Scratchbuilding a 0-8-0 chassis using basic tools

[steverabone]

As I model in S Scale I sometimes need to be a bit inventive (think outside the box if you prefer) when it comes to building models. At the moment I'm constructing an old re-scaled Brassmaster LNWR G2 0-8-0 that I've had lurking in my "to build one day box". It's to run on my small north west of England  London Midland Region layout.

 

The bits I have are the etched parts for the loco body, footplate and also the tender body. I also have sufficient parts for the tender chassis and suitable metal tube for the boiler (I may actually use a card tube instead depending on how I feel later) and some castings that I can adapt for the chimney and dome.

 

What I didn't have was  the chassis for the locomotive. Now most people would give up now but I've already built several chassis for four and six coupled locos using the most basic of tools.

By basic I mean a pair of large scissors, some files, a set of broaches, a very small vice and a soldering iron.

 

What follows is a description of how I built a free rolling eight coupled chassis - I power all my locos with tender drive mechanisms. It will also show how I modified standard Alan Gibson wheels to resemble those strange LNWR spoked wheels.

In the early 1960s I saw quite a number of these locos in service. This is Buxton shed's 49406 seen at Miller's Dale on a freight to Rowsley. The photo was taken by my father on my Brownie Cresta box camera as he knew I wouldn't be able to take it successfully. With a slow shutter speed the image  

This fret holds most of the parts for the loco body.

Whilst this one is for the tender.

There is also an assembled outer tender frame with those distinctive steps

There are no loco frames in the parts I have so these will need to be made. For once good fortune stepped in as the G2s have absolutely identical axle spacings to the Somerset and Dorset 2-8-0s.  I have several etchings for the Alan Gibson S Scale model of the 2-8-0 which  means that I will be able to use the coupling rods to mark out the G2 frames accurately.

 Fortunately I do have photocopies of the etches so after resizing these to the correct dimensions I've used these as a template for the frames.

Two rectangles of brass sheet were cut roughly to the correct size and tack soldered together. The paper template was glued to the brass. Then I drilled a pilot hole for one of the axle holes in the correct position. Next using the S&D 2-8-0 rods the other three holes were drilled. This should mean that axle holes, crankpins and coupling rods should line up automatically. I suspect I may end up splitting the rods into three sections, as with four axles to quarter and tight curves to be negotiated, I think it is unlikely the chassis will run smoothly unless I do. Luckily the Gibson kit comes with four sets of rods so this will be a simple task.

I've now cut out the main frames. The cut outs at the ends of the frames were made using scissors. The three cutouts between the axle holes were made by first scoring the brass along the straight lines with a heavy duty Stanley knife and then snipping the short diagonal lines with scissors. The small sections of brass were then broken loose using a pair of pliers to bend the metal until it broke. The corners were slightly rounded with a file. It's a crude but reasonably effective method. The axle holes will now be opened out with broaches and bearings fitted.

 

TO BE CONTINUED

 

 

Edited October 27, 2019 by steverabone

[steverabone]

PART TWO

The real problem building a LNWR 0-8-0 is, of course, the highly unusual driving wheels. Having looked at the 4mm scale Brassmasters kit (where a etched brass overlay is glued to Gibson wheels) I decided to experiment by printing out an S Scale sized drawing of a G2 onto thin card. The card was then coated with wood hardening fluid and later I cut out the space between the spokes.

The card was then glued on to the correct size wheel with epoxy. A thin smearing of epoxy was later added to the top of the card and also underneath the card between some of the plastic spokes.

 

The basic footplate structure has been assembled. The lower photo shows the valances which are lengths of code 45 rail soldered to the underside of the footplate. The bufferbeam and drag bar have been added and the front and rear spacers and their associated nuts and bolts are in place.

The frames have been strengthened using the same method I adopted for some of my other locomotives; lengths of double sided copper clad sleeper strip have been soldered along the top edge of the frames. This isn't really essential as I could simply have used more frame spacers to get the same results.

Rather than use top hat bearings (which I've run out of) I soldered axle washers to the axle holes in the frame. These were later filed much thinner to allow sufficient side play and the middle two holes were elongated at the top by filing to allow the centre wheels to rise slightly over rough track.

The front of the wheel splashers have been bent up and Alan Gibson wheels have been temporarily fitted to the chassis. When being propelled by a powered tender this revealed (as expected) some spectacular arcing as the backs of wheels touched the back of the splashers. Otherwise the chassis rolls well.

I've now added the brake hangers and the wheel on  the third axle has had its flange filed away. Smaller brake hangers are need between the second and third wheels as the wheel spacing is much closer here.

Viewed from underneath, the flangeless third wheel is more obvious. Careful filing can produce a wheel that doesn't have its tread damaged. To avoid the flangeless wheel cause problems catching on check rails or simply falling off the rail on curves I found it necessary to limit the side-play on both the third and fourth axle to a  minimum.

