NefMoto

Hey guys,

let us start a discusion about running the 551K/Q RS4 software on a 551M S4 ECU. :)


What we know is:

+ The ME7.1 hardware is identical between both ECUs.
+ The 32k CPU ROM binary is equal.
+ EEPROM data structure is also the same.

- The MAF sensor, injectors, K03 turbos and the throttle are different to the RS4, but that is not important for flashing the ECU.

So it must be possible to flash the original 551K/Q file into a 551M ECU.

A obstacle could be the different partnumbers between the original and the new software, which are stored in flash and EEPROM (address hex 1C2 - 1CB).
So I think it is impossible to flash over OBD and must be done in boot mode.

ECU coding can also be a problem.
The 551K ECU can only be fix coded with value "10711" for a manual gearbox and exhaust emissions EURO III.
This is set by "VARDEF" in flash and is also stored in EEPROM (address hex 7A, 7B and 8A, 8B).

The 551Q is coded with "5711"
The 551M is coded with "6611"

I don`t know what happends if the "VARDEF" value in flash, is not the same as the EEPROM data.

A table with all ECU coding variants and their meaning, will be find at page 42 in the Bosch ME7.1 function sheet.

Now we must find somebody with a 551M who do the first step and goes it on... ;)

Later we can adapt the involved maps for a fine tuning.

What do you think?

I know how to do MAF conversions... but i dont know how to make throttle changes. What maps have to be changed for that?

Also, I dont understand how VARDEF restricts engine coding..

All the other ecus have various VARDEFs as well, what prevents me from changing the coding?

Also, what cluster code to use?

Sometime after next wed I plan on swapping my tuned MAF/Fueling/PID into a K-box bin (flashed on an M-box) to see what issues arise regarding the EEPROM, cluster, VARDEF etc.  I'll report back when I do it, but that won't be for a week or so.

Thoughts about this oil temp based rev-limiter mentioned here?
http://nefariousmotorsports.com/forum/index.php?topic=2169.msg20878#msg20878

I have a spare ecu.  The ecu I have is from a A6 2.7t 6mt.  It was 95040-boot-mode reflashed to run M-box.  I have verified that the m-box is on there despite it being the A6 2.7t mt ecu.  4R box I believe, don't quote me.  Anyways, here is exactly what I am going to do.

I am going to swap KRKTE, TVUB, MLHFM, KFMLDMX, KFMLDMN, and MLOFS.  I will change nothing else.  Correct checksum, I'll flash right over my working m-box ECU.  I have a spare ecu so in case something goes truly wrong, no harm in ruining my daily driver.  I'll put the old ecu back and whatever.

Sound good?  I'll do nothing else and report back.

Sounds good.

I love you guys.  ;D

If the 551M ECU cannot be fully converted to a 551K ECU via OBD flashing, then I will fix the NefMoto flashing tool.

Can you please give me details on the codes and information necessary to recode the ecu to "5711" and "10711"?

I assume that's necessary, right?

In what order shoudl this be done?
Correct please:

a)  Swap in spare ecu.
b)  Flash on latest working M-box ecu.
c)  Verify that car starts twice and things are "normal".
d)  Recode 5711.  Recode 10711.  Two steps?  What order?  Login codes?
e)  Flash K-box.
f)  Restart car?

Sound right?  Please assist.

I think you should stick to trying 6711 if you have a 6sp car with ESP, 6611 if no ESP. hopefully rs4boost is wrong about it only accepting the other value?

If he is RIGHT, we need to figure out which coding will work.

Should we be using the 551Q RS4 ECU image instead of the 551K RS4 ECU image? Looking at the software part numbers the Q version is newer than the K version.

I believe the current belief is that the Q-box is the best available box for k04's with no cat's or rear o2's.

Ok guys,

today I will try to change my original 551K ECU coding from "10711" to "6611" or "6711", to test it out.
Then we know whether it works and what happens.

