Compression test has not changed. On a warm engine, remove the plugs, open the throttle plate, install the tester and crank it over until the needle stops climbing.

 

I found a good step by step guide for doing both dry and if needed wet compression test. Have to get my hands on a compression tester. I'll post results when I can get it done.

 

Many auto parts places will rent them for free with a deposit.

 

I did switch the coils and the misfire stays on #3. I also switched the fuel injector and the misfire stays on #3.


Going by the freeze frame data the misfire is on a cold engine. To be honest I cannot feel the misfire. I have had other cars where it was very noticeable if there was a miss.

I'll check the wires going to the coils.

If I don't get it done Sunday it will have to wait until Wednesday.

From different things I have read here and on some Audi forums I'm beginning to think it might be carbon buildup on the intake. The car has almost 120,000 and no records of a cleaning and the records I did get with the car showed some long intervals between oil changes. I got the timing belt done when I bought the car. I've got a dose of Techron in the tank right now even though I know the injectors are pretty clean.

 

I see you changed the plugs and since the misfire occurred before and after, it's probably not the plugs. Unless one was gapped to large and you were lucky enough to install in the same hole. In which case, buy a lottery ticket. [emoji16]

I'm not sure what the mechanics are behind it are, but carbon buildup is more of a problem for the fuel stratified injection engines, like the 2.0T in later model cars. Never have I see that as a concern for the 1.8T.

If you're not losing any oil or coolant, and no cold startup misfires occur, and would not be inclined to think it's a head problem. You can do a quick check even without a compression gauge. Just pull the plugs and shine a flashlight in them. If the crowns look dry, probably not that.

More inclined to say it's the wiring. Either the coil wires, or possibly the injector wires.

On the injector wires, they usually crack around the connection to the injector because of the sharp bend there. May have to pull back the boots (if they themselves have split) to see.

Sent from my SM-G965U using Tapatalk

 

Compression test done and everything seems to be OK
Cylinder #1 160
Cylinder #2 155
Cylinder #3 165 (the one with the misfire)
Cylinder #4 155


I did notice when I removed the plastic cover from the engine that the harness to the coils was trapped under the cover. None of the tie downs were connected. I haven't cut away the insulation in that area yet to see if any wires are damaged. I'm leaving the cover off for now.

All of the injectors are working and all of the spark plugs looked normal. Looked in each cylinder with a flash light and what I could see they all looked the same, black deposits on the piston tops.

When the code comes back I will save the freeze frame data. I'm just using a generic code reader from Harbor Freight, should I be looking to get something Volkswagen specific?

 

The only other thing I can think of to cause a misfire would be an incorrectly gapped spark plug, which is highly doubtful since they're pretty close out of the box.

I'm seriously leaning towards compromised wiring; either coil or injector. When you inspect, if you find any problem, I'd just replace the other wires as well, since they won't be far behind.

ECS sells a kit to replace the coil wires for the Passat/A4 1.8t engines, or you can rebuild them yourself.

As for the injector harness, no one seems to sell rebuild kits for them, so you'll need to build your own.

There are DIYs for both in the Audizine B6 A4 forum.

Sent from my SM-G965U using Tapatalk

 

This is a long shot but since most of the obvious stuff isn't panning out it, you could check the sensor that indicates the timing to the ECM. I think it is the camshaft sensor but hopefully someone will chime in if that's not right. The sensor ring could be bent or dirty and the ECM might be getting a spotty signal from cyl 3 even though the car is not actually mis-firing.

 

I think it is the camshaft sensor but hopefully someone will chime in if that's not right.

Are you talking about the Crankshaft Position Sensor? The CPS going out would probably brick the car but you might be talking about something else.

 

The CPS makes more sense for a mis-fire detection sensor and that wouldn't affect a single cylinder. So, nevermind....

 

I'm going to inspect the wire harness going to the coils. If there is an actual misfire it must be intermittent because there is no evidence that the cylinder is not firing at all. I won't be able to get to that until I have time on Wednesday. I took a look at the ECS harness kit and if any thing looks bad I will probably be going that route. Anything I see will get a temporary repair for now.

