Building and completing my 4mm 3D Printed GNR Coaches

[MikeTrice]

I have been dreading doing the stepboards and puzzling over how to do them as I was worried they could be quite fragile. So will no guarantee of success thought I would try something out.

 

First off I needed a template for a step hanger and a suitable jig to bend them in if I wanted consistency as there were 20 in total. The jig I thought I would make from styrene thus:

 

I made the mistake of radiusing the bottom edge where the bend is formed so this jig will have to be replaced, probably in brass, however I was keen to press on and see if my idea worked.

 

Using some 0.6mm N/S wire a first bend at 90 degrees is formed and the wire placed in the jig:

 

Holding the wire firmly in place the lower fold was formed:

 

Five wire step hangers later, only another 15 to go:

 

I made an assembly jig using an offcut of timber sticking a template onto it and drilling the top location points with a 0.6mm drill. 5 of the hangers are threaded into the holes. A piece of 0.5mm * 3mm brass is cut to length and the lower edge rounded off to try and compensate for the badly formed lower bend:

 

Somehow I got away with it so the lower step was soldered to the hangers:

 

The top step was then aligned and soldered to the hanger relying on a simple butt joint:

 

That seemed to be successful and on removing the assembly from the soldering jig seemed reasonably rigid so I pressed on and soldered the right hand sections of the footboards resulting in this:

 

On removal from the jig I now have a basic unit. It needs tidying up but looks promising:

 

[Guy Rixon]

Looks good. The jig is definitely the right way to go for S-shaped supports. However...

 

What is the full-side width of the supports? 0.6mm scales to 1 11/16" which sounds a bit heavy to me. I wonder if 0.45mm wire would be closer to scale. I've used this size for step supports before and it's strong enough, IMO.

 

Secondly, I presume that the upper leg of the supports goes into holes drilled at the top of the solebars, with the top of the vertical leg representing the part where the full-size support is bolted to the solebar. I.e. a round section is trying to represent a part that was forged flat on the full-size vehicle. There is another way. If one has solebars with some substance - i.e. not a single thickness of etched brass - then the top of the supports can be bent back about 30 degrees level with the bottom of the solebar and then sunk into holes drilled obliquely into the solebar. The flat part of the support can then be represented by microstrip.

[MikeTrice]

Guy, thanks for the comments. You are correct, the top "pin" of the hanger is inserted into horizontal drilled holes in the solebars. I thought I might file the outer face flat giving a better representation.

 

I really had no idea how strong/flimsy such a construction would be which is why I initially picked 0.6mm n/s. I think I have come around to the idea that I can go to 0.45 or 0.5 and still have sufficient strength. I don't have a prototype width for the hangers. A big benefit of using a thinner wire is as you say, it is probably closer to scale, and it will hopefully allow sharper right angled bends without the obvious curving of the 0.6mm wire. The upper footboard actually sits against the lower edge of the solebar so obliquely drilling into the solebar I suspect might not be viable however I like the idea.  So I will probably remake them using thinner wire however there are some useful experiments I can carry out using these prototypes first. I am also curious as to how good Halford's Etching Primer will adhere to them as they are going to be very prone to paint rubbing off. If all else fail I might heve to chemiblack them.

 

This is what the prototype bracket looks like. Mine is a vast simplification:

[gr.king]

If you are not worried by those trifling matters of cost and availability, along with the need for a more sophisticated set of jigs or a lot more faffing around, then thin and suitably narrow flat strip could be used, bent double and strengthened with solder, to form brackets that provide support for both the upper and lower steps boards, which have flat faces, and which can still be inserted into holes drilled squarely into the top edges of the solebars.

[MikeTrice]

Had not thought of thinner strip doubled up. Might well be strong enough.

[gr.king]

Had thought of this sort of fold up, all soldered:

 

[Guy Rixon]

I was thinking that would be a swine to fold up accurately, but actually if you worked out the lengths between the folds and notched it with a triangular file, as if it were half-etched, then it might be quite easy.

[Bucoops]

Looks similar to the steps you get with Jamieson kits.

[gr.king]

Not sure any notching would be needed. I'd probably begin by folding double in the right place to create what would ultimately be the top board support. A sequence of jigged folds, nipped up tight with pliers then seems quite straightforward, according to the thinking process anyway.

[MikeTrice]

It does not have to be one piece even.

 

Perhaps like this:

 

Edited May 12, 2018 by MikeTrice

[Compound2632]

It does not have to be one piece even.

 

Perhaps like this:

bracket.jpg

 

Looks easier (from my armchair). The horizontal parts could be trimmed to length after the whole thing has been soldered up.

[MikeTrice]

I suspect I will stick with my current simple version albeit reduce the wire thickness to 0.45mm. It seems strong enough and will not be very visible when sprayed black.

[PenrithBeacon]

It does not have to be one piece even.

 

Perhaps like this:

bracket.jpg

Wouldn't it be easier to remove the backing component?

[MikeTrice]

Revised jig. Made from offcut of copper clad and 0.5 * 3mm strip. Gapped for 0.45mm:

 

As before first bend formed with pliers and slotted into jig:

 

Bottom leg bend formed with steel rule. Here the bend is in progress:

 

Bend complete as far as the jig allows:

 

A little tweaking with pliers and one supports looks like this. Much finer than previous effort and sharper bends:

 

19 more to do.

