NefMoto > Technical > Tuning > AVK V6 3.0 2003 8E0 909 559E ECU tuning thread
Pages: [1] 2 3 ... 6
« previous next »
Author Topic: AVK V6 3.0 2003 8E0 909 559E ECU tuning thread  (Read 59925 times)
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« on: August 02, 2015, 11:19:25 PM »
I've updated the first post to just include the map packs and tune for easier reference.

Also, THESE CALIBRATIONS ARE STRICTLY FOR INDIVIDUAL USE, ANY COMMERCIAL USE IS STRICTLY PROHIBITED Not that this warning will necessarily stop anyone, but you know what, as a lawyer myself, I know enough ways to at least require people to spend money lawyering up to defend a claim. So while I would likely never collect any damages, if I ever catch wind if these calibrations in a commercial file, trust me, I'll do my best to get my pound of flesh....

MAP PACK *****NOTE THERE ARE TWO HARDWARE VERSIONS BELOW, 0001 AND 0004, THE TUNE BELOW IS HARDWARE VERSION 0001 AND THE .kp IS HARDWARE VERSION 0004.

THE TWO ARE SLIGHTLY DIFFERENT!!! SO BE CAREFUL IF YOU TRANSFER VALUES.

Note in the .kp There are a bunch of duplicates, however I left them in the regular folder system in German and used the "my maps" portion to name them in English for easier reference. Also included the original .bin

This is the product of hundreds and hundreds of hours boys and Girls, hope it gets some use Smiley Smiley Smiley

FROM PAGE 4:

NOTE I HAVE MOVED THE TUNING BINARY AND THE CONTENTS OF THIS POST TO POST 1 FOR EASIER REFERENCE

I added 3 degrees of timing to the optimal timing areas (12x16) and between 3-8 degrees to the 11x16 KFZW areas. I also found that I was pulling timing above 6000+ RPM and about 70% load, so I left the uppermost RPM areas of the maps alone and subtracted one degree of the timing I added for the surrounding cells.

As indicated, I also changed the intake cam timing. In a nutshell, I extended the intake advance. I also found that extending the long runner intake manifold configuration to about 4200-4400 RPM helped torque production in the areas where you drive the most.  

I edited the KFPED maps to provide better throttle response, rescaled KFMOIP (left the map values alone), added some to the high load KFMIRL values and I richened up the peak torque ares to about 0.84 lambda and dialed it back to about 0.87 as I approach redline.

Hardware on my vehicle:

2003 Audi A4, Quattro, 6 speed manual
-2.5 in catless downpipes
-2.25 in X pipe exhaust with straight through resonators and mufflers
-Zingo mod on the airbox as well as additional ducting into the airbox via the fog light

Environmental factors:
-ambient temps 0-10C
-elevation 2500-3500 ft (approx 13.1 to 12.5 PSI ambient atmospheric pressure)
-low humidity
-94 Octane E10 fuel used *******

WARNING!!!!!l As indicated, this is to be considered a Beta file and under NO circumstances should you assume that it is safe to use on your vehicle in it's present form. It is the end user's responsibility to verify that the settings in this file will be safe for use on your own vehicle.

WARNING!!!! I also had a P1685 (or whatever it was) EEPROM checksum error with a previous version. I believe that I worked it out as I haven't had that problem with this revision, but be aware that it is a possibility. I think I resolved the issue, check the bottom post for the explanation.

WARNING!!! You should only consider using this file unaltered if you have a low restriction exhaust. Any exhaust gas reversion resulting from exhaust restrictions may cause detonation and engine damage with the timing levels I have requested.

If you do have the stock cats in place, I would advise pulling 2-3 degrees out of all the timing tables to start (or more) you can always use the original binary I posted  if the motor will take the timing. However I wouldn't advise just throwing the file on and running it on a bone stock motor.

That said, the cam timing, KFPED and Lambda changes should be OK on a stock motor. But as I said above, proceed with caution in any event if you don't have a full exhaust or if you're close to sea level.

Factors to consider with this particular file is that the timing was set for cold, dry and relatively high elevation. As well, I have access to 94 octane fuel. If you live in a hot, humid, low elevation area without access to 94 octane fuel, you should consider dropping the timing back to stock, or near stock levels raising it from there.

