[Tomba]

Have to say it runs beautiful from the clip. Very steady and nice tone from the exhaust.
I can hear/notice the single interrupts you mean. Might be a bit common on these engines. I remember my dads B36 doing the same once in a while. S38/S50 engines don't like to idle for a long time. Spark plugs look very dark after prolonged idle. The heat number of the spark plug is adjusted to the conditions it runs on top power and might influence these light loads to operate flawlessly. How does it run if you have driven it hard for a long time -> clean combustion chambers?

To pinpoint the issue I would monitor the ignition primary/secondary oscilloscope views. Burning time and wave out of the signals could indicate something cylinder specific.
An extraction point before the catalytic converter to a 4-5 gas analyzer could tell you if the mixture is to rich or not burnt completely (high HC values).
Measuring the lambda voltage could help, long lean or rich times after the phenomena could tell you additionally information.

Some oscilloscopes have a functionality to present the crankshaft acceleration by measuring the crank position signal which would help identify the issue.

Unfortunately 1980 Motronic doesn't have these functions from the ECU ;-).

I remember member Stevie mention that after long traffic jams the owners manuals tells you actually to rev the engine higher to clean it.

There is a faint whistling/whining sound, coming from the front/bottom of the engine when I press gas pedal and engine revs -

Maybe alternator bearing. I would first remove all belts 1.(A/C pump, secondary air pump) 2.(Power steering) 3.(alternator-waterpump).
Start engine and check if the noise is present. if not, its in one of the ancillaries (use stethoscope to ID noise), else could be engine related (chain / oil pump).

 

[Tomba]

Do you have access to tinyads or an ICOM?

You should have enough information on the DM33M301.prg to log motor condition when its cold and warm to isolate what the problem is using Testo.

P.S. if you can find the OE Flash file, i can burn it and send it to you.

Hi Myke,

TS has S38B36 = Motronic 1.2 -> D_MOTOR.prg? , B38 = M3.3 -> DM33M301
Haven't got success connecting by Tool32 on M1.2. Maybe I am missing something. Used Bosch KTS300 for these.

 

[Tomba]

Thanks Tomba, a lot of people say this is the way it is supposed to run, but I have my doubts. The video I posted was a cold start, it does pick up a bit of a shake when warm - see below:

You can see my hand and cables shake when I place it on plenum. Partially I think it might be the fact that gearbox oil warms up and allows the input shaft to vibrate more (I do have typical carrier bearing rattle when transmission is warm).

Really sounds like a misfire to me. Without additional measurements it is really hard to pinpoint.
Remove the ignition lead cover and spray some water near it (plant sprayer). If an ignition component is failing it should be visible, engine might stall. Even new components can fail (happened enough times to me)
Suggest to measure the o2 sensor voltage to check if it is running lean.
Remove the spark plugs to visual check them.
Another option is to remove each ignition lead one at a time while engine is running to identify which cylinder has the biggest impact. Be sure to use isolated tools to prevent electric shocks! And don't do it for a long time as it can hurt the catalytic converter!
This is a so called cylinder balance test. Our Bosch ignition tester can do this by a push on a button. The ignition lead/cylinder with less noticeable change is suspected to be the one failing. As good working cylinder has more impact.

 

[Tomba]

If it was a misfire, then it would misfire all the time, no?

Engine conditions change once warmed up.
valve clearance becomes different. Piston rings can behave differently. Expansion off different components.

Have you done a smoke test also with hot engine?

O2 sensor is new, and I have no codes. I did remove the spark leads one-by-one with the engine running, and each unplugged lead made the car run noticeably worse - so I honestly don't think this is a misfire.

Good sign. Seems not cylinder specific.
Do you have the CHECK ENGINE light? Does it go on once you turn the ignition on (engine off) and turns off after the engine runs?

At this stage I would like to know if the engine runs lean, rich or closed loop. Do you have a voltmeter and are able to measure the lambdasonde voltage?

 

[Tomba]

My CEL light seems to be functioning properly. I did stomp code test and got 1444, so it appears there are no codes stored. CEL turns off after the engine starts and does not come back.

The CEL is triggered once the lambda control boundaries (+/-25%) are reached and once the lambdasonde is faulty. 99% sure the engine isn't running to lean or rich as the CEL is off.

