Nick Dunhill's Workshop. Victorian Models GWR no 34

[Nick Dunhill]

As one job draws to an end, another begins. This time we are returning to the neolithic period to make a GWR no 34 in S7.



On first inspection the etches seem to be ok, but you can never tell until we begin nailing them together. The client had been very thorough in sourcing all the other parts required, and also supplied drawings.



As always I began with the rods. They were supplied by Premier Components, but the client requested that I beef up the bosses with an overlay of thin nickel silver sheet, filed to shape and with an added a brass oil pot bung.

The chassis frames and associated components were dressed up, but new stays had to be cut to convert the chassis to S7.





The rear driving axle would run in a fixed top-hat bearing and the front axle in a Slater's cast brass axlebox to add some compensation. It is worth reporting that the centres of the coupling rod and the centres of the etched bearing holes in the frame were a mile out....well around 0.75 mm. This would result in a jammy mechanism if not altered. It didn't matter in this case as we could easily locate the axlebox in the correct location using the rods and some jury axles.



This resulted in a nice free running chassis, which I levelled on a sheet of plate glass, and made the bogie. All runs very nicely, and traverses a 5' 6'' curve with some to spare.



More next week.

 

[Tim Humphreys ex Mudhen]

Hi Nick,

Were the rods or the frames the correct measurement ?

Tim

 

[Nick Dunhill]

The rods were the correct length. 6ft 8 ins 46.6666mm

 

[john lewsey]

Hi Nick , who is the manufacturer the kit.
John

 

[Nick Dunhill]

Hi John, it's Victorian Models.

 

[ianlbsc]

It is showing on Roxey Mouldings pages. After a google search. Not yet available. Cheers, Ian

 

[Nick Dunhill]

We have bashed on with the model this week. I added the ashpan etches to the chassis to stiffen it all up a bit more. Of course because the chassis had been widened to S7, the ashpan rear needed to be widened too. There were a number of additions to the chassis that would be easier if the footplate was added first, so I made that. The etches were all good, and fitted well.





I added the splasher tops and also made some splasher rears so the spokes will not be on view under the boiler. There are splasher top etches in the kit, but it is advised that they fit behind the sides and not on top of them as they should be. I made replacement splasher tops, and this turned out to be essential. The instructions advise the use of smaller 4ft diameter Slater's wheels, whereas the loco had 4ft 1in wheels. The instructions did go on to say that 4ft 1in wheels would fit if the wheels were reprofiled to S7 standards. They did not....they touch the splasher tops, but only by a tiny amount, which would have been much worse if the splasher tops in the kit had been used mounted behind the fold up sides. The wheels have gone away to have 0.8mm skimmed off the diameter, so will fit when they return.

I batted on with the brake rigging (before the wheels were sent away for modification) and the hanger and brake shoe etches, although awkward to assemble, made nice parts. The pull rods supplied had a joggle in them around the wheel centres to match the locos in their full cab state. Mine will have a half cab so the pull rods had to be replaced with straight ones.



There's a whole rear section of the brake system not in the kit. Also the trunnions for the rear brake shaft seem to be in a section of the chassis side frames that are not represented in the chassis etch either. There wasn't the budget in the kit for a radical redesign of both chassis and bogie, so I mounted all the rear section of the brake system on a tube that slots over the bogie pivot. It is all a bit Heath Robinson-esque, but serves the purpose.





After adding the pick ups and sand pipes I went upstairs, back to the footplate. GWR bunkers and tenders are a difficult to make. The kit had a panel for the bunker. It has 'fingers' for the corners, and it is etched in 0.45 mm brass. There is no way that it is possible to form the flares in such thick material without a hammer. That's not my MO, so I cut a new panel out of thinner nickel silver. There is no information in the instructions about the radius of the curved back corners, and hence what the flared top should be. The instructions say 'bend round a drill' without saying whether that is an 0.4 or 20 mm drill. I guessed and got it wrong, so had to cut out another panel.



Radius too big!



Much better!



Fingers flodded with electrical solder and filed to shape before fitting the bunker to the footplate.



Next week I'll finish the coal rails and move on to other fabrications, probably the boiler, as there are no guide lines about where to mount the cab front. I'll build the boiler and work my way back. BTW the coal rails were one single etch with the upstands. I cut them apart, and reprofiled the rail etches so that they looked like half-round.

