Hard start and rought idle/power loss

[t_marat]

Since last November I started having problem with starting the engine in the mornings, especially when it was cold. It would take a lot more cranking (3-5secs) and run rough and stall. What I did was to press the gas and hold at about 1000-1200rpm for about 40-60seconds, then it would idle normally, without roughness. Back then I changed the spark plugs, they were oily. Changed the valvecover gasket. But still had this problem.
Now the weather isn't cold. In the mornings sometimes it starts right up, about 2 secs of cranking. Sometimes I crank for 3-5 seconds, it won't start and I let go. Interesting is that right after that it starts immediately, have to crank for maybe 1.5seconds. This wasn't causing me much problems and I just ignored it. But today engine briefly ran rough and lost power while fully warmed up. This lasted for something like 1-2 minutes, then back again normal. My brother took the car after that and says that car actually stalled when he switched to neutral.

What could this be? During winter holding the rev at 1200rpm for a brief period made the engine ran normally. Because of that I tend to think it is something close to the combustion chamber, as in 1 minute area near it gets warmed up. I already changed spark plugs (Denso, by catalogue), so could this be injectors or ignition coils?

I read postings about fuel pumps. Could this be a fuel pump?

[mamilapon]

Pull the valve out and give it a good clean with carby cleaner.Also use a digital meter to check if youre getting 12v.

[Morgenster]

When you started in cold mornings and managed to without applying throttle, did the idle jump to 1600 and then drop to 1000?
If it just dropped lower than that there's a sensor out of whack. Could be anything from TPS to MAF, etc. I'd suggest checking for vacuum leaks or sensors out of tolerance.

[t_marat]

Mamilapon
By valve you mean ICV? Where from to measure the voltage, at the battery? The battery is OK.

Morgenster,
In winter engine would never start and idle by itself, I always had to keep the gas pedal pressed a little. Otherwise it would ran really rough and stall.
If it was TPS, then I shouldn't have rough running and power loss when the car in gear and going forward? As far as I know TPS is there to tell the ECU to engage the ICV, and has no effect when I am pressing the gas.
Doubt its a MAP, but could be. My previous was dead, changed a more than a year ago.

I tried spraying around carb cleaner. Couldn't see any difference in rpm.

[Morgenster]

What happens after a full DME reset (battery disconnect for 30 min)? Does the idle bounce or move at all when running warm? I'd still try multimeter tests on those sensors including temp sensors on the engine. If those all check out I'd work my way to more mechanical issues like the icv, fuel pump, FPR or filter. BTW: were the plugs oily on the firing side or on top? If they're oily on the firing side things are looking slightly worse.

[t_marat]

The plugs were fully covered in oil. I think it wasn't the firing side, its just the oil accumulated in the spark plug socket, while removing the plug it just covered it all. Today I checked the spark plug sockets, and they are pretty clean (changed valvecover gaskets couple of months ago).
Checked TPS resistance values. They were 1.3kohms to about 4kohms. Bentley says it should be in 1-4 kohms range. So I disconnected TPS sensor and drove like that around for about 30 mins (its 90F/32C in here :) ) Engine did not stall. But if you are say in 3rd gear, release the gas pedal and press it again, then there is some wobbliness in acceleration for a short while.
Went to a local shop to buy a new one and was quoted 150$ for a new one. Very expensive for a simple variable resistor. Anyway of cleaning it?

edit: can a TPS be blamed for rough running (not just idling) and loss of power?

[Morgenster]

Mine has a range of 1,8k to 5k and it works so I wouldn't change it just yet. And you can get them less than half that price anywhere else. It could still be bad and cause rough running if the voltage differential isn't continuous from idle to WOT. Changing engine temperatures can maybe obscure faults with electricals because they do not necessarily show up when cold (or the other way around). Did the reset change anything? Because when I did it some two or three times after changing the valve cover gasket and plugs it ran great all the time.

[t_marat]

I did the reset yesterday, left the battery disconnected for the night. Cleaned the crankcase ventilation connector with a carb cleaner, as it was very dusty (leaking slightly).
So far I did not have any problems, car did not stall and no rough running/power loss. But I guess its too early to say for sure problems are gone.

