How do I make small, inexpensive cast plastic fittings? Pewter?

How easy it is depends on what shape you want to cast. If the object is flat on one side (e.g., water-tight doors), or can be made as a set of components that are flat and assembled later, it's fairly easy.

The major expense is the cost of a can of RTV (Room Temperature Vulcanizing) rubber...get the smallest can you can find. Your local hobby shop may have it, or can order it for you. Price this out first -- then decide if you want to continue!

There are a number of ways to do this; I'll describe one that I've used. I needed to make up a set of 8 steam cargo winches for a WWI vintage freighter. Each was essentially a rectangular frame with various "bits" sticking out here and there.

I decided to break the pattern into 10 parts: base, 2 sides, 2 "cranks", 2 cams, 1 spool, 1 "head" and a gear; these would be assembled using brass rod.

The base was a 1/2 x 5/8 rectangle of .015 styrene, this had two 3/16 x 3/16 x 1/16 blocks of styrene for steam boxes glued on, and some of the steam piping.
The two sides (mirror images of one another) were .015 styrene, properly shaped. A round boss was added as well as a flange around the top and side, and a steam cylinder (a turned spool bedded into a carrier, both from Lucite). Also, more steam piping.
The cranks and cams were 1/32 slices off a 3/16 Lucite rod; the cranks had a couple of notches filed into the edge to become "T" shaped.
The spool was turned from 1/4 Lucite; it had a slice of a 1/4" 10-tooth plastic pinion gear salvaged from a slot car.
The head was turned from wood, 3/16 d x 3/16 long.
The gear was a slice of another, 8-tooth pinion.

None of these pieces had severe undercuts!

All necessary holes were located, and either drilled through (on the thin stuff) or at least "dimpled" to locate for later drilling.

I next took a piece of scrap plate glass (about 3" x 4" -- I had a lot of that left over after the last earthquake), and CA'ed the back of all the parts to it in a reasonably close pattern.

I made a "fence" from cardboard -- just a box without top or bottom; approximately 1 1/4 x 2 3/4 x 1/2 high. This was temporarily taped to the glass surrounding the patterns.

I smeared the inside of the box, the glass, and the patterns with vaseline thinned with a bit of lighter fluid as a release agent.

Mix up a small amount of the RTV; try not to get any bubbles in it. Vibration helps -- try one of the Dremel engravers, or a jig saw -- anything that will vibrate and shake out the bubbles.

Paint a thin coat of the RTV over the patterns...the thinner the better to avoid bubbles. Do it again. After a couple of coats, pour the rest of the RTV into the box level with the top, and let it set up.

Pry the now-solid rubber block off...that's your mould. The "top" of the mould should be nice and flat (from the glass), with the "prints" from the patterns as cavities in the surface.

I used 5-minute Epoxy for the castings.

Again paint the rubber with the vaseline mould release, mix up a small batch of epoxy (post'it note pads make a wonderful disposable palette for this) and fill all the cavities. Since it is clear, you can see bubbles and poke them with a toothpick or pin. Use an old razor blade to level the top of the epoxy (that will be the back of the individual castings). Once the epoxy on the palette is hard, you can remove the castings from the mould. Being rubber, you can flex it to help get the parts out.

Warning...the epoxy does tend to stick slightly to the mould. if there is a bubble in the rubber adjacent to this, the rubber will tear there and leave the bubble exposed. The next casting you make from the mould will have tiny "beads" at this sites, as your casting cavity includes the bubbles which will fill with epoxy. A few of these can be sliced off the parts, but eventually the mould will become unusable.

How long that takes depends on how much care you took to avoid having bubbles in the rubber.
{John O. Kopf}

There are two main mold materials that have very different applications. RTV mold material is great, you can make undercut molds. But, it is expensive and has limited shelf life. Other is plaster of paris. Cheap, but parts must have relief angle, absolutely no undercuts or straight sides. Limited shelf life also, but at price this is not a problem.

