Random misfire MIL- Solved?

[Johnny O]

Okay gang, here’s my noobiness shining again…
You may remember I have been chasing an intermittent “General Random Missfire” with no other codes for quite some time. I replaced the usual suspects and those when a targeted code was available. So cops, plugs, injectors, wires, etc.

Then today months of troubleshooting notes suddenly came together when a new issue appeared.

I’m curious if any of you draw the same conclusion. What follows is a transcript of my “lab notes”. Hopefully will help me now and help others later.

Feel free to TLDR and skip to end.

Original Codes at 168k miles (time of purchase) were general random misfire. Misfire on #4. Misfire on #6.
Replaced plugs with proper NGKs. Observed #4 and #6 COPs showed carbon from arcing. Tested with knock off COP from Amazon. Confirmed faulty COPs. Replaced all with Bosch. All Mils cleared and stayed gone. Went on Great Basin Excursion. After about 2500 miles, #6 MIL reappeared along with General Random Misfire. Fuel Mileage dropped to 10-12mpg.
Treated with Techron twice and all Mils cleared. 150 miles later rough idle and P0300 returns intermittently. Suspect #6 bad injector. Tested with cheapo Amazon injector. Got codes to jump. Installed 6 Bosch injectors. All codes cleared and stayed gone 1200 miles. At 175k P0300 returns and auto clears intermittently, but rough idle not present. Code appears at speeds above 30mph but clears when parked overnight. Treated tank of 87 ETOH free gas with 8 oz of Marvel Mystery Oil. Mil self cleared after a quarter tank. At half tank mil only appeared at freeway speeds. No codes in memory. ?!?!?
Noticed fuel back splash when fueling up. Odd, assumed faulty gas pump. MIL only appears at random at speeds over 40mph and clears when stopped.
Tried driving with Reader on in real-time mode, of course get no MIL when reader is plugged in. Tried fancy expensive modern fuel with fancy snake oil stuff in it. Ghost mil remains.
Gave up chasing it. Assumed is engine wear. Contemplating Homer Simpson Fix.
Weather changed and got near freezing. Ran tank down to fumes, then filled with 87 ETOH free gas and 4oz of MMO. Backsplash at full tank again from different pump…

Epiphany! Should have noticed.

Now suspect some sort of check valve or vapor thingy has failed or bad atmospheric vent is full of desert and muck.

Problem is I am not sure what sort of gizmos are on Bert as research shows a number of possibilities.

Gurus please guide me? If no TLDR. Service manuals and parts fiche not too clear.

Currently have Jeep parked for suspension work, so figure is good time to fix other stuff too. No hurry to get back on road. Can walk to work. Currently have fuel tank however.

 

[Elliott]

Perhaps this is what you are looking for?

HOW TO: fix the KJ fuel burp plague!

Yesterday evening, fueled by turkey, taters, cranberry sauce, and delicious homemade pies of every sort, I finally tackled the fuel burp issue that has plagued my KJ since I bought it. One Liberty write-up has been done, but it was pretty abstract and didn't get much attention. So, since...

www.jeepkj.com

 

[Johnny O]

Perhaps this is what you are looking for?

HOW TO: fix the KJ fuel burp plague!

Yesterday evening, fueled by turkey, taters, cranberry sauce, and delicious homemade pies of every sort, I finally tackled the fuel burp issue that has plagued my KJ since I bought it. One Liberty write-up has been done, but it was pretty abstract and didn't get much attention. So, since...

www.jeepkj.com

I had read this before out of boredom. Ironically the two burps described are the only ones I have experienced. Weather they are linked to my issues with the MIL, or my use of 4oz of MMO, or me never using ETOH gas..or more likely the recent drop in ambient temperatures…

 

[Ksat]

Your jumping misfire code is probably due to a lean (or possibly rich) condition on bank 2 (if this is the 3.7), as both cyls 4 & 6 are on that bank. You can change coils or throw liquid snake oil in the tank all day long and that's not going to help. Hook up a live data reading scanner and look at the fuel trim PIDs to see what they look like.

A lot of us have gas spitting out when we refill. Mine as well, I never looked into it, but I suspect its a faulty spit-back valve. Whatever the cause, I doubt that has anything to do with your problem.

 

[Johnny O]

Your jumping misfire code is probably due to a lean (or possibly rich) condition on bank 2 (if this is the 3.7), as both cyls 4 & 6 are on that bank. You can change coils or throw liquid snake oil in the tank all day long and that's not going to help. Hook up a live data reading scanner and look at the fuel trim PIDs to see what they look like.

A lot of us have gas spitting out when we refill. Mine as well, I never looked into it, but I suspect its a faulty spit-back valve. Whatever the cause, I doubt that has anything to do with your problem.