 

[Nick Dunhill]

I quickly added the coal rails and cracked on with the boiler assembly. The coal rails are tricky to get level and vertically aligned.

The suggestion in the instructions is to solder a circular former for the boiler to the rear of the smokebox front. In this way the boiler mounts inside the smokebox. I really didn't like this as a plan, also the boiler former is way too small. The kit has an extra smokebox former, so I made a base to connect the front and rear smokebox formers, and cut some strengtheners from scrap strip. I formed the wrapper to fit and soldered up.



I decided to make the firebox next. The firebox front is a brass casting to represent the brass cover on the real loco. It has a hole to accept the boiler which is vaguely circular. There is no former for the firebox other than the casting, and the instructions suggest that the builder form the wrapper and solder it to the casting, and somhow the rear of the firebox would be the correct shape and stable. I thought that this was a rubbish plan and made a former by scribing round the rebate in the back of the casting. I formed the wrapper round my rear former and tacked them together. I then tacked the cast front on, and adjusted and adjusted until I had a square and nicely formed firebox. I made a larger firebox former to plant on the rear of the firebox, and soldered a boiler band up against it. This would represent the L section that is bent round the rear of the firebox on the cab front.

As mentioned above, the former in the kit for the boiler is way to small in diameter. I think that the instructions suggest that you roll the boiler and fit it inside the firebox casting. The smokebox has a cast brass ring to represent the cover between the boiler and smokebox rear. The implication is that the front of the boiler wrapper should be soldered inside this ring to form a cylinder, along with the hole in the firebox front casting. The holes in both castings are a different diameter so.....

If you are struggling to follow this, so was I. I think that the designer has never built this kit, and the instructions are a bit of a word salad. There are no construction drawings. I cut the front off the boiler wrapper (remember it is designed to be inside the smokebox) and made some circular formers to the correct diameter. I made a conventional tube for the boiler section and soldered it square to the rear of the smokebox. The boiler telescopes inside the firebox front, but I figured I could make it the correct length when the cab was complete.



I added all the boiler bands I could and cut a section out of the brass finishing ring (between boiler front and smokebox rear) to reduce it's diameter and make it fit.

I thought it would be a good plan to make the cab/water tanks assembly next. The cab/tank sides have to be curved to form the tank front. There is a notch at the top and bottom of the panel to suggest where the curve is formed, and the instructions indicate that you should form the bend round a drill....again the instructions don't suggest what diameter drill to use. I was losing patience with the instructions. I soldered a strip of scrap just to one side of the notches and ran a slitting disc along the edge to form a fold line. I made the 90 deg fold, flooded the back of the fold with solder, and filed the front to the correct radius. The only guide to the radius of the curve is the beading strip that you have to plant on top of the panel. Anyway, I won in the end. The panels for the tank tops need to be narrowed to fit, and they turned out to be too short later. The firebox will be captured between the tanks when the cab assembly is complete, so I checked that the boiler/firebox/smokebox assembly sat level before going any further. Luckily it did.



The cab front is too wide and has to be narrowed to fit (would have been worse if it was too small!) Also the registration slots for the cab front are in the wrong place on the cab sides. I tacked and checked, and adjusted and checked again, until I was satisfied that everything aligned properly (the cab front is also slightly too tall.) The firebox was now captive, and I checked alignment and soldered it to the cab front. I soldered it all up and cleaned away the excess. I pushed the boiler inside the firebox front, and it clattered into the motor. I had to remove the former, shorten the boiler wrapper and push the former back in. When it eventually all fitted I was able to score a line and add a boiler band to disguise the ill fitting join.



I was happy with the model now the main superstructure is complete. I'm lucky that I'm a reasonably competent builder otherwise I would have struggled. Obvoiusly no one bothered to do a test build before putting this kit on the market, or if they did they couldn't be bothered to alter the ill fitting parts. Perhaps they should offer to refund my client for the hours I have spent making all the random parts fit.

Poor kit, but not the only one on the market. I am struggling to hit the budget.

