The earlier R conversions were, as might be expected, all of the pre war engines, yet the GWR drawings I have, although prepared before the first conversions were made, shows AR frames and I am not clear whether the converted Rs received the suspension and frame changes which would be needed to exactly match the drawings.

One may also recall that in those days honours were much more readily 'acquired' than they are these days...