The whole point of the mold positive is to create the molds and associated cores which will be used to actually cast the part. Without further ado, here they are!



Making the mold positive

Before we can make the mold, we need to create the mold positive. This is basically a case half, except we make it out of plastic. It’s not just any case half though – it’s a reproduction of our raw casting, which means that it is dimensionally different than an actual case, and it lacks some of the holes and features that a finished case would have, because those will be machined on the case post-mold.

The process of machining the 1:1 scale case half in plastic takes a CNC machine about 30 hours to do – for one case half. Here are some progress pictures……starting to look like a case!

You scanned a case. So what?

Although scanning the case was a monumental and expensive effort, in and of itself, a Solidworks drawing of a case is pretty much useless for building the tooling needed to actually build a case.

But it’s a start.

This is where the toolmaker comes in. In order to cast new cases, you need to build tooling. Mainly, this is where the bulk of the expense lies in this little project.

Also, and super importantly, the case doesn’t pop out of the mold looking like one that’s in your car. Here are the differences:

  • The raw casting needs to have machining meat on the mating halves, the mains, and just about every surface that will be machined in any way, shape or form. That means that our casting will be thicker in many places than the actual finished case, since it will be machined down later.
  • We need to add draft to our drawing so that the case can be pulled from the mold properly.
  • Our case is sand cast; the original Porsche® case was die cast. Die casting is a little different than sand casting, so our mold is markedly different from a die cast mold.

In short, the Solidworks drawing needs to be worked over by a team of tooling and pattern makers into something that can be used to make a mold. Which is very, very expensive!


Which one, and how?

When we decided to reproduce a Porsche® air cooled flat 6 engine case, the dilemma was which one exactly we should invest in making. It was a hard decision; there are many great cases. We were initially partial to an aluminum version of a 7R mag case, but that has its limitations. Some suggested an RSR case, with its associated head stud spacing.

In the end, we went with a 3.6 964® style case, for several reasons.

  • It was latest air cooled design present, and was carried over virtually untouched to the 993® and was even the basis of the 996® twin turbo and GT2® cases.
  • It has the largest spigot size available from the factory
  • it seems to be the rarest and most desirable case in the aftermarket, and still remains difficult to find used.

We had a 3.6 case on hand, so we decided to use it as a basis for reproduction. Cases, however, are fairly complicated parts to reproduce; it was critical that the case be able to accept standard internals, and thus the dimensions of the case were critical.

The only real way to capture the case dimensions was to have each case half scanned, and so that’s what we did. We scanned each half to 50 microns, and then created a mesh model from that scan data which was then extrapolate into an IGES / STEP file by an engineer. From there, it was converted into a Solidworks .sldprt file, with a full feature tree.

A full feature tree allows us to use Solidworks to change any part of the case design on the fly, while having the software adjust the remaining parameters. Want bigger spigots? No problem, it’s a mouse click away.

To say that this process was insanely expensive and laborious is an understatement. It took a team of people the better part of  to generate a workable .sldprt file from a scanned case half….