89 GTUs revs up and down at stop
 

On my 1989 GTUs, which has been running great for years, two things happening. Engine was fully warmed up and suddenly, the engine revs up and down at stop. I can't tell what the range is because the tach stopped working too. Since they both seemed to happen at the same time, they might be connected but I cannot find any related posts. Appreciate any thoughts from the everyone here.
TIA

Scan error codes...?
Your car has an extensive error code library, so I bet there's a bunch of error codes stored.

Bouncing idle is most likely TPS.

Dead tach is usually an indication there's a problem with the trailing coil pack.


-Ted

The code appears to be a '1'. It's a short pulse followed by 4 seconds off so that's what it looks like to me. The shop manual says trailing pack so that's in line with your suggestion. Does it have to be the pack itself or could a plug wire be bad, resulting in the same code?

It's the pack. The signal for the tach comes off the pack and should be unaffected by the wiring.

The bouncing idle is usually an out of adjustment TPS, but it could also be caused by a vacuum leak.

Be sure all your coil packs have the 10mm nuts on them. Those nuts provide grounding for them. Also its rare for trailing coils to fail. Replace it and see what happens. Your RPMs should come back. After your car is warmed up and at normal operating temps unplug the tps and see if the idle stabilizes. It should. I can say this because I had a GTUs with the same problem. It's very common for tps adjustments and or a vacuum leak if you removed the intake manifolds.

Thanks to everyone for the suggestions. At this point, I cleared the codes and started the thing up. Everything is now fine, no changes made. It's possible that the computer just glitched and shut the pack down? I guess I'll just have to watch it for now.

It could be just an intermittent glitch...
Remember, your car is approaching 25 years of age.
Mazda isn't known as the best in reliability on top of that.
The ECU stores error codes until you clear them out.
All it needs for an error code to be stored is just one wrong signal...
Glad to hear you solved it pretty easily.


-Ted