Well, good luck with this. I'm out of (so far useless :scratch: ) ideas.

Dave

Have you checked the MAP to see if was in spec? If it is, than I'd revisit the repairs you have made to the crank sensor. The oem wiring is shielded from the crank to the ecu because it needs a clean signal from the CAS. I've seen all sorts of aberrant behavior from motors that were getting poor signals to the ecu from damaged wiring.


Paul

i just wanted to update this thread.

i found my long-lost mity-vac vacuum/boost tester and i tested my MAP sensor. the sensor is stuck at 5.0V no matter what vacuum/boost it sees. SO, the sensor is dead and it is making my ECU think it's runng 17psi, hence the fuel cut :)
apparently, before, i was testing wires A and C and should have tested A and B.

thanks for the help guys

That's good to hear.

Hopefully you don't have too much pent-up frustration to deal with.

Dave

i spoke too soon!


the sensor tested bad, so i replaced it with a known good sensor and the problem is still the exact same.........
i'm not sure where to turn to now.

here is my list of CEL codes:

5 - knock sensor
25 - PRC solenoid
28 - EGR solenoid
30 - split air bypass
31 - relief 1 solenoid
32 - switching solenoid
33 - port air bypass
34 - idle speed control
38 - AWS solenoid
39 - relief 2 solenoid
42 - turbo precontrol solenoid
46 - charge relief solenoid
50 - double throttle control solenoid


I don't see how any of those could make me hit fuel cut under 10psi!

another update. i just installed a known-to-be-good ISC and reset the ECU and the code didn't go away...

i am starting to think it is this crappy wiring harness...

The ECU usually throws codes based on incorrect resistance.

I would take a sensor that's throwing a code, check it's resistance alone, then check it's resistance through the wiring harness at the ECU.

Dave

Try connecting your boost gauge to the nipple on the passenger side of the intake instead of Teeing it into the MAP line and see if you get different boost readings when you hit the fuel cut. And while you're at it, use the Mighty-Vac to test your boost gauge if you haven't done that yet. Hook the M-V up to the boost gauge hose at the point where the hose connects to the engine in order to ID any leaks in that line.

when the knock sensor isn't present.. does it show max knock? min knock? or something between?
would excessive knock cause the car to fuel cut as a safety measure?

do you or a nearby friend have a power FC to test with?


it's my understanding then when cancelling the rats nest.. you HAVE to put resistors and not have codes

the ECUs consequence for not seeing the knock sensor, per the FSM, is that it decreases base timing by 2 or 3 degrees. no other consequences.

yeah that's a good idea with the powerFC. i suppose that will have to be my next step.

that's correct about having to put resistors in to avoid codes, but codes don't cause fuel cut, they cause limp mode (and only certain codes at that). apparently the only thing that causes over-boost fuel cut (exactly what this feels like) is the map sensor reading too high...

i have done all of the things you mentioned.

i also moved my boost/vac source for the boost gauge from the passenger side over the the drivers side, then tee'd it off of the MAP signal so i could see what the map was seeing. the boost guage is reading correctly.

it is also worth mentioning that the fuel-cut happens regardless of boost pressure. it will do the same thing at 6psi that it does at 12psi.

So you're measuring an excessively high voltage on the MAP signal wire... try measuring the MAP signal pin on the ECU (key on engine off), measure it with the sensor plugged in and then with the sensor disconnected.

If you measure 5V at the ECU even when the sensor is disconnected, try measuring the MAP signal wire again at the MAP sensor connector. If you measure 0V at the sensor connector but 5V at the ECU, it's likely you have a break in the signal wire somewhere. Many ECUs have a circuit inside them that will cause the signal wire to measure 5V at the ECU if the sensor is not connected, I'm not sure if the stock RX7 ECU has a circuit like this or not.

i repaired and re-wrapped the entire engine wiring harness, hooked everything back up and it still hit the "fuel cut." apparently i still had the knock sensor wire disconnected. I connected the knock sensor wire and it totally went away.

i put resistors in all the emissions solenoids and hooked the knock sensor up and i have no CELs and the car runs perfectly.

So with the entire harness repaired and with a disconnected knock sensor, you would hit fuel cutout.
Reconnect the knock sensor and everything runs fine. Sounds like the FSM might not be telling the whole story.
Was the knock sensor working properly before the harness repair?

yes it was.

I know this sounds simple, but have you tried swapping out your ecu with a friend's?

Long ago, I was working on my FD w/ the battery cables still in place (dumb i know). I accidentally touched the alternator and the ground with a wrench and there was a short. Car started up fine, everything seemed normal. Took her for a ride when i was done w/ my project and the car would hit boost cut almost immediately.

Swapped out the ECU w/ a friend's, problem solved: My ecu was fried and it exhibited no symptoms other than the boost cut. Give it a try (if you haven't already).

yep. tried that.

Well sounds like you fixed the problem by redoing the wiring harness. Nice work! I have a feeling that might be on my list of projects in the future based on the condition of some of the wires I've checked. Hard for me to shell out almost a grand for a new one...

really the problem the entire time was the stupid knock sensor!

So that was probably disconnected even BEFORE you worked on the harness?

Dave

yes that's correct. it has been disconnected since i went back to the stock ECU/harness/coils. (sold my haltech e8 and went back to stock)

however, i did repair SEVERAL wires that were broken or about the break off at the base of their plug (sensor side of harness) and repaired several very weak grounds. prior to this i had a minor 2k rpm hesitation and that is now gone.

My ISC valve was also not functioning due to bad wiring.

also i got rid of EVERY ECU CEL code :)

so i guess it's a good thing that this turned into an involved project, because i've pretty much got a brand new wiring harness now and everything is working as it should.

The effect of the knock sensor being disconnected is a good one to know. Glad this effort wasn't wasted, since you repaired a bunch of other stuff!

Dave

yeah, the Service Highlights Manual says the consequence for no knock sensor is "ign timing retarded," which i assumed was just a retard in base timing, not a huge retard at 6-7000rpms.
nevertheless, it was the knock sensor.

actually, a slightly update: i hammered through the gears a few times and the knock sensor seems to have just died on me. I am wondering if there is any way to remove the knock sensor and trick the ECU into thinking it's there (like with the emissions solenoids/resistors)