ECU Problem?

[neilxjs]

My '89 3.6 has been suffering from an as yet undiagnosed condition causing the revs to fluctuate. When the engine is started from cold it runs perfectly until the engine is switched off, then when the engine is restarted it idles at 1000 when stationary and 1500 when moving. While coasting to a standstill the revs will rapidly oscillate from 500 - 1500 - 500 until the car is stationary whereby the revs sit at 1000. So far the following work has been carried out to try and solve the problem:
Timing and TDC reset; throttle potentiometer, coolant temperature sensor, stepper motor, speed transducer and speed interface replaced; stepper motor reset; and coolant temperature sensor circuit tested from ECU connector blocks. Today, after the car had warmed up I unplugged the coolant sensor and attached a 100 ohm resistor across the connector terminals and the car ran perfectly no matter how many times it was switched off and restarted. Does this sound like an ECU fault? Any ideas would be greatly appreciated as the car is an absolute pig to drive while it is misbehaving in this way.

[poprivet]

Always tricky to diagnose from afar, however just a few thoughts:

As the ecu is in the left hand footwell the connections can suffer from water through a leak or condensate from the A/C.

I assume you have the later black bodied temperature sensor, the early white ones were a large cause of problems.

Worth checking all earth connections in the engine bay, there are many!

High engine revs are normally a result of too much air getting in, either due to throttle body disc not set correctly, stepper motor not set correctly, hose split/perished. I realise you have looked at these already but air is getting in or being let in, somewhere.

The problem can be solved

[ptjs1]

Neil,

My first thought would be air leak somewhere, as Pop has discussed. Very typical symptom on a six-cylinder XJS.

However, you've mentioned that you've replaced a number of items. Is that because you had confirmed that they were faulty or was iy just attempts to cure the problem?

As you've mentioned, a disconnected and "forced-signal" coolant sensor seems to cure the problem which would imply that the problem lies here. Why not try reconnecting the original (or replaced) sensor and monitor the resistance reading with a multimeter whilst the problem is occurring.

Good luck

Paul

[neilxjs]

Thanks for the reply but I have ruled out an air leak after extensive gasket and hose replacement and testing all round the inlet manifold and its appendages with brake cleaner while the engine is running. I think the problem going away when the CTS is substituted with a resistor confirms that it is not an air leak. Could the ECU be misinterpreting the signal from the CTS? I have no idea what goes on in that black box and cannot find a schematic for the Lucas 9CU.

[ptjs1]

Neil,

I would definitely check the resistance readings of the CTS, both at the CTS itself and also at the ECU plug.

It may be that forcing a 100ohm resistance is ok, but I thought the correct figures were closer to 700 ohms at coldish dropping to 300 ohms at Normal operating temp.

I know there has been another instance of similar symptoms which was caused by incorrect seating of the CTS plug.

Paul