History:
86 car, 60 degree V6, 730 & 749 ECM used to run 8F (Turbo Grand Prix) code mask, 2002 Northstar throttlebody, Moates APU1 emulator and socket, manual tranny.

Accidental grounding of the USB plug to neg or B+ wire of cig lighter fried emulator and ECM.

Replaced both, car restored to normal function.

Cleaned up wiring harness by redoing it. Back probed to make sure it was right.

Cranked the car to test the harness. Started fine on 3 attempts with no codes and all sensors registering on TunerPro 5 data dash.

During the third engine start I observed the TPS% jump from .39 to 14% and the idle immediately respond by raising to ~2400 rpm for no apparent reason.

Swapped ECMs and got the same result.
Swapped in two TPS sensors, one of unknown history and the other new but dropped before install. Same result.

Bypassed the harness with three direct wires to the TPS sensor, same result 14% with throttle closed. Above occured about 3 mo ago.

Tried a fourth used TPS today still attached to the used throttlebody and got 8% but suspect that was due to the stock adjustment screw setting.

Probed the terminals and all appear to have correct status, 4.x volts, confirmed ground and sense wire. No volts between ground and frame or sense and frame.

I've also noticed as of late and a few months after the problem developed that I've been having IAC synch difficulty that I do not recall having before whenever ECM power is cut, chip is swapped or a large file change is made via the emulator, resulting in intermittent high idle and an occasional IAC code until the ECM finally synchs. Haven't had a chance to swap IAC motors yet.

I ran 100% meth injection briefly but that was about 6 mo prior to the development of the problem.

The only common parts between the ECMs used are the harness and the Moates chip socket adaptor. I replaced the chip used with the damaged ECM also.

I do recall making a swap between two IAC motor wires that appeared to have been backwards initially before the rewire of the harness but that's it.

No idea where to look now. As I don't believe there is a problem with the TPS which has been modified to adustable in order to get a proper closed throttle reading.

 

I thought the ECU "learns" that car is closed throttle?

Key on engine off at WOT what is the % and voltage shown?

 

I don't recall checking WOT since it shows up normal in the log. I'm pretty certain the issue is with the rewire, as I've tried an additional two OE TPS sensors in the last week that each posted 8%, no doubt due to the stock throttle stop adjustments being in place.

As I mentioned above, after the rewire the motor started fine and posted the proper TPS value at closed throttle and then right before my eyes on the third start, jumped from .39% to 14% within seconds of being started up.

I doubt the relearn theory, if the signal can not be converted by the ECM back to normal closed throttle range, the entire TPS vs AE table will be off, not to mention the high idle that resulted.

My guess is some kind of inductance somewhere in the harness or a problematic wire arrangement between the sensors. The engine grounds are all at a common location.

Two computers, 5 TPS sensors, bypass harness wiring test = same result. The only thing I changed was the harness so that has to be the problem. I'll have to wire it exactly the way GM did in the diagram.