Hesitation/stutter giving gas off idle?

My dad hit me with a question about his IX the other day that I hadnt experienced ever. Apparently if the car it at idle, and his steps on the gas, the car stutters and almost stalls for a second or two. After that, and once the car is revved past 1000rpm the car reacts normal and quick to increases in gas. He said he has stalled the car a few times releasing the clutch pointed up a hill(where you don't usually have 2 seconds to get rolling forward). He also said it's a lot worse when the car is cold.

His idle is solid, his power past idle is solid and smooth. Thoughts?

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

Hmmm, in a carbureted engine I would say your accelerator pump shot is too weak but these are fuel injected engine so no accelerator pump. Possibly something in the main butterfly setup? Dirty carbon tracks? If it was vaccuum, the idle wouldn't be solid, it would hunt.

alan

A little more info. It was running perfectly 3 weeks ago after I did the following work:

New:
fuel filter, air filter, coolant tempt sensor, spark plugs, cap, rotor.

Spray cleaned the ICV, wiped down the intake manifold around the throttle door. Adjusted the TPS.

The car was silky smooth after that. And now this problem suddenly crops up.

Could the Throttle Position Sensor (TPS) have shifted position and be misreading? Could the car think it was "at idle" even though the pedal was depressed a bit?

I know that on my own car that I can tap the gas pedal repeatedly just a tad bit, real quick, and this causes the engine to stumble all over the place. I've always believed this to be normal, the ECU just can't keep up trying to switch out of and back into idle 5 times in 2 seconds. This is however not the problem that he is experiencing. My car jumps off idle well enough, and certainly doesn't suffer from power right of idle.

I wish I could provide a bit more, but since the car isn't sitting in my driveway trouble shooting is hard. I'll try and spend an hour or two with it in the next week. I'm just trying to get as many ducks in a row as I can. I haven't seen this issue before, and I can't surmise what it could be based on what i know of the M20.

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

How about the airflow meter? Could it be a bit sticky?

Quote
Michiel 318iS
How about the airflow meter? Could it be a bit sticky?

Could be. I can check that out as best I can with an ohm meter. But I'm thinking the airflow meter should be flapping around even at idle. It could have a dead spot in the reading though, which is what I can check for.

The ICV is opening correctly, otherwise the idle would be bad. But what would happen if the ICV stayed open? Could this cause an issue with a slightly cracked throttle? I dunno...

It seems like it should be a sensor thing not a mechanical thing. I can't think of a mechanical part that would effect the idle-->open throttle transition that wouldn't also effect idle or an open throttle --> more open throttle.

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

Quote
Earendil
A little more info. It was running perfectly 3 weeks ago after I did the following work:

New:
fuel filter, air filter, coolant tempt sensor, spark plugs, cap, rotor.

Spray cleaned the ICV, wiped down the intake manifold around the throttle door. Adjusted the TPS.

The car was silky smooth after that. And now this problem suddenly crops up.

Could the Throttle Position Sensor (TPS) have shifted position and be misreading? Could the car think it was "at idle" even though the pedal was depressed a bit?

I know that on my own car that I can tap the gas pedal repeatedly just a tad bit, real quick, and this causes the engine to stumble all over the place. I've always believed this to be normal, the ECU just can't keep up trying to switch out of and back into idle 5 times in 2 seconds. This is however not the problem that he is experiencing. My car jumps off idle well enough, and certainly doesn't suffer from power right of idle.

I wish I could provide a bit more, but since the car isn't sitting in my driveway trouble shooting is hard. I'll try and spend an hour or two with it in the next week. I'm just trying to get as many ducks in a row as I can. I haven't seen this issue before, and I can't surmise what it could be based on what i know of the M20.

What made you adjust the TPS?

Quote
rkj

What made you adjust the TPS?

It wasn't reading the throttle position?
The TPS was failing to read the closed (idle) position. This was tested using a multimeter. I adjusted it such that it would read the closed position and WOT, and tested this using the gas pedal not the readily available throttle "handle"(or whatever you want to call it), so it was correct at the time I set it. And as I said, it ran just peachy for 3 weeks.
But of all the work I did, the TPS is the one thing that could "undo" itself if I didn't tighten the screws enough.

