[Lulz_M3]

Originally Posted by themyst

More along the lines of how you can get away with bloody murder with DI... to a point.

Oddly, I never had any of these fuel trim oddities on Procede when I ran it. Only issue with it was timing flatline.

I frequently flatline timing post shift, if I "slip" the clutch during the shift or shift slow it's fine. Annoying though.

[skim7x]

Originally Posted by rader1

Re-install Cobb and log. I'll be willing to bet your AFR will be perfect and your STFTs will be spot on. The PROcede cannot control fueling as it should be and hardware is almost certainly not your problem.

Originally Posted by onesuperboi

Yeah, if you still have COBB, you might as well try it and log your AFR. At least it would help you out and let you know if it is indeed a hardware problem and not tune related.

Actually, I did just that and my AFR's were perfect on Cobb.

Originally Posted by themyst

OP, IIRC, AFR wasn't your issue with Cobb. You had massive timing correction activity it seems. Valve cleaning maybe?

Try disabling NLS and shift normally just for testing purposes?

Yeah the massive timing corrections were why I decided to try Procede... but the AFR problem seems to be a bigger problem haha. I actually had my valves cleaned while I was trying to troubleshoot the timing corrections with Cobb... First, I changed gas stations because I thought it was bad gas... didn't work... then I had intake valves cleaned, didn't work... then I had plugs changed (because I started getting misfires) and that fixed the misfires but not the timing correction. I wasn't sure what else I could do...

Today I reloaded the firmware, reloaded the aggressive maps, reset all the adaptations, disabled NLS, and shifted normally and still had fueling problems... so it's not the NLS...

My car is so weird... lol

[themyst]

Originally Posted by Lulz_M3

I frequently flatline timing post shift, if I "slip" the clutch during the shift or shift slow it's fine. Annoying though.

Being short 10+ degrees of timing after the shift is costing you 30 whp easy.

[themyst]

Originally Posted by skim7x

Actually, I did just that and my AFR's were perfect on Cobb.


Yeah the massive timing corrections were why I decided to try Procede... but the AFR problem seems to be a bigger problem haha. I actually had my valves cleaned while I was trying to troubleshoot the timing corrections with Cobb... First, I changed gas stations because I thought it was bad gas... didn't work... then I had intake valves cleaned, didn't work... then I had plugs changed (because I started getting misfires) and that fixed the misfires but not the timing correction. I wasn't sure what else I could do...

Today I reloaded the firmware, reloaded the aggressive maps, reset all the adaptations, disabled NLS, and shifted normally and still had fueling problems... so it's not the NLS...

My car is so weird... lol

Heh, all you got left is a JB4 to try

[Lulz_M3]

Originally Posted by themyst

Being short 10+ degrees of timing after the shift is costing you 30 whp easy.

Yeah, no doubt. I can feel it fall on its face, usually recovers a few seconds later.

OP, GL man - your issues seem very odd. Can you log knock sensor voltage with PROcede? I'd want to know how much knock is being induced during those lean spikes, you may be doing damage.

[skim7x]

Originally Posted by themyst

Heh, all you got left is a JB4 to try

Seriously... I had Dinan before the Cobb also haha...

Originally Posted by Lulz_M3

OP, GL man - your issues seem very odd. Can you log knock sensor voltage with PROcede? I'd want to know how much knock is being induced during those lean spikes, you may be doing damage.

Yikes.. you think so? Well, I'm not going to be doing any more WOT runs on the PROcede due to these problems... I think I'll likely go back to Cobb after all this... which I'm perfectly happy with... I wanted to see if I was missing out on anything with the Procede... on-the-fly map switching is awesome, NLS is awesome... if Cobb can add those to their software, I'll be 100% satisfied.

[onesuperboi]

You still have timing problems when you just tried COBB?

It's weird that you are having two different issues with two different tunes.

[skim7x]

Originally Posted by onesuperboi

You still have timing problems when you just tried COBB?

It's weird that you are having two different issues with two different tunes.

Exactly... with Cobb I'm having timing problems, and with Procede I'm having fueling problems... Here's a log of the Cobb:

[jpsimon]

Could it be hardware related? Those ae really strange issues?

[Tzu]

The fueling is NOT hardware related as illustrated by the Cobb log. Thats just typical vishnu telling you to spend more money. Although a valve cleaning never hurts.

Are these the stg 3 maps? Those OTS maps maxed out my trims even with OL at 100% and boost at 14psi. They are overfueled for stock twins, IMO. But OP seems to have OL-CL transitional issues with the procede.

What maps are you using for cobb? Have you updated to the newer maps available?

[Lulz_M3]

Skim, have you sent those Cobb logs into Rob Irish?

[jpsimon]

How do you explain his incredibly low timing while running Cobb? Something isn't right and both tunes are doing strange things

Originally Posted by Tzu

The fueling is NOT hardware related as illustrated by the Cobb log. Thats just typical vishnu telling you to spend more money. Although a valve cleaning never hurts.

Are these the stg 3 maps? Those OTS maps maxed out my trims even with OL at 100% and boost at 14psi. They are overfueled for stock twins, IMO. But OP seems to have OL-CL transitional issues with the procede.

What maps are you using for cobb? Have you updated to the newer maps available?

[themyst]

Originally Posted by jpsimon

How do you explain his incredibly low timing while running Cobb? Something isn't right and both tunes are doing strange things

But both tunes are showing two different things. One has fuel anomalies which other people also have and the other just has crap timing.

