Casting Iron
I ran into a not so unique situation when I researched casting iron parts for my car. All the "experts" I contacted who run foundries said that it would cost around $4500 to make a template and mold and another $500 to pour the iron for the three exhaust manifold to block spacers that I need.
They made it sound so damn complicated and difficult that it was quite discouraging. Something sounded wrong to me. There has been sand casting going on for over 2000 years so it can't be that complicated and difficult. Everything that I read about it indicated it was a pretty simple process. Make a wooden template, pack it in a split sand box, add some pour holes and vent holes, pour in melted iron, cool, machine the unwanted parts off.
I researched sand casting on the Internet and came up with some great links from hobbyists who had there own mini foundries. They designed and built their own furnances for typically under $200. All I had to do is find someone who has done that in my area. I figured out that model steam engine builders usually cast a lot of their own parts so they must have a foundry or access to one.
I eventually contacted Rupert Wenig who was gracious enough to offer his help even though he lives 180 miles from my house. He is a member of the Alberta Metal Enthuasists Network which is a great bunch of guys who go out of their way to help you.
I was given very good advice on how to make the wooden template and invited to a "pour" of aluminum and zinc-aluminum about a week later. I figured that I would try to quickly whip up a template out of a 2x6 and see what type of problems I would encounter.
Here is a pdf file of a side view of what the exhaust spacers will look like. They are basically the shape of the exhaust outlet on the heads and 1 7/8" thick. The pdf file shows the parting line for the casting running through the mounting bolt holes.
Some quick rules:
- cast iron will shrink 1/16" per foot from the template size. This means that the template
has to be 5% larger on exterior dimensions and 5% smaller on interior dimensions versus the final product.
The term for this is shrinkage.
- Any area that will be machined after casting should have an extra 0.125" added because the iron will react
with the silica to form a very hard layer and you'll want to machine past it.
- In order to remove the template from the sand without breaking the sand mold, vertical lines are not allowed.
A small angle anywhere from 1-3 degree has to be made into the template.
- No sharp corners are allowed on the template.
The first thing I discovered when making a template was that table saws are very inaccurate instruments for cutting when dealing with hundreds of an inch. I was attempting accuracy of 1.125". I asked the local carpenters at work to thickness plane my 2x6 to 1.125". They said no way, the closest they could get it would be 1 1/8". The result was 1.100" which is too small for my purposes.
I got out my Dremel tool and went to work on one template. I tried steel routing style bits but found that the small drum sanders worked the best. Too good, it was very easy to take too much material off. I figured that I would have more control if I put one of the small drum sanders in my drill press. This worked out very well but definitely indicated the need for better control of movement like found on a mill.
I had a split pattern (2 piece) that was raw wood (should be painted and a layer of hard wax) that I brought with me to Rupert's farm to test.
2 piece wooden template
The templates have small rectangular extensions that run out both sides for holding the sand core plug. The sand core plug is used to make the exhaust runner opening.
Just a note on the template's warpage: right after the 2x6 was planed, it was perfectly flat. Within 3 hours, it had warped slightly concave. Two days, after carving the template, the wood warped the opposite direction and now was slightly convex. A drier piece of wood would of helped.
The Pour at Rupert's Farm!
I drove 3 hours to get there and another 3 to get back, for 4 hours of the most interesting and exciting discussions, demonstrations, advice and castings I have ever seen. What a blast! And like a fool, I forgot to bring my camera to record the event :-( When I go to pour the cast iron pieces, I will definitely pack it though!
Rupert is an excellent host, a natural born instructor who basically gave us a step by step description of everything that he was doing, why he was doing it and the problems that can arise. I was able to have my 2 piece template casted in aluminum and was pleasantly surprised at the quality and finish of the cast part.
First try at casting and it worked better than expected!