Does anyone know if the stomp test needs to read both closed and open 5 times, or is it just open? or just closed?
If it were open and closed 5 times that would be an awesome way to tell if the TPS was adjusted fairly well.

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

Ok, it looks like your culprit is the TPS.
These things get really dirty after all the years they have been in use. The best solution is to open up the TPS and clean it properly; it's probably plugged with grease, oil and tens of thousands of miles of dirt

The quick and dirty solution is to immerse the unit in solvent of some sort, over night, then let it drain out. This is a flaky fix and doesn't always work.
Opening the TPS involves a little work with a file or Dremel tool but it is the best way to do the job right.

Inside the TPS there are two switches; the 'closed throttle' micro switch and the WOT or wide open throttle switch. Cleaning these things up will go a long way towards fixing bad throttle responses where transitions from idle to GO are involved

Do a Youtube and/or Google search for BMW Throttle Position Sensor
Just found this:

[www.youtube.com]

alan



Edited 1 time(s). Last edit at 02/07/2011 08:09PM by alanrw.

+1 for TPS being the most likely culprit. If you are really unlucky, the TPS being previously incorrect was masking a problem with the ECU (aka DME) and now that it's "fixed" it's actually worse. A while back I threw in another stock ECU (just to test it worked) and it was amazing how better the off-idle throttle response was than my own (otherwise fine) ECU.

The signs also make me suspect coolant temp sensor. I realise a new part, but it's still worth checking that it's within spec IMHO.

Quote
nomis3613
+1 for TPS being the most likely culprit. If you are really unlucky, the TPS being previously incorrect was masking a problem with the ECU (aka DME) and now that it's "fixed" it's actually worse. A while back I threw in another stock ECU (just to test it worked) and it was amazing how better the off-idle throttle response was than my own (otherwise fine) ECU.

The signs also make me suspect coolant temp sensor. I realise a new part, but it's still worth checking that it's within spec IMHO.

Just to be clear, the ecu is the brain box above the glove box and the dme is the air fuel meter located on the motor. right?

I'd like to throw my twocents in here on the tps, and I know Peter is going to disagree but I always replace the dam thing just because it's a little thing that wears out and leaks oil into the body and is a constant thorn in the side of running right; it's in a prime place to get oil all the time anyway.

The tps is such a crucial part of the injection I'd much rather bite the bullet and be done with it. Once it's adjusted properly you can pretty much forget about it for good, while you're in there, there are a few adjustments to check; it's all pretty easy and laid out (as I remember) in the bentley. If you have a feeler gauge you should be able, if you're careful, to have things all ship shape in no time at all.

There, I've said it...... Peter

ECU and DME are the same thing. It is the computer above the glove box

The difference between cleaning up the old TPS and buying a new one is simply economics of time and money...I'm all about fixing things rather than replacing them if that is possible

Other than physically cleaning the wiper and resistor track, are there any electrical readings one looks for at the plug connector? Or do you simply clean it and the ECU figures out the rest?

alan

Quote
alanrw
Other than physically cleaning the wiper and resistor track, are there any electrical readings one looks for at the plug connector? Or do you simply clean it and the ECU figures out the rest?

alan

To properly measure the AFM you need to power it up and I don't remember the procedure, sorry.

anybody?

Quote
alanrw
Other than physically cleaning the wiper and resistor track, are there any electrical readings one looks for at the plug connector? Or do you simply clean it and the ECU figures out the rest?

alan

Yep! The hardest part of testing is holding the multimeter probes against the pins
But it's a very simple On/Off type of test.

You have ground and two other pins. One pin will show absolute resistance to ground until it reads idle, and the other pin will show absolute resistance until it reads WOT.
The TPS on these cars only reads 2 positions (and by elimination, it infers a 3rd). It isn't like the air flow meter that has increasing resistance as the gate opens further.

I'll check the TPS again when I get my hands on it. Shouldn't take too long to at least test. More time if I need to adjust it, and considerably more if adjusting doesn't do the trick and I need to clean it

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

Quote
alanrw
Other than physically cleaning the wiper and resistor track, are there any electrical readings one looks for at the plug connector? Or do you simply clean it and the ECU figures out the rest?

alan

I see now that both Rick and I assumed you meant different things. And I'm not sure which you are talking about, but feel free to pick and choose the most appropriate response

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

ECU,DME - is the electronic engine control unit or whatever you call the computer that controls the ignition and injector timing.