 

[adrian]

If you are struggling to follow this,

Just a little - it seems to be a little over complicated to me. I've read the notes but if you hadn't shown the final assembly I'm not sure it would have been achievable. I'm going back to my scratch building - at least then I know what the designer was thinking and if it doesn't work I can only blame myself.

 

[simond]

And despite it being early in the season, the WT “ Master of Understatement” Award for 2025 goes to …

I'm a reasonably competent builder

 

[Overseer]

As we expect from 'reasonably competent builder' Nick the result will be lovely. Building things for fun is one thing, keeping going when not good parts are being built on commission is another. I really admire the effort Nick puts in to build models of the prototype locos rather than just assembling kits.

One question though, why does the bunker have a solid top on it? Surely the back tank was in the lower part of the bunker with the raised filler through the coal space above the tank.

 

[Eastsidepilot]

And despite it being early in the season, the WT “ Master of Understatement” Award for 2025 goes to …

...if the people who design and sell kits were competent builders as well !

 

[Nick Dunhill]

Thanks for those kind words Fraser. That's the rub with doing this kind of thing professionally. You have to come up with a solution to a problem quickly, and keep it on budget.

It has a solid bunker top as the brief is to cover it all with coal after paint.

 

[mickoo]

I know I keep banging on ad nauseum about 3D (less so recently), but without them projects like this would simply be too costly to replace everything that's wrong with traditional tooling and materials.

As tends to happen when working with Nick, it starts out with a single item, but then once you replace that then the adjoing items look incorrect and in a snowball effect you end up with a reasonably hefty bag of replacements winging through the postal system.

There are still some limitations to the resin medium with regards to material thickness, certainly at thin flanges; most of that is down to the print process, less than 0.3 mm and it'll buckle or warp, print at 0.5 mm and it'll be fine and once cured you can trim back to 0.3 mm.

I get a lot of artwork that's to scale and it's a steady email steam to get it changed to something thicker (unprototypical) so that it will at least print in the first place....and then post process if you feel the need to.

 

[JimG]

There are still some limitations to the resin medium with regards to material thickness, certainly at thin flanges; most of that is down to the print process, less than 0.3 mm and it'll buckle or warp, print at 0.5 mm and it'll be fine and once cured you can trim back to 0.3 mm.

Mick,

I ran into that problem a while ago - from the other end. I had drawn up an S scale STL file of a 16T mineral wagon and @ScottW of this parish asked for a copy of the file so that he could get some printed for himself. Scott came back to me saying that the end door detail was not printing properly with the angle iron detail being frilled. This detail was 0.2mm thick and had printed well on my Phrozen Shuffle using Phrozen resin. So I thickened the detail up to 0.3mm and I think Scott got a successful print.

A few months ago I was asked to do some more prints of the same wagon for the SSMRS Parts Department and I dug out my original file and set it up on my new printer, an Elegoo Mars 4 Ultra. I had changed to using Fast Naval Grey resin having seen reports on it in WT and my first wagon print came out with the end detail frilled. I still had some Phrozen Aqua 4 resin, so I changed to that and the 0.2mm door end detail printed perfectly. So it looks as though the resin is a factor on the reproduction of small detail.

I'm just drawing up another wagon for printing and I'm going to experiment with going down to 0.15mm thickness

Sorry for thread drift, Nick.

Jim.

 

[mickoo]

Mick,

I ran into that problem a while ago - from the other end. I had drawn up an S scale STL file of a 16T mineral wagon and @ScottW of this parish asked for a copy of the file so that he could get some printed for himself. Scott came back to me saying that the end door detail was not printing properly with the angle iron detail being frilled. This detail was 0.2mm thick and had printed well on my Phrozen Shuffle using Phrozen resin. So I thickened the detail up to 0.3mm and I think Scott got a successful print.

A few months ago I was asked to do some more prints of the same wagon for the SSMRS Parts Department and I dug out my original file and set it up on my new printer, an Elegoo Mars 4 Ultra. I had changed to using Fast Naval Grey resin having seen reports on it in WT and my first wagon print came out with the end detail frilled. I still had some Phrozen Aqua 4 resin, so I changed to that and the 0.2mm door end detail printed perfectly. So it looks as though the resin is a factor on the reproduction of small detail.

