buffalo

Yes, I keep telling myself that one day I'll build a River... Interesting that you've seen a photo of a Barnum with underhung springs and later style of smokebox front plate. I've not seen one like that, although Russell has a photo of 3221 with original boiler and winged plate, but with the springs above the running plate. It looks like all combinations are possible. Nick

Maybe not as nice as a River, but a Barnum in near original condition is a very fine choice. Did they remove the winged smokebox front plate before the springs were moved? Nick

That's a very nice collection of coaches. I hope you'll find time to finish them off. There are plenty of works photos showing this feature and a few in service. See, for example, Russell's coaches vol 1, Edwardian Enterprise, etc. There's a good example of a recently painted rake of four-wheelers with this feature at Acton on p145 of Edwardian Enterprise, and a 1907 horsebox that may have it on p157. It appears to have started in the early 1880s and continued into the early 1900s. However, it was far from universal and there are many photos showing overall white roofs both at the works and in service. Many photos where the white lead paint has turned grey show no sign of a difference in shade in this area. I get the impression that it is perhaps more common on stock with only a small gap between the lower rain strip and the cant rail. I don't recall seeing any examples of wider brown strips as in the fourth and last photos above. I think it is fair to say that it is a feature seen more often on models than on the prototype. I know that I painted all my Ratio coaches this way before I studied prototype photos more closely. Perhaps some of us are also influenced by the re-appearance of this feature on older clerestoreys and four-wheelers after 1927? Nick

Why? The rocker mechanism takes the drive from the inside motion up over the frame and down outside the wheels. Where would the clearance problem be? See also the larger standard classes where a version of this mechanism with longer levers is used with larger driving wheels. "When this took place I was in the drawing office, and detail drawings of the new locomotives were sent up from Swindon for issue to the shops." The previous paragraph briefly describes both classes and the orders from Swindon, so I think we can assume that he was referring to the first drawings for the 44XX. There is no reason to suspect that those for the the 45XX were any different. Yes, they were the first and last Churchward standard designs to be built at Wolverhampton. Even though they were the smallest of the Churchward standard classes, they were more than six feet longer than anything previously built at Wolverhampton. To get them out of the factory, they had to remove buffers and both trucks, and then reassemble them outside in the yard. Little wonder, then, that without redevelopment Wolverhampton was closed down and production from 4520 onwards reverted to Swindon. The difference in the drawings was not solely the different shapes of the parts. Holcroft says "...they were white prints with dark blue lines..." He goes on to describe how they mounted on linen and cloured to identify individual parts. That he remarked on their form suggests to me that they may have been a quite different style of drawing from what had previously been seen in the Wolverhampton drawing office. Whether this was simply a difference between styles that had evolved separately at the two factories or whether this was the result of innovation in the Swindon drawing office under Churchward is not clear, though I suspect the latter. Nick

I really don't know where Julie got that idea from. For the cast name plates, the words "x signal box" were the norm on the GWR. This was not always the case before the cast plates appeared, and there were, of course, exceptions. As to Box, I've found only a single undated photo in Clark's Historical Survey of Selected GW Stations, vol 2. There's a poor copy here. In the book, the cast plate clearly has three words which are almost certainly "Box Signal Box". Nick

The Cambrian example sounds quite similar to the Wolverhampton stores van shown in fig 260 of Russell's Appendix. Nick

Pure speculation, and quite contrary to the available evidence. They already had two years experience of the prototype No 115 (later 4400) when 2151 (later 4500) was built, and there was a four month gap between 3110 (4410) and 2161 (4500). Holcroft, who worked in the Wolverhampton drawing office at the time explained the thinking that the larger wheeled version whould be "more generally useful". Subsequent experience showed this to be correct. Interestingly, Holcroft also describes the drawings that were sent from Swindon for these engines. His description suggests that their form was rather different to any in his previous experience. On the cutting out and erection of frames, I'd suggest reading some contemporary sources such as Ahrons or Holcroft to find how it was actually done. There is much precision drilling and machining involved. The Wolverhampton staff were certainly not a bunch of country blacksmiths. Swindon didn't have a great deal of experience with outside cylinders when 115 was built, either. These classes were the last to be built at Wolverhampton because Swindon had plenty of work on its hands and the Wolverhampton works had had no new building work since completing the last of the 2021 class a few months earlier. As the new designs were much smaller than the other standard classes, the job was given to Wolverhampton. Even then, they were rather too large for the Wolverhampton works and Holcroft describes in some detail the contortions needed to get them out of the factory. This experience demonstrated that there was no future in new builds at Wolverhampton without serious re-development that was planned, but never took place. One thing I've not yet been able to find an answer to is whether major components, such as the cylinder castings, were made at Wolverhampton or shipped up from Swindon. Nick