Also, I am offering this file under the assumption that the end user possesses enough knowledge to assess whether or not it is appropriate for use on his/her vehicle. If you use this file, LOG LOG and LOG!!!!

Anyway, enough with the legal talk, hope y'all can find some use for this:

EDIT: PLEASE BE AWARE THAT I AM PUTTING THIS FILE OUT IN THE HOPES THAT WHOEVER USES IT WILL BE WILLING TO OFFER THEIR OWN FEEDBACK AND OBSERVATIONS WITH THE INTENT TO IMPROVE ON IT.

********EDIT, I have not done the P1681 EEPROM error workaround on these files. If you want to do it yourself, change the value at address 33A90 from 2D to 0D. I haven't found it necessary as long as I've been bench flashing it, so I've left well enough alone. You may want to change it if you're flashing via the OBD port as I understand that's when it will give you issues.

One last thing, I made up the full .kp map pack on the 0004 hardware .bin , however when I made the tune, it was on the 0001 .bin ( I lost track of which one I was using after, like, the 50th time I revised the file lol) I included a .kp pack with the 0001 version with a bare bone set of maps that relate to the tune.

However, it's not hard to transfer the maps between hardware versions. just line them up side by side and use the "compare map" feature in winOLS to transfer the maps.

FEB 24, 2016: i HAVE ADDED ANOTHER TUNE WHERE I WENT MORE AGGRESSIVE WITH THE CAM PHASING. I feel that it is a better overall tune, but I will leave both versions for evaluation.

APR 20 2016: So my face is a little red... I realized that I had mixed up ZWOPT AND KFZW in my earlier map packs and tunes. I don't know how I didn't catch this earlier, but I edited the map packs to reflect the changes. I also added more maps to each (especially the 0001 HW version).

I'm leaving the two tunes posted as is because in all honesty, what even though the timing maps were mixed up, the end product seems to work. However, just be mindful of the fact that ZWOPT has more timing added than KFZW if you use the tuned .bin as a basis for any of your own stuff.

JULY 31 2017: Included the last revision of my tune before my car got demolished last October. It was a solid file Smiley

EDIT: I included my latest version as of august.
« Last Edit: July 31, 2017, 08:00:55 PM by mister t » Logged
ddillenger
Hero Member
*****

Karma: +641/-21
Offline Offline

Posts: 5640
« Reply #1 on: August 03, 2015, 04:37:56 AM »
I don't have a damos, but I have done a lot with this file. If you have any questions I will do my best to answer them.
Logged
Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your experience!

Email/Google chat:
DDillenger84(at)gmail(dot)com

Email>PM
Jim_Coupe
Hero Member
*****

Karma: +16/-12
Offline Offline

Posts: 663
« Reply #2 on: August 03, 2015, 05:38:54 AM »
Great Keep up the work I will follow this Smiley
Logged
E85oholic
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #3 on: August 03, 2015, 11:46:27 AM »
So I've picked out about 50-60 possible map locations. I'll post up a really rough list that I've made up in word.

Bear in mind that a lot of these are educated guesses based solely on my read of the 3d map. As well, the addresses listed are just accurate enough for me to find the location again) So DO NOT take them as gospel truth.

The first 20 or so, I used the map locator and the factors from Nyet's 2.7T file.

02B97A Inlet cam timing","8x12","16Bit(LoHi)","","°KW","%","U/min","0.0078125","0.0234375","0.25 1155C,

01163E,lambda maps? 14x14 %correction, %, RPM 0.0078125 / 0.75 / 40
01147A lambda maps? 14x14 %correction, %, RPM 0.0078125 / 0.75 / 40
011398 lambda maps 14x14 %correction, %, RPM 0.0078125 / 0.75 / 40
022C2E lambda  12x16  %correction, %, RPM 0.0078125 / 0.75 / 40

025D02 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.0030517578125","0.25","0.0015259021896696422" X ”25CEA” Y “25CCA”

0115A6 KFMIRL","0x14a1c","Kennfeld für Berechnung Sollfüllung","16x16","16 Bit (LoHi)","-","%","%","U/min","0.0234375","0.00152587890625","0.25","3.984375","190.9921875", S"155A6",  X"15586" Y“15566”

