Mixture changes during driving

Scott R · Scott R's Garage

Anyone every have mixture change during driving? In the AM when I start it up it runs great, idles at about 950+/- and then after about 20 min or so it seems to rich up and idle up to about 1100. Any thoughts?

fty · 05-13-2002, 08:57 AM

ARRG I have the same problem....any sugeestions?

Makis · 05-13-2002, 03:05 PM

I have noticed something like this on my S2 and in particular after I connected my O2 sensor back ( I am using the car without a CAT so I can connect/disconnect O2). At idle both closed loop operation controlled by the O2 sensor as well as engine rpm control loop may fight each other. As a result the engine idles off the 950rpm mark. It is possible that a faulty O2 (slow reacting sensor) will cause that. Anybody can confirm these?

AFJuvat · 05-13-2002, 04:44 PM

A couple of possible causes:

1. Gunked up throttle body - take off the intake boot and clean it (brake cleaner is your friend) Also use the brake cleaner on your throttle body - rotate the throttle cam while you are spraying into the throttle body make sure the choke and the mating surfaces are nice and clean. do the same to the air flow sensor - you will be amazed at how much crap gets gunked up in there. After it is clean, you may have to readjust your idle to spec. 800 - 950 RPM at warm idle

2. Poor connection at O2 sensor - check, and clean contacts - DO NOT USE anything containing silicone near the O2 sensor.

3. Bad O2 sensor - replace

4. Small vacuum leak - check your hoses.

hope this helps

AFJuvat

Scott R · Scott R's Garage

What worries my is this has been going on since last years smog check. The tech that "repaired" the problem popped the AFM cover and adjusted the mix 2 points lean. It then passed perfect, but the idle sat at 11000 after that. Even after doing the adjustment procedure in my service manual it still creeps up. So I guess I will pull the intake and AFM and clean them out.

todwic · todwic's Garage

They dink around with your settings in order for it to pass smog, and then they are supposed to put it back to where it runs like a Porsche. From what I gather, our P-cars don't care for it none at all. My advise is pick up a book on Bosch injection systems and fine tune the bits and pieces*. One mechanic that I had to try to smog the thing fried my AMF dinking around trying to get it to pass.

*Disclaimer: all advice contained herein is subject knuckleheadedness, please don't do anything todwic the dumbass sez.

mike944 · 05-14-2002, 11:58 AM

Scott-

How do you know that it is running rich?

do you have an air/fuel meter in the car? if so, is it hooked to the same O2 sensor as the DME, or a seperate sensor?

I have a theory that revolves around either a faulty O2 sensor, or a faulty coolant temp sensor. are you sure the engine goes rich? maybe it is really going lean, usually, going slightly lean causes the rpm to jump up.

A little background on ECU operation here. all ECU's are different, i'm not 100% sure how the Bosch Motronic DME works, so all this info is generic, i hope it helps in some way. During warmup, the engine is in open loop mode. It ignores the O2 sensor. it determines injection time, and spark advance off of programmed databases (also refered to as maps) based on the inputs from the rest of the sensors on the engine, these are: TP (throttle position) sensor, the AF (air flow) sensor, the ACT (air charge temp) sensor, and the ECT (engine coolant temp) sensor. Once the engine warms up (determined by some combination of elapsed time, ECT reading, and valid readings coming from the O2 sensor) Ususally, once valid readings start coming from the O2 sensor, it jumps into closed loop mode. Once the ECU believes the engine has warmed up, it starts to consider the O2 sensor reading, and adjusts points on the maps richer and leaner based on the O2 readings, up to a maximum correction percentage. It goes so far as to believe the O2 sensor before it believes other sensors, and corrects to make the readings from the O2 sensor good, even if the info from the O2 conflicts with other sensors.

20 minutes is pretty long before it kicks into closed loop mode. If your O2 sensor is fouled, it could be reading either too rich, or too lean, causing the ECU to correct for a problem that doesn't really exist, which throws off the mixture. OR, the computer goes into closed loop mode, but the ECT sensor says the engine is still cold. The correction that the O2 sensor needs conflicts with the ECT sensor that says the engine still needs a overly-rich warmup mixture, and the ECU sets the mixture somewhere inbetween normal, and overly rich warmup mixture (which would be somewhat rich)

Not quite sure if that will answer your question, but some of that theory might help you narrow down the problem.

Mike

p.s. as AFJuvat said, silicone will contaminate an O2 sensor. even an extremely small amount will destroy a sensor very quickly. DO NOT use silicone ANYWHERE where vapors can get sucked into the engine, this means intake gaskets, throttle body, AF sensor, intake boots, or air box, or air filter. If there is silicone gasket material in any of these locations, this is probably your problem.

Scott R · Scott R's Garage

Some of that makes sense since I do not really have a warmup process. I start the engine and it kinda finds it way up to the 950 mark. Later on it climbs up. Maybe I should try an engine temp sensor.

Makis · 05-14-2002, 01:56 PM

I think Mike and AFJuvat gave good explanation here. I still think the O2 closed loop operation in conjunction with the ECM idle closed loop control fight against each other causing the idle to move up. The easiest way to confirm this in your case Scott is to just disconnect the O2 sensor when you observing the engine running at 1100rpm. If the engine speed drops to normal levels then you know the problem is related to what I said. In this case I think you have 3 option:
1. Bad O2 sensor.
2. Temperature sensor is giving wrong readings.
3 Mixture screw is set incorrectly at the air flow meter.

A good O2 sensor will try to compensate for case 2/3 above thus causing the idle to move up. In either case I suspect the mixture to be actualy right, just little bit more idle.

Ehab · 05-14-2002, 02:27 PM

Another thing you guys are overlooking is the Aux Air Valve, which is supposed to shut off the extra air once the engine has warmed up. I had the problem of idle creeping up once the engine warms up, and that was frustrating, since it was the Opposite of what's supposed to happen.

I took the Aux AV out, cleaned it, tested it to make sure it closes and opens normally, readjusted the opening when cold (which is done with a little nut that holds the plate in place) ... and now, the engine idles at 1100 when cold, then slowly goes down to 950.

If your AAV is not working, then you might be running rich and lowering your idle to 950 at the beginning, then when the O2 sensor kicks in, it corrects for that and the idle rises to its warm level of 1100 ... which you could adjust with the throttle body screw once you've fixed your AAV.

Good luck.

Scott R · Scott R's Garage

Yuck! I hate taking the idle air valve out! Was not so bad on my 944, but the 951 is a pain.

Makis · 05-14-2002, 03:10 PM

Good point. Don't have an Aux Air Valve on my S2 so I guess this does not apply to me but it will apply to the 2.5 engines.

Scott R · Scott R's Garage

Well it was the O2 sensor. So thanks Makis and Mike944. I actually grabbed a new one on my way home. Although I did not get a Bosch sensor, the Bosch was $80 for the three wire. They had a "Beck Arnley" three wire for $25. It is working great! Car runs smooth and never breaks the 950 mark. Thanks again guys!