89 NA GXL runs solid when cold, but when it warms up, the idle starts behaving erratically. If I leave the car sitting still or stop while driving, the idle will oscillate between something like 1500 and close to 0. Seems to like riding the threshold between almost dying and actually dying, occasionally actually dying. Kinda frustrating to have to sit there and rev the engine constantly at stop lights.

Reading around the forums and FSM, seemed like the likely culprit was the TPS. Found that it was out of spec, and actually needed to be shoved over slightly in order to get it to read right. Fortunately, there was enough play in the mounting to allow that. I believe it's now at the proper settings (used the resistance check method), but the idle problem seems to still exist, unless there's some kind of adjustment period for the ECU to register the changes or something--I only ran briefly after adjusting it and seemed like the problem was still there. Any suggestions on what to try next?

 

The TPS is not adjusted by moving its brackets, but by turning the TPS adjuster screw. This must be done when the car is fully warmed up so that the cold start cam is no longer engaging the throttle plates and holding them open.

After that, it's time to start hunting down the vacuum leak.

 

Well, I think what I started adjusting was the TPS adjustment screw. The screw with the spring around it that clamps together the 2 metal doohickeys on the TPS? Engine was warmed up, but turning that screw only seemed to affect resistance on one of the sensors.

In order to get both to the right spec, I found that I had to wiggle the TPS assebly a little closer to get the DE reading right, and then play with that adjustment screw to get the AB reading right (I think those were the connections I checked, don't have the FSM schematic in front of me at the moment). Bought the car with a recently replaced engine, so conceivably they didn't quite put the TPS back in right.

So I take it that's a "no" to any sort of ECU adaptive period or anything, huh? Any likely candidates for a vacuum leak since this only occurs once the engine is warmed up?

 

Well, further update. Actually took her out today since it was sorta dry, had the same issue with the wandering idle. Turns out either my alternator or battery are bad, wondering if that has anything to do with anything. Furthermore, the wandering idle isn't constant, it only seems to happen if I've been sitting at idle in neutral for a bit, or once I put the clutch in to get back moving at a light or whatever. Does that still sound like a vacuum leak, or would a leak indication be more constant?

 

The ECU is very stupid, it doesn't have any memory or adaption to the TPS. Hell, you could toss it on a 4 cylinder and it wouldn't know the difference.

Only one TPS sensor is effected by the adjustment screw, which is by design. The narrow range TPS is used by the ECU to control accel pump, on/off idle, etc. The full range is used to control the metering oil pump mainly.

Definitely make absolutely sure there are no vacuum leaks, and it would probably be worth pulling the BAC and cleaning it up with carb cleaner.