Building and completing my 4mm 3D Printed GNR Coaches

[MikeTrice]

A number of people have very kindly ordered some of my recently released 3D Printed GNR 6 wheelers: http://www.rmweb.co.uk/community/index.php?/topic/111744-mike-trices-shapeways-shop-gnr-6-wheel-carriages-now-available/page-3&do=findComment&comment=3059646. They are body only kits and require various bits and pieces to complete, including providing a suitable 6 wheel chassis. Some time back I wrote a topic covering 6 wheel chassis: http://www.rmweb.co.uk/community/index.php?/topic/112237-6-wheel-chassis-for-gnr-coaches-oo/ however there have been a few additional third party products that have come on to the market that I should include as an option.

 

So the purpose of this topic is to initially show how I intended the 3D printed kits to be assembled, and then go on to cover the full detailing of them.

 

The kits are only available in Shapeways' Frosted Ultra Detail plastic which is a resin based produce cured via UV light. As the Dia 303 Full Brake is marginally easier I will start with that.

 

As supplied the kit comprises Roof, Sides, Solebars, Ends and Lamps:

 

My photos show the supplied moulding after spraying with a coat of Halfords' Filler Primer (necessary for photography) however before any attempt at painting they need cleaning (for which I do not have any photos).

 

The mouldings arive in a plastic zip-lock bag and are slightly translucent. The moulding process uses wax as a support material and this is removed by Shapeways before shipping the produce however some oil (and potentially wax) may still be present which paint does not like adhering to. The first thing I do is bath the mouldings in a warm bath of water with washing up liquid and leave to soak for 1/2hr then gentle brushing with an old toothbrush under running water.

 

The next process is to degrease further using a bath of either acetone (nail varnish remover) or IPA (Isopropal Alcohol). For the acetone I leave soaking for no more than 5 minutes, but for the IPA they can be left longer as the IPA does not appear to affect the resin. A final brush with a toothbrush completed then left to dry. When dry and clean the moulding will be more a white finish rather than translucent. Note: I was over zealous with the toothbrush and managed to break one of the roof light supports. As a result I have thickened them up on the masters.

 

When dry they can now be given a coat of primer resulting in the starting image above. It would be possible to assemble first before priming but I find it much easier to see what I am doing once primed.

 

The first step is to provide a floor from 40thou styrene. Measure the width at the bottom of an end not including the rebate. It is hard to see in this photo but for the Dia303 this is 27mm:

 

Now measure the length of the solebars. Again for the Dia303 this is 113.5mm:

 

The cut styrene is checked (and adjusted) against the mouldings:

 

The ends needs a little clean up as there may be tab marks present which will not allow the parts to fit properly:

 

Careful cleanup with files (I find emery boards from chemists ideal) removes them:

 

Assembly can now start. Epoxy or Superglue can be used for assembly. I started by using a rapid epoxy but got frustrated with the faffing around so went over to superglue. A piece of melamine faced board was used as an assembly platform with two engineer's squares. The end without steps was located on top of the styrene floor and glued in place:

 

 

As this was a brake vehicle I also superglued a small triangle of styrene to reinforce the end.

 

The first side is now stuck in place. The ends are designed to rebate into the sides which I will try and illustrate in a moment:

 

Now the second side can be added:

 

A poor photo trying to show the rebate (and the reinforcing styrene triangle):

 

With both sides in place t'other end can be glued in place:

Edited March 3, 2018 by MikeTrice

[MikeTrice]

Second attempt at showing the rebate arrangement:

 

With the glue set the floor was sanded to ensure it was flat and remove any odd superglue. Again an emery board was used:

 

The solebars need to be fitted the right way up. Locate the little triangle of moulded nuts and notice which way around they are:

 

Making sure they are correctly orientated one end is butted against the headstock and a drop of superglue applied:

 

The process is repeated the other end and then the middle section fixed in place:

 

Note that this is the correct orientation of the solebar in relation to the body:

 

The second solebar is now fixed:

 

At this point I decided I really wanted to strengthen the top of the ends. Yes I know it would have been easier earlier on but it was actually straight forward (although I will do earlier in future). Some strips of 40thou styrene were superglued in place:

 

They were then roughly clipped to shape and sanded to match the end profile:

 

Some filler was applied to cover the joint between headstock and solebar:

 

Then the filler cleaned up. Again I found a cut off portion of the nail file useful:

 

The body was then touched up with the Filler Primer on the solebar joints and the end seams. I really need to redo this joint, the rest are much better:

 

 

With the roof temporarily put in place. Note that the roof with the raised portion at the end goes with the end with the steps:

 

Next up the third, but it will not be a for a couple of days as I have other commitments.

