My quest to tune my car...

[Crunkmaro]

i hope this is allowed. i just wanted to make a thread that i can update with various changes, issues, solutions, advice etc as my tune progresses.

if its too cumbersome, then delete/move, i won't be offended :-) i'll just have to stick to notebook and pen!

Thanks to you guys, my trans tune/lock-up issue is fixed.

I have posted my starting tune, some stuff is odd (its an ls1 in an lt1 car, some of the tables for the gauges had to be 'manipulated') so if anyone uses the tune (don't know why you would anyway), beware!

Here is the car as of 8/1/2011

ls1, a4 trans, 3600 Torque converter, 2.73 gears, ms3 cam ( 237/242 .603/.609 at 112lsa), longtubes + full exhaust.

Motor other than cam is stock, stock ls1 intake, stock TB, stock 241 casting heads

I am running on a 98 PCM

First step is to get the car running normal at idle...i am going to read some of the idle threads on here (all 50 of them...:p) and i will post up what i changed...

if anyone has any suggestions/opinions at any point, please, by all means, feel free to tell me!




11537

[ScarabEpic22]

What wideband do you have? Thats a pretty stock engine setup, you might have good luck using the Calc.VET tuning method to dial your VE and MAF tables in. Dont know how to do it on a LS1A PCM, it was written for the LS1B PCM (99-02). Not sure if you want to invest in a 99-02 PCM swap or not, depends what your goals are for the car.

[joecar]

The Calc.VET tutorial shows how to rig up a dynamic air temperature pid for the 1998/LS1A PCM.

[Crunkmaro]

i will give that tutorial a full analysis! thanks alot.


still using NB02s...wideband is on order but i have another project to finish first :-/

[Crunkmaro]

okay so i'm going to upload my previous tune prior to adding 10% to ve tables and go try out the calc.vet tuning

in order to get my idle trims set (running RICH), can i utilize the calc.vet method? would i be able to just use the scan tool to command 800, 900, 1000, 1100, 1200 rpms to get trims? i may have missed this portion in the idle tips/tricks thread

[ScarabEpic22]

The Calc.VET tutorial shows how to rig up a dynamic air temperature pid for the 1998/LS1A PCM.

:doh2::doh2: Didnt know that Joe, its been a few months since I read it and Ive only used the Calc.VE tutorial not the Calc.VET one because I didnt have a wideband on the truck I was tuning.

[Crunkmaro]

:doh2::doh2: Didnt know that Joe, its been a few months since I read it and Ive only used the Calc.VE tutorial not the Calc.VET one because I didnt have a wideband on the truck I was tuning.


yeah, theres lots of hidden info in there that i didn't see until i read the WHOLE thing in its entirety a couple times...like where the frick i'm supposed to put the calc_vet.txt file lol

i'll be out tomorrow hopefully getting my fuel trims dialed in.

i just hope i can use this same method to get fuel trims between 800-1000 under control...
if i idle with my windows open my eyes burn from the exhaust fumes

[ScarabEpic22]

yeah, theres lots of hidden info in there that i didn't see until i read the WHOLE thing in its entirety a couple times...like where the frick i'm supposed to put the calc_vet.txt file lol

i'll be out tomorrow hopefully getting my fuel trims dialed in.

i just hope i can use this same method to get fuel trims between 800-1000 under control...
if i idle with my windows open my eyes burn from the exhaust fumes

Yep I had to read it 2 or 3 times before I tried it, always picked up some small fact that was important the next time reading it.

Ugh, burning your eyes at idle? Doesnt sound like any fun, shouldnt be doing that with a mostly stock LS1 + cam setup.

Have you read the idle stickies? LS1 Idle Sticky (http://forum.efilive.com/showthread.php?14439-Idle-Tips-amp-Tricks)

[Crunkmaro]

okay first log...seems to be an issue getting into closed loop?

did a quick log just to see if it was working and i ended up with 0 values as averages in some of the cells in map A...

gonna go try this again

11551

[Crunkmaro]

when i am doing my log i am noticing that my CL factor is always 1...which i'm guessing means i am still in open loop? would a slow o2 sensor prevent me from being in closed loop?

second log...

theres some funky numbers in the ve table in that log...idk whats going on?

11552

[Crunkmaro]

third log. same issue pretty much.

can i use the values i get when i average map A and map B without applying the closed loop filter?

Every time i apply the closed loop filter all my values drop to 0

third log attached:

11553

[ScarabEpic22]

OK start over and log the Fuelsys PID, it will tell you if you're actually getting into CL or not.

[joecar]

CL=1 means Closed Loop if engine is at operating temperature and your have LTFT/trimming enabled.

