Adapter Plate

There are a couple adapter plates available for mating k-series to rear wheel drive transmissions but I wasn’t too keen on the idea of swapping out the stock unit. The RX8 uses a power plant frame to mount the driveline so swapping in a transmission from another car would have meant I had to fabricate transmission mounts and potentially a custom driveshaft. In my mind both creating a custom adapter plate and using a non-RX8 transmission were equally challenging but in the end I figured the more stock parts the better since, lets be honest, I had no idea what I was doing at this point.

I used cardboard and MDF to mock up a prototype. It took a fair amount of trial and error but I was surprised how quickly I was able to put together something that allowed me to bolt the engine to the transmission. Thankfully the RX8 transmission is fairly light (95-ish lbs) but it has to be one of the most ungainly and unbalanced objects I’ve ever worked with. On any sort of flat surface it is constantly on the brink of falling over and was the cause of much frustration considering I pulled the transmission on and off the engine what seemed like 100 times.

Once I had an initial prototype I had to make the decision how I was actually going to make the final product. I had solicited a few quotes to have an adapter plate made based on my prototype but they were fairly expensive and I had so little faith in my design it seemed like a recipe for disaster. I needed a way to iterate on my design and have full confidence that when I handed it over to a machinist it would result in the part I needed. Behold — the 3D printer. I had been interested in 3D printing ever since the technology became more mainstream but I just couldn’t justify the cost when I didn’t have a specific purpose for it. Since I was actually trying to solve a problem and the alternatives were just as, if not more expensive, it seemed like a good excuse to pick one up.

3D printers that are large enough to print adapter plates in a single piece are still incredibly expensive. Slicing parts into multiple pieces, printing them separately and manually fastening them together worked surprisingly well. I completely lucked out and found the coordinates to the RX8 transmission bolt pattern online. With a quick download of FreeCAD I was able to plug in the coordinates and the external dimensions of my prototype to get the first glimpse at what I was up against.

 

As it turns out I figured out that what I was up against was a pretty tough challenge. Leaning CAD software on top of 3D printing on top of adapter plate design was a lot to bite off all at once. Part of the problem was that I was starting from nothing so a quick Craigslist search found me a willing and able engineering student who would take my prototype and build me a 3D model. I knew it wasn’t going to be the final design when I got it back, but it gave me a starting point where I could print and tweak until I arrived at my final design.

Once I had a model I really started to understand the CAD software and 3D printing better because I had a physical artifact (albeit digital) to work with. Rather than starting from a blank canvas I could make small adjustments like moving a bolt hole or changing the thickness. By the end of it I understood enough about the software and the design process that I could definitely make something from scratch if I needed to.

I kept tweaking, slicing, printing and assembling until I finally had the bolt patterns perfect and in the correct orientation. It’s a painstaking process but I must say, it’s pretty cool how precise you can be with a 3D printer. With the bolt pattern finalized it was time to start dialing in the thickness of the adapter plate.

I knew the adapter plate needed to be fairly thick in order to allow for the stock drive shaft to reach the tail of the transmission and for the engine to clear the firewall. In order to accommodate for this I had to fake a custom flywheel. I had access to a k-series flywheel but any thickness over a half inch meant the input shaft of the transmission couldn’t reach the pilot bearing and therefore couldn’t be properly aligned. By taking a few measurements and doing some quick math and I was able to design a spacer which allowed me to push the flywheel out enough so it could still mate with the transmission. This is really the beauty of 3D printing as I was able to take an idea from concept to implementation in about an hour.

I started with a fairly thick adapter so I could test fit it and dial it back as needed. I was amazed by how strong 3D prints are. You can adjust infill, which is the internal grid structure of the object you’re printing, all the way up to 100% (solid object) but even at the standard 30% it was plenty strong. Once printed I could use a tap to thread the bolt holes and fasten the engine to the transmission just as if it were metal. I still had to be careful when maneuvering the engine and transmission around but it was much more durable than I expected.

Once I had completed my design it was time to cut the final product. I was fairly confident it was going to work but it still seemed like a bit of a gamble. The adapter plate ended up being 1 inch thick and cut from 6061 aluminum. As you can see here I opted to use the RX8 starter but fast forward a bit and I actually changed my mind and decided to use the k-series starter. This required me to manually cut a provision for the starter but allowed for one less variable when ordering a custom flywheel.

If I could block out certain portions of my memory containing moments of complete frustration learning CAD and 3D printing it was actually a pretty enjoyable experience. There is something very satisfying about designing and building something completely from scratch. The moments of frustration I experienced were only related to me not knowing what I was doing which pretty much happens whenever you’re leaning something new. In the end I learned some new skills and had something to show for it. I’m not sure if I’ll be using my new found skills to this extent any time in the near future but it’s nice to know I could 🙂