Weird Engine Stumble - 95 318

[Tim Polzin]

A cool day today, high about 5c (40 f). Car is a 95 318i. I had the car in town, and after leaving it sit for about 1/2 hour, started it (started fine) but it stumbled badly anything above idle when moving until the throttle was wide open. At wide open, it ran fine, (hmmm, maybe a lesson there). Anyway, after a couple of minutes of interesting driving, it seem to run fine again. Ended up doing this a couple of times. No engine codes or check engine light.

Is it possible the AFM is sticking closed or not workking properly?

Any thoughts????

Tim

[RS1060]

Man, could be a whole host of things... bad gas, clogged fuel filter, sticking injectors, fuel pump dying. Do you mean MAF (mass airflow sensor)? If so, it could be a possibility. I tend to think it is something fuel related, but these cars do have a variety of sensors that relay info for fuel, air, and spark delivery so I am not altogether sure. Start simple and move forward. Good luck.

[Tim Polzin]

On the M42 engines, they use a door type air flow meter. The mass flow sensors came later. Fresh gas after the problem appeared, drove about 250 km home with no problem, good gas mileage, good (realively speaking) power, no check engine light nor showing any codes.

OK, more diagonsis. Based on Bentley, the good news is the AFM is working properly. Nice steady variation on pins 1 and 2 as the door opens. +5 volts on pin 1. The throttle position indicator shows an even progression of resistance from about 1.6k to a little over 4k ohms.

The problem seems to be only evident just as the engine is warming up. It is ok stone cold, ok at operating temp. When having the problem, it is most evident at part throttle before the second butterfly opens. Idle is rock steady cold and hot, but is rough when the stumble is present.

I checked the O2 sensor, and the heater has continuity, but did not check to see if the O2 heater was actually getting power.

What would the symptoms of no power to the heating element on the O2 sensor?

Thanks...Tim

[Jeron]

This seems to be a common problem with all of the BMW engines. I've seen numerous similar posts and various fixes. I'm not familiar with the M42 engine.

My experience was similar to yours (although on a 97 M3) and it turned out that the problem was most evident during the cold start phase while the air pump was running. Once the air pump went off, the idle cleared up. I first cleaned the ICV and the problem got worse so I replaced it even though it tested well. After replacement my problem is better but I still have to wait about 30-45 seconds after a stone cold start before driving.

Your problem seems to be slightly different than mine since it is not at its worst during a stone cold start. But I would still clean and check your Idle Control Valve. Make sure the resistance is good and that it doesn't drop out while you move the valve. Other options are vacuum leaks and maybe the secondary air pump valve. Good luck.

[Tim Polzin]

Thanks Jeron. Unfortunately, the M42 with OBD1 does not have a secondary air pump.

I think your right about the lean run condition. However, no vacuum leaks that I have been able to find so far. I cleaned the ICV earlier in the summer and the idle speed is rock solid cold and hot, so it seems to be working ok.

I'm still struggling with the impacts of the O2 sensor heater. Here is a quotation from Jared Fenton's article on O2 sensors:

"When an engine has more fuel than it needs, all the oxygen in the cylinder head is consumed and the resulting exhaust gases contain almost no oxygen. This usually causes the sensor to generate a voltage of greater than 0.45 volts. If the engine is running too lean, all the fuel is burned, and the extra oxygen remaining flows out into the exhaust manifold. The oxygen sensor detects the extra oxygen and the voltage signal will drop to below 0.45 volts.. 0.45 volts is usually the voltage where the car is neither running rich nor lean, this is when the car’s mixture is usually perfect.
Oxygen sensors usually generate an output voltage between 0.2 to 0.7 volts. However, Oxygen sensors doe not begin to generate full output voltage until it reaches about 600 degrees F. Some oxygen sensors have a built-in heating element that lets the sensor heat up quicker. Keep in mind that most oxygen sensors work in an open loop until they have reached a certain temperature. What this means is that until it has heated up, the car will not use the oxygen sensor to measure the mixture, but rather use all the other sensors on the car. The sensor will generate a constant 0.45 volts until it has heated up. Once the sensor reaches a certain temperature, it becomes a closed loop and the sensor starts to generate variable voltage."


