Occasional fail to start when engine warm

wow! - no code stored or CEL displayed? You said no spark when failed, is there any indication of flooding at all? The Engine (not coolant) Temp Sensor can fail and folks will may not get a code.

Perhaps the next time it fails, you could spray some kind of coolant (even water if you avoid the HT components) on various sensors that haven't been replaced yet.

you might try posting this at

. for additional input.

Carl

Thanks Carl for your feedback.

There is no sign of flooding at all. It just eventually decides to start, maybe 5 minutes later or sometimes 20 minutes later. Once it has started, then I can immediately stop and restart it always successfully.

My wiring diagram shows what it calls an "Engine Coolant Temp Sensor" and an "Evaporation Thermo-Switch". Are you referring to one of these? I cannot see any other temperature related sensor. Where is it located?

I suppose if the ECU has a routine that says [Do Not Start] if it thinks the engine has overheated, then that could be the problem if the temp sensor has got temporarily stuck out of range. Definitely worth a look.

Regards, Andrew

Nah, that temp sensor (engine - not thermo, though I'm not sure what THAT one is) is to alter the ECU 'map' for cold weather starting/running to replace what us oldtimers called a 'choke'. Usually, when it fails, it continuously feeds the wrong value to the ECU such that the engine stays in 'choke' mode. Gives rough running, sometimes stalling after warm-up and very poor mileage. But I don't think it kills the spark on start. There was an extensive thread at usmb about a similar issue, try searching there. I just can't recall what was found. If the ECU isn't happy, it won't fire the plugs. Something is either giving the ECU bad info - or the coilpack is failing at high temp/intermittently. Have you tried to retrieve a code with a reader? A GOOD reader known to work with ISO OBDII ?

Carl

I JUST remembered. The good news is, a MAF can be intermittent like this

- including the occasional stalling, the bad news is, its very expensive to just throw one on if you're wrong. Um, has this vehicle had a K&N or other brand 'oiled-style' air filter? They can sometimes cause premature MAF failure.

I dunno

Carl

How many miles on the car? How many 3 years ago when the problem first occured? Oh, and has the front O2 sensor been replaced? when?

Carl

Don't know about Suby's but these get fried fairly frequently on GM products. Very similar if not exact same symptoms.

Is the car automatic?

GM had a notorious problem with something called an automatic transmission torque converter solenoid. When the tranny warmed up this 25$ electronic part would start failing, also intermittently. The car would die shaking as the tranny torque converter would stay locked and not disengage when coming to a stop. The distinguishing factor from coilpack/coil module problem was the shaking. If the car shakes before coming to a full stop and then dies it may be that your problem is some solenoid that controls the converter in the transmission (?).

Does it die when idling only, that is, withuot warming up the tranny? If so, then it is not the torque converter solenoid.

M.J.

Forgot to mention, this problem would never show any diagnostic codes (on GM cars), and was a pain in the butt of so many GM owners. It also took aproximately 20 minutes for the tranny to cool, and everything was back to normal until the next episode.

M.J.

No oiled-style air filter AndyH

I had someone who knows about retrieving codes do it and he tells me there was no codes present

I should check the coipack resistances next time this happens.

I think you're right. Something is telling the ECU bad info, so it plays safe and does not fire.

AndyH

3 Yrs ago it would have been about 70,000 Miles Today it is 88,000 Miles

Do not know re O2 sensor, I'll find out.

Andyh

The car is automatic There is a slight shaking when stalling from idle. It has died after a long idle and without warming up tranny (ie parked in driveway), so I guess that probobly eliminates the torque converter solenoid.

Regards AndyH

That would be about the age a lambda sensor would start getting 'lazy' though I suppose it COULD behave as you describe. It is also a fairly inexpensive part that is not too difficult to change. I'd probably change it if I were in you postion. Even if it doesn't cure the problem, it's well over 2/3 its lifespan at least, and much cheaper than a MAF.

I suppose the IAC could use cleaning or something. It's a shame we can't get a code.

Carl

Sound like the lambda sensor should be replaced and I will do that. However as I'm not getting any spark at the plugs, during the can't start period, I'm not hopefull that this will address my main problem. I guess that goes for the IAC too.

Can you see any reason why a lazy lanbda sensor or a dirty IAC might confuse the ECU into not triggering the igniter ?

AndyH

I can't. The system should really go into 'limp home' mode but.....

Again, the most common thing that both will not throw a code(in many cases at least) but yield very poor running conditions is the Engine Temp Sensor.

Maybe if you had a reader hooked up live and could monitor the OBDII output during failure conditions?

Carl

Thanks for all your tips Carl, its really appreciated. I don't suppose you know of some quick and dirty OBDII interface to a notebook's RS232 port? I did hear of such a thing a while ago.

Regards, Andrew

please try searching the matter at

. If you post, mention being 'antipodean' and you may find some mates that hang out there!

I think I recall a site with the schematic for an adaptoer, but then you had to buy the software.

yeah, dig around here;

there are probably others. Check ebay as well.

Carl