The above picture shows the casting just after it came out of the sand and the sprue (fill hole) and risers (hole to let aluminum flow out of the cast) knocked off with a hammer.. It requires machining of the two faces and drilling of the mounting holes. The picture doesn't show the hole made by the sand core. The extension on the top for the sand core plug is hollow and very thin, the missing one on the bottom was broken off. The presence of the extensions are the result of jury rigging a round core into a square hole just to try the template out.
Sand casting box (notice alignment arrows)
I made a sand casting box out of 3/4" plywood that I had around. It is basically two rectangular frames that attach to both sides of a small flat board (painted red in the picture). On either side of the flat board is the halfs of the template. The template halfs are aligned perfectly to each other.
The idea is that you fill one side with sand, then fill the other side, remove the flat board with the template and you have a perfect mould ready for pouring the iron into. There are 1/2" grooves cut into the inside walls of the rectangular frames to help stop the sand from falling out. When the sand is rammed in, it is very stable and you can turn the frames upside down without dumping the sand.
Core mould
To make the center exhaust runners, a sand core plug is made that has the required dimensions. The core mould is basically two pieces of wood clamped together. Down the middle is a hole milled or routed in each piece the shape and dimension of the exhaust runner.
The template will leave indents in the sand for the core plug to sit on. The core mould is used to make the sand core plug. Sand is packed into the core mould and then a 3 1/2" nail is pushed in from both ends for added support. The core mould is seperated and the sand core plug removed.
Center exhaust template
It took 2 weeks to make one half of the center exhaust runner wooden template on a milling machine. It looked excellent! I figured that rather than spend another 2 weeks, I'll just make a mould and then make a template out of fiberglass or some other material.
I had a bucket of polyfilla so I thought that I would put the template into an ice cream bucket and fill it with polyfilla. I coated the template with a paste wax and poured polyfilla to a depth of 2". First problem is that polyfilla isn't made for 2" depth. It bubbles, cracks and takes 3-4 days to dry. It has very little strength and broke into pieces when I tried to remove the template from the "mould".
The second problem was much more serious! I didn't seal the wooden template (with paint or sealer) and it absorbed the moisture out of the pollyfilla and swelled. Since the polyfilla was only on one side, one side swelled. The template warped!. Aggghh!
I dried out the template and glued and screwed it flat against an old 2" thick cutting board. Of course the template cracked. So out came the polyfilla to fill the cracks. I painted the template with some red spray paint I had around.
Next I decided to follow my original plan of using fiberglass as a mould material. After the paint dried I put on 3 thick coats of carnauba wax and started putting on the fiberglass and matting while the 3rd coat was still wet. Carnauba car wax is a release agent for fiberglass.
After picture - center exhaust runner fiberglass mould on left, wooden template on right.
Before picture - pouring fiberglass and matte on template to make fiberglass mould
The fiberglass mould needed work as it was rough in some places and there were air pockets. Out came the polyfilla and sand paper to smooth out the mould. After which I painted the mould red and put on 3 coats of carnauba wax.
Carnauba car wax - release agent for fiberglass among other things...
I mixed up another batch of fiberglass and used cabolic filler for strength. I poured it into the mold. About 4 hours later the fiberglass had hardened and the new fiberglass template was ready to be removed. The first thing I noticed was that the fiberglass had shrunk :-( It looks like it has about 1-2% shrinkage.
It popped out of the mould (not really, gently prying with a chisel) and is close enough in dimensions that I can use. Fortunately, I only need the general outline, as the critical sides are going to be machined off.
I mounted it and the wooden mould on each side of a flat 3/4" piece of plywood that was cut to fit the casting box shown earlier. Some minor touchup, paint and more carnauba wax and it'll be ready for pouring.
As this page has gotten a little long, I've made another page about the pour for the cast iron exhaust spacers and some aluminum side mirror brackets.
If you have ever considered casting your own parts whether in aluminum, brass, bronze or iron, check out one of the hobbyist foundries in your area, you will be pleasantly surprised at the casting process.
Here's some great links on casting iron:
- Eagle Ridge Forge's Tom Butrum
- Australia's Brian Carter's homepage