01779A "KFZWMN", TIMING "12x16","8 Bit","-","grad KW","%","Upm","0.75","0.75","40.0" or maybe ,"0.75","0.0234375","0.25"

027A56 "KFKATI","Kennfeld Korrektur Nachstartfaktor über anzti","12x12","16 Bit (LoHi)","-","-","-","Grad C","3.0517578125E-5","1.0","0.75","1.0","1.0","0x8000","0x8000" OR "3.0517578125E-5","6.103515625E-5","0.25","0.949951171875",

024A13 "KFSZDUB","0x1b1e1","Schließzeitkorrektur in Abhängigkeit von UB","16x7","8 Bit","-","-","ms","V","0.03125","0.1","0.0704","0.53125","6.25","0x11","0xc8"

027118
02720C
027300
0273F4
0274E8
0275DC
0276D0
0277C4 "KFPBRK","0x1e3b0","Korrekturfaktor für Brennraumdruck","10x10","16 Bit (LoHi)","-","-","%","U/min","3.0517578125E-5","0.0234375","0.25",

024635 Pedal vs load vs torque 10x10 0.75 / 0.75 / 40

022AB1 Intake Manifold Changeover 8x12 position I/O, % , RPM  1 / 0.75 / 40

028EE0 "KFLMSKHMAG","0x26c1e","Kennfeld Lambda-Motor-Soll bei Katheizen (Mager -Konzept)","8x8","16 Bit (LoHi)","-","-","%","U/min","2.44140625E-4","0.0234375","0.25","1.0","1.0","0x1000","0x1000"

024A13 "KFSZDUB","0x1b1e1","Schließzeitkorrektur in Abhängigkeit von UB","16x7","8 Bit","-","-","ms","V","0.03125","0.1","0.0704","0.53125","6.25","0x11","0xc8"

014F22 06x16 "KFDYES","0x1f8ac","Lastdynamikerkennungsschwelle","6x16","16 Bit (LoHi)","-","%/seg","%","Upm","0.0234375","0.75","40.0","2.203125","19.9921875","0x5e","0x355"

015562 possible 16x16

0179E4 possible 16x5 series

027434 12x12 unknown 8 or 16 bit

027460 12x10 unknown 8 or 16 bit

027E3A Unknown 06x16 factor 0.001525878 (%) X: 0.25 (RPM) Y: 0.0023435 (%)


THE REST ARE PLACES I MARKED FOR LATER, LISTED: ADDRESS/8 OR 16 BIT/ROWS
There may be duplicate addresses of the ones listed above

011C61 8 bit 8x
011F66 8 bit 9x Huh
0162DA 8 bit 12x
01835A 8 bit 12x
01839A 8 bit 12x
019B2A 8 bit 12x
02252A 8 bit 12x
022A0A 8 bit 12x NOTE, this whole area seems to have a lot of 12x? 8 bit maps
022D52 8 bit 12x
022D5F 8 bit 12x (may be 4 or 5x)
022F4A 8 bit 12x
023302 8 bit 12x
0234BA 8 bit 12x
0239DC 8 bit 12x
023D97 8 bit 12x
023EEC 8 bit 12x
0285FC 8 bit 12x
028788 8 bit 12x
02898A 8 bit 12x
029DC0 8 bit 12x
029DC3 8 bit
021C0 16 bit 12x
0125C0 16 bit 8x
013F50 16 bit 6x
0155A8 16 bit
015CD8 16 bit 12x
015FF0 16 bit 12x torque map
016188 to 016770 16 bit 12x (likely torque request maps)
0166CE 16 bit 12x (lambda corrections?)
01747C 16 bit 12x (lambda?)
018E1C 16 bit 12x (lambda?)
019110 16 bit 12x (lambda?)
All the way to 019D4C
022C0E 16 bit 12x
022D5E 16 bit 12x
023490 16 bit 12x (looks like timing???)
0258D8 16 bit 12x
025C20 16 bit 12x
025F38 16 bit 8x
026230 – 026610 16 bit 8x (4 or 5 tables) then all the way to 026D68
027048 16 bit
027060 16 bit
0270EA to 027892 4 or 5 10x10 maps
027346 16 bit (not to sure if 12x or 16x)
028422 16 bit 12x
028C24 16 bit 8x to 029304 (multiple maps)
029EB4 16 bit repeating maps 8x
02A734 16 bit
02B01A 16 bit 11x (w optimized values) looks like throttle map
02B2F0 16 bit 12x  
« Last Edit: August 03, 2015, 08:47:22 PM by mister t » Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #4 on: August 03, 2015, 11:48:35 AM »
UPDATE:

So after a good 100 hours or so, I've made some pretty good progress in identifying about 120+ individual maps.