 

 

[gobbler]

A very nice looking vehicle.

 

Will looking at how your three axles for the 6 wheels are configured as there are a couple of GER 6 wheelers I'm interested in scratchin'

 

The beading looks especially fine

 

Scott

[N15class]

Looks good, how do you smooth of the ridges on the sides without losing all the crisp paneling.

[MikeTrice]

All pictures so far show the mouldings as supplied with just a coat of Halfords' Filler Primer. Even so some layering can be seen in the lower panels. A fibreglass pencil works wonders followed by another coat of primer. Will add some photos later showing the result.

[PenrithBeacon]

I have often wondered about the issue of the surface finish on 3D prints. I'm looking forward to seeing the 'workshop technology' aspects of building these.

Thanks for the thread.

[gr.king]

Thanks for what you've done so far Mike.

 

When I used Bill Bedford's FUD cab and boiler print to change my Bachmann O4 into an O4/5, I too went beyond the "gentle soap and warm water" degreasing method, seeking to ensure that I had no adhesion problems for glues or paint. After cleaning the print in solvent I did then find that there was a lot of surface texture / layering to hide or eliminate. I employed gentle scraping of the surface and layers of primer to reduce and hide the ridges. It has been suggested to me that solvent cleaning aggravates the problem of surface irregularity by dissolving out more of the trapped wax in the structure of the print, and that Bill's original instruction for use of only soap and warm water was quite adequate. Obviously there's no point discussing this question at length here as it detracts from the intended topic, but advice may well be found, elsewhere on this website, on what is the "correct" degree of degreasing vigour .

 

A few years on, that painted FUD print still seems to be in good order.

[joppyuk1]

Thank you for this thread. I've ordered one of your kits after seeing them in the GNN, and wanted to investigate if they would be suitable for a currently projected layout.

[MikeTrice]

As promised I have rubbed down the "banding" with a fine fibreglass burnishing brush:

 

And after a coat of primer:

 

Not perfect, but I am in danger of losing the panelling detail. Now it might be advancing old age, but I really struggle to see these bands with the naked eye. Bear in mind these photographs are showing a huge enlargement of the mouldings.

 

The third:

 

 

As mentioned previously I sprayed the sides with Halfords' Filler Primer straight away so people could see the mouldings properly as they come. If I were building these from raw mouldings I would use the fibreglass pens BEFORE spraying. Catch 22.

[gr.king]

I think I'd stick to my favoured "really sharp square edge on a straight edged scraper" drawn across each panel to shave the ridges off without abrading the surrounding beading. I imagine the end of a well honed large jeweller's screwdriver would be a convenient width for shaving the flats of those panels.

[PenrithBeacon]

Would the surface finish be improved if these kits were offered in FED?

[MikeTrice]

Would the surface finish be improved if these kits were offered in FED?

Probably. With their current configuration they exceed the bounds for FED/FXD so would need to be split into two mouldings.

I think I'd stick to my favoured "really sharp square edge on a straight edged scraper" drawn across each panel to shave the ridges off without abrading the surrounding beading. I imagine the end of a well honed large jeweller's screwdriver would be a convenient width for shaving the flats of those panels.

I have such an implement. I think I will try stripping the paint off and attacking the moulding with it to see if it works

[N15class]

This is where another 20 years of development is needed before the smooth finish of a coach side is available. I would think the time spent getting the sides and ends smooth makes it an expensive way of getting coach sides.

 

Shame as it would be a good way to get some of the LSWR coaches I would like

[JCL]

I suppose it depends what you like doing and whether you are after a finish like the brass sides. It doesn't take all that long to smooth down, and for me personally, it would be a lot faster than putting together a brass kit.

 

Mike, we have these X-Acto knife blades over here in two widths and one on a slant, I wonder if they would also be useful as they're fairly narrow. If you're interested I could bring some over to try.

 

[MikeTrice]

Thanks but I have one of these babies: https://www.amazon.co.uk/Trumpeter-TSM-9925-Master-Tools-09925/dp/B00B8456MY

[MikeTrice]

My first prototype mouldings had very pronounced banding so as promised I have stripped one back to the base resin using a bath of IPA. In its raw state, after initial cleaning, it is now less transparent but even so it is almost impossible to see any surface finish with the naked eye (or certainly my naked eye):

 

A very light coat of primer was necessary to see what needed addressing:

 

In addition to the band on the upper lower panel there is also bad banding in the upper panels. This has been eradicated on the production mouldings. Here is a primed side together with the Mother of all sharp tools to try and address:

 

The chisel is carefully scraped along the panel removing the high ridges:

 

The vertical panels have also been addressed:

 

After another light coat of primer:

 

The banding is less pronounced and could probably do with a repeat of the process, however is it really necessary? The photographic enlargements show the problem quite well however viewing the real thing it is far less pronounced.

