Tuning has finally begun! =)

[Corday]

Ok, so its been a few years and i've finally taken the time to dig into the tuning-business.. I installed the LC-1 via serial to my EFI Flashscan V2 and it has been calibrated and is working correctly.

Car: 1999 Pontiac Trans AM - 5,7L LS1 - Manual 6spd
Upgrades: Unrestricted intake-lid, Fast 90mm intake manifold + 90mm TB, Patriot Performance stage 2 heads, TR Camshaft 224/224-112LSA, Dynatech LT's, Y-pipe and highflow cats, Magnaflow Catback 3" exhaust, Underdrive pulley, Meziere electric waterpump, AIR & EGR delete.
Current issues: Very low mpg, strange idle (goes up and down without depressing the throttle), overall strange behavior regarding throttle response, emission values are high.

I have read and read and read on this forum and innovates and came up with the following approach:

I started with updating all software (efilive), firmware (flashscan) and PCM operating system to LS1B.

I did some initial logging of the engine with the wideband connected just to see the values with the PIDs recommended in various guides (i.e. SD AutoVE).

I updated the IFR-table as per RHS excel-file. I hope i put the correct values in there, attaching a photo of it. Injectors and everything about the fuel-system is stock.

It would be nice with some feedback from these logs and my tune before i start the AutoVE-logging just to see that im not going to damage anything.

Attached files and descriptions

Log 2 - 11102
Starting the car pre IFR-table update with MAF unplugged. Got a rich AFR of 10,2 and up (avg 13,5).

Log 3 - 11101
Starting the car with IFR-table updated and MAF connected. Got a lean AFR with peak at ~19 but stable idle.

Log 4 - 11100
Starting car with IFR-table updated and MAF disconnected. Got a good AFR of avg 14,33. Idle was very unsteady and 30s after starting the car it dipped low enough to almost die.

Reason why ive been connecting/disconnecting the MAF is because im preparing for the AutoVE-mapping which says to unplug the MAF

Tune - 11098
My latest tune with updated IFR-table.

Excel Capture
11099
Showing the numbers i used to update IFR-Table


Feedback greatly appreciated! =)

[Corday]

Tomorrow i will perform the SD AutoVE and im still not 100% confident everything is working ok.
Does the values from MAP, MAF and NBO2 look ok?

I think i will revert back to stock IFR-B4001-table for the AutoVE, since im not really confident about the new numbers.. They just seem abit high after looking at others tunes with stock fuelsystem.

How far up the RPM-range of the AutoVE-map i created should i aim to hit numbers in? 4800, 5200 or even higher?
Im also abit confused about AutoVE vs Calc.VE. Do they both target the same goal or what is the difference? Should/can i perform both of those at the same time?

If there is anything more i need to log/provide, just let me know

[joecar]

I'll look at your files on my other PC later tonite.

AutoVE first pass, hit say below 4000 rpm, correct the VE table, make sure you're on a sane path, and extrapolate/guessimate the VE table above 4000 rpm based on the trend you saw below 4000; then on the second pass, making sure fuel is sufficiently rich, do above 4000 (upto your redline).

[joecar]

AutoVE: isolate VE (disable MAF), disable CL/trims, and apply wideband correction to it.

AutoMAF: isolate MAF (disable VE) disable CL/trims, and apply wideband correction to it.


Calc.MAFT: isolate VE (disable MAF) and apply selective correction to it, calculate MAF from corrected VE.

Calc.VET: isolate MAF (disable VE) and apply selective correction to it, calculate VE from corrected MAF.


selective correction: BEN correction selected from LTFT (for CL) or wideband (for OL/PE).


Calc.MAFT and Calc.VET can be performed with pure wideband in which case CL/trims must be disabled (i.e. selective correction is now from wideband only), this is what AutoVE and AutoMAF are.

[joecar]

Corday, and welcome to the forum :cheers:

[joecar]

So your fuel rail pressure was measured to be 55 psi...?

Does the measured fuel pressure show variation when you quickly/sharply snap the throttle open...?

[joecar]

What HP do you expect to produce...? This can help to calculate the injector size required.

[Corday]

Thanks Joecar =)

Nope i used fuelpressure from others with stock fuelsystem... I will connect a gauge to mine and check. (For now i reverted back stock values in B4001)

I dont know about HP yet.. Its rated at 325 stock and i've done upgrades all-around but nothing extreme.. A guess would be around 400?

Not that it has anything to do with hp, but i I have also had the entire engine dissembled and bored/honed the cylinders. I've installed JE forged pistons, all new bearings and seals, chromemoly pushrods, stainless valves with titanium retainers and dual springs, Yellaterra rockers, LS2 timing chain, LS7 Lifters, LS2 flywheel, Melling oilpump, LS7 clutch, all bolts replaced to ARP, and its been balanced and blueprinted.