I'm making the Ols file available here if anyone wants to have a look at it and make some contributions or edit what I've done so far.

Bear in mind that this is still a work in progress so some of the map axes are not yet defined. As well, some of the map factors are just guesses. So again, don't take what I've done so far as being verified or complete.

Also, it doesn't look as though all the axes are defined in the cells adjacent to the map. so what I've done is come up with a selection of addresses which gives you a nice grade of values for both % and RPM in 10-12-14-16 cell ranges. It appears that others have used this approach as well. Anyone know if that will be an issue? Here are the addresses I've been using

0287E0 10-100% X 11 / f0.023435 (16 bit)
0287DE 0-100% X 12 / factor 0.023435 (16 bit)
022C1E 8-105% X 16 / factor 0.75 (8 bit)
015586 0-99% X 16 / factor 0.001516 (16 bit)
015566 480 – 6520 X 16 / factor 0.25 (16 bit)
01558A 1000- 6520 RPM x 14 / factor 0.25 (16 bit)
022C12 600 – 6400 RPM X 12 / factor 40 (8 bit)


Also, there are some sections where there are multiple 8x8 maps and I've only highlighted one. I do realize that there are other maps in that area. I just highlighted on in each so I wouldn't forget where the section was. So feel free to save the rest of the surrounding maps.

What do you all think?
« Last Edit: August 23, 2015, 11:02:24 PM by mister t » Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #5 on: August 03, 2015, 06:15:34 PM »
Hmmm, looks like finding maps manually might be a little easier than I first thought  Grin (knock on wood)

So from other's experience, it looks like most maps that you would want for tuning are generally larger, usually 12x16, 14x14, 10x10 16x16 and such.

What I've started doing is going into the 3d view, setting it to 12 rows, then starting on 8 bit, just scrolling and noting the addresses where I see nice smooth or recognizable patterns.

I've managed to ID quite a few spots that are likely maps.

I'll just keep doing this with 8 rows, 10 rows, 14 rows and 16 rows on 8 bit. Then do the same on 16 bit.

I find a good trick is that you only look for the patterns that are completely uninterrupted across the grid. If you spot something that looks recognizable, but it's jagged, then pass it over until you re-adjust the row numbers to see an uninterrupted shape.

Thoughts on this method?
Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #6 on: August 03, 2015, 06:49:21 PM »
I've also noticed something interesting....

The raw 8 bit values all seem to be 0-240, whereas the 16 bit values are 0-65000. ( I assume it has to do with multiplying integer values)

So knowing the x and Y factor values (i.e. that RPM is usually multiplied by 40 or that timing is often multiplied by 0.75), does that mean you should be able to hunch what format certain tables are in.

Take timing for example, you know that RPM and degrees are the X and Y axes. Both of these values will likely be arrived at by multiplying by either 40 or 0.75. So if values you're working with are between 0-240, does that mean that a table with an x and y axes consisting of these factors will use 0-240 values, i.e. 8 bit ones?

(not sure if this makes sense, but that's what my intuition is telling me).
Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #7 on: August 04, 2015, 06:12:52 PM »
So here's the torque/throttle/RPM maps. I just labelled them Torque 1,2,3etc... for now.

It's not really too bad to do these maps manually once you get the hang of it. Just a matter of how much time you have on your hands lol.



