Concerned and aggrivated

[Whippled 496]

First let me say that I am not making this post just to hear (see) my self talk.....I have been having this problem as well as many others have from day one of installing my COS.

This is the OS I am running: 03190003

I love the versatility and abilities this COS gives me with my SD tune save one HUGE problem.......extremely large Lean spikes on throttle tip-in. I see anywhere in the range of 1.35 - 1.55 BEN factors when hitting the accelerator. When i go WOT, it will actually bog for a fraction of a second until it can recover (5 or so frames later). I was about to purchase the road runner but am waiting for a resolution to this....because to be honest.....the thought of crossing over to other software has crossed my mind lately. Everyone i talk to on the other side does not have these issues when running a SD tune. Lean spikes of this magnitude are a REAL concern to me. I could always go back to closed loop, but i bought this software, spent many hoours learning and tuning, and want to be able to tune my truck the way i should be able to. The first major concern of the COS (which is now a stickey in this section) was the ECT issues at start-up...this was resolved very quickly and updates were posted. I think this situation warrants the same attention. This problem has been brought up many times, most of which were done in private meassages and I would really like to at least see some discussion on a resolution.

I am providing a copy of my tune and a log for that tune so you can see what I am talking about. I know this is not a tuning parameter that is incorrect, this has been confirmed as a COS issue. Please help me to resolve this. Acceleration enrichment tables have been discussed as a resolution to this issue. I understand that this tune is not "spot on" by any means, but this does not account for the terrible lean spikes observed (and felt).



Thanks,
Jimmie

[dc_justin]

Just to be fair Jimmie, the other guys have the same problem. ;-) Dewey and Holty both experience it. Dewey tunes around it (raises VE dramatically in the transition area), Holty just deals with it.

[Whippled 496]

Just to be fair Jimmie, the other guys have the same problem. ;-) Dewey and Holty both experience it. Dewey tunes around it (raises VE dramatically in the transition area), Holty just deals with it.


Okay, I was told they did not have these issues. Thanks for the clarification Justin. I dont want to "Deal" with it, i want it to be correct. I dont know a "work around" that would prevent this spike, but i am not a seasoned tuner either. I am welcome to suggestions as a temporary fix, but i feel there should be a change in the code to prevent this all together. Crossing my fingers.....

[dc_justin]

Here's a visual example of this problem.

http://www.marketitright.com/ss/tuning/transient-blip.gif

[joecar]

W496,

What Justin said, this is a common thing... some people get creative with the VE table.

Thinking Out Loud (TOL)... could the lean spike be "softened" by:
- ramping CFOL up earlier...?
- enabling PE at a lower MAP...?
- zeroing the RPM delay (oh, I see the delay is zero anyway)...?

Justin, do all the PE delay bypasses have to be met for the PCM to bypass the delay,
this is how it works, right (I just drew a blank and I need more coffee... :hihi: )...?

Cheers,
Joe
:cheers:

[dc_justin]

W496,

What Justin said, this is a common thing... some people get creative with the VE table.

Thinking Out Loud (TOL)... could the lean spike be "softened" by:
- ramping CFOL up earlier...?
- enabling PE at a lower MAP...?
- zeroing the RPM delay (oh, I see the delay is zero anyway)...?

Justin, do all the PE delay bypasses have to be met for the PCM to bypass the delay,
this is how it works, right (I just drew a blank and I need more coffee... :hihi: )...?

Cheers,
Joe
:cheers:

Unfortunately, it's not something that can be solved with the current calibrations. Check the screenshot I posted. Commanded AFR of 11.25 (on the dyno). Spiked up to 15:1 before starting to settle back down.

[joecar]

I see, it spikes regardless.

[Whippled 496]

I have tried some VE table creativity but it does not do much to help and ultimately results in an extremely rich idle........so i changed it back. It would seem that if (as Joe mentioned) this is a "common" issue, it would have been addressed by now. Please help Ross/Paul!!

[Delco]

the VE table looks way out of whack for a whipple combo , the problem you are experincing is due to a couple of factors , one is the software never had a tps enrichment table like a pump shot , and the other is the amount of air the whipple forces in on the throttle crack.

having tuned a huge number of whipple combo's I can tell you the low rpm VE under boost should have a huge lump in it whereas yours is a flat line - not what the engine is seeing.

on toad tunign makes the VE table look very differnt to what the VE table looks like under steady state load on the dyno.

