30EE Superknock Codes / bank AFR 1 goes super lean / Running E60 blend

[mfish]

I just recently starting running E60 in my car. I have brand new Bosch spark pugs torqued to proper specs, Bosch pre-cat O2 sensors about 2 months old, fuel-it stage 2 pump that's maybe 2 years old and has less than 20K miles on it, latest version Eldor coils that are about 1 year old with about 10K miles on them, OEM BMW fuel injectors replaced about 1 year ago with about 10K miles on them so they're probably index 11 or 12. They should be coded correctly as the dealer installed them. All vaccum lines replaced about 6 months ago and every hose clamp on the intake tract is nice and snug. HPFP was replaced by previous owner under warranty maybe 6 or 7 years ago.

So hardware should be tip top unless something failed prematurely. I've also tried swapping coils around with no luck either. We've been through several revised custom maps with no success either.

I get heavy timing corrections mostly on cylinders 1 and cylinders 5 but sometimes other cylinders too and occasionally at the top of the pull.

Then about 2 weeks ago I started getting misfires on cylinders 1 and 2. Shortly thereafter I started getting superknock codes on cylinders 2 and 3. I've also noticed since running Ethanol (but this wasn't the case on 93 octane) that bank 1 is always a hair (0.1 - 0.3) richer than bank 2.

And sometimes my bank 1 AFR goes ridiculously lean and the pump pressures sky rocket when this happens.

Has anyone had similar experiences and fixed it. I'm just perplexed that with all new hardware this is happening and I'd rather not just start wasting money and randomly throwing parts at it.

Logs below. Please help!!

3 upshifts with corrections - http://www.datazap.me/u/mfish123/3-upshifts-33
super lean AFR during pull - http://www.datazap.me/u/mfish123/vib...14-17-18-22-23
super lean after upshift 1 - http://www.datazap.me/u/mfish123/ups...ata=5-17-18-23
super lean after upshift 2 - http://www.datazap.me/u/mfish123/lea...14-17-18-22-23
3 previous 93 octane logs - http://www.datazap.me/u/mfish123/2-upshifts-1-3rd-gear-0?log=0&data=18-19-24

[Sgop335]

Strange one. Clearly bank 1 is getting very lean. Could be o2 sensor? Tune? Another thing havent seen stfts =0 at boost...
could it be bad ethanol?

[notmpwr]

Usually when the O2 sensor goes bad it will default to 14.6, full rich or full lean. If it's making corrections and is responding quickly it's probably ok. I would say the full lean could be caused by very rich condition but since the widebands are lsu 4.9 and not 4.2 I'm not sure how the react the full rich conditions and the fact that you are getting knock associated with it makes it sound like it is indeed lean.

What plugs (part number) and what are the gapped to?

[subaru335i]

Daaamn thats a lot of knock retard. I would stop doing pulls until you figure it out...

[notmpwr]

I would do a plug pull with a fresh set of plugs to get a better idea of what's going on.

[mfish]

Strange one. Clearly bank 1 is getting very lean. Could be o2 sensor? Tune? Another thing havent seen stfts =0 at boost...
could it be bad ethanol?

I'm baffled too. It's been like this for the last 2 or 3 fill ups. The one gas station near me with E85 isn't close. So I've been filling up for five gallon gas cans with it and storing it in my shed and then using it the next time I fill up with some 93 octane. It only sits in my shed for 1 to 2 weeks tops. Could that be the smoking gun or is that too short of a time period for it to start going bad?

[mfish]

I would do a plug pull with a fresh set of plugs to get a better idea of what's going on.

I appreciate the advice but I literally changed all six plugs less than six weeks ago

[mfish]

Usually when the O2 sensor goes bad it will default to 14.6, full rich or full lean. If it's making corrections and is responding quickly it's probably ok. I would say the full lean could be caused by very rich condition but since the widebands are lsu 4.9 and not 4.2 I'm not sure how the react the full rich conditions and the fact that you are getting knock associated with it makes it sound like it is indeed lean.