So, if the sensor heater were not working on a cold say, would the ECM thinking that the O2 sensor was hot, think it was getting a signal of an overly rich mixture and try to lean out the injection, up until the sensor got up to operating temps from the exhaust gases?

Tim....

[Jared at Pelican Parts]

Um in theory, you are correct, however keep in mind that I wrote that article as sort of a universal guide as to how oxygen sensors work. It varies from car to car. Your car may be different.

Did you run the codes on the car? That is your best indication as to what the problem may be...

[Tim Polzin]

Did you run the codes on the car? That is your best indication as to what the problem may be...

Thanks Jared. That is part of the problem, it is not generating any codes. The only one appearing is 1444 which I believe is "no faults".

I also just checked the O2 sensor circuit out, all is well there, heater and all.

I'm kind of grasping at straws here. It would be nice if it did draw a fault code.

Cheers..Tim

[Jeron]

I know I've seen an article about a warm start problem before I cant find it. I did a little research yesterday and all I came up with was to check the coolant temp sensor. Its an easy check and worth a shot.

Also, is the problem on every warm start or does it only happen if you didn't let the engine finish its cold start phase before shutdown?

[Jared at Pelican Parts]

Jeron is right, the coolant temp sesor can affect the way the car runs when cold. If the sensor is bad, the car sometimes wont even start..

[Tim Polzin]

Thanks guys, the coolant temp sensor will be the next one to check. So, if the sensor says the engine is warm but actually is not, the ECM will try to lean out the fuel? Is there a period on start-up that this sensor is ignored?

The weather is warmer again, and the problem seems to have disapeared, or at least mitigated to a point of not being obvious. They're predicting cool temps by the weekend again.

Jeron, to answer your question, the problem appears regardless of the heat cycle. It will appear when the engine is started cold as well started after sitting long enough to cool down (20-30 minutes) after running at normal operating temps, again with an ambient temp of around 5c or colder. It causes some real driveability problems, to the point of stalling on part throttle. On wide open throttle the engine seems to perform normally, and on higher RPM (above 3500) seems to work better during this period

You guys are great....Thanks

Tim

[Jeron]

Just so we are clear...

It happens on cold and warm starts? The only precursor is the <=5c ambient temperature? After the engine reaches normal op temp the problem disappears?

[Tim Polzin]

Yes, sort of. When the engine is cold, nothing wrong. When the engine is at normal operating temp, nothing wrong. The problem persists at cool air temps for perhaps for 2 to 5 minutes when the engine cycles through from cold to normal operating temps.

When the engine is restarted after sitting for a while and the problem reoccurs, the temp gauge will be in the "blue" range, but certainly not stone cold. If I just stop for a couple of minutes, and restart when the engine is at normal operating temp, the problem does not reappear.

Tim

[kocio67]

I would defenetilly check fuel pressure.
Lost of pressure after sitting for some time may cause a lot of strange problems. Bentley should have specs on that.
Let us know.

[Tim Polzin]

Originally posted by kocio67
I would defenetilly check fuel pressure.

Where is the best place to check pressure on an M42? Bentley was not explicit. Perhaps pull a line at the output of the fuel pump and insert a tee?

Tim

[kocio67]

According to BMW you should check it somewhere between fuel filter and pressure regulator (some adaptors might be needed).
Operating pressure is about 3 bar.
After switching engine off (after 20-30 min) pressure should not drop more than 0.5 bar. If it does there are some possibilities:
-pressure regulator
-small leak (hoses, etc)
-non return valve in fuel pump.

BTW - when engine is warm (hot) and fuel pressure drops to much ( when engine off) fuel warms up and with lower pressure starts to vaporize going to gas state instead of liquid. When you try to start engine immediately after you stop it it will work fine because fuel did not vaporize yet. After some time - usually 20 min you have gas insted of liquid inside fuel lines and here the problems begin.
After siting few hours even if the pressure goes noticably down fuel goes back to liquid state (because cools down) and car starts ussualy fine- not perfect but fine.
All depends how fast pressure drops down and what is temperture in the engine bay.
Hope this will help a bit.
Tom