[OpenFlash]

Hi guys,
There is still some talk about Ignition advance flatlining on 6AT cars. We've always felt that there is a mechanical reason for this (or from running too much boost/too little octane). And not something software-induced. Some have argued that this problem is inherent with the newer BMW DME software. Below is data that suggests otherwise.

This morning, I recorded a bunch of back-to-back 3rd to 4th gear runs in Mexico. Running 17-18psi with 0% ignition correction. On 91oct+meth. So basically, a pretty aggressive tune with very little rest between runs. Over the course of a few minutes, I managed to get 16 3rd-4th gear pulls in. Sometimes manually upshifting at ~6500rpm. Other times letting it upshift automatically at 7000rpm. Out of those 16 runs, timing didn't flatline (go to 0deg) once in 4th gear. In fact, in 12 of those 16 runs, it went to max timing (12.5 deg). And in the other 4 runs where it ran into knock correction (no doubt due to the back to back pulls), timing dropped to 6-8 deg.

Car Details:
2007 6AT Coupe
MSD81 with newest (3 week old) DME software
New Injectors (replaced 3 weeks ago)
Fresh Plugs
New Coils
FBO with PWM Meth
Procede 3-17 Stg 3 Maps (with Dynamic Meth Targeting)









In the past, this car has had episodes of ignition advance flatlining during the 3rd-4th gear upshift. Since then, it has had it's injectors replaced as the original 6yr old injectors leaked and had compromised compression seals. The DME software was also updated and we made sure that all the new injectors were coded properly (thanks to East Bay BMW). As many know, ignition advance flatlining become more common after a recent BMW software update and fuel injector recall. I suspect those who didn't qualify for free replacement injectors (like this test car) and just got the software update alone, didn't operate as well as it should have. Of course, the worn compression seals on the original injector didn't help either.

It's also worth noting that our single turbo shop car also had it's original 6yr old injectors replaced recently. Prior to the replacement, the car didn't exhibit any obvious issues. Idle was smooth, no misfire, etc,. But there was a AFR target deviated between bank1 and bank2. And the car would trigger lambda trimming codes from time to time. It was also down on compression by 20psi on 2 cylinders (due to worn injector compression seal). Replacing the injectors (at our cost unfortunately since our early build 2006 car didn't qualify for the free injector recall), solved all of those issues and restored 60whp at 18psi of boost (compression returned!). Post NLS shift ignition timing was also more consistent. Cold start was also improved.

So for all those early build 2007 cars that DIDN'T qualify for the injector replacement, I suggest biting the bullet and getting them replaced. Your car will drive much better. And for those who only go half of your injectors replaced, we recommend replacing the other half as well. It seems as if BMW were pretty stingy with replacing injectors that were known to be faulty. Their goal was to fix the issue "well enough" by only replacing those injectors that were exceptionally bad and leaving those that were merely bad. But what is well enough for stock power isn't well enough for higher power levels. As always, these are our opinions based upon what we've learned from our sources and our testing.

Hope that helps!

Cheers,
shiv

[Ilma]

That makes sense in that these problems started appearing precisely when the injector recall took place.

In my mind it was directly correlated.

We have been blaming the software all along.......but in my case, only two injectors were replaced so hardware is a likely candidate.

The fact that the problem happens on multiple tuning platforms also reinforces this.

It would also explain why some people get this issue while others don't.

The hardware is the wildcard that can vary car by car. The software would be identical for everyone.

You just reinforced my recent thinking on where to look to solve the problem.

Thanks for taking the time to do this.

[2Times]

That sir is pretty damn good info. Thank you.

[themyst]

Definitely feasible what you're suggesting shiv. Since I'm doing plugs during the RB install, I will write down all my injector codes and get someone local to check with INPA.

I still believe its tune related, but no harm doing the above to confirm it is not a hardware issue. If it is hardware or coding related, I'll be the first to apologize and send you a fruit basket.

Just to point out, I have a 01/09 build MSD81 with 32k miles so my injectors are fresh and never replaced. Car idles great starts right up, boost builds instantly and throws no codes. Just flatlines. I'd buy the idea my plugs may cause the issue if this weren't happening with 13k miles on the odometer.

