GT3 High Idle Speed - With A Difference

[Zulu10]

Can anyone on here think of a possible cause of this aberrant behaviour on a 1997 Esprit GT3 with Delco fuel injection, please:

When the air conditioning is switched off, the car behaves impeccably, and idle is sweet and stable.

However, when the air conditioning is switched on, the idle speed will rise, inconsistently to somewhere between 1500 and 3000 rpm. Very occasionally this will be accompanied by a 'Check Engine' amber caption which clears within seconds.

Ordinarily, for an idle speed problem I would be looking at the IACV and for air leaks.

However, prior to recognising that the problem only occurs when the air con is running, I had already bought a new IACV, and it made no difference.

 

Regarding air leaks: when the rubber pipe feeding the IACV is artificially compressed, then the engine can be brought almost to the point of stalling, so that suggests that the IACV is pretty much the only source of intake air.

So those two facts suggest to me that the idle system is doing what it should, and indeed it does whilst the air con is off.

 

The air con is such a simple system that interacts with the ECU by requesting compressor on, in response to which the ECU pulls in the relay that provides power to the compressor clutch, that I can’t see how a problem can occur.

Interestingly though, once air con is switched off, any attempt to restart it is ignored and only actioned once the ignition is cycled through off.

 

So, what can the air con do to affect the idle? My shortlist is:

EMC/RFI – could the air con clutch generate interference?

Direct electrical cross feed due to wiring loom problem? Unlikely because car has only done 21,000 miles and in general the loom seems to be in excellent condition.

Vacuum pipes? How can a loss of vacuum, for example at the dashboard, adversely affect the engine idle on an engine whose only vacuum control is of the wastegate?

ECU Problem? Could the code somehow be corrupted such that the ECU behaves erratically?

Bad earth causing loss of reference voltage. But the IACV and throttle pot are both connected directly to the ECU from which they receive their reference, so that seems unlikely.

 

Does anyone else have any other theories, please?

 

[fjmuurling]

In the AC request circuit there is also a low pressure switch.

If the switch behaves erratically (due to faulty switch (connections) or just enough pressure in the system) the AC will be switched ON - OFF - On etc (there is some debounce in the software of the ECU). 

This can explain the erratic behaviour of the Idle speed, as the ECU wants to set the idle speed for when the AC is ON / OFF

[Zulu10]

That's an interesting point, but wouldn't that just cause the ECU to jump between the 'no load' idle speed of 950 rpm and the 'AC on' idle speed of 1050rpm? Whereas in this case it's going right up to 1500 to 2500, and sometimes 3,000 rpm.

[jonwat]

As its throwing a CEL I'd definitely read the codes to get a better idea of the problem 

[Bibs]

A reset of the IACV wouldn't go amiss too. 

Have you got Freescan/Espritmon? @internets can help with the required cable. 

[Chillidoggy]

As a new owner, I recommend your first purchase should be the Lotus Service Notes/parts manual. A mine of information.

[Erikl]

ac clutch on its way out? Dragging /loading engine?  Then sticking and wont release until you turn car off?  

I think the fact that AC wont come on unless you turn car off is important piece of the puzzle.

I am not any kind of AC expert.

[Escape]

Does the AC still work as it should? If not that definitely points to a problem in the AC system that's effecting the engine at idle.