Misfire Cylinder #1 Jon's '97 328is

Okay as many of you know I have a recurring misfire in cylinder #1, and decided to give it a thread of it's own. It comes and goes but the CEL is on most of the time. Granted, the car does not run rough or feel as though it misfires. Acceleration, idle, start up, and fuel consumption (24ish mpg in stop and start traffic most of the time) are all normal. Once in a while the car will actually misfire, and run rough enough to be physically felt. Here's where it get strange...turning the motor off and restarting the car clears the rough run. Sooo, my train of thought is it isn't a valve since the restart eliminates the rough run. Jared suggested it was the CPS so I started looking into that since it made sense but my PO was incredibly sharp (forum member C Sean Watts), who also thought of that. Links below to what Sean tried and I will add my own findings as well. I was a little dismayed to find the car has been doing this for the better part of two years, but it is what is.

His thread on the misfire...

Also since it seems BavAuto coils are generally reviled around here and I have them here is his thread on installing them in what is now my car.

He also had two other things in the paperwork he gave me that I noticed. A BMW shop in Raleigh suggested the cause was the European computer set up (car is Euro Delivery, not Euro-Spec so there are Euro features on the car like long & short wave radio ) and the US coils. They suggested he reflash the computer to US standards. He refused on the grounds or belief that US cars have a governor in the CPU and since this car was European it did not. It sounds like reaching to me anyways but I'm giving you all I got on this one.

The other thing was an injector spray test done at a different BMW shop in Raleigh. It showed all normal EXCEPT a slightly light spray pattern in, you guessed it, cylinder #1. But according to the tech it should not have caused the CEL or the misfire. In his (tech's) pattern very slightly off on the light side, pattern not off enough to cause misfire. And just to be sure, Sean swapped that injector to cylinder #2 and the misfire code stayed on #1. I have checked plug wires and boots, and am currently going to try a compression test at some point coming up. I expect nothing to really come of the comp test, but will try to get to it sometime coming up. Any thoughts/observations/ideas are welcomed.

switch the coil packs with another one.

and the injectors if you can too.... see if it follows

ok. so, if the coils, injectors, and plugs have all been swapped around like a chinese fire drill and the issue stays @ #1 cylinder, then i say it is related to #1 cylinder.

the crank position sensor has been replaced with no improvement, but what about the cam position sensor?? it's what actually tells the pcm what cylinder is supposed to fire and when.

also, check the condition of the trigger wheel. if it seems to set on *only* cyl 1 then i would check that as a possible cause.



df

df, methinks you have hit on something. If there is something interfering with the cam PS picking up the precise location of the #1, that could be the cause. I can't recall how exactly the sensor "sees" the cam. Is it a toothed wheel like on the crank? If there's some debris lodging between the teeth (like what happens with an ABS wheel speed sensor) it could be confusing the signal.

Would a faulty cam position sensor result in its own code? I like this train of thought. I was about to do stupid stuff like pressure test the fuel rail. Also if the cam position sensor were faulty would it still fall in line with the intermittency of symptoms I'm experiencing or would it be more frequent and noticeable?

Originally Posted by SCJon

Would a faulty cam position sensor result in its own code? I like this train of thought. I was about to do stupid stuff like pressure test the fuel rail. Also if the cam position sensor were faulty would it still fall in line with the intermittency of symptoms I'm experiencing or would it be more frequent and noticeable?

ummm.....yesss??
























well, defiinately maybe!!

df

Just laughed out loud at the business license office on that one. Thanks df!
Will start looking into this area when I get a chance. Hopefully this weekend.

Originally Posted by SCJon

Would a faulty cam position sensor result in its own code? I like this train of thought. I was about to do stupid stuff like pressure test the fuel rail. Also if the cam position sensor were faulty would it still fall in line with the intermittency of symptoms I'm experiencing or would it be more frequent and noticeable?

When the cam position sensor on Bebe went south a couple years ago, it threw a knock sensor code!

I always advise to change both position sensors if you change one. They are reasonably cheap, and if one is borderline, the other probably is, too. However, if it IS your CPS, the fault is probably dirt/metal on the wheel it "sees". It wouldn't hurt to pull the sensor, test it, and clean it and the hole it goes in, plus anything else you can reach down that hole. The Bentley lists a resistance figure for testing the CRANK position sensor (1280 ohms + or - 10% IIRC), but not the cam position sensor. I don't know if they're similar or not.

The Cam position sensor looks at the lobes on the plate that sits just behind the vanos gear on the intake cam. Intake cam is the lower one in the photo.

When mine went bad I was down a bit on power and had some random rough running. I also had the CPS fault code though.

If you're going to pull the sensor to check for debris on the sensing face you might as well replace it for the amount work it takes to get to the point where you can check it.

Alright, I just learned I have an uninterrupted Saturday to do whatever I want to my car.
My wife will be in RI until 9PM that night, so here's my options and and parts on hand list what do you guys think I should try to get done first and in what order?
Parts on hand
Transmission filter and pan gaskets (have no leaks, this is maintenance).
Giubo and CSB
Ignition coils and boots
Entire cooling system but don't have the expansion tank yet
RTABs
Cabin filter, and full supplies for an oil change on hand also.
New door handles
Power lock actuator for passenger door

Lock actuator, Door handles, Cabin filter and oil change, Ignition coils and boots, Giubo and CSB. That would probably take up my entire day, including at least one SNAFU, which is typically what happens to me.

That would leave RTABS, Cooling system, and transmission fun for another time, probably an entire weekend.