Also when doing a boosted application like you have we alwasy use a boost referenced pressure regulator so we get a psuedo pump shot from the regulator on large map changes.

it is not EFILIVEs CO3 issue but a GM coding issue really

I have attached a example of what I would expect the VE tables to look like , do not use in your car as it wont be right but more for your interest

[dc_justin]

I agree about the steady state tuning to an extent. The graph I posted up was after a few hours of steady state tuning that truck on a dyno dynamics. Had VE nailed down at all load points from 1200-4400rpm (yeah, that was "fun" in boost at full load, 2nd gear). Problem is still very apparent.

Are we certain that there is no accel enrichment in the PCM? I've read a GM-produced LS1 document that stated otherwise. PDF was posted here at some point a while back...

[Whippled 496]

the VE table looks way out of whack for a whipple combo , the problem you are experincing is due to a couple of factors , one is the software never had a tps enrichment table like a pump shot , and the other is the amount of air the whipple forces in on the throttle crack.

having tuned a huge number of whipple combo's I can tell you the low rpm VE under boost should have a huge lump in it whereas yours is a flat line - not what the engine is seeing.

on toad tunign makes the VE table look very differnt to what the VE table looks like under steady state load on the dyno.

Also when doing a boosted application like you have we alwasy use a boost referenced pressure regulator so we get a psuedo pump shot from the regulator on large map changes.

it is not EFILIVEs CO3 issue but a GM coding issue really

I have attached a example of what I would expect the VE tables to look like , do not use in your car as it wont be right but more for your interest

Thanks for the input DELCO, like i mentioned before I realize my tune is not dead on by any means. All i know is that this phenomenon was not occuring when running the stock OS / MAF tune. The transitions to commanded AFR were made within 2 frames. Lean spikes did not occur until the throttle was released, opposite of how the COS tune works with the rich spikes when throttle is released. I had planned on getting the road runner to help with tuning the VE table because of just how difficult and time consuming it actually is to hit even 1/3 of those cells by driving around......especially in my area where the traffic is always terrible. Believe it or not, aside from a few stray cells, my BEN MAPS were within +/-.07 through out the majority of my standard driving range of the VE table. I know enough about tuning to see when something is wrong but no where near as much as most of you guys know, which is why i started this thread....maybe if the problem is with GM, then we could put our heads together to come up with a work around. If not, i was hoping like Justin has been mentioning, the enrichment acceleration tables could be added.

[joecar]

From the tune file it looks like you already have a boost referenced regulator.

[Whippled 496]

From the tune file it looks like you already have a boost referenced regulator.

Yes sorry, forgot to mention that, thanks Joe. It is a vacuum referenced return style fuel system with the regulator on the fuel rail.

Why is it that these spike do not occur when using the stock OS / closed loop tune?

[joecar]

I've been sifting thru my COS5 logs, and I don't see the lean spikes (I'm NA)...

Are the lean spikes on tip in only occurring with boosted applications...?

I remember seeing that hquick/Howard is seeing the lean spikes (he's got boost).

[Whippled 496]

I've been sifting thru my COS5 logs, and I don't see the lean spikes (I'm NA)...

Are the lean spikes on tip in only occurring with boosted applications...?

I remember seeing that hquick/Howard is seeing the lean spikes (he's got boost).


Not sure Joe, i know i do NOT have to be into boost to see the spikes though. Any positive throttle position gives me lean spikes. Here is a shot of a very slight throttle increase that results in the same lean spike. It happens whenever i press the pedal.

http://www.pasadenatownhall.com/pictures/lean.JPG

[Delco]

Would like to see that doc , I have the GM calibrator tool for the early stuff with every calibration variable , all 4000+ of them , nothing that looks or documented as a pump shot or delta enrichment.

Wall wetting calculations may allow it to be fudged though.

Trust me Ross and I have looked for years as a transiant acceleration factor for rnleanment and enrichment would be great - especially on big cammed stuff and blown combo's

Looking at that log then engine is lean for a long time after the throttle movement , I would be looking more at the mapping of the VE

[Whippled 496]

Would like to see that doc , I have the GM calibrator tool for the early stuff with every calibration variable , all 4000+ of them , nothing that looks or documented as a pump shot or delta enrichment.