I think we'll just have to agree to disagree here. What you call a "huge cut away", I see as appropriate shaping. Perhaps you should look at the larger standard two-cylinder classes such as the 28XX, 3100, 43XX and 42XX. Yes, they are larger but, in each case, the motion bracket is roughly in line with the centre of the leading driver and is shaped as needed to clear the wheel. I cannot see that this would have been a major problem in designing these smaller classes. I'm also not sure, but I thought the main plate frames ended at the rear of the cylinder castings on all of the standard classes. The forward extension was bolted to the main frames in this area and to the back of the cylinders. The potential weakness of this forward extension was recognised at the time and led eventually to the provision of stay bars between smokebox and running plate. In the early types it was also the reason for the large curved plate linking the motion brackets. This was apparently intended to support the boiler in the event of front-end collapse following a collision. This is all covered by Holcroft and it is interesting to note his comment that the 44XX and 45XX had the weakest front ends. Presumably because they were scaled down versions of the larger standard types. Yes, the figures are in my post #5. The proportion of the total weight carried by the front and rear trucks is essentially determined by their springing. Remember also that both trucks are in different positions on the two classes. Nick

This is GWR standardisation in action. Yes, they share con rods and valve motion but, perhaps more importantly, they also share the same cylinder/valve/saddle castings so the overall dimension from cylinder centre line to driven axle is the same 10'6". The same was true with the longer 30" stroke cylinders of the 3100, 43XX, 2221 and 42XX classes where this dimension was 12'. This gives the fixed points from which the rest of the frame design proceeds. Thereafter, the idea is to place the remaining axles in positions that achieve the best possible weight distribution within constraints imposed by other parts of the design. It didn't have to be a perfect distribution, though, and many weren't. That the simple reversal of the distances between the driving and outer axles led to an equal load on each axle was, perhaps, fortuitous and had the added benefit that the same rods could be used. As to the relative positions of the knuckle joint and big end, clearly it is easier to remove the 6' rod on the 45XX, but I do wonder just how often only that rod needed to be done. In most cases it's far more likely that all rods would have needed to be removed for maintenance. That said, I believe the 44XX is the only one of the standard two-cylinder classes with the knuckle masked by the con rod, although most Saints and Stars were built this way. You make frame design sound more like blacksmithing than engineering. Yes, the motion bracket may have provided a constraint on the design but, as probably one of the least expensive of the standard components, I would have thought that an alternative design could have been developed if it really was a problem. Nick

Actually, the situation was quite the opposite. The S&DJR had come into existence as a 'joint' railway leased by a committee of the MR and LSWR in 1875/6. At that time, the S&D was in a dire financial situation and the agreement between the MR and LSWR had been set up very quickly in eight days primarily to avoid the B&E or GWR gaining control. The early years of the twentieth century was a period of reasonable profits and substantial growth in goods traffic. In 1914 they decided to split all of the Traffic Dept goods wagons between the two parent companies. It appears to have been more of a 'rationalisation' than a response to any particular problems. In all, 1136 wagons and vans were divided equally between the two companies with 221 'service' vehicles remaining in the joint ownership. The latter, with some additions, mostly retained S&DJR livery until 1930. The service vehicles included 14 8T open peat wagons, 65 10T loco coal, 24 ballast, 14 miscellaneous, 41 brake vans and 63 'service duplicates'. In most cases, the two companies received vehicles of their own design, or based on their own designs. Each company received a hundred covered vans. Other than the 'road vans' discussed here, it appears that the remainder of the covered vans were a mix of MR and LSWR designs. The fact that over fifty of the MR design eventually passed into SR ownership suggests that the Midland types were in the majority. Although the two photos show vans labelled as 'road vans', I do wonder whether all were. We seem to have two possiblities, either all were labelled in this way and the S&DJR had a quite different understanding of the term, or only a few were and these were used for small deliveries to stations... Nick

Hi Mikkel, That photo and a drawing of the S&DJR example (No 1038) also appears in Bixley et al. An Illustrated History of Southern Wagons, Vol 1. Fifty were built for the S&D at Derby in 1896 and became Midland property in 1914. Similar ones were built in 1899 by SJ Claye and others at Highbridge 1903-12. Fifty six were passed to the SR, becoming diagram 1404. They were 16'6" OH and 10' wheelbase. There was also an earlier type, built 1893-4, with X bracing either side of the door. These were 16' long and had 9' wheelbase. Eventually those that passed to the LSWR became SR diagram 1403. The caption for the latter says Nick edit: ps the last of the early ones was scrapped in 1928 and the later type in 1933.