I tried to do a zero relief cast of a hubcap that had a short, straight cylinder. Got the mold off the part okay. Then, put lots of release agent in mold. Still, cannot get resin part out of mold. Resin castings do not shrink!

For metal parts, you may be able to get away with straight side sections in simple parts, as metal does shrink a bit.

But basically, we have the dilemma of an excellent material (RTV) that is expensive, or a cheap material that limits the type of parts you can cast.
{Don Stauffer}

Other methods of mold-making and casting include the quick and expedient method of modeling clay (oil clay) and superglue. If you need a duplicate of an existing part that isn't too large or complicated, it can be made of cyanoacrylate(sp?). Get some modeling clay (the grayish sometimes oily stuff that never hardens), work it up until its fairly pliable, then simply press the part to be duplicated into it to make a simple negative impression. You might need to try it a few times to get a clean and clear impression. Then, drip in a bit of CA glue over the whole surface and then a bit of filler (microbeads, or for both filler and quick setting, baking soda, which flash cures the CA). A few more layers of CA and filler will build up the object to the desired thickness. However, it works best if you don't rush the process with very thick layers, as they will be very slow to cure, even with soda or accelerator. Then, pull the clay away from the object, and with a bit of paint thinner clean off any clay that is still sticking to the object and there you go.
{Bev Clark/Steve Gallacci}

The brand I use is from a German company, but I am sure you get the equivalent anywhere else, too.

It comes in two components. One is the Rubber, the other is a small flask with the vulcanizing agent (basically a "starter" for the chemical process).

To make 2 part mould, I do this:

I use a glass plate as base (smooth a plane). On this, I build a box from Lego-blocks in the size I want. Then I place a dab of play-doo (or glazing-putty - what ever you call that stuff, I think it is chalk powder mixed with linseed-oil). In this I bed the part I want to duplicate to the separating line.

For pieces that were cast, I use the still visible lines from the original cast. Now there should be about half of the part showing, the other half being embedded in the clay/dough/putty-stuff. Around it are the Lego walls. The thickness of the RTV should be about one inch, so make the Lego box big enough, but not too big to waste RTV.

Then I brush on a separating agent on everything (glass, Lego, the part, not the putty).

Now I mix the RTV with the "starter". You need to work fast here, because it will start to cure after only a few seconds (maybe 10 or so). Then pour the stuff into the Lego box. Make sure you don't get any air bubbles clinging to any surface. I use a toothpick to prod any dents and holes. Let everything sit for at least 2 hours. This is only a suggestion - better to leave everything untouched over night. Use the residue of the RTV in the mixing container for reference. When you are sure that the RTV has cured, tare the box-box apart and pull the mould from the glass plate. Take off the putty, but leave the part inside the mould.

Turn the mould over, reassemble the Lego box around it, but twice as high, brush everything with separating agent, and redo the process for the 2nd half of the mould. Remember the 1 inch thickness rule.

After curing you have a 2 part mould. Wash off any separating agent. Cut channels for the pewter to pour (be generous, make a real nice funnel, but remember the 1 inch rule!), include pipes for air to escape. Use the frames from a plastic model for reference.

You can test the mould by filling it with water and then put it in the freezer over night. The Ice-part you get shows you where air was trapped – you need to cut another pipe or air channel there.

To make a real test casting, dust the insides of the moulds with talcum powder, assemble the mould, place it between 2 pieces of plywood and clamp it (maybe in a vise) hard enough to hold, but without warping the rubber. Then pour pewter. Pour slowly to give the air time to escape, but do it in a steady flow. The pewter needs about 5 minutes to solidify.

Between castings, let the mould cool off to room temperature. Don't forget the talcum!

The cast should be good for actually hundreds of castings. If something breaks off, use CA glue to repair it.

Once you have done this a few times, you can easily make multi-part moulds.

