RX8 5.3 Turbo
I am working on the second version of this swap since I spun a bearing and the turbo was undersized, but here is the beginning.
The car is a 2004 Mazda RX8. I picked up the car with a "blown motor" for cheap, but it turned out to just need to be de-carbed and I was able to drive it off the trailer and used it for about 3 weeks.
But now it was time to start the build. The plan was to do a LM7 with a 4l60e and a turbo.
I made mounts similar to the CXRacing kit. I got the motor as low as possible by using a couple magazines for spacers and setting it on the subframe. I then tacked everything in place and removed for extra gussets and welding.
Turbo Piping was made by modifying truck manifolds and with 2-1/2" pipe. I didn't want to wait for bends so I just pie cut everything. I should have just got the bends after how long it took to weld everything. And after more research I should have used 2" for the crossover.
Turbo is a H1C from a Cummins diesel. This turbo is too small for the application, but it was what I had at the time so I tried it out. It made boost quick but dropped pressure at the top end.
Here is the engine installed and a Ebay SBC radiator with custom brackets. Intercooler piping is 2-1/2". I also made a filler and port for the steam tube.
Tuning was done with HPTuners using a 3 bar SD OS. Testing was going well until a tick developed on the motor. I took it out for another run thinking it might be lifter noise, tick got louder and engine seized.
The car sat like this since September until a few weeks ago when I started on version 2.0
Next post will have the latest engine details.
Dave
I installed new bearings and a LS9 cam, I also plugged off the DOD ports on the block with set screws and a little ultra black.
I also got a new front cover and timing sprocket for the camshaft.
GT45 Turbo was purchased off Ebay and was much better suited for this engine.
I wanted to change the turbo location and keep all the hotside pipes hidden so I decided to place it where the alternator was located.
I used the truck manifolds again but after seeing the final manifold I wish I had just used elbows and made one from scratch. If this location works I might rebuild it later.
Crossover was built with 2" mandrel bends and the wastegate is located beside the oil pan.
Downpipe is 3" mandrel bends
Turbo drain was in a tight location so I clocked the turbo 20 degrees and with a 30 degree cut on the pipe this allowed for a 10 degree slope on the drain.
This is what I have done so far, I think I am just about ready to pull the old motor since the mock up is complete. I also plan on re-wiring the engine so the PCM in inside the car and I need to run new fuel lines to allow for a boost referenced regulator.
Dave
I might test something out in sketchup to see how it would look with a flange and pipe weld elbows and how much work it would be to build.
Dave
I like the manifold design, I'd make the #2 cyl just join the log like the rest and make the turn into the turbo constant radius
look at the old turbo buick v6 manifolds very similar design and they work pretty good.
nice project! the turbo Fds are all the rage but the little RX8 is affordable and has a very nice setup!
Trending Topics
The Best V8 Stories One Small Block at Time
Here is the passenger side manifold.
And this is the whole kit.
I have the old motor ready to pull so it should be out in a day or two.
Dave
I also opened up the motor since I needed the oil pan and had to un-seize the motor to remove the torque converter.
Hmmm... I think I can guess which bearings are the problem ones.
I think I found the problem lol.
I am only running the LS PCM, I built a module to run the RX8 gauges and electric power steering over CAN-Bus
I started finishing the motor assembly, but I realized that I only ordered 2 lifter trays when I need 4.... So I only have one head installed and I have to wait for the other trays.
I also wanted to flow test my 42# eBay injectors to see if they are matched. So I made a very cheap test rig with a 2L bottle, some air fittings and hose and a micro controller with a relay. I set the air pressure to 35PSI and have the controller run for 30 seconds with a 75% duty cycle. When it is done I just mark the bottle and run the other injectors to see if they flow the same volume. They all seemed to be pretty consistent.
I also modified the valley cover to finish the DOD delete. Basically I cut a hole in the small port coming off the pressure sensor port and TIG welded both sides shut. So now oil only goes up to the sensor and doesn't continue in to the ports in the cover. And then I just removed all the solenoids so its just an aluminum cover now.