TPS - Throttle Position Sensor mounted at the bottom end of the throttle plate shaft is not so much a sensor as simply two switches. With throttle closed one switch closes to tell the ECU that the Idle Control Computer should take over. With throttle wide open the other switch closes telling the ECU to run its Wide-Open map. Anywhere in between, both switches are open and the ECU relies on other sensors to tell it what's going on.

AFM - Air Flow Meter is the flapper valve between the air filter and throttle plate. As the engine pulls in more or less air, the spring-loaded flap in the AFM swings a wiper arm along a potentiometer track. The resistance variation is NOT linear, so you cannot test it just by measuring resistance while swinging the flap.

Here is an article explaining how to test the AFM (copied from elsewhere on the web).

Quote
Ferdinand
ECU,DME - is the electronic engine control unit or whatever you call the computer that controls the ignition and injector timing.

TPS - Throttle Position Sensor mounted at the bottom end of the throttle plate shaft is not so much a sensor as simply two switches. With throttle closed one switch closes to tell the ECU that the Idle Control Computer should take over. With throttle wide open the other switch closes telling the ECU to run its Wide-Open map. Anywhere in between, both switches are open and the ECU relies on other sensors to tell it what's going on.

AFM - Air Flow Meter is the flapper valve between the air filter and throttle plate. As the engine pulls in more or less air, the spring-loaded flap in the AFM swings a wiper arm along a potentiometer track. The resistance variation is NOT linear, so you cannot test it just by measuring resistance while swinging the flap.

Here is an article explaining how to test the AFM (copied from elsewhere on the web).

What would we do without you and Peter....... things would go a lot slower, I know that!

Quote
Ferdinand
ECU,DME - is the electronic engine control unit or whatever you call the computer that controls the ignition and injector timing.

TPS - Throttle Position Sensor mounted at the bottom end of the throttle plate shaft is not so much a sensor as simply two switches. With throttle closed one switch closes to tell the ECU that the Idle Control Computer should take over. With throttle wide open the other switch closes telling the ECU to run its Wide-Open map. Anywhere in between, both switches are open and the ECU relies on other sensors to tell it what's going on.

AFM - Air Flow Meter is the flapper valve between the air filter and throttle plate. As the engine pulls in more or less air, the spring-loaded flap in the AFM swings a wiper arm along a potentiometer track. The resistance variation is NOT linear, so you cannot test it just by measuring resistance while swinging the flap.

Right. You can not prove that it is in proper working order by just measuring the resistance while swing the flap. However you can prove it's not in working order by swing the flap. A few years ago I replaced mine because there was a dead spot half way through the door swing. It may not be linear, but one should be able to graph the output without using right angles ;-)

Naturally a dead spot is not the only way it can malfunction, so as you said, not finding a dead spot does not prove that it is functioning within spec.

Quote

Here is an article explaining how to test the AFM (copied from elsewhere on the web).

Awesome site. Nice example of a graph without right angles ;-)

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles

Quote
Earendil

Quote
Ferdinand
ECU,DME - is the electronic engine control unit or whatever you call the computer that controls the ignition and injector timing.

TPS - Throttle Position Sensor mounted at the bottom end of the throttle plate shaft is not so much a sensor as simply two switches. With throttle closed one switch closes to tell the ECU that the Idle Control Computer should take over. With throttle wide open the other switch closes telling the ECU to run its Wide-Open map. Anywhere in between, both switches are open and the ECU relies on other sensors to tell it what's going on.

AFM - Air Flow Meter is the flapper valve between the air filter and throttle plate. As the engine pulls in more or less air, the spring-loaded flap in the AFM swings a wiper arm along a potentiometer track. The resistance variation is NOT linear, so you cannot test it just by measuring resistance while swinging the flap.

Right. You can not prove that it is in proper working order by just measuring the resistance while swing the flap. However you can prove it's not in working order by swing the flap. A few years ago I replaced mine because there was a dead spot half way through the door swing. It may not be linear, but one should be able to graph the output without using right angles ;-)

Naturally a dead spot is not the only way it can malfunction, so as you said, not finding a dead spot does not prove that it is functioning within spec.