What plugs (part number) and what are the gapped to?

Using the OEM Bosch ZGR6STE2, not gapped. Here is the exact link:
https://www.amazon.com/dp/B00804F080...a_ZqWwzbVZMD6R

[mfish]

Daaamn thats a lot of knock retard. I would stop doing pulls until you figure it out...

I know right. I'm getting nervous but the only way to see if I fix it is to go WOT.

[RSL]

Fuel pressures rise and it goes lean nearly the same time, hard to tell what is causing what or if they're all in relation to something else happening (super knock code triggering a safety). Everything goes flat when the fit hits the shan there: torque output, etc. and STFT go 0 on open loop.

Misfires cause lean AFRs on these, so still thinking it may be plug related...like you may be long past the point of using stock heat range plugs. If the tips (all 3 on stock plugs) aren't in the correct heat range (suspect too hot), you'll have issues = misfires = lean reading AFR = DME intervening. Like was said, pull the plugs and read them. You may find they're running too hot and bank 1 would probably be the first to indicate it (and misfire), but even if they don't show signs, I'd still replace them with at least 1 range colder

Have you done a ECA on that gas? If it's not plugs, consider lowering the E85 content and backing off the timing up top 1-2° and see if it stops. My $.02 is on plugs though

[GermanBoy]

Did you Tune it by yourself?

[mfish]

Fuel pressures rise and it goes lean nearly the same time, hard to tell what is causing what or if they're all in relation to something else happening (super knock code triggering a safety). Everything goes flat when the fit hits the shan there: torque output, etc. and STFT go 0 on open loop.

Misfires cause lean AFRs on these, so still thinking it may be plug related...like you may be long past the point of using stock heat range plugs. If the tips (all 3 on stock plugs) aren't in the correct heat range (suspect too hot), you'll have issues = misfires = lean reading AFR = DME intervening. Like was said, pull the plugs and read them. You may find they're running too hot and bank 1 would probably be the first to indicate it (and misfire), but even if they don't show signs, I'd still replace them with at least 1 range colder

Have you done a ECA on that gas? If it's not plugs, consider lowering the E85 content and backing off the timing up top 1-2° and see if it stops. My $.02 is on plugs though

I'm not sure what came first. Is the superknock the initial cause in this chain of events? Or does bank 1 running persistently richer than bank 2 cause the DME to try to lean out bank 1 out a little, which then in turn causes the superknock. Either way, assuming the DME is purposely shutting down the cylinders due to superknock, I believe that is what causes the super lean condition and if the injector basically shuts down and the fuel has no where to go that's why the fuel pressures then shoot up.

The plugs look good. Below are some pics. I'm not opposed to trying colder plugs but it's odd that only a couple of cylinders correct heavily and only a couple throw the superknock codes. Odd that all cylinders don't exhibit similar behavior but like you said maybe there is something about bank 1 that would cause it to exhibit symptoms first.

I haven't done ECA on the gas but that would be a good test to rule out variables - I'll order an analyzer kit. I have tried slightly lower E85 content and lowering the timing a bit but neither helped much.

I'm still wondering why bank 1 always runs a tad richer regardless of conditions. It even runs a bit richer when idling. And that this is also the bank that goes super lean when I have the issue. The pre-cat O2 sensors are only 1 month old but who knows maybe one or both was defective out of the box or the mechanic damaged them on the install??

[mfish]

I just swapped pre-cat O2 sensor harnesses (same tune / same tank of gas / no other variables changed) and now bank 2 AFR is reading slightly richer (since it's being fed from the O2 sensor that is actually on bank 1). The wire isn't long enough from the sensor on bank 2 (in the exhaust) to safely route it behind the engine so I don't want to drive and log like this but I wouldn't be surprised if I'd now get the same behavior on bank 2.

In any event I'm going to get a brand new sensor installed on bank 1. Maybe it was defective or the mechanic damaged it slightly on install?

Does this sound like a plausible fix?