It's the lock actuator I'm kind of avoiding, I've heard it's an awful task, but you are right. That and the ignition coils are the only real issues I have, the rest is fun stuff. Advice noted and taken.

I've done the driver's side. As long as you don't strip the three torx bolts you'll be fine. It's tough because you can't see in there without mirrors. After feeling it out with my hands for a few hours I got the hang of it.


Giggity.

Originally Posted by SCJon

Ignition coils and boots
Cabin filter, and full supplies for an oil change on hand also.
Transmission filter and pan gaskets (have no leaks, this is maintenance).
Giubo and CSB
Entire cooling system but don't have the expansion tank yet
RTABs
New door handles
Power lock actuator for passenger door

this would be my order. I've done the door lock actuator. First time took me about an hour, cause nothing would line up, second time was about 30 minutes, and I had to do that one twice cause of stupid human error.... but, everything lined up quickly and nicely.

I want to see those new (to you) coils in to see if that stops your misfire issue.

Thanks Patrick, I am very curious about the coils too, that will be one of the things I definitely do. Probably that and the actuator.


Sent from BimmerApp mobile app

Okay last night I swapped all of the Bavauto coils out and replced them with the Bremi M52 coils I got from Derek. The boots I had in were brand new, so I kept those and cleared the codes stored, which were (I have a Peake reader, not an OBDII)
1. Misfire cylinder #1
2. Catalytic converter below threshold cylinders 1-3 (this accompanies the first code about half the time)
3. secondary air flow low cylinders 4-6

I cleared them all and ran the car about ten minutes with no codes popping, however I was unable to drive it with the torrential rain we received last night. We are having a tropical storm tonight and tomorrow so I doubt I will be able to do much until Saturday since I live on a dirt road. Don, if you read this how is that storm? I imagine you're in the middle of it now.

The car seems much stronger than it did with the previous coils so I was a little shocked by the engine seeming that way. It has a definitive growl now it did not have with the Bavauto coils. That was an odd effect that I did not expect, and maybe it was just because I had it in the garage when I ran it (with the garage doors open, of course).

It's back. CEL is on again at 47 miles from the coil swap. Next up injectors from Karl's eBay guy. Anyone have any other ideas?


Sent from BimmerApp mobile app

I assume you did plugs too right?

If so, then yeah I'd look at fuel issues.


Sent from BimmerApp mobile app

I didn't, the plugs only have 10k on them. Didn't even think of it since I was convinced it was those coils. I'll pull them tomorrow and have a look at them. Possibly replace #1. Frustrating to have to drive 50 miles to find out what doesn't work. Argh.


Sent from BimmerApp mobile app

If replace the plugs too. After 10k the plug edges might not be as sharp causing the coils to have to work harder to produce the spark.

How old is your battery?


Sent from BimmerApp mobile app

I will give that a shot as well, between my 3 cars I have to order I the future 16 spark plugs, 19 quarts of oil, 23 quarts of ATF, 3 air filters and 3 oil filters. Yikes. Amsoil is going to love me.

Not sure how ole the battery is, never even thought of that. I'll check it out. My Saturday got ruined by work, went from nothing to booked all day.

Edit I wonder about the plug wires too.


Sent from BimmerApp mobile app

Jon, I think you need to look at the knock sensors. The 328 has two of them, one that monitors the forward bank and one that monitors the rear bank of cylinders. (they come as a pair in a single harness with one connector to mate to the engine harness.) The knock sensor is what determines if you have a misfire. The DME is monitoring the crank and cam position sensors to determine the position of the piston and the valves to determine when and how long to fire the fuel injectors and when to fire the coil pack and create the combustion in a given cylinder. In parallel with that it's monitoring the knock sensors to determine if the explosion from combustion happened just following firing the coil pack or before. If the knock sensor doesn't see the explosion signature from combustion, the DME flags that as mis-fire. If it sees combustion before it fired the coil pack then the DME adjusts fuel and timing to try to eliminate the pre-ignition condition. If everything is correct the DME adjusts fuel mix and coil pack firing to get the max performance without pre-ignition.
If you've got a bad knock sensor it can cause the DME to flag a mis-fire when there is none. If the knock sensor gives the DME an out of timing signal, the DME is going to adjust fuel and timing to try to eliminate a non-existent pre-ignition, and it could cause the DME to set fuel and timing to a very poor performance setting.
Can't guarantee this is it, but knock sensors shouldn't be overlooked when trying to track down a mystery mis-fire CEL. While the BAV Auto coils don't have a very good rep, they may have just been a minor factor in this and thrown you a red herring.

Thanks Don that is a terrific idea and an incredible explanation too. I had an idea what a knock sensor did but now I feel much better about it. Given a couple other things the car does, I feel like this may be the issue. It does have a smell upon start up that makes me think the fuel could possibly be off but is ever so slight. Which of course fits PERFECTLY with what you mention in relation to the self correcting. I feel as though I am a little down power wise from where I should be too, fitting in again. Is there a way to test the knock sensor when I pull it or do I have to replace it to find out?


Sent from BimmerApp mobile app

I believe the knock sensors are piezoelectric accelerometers. If you had an oscilloscope you could get an idea if it was working by tapping the sensor and looking at the output signature. You might be able to see something on a DMM but it would depend on its sample rate and I think you would be able to judge if it was completely dead. If its damaged that would be much more difficult to judge from a crude test like this.

Thanks Don, I'll just replace it, since I think it's damaged and not destroyed. They're $66 on Pelican, so it's not too bad to just replace it. How do you know so much about the electrical parts on these cars? It's amazing the amount of things you know about them! : wow:


Sent from BimmerApp mobile app