I hope this helps
{Thomas Boegel}

There are a variety of resins to choose from, I don't think there is a lot of {practical difference between them. The one I have sitting on my shelf is {called "Alumilite". I can't remember if it is polyester resin or epoxy resin, {but it's one or the other. "The other one" is used in fiberglass and is rather {sticky and not really suited for our purposes!

The stuff comes in two separate bottles, rather like RTV in fact. The Alumilite is mixed {fifty-fifty (the mix isn't all THAT critical however) from one dark brown liquid {and one clear liquid, the brown one is liquid like water while the clear one {is a bit thicker, like shampoo or something. When mixed the stuff is quite {liquid but takes only a minute or two to start hardening, so speed is of the {essence :) In five minutes or so it has set rock-solid and can be removed {from the mould. RTV moulds combined with resin is capable of producing {incredibly detailed cast parts - but if you've built models made from it you {know that already!

The cured Alumilite is quite strong, but has a certain flexibility; it is not brittle. {It is yellow in colour, other resins I have seen have a more greenish tint. {The reaction between the two components produces a certain amount of heat, {be a little bit careful when removing large pieces cast from it... {Final note: Largeish parts cure well on their own, but if only a very small {amount of resin is used it may actually need a visit into the stove at 50-75 {degrees Centigrade to kickstart the process. (I've moulded airplane ejection {seats in 1:72 scale without problems with this, so it should not be much of {a practical problem.)

I have seen resin for sale in a couple of well-stocked hobby shops here in Oslo, {so it shouldn't be impossible to get in Germany either. It is somewhat expensive, {but a two-bottle set will go a long way; however the shelf life is limited to a {couple of years. Try to find a hobby shop that is run by a serious plastic modeller, {preferably of the air or armour persuation; those guys are practically married {to the stuff and should be able to point you in the right direction. Or you might {try to contact the local IPMS club!
{Staale Sannerud}

RTV can be just one part moulds for long slim items that have little or no undercuts. When I took a mould for the dragon on the rudder of my Green Dragon, I used a six part mould which cost a fortune in RTV.

I used Lego bricks for a dam and plenty of plasticine to mask the areas I didn't yet want to flood with RTV. Gradually more sections of Plasticine were removed and RTV flooded their voids until the whole dragon was covered. RTV doesn't easily stick to itself, however a spray over with wax spray first guarantees the mould will come apart easily.

I once used RTV to make a mould of a Thames barge's hatch covers which were covered in linen to simulate canvas tarpaulins. I was somewhat skeptical that it would work and would not just become a rotten mess and ruin the hatch cover of three weeks in the making. To my complete and utter astonishment it worked and the RTV came away quite easily (I waxed the patterns first) , leaving a beautiful mirror imprint. After glassing, the resulting covers were an exact copy of the wooden originals and once painted it is hard to tell what they were made of..

I the case of the dragon mould I used tinman's solder to run in it and was rather lucky on my first go. The RTV could have equally taken pewter, white metal and several other metals that took about 5 - 600°F to melt. I naturally dusted the RTV moulds with graphite powder first as this aids the flow of the metal a bit and helps to leave a great finish on the cast item.

There are a couple of problems with the rubber though as it degrades with time plus it disintegrates after about 90 or so pourings. Vulcanising rubber is far better for manufacturing batches.
{Mike Taylor}

I have to disagree with you in this, Mike, I really think that you have been using the 'wrong' RTV in that case :) They come in a very wide variety, some are very stiff, some very soft, some can take a lot of heat and can be used for moulding molten metals, some can not. Take a look at the resin parts that are manufactured for the aftermarket tank/plane modelling market; they have undercuts left, right and centre.

It also depends on the material you are using for the part itself, of course. Epoxy resin (or is it polyester resin???), used for the aftermarket parts, is very rigid and strong when cured and can take the kind of punishment it gets when extracted from a one-piece mould. Pewter could be a bit on the soft side I think.

