High RPM Cylinder 6 Misfire

[carabuser]

I'm completely uneducated on the N55 but my understanding is that it was developed by Bosch and not Continental like the MSD8x. So variable naming standards will vary hugely. I do have some random A2L files for various different N55 offerings but without knowing the naming it's hard to know what parameters do.

I have looked at the S55 FR and it does appear very similar to the layout for the N55. There is a parameter called "CDMD" with the description "Switch off code word DMD (EURO coding), CD..=0 -> no diagnosis". In the N55 ROM "7982780B" that parameter is at address 0x801B59BB, but as usual with leaked A2L files, there isn't one for your 98G0B.
That looks as close as any other parameter to being a straight up disable. There certainly very little in common between the N54 and N55 code. The S55 would be your closest relative for R&D.

Maybe contact Jake on here about it. He will undoubtedly have much more N55 info.

 

[houtan]

@carabuser , thank you.

It’s funny you say they are very different because we found a few tables leveraging N54 information in the past only to find out now it was probably complete luck! Ha ha. But we have luck leveraging, the F series tables, which are more similar like you mentioned, so that makes sense.

I will reach out to Jake, thanks for that recommendation. I have reached out to MHD multiple times in the past to try and get help on this issue but I’ve been told they don’t have the bandwidth as they have higher priority items to focus on. But maybe Jake can help!

 

[houtan]

I'm completely uneducated on the N55 but my understanding is that it was developed by Bosch and not Continental like the MSD8x. So variable naming standards will vary hugely. I do have some random A2L files for various different N55 offerings but without knowing the naming it's hard to know what parameters do.

I have looked at the S55 FR and it does appear very similar to the layout for the N55. There is a parameter called "CDMD" with the description "Switch off code word DMD (EURO coding), CD..=0 -> no diagnosis". In the N55 ROM "7982780B" that parameter is at address 0x801B59BB, but as usual with leaked A2L files, there isn't one for your 98G0B.
That looks as close as any other parameter to being a straight up disable. There certainly very little in common between the N54 and N55 code. The S55 would be your closest relative for R&D.

Maybe contact Jake on here about it. He will undoubtedly have much more N55 info.

So we have table cdmd. It’s a full disable or enable table, which is different than what was mentioned here where it can be disabled at a certain RPM. So i would still like to find the table by RPM, but I’m going to give the CDMD table a try to see if it actually fully disables misfire detection.

 

[houtan]

Just want to say thanks. With the help of a forum member in the N55 section, we found the misfire disable by rpm and gear and it works. I also tested CDMD which disables misfiring detection completely, and that also worked.

Log: https://www.spoolstreet.com/graphs/log-11-14-2023-22-16-22.31829/?series=3,12,14&zoom=-1,78

I also noticed another thread pop up discussing misfires and a proposed cause was discussed. Post 36 here: https://www.spoolstreet.com/threads/unsolvable-cyl-5-misfire-tried-everything.8974/page-2

Has anyone tried adjusting the “air mass threshold release KNK” table to see if that is the cause of your misfire?

 

[JohnDaviz]

Thank you. It would be nice to know why this even happens but my guess is we will never get an answer. In the meantime, we are going to look for that table for the E series N55.

If the issue is solved with disabling misfire detection above a certain RPM (for me it was above 6300) it is called phantom misfire.
In my case it came from the stock DMFW as it was completely out of spec when i changed it to the S55 FW.

 

[TOR COILS]

If the issue is solved with disabling misfire detection above a certain RPM (for me it was above 6300) it is called phantom misfire.
In my case it came from the stock DMFW as it was completely out of spec when i changed it to the S55 FW.

same

 

[135boost]

Sorry to mess up this too but knock detection on my knowledge is 2 stage and disabling it above some rpm takes only first stage away.

That second phase do all kind of wierdness like bank to bank superlean conditions.. on my tune it went to over 17:1 whatcit seives cause misfires.. and what trig misfire detection

 

[houtan]

If the issue is solved with disabling misfire detection above a certain RPM (for me it was above 6300) it is called phantom misfire.
In my case it came from the stock DMFW as it was completely out of spec when i changed it to the S55 FW.