O2 sensor in original position: http://www.datazap.me/u/mfish123/o2-...&data=15-16-23

O2 sensor harnesses reversed: http://www.datazap.me/u/mfish123/idl...&data=15-16-23

[RSL]

I'm not sure what came first. Is the superknock the initial cause in this chain of events? Or does bank 1 running persistently richer than bank 2 cause the DME to try to lean out bank 1 out a little, which then in turn causes the superknock. Either way, assuming the DME is purposely shutting down the cylinders due to superknock, I believe that is what causes the super lean condition and if the injector basically shuts down and the fuel has no where to go that's why the fuel pressures then shoot up.

The plugs look good. Below are some pics. I'm not opposed to trying colder plugs but it's odd that only a couple of cylinders correct heavily and only a couple throw the superknock codes. Odd that all cylinders don't exhibit similar behavior but like you said maybe there is something about bank 1 that would cause it to exhibit symptoms first.

I haven't done ECA on the gas but that would be a good test to rule out variables - I'll order an analyzer kit. I have tried slightly lower E85 content and lowering the timing a bit but neither helped much.

I'm still wondering why bank 1 always runs a tad richer regardless of conditions. It even runs a bit richer when idling. And that this is also the bank that goes super lean when I have the issue. The pre-cat O2 sensors are only 1 month old but who knows maybe one or both was defective out of the box or the mechanic damaged them on the install??

It's a pretty safe bet the super knock code is causing limp/open loop and I can't remotely imagine the DME would trims AFR to the point of knocking.

Hard to tell from the pics, but nothing stands out as obviously wrong. There may be other issues, but I think you're on the wrong plugs regardless. They can work in a pretty wide range of power depending on combustion conditions, but once you're out of the range they're comfortable in, plugs are completely unforgiving. It's happening at high RPM where there is significantly less margin since cycles are so short. The time between peak temperatures are closer together and the time they have to cool is shorter.

Not odd at all, bank 1 and particularly cyl 1-2 will probably always let you know first. Other things can impact it, but if something is happening consistently in cyl 1 or 2 and hardly or nowhere else, I pay attention.

Always start with what you did last before the issue popped up. If this all started right after or not long after you moved to E60, there aren't too many things to rule out. The tune needs work for the mix, the E85 isn't up to snuff for the tune or some combination of both tune and fuel isn't working well with hardware (duh ). I'd start with the tune/gas/plugs first since they're easy/cheap.

I just swapped pre-cat O2 sensor harnesses (same tune / same tank of gas / no other variables changed) and now bank 2 AFR is reading slightly richer (since it's being fed from the O2 sensor that is actually on bank 1). The wire isn't long enough from the sensor on bank 2 (in the exhaust) to safely route it behind the engine so I don't want to drive and log like this but I wouldn't be surprised if I'd now get the same behavior on bank 2.

In any event I'm going to get a brand new sensor installed on bank 1. Maybe it was defective or the mechanic damaged it slightly on install?

Does this sound like a plausible fix?

O2 sensor in original position: http://www.datazap.me/u/mfish123/o2-...&data=15-16-23

O2 sensor harnesses reversed: http://www.datazap.me/u/mfish123/idl...&data=15-16-23

I suppose it's possible that sensor reads slightly leaner, but it certainly doesn't look damaged. It is reading and trimming. The only thing you've done is tell the DME to trim bank 2 based on bank 1 conditions and vice versa, which isn't a good idea. There are separate knock sensors for the banks, swapping O2 sensors isn't going to effect them, and if an issue comes up, the DME will take fueling actions on the wrong bank. Put them back in their correct orientation.

Don't just throw parts at it. Go back and look at what changed when the issue started.

[notmpwr]

I would switch to some 5992 ngks regardless of what the issue may be.

[mfish]

It's a pretty safe bet the super knock code is causing limp/open loop and I can't remotely imagine the DME would trims AFR to the point of knocking.

Hard to tell from the pics, but nothing stands out as obviously wrong. There may be other issues, but I think you're on the wrong plugs regardless. They can work in a pretty wide range of power depending on combustion conditions, but once you're out of the range they're comfortable in, plugs are completely unforgiving. It's happening at high RPM where there is significantly less margin since cycles are so short. The time between peak temperatures are closer together and the time they have to cool is shorter.