[Morgenster]

You might get lucky or not. But if the problem gradually returns it's most likely a DME sensor out of tolerance. Doing the reset right after plug changes or any engine work can help the engine adjust to different running conditions, so if you didn't do this when you changed the gasket and plugs it could stay good.

[t_marat]

Problem is back, nearly stalled this evening (even though it drove wonderfully during the day). Again the same thing, try starting the engine, spin for about 4 seconds, give up. Right after that it starts immediately.
I don't think engine is stalling due to ICV. Last year I had the same problem with engine stalling, cleaned and changed ICV to no help. Changed MAF and problem was gone.
Last year I also had a fuel pump failure. It might be unrelated but it failed at exactly the same time as engine stalled. I switched to neutral and was taking a left turn.

[Morgenster]

Problem is back, nearly stalled this evening (even though it drove wonderfully during the day). Again the same thing, try starting the engine, spin for about 4 seconds, give up. Right after that it starts immediately.
I don't think engine is stalling due to ICV. Last year I had the same problem with engine stalling, cleaned and changed ICV to no help. Changed MAF and problem was gone.
Last year I also had a fuel pump failure. It might be unrelated but it failed at exactly the same time as engine stalled. I switched to neutral and was taking a left turn.

Well, what you can take home from this is that there's probably no mechanical problem. Injectors and coils should be fine. However, when the fuelpump failed was it replaced?
Likely culprits are: MAF, O2 sensor, coolant temp sensor, air intake temp sensor, TPS or crank position sensor on the way out (unlikely though). If the problem occurs only when starting the car 'cold' and gets better with engine coming up to temp the first two sensors shouldn't be the issue. If it's the inverse I'd start looking at the O2 sensor. This is where it gets hard to diagnose.
In the end, when mine started acting up the problem was only there when starting cold and it got slightly better when warmed up. The engine would would drop idle below 400 rpm when I let go of throttle and then correct itself to 600 when cold which is still too low but sufficient. Sometimes this would make the engine stall. I cleaned the intake parts, reattached them firmly, and last but not least: I changed the position of the screw blocking the throttle plate (leftside on throttle body). It changed resistance on the TPS from 1.8 to 1.7 to 1.65 KOhms (when closed) and after that reset things have been fine ever since. This makes me believe that tolerance for the TPS is around those values. If you're over that value a new TPS is better. In my case I think it was tampered with to keep LPG flowing in the manifold at idle because the mixer was set up in the wrong position.

[t_marat]

I am having problem starting the engine, be it cold or warm. In winter it would idle very rough and stall; I had to keep the revs at 1200 for 40-50 seconds and then it would go fine. Even when warmed up.
Now when its summer and very hot, it sometimes starts fine, sometimes idles rough, but doesn't stall right after starting. But now it stalls when its fully warmed up. If I just press the gas (without moving the car), rev the engine to 3000-4000 rpm and let go, it won't stall. It stalls, or comes very close to stalling when I switch to neutral. Say rev is 3000rpm in 3rd gear, and I switch to neutral.
Today after parking the car I disconnected the MAF. As far as I know engine should come close to stalling then recover. In my case exactly the opposite happens. When I disconnect, no reaction. But when I reconnect it, engine goes to near stall, and recovers. I did this several times, every time when I reconnect the MAF, it nearly stalls and recovers.
Last year I had exactly the same problem with engine stalling when warmed up. Changing the MAF solved it back then.

[shogun]

Pelican parts has a tech section for all BMW, maybe you check these articles, some might be useful for you such as erratic idle
http://www.pelicanparts.com/bmw/techarticles/tech_main.htm

[J.DeFeo]

My car does the exact same thing, I suspect a leaking fuel injector. The fuel injector leaks a bunch of fuel into one of the cylinders, causing it to flood. Since your problem is solved by holding the RPMs at 2k or so for a few seconds this seems pretty likely. The flooded cylinder burns enough fuel to start firing normally.

