Volkswagen Passat Forum banner

1998 1.8T AEB: Misfires on Cold Start

13K views 17 replies 8 participants last post by  vbora01 
#1 ·
Hey everyone,

So I've searched around the forums about misfire problems, but none of the threads which I have read have really pinpointed my specific issue. It is a 1998 1.8t AEB about to hit 200,000 miles and I'm keeping her alive.

Problem: About once a month (extremely intermittent, and rarely occurs) I will start the car and it will begin to run extremely rough. While it is running this way, the CEL starts blinking. After about 30 seconds (or even shorter than that) the engine completely smooths out and the CEL solidifies. I have run VAGCOM and the ECU is registering "multiple random misfires," and I'll be checking later tomorrow if it is a specific cylinder. This only happens during some kind of change of season, or when the car sits for an extended amount of time (maybe about 1-2 weeks.)

What am I looking at here? Has anyone else experienced something similar? Coils, ICM (did check it with an multimeter and it had resistance at each pin), battery, ECM, which component would do that? Is there a way to diagnose a coil involving a multimeter?

Appreciate any input on this.
 
#2 ·
I gotta be honest. I have a 98 AEB with 240k and about 2x year my car does the same thing it has since about 120k. Mine runs rough (feels like 3 cyl) but I don't think CEL always flashes...sometimes does. It always self corrects and has never gotten any worse. Therefore I haven't done any further diagnosis. Every single time it happens if I drive away it will drive fine and not misfire again for months. In the mean time I have needed to replace my ICM (~160K) and that didn't change a thing. I do my plugs about every 40-50k and I still have all my original coil packs.
 
#3 ·
I'll see what the codes report, and if its the common single cylinder misfire (which would probably indicate the ICM) I might go about replacing it. All the parts on the car are pretty much original, an older lady drove it most of its life. I'll double check the service records too, and see if ignition was ever an issue.

Thanks for the input, it would not be a VW if it didn't have some kind of quirk
 
#4 ·
Inspect every rubber hose and plastic fitting under the hood. A car that age and with that many miles could easily have a vacuum leak. I wonder if a check-valve is occasionally sticking and causing the intermittent rough cold start idle?

Check the ignition system ground connection on the valve cover between cylinders 1 and 2. The pen tip is pointing at the connection.


A Techron treatment to clean the fuel injectors couldn't hurt. Techron is available at most auto parts stores.
 
#5 ·
Bad ICM or bad coils, in my experience, do not correct themselves nor are they intermittent. Once they start to crap out they then go rapidly downhill.

My 2 cents; It sounds like a coolant temperature sensor (CTS) that's on the fritz. Changing the CTS on the AEB engine is a piece of cake. Very easy access at the rear of the engine up against the fire wall.

A faulty CTS will cause your symptoms. It did with my car - twice over the last 230,000 mis. It is an easy 20 minute repair with a $25 part (plus O ring). I beleive the OEM CTS is held in place with a plastice "C" clip that readily snaps in two when attempting to remove. So when you buy the replacement CTS be sure to get a C-clip. The CTS comes with an O-ring but not a replacement clip.

I'd be curious to learn if your problem does get fixed by R/R the CTS.
 
#6 ·
I plan on making a vacuum leak detector (as seen on the forums somewhere,) because like I said before, nearly all of the hoses and parts under the hood are original. The indy shop that took care of the car for her replaced the check valves (believe there are 4 or 5?) I'll double check to see if they replaced all of them. I'll also check out that ground connection. Techron is definitely in the plans. I've also got a set of BKR6EK plugs goin' in today.

Wouldn't the CTS throw a code? I'm not sure about the AEB, but do they have two CTS' (the one for the meter in the instrument cluster, and one for the ECM?)
 
#7 ·
Mileage: 319870km/198758miles
-------------------------------------------------------------------------------
Address 01: Engine Labels: 058-907-55x-AEBN.lbl
Part No: 8D0 907 557 P
Component: 1.8L R4/5VT MOTR HS D02
Coding: 06001
Shop #: WSC 05314
VCID: 3D7FC86F1DEA

4 Faults Found:
16684 - Random/Multiple Cylinder Misfire Detected
P0300 - 35-10 - - - Intermittent
16686 - Cylinder 2: Misfire Detected
P0302 - 35-10 - - - Intermittent
16684 - Random/Multiple Cylinder Misfire Detected
P0300 - 35-10 - - - Intermittent
16686 - Cylinder 2: Misfire Detected
P0302 - 35-10 - - - Intermittent
Readiness: 0000 0000

Since its Cylinder 2, is it safe to assume ICM, since thats the one most commonly associated with ICM failures? I just don't think it would be a coil. Good call on the ground though, that little bolt was loose and the wire was moving around. Changed out the plugs, and all of em showed normal wear and tear, nothing out of the ordinary. The PCV system looks airtight, as well as all the vacuum lines, some of em even look brand new!
 
#8 ·
Clear the codes.
Swap coilpack #2 with coilpack #4.
Put the car back into use.
Scan again and see if the misfire returns and if it does, where it returns.
If it returns on cylinder #4, suspect that coilpack.

Yes, a failing ICM often loses the channel for the #2 cylinder first, but not always.

