Please select from the menu at left for how-to's and articles about the MR2 turbo.
The first modification done to my 1991 MR2 turbo was the installation of an electronic boost controller, the APEXi AVC-R. Boost was upped from the factory value of 11.2psi to 14.2psi (or 1.00kg/cm^2). A dyno run was done and the engine outputted 182.3HP and 201.0ft-lb of torque to the wheels. This boost controller works very well, and has two settings. I currently have then set to 1.2kg/cm^2 and 1.35kg/cm^2 (19.5psi). To convert from kg/cm^2 to psi divide by 0.0703. The APEXi AVC-R is setup with the user specifying a starting duty cycle for the solenoid valve which controls boost. Installation involved running some wires from the ECU (in the trunk) to the passenger compartment (switched +12 volts, injector signal, ground etc.). It is best to use a shielded cable, in my opinion. The solenoid valve was installed near the stock wastegate actuator. The stock VSV was bypassed - hence the ECU no longer has control over turbo boost pressure. The APEXi AVC-R displays current boost (it includes a pressure sensor that is T'd off the intake manifold), and current injector duty cycle. Upping the boost to 14.2psi was a noticeable change in performance - what fun!
The next modification was the installation of a K&N FIPK. The best part about this modification is the wonderful induction noises that the driver now is exposed to. The stock air box does an excellent job of quelling these noises. I really prefer to hear them. I have no idea how much HP was gained or lost, but the sound was well worth it.
Next was the installation of a custom 3" exhaust from the stock primary catalytic converter back The secondary converted was replaced with a pipe. This particular exhaust was quite loud, but livable in my opinion. No idea on HP gains here. I eventually purchased a Gtech performance meter which, I find to be, very useful. Installation was OK, except I broke a stud on the catalytic converter. I ended up removing the converter, bringing it to a shop, and having it removed. I also had a shop remove the stock exhaust, as I could not life the car up high enough to remove it. Installation of the exhaust was easy, and removal of the aftermarket exhaust is much simpler than the stock one.
I then installed a Blitz BOV and vented it to atmosphere. What a great sound! I did notice a good amount of difference with this valve over the stock one - although I believe the stock one may have been leaking. Between shifts, boost was available much sooner. Installation was straight forward.
Next up was a Greddy intercooler. At this point I had installed temperature sensors in the intake pipes before and after the intercooler. I have graphs showing the temperatures from the stock IC and from the Greddy IC. Suffice it to say, the Greddy IC is far better. It takes much longer for the Greddy IC to reach high temperatures - but it also takes longer for it to cool during cruising as compared with the stock IC. This makes sense, since the Greddy IC is much thicker. Removing the stock IC was a little of a pain but it eventually comes out. Installation of the Greddy IC is easier than removing the stock IC.
I then removed the stock turbo and sent it to Performance Technique where they installed a TO4E-50 trim wheel. Removal of the turbo takes some time - I just followed the service manual. After installing the upgraded turbo there was a good difference in performance. Again I do not have numbers here. Where the stock turbo would 'run out of breath' past about 5000RPM at 14.2 psi of boost, the upgraded holds it rock solid to redline. I felt comfortable (and so did the ECU) upping the boost to 1.10kg/cm^2 or 15.6psi. I tried going to 17psi, but the ECU would go into 'bad gas mode' (see TVIS/VSV LEDs). With the different turbo in place, the Blitz BOV would not flutter during shifts. I was told that this could be back for the turbine, so I vented the Blitz BOV back into the intake, and it no longer fluttered. The sound was very muffled, but can still be heard. This can be seen in the following picture, as well as an installed SPAL fan, Greddy Intercooler, water injection, and FIPK. The wires coming out of the heat shield over the turbo are for two exhaust gas temperature probes.
To add more fuel, I installed Supra Twin Turbo fuel injectors. Installation involved moving the throttle body out of the way (no need to disconnect coolant lines). It is important to change the fuel injector O-rings and the 8 insulators. I found that the engine would run OK without any other modifications, just more rich (idle was OK), but according the the Gtech performance meter, power was down.