[OpenFlash]

Originally Posted by themyst

Definitely feasible what you're suggesting shiv. Since I'm doing plugs during the RB install, I will write down all my injector codes and get someone local to check with INPA.

I still believe its tune related, but no harm doing the above to confirm it is not a hardware issue. If it is hardware or coding related, I'll be the first to apologize and send you a fruit basket.

Just to point out, I have a 01/09 build MSD81 with 32k miles so my injectors are fresh and never replaced. Car idles great starts right up, boost builds instantly and throws no codes. Just flatlines. I'd buy the idea my plugs may cause the issue if this weren't happening with 13k miles on the odometer.

I cant speak for your particular tune. I don't even know which one you are running these days. All our testing has been done on Procedes. Last I recall, you were stacking a jb4 on top of a Cobb tune. Getting flatlining and somehow blaming it on the Procede when opportunity arose

[Joshboody]

in my opinion boost could play a lot into this... Higher the boost before shifting more heat and more chance of ignition correction. Cobb is targeting lower ign, so any correcton will look more severe then Procede with higher targets.

Hey Shiv in your opinion woud a leak down test find bad couplers, or do you need the high pressures of a compression test? The latter seems a little involved on this car.

[ppatel223]

Lord Shiv of N54 is always ahead of other tuners!!!!!!!!!

[onesuperboi]

Originally Posted by ppatel223

Lord Shiv of N54 is always ahead of other tuners!!!!!!!!!

Just had to quote this...

[Ilma]

Originally Posted by Joshboody

in my opinion boost could play a lot into this... Higher the boost before shifting more heat and more chance of ignition correction. Cobb is targeting lower ign, so any correcton will look more severe then Procede with higher targets.

Not really.......as Myst pointed out in another thread, I could flatline running anywhere from 14 psi to 18 psi.

Lowering boost did not seem to eliminate the issue.

I never thought it was knock related since it did not follow the typical 3-4 degree retard correction that is the classic DME pattern. It just dumped an entire 10 degrees of advance right after the shift as if the DME didn't like something else that was going on.

And it only would happen on the 3-4 upshift during runs of less than 7 seconds duration on a relatively cool engine.

It seems to be correlated to the shift mechanism somehow.

I have logs where I alternatively tried downshifting into 4th from 5th on the highway to replicate the pattern, and could pull for 12 seconds at a time with full timing and no drop.

It's only on the 3-4 upshift for some reason.

That doesn't sound heat or knock related to me.

[Ilma]

Originally Posted by themyst

Definitely feasible what you're suggesting shiv. Since I'm doing plugs during the RB install, I will write down all my injector codes and get someone local to check with INPA.

I still believe its tune related, but no harm doing the above to confirm it is not a hardware issue. If it is hardware or coding related, I'll be the first to apologize and send you a fruit basket.

Just to point out, I have a 01/09 build MSD81 with 32k miles so my injectors are fresh and never replaced. Car idles great starts right up, boost builds instantly and throws no codes. Just flatlines. I'd buy the idea my plugs may cause the issue if this weren't happening with 13k miles on the odometer.

Hmmmmm......seem like it's not necessarily the age of your injectors because when I had the recall done my car was about the same mileage as yours is now.

That's why I suspect it has something to do with the new software and coding of the injectors.

That was the only thing that was changed before the problem started so it must have it's roots there.

[themyst]

Originally Posted by shiv@vishnu

I cant speak for your particular tune. I don't even know which one you are running these days. All our testing has been done on Procedes. Last I recall, you were stacking a jb4 on top of a Cobb tune. Getting flatlining and somehow blaming it on the Procede when opportunity arose

You know quite well exactly what I'm running considering you were actively responding to me just yesterday in a help-me thread.

Anyway, it's quite simple to check on the injector coding. If coding is the issue, you're a genius; if it doesn't, you're wrong (as I suspect).

PS - I didn't necessarily blame the Procede if you'd review our interactions; it's the fact you claimed it was my basic meth kit and that a PWM meth kit is the only way to fix it. Now you acknowledge otherwise. That kind of sales tactic along with deception and unwillingness to acknowledge issues are the issue here.

[themyst]

Originally Posted by Ilma

Hmmmmm......seem like it's not necessarily the age of your injectors because when I had the recall done my car was about the same mileage as yours is now.

