Chasing An Engine Misfire

[Majoraslayer]

Recently during the polar vortex, the temperature here locally reached -16. I work second shift, so I had to drive home in that. When I started the vehicle, it made a loud almost scratchy sound in the engine until it warmed up. Any vehicle is going to hate running in that kind of weather, and after warming up the sound would go away and it would run normally. No drivability issues, I just thought I would note it.

Fast forward to yesterday, I was driving around town and not having any issues. I hopped on the Interstate for a shortcut home when my CEL started flashing. When I pulled off the exit and came to a stop at the stoplight, I noticed it was idling very rough. At take off it would bog down terribly.

Today, I connected my laptop to the OBD2 port and ran diagnostics on it. The first error I had was for a misfire in cylinder 3 (P0303 I believe). My software can monitor misfires in individual cylinders, so I was able to watch the misfire count live. No other cylinders were reporting a misfire. I tried swapping the #2 and #3 coils to test if the problem was the #3 coil. Defying all logic, I then started reading a misfire on cylinder #1. I then swapped what was originally the #3 coil with the #1 coil; now I was showing a misfire at cylinder #2.

At this point I was completely perplexed, so I decided it was a good time to install all new plugs (AC Delco 41-103). When I put it all back together, I installed all of the coils in their original spots. I fired it up, and cylinder #1 was misfiring again just as bad. Finally, I swapped the #1 coil with the #4 coil, but the problem stayed at the #1 cylinder.

WTF is gong on with this stupid thing? I appreciate any advice!

If its a leaky gasket of some sort, I should note that each time I registered a misfire it was only on one cylinder at a time.

Earlier I rented a compression tester kit; I had never tested compression before so I just removed the #1 plug, connected the tester, and fired it up. Note that I didn't let it warm up first or disconnect all of the coils, so I was testing it with the engine running. I tested the #1 cylinder that continues showing issues, then tested #4 because it hasn't show any issues yet. Cylinder #1 showed 90 lbs, and Cylinder #4 showed 100 lbs. After some research tonight I learned the proper way to do it, so tomorrow I'm going to check compression on every cylinder and post my results here.

If the problem isn't the cylinder head, spark plugs, or ignition coil, the only things I can think of it could be are some sort of sensor (O2 seems to be operating within tolerance according to logs), or a problem with the fuel system. If it is the fuel system, how do I get to the injectors, and once I do how do I diagnose them? Is it possible to clean them?

 

[IllogicTC]

[video=youtube;DjL_Wefla0s]http://www.youtube.com/watch?v=DjL_Wefla0s[/video]

Mayo made this video on how to help diagnose this issue.

To get to the injectors, download and consult the service manual package available for free here. Find the year relevant to your needs and download the package, Mooseman was very generous in making these available, take full advantage!

Also, check the intake manifold bolts. This will require some disassembly (such as removing the PCM module), but it may be worth it. It's a rather common issue on the I6 for the intake manifold bolts to come loose over time. Funny, too, because the exhaust manifold usually rusts up where you can't get them loose.

I would start with the tests outlined in the video and double-checking the intake manifold bolts before dipping into injectors, both are easy to do before we get into playing around with the expensive stuff.

Are there any other error codes that show up, or just the misfire?

 

[Majoraslayer]

[video=youtube;DjL_Wefla0s]http://www.youtube.com/watch?v=DjL_Wefla0s[/video]

Mayo made this video on how to help diagnose this issue.

To get to the injectors, download and consult the service manual package available for free here. Find the year relevant to your needs and download the package, Mooseman was very generous in making these available, take full advantage!

Also, check the intake manifold bolts. This will require some disassembly (such as removing the PCM module), but it may be worth it. It's a rather common issue on the I6 for the intake manifold bolts to come loose over time. Funny, too, because the exhaust manifold usually rusts up where you can't get them loose.

I would start with the tests outlined in the video and double-checking the intake manifold bolts before dipping into injectors, both are easy to do before we get into playing around with the expensive stuff.

Are there any other error codes that show up, or just the misfire?

Actually, that video is exactly where I started. In fact I've watched it about 15 times. That's what I'm saying; following the logic of the video, when I swapped the #3 coil (source of the original misfire) with the #2 coil, the problem should have either followed the coil to the #2 cylinder if it was a coil problem or stayed in the #3 cylinder if it was a plug problem. However, instead of following any logic whatsoever, the problem completely moved to cylinder #1, a cylinder I hadn't yet made ANY changes to. My problem didn't make any sense whatsoever.

With new plugs in every cylinder now, it doesn't matter which coils I swap out; the misfire stays in cylinder #1. That's why I'm hoping that I'm not losing compression in cylinder #1, or its as simple as unclogging a stuck injector.

 

[Majoraslayer]

Oh, also the misfire code was the only code I had. Like I said, I have an OBD2 PC cable and software that lets me monitor the misfires live, so I was able to watch the misfires. Cylinder 1 doesn't seem to be firing at all; it counts all the way to 190 and restarts the count consistently.

 

[RayVoy]

Just trying to understand the sequence of events.

As I read your posts, you decided to check the compression.

Checked the compression with the engine running,

And then had mis-fires.

Is the above correct?

 

[Majoraslayer]

No, I had misfires first. Checking the compression was the last thing I did.

I just finished checking compression again, this time with the engine warm, all plugs removed, the fuel pump relay disconnected, and the throttle fully open while cranking. I got an exact reading of 170 on every cylinder with no leak-down.

Just to be sure, is cylinder #1 the cylinder closest to the firewall, or the closest to the accessories? I swapped coils and plugs on the cylinder closest to the accessories and still show a misfire at cylinder #1 according to live data. I'm showing a misfire HISTORY at cylinder #2, but the live data is only showing misfire at cylinder #1. I just drove it a little bit, and the code it threw was for cylinder #2.

First things first, I guess I should make sure my cylinder numbering is correct, then try swapping things around one more time. From front to back, does the engine refer to each cylinder by the firing order number, or cylinder numbering (i.e. front cylinder #1, next #2, etc.)?

 

[triz]

Check the coil pack on no.1 and make sure it's sitting properly on the plug. Sometimes the plastic trim where the wire runs just up front of the coil pack keeps it from sitting properly. I had to remove that plastic trim and then reinsert the coil pack. No more misfire.

 

[Majoraslayer]

Alright, so I took a different approach. It seems my software for reading it was buggy, so I decided to go the old-fashioned way and use the codes as a reference. I cleared the codes and went for a test drive; the code I was showing was P0302, which is a misfire on cylinder #2. So, I swapped the coils between #2 and #4, then cleared codes and test drove again. Sure enough, the new error was P0304, cylinder #4 misfire.

Thanks for all the help everyone!

 

[RayVoy]

First things first, I guess I should make sure my cylinder numbering is correct, then try swapping things around one more time. From front to back, does the engine refer to each cylinder by the firing order number, or cylinder numbering (i.e. front cylinder #1, next #2, etc.)?

The #1 is the cylinder closest to the front of the truck and they number 1 to 6 as you move front to rear. It is the physical position that is always referenced.

 

[Majoraslayer]

It was apparently my software that was showing a misfire in cylinder #2 as cylinder #1. It was odd because I was showing a misfire history for cylinder #2, but the live data was only showing misfires in cylinder #1. Going by the codes, I was able to confirm it was the #2 coil. I replaced the coil, and everything is running great now. Hopefully the other coils don't start failing as well; they're all the same age.