 

The coil pack wires are a notorious issue on Passats. I had to repair the wire jacket on mine to prevent grounding out the respective coil pack wires

 

I didn't get a chance to check out the wiring harness today. I have gone about 100 miles since I last reset the code and removed the plastic engine cover. The code had been coming back after about 40 miles. I'm sure there is some problem with the wires, just not sure if it has to do with the harness having been caught under the edge of the cover or the heat of being under the cover. I'm still going to unwrap the harness, at least the wires to cylinder 3. I'll update when I know something.

 

The engine light came back on after 100 miles.
This is what I got as freeze frame from my code reader.
DTCFRZF P0303
FUELSYS1 OL
FUEL SYS2 N/A?
LOAD_PCT(%) 51.4
ETC( F) 64
SHRTFT1(%) 0.0
LONGFT(%) -0.8
RPM(/min) 1359
VSS(mph) 6?
SPARKADV -11
IAT( F) 52
MAF(lb/min) 1.455
TP(%) 16.5
AIR_STAT OFF

The light came on as I was backing out of my driveway yesterday morning.

This morning I peeled back the rubber boots on all the plugs. No obvious signs of any breaks in the wires. I removed the outer wrap on the wires to #3 and couldn't find anything going back about 5 inches from the plug.

Not sure what my next step should be. I'm willing to change the harness with the ECS kit, just not sure what to do if that doesn't fix the problem.

 

I would suggest getting a harness kit, 4 new coils and spark plugs and see how that goes. Depending on how many miles you have on it, it may be time anyway especially in the area of the country you live. Cold weather can make wires brittle when they are old.

 

Going to get the harness kit, coils were replaced 600 miles ago by Volkswagen under recall and plugs were replaced 300 miles ago.

 

Really need that new harness now. Since trying to check the wires the other day I am now getting P0300 and P0304 along with the original P0303.

When I get the harness and get it installed I'll take pictures and try to keep track of how long it takes to get done.

 

MAF reading is really low for 1359 rpm unless the throttle was closed (which it should not be at 51% load).

 

The 1.8t is notorious for the harness issues, specifically where the wires make the 90* turn into the connector to the coil pack. If you pull back the book on the connector you'll probably see that the wires are brittle and even some small broken strands, also the wire that connects to the ground on top of the valve cover is probably in need of replacing. You can do the whole harness but if you're handy, get new connectors (I get them from a a pick and pull) and rewire them yourself, do one at a time so you don't mix them up. I wrapped them in that heat shield wrap and going on 3 years now with no issue)

 

I was able to install my new wire harness today.

The worst part of the job was stripping the wire and crimping the connectors. Even with a good crimper the space where you need to work is very limited. Also the little seals that go around the wires and into the plug are virtually impossible to push into the plug.

Start to finish including clean up was a little under two hours. That also included time to stop and take pictures which I will be posting later in write up of the procedure. The instructions that are on the ECS site are very good and include many pictures. All of their crimps are perfect, I was not able to achieve perfection but got very workable results.


One tip I can offer, based on reviews on the ECS web site about having trouble with the crimps on the very thin signal wires. On the signal wires I doubled the length of the stripped part of the wire and folded it back on itself and twisted to give the connector more to grip. I have no fear of the crimps coming apart.


I've only put 15 miles on the car since I finished. It does sound better and only time will tell if the codes come back. I hope this is the end of this problem


Thanks for all the advice.

 

IF the misfires return, which hopefully they won't, I would go to the other end of those wires; at the ECU connector. Something that I have done for preventative maintenance is to disconnect the battery, remove the ECU box cover, then disconnect and spray contacts and receptacles with electrical contact cleaner. Connect and re-connect a couple of times to wear through any contamination. A small amount of Oxidation at an ECU pin can cause critically-high resistance.

 

HELP!