That's why I suspect it has something to do with the new software and coding of the injectors.

That was the only thing that was changed before the problem started so it must have it's roots there.

Easy enough to test this theory. If you have access to a K-CAN cable, follow vasilalov's injector coding DIY and get the injector codes. I believe all you need to do is pull the engine cover off with no other modification.

[hotrod182]

On the 08 N54 Sedan, I had new injectors and programming last year.
Here is a pull to 150mph. 91 octane + Meth.

[themyst]

Originally Posted by hotrod182

On the 08 N54 Sedan, I had new injectors and programming last year.
Here is a pull to 150mph. 91 octane + Meth.

I remember you used to flatline poorly as well. Did the new injectors and coding fix your issue?

[OpenFlash]

Originally Posted by themyst

You know quite well exactly what I'm running considering you were actively responding to me just yesterday in a help-me thread.

Anyway, it's quite simple to check on the injector coding. If coding is the issue, you're a genius; if it doesn't, you're wrong (as I suspect).

PS - I didn't necessarily blame the Procede if you'd review our interactions; it's the fact you claimed it was my basic meth kit and that a PWM meth kit is the only way to fix it. Now you acknowledge otherwise. That kind of sales tactic along with deception and unwillingness to acknowledge issues are the issue here.

You can believe what you want to believe. My goal is not to make you change your mind. But rather help others who are willing to learn instead of pointing fingers with ill-conceived accusations. I think I have a pretty good understanding and testing methodology. And results to back it up. You, on the other hand, are more interested in calling me names and updating the other forum when you think you caught me in a "lie". Ask yourself if you have enough knowledge to distinguish a lie from your inability to understand the facts. Fwiw, poorly timed methanol injection can easily have the same effect as a bad injector. Both are injection issues and both can trigger ignition timing aberrations as well as changes in fuel modes. We've learned a lot about this with the single turbo application. Making 700hp means that everything must be even more tightly controlled as far as the DME is concerned.

[themyst]

Originally Posted by shiv@vishnu

You can believe what you want to believe. My goal is not to make you change your mind. But rather help others who are willing to learn instead of pointing fingers with ill-conceived accusations. I think I have a pretty good understanding and testing methodology. And results to back it up. You, on the other hand, are more interested in calling me names and updating the other forum when you think you caught me in a "lie". Ask yourself if you have enough knowledge to distinguish a lie from your inability to understand the facts. Fwiw, poorly timed methanol injection can easily have the same effect as a bad injector. Both are injection issues and both can trigger ignition timing aberrations as well as changes in fuel modes. We've learned a lot about this with the single turbo application. Making 700hp means that everything must be even more tightly controlled as far as the DME is concerned.

Hey Shiv, thanks for that well thought-out response. If you'd kindly review my recent statements, I said that it's an interesting theory, one I will be checking in a week.

And if your PWM meth kit is so wonderful, explain why we flatline on pump gas no meth? And explain why even those on PWM kits flatline?

Like I said (since you have a short selective memory) if the injector coding is the issue I'll be the first to acknowledge that you found the solution.

[BMW-335i-07]

So for all those early build 2007 cars that DIDN'T qualify for the injector replacement, I suggest biting the bullet and getting them replaced. Your car will drive much better.

any one have this bullet? and on witch cars can this be done? thx...

[SIKH335]

It truly is sad when people can't see that they are being helped in everyway possible. Suggestions to fix there problems are taken as sales pitches rather than an option of various options that could be a fix. I'm Glad people like Shiv and the Vishnu team are still pushing forward with this platform. Every new item learned benefits everyone including those who are haters and like to live on every piece of gossip (ohhhh Mmmm geee). I find myself laughing at people for their lack of a life. But am again glad we are pushing forward and understanding this motor better every day

[themyst]

Theory update:

My injector coding was actually spot on.

[Jeff@TopGearSolutions]

Don't forget the compression seals that go bad.

[themyst]

Originally Posted by Jeff@topgearsolutions

Don't forget the compression seals that go bad.

First step:

Check compression again

Second step:

Replace valve cover gasket

Third step:

Change cylinder 6 injector

If that fails:

Leave car in the ghetto with windows down and key in the ignition and walk away

[Jeff@TopGearSolutions]

Profit? By Honda?