[t_marat]

My car does the exact same thing, I suspect a leaking fuel injector. The fuel injector leaks a bunch of fuel into one of the cylinders, causing it to flood. Since your problem is solved by holding the RPMs at 2k or so for a few seconds this seems pretty likely. The flooded cylinder burns enough fuel to start firing normally.
does your car stall, or come close to stalling when warmed up? or run rough with power loss.
when engine starts running rough and looses power, switching ignition off and back on brings it back to normal. so does this make it definitely electrical? I guess it rules out TPS sensor, right?

[Morgenster]

Changing the MAF solved it back then.

New or used?
I still think it's electrical and one of your sensors is out of tolerance.

[J.DeFeo]

does your car stall, or come close to stalling when warmed up? or run rough with power loss.
when engine starts running rough and looses power, switching ignition off and back on brings it back to normal. so does this make it definitely electrical? I guess it rules out TPS sensor, right?

I just realized we have different engines, but here's how it goes for me:

Cold Start: Engine cranks a couple times, engine fires up. One piston is clearly not firing, causing extreme vibrations. If the gas pedal isn't held, the engine will stall within seconds. After maybe 7-10 seconds of holding at 2500 RPMs, the engine smooths out and idles normally. Occationally, the idle is at 1000 RPMs (instead of 850 like it should be) and increases as the car gets warmer. The idle usually tops out at about 1500 RPMs. Flooring it in 1st gear, getting the RPMs above 5000 for a few seconds solves this and the car then idles normally for the rest of the trip.

Warm Start: Engine starts right up, no problems. Sometimes has the high idle problem.

[t_marat]

New or used?
I still think it's electrical and one of your sensors is out of tolerance.
It was a used MAF. The one that I removed from the engine had obvious signs it was openned before and no thin film (wire) inside.

[t_marat]

I just realized my car started to make a puffing sound out of the exhaust. The sound that usually very old cars do. I don't know how to call it in English, may be backfire?
It only does it if I am braking with engine. It never does that in idle, acceleration or steady pace, only if I release the gas pedal completely while in gear. What could be the reason?

[shogun]

Not sure if that helps, but read the problem and how it was solved
http://www.bimmerboard.com/forums/posts/360519

[t_marat]

Not sure if that helps, but read the problem and how it was solved
http://www.bimmerboard.com/forums/posts/360519
I will try that tomorrow, disconnecting the O2 sensor.

Is it possible to clean the MAF by spraying carb cleaner on it? I know I should not touch it, too delicate.

Can you hear how air moves inside the air intake manifold? I mean a shshsh sound with hood open. I can hear it if I rev to 2000, higher than that engine sound becomes louder. no such sound while in idle. I looked where I could, sprayed carb cleaner on hoses and under the intake manifold, but it did not see any increase in rpm. I always had this sound and thought it is normal. I think it is the sound of air moving through intake manifold. am I right?

[shogun]

I have never cleaned my MAF, it is self cleaning.

When the air-mass sensor is working normal, the hot wire does not glow. To burn off any deposits and dirt, the control system will heat the wire red hot to 1000 degrees Celsius (1800 F) for about 1 second, after you shut off your engine. This is the cleaning process and only performed if the engine is revved up to at least 2000 rpm. So starting your engine, let it idle for a certain time and shut off means no hot wire cleaning.

That is the 750 MAF, not sure if you have the same system
http://bmwe32.masscom.net/sean750/MAF_v12/MAFtroubleshooting.htm

Here's a link you might find interesting,all though it's a Porsche 944 AFM, he's desribing it is a MOTRONIC set up from around the same era as your E34, so they should be very similar our BM's.


http://frwilk.com/944dme/afm.htm

[t_marat]

I have never cleaned my MAF, it is self cleaning.

When the air-mass sensor is working normal, the hot wire does not glow. To burn off any deposits and dirt, the control system will heat the wire red hot to 1000 degrees Celsius (1800 F) for about 1 second, after you shut off your engine. This is the cleaning process and only performed if the engine is revved up to at least 2000 rpm. So starting your engine, let it idle for a certain time and shut off means no hot wire cleaning.