[joecar]

I'm out of town this week so I can't view your files.

[Crunkmaro]

okay this is a picture of map A before filtering anything except for low cell counts:

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapacalcvet1b4filter.jpg

this is a picture of map A after using the calc_vet filter

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapacalcvet1afterfilter.jpg


this is a picture with the CL filter...

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapacalcvet1afterfilterandCL0.jpg

this is a picture of map B before filtering anything but low cell counts

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapbcalcvet1b4filter.jpg

this is a picture of map b after calc_vet

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapbcalcvet1afterfilter.jpg

this is a picture of map b after CL filter

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapbcalcvet1afterfilterandCL0.jpg



This is what my CL filter is set up to look like...per tutorial:

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofCLfilter.jpg


but then when i change calc_vet filter to EXCLUDE all frames where calc.cl is less than 1 like so...

http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofCalcvetwithclfilterincluded.jpg

which when i apply to my map A...i end up with a map like so:




http://i87.photobucket.com/albums/k145/crunkmaro/EFILIVE/picofmapacalcvetaftercombofilter.jpg

[joecar]

Ok, change the filter to include CL=1 (or to exclude CL=0).

Think about what CL stands for... :)

[Crunkmaro]

Ok, change the filter to include CL=1 (or to exclude CL=0).

Think about what CL stands for... :)


:-) it made sense to me, but unfortunately, so does mixing dr. pepper and moscato wine....

thanks a lot. i now know i can trust those values (relatively speaking of course) to make minor tweaks until my wb02 situation is resolved.

<3

[Crunkmaro]

next step is getting a decent log. the logs i have only have about 30 miles or so of travel on them (which is eqv to an hours worth of driving around here). they don't look like any of the logs that the other threads have in them.

is there a safe "goal" for counts per cell? some sort of target number i should achieve in order to get a better average?

[joecar]

Try to get better than 10 hits per cell.

Operate the throttle is a smooth steady manner, this produces better data (the filter throws away less of it).

[Crunkmaro]

my question now is about smoothing...


i took a log, did the copy/paste with labels in the ve main and the b5001 maf table (multiply with labels in that table)

in the ve table, there were tons of spikes everywhere, should i utilize smoothing in order to get the curve back?

[joecar]

Leave ripples/sawtooth patterns alone. Push down tall spikes and pull up deep wells.

Post screenshoots.

[Crunkmaro]

this is a copy of that scan:

11659



this is a copy of the tune that i'm starting with:
11660

[Crunkmaro]

so after a log and filtering i have a calcvet map that looks like this:

11665

this is my stock ve table values:


11661

this is my stock ve 2d graph

11663

this is my altered per map A ve table data:

11662


this is what the graph ends up looking like

11664

[joecar]

Can you post a screenshot of the difference between the old and new VE tables... do it like this:

open the edited file as the calibration file, open the previous file as the alternate file, goto table B0101, and post screenshots of each of the 3 differences (the buttons labelled +/-, %, [%]).

[ sorry, I'm still in Palm Springs so I can't view your files yet ]

[Crunkmaro]

table b0101??> charge temperature blending?

maybe its supposed to be the ve table. ill get that up:'





11670

11671

11672



i smoothed out some of the huge peaks and dips.

[joecar]

B0101 is the main ve table, isn't it...?

Your differences pics show a large difference before->after...

[Crunkmaro]

B0101 is the main ve table, isn't it...?

Your differences pics show a large difference before->after...

maybe its different in the 98 OS? i know a couple of the tables mentioned in the idle tutorial don't exist in my 98 OS.

the differences are insane...thats why i am so leery about this. the car runs VERY rich at idle (800 idle x 55 MAP) to the point where i can see and smell fumes while sitting at a stop light in 100 degree weather

[joecar]

B0101 is the Main VE table in the 1998 file... if you're doing the Calc.VET tutorial (MAF'd and CL/trims) then you should be pasting into B0101.

[joecar]

I see the ETC warning when I open you .tun file... have you been flashing this file, does the engine run ok...?

[maudyZ28]

it should run ok, I get the warning too Joecar with my 99 Camaro. Has cable throttle anyhow.

The VE do look different but not in a bad way, the stock 98-99 VE tables i found were quite close at high MAP/High RPM but way over at low RPM.
Just doing Auto VE with my wide band for my Heads cam camaro (253 stock ls2 heads with TSP 228R cam), i'm struggling with idle also a bit, your must be much harder being a larger cam??
Are you in CL when it smells rich?? The AFR should be fluctuating about 14.7 ish so the fuel mix should be right?

[joecar]

Ok that's good, I just wanted to make sure it was ok.