I've worked on it since 2004 except for 2009 and 2010 because of health issues.. Im very serious with this car and i really want it to run good.. I have spent countless hours on the mechanics and appearance.

Since end of 2008 the tune has been all that remains to get the car on the road. But ofcourse a mountain of problems arose. Firstly because there are no tuners where i live. Noone even knows what a car like mine "should" run. Its also very hard for me to express to you how the car acts, because that demands english words that i do not know.. Like surge and stall etc etc.

I feel abit dizzy scrolling through the forum trying to form an "image" of the complex PCM and how everything is controlled and works together.. Let alone trying to learn all the abbreviations used :) So know that i am wholeheartedly happy for any help i get with the tuning process..

Ok, so on to the relevant stuff!

I just idled the car up to normal temp and after that i flashed in the SD-AutoVE-tune and disconnected the MAF. It barely manages to idle.. It goes down real low, then up again (to try to "make up" for it i assume), but then dips right back down to almost die (400-600rpm).
11108
Check frame 204 and onward and you can clearly see what happens.

With AutoVE-tune and MAF connected - same thing happened.
11109
First bad "dip" is at frame 204

Flashed back to the stock tune and that made the car idle better. It stayed at ~800 which is about the lowest it can manage without dieing. However, looking at frame 832 makes me worried, because when i push throttle to 20% the AFR goes sky-high to ~21 (or 1,4Lambda).
11110

Can you make anything out of these logfiles?
I have a friend coming over tomorrow to look at the BEN-map on the computer while im driving for the AutoVE process, but now im not even sure if i can/should? Would you?

[joecar]

OK, may be better start with stock tune.

[joecar]

Since you have MAF and NBO2, try this to see if you can get it to run better:
A-New-Twist-on-CALC.-VE-Table..Computing-the-Entire-VE-Table (http://forum.efilive.com/showthread.php?15236-A-New-Twist-on-CALC.-VE-Table..Computing-the-Entire-VE-Table).

That tutorial has you moving away from AFR and thinking in terms of EQR and Lambda;
in your tunetool go Edit->Properties and set Fuel units to EQR and VE units to g*K/kPa.

(in the scantool, make sure that MAP has Metric units (kPa)).

[joecar]

You will learn the acronyms and lingo from the forum :cheers:

We hope your health has greatly improved :cheers:

[Corday]

Btw, AC/belt is removed so its not the ac-clutch causing the idle problems.

Been pondering abit about my NBO2 values.... In the charts i dont see them popping up and down as i have seen in others logfiles and read that they should?

[Corday]

Thank you =) Im better now, was in a bad car accident but luckily survived!

I will check that guide out carefully and start setting up the software for it :]

[Corday]

Phew! Just got back from a 55min log-session! :w00t:

I followed every step of http://forum.efilive.com/showthread.php?15236-A-New-Twist-on-CALC.-VE-Table..Computing-the-Entire-VE-Table. and think it all went perfect! :good:

To be on the safe side however, i decided to up all values in B0101 (Main VE) by 15% as advised in the AutoVE-guide, since i noticed it was running real lean on the stock table.

I hope i got sufficient cellcount in the maps. Did my best to hit as many cells as possible for 50+.

I'd really appreciate some feedback on how it went - since your the best judge of that and not me.. =) Im just happy nothing broke :D

Complete log session

11114

Resulting Calc-VE-MAP (Cellcount)

11117

Resulting Calc-VE-MAP (Average cell-values)

11115

Resulting Calc-Selective BEN (Average cell values)

11116

[WeathermanShawn]

I'll look over your log in more detail this evening.

My first comment is that you need more digits in your MAF SELBEN Table. You have the wrong idea of adding +15% to the VE Table. Your confusing that with AUTOVE which is SD (Open-Loop) tuning.

You need to multiple the SELBENS against your MAF Table B5001 to get accurate fueling, not the VE Table.

I'll comment more later..Good logs though.:)

[Corday]

Thanks alot for looking Shawn :cool:

Hope i didnt ruin the test because of the B0101-change tho? :((
Hmmm how do i get more digits in the SELBEN table? Longer logsession or different driving technique? I take it you mean more digits in the cells pre 2750 and post 9375? If so - i noticed that oddity before i headed out... You have numbers in the cells down to 2000Hz.. Why is it that i have no numbers below 2750? Could it mean a defective MAF or something else?

