944 Turbo idle mixture rich - Cycling valve?

[tref]

Gentlefolk,

The idle mixture on my 220 Turbo is rich... before I look at fiddling to correct it, is there any possibility this could be a side effect of a faulty cycling valve?

 

[944Turbo]

Wouldn't have thought so, but guess it could be if stuck closed. Faulty temp sensor might be a problem if its warm and rich.
Tony

 

[tref]

Hmmm... good point... presumably I should be able to test the resistance of the temp sensor... can any-one point me in the direction of figures?

Thanks,

Tref.

 

[sawood12]

Not sure about the temp sensor. When idling the mixture is not controlled by the ECU map but the position of the AFM barn door flap at idle so is not sensitive to temp. The air temperature is sampled upon startup and used to correct the maps for partial throttle positions that are controlled via the ECU. It could be that your AFM signal has simply wandered, which can happen as the variable resistor becomes worn and dirty and the resisance changes, or the barn door spring weakens through age and use so your barn door settles at a different position at idle RPM. I'd look to troubleshoot your AFM first. I did see the video of someone refurbing their AFM on YouTube a while back. A quick search should find it.

 

[tref]

Oh... I thought at idle the barn door was shut and the air volume was down to the iddy-bitty bypass bleed-screw thingy?

 

[sawood12]

The bleed screw thing is used to balance the barn door so you can adjust the air fuel mixture at idle. There is a small channel that air can pass through that goes around and behind the barn door to vary the pressure differential across the door and balance it against the incoming airflow such that the output voltage gives you the correct mixture. You adjust the screw to control the air flowing through this small air passage to change the position of the barn door at idle.

 

[tref]

Sorry - I'm getting my barn door confused with my throttle butterfly[&:]

 

[tref]

Can some-one confirm that this is where the Lambda sensor should be, if my car has one, which aparently it doesn't?

 

[edh]

ORIGINAL: tref

Hmmm... good point... presumably I should be able to test the resistance of the temp sensor... can any-one point me in the direction of figures?

Thanks,

Tref.

Clark's Garage suggests it will run lean if the temp sensor is bad.

"On startup when the engine is cold, the DME temperature sensor tells the DME computer that the engine (coolant) is cold. This causes the DME computer to supply a much richer mixture to the engine (a hot engine atomizes fuel better). When the DME temperature starts to fail, it normally sends a higher than normal temperature signal to the DME computer. The DME computer in turn supplies a leaner mixture to the engine than that which is normally required. So, the engine runs rough (lean) and if it gets bad enough, may not run at all. Check the cold resistance of the DME temperature sensor"
http://www.clarks-garage.com/shop-manual/elect-19.htm

 

[edh]

AFM testing covered here

http://www.clarks-garage.com/shop-manual/elect-22.htm

 

[rcsalmons]

My car was v.rich at idle and low revs - turned out to be the ECU...ouch!!

Rich

 

[edh]

Tref

These people will test and repair ECU's
http://www.atpelectronics.co.uk/product.htm

I have a spare (early) ECU that works fine if you need one.

 

[Diver944]

ORIGINAL: tref

Can some-one confirm that this is where the Lambda sensor should be, if my car has one, which aparently it doesn't?

If you have one, it is on the other side of the crossover pipe (to the right of this picture) on the upstroke and before it goes into the turbocharger. If you don't have one there will be a bung in its place

 

[tref]

Thanks Paul(and all others)... I'll have to crawl under a bit further then! So what is that plug/bung I have pictured for then??

Regards,

Tref.

 

[sawood12]

I didn't think the car ran in closed loop at idle? Closed loop is only used at partial throttle isn't it? - it's not even used at WOT. I'm pretty sure that at MOT's the mixture is measured at 4krpm for cars with Lamda sensors and at idle for cars without.

 

[944Turbo]

My car had a failed temp sensor and was running very rich - basically it thought the engine was cold so was injecting additional 'choke' fuel even when it was warmed up - this was apparent all through the rev range.

The bung is further up in the vertical wall of the downpipe (i think) from the turbo, I think only uk cars with cats (90ish on) had a lambda from the factory and it works at tickover through to around 4,000 rpm where the full throttle map takes over. There is a connector at the rear of the cam cover that has a looped wire on non lambda cars. On lambda equipped cars there will be a plug with a cable running down to the sensor.

Apparently the bung in non lambda cars is very tricky to extract.
Tony

ps the one in the pic is for the emmisions 'sniffer' pipe that allows the mixture to be checked pre cat.

 

[tref]

And the winner of the best suggestion is...

Thanks Tony, temperature sensor it is!

I didn't initially suspect it, as it read the correct resistance cold, and given the need to take off one of the intercooler pipes to get to it, I left it disconnected to run it up to temperature... well actually I didn't because I couldn't stand the black smoke and rough idle... which suggested it was obviously working. Well not well enough, the 80C resistance was a about 100 ohms out of tolerance I believe (have forgotten now). Suffice to say, swapped out for one from an old engine I had lying around, and all is well, MOT passed.

Thanks to every-one for the suggestions and links...

Regards,

Tref.

 

[tref]

Oh, and I did find another blanked off bung further up the side of the engine, which I presume is for the lambda sesnor mine doesn't have... so I think we can categorically state that mine is NOT closed loop at idle[]

 

[944Turbo]

Is there a smug smiley?

Glad it is sorted.