[Crunkmaro]

I see the ETC warning when I open you .tun file... have you been flashing this file, does the engine run ok...?


yeahh its been like this ever since i got the car runnign with the first file. i've flashed maybe 30 times? that warning shows up every now and then.



it should run ok, I get the warning too Joecar with my 99 Camaro. Has cable throttle anyhow.

The VE do look different but not in a bad way, the stock 98-99 VE tables i found were quite close at high MAP/High RPM but way over at low RPM.
Just doing Auto VE with my wide band for my Heads cam camaro (253 stock ls2 heads with TSP 228R cam), i'm struggling with idle also a bit, your must be much harder being a larger cam??
Are you in CL when it smells rich?? The AFR should be fluctuating about 14.7 ish so the fuel mix should be right?

my idle sits pretty still for the most part, i am in closed loop, it was much much worse initially before i started doing calcvet...





so that zigzaggy pattern is normal? i'll flash it in and see what happens lol

[Crunkmaro]

okay back from vacation.

back to work lol


i'm going to post up my 'progress' so to speak...

tune 01 is after calcve, i pushed up somedips and pulled down some spikes but it is pretty much exactly what i scanned.

11741


now here is a log after that tune was flashed in, and ltfts i forgot to clear...

11742



i used that to modify maf calibration and the ve tables once...i also checked out the idle tutorial and added spark and air to the corresponding tables(per tutorial) i also added to the startup friction table in order to get my car to start on the first crank versus having to pedal it to start...i noticed that now i have surge issues as well as stalling out a couple times in low speed manuevers (bumper to bumper traffic)

here is my updated tune...

11743


i did a log afterwords...LTFTS and selbens seem to be settling down

11744


i used that log to modify the ve table and maf calibration once again...also decreased startup friction tables by half...the surging isn't so bad but on warm startups, going from Park to R/D causes the car to stall... this is tune 03...

11745


the log from that is in the next post...ltfts and selbens are slowly but surely approaching zero with occasional spikes...


TBC....

[Crunkmaro]

here is the log after flashing in the third tune...


11746


its getting better...

warm startups still have the same issue...as soon as i go from P/N to Drive gear car dies out...now to check out the idle tutorial and check out the spark section.

i've also utilized the logging off sae.maf vs ECT temps in order to determine desired airflow...

[Crunkmaro]

quick question to whoever is looking out there..

in the idle tutorial, its mentioned to log sae.maf vs ect

now, those values go in both the P/N as well as INGEAR columns correct?

[joecar]

They go into whichever column matches the gear lever (i.e. P/N or in-gear).

[joecar]

You will need wideband for the non-closed loop areas of VE and MAF.

In the meantime, you can take the trend and apply it up there.

You can locally smooth the VE table, fill in the inverted spikes, don't do more than that.

How does it run, do you notice any difference.

[Crunkmaro]

They go into whichever column matches the gear lever (i.e. P/N or in-gear).


hahaha...:p



You will need wideband for the non-closed loop areas of VE and MAF.

In the meantime, you can take the trend and apply it up there.

You can locally smooth the VE table, fill in the inverted spikes, don't do more than that.

How does it run, do you notice any difference.


it runs a LOT better. the hesitation upon applying the throttle is gone, the car pulls smoothly instead of that "sunken stomach" feeling i used to get when there'd be a noticeable loss of power in the curve.

calcvet has made a HUGE difference.

until i get my wideband in (unfortunately i have to wait until my twin turbo project is done before i can drop the cash on that) i want to focus more on driveability...

the only problems i have as of now is intermittent idle surging.

i said in an above post that after the 03 file was flashed, i was having an issue with the car stalling out. i had done the idle logging to get my desired airflow in...but i had only transfered that information into the p/n column. i did another log with the car in gear (hooray for parking brakes) and utilized those values for ingear column. i went out for a quick ride and the car seemed to be doing much better. *knock on wood*

i kept reducing down the startup airflow table until i know its too low (car won't start normally)...now i should be able to up it ever so slightly so i can get a nice clean startup first time around.


w00t

[joecar]

Cool, good job :cheers:

[Crunkmaro]

Cool, good job :cheers:


well...more like good job to you.


i pretty much went through blackflag's thread to get my stuff figured out...Shawn's (rip) and yours comments are the only reason i figured out how to use a filter, how to set up a map, how to log, how to change tables without manually doing them etc etc etc.



i'm done for the night...updated my tune file to raise up the start up friction table a bit so i can get easy starts, now i'll drive the thing for a couple days and see how it reacts and learns.



Ps: this forum beats any journal/diary i've ever kept