SOLVED: 2003 X5, 3.0l, runs rough, reset on restart?

[upallnight]

It would help if the poster would post all pertinent info when they post their problems. Nowhere in his first post did he posted that he had an SES light come on. That is why in my reply to his first post I asked him if he had a working SES or CEL light. That is when he replied that the SES light came on and he scanned it and came up with a misfire code. The more info that a poster can provide the more accurate a guess on our part. After all, we are not there to see, listen and perform rudimentary tests so any advice on our part is just an educated WAG (Wild Ass Guess).

[cmelkoni]

My apologies. I thought I had posted all pertinent information and had just assumed that describing the OBD code for cylnder #2 misfire implied that my SES light had come one. That was my fault.

Here's the latest update: I was so sure that this was a failing coil pack that I replaced the coil on Cylinder #2 on Friday. I then took a weekend trip and didn't even get out of the gas station before it happened again. Same thing as before, always when stopped and idling, never while driving. In this case, it happened twice, once while idling at gas station, second while idling at friends house after starting it to warm up. As before, idle drops slightly, can hear engine running different, SES light comes on. Step on accelerator and car is a shuddering mess with barely any power. Shut off ignition, turn back on, and car is back to normal.

A little more codes thrown this time though. Now I'm getting:
P0300: Random/multiple cylinder misfire detected
P0302: Cylinder 2 misfire detected
P0306: Cylinder 6 misfire detected

I also had a cylinder 5 misfire code before I erased them the first time. So, multiple cylinder misfires now instead of just cylinder 2. Really wish I had the other codes before I bought a new coil pack.

Now, I realize there are numerous issues that could cause random cylinder misfires: fuel delivery, vacuum leak, bad computer, etc. But, I'm hoping that these 2 symptoms can narrow down the list:

1. Only happens when idling and stopped. (Alternatively, it could be just at low RPM's for an extended period of time which is what idling is essentially)
2. Ignition off then on immediately solves the problem.

I would appreciate anymore insight someone could give.

Please let me know if you need more information and I will be happy to get/share.

[upallnight]

Would be nice if you could monitor live data with the engine running. If you could I would look at the short term and long term fuel trim in Bank 1 & 2. Multiple misfires could be due to a vacuum leak. A vacuum leak will show up in the fuel trim numbers. The short term will be + because it is adding fuel to the engine and the Long term will be - because it will be lean due to the vacuum leak. Vacuum is always highest when the throttle body is close when the engine is at idle and lowest when the throttle is open.

[cmelkoni]

uppallnight,

Thanks for the suggestion. My scanner does allow me to check live data. Bank 1 has positive Long Term and negative Short Term. Bank 2 has the opposite. Here are some numbers at idle and 2500 RPM:

Idle
Bank 1 LT: +5.5
Bank 1 ST: 0.0 to -3.0
Bank 2 LT: -0.8
Bank 2 ST: 0.0 to +3.0

2500 RPM
Bank 1 LT: +5.5
Bank 1 ST: 0.0 to -3.0
Bank 2 LT: -0.8
Bank 2 ST: +4.0 to +7.0

What it looks like is that upon going to 2500 RPM, both Short Term trims change, but Bank 1 quickly goes back to the same value as at idle while Bank 2 stays more positive.

From your post, it looks like this is pointing to a vacuum leak. Would appreciate your opinion/direction.

[cmelkoni]

Another interesting point is that the TPS sensor doesn't seem to be acting as I would expect it to.

TPS is at 0.0% when at idle. When not moving (in neutral), I would expect the TPS % to smoothly follow the accelerator pedal as I step on it. Instead, a change doesn't register until I push the pedal halfway down and the RPM's are pretty high. If I peg the pedal, I can briefly see the TPS % in the 90's, but it is not steady.

I took the car out for a test drive and watched the TPS%. It does change and I can see it occasionally in double digits, but I couldn't see any pattern.

This may be just how the TPS operates on the E53 but I thought it was worthwhile to note.

[upallnight]

Originally Posted by cmelkoni

uppallnight,

Thanks for the suggestion. My scanner does allow me to check live data. Bank 1 has positive Long Term and negative Short Term. Bank 2 has the opposite. Here are some numbers at idle and 2500 RPM:

Idle
Bank 1 LT: +5.5
Bank 1 ST: 0.0 to -3.0
Bank 2 LT: -0.8
Bank 2 ST: 0.0 to +3.0

2500 RPM
Bank 1 LT: +5.5
Bank 1 ST: 0.0 to -3.0
Bank 2 LT: -0.8
Bank 2 ST: +4.0 to +7.0

What it looks like is that upon going to 2500 RPM, both Short Term trims change, but Bank 1 quickly goes back to the same value as at idle while Bank 2 stays more positive.