That is the 750 MAF, not sure if you have the same system
http://bmwe32.masscom.net/sean750/MAF_v12/MAFtroubleshooting.htm

Here's a link you might find interesting,all though it's a Porsche 944 AFM, he's desribing it is a MOTRONIC set up from around the same era as your E34, so they should be very similar our BM's.


http://frwilk.com/944dme/afm.htm

I have read in Bentley that it is supposed to clear itself. its just I remember reading about some MAF cleaner on this board. MAF on my engine is principally the same as in that link, but it has only 4 pins.

I noticed that whenever car starts running rough the exhaust gets more black and clears out as soon as engine starts running normally. smoke isn't very black, but has a tint of blackness. one cylinder not running for some reason?

Other thing. after some hard accelerations engine started making a ticking/knocking sound. sound was coming from the valve cover/intake manifold area. got scared with a thought of piston hitting the valve, but the sound was not metallic, more like something plastic hitting plastic. plus car run normal, with no loss of power. it wasn't fan, tried spinning it with engine off, no obstruction. sound was rather loud, could hear even with hood closed. after a while ticking disappeared. could this be injectors? maybe from time to time one of the injectors starting to leak badly, fouling the spark plug?

[Morgenster]

When's the last time you used injector/valve cleaner in the tank?

[t_marat]

Last year. But after using the cleaner, I did not see any difference in how the engine ran.

[Morgenster]

Last year. But after using the cleaner, I did not see any difference in how the engine ran.

Maybe now you will?
I'm not too sure about your problem, but with the added info on the piston I'd say take a look at the injectors. I never get black smoke, ever. So one cilinder at least is not firing on startup and maybe cuts out once in a while.
However, that won't explain why your engine runs fine right after a reset and starts coughing after a while. How long did the effect of the reset last?

[t_marat]

Maybe for a day I think.
Coughing is also strange. Sometimes there is no coughing, sometimes it starts coughing badly (but again only when I am braking with engine).
Somebody here told me it could be a crankshaft sensor. Can it go bad from time to time, especially when its hot? If it isn't functioning properly then I think fuel pump won't pump as it should and ignitiion will also be wrong (pinging). It kind of fits perfectly.
But it doesn't explain the knocking noise.

[Morgenster]

Maybe for a day I think.
Coughing is also strange. Sometimes there is no coughing, sometimes it starts coughing badly (but again only when I am braking with engine).


Are things getting worse?
Engine braking should cut off fuel and spark so the engine shouldn't cough when you do that!

[shrike071]

It could also be the DME itself going out. Electronics do all sorts of odd things when they begin to fail...

[t_marat]

how strange it is, but it coughs only while braking with engine. meaning releasing the gas pedal while in motion in say 3rd gear. by coughing I mean the puffing sound that comes out of exhaust (?).

[t_marat]

it is becoming worse. Mostly starting the engine takes 3-4 attempts; it just won't fire up; or starts up, but rpm is close to 0, pressing the gas doesn't make any difference. and more strange is that sometimes it starts right at the first try.. Also engine pings a lot more than it used to. But it could be related to temperature, it is really hot (around 100F, 35-38C)
I am inclined to change the crankshaft sensor. could it be the culprit? does it do things like this?

[Morgenster]

Wouldn't hurt changing it. Did you try a second reset?

[t_marat]

I used an injector cleaner called Gunk Carbuerator and Injector Cleaner. It definitely made a difference. For the past 5 days I didn't have trouble starting the engine, it would start immediately. Though, I can't say it is gone totally, yesterday I had to do a second try as it wouldn't start on first attempt. But that was the only hard start in the past five days.
However, problem with stumbles on acceleration is still there. Today tried resetting the ECU. Till afternoon it was fine, but then it stumbled again. And I don't really think resets have any effect, because problem seems to appear randomly. For example on the evening car drove without problem, even though during afternoon it had problems.
One thing, when engine starts stumbling, flooring the pedal helps. It starts accelerating normally. And as I wrote before during hiccups like that engine idles rough, and exhaust turns into black (not very black). Problem seems to appear mostly when engine hasn't fully warmed up. Mostly, not always.
How can I disconnect the oxygen sensor? Bentley says the socket is on the left side of the engine, under the starter. Looked into that area but there are some many things in there, couldn't find the connector.

