Uneven acceleration and fuel consumption gauge swinging on my e36 320i M52 (oct.'95)

Hello all!

I need your help. I have a problem. My e36 320i M52B20 (oct.'95) (275000 kilometers) accelerates unevenly most of the time (about 95%) and during this the fuel consumption gauge sways - rocks up and down evenly about once every second or less for a few liters (metric unit - about two pints) in each direction.
(I am not from an English speaking country so I hope "fuel consumption gauge" is the correct phrase - the thing that shows fuel consumption under the RPM meter).

Of course, the gas pedal is kept perfectly still during this, so the problem isn't due to my leg shaking or anything like that

The previous owner of the car and I studied the problem quite extensively. The phenomenon is not due to any external factor, e.g. wind, road. It happens in all situations. It does not matter whether the car is cold or at the working temperature (although when I have been driving it for 1 hour or so, the acceleration is higher and the fuel consumption much lower, but that is probably normal and due to engine materials being evenly hot and properly dilated)

1. It happens when the gas pedal is depressed from about 1/4 to 3/4 of the way. At low acceleration and at full throttle the acceleration is even, however I think it is not high enough for this engine.
2. The biggest differences in acceleration are seen between 2000 and 4000 RPMs. Above that it is much less pronounced.

Follow this link to see the video, made by the previous owner of the car around 2 years ago:

During this experiment, he held his foot perfectly still on the accelerator. The rocking of the fuel consumption gauge can bee seen above 60 kilometers per hour (above 2000 RPMs, driving in the third gear):
http://tinypic.com/player.php?v=6oh28l&s=5


3. While driving at a steady speed and same gas pedal position (again on a straight road and not on a hill or anything), the car very finely but distinctively accelerates and decelerates, again the consumption gauge mirroring the effect.

This may be important, however it is a recent issue and was not present for almost two years since the problem was first seen: A couple of times when I started the car, one cylinder, I think, didn't work, so the problem could probably be a spark plug, spark plug cable or the top of the thing that divides the current among the spark plugs, right?

This also might be important: Sometimes when I press the pedal, the car doesn't immediately start, but waits and then slowly starts to accelerate.

I don't know if it is related or if it is normal for this engine, but below 3000 RPMs the acceleration is in my opinion very low. Above this barrier, the car is a rocket compared to below it.

I have to ask the previous owner, if he replaced something and after this he noticed the problem or if it couldn't have been related to anything.

- I replaced the rubber hose that connects the airflow meter with whatever is behind it.
- The car didn't show any problems when my mechanic hooked it up on his tester. The spark plugs looked normal and the cylinders didn't show any unusual deviation from normal data.
- The airflow meter was replaced by the previous owner.

My mechanic thinks that the problem could be that somewhere there is air going in through some crack. That would cause the vacuum to drop and would cause the problem in his opinion.

Any comments and ideas will be greatly appreciated! Thank you in advance!

Bye

Hi, did you ever figure out the problem? I realise it was a long time ago! I'm having the same issue. I can't find any other reference to it on google...

Originally Posted by slo_e36

...
I don't know if it is related or if it is normal for this engine, but below 3000 RPMs the acceleration is in my opinion very low. Above this barrier, the car is a rocket compared to below it.

...

This comment makes me think it has something to do with the VANOS. The intake cam is advanced with oil pressure from ~1000-3500rpm for more torque and then retarded above that rpm. If the VANOS piston seal is worn, the cam may not be able to advance with poor performance as an effect. At high rpm, it's back to normal. Here is the raather complicated but inexpensive repair procedure. Beisan has sold many kits in the US to fix this issue.
http://www.beisansystems.com/procedu..._procedure.htm

Hi. Yes, actually I may have figured it out. The problem was either in one (or both) of the lambda probes (I hope this has the same name in English as it does in my language. I cant find the thing on BMWfans to show you the picture of what I mean) or something downstream. I disconnected both probes and the problem was gone. The car accelerated wonderfully and contrary to my expectations, the fuel consumption was actually a bit lower, which in itself indicates a faulty lambda probe. I stopped my investigation at this stage. I should have tried disconnecting only one lambda probe at the time and see if the problem persists, but since the problem was gone, I left it as it was. I hope this helps you.

Bye

Originally Posted by rossjp

Hi, did you ever figure out the problem? I realise it was a long time ago! I'm having the same issue. I can't find any other reference to it on google...

Thanks for your reply,

Lambda - it's the same word in english, yes, from the greek letter, though some call it the O2 (oxygen) sensor.

I'll look into this - I have a lead to connect my computer up and read codes, maybe it will give me more information.

With mine it is quite a small effect, on full throttle it's not noticable, only when you are on part-throttle or cruising on light throttle it just feels like you're lifting off then putting your foot back on, then off again. I don't have cruise control, I just mean holding my foot totally still on the accelerator.

On the MPG / fuel consumption meter it looks like it's cycling between 3 levels - It's like the ECU is saying "setting A isn't correct, try setting B" then "setting B isn't correct, try setting C", "setting C isn't correct, try setting A" and just looping through that.

I wouldn't say that below 3000 rpm mine is lacking in power, which is different to your symptoms.

I suspect either lambda (one or both), air meter (which I know wasn't the problem for you) or an air leak on the intake manifold.

Ross

Your English was fine, and mostly better than many native speakers here in the States...

Glad you got it sorted out. A faulty Lambda probe (we call them O2 sensors) should give a Check Engine Light, though.

Originally Posted by slo_e36

Hi. Yes, actually I may have figured it out. The problem was either in one (or both) of the lambda probes (I hope this has the same name in English as it does in my language. I cant find the thing on BMWfans to show you the picture of what I mean) or something downstream. I disconnected both probes and the problem was gone. The car accelerated wonderfully and contrary to my expectations, the fuel consumption was actually a bit lower, which in itself indicates a faulty lambda probe. I stopped my investigation at this stage. I should have tried disconnecting only one lambda probe at the time and see if the problem persists, but since the problem was gone, I left it as it was. I hope this helps you.

Bye

Hi there, sorry but it wasn't clear to me: did you change the O2 sensors or you left the car without them? I am having this problem since I got the car 2 years ago, but as it wasn't that bad and I couldn't find the cause, I stopped bothering. But now that I found your solution, I'm interested in solving it

Originally Posted by shiga

Hi there, sorry but it wasn't clear to me: did you change the O2 sensors or you left the car without them? I am having this problem since I got the car 2 years ago, but as it wasn't that bad and I couldn't find the cause, I stopped bothering. But now that I found your solution, I'm interested in solving it

Well, I just changed the O2 sensor and it's running great now! Thanks!