Baffling fault - overheating plus rough idle

[CMF_Red Menace]

OK, here are the symptoms:

- Car developed a high (but smooth, I think) idle. Not sure of exact RPMs, but it was very high - 2000+. Only happened twice, with a hot engine.
- The second time, 3 days after the first, it happened in traffic, and the car overheated dramatically within two minutes. I got the car towed.

Based on high idle and overheating, I changed the head gasket, even though there was no sign of cross-contamination of oil and water. Then I flushed the cooling system - except in a moment of ludicrous stupidity, I flushed a breather hose by mistake and literally filled the engine. I've subsequently flushed it through with a lot of oil. I know, it was a stupid thing to do...

Thereafter, I queried the ECU, which said that the airflow meter is the cause. I've done the soldering fix (and, ominously, extracted a little water from the sealed circuit box in the process) but the idle remains rough.

These are the facts:

- Airflow system has been misreading at some point, either because it is faulty or because it is water damaged.

- Coolant temperature sensor is within limits - haven't tested the wiring back to the ECU, but the ECU doesn't complain.

- Engine overheated - shouldn't the fan have kept this under control? I switched the engine off just before the gauge hit the red, but it was rising very fast.

- Head gasket wasn't the problem, nor the thermostat, and the water pump is good.

If it wasn't for the overheating, I'd assume it was the throttle body as everything else points to it. If it wasn't for the ECU code saying AFM not coolant, plus the temp sensor reading OK on a voltmeter, I'd assume it was a bad temperature sensor, telling the engine it was at -40C, fuelling accordingly and refusing to turn on the fan.

I'm confused. I'm also obviously stupid for filling the breather with a hosepipe, but there's always one idiot.

 

[CMF_frank2]

red menace

iirc, the fan kicks in if you join the 2 wires on the coolant sensor, that would test the circuit eh ?.
and fast idle problem may be the idle control valve ?

frank

 

[CMF_Red Menace]

I'll dig out a low-value resistor and ground the temperature sensor line through it (don't fancy shorting it for fear of burning the ECU; as far as I understand, it puts a constant voltage across it and measures the current to determine temperature).

Thanks for the advice.

 

[CMF_Red Menace]

Update:

- The Haynes manual is misleading; the photo that points to the engine temperature sensor (for the ECU) and makes clear that this is the sensor, not the sender (for the dash), points to the sender. Hence I'm not yet sure if the engine temperature sensor is bad or not. I shorted its plug out with the ignition on but with the engine off; the fan did not spin. I'll try again with the engine running during the week.

- Having re-soldered the air flow meter, I cleared the fault codes and they have stayed clear. I don't think that the AFM is at fault; in fact, I ran the engine without it and it idled quite well for a while and then abruptly came to a stop.

- Running the engine in diagnostic mode shows that it tends to run lean (check light on more often than not) at idle.

- It drives fine; no obvious signs of power loss &c (though I'm driving it gently in any case).

- Idle can be high, it can be low, it can be OK. Typically, it's absolutely normal when the engine is first started, and starts to wander as the temperature gauge starts to move. Thereafter it will probably idle for five minutes or so, increasingly roughly until the point where it will stop if I don't rev it.

I'm wondering about the idle air control valve. I could see that if it was sticking shut, the idle would be low. If it stuck open, the idle might be high.

 

[CMF_Red Menace]

Update again:

- With the engine running, the fan comes on when the ECU detects open circuit (unplugged coolant temp sensor). Subsequently, I shorted the plug; the fan stayed on (I guess it might have been spinning down, but it didn't particularly appear to be). Those two both being true, I can't see the room for a fault on the coolant temp sensor that would allow the engine to overheat - so I'm confused as to how the overheating occurred.

- After shutting down the engine, I can smell fuel in the inlet manifold if I open the throttle body. I guess that's pretty normal.

- I was wrong about the idle being OK when the engine first starts. Idle is higher than one would expect on first start, higher than one would expect from the fast idle thermowax. With time, it settles back towards normal, but then the idle just keeps falling until the engine starts to miss and cuts out.

- With the inlet air control valve disconnected, it still doesn't run right; the idle still fluctuates.