Quote

Here is an article explaining how to test the AFM (copied from elsewhere on the web).

Awesome site. Nice example of a graph without right angles ;-)

I think there Are tests you can do to the AFM but I'm pretty sure for Any electrical test the thing has to be powered up, anything else is a waste of time.

It's been my experience just lifting that black cover you can see if the track is wiped out, no electrical test needed really.

Rick



Edited 1 time(s). Last edit at 02/08/2011 03:29PM by rkj.

The article Ferd linked does an excellent job of explaining why the resistance test is not good enough and how to do a 'live' voltage test that is accurate.

Is there a lot of variation of that part thru the model years? Are they calibrated differently? I bet they have a potload of them at the Pick A Part.

alan

The AFMs are specific to the engine models they go on. Other than that, I would thing that one from an M20B25 with Motronic 1.3 would be the same through all years. There would be differences for the M20B25 Motronic 1.1 or the M20B27 Motronic 1.0 though.

If you just matched like engine and Motronic with the same, there should be lots of them at the wreckers.
Fortunately, since they are passive devices, the only thing that gets worn or out of spec, is the carbon track. That can be taken care of by adjusting the wiper to run on an unscored part of the track

This is all about the AFM...the TPS should be the same for most models.

Quote
alanrw
Is there a lot of variation of that part thru the model years? Are they calibrated differently? I bet they have a potload of them at the Pick A Part.

That article actually describes the testing of an Air Flow Meter from a Porsche 944.

When I opened up mine from my '86 325, it looked pretty much the same.

Quote
Archeo-peteriX
The article Ferd linked does an excellent job of explaining why the resistance test is not good enough and how to do a 'live' voltage test that is accurate.

I can't find an explanation of why anywhere in the article, only a statement that a resistance test "will jump about meaninglessly".
This would defy the experience I've had with the M20 AFM. I have certainly placed a multimeter across the pins and seen a smooth and steadily increasing in resistance as the flap opened.

Could someone explain this to me? The flap is just a potentiometer (something that causes a change in resistance). The Current(I) is constant, and the Voltage (V) will vary will the resistance (R).

Given V=I*R.
Why would it matter which you measured, V or R, as long as I is constant?

Also, afaik a multimeter measures resistance the same way you would in this test, just using a lower voltage and possibly lower current. The principles are the same. So I'm not sure why it would bounce around in one scenerio and not the other?

So, does anyone have an explanation for this?


(Yes, I'm allowed to sidetrack my own discussion, especially for educational purposes )

-----------------------------------------------------------------------------------
1989 - E30 - M20 - Manual. Approximately 270,000 miles
2000 - E46 - M52TUB28 - Manual. Approximately 110,000 miles




Edited 1 time(s). Last edit at 02/09/2011 09:29AM by Earendil.

Quote
Earendil

Quote
Archeo-peteriX
The article Ferd linked does an excellent job of explaining why the resistance test is not good enough and how to do a 'live' voltage test that is accurate.

I can't find an explanation of why anywhere in the article, only a statement that a resistance test "will jump about meaninglessly".
This would defy the experience I've had with the M20 AFM. I have certainly placed a multimeter across the pins and seen a smooth and steadily increasing in resistance as the flap opened.

Could someone explain this to me? The flap is just a potentiometer (something that causes a change in resistance). The Current(I) is constant, and the Voltage (V) will vary will the resistance (R).

Given V=I*R.
Why would it matter which you measured, V or R, as long as I is constant?

Also, afaik a multimeter measures resistance the same way you would in this test, just using a lower voltage and possibly lower current. The principles are the same. So I'm not sure why it would bounce around in one scenerio and not the other?

So, does anyone have an explanation for this?


(Yes, I'm allowed to sidetrack my own discussion, especially for educational purposes )

Technically speaking, you are correct but real world experience seems to dictate otherwise. I'm guessing it's because most of our AFM carbon tracks are no longer smooth and give nice clean resistance readings. If the voltage test is done, it will likely hop around a bit too but be a little more meaningful as the range of volts vs the range of ohms will be easier to gage.

If yours is reading nice and clean on the resistance check, then look no further; close it up and don't touch it; you have a golden one