ECU for a 1970 Toyota Corona
Posted: Thu Sep 05, 2019 3:10 am
Well, that title might be slightly misleading. I'm actually interested in an ECU for a Toyota 1JZ-GTE that we stuffed into the Corona. (The Corona is pretty much a micro-car. The result is absurd. This is a Lemons car that we first raced with the original engine and two speed automatic transmission, winning the IOE prize.)
The 1JZ-GTE is a inline six cylinder engine with twin turbos. It was only available in Japan, and this particular engine came from a Soarer.
Unlike the later versions (VVTi and 2JZ) there isn't any complication with the turbo control -- the two turbos are symmetrical rather than sequential.
It has port fuel injection, with six injectors controlled as three pairs. Which is slightly annoying because a pair probably pushes the limit of typical injector drivers.
The 1JZ uses a VR crank sensor with a twelve tooth wheel on the crank, and two VR cam sensors each sensing a single tooth. It appears that the two cam sensors are redundant in this engine, perhaps as a legacy of an earlier generation that had multiple wheels in a distributor to simplify the spark logic. Or perhaps they were already planning on variable valve timing and put sensor on both intake and exhaust cams.
The idle control valve is a stepper motor. Unusually it requires four low side drivers. Apparently the extra coils and wiring was cheaper than a pair of H-bridges in the ECU.
The engine has two resonant knock sensors and a single heated oxygen sensor. It uses a unique 2.2 bar MAP sensor that may have a offset. There are two thermistor temperature sensors, water and intake air.
The throttle position sensor is a single wiper type with an idle contact (contact to ground when the throttle is closed).
Right now my goal is to construct a plug-and-play ECU that allows us to swap back to the factory ECU with no modification. That means supporting two (or three) VR inputs, three injector channels for injector pairs (which implies a current of 2.0 to 2.2 amps), and six logic level ignition outputs to the stock igniter. Secondary goals are to have idle control working (four low-side drivers) and some pretense at knock sensing.
We can easily add a second MAP sensor and a wideband oxygen sensor, which won't impact the original ECU.
My background is in EE and CS, with a broad range of experience. I've done quite a bit of open source software, and designed a number of boards at the hobby level.
The 1JZ-GTE is a inline six cylinder engine with twin turbos. It was only available in Japan, and this particular engine came from a Soarer.
Unlike the later versions (VVTi and 2JZ) there isn't any complication with the turbo control -- the two turbos are symmetrical rather than sequential.
It has port fuel injection, with six injectors controlled as three pairs. Which is slightly annoying because a pair probably pushes the limit of typical injector drivers.
The 1JZ uses a VR crank sensor with a twelve tooth wheel on the crank, and two VR cam sensors each sensing a single tooth. It appears that the two cam sensors are redundant in this engine, perhaps as a legacy of an earlier generation that had multiple wheels in a distributor to simplify the spark logic. Or perhaps they were already planning on variable valve timing and put sensor on both intake and exhaust cams.
The idle control valve is a stepper motor. Unusually it requires four low side drivers. Apparently the extra coils and wiring was cheaper than a pair of H-bridges in the ECU.
The engine has two resonant knock sensors and a single heated oxygen sensor. It uses a unique 2.2 bar MAP sensor that may have a offset. There are two thermistor temperature sensors, water and intake air.
The throttle position sensor is a single wiper type with an idle contact (contact to ground when the throttle is closed).
Right now my goal is to construct a plug-and-play ECU that allows us to swap back to the factory ECU with no modification. That means supporting two (or three) VR inputs, three injector channels for injector pairs (which implies a current of 2.0 to 2.2 amps), and six logic level ignition outputs to the stock igniter. Secondary goals are to have idle control working (four low-side drivers) and some pretense at knock sensing.
We can easily add a second MAP sensor and a wideband oxygen sensor, which won't impact the original ECU.
My background is in EE and CS, with a broad range of experience. I've done quite a bit of open source software, and designed a number of boards at the hobby level.