One thing im curious about, does my NBo2s look as if they are working ok? I think the NBo2-curves look abit strange, however i will admit im no expert in the field=)

[WeathermanShawn]

Your NB's generally look O.K. at cruise. There are some points in your Log where I think you might want to see more amplitude..normally that is just a sign that perhaps your O2 sensors might need replacing. But, it could also be location..headers?

But, overall they averaged out O.K. (Trims, Max, Min, AVR).

By Precision, I mean more digits in MAF SELBEN Table B5001. (See Attachment). On the Map Properties folder, there is a 'Tab' called "Precision". Just increase that to 2 or 3 digits. Are you then multiplying (with Labels) into your Table B5001 in your Tune? If you are, then your next log should all have SELBENS of 1.00. At that point your VE Table should be fine. You didn't hurt anything by adding +15% to the VE Table..the magic of the CALC.VET formula is that it does all the work for you. Its your MAF Calibration at this point that needs to be precise. Let the math do the heavy lifting on the VE Table.

Once you get your SELBENS to ~ 1.00 read the Advanced CALC.VET Summary Notes for techniques on smoothing the MAF and VE Table: http://forum.efilive.com/showthread.php?16280-CALC-VET-Summary-Notes

You actually did pretty good for your first log. You just need another 1-2 logs..:)

[Corday]

Yay cool! =) Big relief
Ops, ok added 3 digits to Precision :doh2:
I didnt make any changes to the tune yet, but will update it now that you have looked at it :cheers:

1. Filtering + "paste with labels" for B0101
2. Filtering + "paste and multiply with labels" for B5001
3. Ready for next log-session

Is that order correct? Or should i skip B0101? Should i change back to the stock values instead?

Hehe, I just discovered the "zoom out" button for the charts... Now i also see the NBo2's popping up and down :D

[WeathermanShawn]

Yay cool! =) Big relief
Ops, ok added 3 digits to Precision :doh2:
I didnt make any changes to the tune yet, but will update it now that you have looked at it :cheers:

1. Filtering + "paste with labels" for B0101
Yes..
2. Filtering + "paste and multiply with labels" for B5001
Yes...
3. Ready for next log-session

Is that order correct? Or should i skip B0101? Should i change back to the stock values instead?
Ideally it would be best to use the B0101 Values without the +15%..But I would use the latest logged Values if possible..

Hehe, I just discovered the "zoom out" button for the charts... Now i also see the NBo2's popping up and down :D

It sounds like you know computers and I think you will be one of those who will successfully tune your vehicle quickly.

Concerning the MAF. There is nothing wrong with your Values. Mine only starts at 2000 Hz because my elevation is 5300 feet above seal level. That lowers the MAF scale vs yours. All is good..

[joecar]

The Calc.VET tutorial calculates a new VE based on the corrected MAF;

hit cells of the new VE overwrite the cells in the previous VE, you have to blend the surrounding non-hit cells...

i.e. if you notice that the new VE hit cells are a 10% increase over the old cells, then you should apply the same increase to the surrounding non-hit cells (i.e. blend)...

but do not necessarily smooth the VE table (there are some ridges caused by things like header resonance).

[joecar]

During a logging session, try to keep throttle constant (or almost) constant for long periods of time.

Also, when varying throttle, try to do so in a slow progressive manner, this gives better data, so the filter will exclude less data.

But, do keep safe while driving doing this.

[joecar]

From your log, you can see:
- DYNCYLAIR_DMA (VE) and CYLAIR_DMA (MAF) diverge,
- CALC.VET and CALC.VEN diverge.

After you apply the Calc.VET tutorial to B0101/B5001 you will see those converging.

[Corday]

Thanks for all help! I have successfully added the results from my first Calc.VE-run to B0101 and B5001.

I am abit confused about the looks of B0101 after the changes.. Because the cells that was not affected by the Calc.VE are alot higher than the new values... I dont know how or if to use smoothing of the other cells?

Uploading a screenshot of the surface map that shows what i mean and the tune-file itself :)
Also, does B5001 look ok?

11731
11732

[joecar]

Hi Corday,

The cells that were not effected should have not moved;

B5001 looks ok except for that dog-leg.


Can you also post the .tun file you logged with and the log file.

[Corday]

Here is the .tun-file that was used just before heading out on the Calc.VE-run and the resulting log-file :)

11738
11739

The other cells didnt move in B0101, they just differ alot from the Calc.VE results.. Maybe you remember I added 15% to the entire B0101 before Calc.VE because i got AutoVE and CalcVE mixed up... I was told it didnt ruin the results, but i assume it contributes to the high peaks i see now when inserting the results from Calc.VE.

[joecar]

Ok, try this, go back to before the 15% increment to B0101, and add say 5% instead, and redo Calc.VET, I think you will get a better result like this.