[themyst]

Originally Posted by skim7x

Seriously... I had Dinan before the Cobb also haha...

Yikes.. you think so? Well, I'm not going to be doing any more WOT runs on the PROcede due to these problems... I think I'll likely go back to Cobb after all this... which I'm perfectly happy with... I wanted to see if I was missing out on anything with the Procede... on-the-fly map switching is awesome, NLS is awesome... if Cobb can add those to their software, I'll be 100% satisfied.

You could add a WOTbox for NLS shifting on the AP.

Also try an INPA adaptation reset. It made my car drive terribly overall for a few hundred miles but once it readapted car is running great no issues. Not sure if it's the tune or the adaptation resets but it can't hurt to try.

[Lulz_M3]

I'd much rather have no timing than lean fuel spikes... I honestly cannot believe the PROcede doesn't go into some kind of an auto protect mode.

[Tzu]

Originally Posted by jpsimon

How do you explain his incredibly low timing while running Cobb? Something isn't right and both tunes are doing strange things

I havn't played around enoguh with Cobb to really have a complete understanding of how it functions to give map advice. I've only loaded it and logged, and know that it works fine on my setup. Guys like Themyst, Enrita and Vas are much more experienced.

But as Themyst stated, two different anomolies. Though Skim isnt the first one I've seen to have negative timing. The only solutions I've seen were newer updates and bug fixes.

Originally Posted by Lulz_M3

I'd much rather have no timing than lean fuel spikes... I honestly cannot believe the PROcede doesn't go into some kind of an auto protect mode.

The reason why lean-run protection won't work correctly as I see it, is the transition from OL to CL in many logs illustrate a lean spike at MBT. If you set lean-run to 15AFR (~1.05 lambda), it will trip quite often at the transitional spike. This is something that cannot be tuned out (as far as I know) with piggies. Narrowing the observation window to appear only after the spike (maybe throttle position/load/boost thresholds) just defeats the purpose. Kind of a catch 22.

[BrianMN]

Note- those lean spikes in the logs are not actual AFR. An independent wideband can (and has) confirmed this for me time and time again. Tip in and post shift lean spots are applying bias. Perhaps it would help if people more knowledgeable would clear the air rather than the typical "proceed can't control fuel properly" song and dance.

[rader1]

Originally Posted by BrianMN

Note- those lean spikes in the logs are not actual AFR. An independent wideband can (and has) confirmed this for me time and time again. Tip in and post shift lean spots are applying bias. Perhaps it would help if people more knowledgeable would clear the air rather than the typical "proceed can't control fuel properly" song and dance.

No. Skim isnt having "tip in" spikes. He's seeing sustained lean conditions with his STFTs maxed at peak torque mid-pull. He's experiencing what you described during and IMMEDIATELY post shift, but it settles back to about 17:1 and stays there.

[themyst]

Originally Posted by BrianMN

Note- those lean spikes in the logs are not actual AFR. An independent wideband can (and has) confirmed this for me time and time again. Tip in and post shift lean spots are applying bias. Perhaps it would help if people more knowledgeable would clear the air rather than the typical "proceed can't control fuel properly" song and dance.

Why is the calculation not corrected and why does the AFR not hit target immediately if those examples aren't an onset lean condition? AFR trace on countless dyno graphs has proved this over and over again.

[themyst]

Originally Posted by Tzu

I havn't played around enoguh with Cobb to really have a complete understanding of how it functions to give map advice. I've only loaded it and logged, and know that it works fine on my setup. Guys like Themyst, Enrita and Vas are much more experienced.

But as Themyst stated, two different anomolies. Though Skim isnt the first one I've seen to have negative timing. The only solutions I've seen were newer updates and bug fixes.



The reason why lean-run protection won't work correctly as I see it, is the transition from OL to CL in many logs illustrate a lean spike at MBT. If you set lean-run to 15AFR (~1.05 lambda), it will trip quite often at the transitional spike. This is something that cannot be tuned out (as far as I know) with piggies. Narrowing the observation window to appear only after the spike (maybe throttle position/load/boost thresholds) just defeats the purpose. Kind of a catch 22.

All I know is that is some weird chiz and not the first time I've seen it. I think DME versions also play a factor in how some of these cars behave with various tunes. MSD80 vs MSD81?

[Lulz_M3]

Originally Posted by BrianMN

Note- those lean spikes in the logs are not actual AFR. An independent wideband can (and has) confirmed this for me time and time again. Tip in and post shift lean spots are applying bias. Perhaps it would help if people more knowledgeable would clear the air rather than the typical "proceed can't control fuel properly" song and dance.

Sorry, don't know much about PROcede but I understand AFR. From what I recall the PROcede calculates AFR, so why would it show such a crazy lean spike?

[Tzu]

Originally Posted by BrianMN

Note- those lean spikes in the logs are not actual AFR. An independent wideband can (and has) confirmed this for me time and time again. Tip in and post shift lean spots are applying bias. Perhaps it would help if people more knowledgeable would clear the air rather than the typical "proceed can't control fuel properly" song and dance.

I'm not trying to start any arguements here. I'm just speaking about my own testing with sniffers shoved up the the tailing end of the midpipe. My outcome is different than yours, and I try to account for it in the AFR maps. Not a huge deal.

There was a rather large thread about how the AFR calculations were refined over time, and that theyre pretty on spot right now. If that is a fact, then the spikes at WOT onset are as well. Something has to be taken as gospel here.