Subject: '86 RX-7 ECU Wiring Diagrams
From: Paul
Date: 8/22/2000, 9:48 AM

Perry Mick wrote:

Attached is a zip of the schematic for the use of the '86-'88
Mazda ECU.

The '89+ and certainly the 93+ were a different computer and
I have no knowledge of those.

Basically the computer takes in 10 inputs, including the
critical crank angle sensor.  Loss of any of the inputs except
crank angle sensor has a fallback, or limp home, but I have not
tested all of these for failure to see how the computer responds.
I do have the two-led readout from the computer installed in the
cockpit, so if the ECU indicates a failure of any of the input
sensors, I will know which one it is (as I'm either descending
with little or no power and looking for a place to land, or flying
normally).  I do know that failure of the oxygen sensor is a
non-event.

The computer outputs control several of the emissions functions
that I removed.  Removing these had no effect, the computer is
still trying to control them.  The computer still operates
the engine just fine controlling only the injectors and ignition.
I retained the stock atmospheric pressure sensor and that is what
I use for altitude compensation.  It is located in the cockpit
behind the passenger's head along with the computer itself.
So I have a single level throttle control, no mixture control,
the computer takes care of that.  All I can say is it purrs along
just as well at 12,000 as it does at sea level.  I have no idea
where the programmed cutoff in altitude is, if there is one.
I do have a manifold pressure gauge installed so I can judge
relative engine power with throttle and altitude changes.

Someone has mentioned that the ECU turns off the engine if it
detects a overheated condition.  The only input that I could guess
it would look at for that would be water temp, and it picks that
up near the water pump.  Now I may be wrong, but if the engine
overheats due to lack of coolant I don't think the sensor near
the water pump would detect that.  The other coolant sensor is in
the upper rear side housing, which is a direct measurement of engine
temp, and that just goes to the coolant temp gauge.  Maybe it's
the '89+ cars that have the computer that shuts the engine off, or
maybe this is just an urban legend.  Nevertheless, I will test
that by simulating an overtemp electrically into the computer input.

The only other temp related inputs are two air temp inputs, one
is air temp at the airflow sensor and the other is air temp at
the dynamic chamber.

One thought I had was to use two stock computers, one computer
controlling two coils with leading timing only for the leading
plugs, the other computer controlling two coils with leading
timing only for the trailing plugs.  This way one would have
a redundant ignition system.  But it would take quite a bit
of experimenting and rewiring.  May do it eventually on the
test stand I'm building.  But you would still have only one
crank angle sensor, so you can't eliminate all sources of failure.

Use at your own risk.  No warranties express or implied.

--
Perry J. Mick
Duckt N7XR
http://www.ductedfan.com


The Aircraft Rotary Engine Newsletter.              Powered by Linux.

http://home.earthlink.net/~rotaryeng/            http://www.linux.org