The resin used is also a two-component self-hardening stuff, by the way, rather like the RTV itself. This means that the RTV used in the moulds doesn't have to withstand high temperatures; there may be a connection between this property and their softness for all I know. In other words, if the RTV is designed to withstand high temperatures it may also have to be so hard that one-piece moulds are impossible.
{Staale Sannerud}

Staale my friend..
My many thanks for disagreeing with me as though your note I have discovered that there are more types of RTV than the one I have been using for over 20 years now. My supplier only stocks the one that is heat resistant (up to a point) and labels it as RTV rubber and it is this type that I know about to quite an extent. It will take virtually any type of material without detriment. I say 'virtually' as I haven't tried every medium yet, but keep thinking them up nevertheless.

Once cured, the rubber is quite hard and I don't like bending it about too much just in case it cracks. I haven't had a mould crack on me yet but this stuff is not cheap and there is always a first time.

When I said that my type of RTV doesn't like undercuts all that much, I should have added that continued removal of cast items tends to tear the mould fractionally. I appreciate that resin based models do have quite large undercuts and have indeed moulded military vehicle parts for a friend. I have also used the same moulds to make metal parts too with none or little detriment to the rubber. I haven't done any volume runs with the stuff so cannot say how long the mould will last and have only the run life as advised by the chap who sells it to me. He says 90 or so reproductions per mould and has always been fair to me so I have little evidence not to believe him.

Pewter is quite soft when it emerges from a mould but this metal age hardens and work softens unlike most of the rest of metals. I like using pewter because it flows really nicely and one can get a very thin and intact casting from it without flow run-outs or air pockets impeding the flow.
{Mike Taylor}

At least with slow-setting stuff like resin, you can simply put the filled mold in {a bucket tied to a string, and whirl the whole shebang around over your head {for a couple of minutes. Quick and dirty, but it works :) I guess pewter sets too {quickly for this to work, though!
{Staale Sannerud}

I am trying my hand for the first time at resin casting (fifteen 1/350 Avengers for an escort carrier). I am still in the prototype stage. Essentially, what I am doing is casting a five piece model kit for each plane. Since I am a beginner at casting, I am using a 1 piece, open face mold. I am using the Pour-a-Mold and Pour-a-Cast product from Blick.

Measuring is critical. I bought a pack of 100 graduated medicine cups from my pharmacist to use in mixing. The instructions note that a 3% error by volume can result in a failure to cure. So far, I have only had one failure with the mold material and none with the casting material. I have used the Leggo method of mold construction. I found some sulfur-free clay to use to hold the Leggo box in place. I read that the sulfur content of some clays can mess up the curing process.

The cast material is fun, but difficult to work with. You only have about a minute working life and it changes from clear to snow white in just a few seconds. I can remove the cast within five minutes of pouring. When removed, the cast pieces are very flexible, and even stretchable at this stage. After removing the pieces I arrange them on wax paper and make sure that I remove all undesirable bends and warps. In about twenty minutes I a have a nice little model kit with sharply defined wings, fuselage halves, and tail surfaces. The casting material works very well with CA glue.
{David L. Miller}


" SiliconeKit" (as available in Belgium)

I simply use a tube of SILICONEKIT from a DIY. It works great on small parts if you have the patience to let it cure a few days (1/4 inch). (It is the silicone sealer that is used in bathrooms, kitchens. It leaves an acetic smell when curing. It is readily available in DIY outlets here in Europe. I use the cheapest, non-branded and it works very well, as long as you do not forget the mold release (mineral oil work fine), as it sticks to anything. In fact it is also used to glue glass panels.)

I fix the master to a glass plate with modelling clay, build a wall around it, apply some fine sewing machine oil on part, walls and glass, squeeze in a small quantity of transparent SILICONEKIT, coat the part with a layer, eliminate air-bubbles and then complete the mold.

The kit cures to a flexible material that is nearly as resistant as RTV at a fraction of the cost. I use it a lot to copy scratchbuild wheels for trucks and cars.
{Johan Dekimpe}

Back?