From your post, it looks like this is pointing to a vacuum leak. Would appreciate your opinion/direction.

One area that I would look at is the rubber bellow boot that connects the throttle body, Idle Control Valve to the MAF. The bellow portion of the boot that connects to the IAC normally tears. If you haven't replace it recently, take a look at that.

[cmelkoni]

Thanks. I will check that boot and other intake components for a leak as soon as I can.

Any thoughts on the TPS sensor? Live data of the Absolute TPS position just doesn't seem right to me. Again, here's what is happening:

TPS is at 0.0% when at idle. When not in gear, I would expect the TPS % to smoothly follow the accelerator pedal as I step on it. Instead, a change doesn't register until I push the pedal halfway down and the RPM's are pretty high, at which point the TPS % will be in the teens but not steady. If I peg the pedal, I can briefly see the TPS % in the 90's, but it jumps around.

I took the car out for a test drive and watched the TPS%. It does change and I can see it occasionally in double digits, but I couldn't see any linear pattern.

[StephenVA]

The signal you are reading is what is called a processed value. It will NOT give you a lab scope line wave/steps.

In order to get a true value of TPS readings you will need a lab scope. But there is an alternative. Just read it with the engine off but key on through the OBDC port with a scanner. It will put you in the ball park to see if it functions. When they fail the volt values jump all over so there usually no codes being set, just input the engine management systems can not comprehend so it does stupid stuff. Most shops choose the "when in doubt replace" methodology as it is after all 15 years old and only has a 100 thousand mile cycle life.....

If you purchase the BMW laptop software you can see the values live and any captured failures
Helpful?

[cmelkoni]

StephenVA,

Thanks for the reply. I believe what you are saying is that the "ABSLT TPS %" value from my Actron scanner when the car is running is not a real-time value but is processed. You therefore won't see the expected smooth transition from about 5% at idle to 90% WOT while the car is running. Correct?

I believe you are also saying that you can see the expected smooth transition if you scan with the key on/engine off. However, in this case, throttle must be manually actuated.

Please confirm above.

Yes, I do have BMW laptop software, DIS V57 with K-line cable. I just haven't installed and used it yet which I will do tonight.

[upallnight]

Originally Posted by cmelkoni

My apologies. I thought I had posted all pertinent information and had just assumed that describing the OBD code for cylnder #2 misfire implied that my SES light had come one. That was my fault.

Here's the latest update: I was so sure that this was a failing coil pack that I replaced the coil on Cylinder #2 on Friday. I then took a weekend trip and didn't even get out of the gas station before it happened again. Same thing as before, always when stopped and idling, never while driving. In this case, it happened twice, once while idling at gas station, second while idling at friends house after starting it to warm up. As before, idle drops slightly, can hear engine running different, SES light comes on. Step on accelerator and car is a shuddering mess with barely any power. Shut off ignition, turn back on, and car is back to normal.

A little more codes thrown this time though. Now I'm getting:
P0300: Random/multiple cylinder misfire detected
P0302: Cylinder 2 misfire detected
P0306: Cylinder 6 misfire detected

I also had a cylinder 5 misfire code before I erased them the first time. So, multiple cylinder misfires now instead of just cylinder 2. Really wish I had the other codes before I bought a new coil pack.

Now, I realize there are numerous issues that could cause random cylinder misfires: fuel delivery, vacuum leak, bad computer, etc. But, I'm hoping that these 2 symptoms can narrow down the list:

1. Only happens when idling and stopped. (Alternatively, it could be just at low RPM's for an extended period of time which is what idling is essentially)
2. Ignition off then on immediately solves the problem.

I would appreciate anymore insight someone could give.

Please let me know if you need more information and I will be happy to get/share.

You have misfires with injectors shut off. The computer is not sending a signal to the injectors to open so that cylinder is basically dead. Turning the engine off and back on just reset the misfire counters and the computer is sending a signal to those injectors to open again, until x amount of misfires has occurred and the computer will shut off the injectos again.

Reviewing the short and long term data that you posted doesn't show anything that will cause the computer to turn off the injectors.

In a computer controlled engine, a misfire is defined as a cylinder not producing the same amount of power as the other cylinders. The computer can determine this by the rpm of the engine once the cylinder is on the power stroke. In the old day, misfires were limited to spark plugs, ignition coil, spark plug cables and the distributor cap and rotor. I think you have something else wrong and not just a lean condition.