Sunday, December 27, 2009

How I made my third batch of open wagons

Learning from the experiences of making the previous two batches of open wagons (see How I made an open wagon 1 and How I made open wagons 2) and from making the cattle wagons (see How I made two cattle wagons), I decided to go into mass production, using a mould and resin castings.

The prototype
During a visit to Llanfair Caereinion Steam Gala I took several picture of a couple of open wagons from various angles.
My models would be 'inspired. by these prototypes rather than accurate scale models.The chassis
As previously, Hartland Loco Works wagon chassis formed the basis for each wagon. These are very cheap (around £10) and very easy to construct - they just click together. The only modification I made was to remove the bolt detail from each end of the deck with a sharp craft knife:
The underframes are not strictly accurate for British rolling stock and one day, I will produce my own, but for now these are quite acceptable.

The dimensions for the wagon sides were based on the size of the Hartland wagon base (ie the red part in the picture above). The length of the sides and ends were increased by a couple of millimetres to allow for shrinkage.

The masters
Using the experience gained from making the previous two batches of wagons, I constructed one side and one end from plasticard.

A 147mm x 33mm piece of 2mm plasticard was cut for the side. This was scribed at 7mm intervals to represent the planking.
Three 27mm x 4mm pieces of 1mm plasticard were cut and glued in place to represent the strapping for the hinges.Two 25mm x 4mm pieces of 1mm plasticard were fixed in place midway between the hinge strapping for the intermediate strapping.Two 6mm x 7mm rectangles were cut from 1mm plasticard for the latch-plates. These were fixed to each top corner. A small strip of 1mm plasticard was fixed to the ends and the edge rounded with a file to give the impression that the latch plates were bent around the ends of the sides.Hinge details were made from small pieces of 1mm plasticard.Finally, nuts and bolt-heads from Cambrian Plastics were glued into place using the photos of the prototype as a guide (see How I made the cattle wagons to see my technique for fixing rivet and bolt detail)

The Ends
A 93mm x 47mm piece of 2mm plasticard was cut out. A 27mm x 8mm recess was cut out midway along the base to fit over the coupling bracket on the chassis. A 33mm x 3mm strip was removed from each side to accommodate the wagon sides when the wagon is constructed.

Planks were scribed at 7 mm centres.
Two 40mm x 6mm straps were cut from 1mm plasticard and attached either side of the slot for the coupling plate. Two 40mm x 3mm pieces of 1mm plasticard were then glued vertically down the centre of the two straps to form T brackets.
Two 40mm x 3mm straps of 1 mm plasticard were fixed 2.5mm from each side.

Cambrian Plastics nut and bolt heads were added, together with coach-bolt heads, using the prototype photos as a guide.
Choosing the right materials
 Before blundering into the world of moulds and casting, for a modest fee, I downloaded the Garden Railways introductory guide to resin casting. At $5.95 I feel it is well worth it for, if nothing else, it gave me sufficient background knowledge and some confidence to at least make a start.
Following suggestions from the G Scale Central forum, I contacted Tomps (http://www.tomps.com/) asking their advice as to which of their baffling array of products would suit my needs. Their medium-hard RTV silicone plus catalyst from their value range was purchased for making the mould:

Their polyeurethane fast cast resin and hardener was also purchased following their guidance. I went for the smallest quantity available in case it all turned out to be a disaster. I would also suggest buying the slower (extended potlife) version as it gives more time to 'paint' the mould (see below).

I also bought a bag of grey Fillite filler to bulk out the resin, though I have heard that talcum powder can be used as an alternative.

Making the mould
The masters were fixed firmly to a piece of polycarbonate glazing sheet with clear Bostik adhesive. I found to my cost that it is very important to make sure the adhesive is spread evenly otherwise the latex will seep in under the masters producing a ragged mould.Using clear polycarbonate enabled me to spot any gaps underneath the masters. A margin of around 1cm was left between the two masters and the sides of the mould-box.1.5cm wide strips of of 2mm plasticard were fixed around the masters to form a mould box. These were held in place with hot glue. Again, care was taken to ensure that there were no gaps.A 20:1 mix of RTV silicone and hardener was measured out using a set of electronic scales as the two components have to be measured by weight. The advantage of electronic scales is their precision and also being able to zero the reading once the mixing vessel has been placed on the scale.
The mixture was stirred with a chopstick in an old tin can. I realise now that I needed to spend a lot more time making sure the hardener was fully mixed. This can be judged by the uniformity of colour as the blue hardener mixes with the white latex. Unknown to me there was some unmixed latex at the bottom of the can but fortunately there was sufficient mixed latex at the top of the can to fill the mould. I have learned a lesson about latex mixing for future projects! A good suggestion I found on YouTube is to mix in one tin and then transfer to another tin for final mixing - this helps ensure any unmixed latex at the bottom of the first tin becomes top of the second tin.After giving the masters and mould box a light dusting of talcum powder, the latex mould mixture was poured very gently and carefully into one corner of the mould-box, taking about ten minutes to help prevent air bubbles forming. The mould-box was tapped vigorously for several minutes after filling to encourage any bubbles to rise to the surface. The mould was left overnight to harden and then carefully removed from the mould-box.