[MikeTrice]

Wouldn't it be easier to remove the backing component?

Depends on the thickness of the strip. I suspect it is still needed for strength.

 

Anyway carrying on with revised first design.

[gr.king]

Sticking (wisely perhaps) to the simplicity of the basic wire Z brackets for the lower stepboards, there are still several positions along the length of the upper board and solebar in which it is possible without fouling wheels or spring gear to place hidden L shaped brackets with the horizontal leg soldered under the board and vertical leg rising behind the solebar, glued to the latter as well as fixed into holes drilled in the floor.

 

Might it be worth sticking to the thicker, stiffer wire for the end bracket on each lower board, the corners being most vulnerable to mishandling and enjoying least support from adjoining brackets?

[MikeTrice]

Sticking (wisely perhaps) to the simplicity of the basic wire Z brackets for the lower stepboards, there are still several positions along the length of the upper board and solebar in which it is possible without fouling wheels or spring gear to place hidden L shaped brackets with the horizontal leg soldered under the board and vertical leg rising behind the solebar, glued to the latter as well as fixed into holes drilled in the floor.

 

Might it be worth sticking to the thicker, stiffer wire for the end bracket on each lower board, the corners being most vulnerable to mishandling and enjoying least support from adjoining brackets?

Interesting ideas.

[MikeTrice]

A few posts back I bemoaned the lack of accurate GNR coach transfers. With no obvious resolution in sight I decided to have a go at producing my own the results of which are shown below:

 

 

 

To say I am pleased with these is an understatement. In best teasing fashion I will reveal how I did them in a later post ;-)

 

[Guy Rixon]

Wow! The transfers look just about perfect. A proper, gilded look, and no sign of carrier film at 2x magnification under strong light. And they look like they are blocked in blue to the left as well as the stronger blocking to the right. Can't really get better than that.

[MikeTrice]

First off here is a prototype transfer so show what the lettering should be like. The main letter form should be gold leaf but on this transfer has been overpainted cream:

 

When reduced to 4mm scale the letters are a maximum of 1.2mm high over the "gold". Producing accurate full colour reproductions would be time consuming and I would argue would not be seen by the naked eye. For this reason the lettering has been simplified into three elements, the "gold" figure, the blue shading and a black drop shadow (exactly as observed by Guy above).

 

With this in mind Inkscape was used to provide a crude letter form:

 

I used the standard fonts choosing those that I considered nearest so for most of the lettering used "Arno Pro" and for the numbers "Century Schoolbook". Note that the "Gold" is actually yellow simply so it will stand out when printed and applied.

 

I then added borders around the letter groups representing the dimensions of the panels:

 

The various panels and letters were duplicated and a few spares added that I thought might be problematical:

 

In order to print the transfers on an inkjet printer I need to be able to print on a white backgound and then either cut the resulting letters out or paint out the unwanted white. Instead I decided to print the teak background as part of the transfers so before I could continue I needed a painted teak grained sample that matched the painted coach. This was painted using exactly the same technique as the original coach:

 

The sample was scanned and cropped. I knew from previous experience that trying to match a printed colour to a painted one would be difficult. Fortunately if the panels were slightly out it would not matter, but I needed to get close so I produced three variants of the scanned image with increasing amounts of yellow present and imported them into Inkscape:

 

I initially printed this out on some vinyl labels but they were not a success when applied due to the thickness of the label. Instead I needed to print them on white decal paper. Now in the past this has not always been successful and I was not overly hopeful but pressed on regardless. The decal paper used was bought on ebay some time ago:

 

The decals were printed on my Canon iP7250 and came out very well:

 

When the ink was dry the whole print was sprayed in Humbrol Acrylic Satin Varnish and when that had dried brushed with Klear.

 

In the past when printing on white decal paper I have had problems with the cut edges of the transfers bleeding when placed in water. Fortunately it did not happen with these and they went on extremely well. I did use Microset in order to ensure they were stuck down firmly and on drying have since been given a coat of Klear to seal them to the coach.

 

Hopefully, now I can continue with the weathering as and when time permits.

 

In conclusion I am really happy the way these came out so I am now in a position to produce as many GNR transfers as I need and extremely cheaply.

 

 

[chris p bacon]

Looks really good Mike.

 

I'd found the Arno Pro lettering by chance when looking for another typeset, but was still undecided about lettering.

[MikeTrice]

The full brake has now been weathered with the oil and pin washes. Happy with this:

 

Not so happy with the roof, too streaky:

 

Should have been airbrushed so I suspect I will clean off the streaky finish and spray the roof to give a smoother finish.

 

[MikeTrice]

Oil wash applied with airbrush. This is more like the effect I was after. A sooty finish with the white just showing through:

 

[MikeTrice]

Current state of play following a coat of Humbrol Acylic Spray Matt Varnish:

 

[Compound2632]

The epitome of period dinginess. I do like the way the previously rather garish solebar has been blended in to the overall worn and grimy teak look.

 

Just need those footboards for decency now!

Edited May 26, 2018 by Compound2632