Not odd at all, bank 1 and particularly cyl 1-2 will probably always let you know first. Other things can impact it, but if something is happening consistently in cyl 1 or 2 and hardly or nowhere else, I pay attention.

Always start with what you did last before the issue popped up. If this all started right after or not long after you moved to E60, there aren't too many things to rule out. The tune needs work for the mix, the E85 isn't up to snuff for the tune or some combination of both tune and fuel isn't working well with hardware (duh ). I'd start with the tune/gas/plugs first since they're easy/cheap.

I suppose it's possible that sensor reads slightly leaner, but it certainly doesn't look damaged. It is reading and trimming. The only thing you've done is tell the DME to trim bank 2 based on bank 1 conditions and vice versa, which isn't a good idea. There are separate knock sensors for the banks, swapping O2 sensors isn't going to effect them, and if an issue comes up, the DME will take fueling actions on the wrong bank. Put them back in their correct orientation.

Don't just throw parts at it. Go back and look at what changed when the issue started.

Thanks for the detailed feedback. I ordered a set of NGK's. Hoping to get them today or tomorrow. I'll properly/carefully gap them and install and see if that does the trick. If not, I'll move on to testing the E content in the gas I'm using.

I did switch the O2 sensor harnesses back to the original / correct banks. I was more less just trying to isolate the hardware to see if one sensor inherently read a tad richer than the other which seems like it does. But hopefully the DME can compensate with trims or the readings are close enough that it's not having any negative impact.

In the meantime, I threw in my old coils which seemed to be working fine at the time I replaced them (I just replaced them as preventative maintenance) and like I said, I already swapped around the new coils. I got the same behavior on the same cylinders with the old coils installed and the new ones swapped around so I'm pretty sure I can rule out the coils.

Really hoping the NGK's do the trick. Fingers crossed!

[shushikiary]

When you life off the throttle and let it go to full lean does bank one take longer to get to full lean than bank 2?

Likely your bank1 is slightly richer because you have a minor leaking fuel injector in that bank.

The fact that your STFT's are 0 when the AFR gets that high is very telling that the DME is likely no longer in closed loop at that point, and also leaning out the system due to detecting super knock.

If your plug tips are getting too hot they can easily cause super knock as they are the hot spot for the detonation to start at, so I think going to the NGK's is a good idea, or maybe even the NGK's that are two steps colder if you really want to be sure.

If it was me the next thing I'd be doing is sticking a bore scope down that cyl just to have a looksy.

[mfish]

When you life off the throttle and let it go to full lean does bank one take longer to get to full lean than bank 2?

Likely your bank1 is slightly richer because you have a minor leaking fuel injector in that bank.

The fact that your STFT's are 0 when the AFR gets that high is very telling that the DME is likely no longer in closed loop at that point, and also leaning out the system due to detecting super knock.

If your plug tips are getting too hot they can easily cause super knock as they are the hot spot for the detonation to start at, so I think going to the NGK's is a good idea, or maybe even the NGK's that are two steps colder if you really want to be sure.

If it was me the next thing I'd be doing is sticking a bore scope down that cyl just to have a looksy.

Thanks for the advice. When I get the superknock, which occurs on cylinders 2 and 3, typically only bank 1 goes full lean. Occasionally bank 2 will go slight lean but not nearly as much as bank 1. See: http://www.datazap.me/u/mfish123/vib...&data=17-18-23

When I switch the O2 sensor harness then bank 2 read slightly richer (it was pulling the input from the o2 sensor that physically remained in the bank 1 exhaust. So unless I'm missing something it seems to me that the O2 sensor physically in bank 1 reads slightly richer. And therefore I don't think it's pointing to a slightly leaking injector.