Wall wetting calculations may allow it to be fudged though.

Trust me Ross and I have looked for years as a transiant acceleration factor for rnleanment and enrichment would be great - especially on big cammed stuff and blown combo's

Looking at that log then engine is lean for a long time after the throttle movement , I would be looking more at the mapping of the VE

It just takes a little while for it to settle back and I am still working on dialing in the VE. The same engine parameters that result in a stoich AFR will result in a lean one after tip-in which makes AutoVE even more tough. Again, i realize my tune is not dead on yet....this does not change the fact there is a lean tip-in problem with this COS as represented by several other tuners here who's tunes are much more inline than mine is and have bunches more experience than i do. What i know is that it goes WAY lean everytime i hit the pedal....everytime......regardless of RPM, MAP, MPH, or what g/cyl is in the VE cells.

Why do these lean spikes happen in the COS BUT not when using the 1bar stock OS/MAF tune?

[slowhawk]

I can't read your tune at work but I've tuned many Whipple and Turbo Trucks and never have a problem with a lean spike or richening problem. I use COS3 2bar,no MAF. I'm guessing the PE throttle position table is different than the way I set it.

[Whippled 496]

I can't read your tune at work but I've tuned many Whipple and Turbo Trucks and never have a problem with a lean spike or richening problem. I use COS3 2bar,no MAF. I'm guessing the PE throttle position table is different than the way I set it.

Well if you get a second to peak at it, that would be great!! If you can fix it, lunch is on me!!

[slowhawk]

Stop smoothing out the VE table!!!!tune it to what it wants.It will have hills,spikes and valleys when tuned in.

[Whippled 496]

Stop smoothing out the VE table!!!!tune it to what it wants.It will have hills,spikes and valleys when tuned in.

So you were able to look at it? LOL, i know.....i was just trying to smooth the areas I have not hit. Do you use the road runner to tune your VE?

[slowhawk]

So you were able to look at it? LOL, i know.....i was just trying to smooth the areas I have not hit. Do you use the road runner to tune your VE?

No,auto VE. Doesn't take long.If you have lean spot's auto VE those spots and transfer them.Don't worry about the pretty looking VE table..

[Whippled 496]

So is this really the concensus? The VE table is not inline and thats why the lean spikes are happening? Justin, how inline is your VE table? Why is this only happening to a few people? Is it a 2003+ thing maybe? Justin, your truck is a 2003 right? Mine is a 2003. Maybe its a 2003+ COS issue?

[slowhawk]

If your doing SD without a MAF the VE never looks smooth.My trucks OLSD VE table looks horrible for smoothness but the truck runs perfect on a stock 5.3 with a 216 cam and 7lbs of boost from a Turbo.I have 10k miles on it while towing and my wife daily driving it. Just auto VE it and DO NOT smooth it.

[Whippled 496]

If your doing SD without a MAF the VE never looks smooth.My trucks OLSD VE table looks horrible for smoothness but the truck runs perfect on a stock 5.3 with a 216 cam and 7lbs of boost from a Turbo.I have 10k miles on it while towing and my wife daily driving it. Just auto VE it and DO NOT smooth it.


I am going to do that (keep doing that). Please help me understand how this will prevent the lean spikes i am seeing though. Please dont take this the wrong way...I am not trying to be difficult..:) I am just a little ignorant is all. How is it that no matter where i am in my RPM range, MAP, or MPH (its not that far off everywhere)...when ever i touch the throttle it goes lean....same with Justin who has his VE nailed down tight and same with various others here. I understand my VE needs work, but this cant account for these spikes that happen all the time. I just know there is something more to this lean tip-in. What year is your truck? Have you tuned any 2003+ 1024k PCMs that were supercharged? Like i said, maybe its a COS year specific problem.

[hquick]

Glad I found this thread.
As Joe mentioned...I too am having these issues...SD or running a MAF.
I have run probably in excess of 300 logs and autoVE'd numerous times.
Mick leant me his RR so I used RTACS and hit a good amount of cells on some long 5-6 hr drives. I've been/am still trying to dial in the MAF using RTACS.
I still have the lean spikes...exactly the same as Whippled496....any speed...any RPM as I jab/touch the accellerator.
I've tried going back to using 1bar and a MAF via a 'PRO's' suggestion but feel its worse now than before. In the morning...sometimes as I try to take off (normally) it will go VERY lean (18+) and almost stall.
I have a 98 Suburban with the Whipple and marine intake and using 42lb injectors.
Replaced the fuel pump with a Walbro 2 days ago.