The engine light came back on after 90 miles. Got into the car after work and started it up to let it warm up a little before I headed home and the light came on almost immediately.
This is what I got as freeze frame from my code reader.
DTCFRZF P0303
FUELSYS1 OL
FUEL SYS2 N/A?
LOAD_PCT(%) 31.4
ETC( F) 100
SHRTFT1(%) 0.0
LONGFT(%) -0.8
RPM(/min) 1257
VSS(mph) 0?
SPARKADV -5
IAT( F) 81
MAF(lb/min) 1.017
TP(%) 14.9
AIR_STAT UPS

So changing the wire harness has not solved my problem.

Going to try ylwagon's suggestion of checking the ECU connector.
What is the recommended contact cleaner to use? I know from past experience that some of them are not very friendly to some plastics.

If that does not fix the problem where do I go next? I know that the injector works but I'm not sure that the wiring to the injector works properly. how do I check to make sure the harness to the injector is good? wires look good but I'm not sure that it works all the time.

PZ mentioned the MAF reading the last time I posted the Freeze Frame results. Should I be looking at the MAF? But why would that only effect cyl. 3?

How do I know if the misfire is continuous or intermittent? It has taken as little as 20 miles for the code to come back or as much as 100 miles. I'm only getting the solid light not the flashing light.

I have some other repairs that I want to make to the car but I don't want to spend any more money until I can get this situation fixed.

If a different code reader would give me more information as to what the real issue is I would get one. Right now I'm using a generic Harbor Freight model 99722 that I've had for over 10 years.

 

I'm coming into this show way late...

After reading the thread, I think you might have a mechanical problem, PZ and ylwagon see if you concur.

You've swapped out the coil packs and injectors, now you've gone through the task of doing injector harness work and the misfire still is present.

I see that you've done something with the plugs, but did you actually change the #3 cylinder plug to make that was faulty?
I'm assuming so.

I think you might have a problem with your piston rings or a valve or even maybe carbon build-up.
Here's my .02 cents worth.

There is a possibility that one of the valve cam followers (on the effected cylinder) might be collapsed. If the cam follower is collapsed the valve won't open far enough to get the proper amount air/fuel for a full power cycle on that cylinder.

Maybe even a broken or cracked valve spring (but not likely).

I see you have a high number for compression on that cylinder. How did you test, dry or by giving a squirt of oil in the cylinder?
If you did the wet compression test by squirting oil into the cylinder there is a chance you could have a broken piston ring.

Another possible cause is having an alien piece of carbon build up on the edge of the valve or piston and that is causing some kind of pre-ignition.

I also see that you are questioning what to use to clean the ECU connector. CRC products are very good and would be the one I use.

https://www.crcindustries.com/products/lectra-motive-174-electric-parts-cleaner-19-wt-oz-05018.html

https://www.crcindustries.com/products/co-174-contact-cleaner-10-fl-oz-02017.html

It's definitely worth a shot to clean the connections, but from a reliability stand point, our cars are pretty reliable when it comes to wire harnesses.
Yes other things fail and are known to fail but the engine harness is pretty solid. Unless some one was cutting and splicing on the car before you I would lean towards ruling that out. Don't ignore it but I would be looking at mechanical things at this point in time.


The other thing I wanted to point out, getting a VCDS scan or investing in the VCDS software I think would help you pinpoint issues more reliably. VCDS is VW specific so it's going to give you the most accurate information.

 

I have changed positions of the plugs, coils and injectors.

The compression test was done dry.

I am assuming that when you are talking about a valve cam follower being collapsed that you are indicating an intake valve. How can I check that?

I did not do a leakdown test. I was not sure from instructions that I found online how to get cylinder into correct position to make sure valves were closed.

Just wanted to make clear it was the ignition harness that I replaced not the injector harness.

If I did not mention it earlier in this thread I had the timing belt replaced when I bought the car by a local VW specialist. The code existed before that work was done.

What has me confused is the various amounts of time and mileage that it takes for the code to reappear. I would think that if the misfire was chronic that the code would come back immediately.