I agree wtih your take STFT's going to 0. I could be wrong but I think it's just straight up shutting down the cylinders effecting and completely stopping spraying fuel. The fuel pressure goes through the roof at this time since the injecotrs are probably on lock down and the fuel has nowhere to go. Who knows- maybe the DME stops sending voltage to the coils on those cylinders too.

No one local had the NGK 95770 (replacke part # for 5992). I did in store pick up and hoping they come in tomorrow. If it doesn't alleviate the issue completely but helps (e.g. no more super knock codes and timing corrections are reduced but not eliminated), I'll try going an additional step colder. Do you know the part # for the recommended 2 steps colder than stock plugs and what the correct gap on them is?

I'm really hoping there is no cylinder or piston damage but if all else fails I will ultimately have to have a look in there and/or do a compression test.

[shushikiary]

the plugs part number: NGK 97506

When I got them they had a gap of .028 on them, and because I'm using the precision race works coils I left them at that. I'd gap them down to .022 or .018 depending on your power level on stock coils.

[RSL]

I run .030" on stock ignition (S55 plugs @ stock gap, Eldor coils) at about 500whp (RR600s). I'm either lucky or people don't even bother to try anything other than the interwebz tells them. Run the widest gap you can without misfires, simple. Don't need them, but I'm semi-tempted on the PRI coils just to see if I could run .035"

[mfish]

the plugs part number: NGK 97506

When I got them they had a gap of .028 on them, and because I'm using the precision race works coils I left them at that. I'd gap them down to .022 or .018 depending on your power level on stock coils.

@shushikiary Looks like the one step colder NGK current part # is 95770 (the ones I got) and those replaced 5992.

I'm seeing NGK 97506 are the 2 step colder plugs - do you think I could further benefit from these at my power level?

[mfish]

I run .030" on stock ignition (S55 plugs @ stock gap, Eldor coils) at about 500whp (RR600s). I'm either lucky or people don't even bother to try anything other than the interwebz tells them. Run the widest gap you can without misfires, simple. Don't need them, but I'm semi-tempted on the PRI coils just to see if I could run .035"

@RSL - I see you're at a higher power level than me with upgraded turbos. As far as the 2 step colder S55 plugs did you at one point try the NGK 5992 / 95770 and find less than ideal performance and therefore went with a 2 step colder plug.

I know there are varying opinions. Just trying to figure out at what level is 1 step colder "not enough"

[mfish]

@notmpwr @RSL @shushikiary - Thank you all for the suggestions on the colder plugs! I went with the NGK 95770 and no more misfires or superknock codes! I meticulously gapped with feeler guages and used the tool on a wire guage to properly move the electrode. Plugs properly torqued with a torque wrench. And trust me I beat her up good on the highway with back to back pulls to put her through her paces.

I am still getting some fairly heavy corrections but granted this tune is still in progress and got put on hold until I could sort out the misfires/superknock. We're thinking the low pressure pump dipping into the 50's is contributing to that so it's a work in progress.

I have the car from hell. An IJE0S, AT xdrive. If there is a gremlin I have it / had it - torque limit 4 in multiple gears especially 5th, the car always had crappy post shift timing, it reacts oddly and in different ways to some table changes vs. other easier to tune cars, etc. Things are improving with is good.

See log below. More tuning tweaks needed and it was hot out and this was after multiple pulls with high IAT. Moving in the right direction:

http://www.datazap.me/u/mfish123/cle...12-13-14-23-27

[RSL]

@RSL - I see you're at a higher power level than me with upgraded turbos. As far as the 2 step colder S55 plugs did you at one point try the NGK 5992 / 95770 and find less than ideal performance and therefore went with a 2 step colder plug.

I know there are varying opinions. Just trying to figure out at what level is 1 step colder "not enough"

S55 plugs are only 1 range colder. Bosch heat range numbering is backwards from most, so Bosch 5 = 7 on NGK. If the NGK 7s are working well for you, leave them be now until they're not. Glad that resolved the issue.

Edit: now ECA that gas

[RSL]

Quickly looked at the log, timing corrections are post-shift and none through the pull. That's probably a tuning issue, not gas.