Michael Harris, Great Western coaches from 1890 has: Current in 1890-1895: 1st Crimson Plush, 2nd and 3rd Rep, smoking compts in leather c1896: 1st 'Fancy' moquette or blue cloth, 2nd velvet terry or rep, 3rd fawn rep, smoking leather c1900: 1st dark green cloth or leather, 2nd brown and white star motif moquette, 3rd dark red and white star, smoking leather. Nick

You don't say where you found it, but a quick look in Wikipedia found with a reference to Bradley, D.L. (1980). The locomotive history of the South Eastern and Chatham Railway. London: Railway Correspondence and Travel Society. ISBN 0-901115-49-5. p102. Nick

Lamp bases: IKB made them as white metal castings so they may still be available from the BGS. However, every supplier of cast lamp tops has a different idea of how big they were, so they might not suit another make. I'd go with Mikkel's suggestion, but tubing might be easier if the lamps have a mounting spigot underneath... Nick

Possibly. The first vending machines to become widely used on railway stations appear to have been for postcards around 1880. There are several variations of the Nestle machines, including single and double, square and round column pedestals, metal and glass fronts. Notably, the small wording between "Nestles" and "Chocolate" varies between "milk", "swiss milk" and none (usually on the later solid front types). All of these variations may be indicative of dates, but I've yet to work it out. The one in the 1909-10 photos of Camerton is, as far as I can tell, identical to Robin's example from York. As you might guess, I've looked into this before. So far, I've not seen a clear photo showing any of these machines before about 1905. Although Société Farine Lactée Henri Nestlé produced milk chocolate from the 1870s, several sources suggest it was not widespread in Britain until after the merger with the Anglo-Swiss Condensed Milk Company in 1905. Nick

According to Robert Opie in Sweet Memories (p113), Nestle chocolate appeared after 1905. The machines appeared about the same time and there was even one of the same glass fronted type at Camerton by 1909. Nick ps. that book is a good source for advertising posters.

Thanks, Tony. I'd missed the point that you were talking about 'fixing' defects in the photographed model, rather than just 'improving' the photo. You can tell it's a long time since I did any darkroom work, I can't even remember the proper terms for the techniqes. I fully agree with you about review photos showing the model as it is. As a very average photographer, I too marvel at what those guys produce. Nick

I can almost sympathize with that view, Tony, but surely in the old days we used to do similar things in the darkroom? Maybe some techniques were not widely used in model photography, but faded backgrounds by varying exposure, adjusting contrast of different parts of an image and even creating a single print by masking multiple negatives were all part of the darkroom craft. These probably are skills that are in danger of disappearing simply because modern technology makes the tasks more convenient and more controllable. It's also worth remembering that our eyes and brains effectively build up what we see from multiple images so that we appear to be seeing the world with a much deeper DoF than we can achieve with a camera. Is the idea of overcoming some of the severe limitations of any camera system such a bad thing? I'm often puzzled when folk say things like this. Most of what is said here and in other topics revolves around the pleasure to be gained from kit and scratchbuilding, and encouragement to others to have a go. The message is that if you can develop the skills then you are likely to be able to get much more out of modelling. Of course, that message is not for all and there may even be folk with the relevant skills who find little interest in using, perhaps because they spend all day using those skills and want something different from their modelling (hence my preference for the mechanical over the virtual when it comes to making things). It's all down to personal choice. Nick

Yes, just different skills. How does that differ from hand drawn etch artwork or hand-made masters for casting? No motor skills? You have to be joking? Again, it's just different skills and, for many of us, additional skills. I can see little evidence of a 'loss of scratchbuilding' or kitbuilding. Those who have them will retain older skills even if they adopt new ones. New technologies are more likely to attract others to have a go and, once the limitations of any one technology are appreciated, they may try others. Nick

