PDA

View Full Version : Will be starting on my friends tune soon



Spdfrk1990
January 22nd, 2008, 02:49 PM
I wanted to make this thread to list things along the way and to make sure im doing everything I need to do. The car is a 01 TA ws6 6-speed so very similar to my 01 SS. He just ordered the same NGK wideband that I have also.
Specs on the car are
Comp 224/228 581/588 113lsa
Pacesetters/catless y/catback spintech
Pulley
Slp Lid
Slightly ported TB
Everything else is stock as far as I know.

Friday we just took it to a shop for a dyno and tune. All the shop did was a WOT tune just adjusting the spark and PE table basically. So now once he gets the wideband I want to run auto VE on it and log the maf and follow the idle tutorial on here and adjust the spark some more. He said the issue with idle is only on a cold start and it will idle at 500rpm for a min and want to stumble.
Ok so now you know what im working with I guess my question is where is the best place to start. Should I run the auto ve first up for 4k rpm getting as many cells as possible? Then where should I go from there should I then make a table for the maf and tune it all the way to redline. Then follow the idle tutorial? I will also want some help on adjusting the spark that is one thing im not too sure about.

Spdfrk1990
January 22nd, 2008, 02:52 PM
Here is some tune files and a few logs I did after the tune on friday.
The first tune file is the one b4 the shop touched it and supposely it was a bolt on tune some other shop had done.
The tune2 file is the one after we got the dyno and tune on friday.
The logs are after the tune on the way back.

Eatinstang4life
January 23rd, 2008, 04:54 PM
33 veiws and no posts. Come on guys we need to get this thing tuned and dont know here to start help us out!

Spdfrk1990
January 23rd, 2008, 05:00 PM
Well im sure the autoVE we will want to do first but it would be nice to see what some of the pro's think lol.

ssheets
January 23rd, 2008, 06:28 PM
I wish I could help, but I'm behind you in the learning curve. I will be watching this thread though.:)

5.7ute
January 23rd, 2008, 07:23 PM
I am no expert but this is what I would do.
First log some wot runs to see what AFR & timing the dyno tune is giving you. Once you have done autoVE the PE table will be miles out. The timing table may also be referencing the wrong cells.
Make a map using the B3618 PE/RPM table with wideband AFR as the data. This is what you will want to enter into the B3618 once auto VE is complete.
Double check that the IFR table matches your injectors.
Change B3618 to something sane across the board like 12.5 to 1
Then go through the Auto VE process & get the VE table dialled in.
Enter the B3618 data from the log with the dyno tune.
Dial in the maf.
The timing from the dyno log can now be added into the correct cells. Work up to this timing from a couple of degrees lower keeping an eye out for KR.
This should get you to a better base tune utilising the timing & AFR's you paid for on the dyno.

Spdfrk1990
January 23rd, 2008, 08:10 PM
Ill want to do the VE first tho cause like you said after I do the VE the PE table will be miles out right. Well the timing on the dyno he didnt do much if you down load those tunes and look at the b4 and after it isnt much.

5.7ute
January 23rd, 2008, 08:22 PM
The map with the old dyno tune was only to save the PE AFR settings he found on the dyno. If he didnt do any real work with this or the timing I would be starting with a fresh, unmolested tune.

Spdfrk1990
January 23rd, 2008, 09:16 PM
The first tune I have listed there I saved from his car before it ever went to the dyno. The tune that has 2 after the name is the one that was done on the dyno if you do a compare between the 2 you will see what was done on the dyno. I havnt compared the other tune to a stock one yet but its supposely just a bolt on tune he just worked off it when the car was on the dyno. Soo yeh basically no real work was done just some work to get the HP numbers my friend wanted.

joecar
January 24th, 2008, 05:27 AM
I agree with 5.7ute...

Basically:
- start with first tune file;
- calculate injector flow rate table (if appreciably different, paste in the new values);
- do AutoVE tune (for your mods, increase VE table by 8% rather than 15%);
- do idle tutorial showthread.php?t=4661 (http://forum.efilive.com/showthread.php?t=4661)
- adjust CFOL B3605 and/or PE B3618 to get the best throttle response (street) and/or best peak TQ/HP (track);
- check to see what if any knock is occuring, and take care of it (check that actual AFR is appropriate and/or check that timing is not too high);
- enable MAF and do AutoMAF tune (if planning on running MAF);
- enable closed loop and LTFT's (if planning on running CL/LTFT's) and check LTFT variation.

I would then continue to leave COTP (cat over temp protection) off as this just plainly wastes fuel.

Also: when doing anything with the wideband (or, actually anything where you want high resolution), log no more than 24 channels to get 10 frames/second (i.e. each 1st priority pid gets logged 10x/sec); more than 24 channels logs at a slower rate (5 frames/sec or less).

Cheers,
Joe
:cheers:

Spdfrk1990
January 24th, 2008, 06:54 AM
Thanks Joe,
We were planning on running CL with the MAF and the car is on stock injectors currently and it does not have any cats. Im a little confused on your last statement but basically it looks like dont make the logs too long?

joecar
January 24th, 2008, 07:59 AM
Thanks Joe,
We were planning on running CL with the MAF and the car is on stock injectors currently and it does not have any cats. Im a little confused on your last statement but basically it looks like dont make the logs too long?Oh, what I'm trying to say is: with AutoVE you want 10 frames/sec... so you have to limit the number of pids logged such that no more than 24 channels are used.

If your pids occupy more than 24 channels, you will get only 5 frames/sec or less; this means you may miss some transitions, and your AutoVE attempts will not be as accurate.

It has nothing to do with the length of a log, but it has to do with how often each pid is updated: at 10 frames/sec each pid is updated 10 times per second; at 5 frames/sec each pid is updated 5 times per second.

ssheets
January 24th, 2008, 08:03 AM
Oh, what I'm trying to say is: with AutoVE you want 10 frames/sec... so you have to limit the number of pids logged such that no more than 24 channels are used.

If your pids occupy more than 24 channels, you will get only 5 frames/sec or less; this means you may miss some transitions, and your AutoVE attempts will not be as accurate.

It has nothing to do with the length of a log, but it has to do with how often each pid is updated: at 10 frames/sec each pid is updated 10 times per second; at 5 frames/sec each pid is updated 5 times per second.That's a great explanation. :)

Spdfrk1990
January 24th, 2008, 08:19 AM
Thanks Joe someone else explained to me what you meant also. I try to keep only the PID's I need at the time.

ssheets
January 24th, 2008, 09:13 AM
Thanks Joe someone else explained to me what you meant also. I try to keep only the PID's I need at the time.
That's been one of my challengs. Learning what PIDs you really need for each logging task? Some stuff you just want, but you can't afford the overhead.

Another thing that I'm trying to figure out is exactly what driving conditions along with the PIDs to perform for specific types of tuning logs. I guess it makes since to drive it under the same conditions your tuning for. I was looking at WOT stuff, but then ran into my idle issues so I'll work on that first.