The AEB (like the others) has a CTS located at the back of the head which is two senders in one part. One side sends to the cluster for the temperature gauge and the other side sends to the ECM. A failing CTS doesn't always throw a code. If the data it's sending is wrong, but plausible, the ECM will believe it. One common failed CTS symptom is hard hot starting. The hot engine gets flooded because the confused ECM is delivering rich cold-start fuel mixtures.

There is another temperature sensor near the output of the radiator. This one is used to activate the secondary electric cooling fan when the cooling system needs to increase capacity. This one doesn't contribute to engine management decisions and won't effect the way the engine runs (unless it allows you to overheat).

Good idea about building a boost/vac leak tester. Sometimes that test will uncover leaks you'd never find otherwise.

You may also want to clean the MAF and check the MAF data. For the first 30 seconds or so of cold start, the O2 sensor isn't heated up and on-line. During this time the engine is running open-loop based on MAF data (probably with learned trims applied). Once the O2 sensor comes on line, the system goes closed-loop and the fueling gets more accurate.
 
#9 ·
Thank you for all the help, I really appreciate it. I'll swap the coils out, and see what happens (although it may take weeks to see symptoms again, cause it only happens every once and a while.) I'll also keep my eye on the CTS, but will first diagnose either the coil pack or ICM.

Just for clarification, where is cylinder 2 located? Is it the second one from the front of the car, or the second to last one?
 
#10 ·
The cylinders are numbered from the front of the car to the fire wall. Number one is closest to the radiator.

One thing about ICM equipped cars is, it's never a bad idea to have a spare ICM in the glovebox. If the ICM fails completely, it strands you. If you replace this ICM with a new one, toss this old one in the glovebox as a "get me home" spare. If swapping the ICM doesn't solve the problem that leaves you with a fully functional backup ICM.

The ICM is mounted to a metal heat-sync embedded in the top of the airbox. The ICM generates heat as it works and needs to be kept cool or it will fail prematurely. When you replace it, clean off the surface of the heat-sync and apply a fresh super thin layer of thermal conductive paste. This helps heat transfer from the ICM to the sync. The paste can be found at some auto parts stores and most computer stores. It's the same principal used when they attach heat syncs to chips in a computer.
 
#11 ·
I re-pasted the heatsink and ICM with some thermal paste I had laying around earlier this year (maybe July.) I made sure I did not use a lot. I bought some MAF cleaner today, will switch out the coils too when I get the opportunity. The CEL turned off, and I'll be clearing the codes today. I'm looking at getting a new Huco ICM at some point for the whole reason of it being a spare as well, though it could already be a problem (yet to be determined.) If it looks like neither the ICM or a coil, the CTS is my next bet (although I could just replace it now for preventative maintenance reasons cause it's cheap.)

I wonder if this is the original ICM that the car was manufactured with, then I'd be really impressed considering the volume of failures I have read about around here. Got my madre scanning the service records at work and sending them to me in a PDF file (previous owner was meticulous with these, thank god.) Additionally, thanks for all the insight Steve.
 
#12 ·
Misfire

If you have a bad pickup coil it will produce all kinds of errors (too many). Instead of swapping just replace them all. You can find a set of 4 on ebay for $69 bucks (with shipping). Switch them out and then clear out all the codes. This should solve your problem.
 
#14 ·
If you have a bad pickup coil it will produce all kinds of errors (too many). Instead of swapping just replace them all. You can find a set of 4 on ebay for $69 bucks (with shipping). Switch them out and then clear out all the codes. This should solve your problem.
Following this and wonder what is meant by "pickup coil"? Coil pack? The AEBs are not, to my knowledge, known for coil pack failures. Personally, I would much rather test and find a bad coil pack and replace the one bad pack w/ a quality part than replace all the other good parts with cheap ebay parts.

On the other hand the ICM is known to fail.
I 2nd the notion to grab a new ICM. Depending on where you drive a spare is always a good idea. When mine failed the car was inoperable...thankfully I had a spare and was pack on the road after a 10 minute fix. I agree w/ the idea to swap out the ICM for a new one and carry the current ICM as a spare.
 
#15 ·
Sifted through every single repair documentation the previous owner had. There is no indication of the ICM or coilpacks ever being replaced. I'll probably purchase a spare when I get paid next. I already spent money on a "newer" ko3 and motor mounts from a car with 35k miles, :). I ran a bottle of Techron at half a tank, and am on a second one currently (again at half tank.) The car usually stuttered occasionally, meaning it was dipping below idle rpm's when I pushed in the clutch for a stop, and its not as bad as it used to be (it was not horrible to begin with.)

Is the fuel pump supposed to make an apparent humming noise around the back passenger wheel? I've never noticed it inside the car, but I walked around the car yesterday checking the car out and I heard a humming which I assume is the pump. When I use to drive a 2001 Jetta, you could only really hear the pump when you opened the car door and it was priming/pressurizing the lines.
 
#18 ·
Can rule out the CTS, it was replaced in 2008 (unless it went bad really quickly.) It was done at the same time the previous owner had the complete PVC system redone, the timing belt done, and the heater hoses replaced.

Doing a little more research, but I might be kind of leaning towards a complete ICM delete if it saves me from being stranded someday if my intermittent issues become permanent. Not sure yet, just a consideration.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top