Thanks. The misfire I am getting feels exactly the same as the phantom misfire due to the stock DCT flywheel. I already have an M2/3/4 flywheel installed.

The phantom misfire comes back when I start getting close to 600whp.

 

[135boost]

Thanks. The misfire I am getting feels exactly the same as the phantom misfire due to the stock DCT flywheel. I already have an M2/3/4 flywheel installed.

The phantom misfire comes back when I start getting close to 600whp.

There is variation on air mass calibrations, ve calibrations etc byt that second phase knock logic kicked in some 200+ times in my engine with trigger value some 400g/s air mass, wot, 6000rpm, and rich condition.
Every time there was indication only on misfire, every time bank afr jumped to 16.5-18 afr and if allowed recovery without restart and continue(with risk) wot, cylinder was wakening after 0.8sec window.. then continuing wot, it came back after some 1.3 sec...

 

[houtan]

There is variation on air mass calibrations, ve calibrations etc byt that second phase knock logic kicked in some 200+ times in my engine with trigger value some 400g/s air mass, wot, 6000rpm, and rich condition.
Every time there was indication only on misfire, every time bank afr jumped to 16.5-18 afr and if allowed recovery without restart and continue(with risk) wot, cylinder was wakening after 0.8sec window.. then continuing wot, it came back after some 1.3 sec...

I will admit, I don’t fully understand everything you’re saying, mostly because my knowledge is much less than yours.

Here is my latest log with misfire above 5k disabled and I dont appear to have the lambda fluctuations you mentioned. I am also e series n55. https://www.spoolstreet.com/graphs/log-11-14-2023-22-16-22.31829/?series=3,8,14,16,40&zoom=-1,78

 

[135boost]

You are on n54 pages and you have a n55 with valvetronic engine and different dme ??
Did i get it right?

 

[houtan]

You are on n54 pages and you have a n55 with valvetronic engine and different dme ??
Did i get it right?

Haha. Yes. Different DME, different motor, but our hpfp is the same!

We have leveraged information for the N54 for the last 10 or so years to help our platform progress. Already in this thread, we figured something out that we’ve been trying to do for the last two years with regard to misfire disable by gear.

There is not a lot of E-series N55, and we have much less aftermarket support , so we get a lot done on here.

 

[pbondar]

If the issue is solved with disabling misfire detection above a certain RPM (for me it was above 6300) it is called phantom misfire.
In my case it came from the stock DMFW as it was completely out of spec when i changed it to the S55 FW.

Is that the DCT M2/M4 DMFW….?

I have the Luk version which I believe is the OE supplier..

 

[135boost]

I didn't mean to be offensive, just try to understand the issue and sure, n54 is more common very high horsepower base than n55 so i get your poit to write it here.

Not knowing n55 dme logic, there can be hidden so wierd things not to be found without dme fr and stupid or insane amount of digging and tryalerror methods to figure out that misfire thing.

Those wierd things is something they at bmw must use drugs to invent and because total map and checkbox quantity on a dme can exceed 100 000 variables or maps, how to resolve everything.

Fr of 2 megs bin is 14000 pages of pdf, b58 has 8 megs bin file..

So at the end, hope you find that if it's software related because different dme is different maps and locations and for sure valvetronic is making that even harder.

I keep my fingers crossed and follow this progress

 

[wheela]

Can someone please show what tables you use to turn off misfire detection? As @wheela mentioned I have a similar issue when I creek up to 600whp on my e series DCT n55.

I have tried to disable all cel but it still happens so my guess is the table is not correct. I have also softened cyl 4, 5, 6 knock detection at high load and also average knock detection for those cylinders. Same issue.

I don’t have an actual misfire, it just throws the code which on my dme is 65536 in the log.

@houtan have you mentioned in the past that m235i n55 guys don't have this same phantom misfire as you? I just came across this post - turns out m135i and m235i n55's have a different crank than the n55's in regular 135i or m40i n55's.

This, along with the fact that changing fly wheels can buy you some more room may support my crank twist/harmonic theory as to why the n55 powered DCT 135i's run into this phantom misfire around 600whp... Different crank, or different fly wheel, different twist/harmonics characteristics to mess with the crank position sensor readings ...