Maybe it is the fuel pump delivery, although I would expect hard starting at cold (more fuel necessary). Volume flow should be 2.1 liters/minute at 3.0bar back pressure. To do this, take the hose from the fuel pressure regulator and activate the fuel pump for 1 minute. Then measure the amount of fuel collected.

Is it possible that old fuel is still in the gas tank? Since it wasn't driven much... ?

I do have a voltmeter, but O2 sensor connector is isolated - how do I measure it? Do I need to peel back the connector cover and check while running? Do you have the location of the pins I need to be checking and what are the values supposed to be?

You could measure this at the ECU connector or at the lambdasonde connector self once the car is running.
ECU pin#10 and #28. lambdasonde connector grey and black wire. If you remove the rubber protection you should be able to insert pins to measure at the terminals.
After starting I would suspect 0.45V and once warmed up it should move between 0.1 and 0.9V. Slow at idle, faster once the engine speed goes up.

Maybe besides the smoke test do a compression and leak down test on all cylinders just to be sure with warmed up engine.

Do you have brake cleaner? try to spray this near the intake manifold parts. Once a vacuum leak is present the engine will behave differently due to additional fuel (brake cleaner) being entered to the engine.

Sa far I am quite clueless..

 

[Tomba]

I have a feeling the tune might have something to do with it as well.

Hi Myke, from all the tunes I have seen none off them adjust the idle fuel correction map, ignition map, lambda control or idle speed. Only adjustments (retarded ignition) to partial low rpm/load ignition area to prevent sudden ignition jumps that cause on/off behavior on these cars from idle to part load (pure drive ability improvement). Once the idle switch is active, partial maps are not used. Of course wouldn't hurt if we swap it for a check.

At this point TexaZ3 isn't able to perform a certain amount of measurements to identify the problem. From the other side of the planet I can't be of much more assistance. At this stage the recommended measurements are open, I would suggest you take it to a specialist or invest money in tools to measure it on your own.

open points

• ignition oscilloscope measurement (primary / secondary)
• cylinder compression test and leak rate results (check if hot engine changes these numbers) health check. Although good service is provided in the past you can't take it for granted it is in good condition. Even the best maintained cars can fail.
• fuel delivery check (2.1 liter / minute @ 3.0bar back pressure -> measured from return line at 12V @ fuel pump)
• Check voltage at the fuel pump, poor electric contacts can increase resistance and thus performance to the fuel pump.
• 4 gas analysis of the fuel mixture after/before catalyst. Might consider to measure the cylinder individual ones as extraction points are available for each cylinder.

Good luck! Keep us posted.

 

[Tomba]

• I can check compression - I do have the gauge. Shoudl I do this with all plugs out on a cold engine, or run it first, take all plugs out, and check this way?
• Previous owner did a compression test about 1k miles ago, and the numbers came as follows. I do have records and pictures of these readings, so I believe they can be trusted:

I should heat it up remove the plugs and check it with hot engine. Once cooled down (at least 6 hours) perform again.
Did the rough idle happened directly or got worse by time

- How would I check the voltage at the fuel pump, could you walk me through it?

In the trunk there is a cover which you can remove with (4?) philips screws. Fuel pump wire color should be green/purple.

I honestly think my whistling sound and my issue with engine vibration is related somehow. I think once I figure out what causing the former, the latter will resolve on its own.

Did you already remove the belts to make sure it isn't coming from ancillaries?

For the record - yesterday was a warmer day, and I ran A/C - the car idles WAY smoother with A/C on. Way smoother. So the question is - what changes for the idle when A/C is on? Some sort of different DME program?

There is a wire going to the DME to tell the ECU the A/C is on. Assume idle speed is lifted and maybe different control settings (PID) are used. But I am not sure about it. I have noticed this also on other 3.6.

I did balance ITBs. It made no difference. I really don't think that air supply is my issue - if anything, I think its the air-fuel mixture that is the problem. I bought the Carbtune tool, so I could re-balance the ITBs at any time.

I would suggest to replace the bearings as well just to make sure.

 

[Tomba]

Well good it is resolved, still strange your CEL won't go on as I would expect the lambda regulation boundaries should have been reached.