Computer questions

So I posted about this back in August but have since done a ton of troubleshooting and have more info. 1996 Jeep Cherokee sport 4.0 automatic trans. Torque Converter Locks up and unlocks at the opposite time as it should. I checked this with a test light coming out of the TCM to the torque converter lockup solenoid to verify what I was feeling. During warm up and while coolant is still cold torque converter will lock up completely even at a fairly low speed while still accelerating (above 40) once the Jeep is up to temp which sometimes takes a while the torque converter will unlock while cruising on flat ground at 55 or 60 etc (light off) and then will lock up (light on) when under load such as going up a hill or trying to accelerate. all gears shift fine



this is completely backwards. and I cannot seem to figure out why it would do this. could the TCM be sending reversed signal or would it be the PCM sending to the TCM? I replaced the TCM, TPS, changed solenoids, adjusted kickdown, and now am seeing that the signal is in reverse. I do believe the computer is a replacement. should the computer be able to communicate with the TCM? wiring would cause just intermittent signal not a full reversal. because its only sending power to a single solenoid I can only see it being the computer.



I do have a trans cooler, and a large aluminum radiator but from my understanding that would only slow the engagement of TC due to increased warm up time.

Hm. Has it always done this, or has it only started doing this recently?

Given how 96s are a transition year, it wouldn't surprise me at all if if the logic polarity for this signal got reversed as part of the 97+ updates. What leads you to believe that the TCM has been replaced?

The TPS signal was backwards on the older Renix jeeps, and you had the dual output TPS providing a normal 0-5 volt output to the ecm and a 5-0 volt output to the tcm. The tcm connector on the older ones was grey, versus green on the newer. In 98, the connector changed shape.

This still screams bad TPS input to me. I would measure the TPS signal at the TCM itself while you move the throttle and verify that it's changes smoothly from about 0.5 to 4.8 (can vary, but doesn't go full 0-5 volt range). It's not unusual to have a bad ground at the TPS, and to add a new ground wire to the manifold.

Is it shifting okay otherwise? About the right speeds and firmness of the shift seems right? I'd probably still adjust the thottle-body to trans cable anyway.

Also check the brake light signal, because if the trans computer thinks you have your foot on the brake it will unlock the torque converter.