165F0 TORQUE 5 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.003056","0.25","0.001525" X ”16772” Y “1678C”


16470 TORQUE 4 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.003056","0.25","0.001525" X ”16772” Y “1678C”


162F0 TORQUE 3 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.003056","0.25","0.001525" X ”16772” Y “1678C”


16170 TORQUE 2 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.003056","0.25","0.001525" X ”16772” Y “1678C”


15FF0 TORQUE 1 "KFPEDR_relative pedal 16 Bit (LoHi)","-","%","U/min","% PED","0.003056","0.25","0.001525" X ”16772” Y “1678C”
« Last Edit: August 04, 2015, 06:34:06 PM by mister t » Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #8 on: August 04, 2015, 06:20:11 PM »
Target cylinder filling

115A6 TARGET CYLINDER FILLING KFMIRL" ,"Kennfeld für Berechnung Sollfüllung","16x16","16 Bit (LoHi)","-","%","%","U/min","0.0234375","0.00152587890625","0.25","3.984375","190.9921875", Z"155A6",  X"15586" Y“15566”
Logged
nyet
Administrator
Hero Member
*****

Karma: +607/-168
Offline Offline

Posts: 12268


WWW
« Reply #9 on: August 04, 2015, 07:46:16 PM »
Quote from: mister t on August 03, 2015, 06:49:21 PM
The raw 8 bit values all seem to be 0-240, whereas the 16 bit values are 0-65000

What is 1<<8?
What is 1<<16?
Logged
ME7.1 tuning guide
ECUx Plot
ME7Sum checksum
Trim heatmap tool

Please do not ask me for tunes. I'm here to help people make their own.

Do not PM me technical questions! Please, ask all questions on the forums! Doing so will ensure the next person with the same issue gets the opportunity to learn from your ex
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #10 on: August 04, 2015, 09:54:27 PM »
Quote from: nyet on August 04, 2015, 07:46:16 PM
What is 1<<8?
What is 1<<16?

That's what I was getting at with the 'integer' comment earlier. (I'm not sure if I used the term right).

By the way, I wanted to go on record as thinking you for posting those CSV files with all the factors and such.

That was the one document that brought it all together for me.
« Last Edit: August 04, 2015, 09:58:08 PM by mister t » Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #11 on: August 23, 2015, 09:54:01 PM »
BUMP, I've posted an OLS file of my work so far in post #4

There are about 120 individual maps in it so far  Grin  
« Last Edit: August 23, 2015, 09:55:34 PM by mister t » Logged
mister t
Sr. Member
****

Karma: +74/-18
Offline Offline

Posts: 343
« Reply #12 on: August 25, 2015, 08:54:52 PM »
Bump, anyone have any feedback?
Logged
metronet
Full Member
***

Karma: +5/-0
Offline Offline

Posts: 73
« Reply #13 on: August 26, 2015, 02:30:14 PM »
I am learning using the .ols file you posted. Thanks for sharing your work so far.

I found it useful to plug-in the parameter names like KFWKSTT & KFMSNWDK etc...

Here is what I found for ESKONF:

Code:
0C C0 B3 F0 00 02 F3 FF AA FA 55 55 33 -> (0x1121D)
0C C0 B0 F0 00 02 F3 FF AA FA 55 55 33 -> (0x1122A)
0C C0 B3 F0 00 02 F3 FF AA FA 55 55 33 -> (0x11237)
0C C0 B0 F0 00 02 F3 FF AA FA 55 55 33 -> (0x11244)
0C C0 B3 F0 00 02 F3 FF AA FA 55 55 30 -> (0x11251)
0C C0 B0 F0 00 02 F3 FF AA FA 55 55 30 -> (0x1125E)
0C C0 B3 F0 00 02 F3 FF AA FA 55 55 33 -> (0x1126B)
0C C0 B0 F0 00 02 F3 FF AA FA 55 55 33 -> (0x11278)

Anyone knows what the bit pairs are for byte #2 and byte #12?
I am guessing they are for different transmission coding and with/without the electric vacuum pump beside the ABS module?
« Last Edit: August 26, 2015, 07:11:51 PM by metronet » Logged
Jim_Coupe
Hero Member
*****

Karma: +16/-12
Offline Offline

Posts: 663
« Reply #14 on: August 27, 2015, 12:25:59 AM »
Is there no Damos for this already?
Logged
E85oholic
Pages: [1] 2 3 ... 6
  Print  
« previous next »
 
Jump to:  

Navigation
Personal Tools
  • November 23, 2024, 09:54:51 AM
  • Welcome, Guest. Please login or register.
    Did you miss your activation email?

    Login with username, password and session length
Search
 
Advanced Search
About Disclaimer
All Rights Reserved 2010 Nefarious Motosports
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.026 seconds with 17 queries. (Pretty URLs adds 0s, 0q)