Ok cool. I can attest to something running rich just by the smell of the exhaust. OP is just me wondering if there was some sort of a pressure issue with fuel delivery. Hence why I wondered if the new to me “burp” was a new symptom. That part of the post is only on the last two fill ups.

Anyway, as I am new to fuel injection, what controls the fuel air mix issues on just one bank? I’m not very clear on how that end of the system actually works. As I read about it, I’m even more confused.

 

[Ksat]

"Anyway, as I am new to fuel injection, what controls the fuel air mix issues on just one bank? I’m not very clear on how that end of the system actually works. As I read about it, I’m even more confused."

There's a pretty steep learning curve, LabRat, to understanding how computer controlled systems work. If you have the time/inclination, there are some top notch videos on the topic on YouTube by authors such as "Go Tech," "Scanner Danner," and "The Trainer." If you decide to watch any of them, be sure to go to their respective channels and search their videos from oldest to newest, that way you're not jumping into advanced stuff right away.

To help answer your question, the 3.7 is in a "V" configuration. There's 3 cylinders front to back on the driver's side (bank 1) and 3 on the passenger (bank 2). Each bank has 3 respective injectors on them. The computer controls bank 1 injectors as one group and treats bank 2 as another group. Referring to the image I uploaded, that means injector 1, 3 and 5 are all sent the same signal from the computer on how much fuel they should let through and 2, 4 and 6 are sent another signal on how much fuel they should allow through. What signal the computer sends each bank of injectors depends on input it gets from various sensors.

Now, a misfire basically means a cylinder isn't firing. Most people seem to automatically assume the main (if not only) reason for misfire is something electrical- a coil has failed or spark plug is fouled- but the truth there are other causes for that condition as well, such as a lean condition (too little fuel in a cylinder). If, for some reason, the computer commands a given injector bank to reduce fuel flow too much, you will get misfires on that bank. In that situation, misfires will be transient and hop around from cylinder to cylinder on that given side. The computer fights a losing battle as it tries to calculate which particular cylinder to blame when the truth is it's none of them specifically.

When you said you had a misfire on both 4 & 6 and had already changed the coils/plugs, I noticed those cylinders were on the same bank, so I thought It might be fuel control problem. If you had a scanner that could read what's called live data, you would be able to see the amount of fuel the computer is telling each injector bank to throw in. If that amount is off (too little or too great, even) you would know there's a problem that needs to be addressed.

 

[DadOSix]

To add just a bit —-

If you smell rich at the exhaust, you are dumping excess fuel into the engine.

Excess fuel either goes out the exhaust, or ends up in the crankcase. Does your oil smell of fuel? Oil level rising on the stick?

FI is computer controlled. So it relies on a bunch of data from map, iat, temp, o2 front and rear, right and left.

Possible culprits involve all of the above, including a leaky / stuck / dirty injector.

Without a higher end scanner to look at the real-time PIDs, you are going to spend a lot of time chasing this unicorn.

You may have it narrowed a bit with the codes pointing at bank 2.

But - before throwing more parts at Bert, find a buddy / shop with the right scanner to help your diagnostic.

 

[Johnny O]

Thanks guys.

I spent the weekend studying EFI in these things. Your posts really helped! The good news is I do have a code reader that lets me do real time reads. Wasn’t quite sure what to do with the data I was seeing.

At the start there was both bad plugs and COPs. I’m in ******** troubleshooting mode now. Had to eliminate the easy stuff first. Both posts gave me some more information to fill in the gaps as I figure this out.

I’ll check back in on this thread as I discover more

 

[Ksat]

If by "real time reads" you mean your scanner can give you live data, post all the parameters here and I'll look it over.

 

[Johnny O]

Will do this weekend, thanks!

 

[RABustaman]

To add just a bit —-

If you smell rich at the exhaust, you are dumping excess fuel into the engine.

Excess fuel either goes out the exhaust, or ends up in the crankcase. Does your oil smell of fuel? Oil level rising on the stick?

FI is computer controlled. So it relies on a bunch of data from map, iat, temp, o2 front and rear, right and left.

Possible culprits involve all of the above, including a leaky / stuck / dirty injector.

Without a higher end scanner to look at the real-time PIDs, you are going to spend a lot of time chasing this unicorn.

You may have it narrowed a bit with the codes pointing at bank 2.

But - before throwing more parts at Bert, find a buddy / shop with the right scanner to help your diagnostic.

Wow, thank you! I was just about to post about the exact same (occasional) multiple misfire issue. I need a better scanner or a mechanic who knows. So hard to find the latter lately!

 

[RABustaman]

There's a pretty steep learning curve, LabRat, to understanding how computer controlled systems work. If you have the time/inclination, there are some top notch videos on the topic on YouTube by authors such as "Go Tech," "Scanner Danner," and "The Trainer." If you decide to watch any of them, be sure to go to their respective channels and search their videos from oldest to newest, that way you're not jumping into advanced stuff right away.