[Morgenster]

I used an injector cleaner called Gunk Carbuerator and Injector Cleaner. It definitely made a difference. For the past 5 days I didn't have trouble starting the engine, it would start immediately. Though, I can't say it is gone totally, yesterday I had to do a second try as it wouldn't start on first attempt. But that was the only hard start in the past five days.
However, problem with stumbles on acceleration is still there. Today tried resetting the ECU. Till afternoon it was fine, but then it stumbled again. And I don't really think resets have any effect, because problem seems to appear randomly. For example on the evening car drove without problem, even though during afternoon it had problems.
One thing, when engine starts stumbling, flooring the pedal helps. It starts accelerating normally. And as I wrote before during hiccups like that engine idles rough, and exhaust turns into black (not very black). Problem seems to appear mostly when engine hasn't fully warmed up. Mostly, not always.
How can I disconnect the oxygen sensor? Bentley says the socket is on the left side of the engine, under the starter. Looked into that area but there are some many things in there, couldn't find the connector.

O2 Very hard to get to unless you're especially limber. Black exhaust pretty much tells you your car is literally choking. So that means either intermittently missing spark or FUBARed air/fuel ratio. Not a mechanical problem IMHO. Intermittent (both cold and warm) points to electrical problems. I still think O2 sensor or just maybe intermittently failing MAF or TPS. A diagnostic trick in case you don't have good known replacement parts and you think failure is intermittent: hook up a multimeter to the wiring on these sensors and check what happens on the multimeter with the engine running until it happens. If you're really handy you can hook it up so you can keep it in the car and drive.

EDIT: almost forgot: when you tested the TPS, did you test resistance range in motion from fully closed to open and back? Also, testing the sensors might not be enough, you might have to test there respective connectors at DME level to check for intermittent circuit breaks.

[t_marat]

O2 Very hard to get to unless you're especially limber. Black exhaust pretty much tells you your car is literally choking. So that means either intermittently missing spark or FUBARed air/fuel ratio. Not a mechanical problem IMHO. Intermittent (both cold and warm) points to electrical problems. I still think O2 sensor or just maybe intermittently failing MAF or TPS. A diagnostic trick in case you don't have good known replacement parts and you think failure is intermittent: hook up a multimeter to the wiring on these sensors and check what happens on the multimeter with the engine running until it happens. If you're really handy you can hook it up so you can keep it in the car and drive.

EDIT: almost forgot: when you tested the TPS, did you test resistance range in motion from fully closed to open and back? Also, testing the sensors might not be enough, you might have to test there respective connectors at DME level to check for intermittent circuit breaks.
Why didn't I just disconnect the MAF when the engine was running rough, should have checked that long ago.
I did test the TPS in motion with digital multimeter. Seems to be normal, linear.
I don't think its circuit breaks. If that would be the case, flooring the pedal should be of no use, correct? or not?
tomorrow going to have the intake manifold taken out and look for vacum leaks. along the way will try running without o2 sensor.

[t_marat]

Had the intake manifold removed and inspected for vacum leaks, none to speak of. intake gaskets had been replaced, hoses also inspected, and it turns out I have no air leaks. a really comforting thing.
tried disconnecting the MAF, but engine keeps stalling.
After that disconnected the O2 sensor. Got disappointed as engine was idling a little rough. Went for a drive, first thing I noticed there was no pinging. Lately engine had been pinging a lot. Drove around for about an hour, had no problems! interesting is that pinging dissapeared as soon as the O2 sensor was disconnected. Could the pinging be also because of the O2 sensor?
Should have checked the O2 sooner, especially it was advised as one of the possible reasons.
Thanks everybody for help.

[Morgenster]

Well I guess it could: fake signals from o2 could maybe have the DME make the mixture leaner and heating he cilinders somewhat more than average. Let's hope the o2 fixes the problem.

[Podmore]

Sounds like a burnt or leaking / sticking exhaust valve to me

[Morgenster]

Wouldn't a burnt valve give consistent poor performance?

[shurton]

I had similar symptoms with my other car as far as the starting goes.
I would crank it for 5 seconds and it wouldn't start. Then on the second try it would start right away. It turned out that the fuel pump wasn't holding the pressure in the fuel line. Replaced the fuel pump and fuel filter and problem was solved.