The lengths of copper clad sleeper soldered to the brake hangar cross wires will be used to provide current pickup form the outer axles. The copper is gapped to give isolated sections of copper behind the wheels.

The chassis is now complete and rolls smoothly. The rods are in three sections with the S&D 2-8-0 rods cut to overlap at the crankpins. I placed two very small washers over each of the crankpins to ensure that the back of the rods don't rub against the card spokes.

The wheels after painting one side of the chassis black. The loco chassis has now been tested with a powered tender and rolls smoothly around the layout, including a section of 36" radius curves.

 

Hopefully, this description might encourage others to have a go at scratch building a chassis. It would have been even easier if I had used Markits wheels with S Scale axles as these have automatic quartering.

 

Edited October 27, 2019 by steverabone

[steverabone]

Part Three

I've now assembled the tender. The power unit has a single large Mashima motor with a long drive shaft and drives two gear boxes on the outer axles of the tender. The profile of the motor and gearboxes is sufficiently low as to allow the etched brass tender to be built as intended.

Construction of the tender body was not that easy as the flared upper part of the sides and ends have to be assembled from five different parts soldered to the top of the sides and ends. The Midland Johnson tenders for Alan Gibson are considerably easier to assemble.

Edited October 29, 2019 by steverabone

[Lacathedrale]

I found your original premise using the Royal Scot tender drive units to be really unique - I gather that it's less a change in requirements and more a refinement of the solution that you've now gone to a more standard motor arrangement? It seems that the arrangement of a motor in the tender with a universal joint powering a gearbox in the loco is a pretty standard solution in 2mm - but I gather this is more a concern around the space available than a need specifically to drive the loco driving wheels. Certainly the premise would appear to make something like a single loco more approachable!

 

I don't know why I've always just had it taken as a given that the loco wheels 'should' be driven. Presumably in the way that steam power 'should' be the driving force, I hear you say :)

 

It's really interesting how that when painted black the paper façade on the wheels is really effective. I imagine a dry brush with a very slightly lighter shade of black (or brown, muck) would highlight the edges more and bring them forward visually?

 

 

 

 

[steverabone]

The reason I changed from using Airfix Royal Scot tender drive units to power my locos was twofold.

 

The first was that I wanted to motorise a Stanier LMS tender which has a longer wheelbase than the Midland Railway tenders. So whilst the wheelbase of the Scot is almost spot on (in S Scale) for a MR tender it is far too short for an LMS tender. The second reason was that I wanted to improve slow running which could be a bit patchy with the Airfix units. After looking at various solutions I eventually came up with the idea of using two Mashima 1620s and Comet 1:38 gearboxes. More recently I have used Mashima 1624 motors with Markits gearboxes driving the outer axles of the tender. The result is a very powerful and slow running mechanism so all my recent tender locos have this two motor arrangement.

 

The drive in the LNWR 0-8-0 was originally in one of my Midland tenders. It has a larger Mashima motor with a much longer drive shaft. The gear box mountings have been modified so that the motor drives two gear boxes.

 

You might ask why use tender drive mechanisms rather than having a motor in the loco? Much of my earlier modelling was of German prototypes where  locos were driven by the tender with the loco free rolling. Particularly if you are using Gibson wheel without automatic quartering and with relatively flimsy crankpins and plastic centred wheels I have found it a lot easier to set up the loco chassis as a free rolling one. It also means that there is no need to worry about cutting out the bottom of boilers etc. to accommodate a motor. Given that our steam locos are powered by electricity I feel that it is irrelevant where the motor is, as long as the driving wheels and valve gear rotate freely so the appearance of the loco is not compromised.

 

A further advantage is that you can pack the tender with lead and it doesn’t matter whether the locos is a 4-2-2, a 4-4-0, a 0-6-0, a 2-6-0 or a 4-6-0 its running characteristics will be exactly the same.

 

Your point about weathering the wheels is spot on and will happen once I’ve finished the model.

 

This page on my website describes the tender drives in more detail:

http://www.steverabone.com/sscalewebsite/indexpage.htm

Edited June 15, 2020 by steverabone

[Regularity]

I use a GW models tool for quartering, and have had a grand total of fewer than 1 problems with quartering since.

 

Quote

 

I don't know why I've always just had it taken as a given that the loco wheels 'should' be driven.

 

If "it" works for you, then whatever "it" is, is the right solution.

Edited October 29, 2019 by Regularity

[steverabone]

A bit more progress with the cab, smokebox and firebox parts assembled and a boiler temporarily in place. The boiler is a plastic waterpipe with about six layers of card wrapped around it and secured with layers of epoxy resin. The card is then  hardened with wood-rot fluid. It slots into the smokebox wrapper at one end and at the other end is located over a Swiss 10 cent coin glued to the front of the firebox. Amazingly the coin was a perfect fit inside the tube.

