Pages: 1 2 [3]
Author Topic: Audi C4 S4/S6 and RS2 4A0907551X and 8D0907551X  (Read 35375 times)
vwnut8392
Sr. Member
****

Karma: +18/-7
Offline Offline

Posts: 271


« Reply #30 on: January 17, 2013, 06:40:52 PM »

Acki-if you used this xdf and improved upon it, why not update it with your findings?



thank you ddillenger, atleast someone thinks like i do instead of just pointing out the faults and telling me to do it myself. has everyone forgot about team work???
Logged
Acki
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 75


« Reply #31 on: February 07, 2013, 12:30:14 PM »

Because to create a XDF with all calcs isn't so easy - at the moment - for me.
In the moment I use a WinOLS Demo and there it's much easier to change axles and to save the values etc.
I would have to do it manual for all tables. This is not the problem, but I won't work on/with a 5cyl.

Do you have maybe some links where I find a easier way "into" the XDF format?
From PRJ the mapfinder works also very well but it need a little bit time to create the labelfile.
Till this I work on OLS and then I will use the findings.

(it's also a lack of time...)
Logged
vwnut8392
Sr. Member
****

Karma: +18/-7
Offline Offline

Posts: 271


« Reply #32 on: February 13, 2013, 12:09:09 PM »

the map finder from PRJ is about as good as it gets for us M2.3.2 guys for now. i have actually been adding all scales to the maps i have fully fleshed out now manually with the help of winOLS. i will post them once i have all the conversions done and it all works right. for me this has to be the most confusing version of motronic i've ever worked with, its like bosch just took an N/A ECU for a 4 cylinder and slapped a boost control board in there and just added code for a 5th cylinder, like just slapped it together and said "here ya go audi, have fun with this one". but looking at it more and more this version of motronic seems capable of running anything from a 4 cylinder to a V8 with code modifications. i've been comparing it to the audi 3.6 V8 ECU on a software and hardware level as well too.
Logged
LHN86
Full Member
***

Karma: +5/-1
Offline Offline

Posts: 63


Engineer


« Reply #33 on: February 17, 2013, 09:04:08 AM »

Hi!

Nice work. Have you found any rev-limiter in the maps?
Logged

BMW E30 M3 S54
IamwhoIam
Hero Member
*****

Karma: +44/-101
Offline Offline

Posts: 1036


« Reply #34 on: February 17, 2013, 02:59:44 PM »

the map finder from PRJ is about as good as it gets for us M2.3.2 guys for now. i have actually been adding all scales to the maps i have fully fleshed out now manually with the help of winOLS. i will post them once i have all the conversions done and it all works right. for me this has to be the most confusing version of motronic i've ever worked with, its like bosch just took an N/A ECU for a 4 cylinder and slapped a boost control board in there and just added code for a 5th cylinder, like just slapped it together and said "here ya go audi, have fun with this one". but looking at it more and more this version of motronic seems capable of running anything from a 4 cylinder to a V8 with code modifications. i've been comparing it to the audi 3.6 V8 ECU on a software and hardware level as well too.

Not sure how many injector drivers are available on the PCB but I know that the number of cylinders is user-configurable directly in the file.
Logged

I have no logs because I have a boost gauge (makes things easier)
prj
Hero Member
*****

Karma: +915/-428
Offline Offline

Posts: 5843


« Reply #35 on: February 17, 2013, 03:48:13 PM »

Not sure how many injector drivers are available on the PCB but I know that the number of cylinders is user-configurable directly in the file.

It's not that simple. You can change that constant if you want, it's only used for two things.
The answer lies in how spark is output in this ECU, which you seem to have no idea about Wink
« Last Edit: February 17, 2013, 03:50:07 PM by prj » Logged

PM's will not be answered, so don't even try.
Log your car properly.
IamwhoIam
Hero Member
*****

Karma: +44/-101
Offline Offline

Posts: 1036


« Reply #36 on: February 18, 2013, 11:59:19 AM »

It's not that simple. You can change that constant if you want, it's only used for two things.
The answer lies in how spark is output in this ECU, which you seem to have no idea about Wink
There are many things I have no idea about, but spark output was not being questioned here. Thanks anyway.
Logged

I have no logs because I have a boost gauge (makes things easier)
prj
Hero Member
*****

Karma: +915/-428
Offline Offline

Posts: 5843


« Reply #37 on: February 19, 2013, 03:28:04 AM »

There are many things I have no idea about, but spark output was not being questioned here. Thanks anyway.
Well, clearly it was hinted at, because you guys think that changing a single constant which tells the ECU degrees between cylinders somehow will change everything...

As I said, it's not that simple, the ECU is full of code like in the image I attached.
It's all hardcoded. No, it can not run an arbitrary number of cylinders, this is compile time. Especially with the I5 ECU, which has a ton of special code written for it.
Note the 5 paths...
Logged

PM's will not be answered, so don't even try.
Log your car properly.
Acki
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 75


« Reply #38 on: February 19, 2013, 11:24:44 AM »

Anybody tried to take a look with Keil C51 on this stuff?
The assembler is pretty useless because you find no step in and knowledges from other languages (C etc) are not helpful because assembler means every operation has to run via some register and this you must find/know Sad

However, the V8 motronics are not able to run I5 because only 2 spark outputs etc.
But with the I5 you can run everything till a V10 (wasted spark configuration) when you know how too modify the code.

Logged
prj
Hero Member
*****

Karma: +915/-428
Offline Offline

Posts: 5843


« Reply #39 on: February 19, 2013, 11:38:24 AM »

But with the I5 you can run everything till a V10 (wasted spark configuration) when you know how too modify the code.
No you can't, you might as well write the ECU from scratch.
The entire ECU is built to run 5 cyl and 5 cyl only. There is a lot more than one chip of the software.

There are 4 main processors on this board, two of which are read-only and one of them has code specific to I5.
Really, just trust me on this, I know a thing or two about all the components in this ECU.

Forget Keil, use IDA Pro to disassemble and ASEM51 to assemble.
Acki, e-mail me, I will help you with logging your Audi V8.
« Last Edit: February 19, 2013, 11:41:01 AM by prj » Logged

PM's will not be answered, so don't even try.
Log your car properly.
Acki
Full Member
***

Karma: +1/-1
Offline Offline

Posts: 75


« Reply #40 on: March 12, 2013, 05:58:59 AM »

Ok thanks. But V10 is 2xI5, add extra drivers for injectors and coil (in parallel) and it's running.
Megasquirt for example makes this viseversa. V10 wasted spark setup is used by the I5 COP guys Wink

I worked yesterday a little bit on the I5 XDF, it contains the same "bad" information like some OLS file I found months ago.
Most is not correct...
Logged
vwnut8392
Sr. Member
****

Karma: +18/-7
Offline Offline

Posts: 271


« Reply #41 on: March 15, 2013, 12:00:20 PM »

Well i worked with the URS4/URS6 4A0 907 551AA chip set a little bit and came up with these XDF's. I know their not perfect but their a step in the right direction. the category labeled "Differences by comparison" are differences i found between comparing the 551C RS2 binary and an MTM stage 1+ file. those maps are different in some way in comparison to the original 551AA file. i've been trying to figure out the axis's for the boost chip with little luck. also i think i have the XDF for the boost chip setup so it will correct the checksum. Im not sure if it actually works as i havn't tested it. Please use these with caution as they aren't perfect and revisions/input to them is much appreciated.
Logged
Pages: 1 2 [3]
  Print  
 
Jump to:  

Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Page created in 0.021 seconds with 16 queries. (Pretty URLs adds 0s, 0q)