Here's my current tune and a log from then drive home today...RTACS'ing the MAF.

[Whippled 496]

Glad I found this thread.
As Joe mentioned...I too am having these issues...SD or running a MAF.
I have run probably in excess of 300 logs and autoVE'd numerous times.
Mick leant me his RR so I used RTACS and hit a good amount of cells on some long 5-6 hr drives. I've been/am still trying to dial in the MAF using RTACS.
I still have the lean spikes...exactly the same as Whippled496....any speed...any RPM as I jab/touch the accellerator.
I've tried going back to using 1bar and a MAF via a 'PRO's' suggestion but feel its worse now than before. In the morning...sometimes as I try to take off (normally) it will go VERY lean (18+) and almost stall.
I have a 98 Suburban with the Whipple and marine intake and using 42lb injectors.
Replaced the fuel pump with a Walbro 2 days ago.

Here's my current tune and a log from then drive home today...RTACS'ing the MAF.

Well, i didnt expect that. So you have these issues even running the stock OS/ MAF tune as it came from the factory? The only time mine does it is in SDOL. It never did it when i was running the factory OS/Tune. I just love the SD so much i dont want to go back....i want to fix it but dont know how.

[hquick]

Sorry...I should have eleaborated further...I forgot! :)

I'm using an LS1 PCM and a 2002 op system.
But it is a standard op system 12212156.

[slowhawk]

I've done the same OS files. You guy's can shoot your tune file to slowhawk@comcast.net and I'll look them over and see what you guy's have different from me.Make sure you include all the engine mods.I've done a few whipple's with no problems but I know the lean spike you guy's are talking about.Was tuning a blown 416ci Vette yesterday and it was doing the same thing.Added full in those cells(alot of fuel) and it smoothed out perfect.

[hquick]

Hey...thanks for that.
I've had so many people look over this thing...the more the merrier.

Thanks (you have mail)

[Whippled 496]

Hey...thanks for that.
I've had so many people look over this thing...the more the merrier.

Thanks (you have mail)

Email sent! Thanks!

[izaks]

You have PM

[Whippled 496]

Cool, i will try these suggestions Izak. I will let you know what i find.

[Aloicious]

Y'know, I'm getting the exact same problem, and I'm N/A, lean at all throttle tip in, and I also get rich spikes on decel. :shock: and its screwing with my AutoVE.

I am thinking I may have some heatsoak problems, so I'm trying to work on that...but any info people have would help out, I'm going out of my mind trying to get this sorted out.

edit:
I'm running a 1996 Vortec L31 motor with a ZZ4 cam, MPFI intake running 24# SV1 injectors (flow tested at 259CC/min @43.5psi) running @ 45psi. using a 122000411 PCM with a 2002 Xvan L31 base tune on a 20020003 COS.

I'm also going to try SSpd's post about how he fixed the rich decel on his SVO injectors by altering the B3701 table. but I wanna get this heatsoak fixed tommorow.

[Whippled 496]

Y'know, I'm getting the exact same problem, and I'm N/A, lean at all throttle tip in, and I also get rich spikes on decel. :shock: and its screwing with my AutoVE.

I am thinking I may have some heatsoak problems, so I'm trying to work on that...but any info people have would help out, I'm going out of my mind trying to get this sorted out.

edit:
I'm running a 1996 Vortec L31 motor with a ZZ4 cam, MPFI intake running 24# SV1 injectors (flow tested at 259CC/min @43.5psi) running @ 45psi. using a 122000411 PCM with a 2002 Xvan L31 base tune on a 20020003 COS.

I'm also going to try SSpd's post about how he fixed the rich decel on his SVO injectors by altering the B3701 table. but I wanna get this heatsoak fixed tommorow.


Are you running COS3? Is everyone who has this problem running COS3? Wonder if COS5 people have this problem?

[Aloicious]

Are you running COS3? Is everyone who has this problem running COS3? Wonder if COS5 people have this problem?

yeah its a COS3, however, I was running the same COS 3 before I swaped to MPFI and swapped cams, and never had the problem. so I'm thinking at least in my case, it is problably some overlooked injector setting, and/or the heatsoak...it may be a simmilar issue for you as well, but just amplified due to the boost. I'll let you know if I find anything out.