 

VCDS will give you more information and can graph lots of data in real time as you drive or while you are idling. You can get a generic OBD11 reader for $10-$20 and use the 'lite' version of VCDS for free. This will give you some but not all of the capability of the full version. If that isn't enough, you can license the full version of software for $99 which I think will works on up to 3 VIN's. I don't know the capabilities of the lite version as I have always had the full version. You might be able to find someone local who the full VCDS with an unrestricted number of VIN's if you ask around.

I can't remember if it's on VCDS or one of my other (non-VW) cars but there may be counters that show the # of mis-fires for each cylinder. If VW does support that, it's the kind of information that VCDS would easily show you.

 

AndreasPassat brings up many possible things that could cause a misfire, but I lean more towards electrical because of your statement in the original post: "The engine runs smooth and sounds normal". That and normal compression tells me that the engine is likely good mechanically. Now, a couple of other causes might be an air leak at the #3 injector seal, or an intake manifold leak. I would also check all of the vacuum lines and the notorious check-valves, and any rubber ducting downstream of the MAF.

VCGS is the preferred way to capture what's going on in real time. I paid about $350 a VCDS tool several years ago, It's around $200 now. For instance, I wanted to check what my MAF sensor measured, from idle to 6,000 RPM, wide-open with my 3.0 A4. The resulting log-file proved that the sensor output increased smoothly with engine speed, plus a visual graph showed the same. You can set the log file to watch the misfire counter fields, and when they occur can be compared to other things like engine speed, throttle-angle, coolant temp, and so-forth. You should also check the fuel trim fields, which if lean (fuel added), and the MAF response looks normal, then possibly low fuel pressure.

 

Thanks for extra info.

So plugs, injectors and coils can be ruled out with fair certainty.
Doing a dry compression test in your case confirms that the piston rings are more than likely in decent shape. So we can probably rule that out now.

The cam follower is the hydraulic piece between the cam lobe and end of the valve.
If a cam follower is collapsed it will cause the intake valve to not lift up as far, thus not bringing in enough air/fuel for a properly charged cylinder for the power stroke. (potential mis-fire)
Your engine has the 5 valves per cylinder, 3 intake valves and 2 exhaust valves. Each valve has a cam follower on it.
Conversely same holds true with the exhaust valves, if the cam follower(s) are collapsed the exhaust valve will not lift up far enough to let the burnt fuel mixture out, again (potential mis-fire).
There is a very rare chance one of the cam followers is actually cracked or broken, again I want to stress that is very rare. But something that should be checked.

The cam follower gets pressurized with oil when the engine starts and takes up mechanical slop between the cam lobe, cam follower and valve stem. It makes for a much quieter valve train assembly.
Doing a leak down test right now I don't think will net you very specific results. Your compression levels are not stellar but they are all about the same so that tells you that your pistons, rings and valves are sealing fairly decent.
If they weren't you'd have a significantly lower number probably with variations.

Checking for a collapsed cam follower could be checked with one of the bendy type bore scope camera's that can enter through the spark plug hole.
If one of or multiple valves are not lifting up far enough it would be obvious as compared to another known good cylinder. You could remove the cams and pull out all of the cam followers and have them checked at a auto shop.
I think at this point in time that's what I would do. Removing the cams on the 1.8 is fairly easy. Once the cams are out the cam followers can be easily removed by hand to have them checked out.

Part of me wants to say try swapping the cam followers to another cylinder and button everything back up and see what happens, the problem with that is if the problem moves to a different cylinder, you've identified the problem, but then you have to tear apart everything again. And tearing stuff apart just because is not very much fun. So I would not recommend that.

As far as the code coming on in a timely manner or not, a malfunctioning valve train could exhibit the exact symptoms you have.

 

ylwagon has a good point about the o-ring seal on the injector and or manifold gasket.
You can test that theory by using something like WD-40 and spray lightly right around where the injector is pressed into the manifold. If the engine speed increases when spraying that immediate area then you know you have an air/vacuum leak.

 

The negative LTFT in both scenarios could be a clue, but I’m not positive what it says. Possibly that you’re losing air from a leak between turbo and engine, or that you’re MAF is faulty?