I'm just drawing up another wagon for printing and I'm going to experiment with going down to 0.15mm thickness

Sorry for thread drift, Nick.

Jim.

That's interesting information, the PA4 seems better at finner detail than FNG but I wonder if it then suffers in surface smoothness.

I tried some Elegoo ABS like resin and it was much smoorther than FNG but the details were less crisp.

Upshot, choice of resin is important for the object you're printing. I've used exclusively FNG since Big Train Jim mentioned it but I've just got a new bottle (Phorozon Aqua as it happens.....does that mean it's water washable? Finding that information illudes me right now, not that I'm a fan of water washables as they tend to be less stable that the alternatives) touted by a skilled 1:32/1;24 aero CAD/3D modeller, the quality of finish on those is astounding; I've yet to find the time to dial it in though.

I probably wouldn't draw something 0.2 mm thick so perhaps have not come across this issue with FNG, certainly not as a flange, I do go as low as 0.1 or 0.15 as a step level so to speak. The piano front hinge plates are 0.15 mm thick and print just fine.

A big down side to detail this fine is painting, hit it with a rattle can and some of this detail risks get flooded out. I know Nick will send the model to Warren or Paul and both lay down very thin coats; that means I can go much finner with the detail.

 

[Nick Dunhill]

Looking forward to getting those mate. No problem with thread drift.....

 

[Nick Dunhill]

Well I got my head down and pushed for the finishing line this week. What could possibly go wrong? My attention turned to the cab. The roof went on and I assembled a typical Dean backhead using castings from Scorpio and somewhere else, possibly Warren Shepherd.





Then I moved towards the front of the loco and did the handrail. It's always tricky to find handrails knobs that are exactly long enough to bridge the step between the boiler and smokebox, but I lucked out! It is also tricky to make the handrail symmetrical.



Then the 3D turned up and I glued it on.



The dome is there for proportions, and is actually off a CR loco. But wait, the chimney looks too high, and out of proportion I thought. Let's do a bit of measuring and compare with the drawing I have (not a GA, more like a poor quality Skinley.)

OM(F)G!!

The boiler as I'd made it, as supplied in the kit, as designed by the manufacturer, was 2 mm too small in diameter. The boiler diameter is dictated by the hole in the front of the smokebox. The hole is in a brass casting there to represent the brass cover on the throat plate, so is fixed. The boiler is too small, and by association the smokebox is also too small in diameter and height.

I thought the job was finished, but a defeat was clawed from the jaws of victory by the bloody kit designer. I spent a whole evening trying to convince myself that it was ok, that it would look fine when it was painted, that the client wouldn't notice. But I bloody did, and I knew it was wrong, and so this morning, after a patchy night's sleep dreaming of boilers, I got up super early and I made another boiler tube of the correct dimensions. It was only 2 mm bigger in diameter than the kit junky one, but what a difference. I offered it up to the rest of the body and CLICK! All the proportions of the real loco in my photographs appeared on my model. I have spent a very long day today scratchbuilding a boiler and smokebox for free. The budget was exhausted so I had the choice of either doing it for nowt or sending out a crap model.









Just goes to show what a massive difference 2 mil makes!

So in conclusion this kit goes in the Junk category for me. It is resting in the 'Scrap bin o'shame' with the Alphagraphics SECR P class, the ACE NBR Atlantic and anything designed by Meteor.

 

[40057]

Tough decision for you. But 2 mm diameter means a heck of a lot more volume, and that’s what the eye sees. Looks very different, so on that score worth the work.

 

[Ian@StEnochs]

Nick,

Super work as always.

In my book there is only two ways with kits.

1, Assemble the kit, warts and all. Usually fairly straightforward but needs the builder to do a bit of bodging and ignoring the dimensions on the scale drawing/GA, which is probably not supplied anyway. It may not look exactly like the prototype but it should be recognisable but don’t look too closely.

2, Build a model of the prototype incorporating the bits from the kit which make the task easier. This requires a more informed approach, a bit of research, a GA, lots of photos plus substitution of incorrect parts and perhaps complete replacement/scratch building of a good part of the supplied kit. The completed model represents the original very closely.

I am very pleased that virtually none of the prototypes I want to model are available as kits but you as a professional builder don’t have that luxury.

Ian