Whenever I see statements like this, I can only assume the writer has no knowledge of either approach. Do you really believe we are all born with CAD and 3D modelling skills and just have to push a couple of buttons to produce a finished model? Traditional scratchbuilding involves research, design, production of parts, assembly and finishing. All require different skills. Research is much the same however we intend to produce the model. At the design stage, computer-based tools give us a way of producing shapes that are more dimensionally accurate than those produced from hand drawings. They may also allow us to check that parts fit together before they are produced. All of this requires at least as much skill in manipulating the software as hand drawing. A Cameo cutter merely replaces the cutting skills used in the production of parts. All very useful, though you won't get very far as a modeller without some hand cutting. As to 3D printing, again that only replaces some of the part production and assembly phases. In both cases, the result still needs a fair amount of finishing to compare with a good hand built model. Personally, I've spent a great deal of time working with various forms of 3D modelling and CAD though I prefer to exercise mechanical skills in building models. Because I do have some of those skills, I know that computer aided methods are certainly not for dumbos. Nick

Looking good, Dave. I agree with Adrian about getting the rodding in early. I didn't and the need to dig out the ballast every six feet is putting me off getting on with it. The GWRJ article is very good, particularly on the dates of the various length requirements and the layout of compensators. However, I found it disappointing on the earlier round rod types. There aren't many photos or drawings of the early roller boxes and no mention of what was used before the 'A' shaped type. The author is not alone in this as I've not seen any mention of what, for want of a better term, I call the tubular type. These appear to date from the 1870s, or earlier, and are seen in a number of photos, e.g. at Rattery box (p41 in The Great Western in South Devon) or at Brislington (p9 in Through Countryide and Coalfield). I'm sure some of these would have survived into the Edwardian era, but you're probably safe with the 'A' shaped rollers. Bearing in mind that there would be at least one crossover or turnout + trap off-scene to the left, the layout is probably complex enough to warrant a small signal box. It looks like you need about seven levers, including FPLs before considering signals. Signalling might be minimal, relying mostly on handsignals especially if operated one engine in steam. Of course, there are examples of both minimalist and full works approaches on the GWR. It would be interesting to see what Mike (The Stationmaster) has to say on this. Nick

Well, I'm not a chemist and it's getting on for 50 years since I did A level chemistry and last thought about vapour pressures, so I could well be wrong. However, I'm wary of believing that significant levels of vapour are released at 330o below it's melting point. As to 'present in most low melt solders', the last time I looked into this I only found a few higher temperature lead-based solder alloys with, usually, less that 0.5% antimony. However, it is found in quantities up to around 5% to aid flow in some lead-free solders. That said, I understand that some white metals used for casting may contain significant amounts of Antimony. Yes, Carr's 70o solder does appear to be Wood's Metal, though I think you should check the ingredients. Wood's Metal is normally 50% Bismuth, 25% lead and 12.5% each of Tin and Cadmium. As to sublimation of Cadmium, I was aware of oxidation issues in the liquid state, but was under the impression (perhaps wrongly) that sublimation was really only significant at low pressures. Wood's metal is also a eutectic alloy which, IIRC, means that it remains an alloy, not a mixture of separate metals in the liquid phase. Personally, I prefer to use C&L's 100o solder which is a Bismuth, Lead, Tin alloy with no Cadmium. Like you and many others, I'm quite happy to use a normal iron on whitemetal, but I have been soldering for well over fifty years. I am, however, nervous about recommending this to a beginner, particularly if they don't have someone to show them how to do it. I still have some well-melted parts of a K's kit that was my first attempt at around 12 years old, though at that time I hadn't discovered low-melt solders Nick

But, if you had tried it, you would know that it was nonsense, unless you use an underpowered iron or too small a tip. The molten solder boils the flux on contact and that starts the local heating necessary for the solder to bond. Provided the iron can maintain the tip temperature well above the solder melting point, there's no chance of 'freezing'. Oh come on, they aren't really that different I wish you'd tell us some believable reasons for not carrying solder on the iron! As to reasons in it's favour, how about speed, convenience, not needing to spend time cutting up and placing small pieces of solder, and being able to solder at any angle without the solder fragments moving from where you want them. Nick

No problem, do what Churchward did and put it between the frames. That way, you won't need to model it until you've had a bit of practice Nick

Well said, Arthur! As you say, both methods work and it's all down to personal preference. Sometimes I use one, sometimes the other. Whilst I agree with much of what Kenton has to say about soldering, I do wish he would stop repeating this nonsense about 'boiling' solder. Soldering irons are just not hot enough to boil off any of the constituents of any modelling solder. As I've said more than once before, the idea that you shouldn't carry solder to a joint is only appropriate with flux-cored solders and we should only be using them for electrical work. Nick