EDIT: I should mention that I've also been smoothing the VE table, but I'm going to do a few drives today w/o smoothing like was mentioned earlier and see if it makes a difference.

[hquick]

Well....I've just switched back to COS3 because the stock op system just had the truck running like crap.
This morning I ran the STFT BEN correction method and got ALOT of the table sitting at 1.0.
After a couple more passes and narrowing down the filters a little...I'll try bringing the MAF back into the equation and see if I can get rid of that lean spike.

[dc_justin]

Would like to see that doc , I have the GM calibrator tool for the early stuff with every calibration variable , all 4000+ of them , nothing that looks or documented as a pump shot or delta enrichment.

Wall wetting calculations may allow it to be fudged though.

Trust me Ross and I have looked for years as a transiant acceleration factor for rnleanment and enrichment would be great - especially on big cammed stuff and blown combo's

Looking at that log then engine is lean for a long time after the throttle movement , I would be looking more at the mapping of the VE

The document that made mention of an "Acceleration Mode" just popped back up in the search today:

http://carprogrammer.com/Z28/PCM/LS1/Holden_LS1.pdf

Looks to be on page 95.




Acceleration Mode

When the driver presses on the accelerator pedal,
air flow into the cylinders increases rapidly, while
fuel flow tends to lag behind. To prevent possible
hesitation, the PCM increases the pulse width to the
injectors to provide extra fuel acceleration. The
PCM determines the amount of fuel required based
on throttle position, coolant temperature, manifold
air pressure, mass air flow and engine speed.

[hquick]

Is that Holden specific perhaps? (like lean cruise is).
I have a buddy who works in Holden writing the cal's....I could ask him...if I knew WHAT to ask him???

[hquick]

Well...I asked the question of my GMH buddy.

Is this something that's actually used in the fueling of the Holden vehicles?


Acceleration Mode
When the driver presses on the accelerator pedal,
air flow into the cylinders increases rapidly, while
fuel flow tends to lag behind. To prevent possible
hesitation, the PCM increases the pulse width to the
injectors to provide extra fuel acceleration. The
PCM determines the amount of fuel required based
on throttle position, coolant temperature, manifold
air pressure, mass air flow and engine speed.

His reply.


"Yep, thats transient fuel, which is what I used to look after as well.
Its a horribly complex algorithm in our software, but the inputs are basically what was described there".

[Whippled 496]

Well...I asked the question of my GMH buddy.

Is this something that's actually used in the fueling of the Holden vehicles?


Acceleration Mode
When the driver presses on the accelerator pedal,
air flow into the cylinders increases rapidly, while
fuel flow tends to lag behind. To prevent possible
hesitation, the PCM increases the pulse width to the
injectors to provide extra fuel acceleration. The
PCM determines the amount of fuel required based
on throttle position, coolant temperature, manifold
air pressure, mass air flow and engine speed.

His reply.


"Yep, thats transient fuel, which is what I used to look after as well.
Its a horribly complex algorithm in our software, but the inputs are basically what was described there".


Sounds good to me....ROSS/PAUL can you go ahead and throw that table in there for us real quick? :)

Seriously...THAT....is what we need! Good find Justin, just hope something can come from it.

[dc_justin]

Sounds good to me....ROSS/PAUL can you go ahead and throw that table in there for us real quick? :)

Seriously...THAT....is what we need! Good find Justin, just hope something can come from it.


I really hope we can get this, but there are also other big fish to fry... T42 full flashing would be great and there are constantly E38 and E67 updates that really are beneficial.

I've asked for the transient fuel stuff 5 times over the past year and I know that if Ross had the time to get to it while still staying on top of the new stuff, he would have. Hopefully he'll get out from under the huge workload for enough time to hack the Gen-IIIs some more. :)

[hquick]

Any news/comments from Ross or Paul on this one?

[Aloicious]

Any news/comments from Ross or Paul on this one?

yeah, a transient fuel table would solve alot of issues. I'm interested to hear about this as well.

[Highlander]

any updates?

[GMPX]

Some have been adding these calibrations as .cax add-ons. Look for that group on the forum.

Cheers,
Ross