To help answer your question, the 3.7 is in a "V" configuration. There's 3 cylinders front to back on the driver's side (bank 1) and 3 on the passenger (bank 2). Each bank has 3 respective injectors on them. The computer controls bank 1 injectors as one group and treats bank 2 as another group. Referring to the image I uploaded, that means injector 1, 3 and 5 are all sent the same signal from the computer on how much fuel they should let through and 2, 4 and 6 are sent another signal on how much fuel they should allow through. What signal the computer sends each bank of injectors depends on input it gets from various sensors.

Now, a misfire basically means a cylinder isn't firing. Most people seem to automatically assume the main (if not only) reason for misfire is something electrical- a coil has failed or spark plug is fouled- but the truth there are other causes for that condition as well, such as a lean condition (too little fuel in a cylinder). If, for some reason, the computer commands a given injector bank to reduce fuel flow too much, you will get misfires on that bank. In that situation, misfires will be transient and hop around from cylinder to cylinder on that given side. The computer fights a losing battle as it tries to calculate which particular cylinder to blame when the truth is it's none of them specifically.

When you said you had a misfire on both 4 & 6 and had already changed the coils/plugs, I noticed those cylinders were on the same bank, so I thought It might be fuel control problem. If you had a scanner that could read what's called live data, you would be able to see the amount of fuel the computer is telling each injector bank to throw in. If that amount is off (too little or too great, even) you would know there's a problem that needs to be addressed.

Wow, thank you! I was just about to post about the exact same (occasional) multiple misfire issue. I need a better scanner or a mechanic who knows. So hard to find the latter lately!

 

[Johnny O]

Since all the top end stuff is new, I’m Easter egging O2 sensors and the like.

This issue has become vendetta!

 

[RABustaman]

Since all the top end stuff is new, I’m Easter egging O2 sensors and the like.

This issue has become vendetta!

HaHa! Easter egging O2s, Hahaha!

 

[Ksat]

Wow, thank you! I was just about to post about the exact same (occasional) multiple misfire issue. I need a better scanner or a mechanic who knows. So hard to find the latter lately

There are a number of Android and iOS apps that will allow you to use your mobile device to view live data info (and more) from your vehicle. You just need to purchase a separate dongle that plugs into the OBD 2 port that will work with them (usually listed as ELM 327 compatible). Many of the apps are free or low cost (Torque and Torque Pro being among the more popular) and the dongle is usually around $20.

If you wanted to buy a dedicated scanner that will read live data and mode 6, you can find them for probably around $60 or so to start. I've had good luck with Autel brand and, conversely, have heard a lot of bad things about the Innova brand.

 

[RABustaman]

There are a number of Android and iOS apps that will allow you to use your mobile device to view live data info (and more) from your vehicle. You just need to purchase a separate dongle that plugs into the OBD 2 port that will work with them (usually listed as ELM 327 compatible). Many of the apps are free or low cost (Torque and Torque Pro being among the more popular) and the dongle is usually around $20.

If you wanted to buy a dedicated scanner that will read live data and mode 6, you can find them for probably around $60 or so to start. I've had good luck with Autel brand and, conversely, have heard a lot of bad things about the Innova brand.

I have an innova. It's a few years old now, most of the options are not out of the box usable. I have to buy something. Since I'm not sure I don't buy it. I'll check out the phone apps and the Autel Thanks!

 

[Johnny O]

Sooo…no abnormal data. Been to three shops. Still get a random, self clearing “random multiple misfire” MIL. No solution. Only symptom is exhaust smells a tad rich only when the mil is present.
Honestly kind of gave up on it.
Then the cold weather arrived. MIL self cleared. No data in the history. Exhaust doesn’t smell of gas.

Idle is fine. Runs great.

Drove down to the VA hospital via surface streets and lo and behold, once fully warmed up the MIL reappeared. Parked for an hour it cooled down quick due to sub zero weather. Get back in and start her up and the code clears…

Hmmm.

Drove around a bit more and once warmed up. it reappears and the fuel smell returns to the exhaust but idle is not noticeably rough.

Tried to see real-time data on my reader and it the Mil auto cleared and was no history data…

Methinks my ghost is a short that only grounds out when all the metal is hot, or a sensor that prefers to work cold.

Wondering if it is the one on the exhaust closest to the aft end of the forward prop shaft.

 

[Ksat]

Unplug both O2 sensor electrical connectors and see if anything changes.

 

[Johnny O]

Unplug both O2 sensor electrical connectors and see if anything changes.

Yep, as soon as we get a break from the freezing fog.

 

[edekok90]

Johnny 0,

Did you ever get the issue resolved? If so, what was your issue. I live in NE too and have similar problems with it be cold and running fine to warm and having misfires.