Consequently the APEXi AFC was added to adjust the air flow signal from the AFM to the computer to compensate for the larger injectors. I then upped the boost to 1.20 (17psi) and the ECU seemed happy with it. The AFC is a really neat device and certainly fun to play with. I tuned the car by starting out with the AFC zeroed out, and doing Gtech HP runs. I then leaned out the mixture until either the ECU would go into bad gas mode, or the power would stop rising. While doing this I watched the voltage from the O2 sensor and the exhaust gas temperatures. I then richened things up to be more safe. I ended up with a low graph of just -10% across the board, and I high graph with it starting at zero, going down to about -8 (5000RPM) and then up to about -4 (7000RPM). This seemed to be a good combination that worked for me at about 19psi of boost.
I then added Water Injection, with the hopes to up the boost further. Unfortunately the turbo would not push more than around 17psi of boost. I thought that the primary converter was clogged and removed it. It was clogged big time! I gutted it with a large drill bit and screw driver (kinda fun actually!). Reinstalled the converter and was able to run 20psi of boost without a hitch! I settled on a dual solenoid water injection setup, with a total of ~650cc/min of water under full boost. I then decided to go ahead and try 22psi, with gave more power, but with the onset of cooler weather, this causes (I'm guessing) a large ping, which caused the knock sensor to blow. I replaced the sensor with a Toyota one, and lowered boost to 19.5psi. I feel that without the water injection, I would not be able to run more than 17psi safely. I'm currently using NGK copper plugs gapped to .0025" - any more and the engine will miss under boost. This is with an MSD 6a ignition box. The TO4E-50 performance techniques upgraded turbo will push at least 22psi of boost with the converters gutted. I'm not sure what I will do come emissions time - might have to replace the converter then.
I've taken the MR2 to Summit Point Raceway and ran a 20minute session on the track with an instructor. It was incredibly fun, and I really want to do it again. For this event, I lowered the boost down to 17psi, kept the same AFC map, and water injection setup. The car ran fine for the 20 minute session, but when we came into the pits and started to idle, the temperature gauge shot up, and coolant started to come out of the overflow. I shut it down and let it cool, added coolant, and it's been fine ever since. Since then I've replace the coolant, and all seems OK, although I havn't (and can't) push it as hard as a 20 minute track session since then. I highly recommend a track session with an instructor, it was very fun and I feel that I learned a lot.
Check out the new dyno page. Better 'tuning history' section coming in the near future.
Engine was rebuilt using ross racing pistons, and Cleavite bearings. The Stock rods are still used, but they were shot peened. Check out the engine build up page for more information.
The MR2 turbo has been sold for about 1 year now. At the time I had installed an AEM engine management system to better control detonation and to control the Supra 540cc/min injectors. This incredible system allowed absolute control over every possible tuning aspect of the 3SGTE engine, including knock retard, ignition timing, warm up maps, and all variants of fuel maps. It was a great learning experience for me, and I enjoyed constantly tweaking variables. I would say that it took about 15 months to get the tune right in all weather conditions. Oddly, full throttle was the easiest to tune, it was the part throttle operation that took the most fine tuning to get the mixture and timing correct to offer good toque and fuel mileage. I used a wide band sensor and a GM MAP sensor along with the AEM. The MAP sensor is required since the Toyota sensor doesn't register boost above about 16psi. The wide band sensor coupled with the AEM is a great combination since the AEM can use the wide band sensor for closed loop mixture control. You specify a map that contains load on the y-axis and RPM on the x-axis. For each cell you specify the fuel mixture you want to run - super cool.
I know that the the new owner is enjoying this wonderful car. I've moved on to the BMW world with a twin turbo charged BMW 135i - you didn't think I would get a non-turbo engine did you?
|< Prev||Next >|