 

Here is one of the prototype mouldings that has been finished in teak livery and other than the upper panel fault (which is not present on the latest mouldings) the lower panels look pretty good almost adding to the grained wood effect:

[gr.king]

Good either way I'd say, the amount of preparation necessary being a matter for the builder's consideration. Do you think the door top vents would be improved were the slots more deeply recessed, or might the material between the slots then just fall to pieces?

[Joseph_Pestell]

My first prototype mouldings had very pronounced banding so as promised I have stripped one back to the base resin using a bath of IPA. In its raw state, after initial cleaning, it is now less transparent but even so it is almost impossible to see any surface finish with the naked eye (or certainly my naked eye):

IMG_3724.JPG

 

A very light coat of primer was necessary to see what needed addressing:

IMG_3726.JPG

 

In addition to the band on the upper lower panel there is also bad banding in the upper panels. This has been eradicated on the production mouldings. Here is a primed side together with the Mother of all sharp tools to try and address:

IMG_3727.JPG

 

The chisel is carefully scraped along the panel removing the high ridges:

IMG_3729.JPG

 

The vertical panels have also been addressed:

IMG_3731.JPG

 

After another light coat of primer:

IMG_3741.JPG

 

The banding is less pronounced and could probably do with a repeat of the process, however is it really necessary? The photographic enlargements show the problem quite well however viewing the real thing it is far less pronounced.

 

Here is one of the prototype mouldings that has been finished in teak livery and other than the upper panel fault (which is not present on the latest mouldings) the lower panels look pretty good almost adding to the grained wood effect:

IMG_3739.JPG

Totally agree with that. The stepping of the 3D print helps to create the teak effect very well. I thought that might be the case after your first pics.

 

For other stuff that needs smooth sides, I doubt we shall be waiting 20 years for the technology to resolve that. 20 months more like it at current rate of progress.

[MikeTrice]

Do you think the door top vents would be improved were the slots more deeply recessed, or might the material between the slots then just fall to pieces?

Yes, exagerating the details is worthwhile, however in the photos in question these are of the prototype with quite a few layers of primer.

[gobbler]

Your last picture of the teak painted vehicle, I think the printing process used enhances the wood grain effect. I at least think this is pleasing to my eye.

 

Can't wait to see the completed model

 

Scott

[MikeTrice]

As promised, the Dia 245 Full Third.

 

The parts supplied:

 

Note that in this instance I have reinforced the top of the ends with 40 thou styrene before assembly (profile still to be filed). Assembly of the body largely follows the process described for the Dia 303 brake.

 

Measure the bottom width of the ends, in this case 27.5mm:

 

And measure the length of the solebars, in this case 126mm:

 

So using a 40 thou styrene floor 27.5mm * 126mm the body is assembled up to the fitting of the end with steps.

 

The end with steps has two lower steps which due to the limitations of the 3D printing process interfere with the rebate of the sides:

 

 

There are two ways of dealing with this. Carefully thin the backing plate of the step, or as I have done here, file a small recess in the sides:

 

With that done the end should now fit in place:

 

 

With solebars attached and roof temporarily put in place we have this:

 

 

 

 

 

[MikeTrice]

I have a confession to make. The full third was nearly assigned to the bin.

 

I decided to try some Gorilla Glue Superglue Gel on the assumption is would be gel like in form. Having applied it to the joint between sides and floor it all leeched through resulting in the following mess:

 

Careful scraping and sanding and I managed to get it back somewhat:

 

And with a final primer coat:

 

It could be a lot better and I really should file off the beading and replace with styrene however if I only photograph the better side I might get away with it. Damn I have just given the game away.

 

[Western Star]

I decided to try some Gorilla Glue Superglue Gel on the assumption is would be gel like in form. Having applied it to the joint between sides and floor it all leeched through resulting in the following mess:

 

The packaging for the Gorilla Glue here - not the gel product - carries a warning that the product expands during the curing process.

[MikeTrice]

Apologies for not being able to report any progress. Unfortunately events outside my control have stepped in to clobber what little modelling time I get for the moment, so there will now be a short intermission. Normal service will be resumed as soon as possible (and time permits).

[gr.king]

It happens.....