 

The age of the Brassmaster etched parts is obvious because the smokebox wrapper is too long on each lower edge by about 2mm ( a photo on the Internet of a 4mm version shows the same issue) and there are no half etch lines on parts that need folding. A score line with a craft knife helps here.

 

Edited June 15, 2020 by steverabone

[Lacathedrale]

Presumably the layers of paper are to achieve the correct diameter? I've never tried scratch-building a loco so I'm sure there's an obvious answer - but would it not have been simpler to roll a boiler from brass sheet?

[Regularity]

5 hours ago, Lacathedrale said:

Presumably the layers of paper are to achieve the correct diameter? I've never tried scratch-building a loco so I'm sure there's an obvious answer - but would it not have been simpler to roll a boiler from brass sheet?

 

On 29/10/2019 at 18:12, Regularity said:

If "it" works for you, then whatever "it" is, is the right solution.

But i repeat myself...

[steverabone]

You are correct in saying that the layers of thin card are wrapped around the plastic tube until the correct external diameter of the boiler is reached. I actually found this a remarkably easy way of making a perfectly round boiler without any of the problems of rolling a brass boiler from sheet - it's very easy to get creases in the brass and as I don't having rolling bars my method is far easier. It took less than  an hour to cut the tube to the right length and roll the layers of card around it, followed by a coating of wood hardening fluid. Because each layer of card has a thin coating of epoxy resin (I use Devcon) underneath it the resulting boiler is extremely rigid. The coating of wood hardener makes the card almost as strong as metal but is far easier to drill holes for handrails than sheet brass.

 

I've used almost the same method to make taper boilers for an S Scale LMS 8F and Class 5 built from Alan Gibson kits that had no boilers provided with the brass and nickel silver etchings. First I built up the smokebox layers as for the 0-8-0. To make the taper section of the boiler I first wrapped 10 mm wide strips of card (usually about 3 or 4 layers) around the plastic tube at the firebox end and another 2 layers about the mid point. Then I cut a piece of card long enough to cover the entire boiler barrel from the smokebox to the firebox and roll this around the boiler until the required diameter at each end is reached. The layered strips automatically provide the correct taper. Any imperfections or gaps are simply filled with a smearing of epoxy resin used as a filler.

 

These methods may not be the "approved" way of doing things but once the coat of paint is on I doubt anybody could spot the difference. In case anybody doubts the longevity of card I have several rakes of Midland Railway coaches made entirely from  card nearly twenty years ago and the still look as good as when first built.

 

 

 

[Lacathedrale]

Please don't think I was saying they were wrong - I was literally just curious! (As regards to card, I've just come across your first article about it in the 2000's SSMRS newsletter)

[steverabone]

19 minutes ago, Lacathedrale said:

Please don't think I was saying they were wrong - I was literally just curious! (As regards to card, I've just come across your first article about it in the 2000's SSMRS newsletter) 

 

 

I didn't for one moment think you were saying I was doing things wrong! As you can probably guess I'm a real bodger as far as modelling is concerned so I often build things in unconventional ways - look at my fondness for tender drive rather than loco drive!

 

[Regularity]

The late Norman Pattenden wrote an article about building his MSWJR 2-4-0, entitled "How I made No. 10". It is in an old Newsletter somewhere, I would guess it to be in the first half of the 1990s, but I might be completely wrong.

 

Anyway, in this article, he expounded on rolling a boiler from 0.010" nickel silver sheet, and making sure it was round by tying it with iron wire around circular discs and soldering it, whereupon any kinks disappeared. 

[Regularity]

Alternatively, College Engineering supply brass tube in increments of 1/8" (scale 8") diameter. If you want something in between, take the smaller size and cut a slit lengthways, and open it out slightly.

(Usual disclaimer.)

[steverabone]

I've been able to spend quite a bit of time on my G2 0-8-0 and have now completed the construction of the loco and sprayed it in grey primer.

This photo shows up all those niggling faults but overall I'm pleased with the result. The chimney is an S Scale Society chimney (I think) reduced in height and narrowed at the base using files. The dome is a Gibson MR one with the top rounded more to suit the 0-8-0's profile. The smokebox door is a disc of card made into a dish- shape with different sized circle underneath it. The less than perfect handrails around the smokebox could do with a bit of straightening! Lots of bits of brass wire and microstrip have been used for the rest of the details. The most important thing is that the loco runs nice and slowly, even before I've added any pickups to the loco wheels and will pull a good sized train.

 

 

[Regularity]

Very nice, Stephen.

Undoubtedly one of your best works.

[steverabone]

 

Well the model is completed apart from weathering. I've numbered it 49428 after one I saw at Carnforth shed in 1962. The only photo I have of it doesn't show which crest it carried on the tender but I've decided on the late crest as several G2s that survived into the 1960s did receive one.