SeaGull

My car failed inspection due to high CO and low lambda. These values were tested at 2500rpm.
The engine is running rich, it smokes, stinks, I think it missfires, and the sparkplugs are black.
There are no fault codes in the Motronic. The lambda-loop is working.
I think the distributor, wires, and spark plugs are OK. I have measured the sensors as described in the Workshop Manual, engine temp. OK, oil temp. OK, throttle pos. OK, MAF OK.
There are no description of how to test the knock sensors, but there is a 'knocking' pin, 19, in the diagnostic connector. If I measure the voltage on this pin relative ground pin, 10, at idle, I read a stable full voltage. Anyone know how this pin works?

SeaGull

If it is a knocking problem, is it possible to remove the bolts and mechanically isolate the sensors from the engine, still electrically connected?
I get no fault-codes if I electrically disconnect the sensors.
If I disconnect the hall-sensor I get a fault-code, but the car runs exactly the same.

mfyoung1086

Have you tested the O2 sensor yet? Get a torch, heat it up, and measure the readings to see if its OK

Also may seem like a simple test but have you checked your timing?

SeaGull

I have a Innovate wide band lambda, so I can see the loop working, but the value is very unstable (due to missfiring?). Maybe I should buy a new standard lambda sensor.
"Also may seem like a simple test but have you checked your timing?"
Ignition timing? (valve timing?) With a stroboscope light? I have thought about it, but I don't know what to look for.
I have not used a stroboscope light in many years, since the contact breaker distributor days.

968 GUY

I can't see what year and type car yours is, but on my 87 951, you take the rubber plug off at the distributor housing and fire the timing light in and look at the marks.Just did this myself recently.

SeaGull

My car is a 968 -93, and it has this rubber plug.
What should I see at idle, if hall and knocking sensors are OK? At 2500rpm?

968 GUY

Ah a 968. I have 2, so maybe I can help a little. Although I haven't had a problem like you describe. I just checked the timing on both 968's today.
First, do you have a timing light? If not borrow one, or buy one if you think you may use it often.
You may get lucky and see the mark on the cam drive gear without putting white marking first.
I had to hand turn the engine til I saw it then make sure it was at top dead center.

If you don't have a shop manual...get one, that and the parts catalog are very helpful in these situations.

There is an open square hole at the left rear of the engine that should show a top dead center mark when the mark on the cam drive gear is lined up on the gear cover.
Also check Clarks Garage.. very helpful.

When you think the two marks are lined up, then you can prep the car to start it and fire the timing light on the inspection hole , where the mark on the cam drive gear should line up with the mark on the gear cover.

968 GUY

At 2500, the timing probably wont be in the window any more.
I guess check for standard stuff. Vacuum leaks, vacuum hose off your fuel pressure regulator, idle control valve not working properly, fuel injectors sticking, bad O2 sensor...think you said you checked temp. sensor, but you know there are 2, one is the sender to gauge, the other to ECU. Also 2 knock sensors if I am correct. Bad fuel pressure reg., bad wires, plugs, dist. cap and rotor, coil.

SeaGull

I tested for vacuum leaks with a empty can, bicycle tube valve, and a bicycle pump.
From the Work shop manual: "build up a positive pressure of approx. 0.5Bar in the intake system", "Leaks will cause the pressure in the intake system to drop quickly".
If I pump up the pressure, waits one minute, and then pulls the oil dip stick, there is still some pressure left. This should be OK, right?
I have also measured the MAF signal. At idle the signal is about 2.4 volts, at 4000 rpm the signal is about 3 volts. Are these values reasonable?

gtroth

Hmmm. Response of the Innovate wideband lambda sensor is a mystery, at least to me. Maybe educate us a little on how that replaces the standard one? MIght be worth testing with the original as you say...

On my C4, high CO failure was fixed by known good used MAF, so you might want to be doubly sure yours is good.

I'd also check fuel pressure. If it's way high due to FPR failure, it can't lean out properly.

SeaGull

I bought this wide band lambda due to this problem, the engine is not running smooth.
This Innovate wide band lambda has two programmable analogue outputs, a analogue output can be configured to simulate a narrow band lambda.
I have tested fuel pressure.
I went to the workshop, they could not find the problem (I could not leave the car there, they only got two hours), but they think a new catalytic converter could help me pass inspection.
I use the car for track days, and they told me track days is not good for the catalytic converter.
Yes, this MAF, it's alive, but is the signal correct?

SeaGull

Still having problems with my car.
Changed to a new OEM Bosch oxygen sensor, but the engine is still running rich.
I measured the sensor signal with a Picoscope, se pics.
It looks like barbwire, is this a measure problem? Other scope pictures on the internet does not look like this.
At idle (left picture) the signal level (should be 0.45V average) looks OK, but at 2500rpm (right picture) it looks high and strange, why?

SeaGull

I now have an adjustable FPR.
If I reduce the fuel pressure to less than 2Bar (normal=3.3Bar), the engine is still running rich!

kev951

well did you change your plugs?? also those afm's are notoriously ****ty.. you can get a cheaper volvo afm for half the price of the porsche one. works great .. as far as the reading.. you need to widen the time scale for a clearer reading. then induce a vacuume leak or ad propane and look for changes in the pattern.

gtroth

I suppose the DME is broadening the injector pulse width to compensate for the lower fuel pressure and keep the stoociometry where it thinks it needs to be. Did it lean out at all?

I am thinking:
1) DME thinks there is more air than there really is, so need more fuel to go with it (MAF signal wrong)
2) DME thinks the air temperature is colder (denser) than it really is so need more fuel to go with it (MAF air temp sensor, NTC1, wrong)
3) DME thinks the engine temperature is cold and that cold enrichment is called for (engine block temp sensor, NTC2, wrong)