 Casting
The two components of the polyurethane resin were mixed in equal proportion using the electronic scales.I used some cheap plastic containers for measuring and mixing keeping one exclusively for measuring component A and the other measuring B. This latter was the container into which the filler and part A were poured for mixing. I found that around 15 grams of each component together with a couple of teaspoons full of filler was sufficient for casting each end and side with some resin left over for general gap filling (see below).

Again, experience has shown that the more slowly and gently the resin is poured into the mould, the fewer air bubbles are formed. Inevitably, airlocks formed in the bolt and and rivet detail, leaving dimples rather than neatly moulded bolt heads.

I've tried several ways of dislodging the bubbles, sprinkling talc into the mould before casting (leads to a build-up of talc in the recesses), tapping the mould (seemed to make no difference), vibrating the mould using an upturned orbital sander (separated out the two components of the resin), omitting the filler. If I could afford a vacuum chamber I would have used one, but it turned out that the most effective approach was to let the resin flow gently into the mould.

PS - Following some extremely helpful advice from a fellow modeller on the G Scale Central discussion forum, I've found out how to overcome the airlocks in the rivet and bolt detailing. Before pouring the resin into the mould, I now 'paint' the inside of the mould with resin, using a small paintbrush. This whetting process helps ensure that all the recesses are filled before the rest of the resin is poured in. It has to be done quickly, however, as the pot life for this type of resin is around 3-4 minutes. I'd suggest buying 'extended pot-life polyurethane resin' which has a pot life of around 6 minutes.

After only 20 minutes, the resin was set sufficiently for the pieces to be removed from the mould and the process could be repeated for the other side and end.

Constructing the wagon
To fill the holes in the deck of the Hartland wagon, parcel tape was stuck to the upper surface.

The deck was then upturned and some of the excess resin not needed for casting the sides and ends was poured into the holes. Once the resin had set, a sharp knife was used to remove any spillage.

A flat needle file was used to remove flash and tidy up the edges of the castings. Then gaps between the planks were scribed on the back of the castings with a scriber:

The ends were fixed in place first with epoxy resin. I find the postman's red elastic bands just right for holding the ends in place while the glue sets. Coffee stirrers were used as temporary spacers to keep the ends from folding in on themselves.

After about half an hour the sides were epoxied in place - with a couple of elastic bands to hold them in place.
Four more wagons later, the whole batch were spray painted with grey Plasticote primer, and left for a couple of days for the primer to set.

I then picked out the strapping with black acrylic paint:

and fine chain was superglued to the latch plate and to the topmost bolt head on the end strapping.
The underframes were then slotted into place and the couplings screwed on. Metal wheels were added as I find these give the wagons much-needed additional weight.


The wagons were given coal loads. A 140mm x 85mm rectangle of 3mm ply was cut and reinforced longitudinally with two lengths of 8mm square stripwood. Two pieces of stripwood were glued near the centre.

This is to make the removal of the load easier when running back down the line as empties. By pressing down on one end, the load flips up and can be removed.

Real coal was crushed to more appropriate sized pieces with a hammer and glued to the wooden platform with PVA.

Finally, the wagons were weathered. The underframes were given a wash of rust reddy-brown - which was then wiped off with a paper towel to leave 'rust' in the recesses. A wash of black acrylic was applied to the body and the excess removed with a paper towel. Dark and light brown acrylics were then lightly sprayed on from spray cans.

In time, I plan to cast some cosmetic W irons to Anglicise the Hartland axle boxes and hangers.

4 comments:

Brian said...

Excellent work!

-Brian

GE Rik said...

Thanks Brian. There is still a bit of work to be done on them before I'm entirely happy, but they do look OK rattling around the garden.

Unknown said...

Perfect as always! thank you very much for you detail comments and pictures.
Did you not think about to use same LESS hardening stuff, it will prolong the time to harden.
And to get those pesky air baubles out... a hairdryer or an ventilator?
If it is just aggravating the surface, you are home free, just like concrete, atgetate it.
Some how the trembling will make the material thinner and cause the air to rise up.

With best regards and thanks for sharing: Best Igor.

GE Rik said...

Thanks Igor
I can't remember clearly, but I think reducing the amount of hardener didn't make any difference to the pot life of the resin. I do remember one batch not setting properly but that might be because I didn't mix it thoroughly. Painting the mould first with resin was effective at reducing air bubbles. I never did try using resin with extended pot life. I think that would be a good solution.

All the best. Keep modelling....... 👍

Rik