If the 551K software not accept an other coding by VCDS, we must change "VARDEF" in flash and look what happens.
In this case I would suggest, jibberjive could insert a changed "VARDEF" in his software to try it out.

Regards

For reference


wow that sucks. acrobat reader blows ass. why doesn't copy/paste maintain whitespace?

(http://www10.pic-upload.de/02.08.12/f8olicv5qdqs.jpg) (http://www.pic-upload.de/view-15412904/VARDEF.jpg.html)

Hi guys,

today I have tried to recode my original 551K ECU with VCDS to "6611" and "6711",

but there is no chance to coding this ECU to something other than the original "10711" code.

After a try, there is immediately the old "10711" code stored and activ.

So I would suggest, jibberjive could try to change the "VARDEF" in his 551K software, to the coding value which is actualy stored in his ECU, and then he flash it.

So there can`t be a conflict between "VARDEF" in flash and coding in the EEPROM.

(http://www10.pic-upload.de/02.08.12/wioii42vb7nl.jpg) (http://www.pic-upload.de/view-15412964/20120802_131700.jpg.html)
 

Is there no VARDEF in M/H/T box etc?

What allows other boxes to have codings other than their respective VARDEF?

Also, is there a problem if its 10 rather than 6?

Do you think it will still pass US smog?

Perhaps 10 is ok?

551M VARDEF is 6611, but I don`t know why the 551M can be recoded and the 551K not.

There must be other stuff in the software that allow or deny recoding the ECU.

One differnet by coding is that EURO III cars have no evaporation test for the gas tank, how in the US.

Not to insult your intelligence, but did you try the recode with the engine running or engine off/on bench. The ECU will not accept the recode with a running engine. My assumption is that the recode isn't even necessary though.

You will not pass OBD emissions with unsupported readiness. Some states will allow one test, so the K box should be acceptable for some, but the Q box could cause problems with emissions testing.

Depending on the state, agreed. The irony is that in CA, the testing standards are so insanely crazy that a lot of test stations simply don't care about a lot of things they should, because if all test stations were 100% anal retentive, about 1% of used cars would actually pass. LOL @ CA.

Of course in my car and NOT by a runing engine... :D
I know how a ECU`s must be recoded.
Also tryed with a valid login. ;)

btw
I have posted new diagnostic and logging stuff here:

http://nefariousmotorsports.com/forum/index.php?topic=2349.0title=

 

Any news here?

Whats the actual status with jibberjive`s test?

I will have my car running again by next week and will definitely do some testing.

Nice to hear thats going on.
Can`t wait to read some good news...

Hint: There are two coding tables within the ROM that list any and all valid codes that the ECU can accept. Altering this table will allow you to both change what a coding does, and what coding is allowed. The code table is often followed by a table that defines what flags get used for each softcode.

Is the coding table located in the first 32k of the ROM?  Is this what Thomas has been saying all along?
Thomas:  My m-box car was originally an a-box.  I run code 06611 on my m-box tuned rom.  Is there any way you can modify the correct location in the first 32k to permit 06611?  If I am not understanding the structure here, is there a diagram which describes the relationship between the EEPROM, CPUROM, and 1024kb ECU-ROM?  Haven't you already verified that the EEPROM and CPUROM are identical?  This is why I'm confused by previous statements.  Does the EEPROM or CPUROM need to be modified for the flash to work?  It sounds like the answer is yes, but the guy above me thinks there's a table for permissible-codes....

concerning coding:
551K you can try 05711.


BR

The 32K CPUROM can`t be changed, modified or flashed. It is a fix binary on the CPU.
We have found that the 551K/Q CPUROM is equal to the 551M and most other newer VAG ME7.1.
So there is no need to make changes on the CPUROM.

I had said that the EEPROM structure is identical between both ECUs (VAG ME7.1), but parts of the content is ECU and car specific (WFS, login, coding, partnumber, VIN...).
The differents must be checked out and adjusted.
Coding is stored in the EEPROM and in the "VARDEF" value of the 1024k flash.
Both we can change, but we have no idea what happend when we do this.

I Believe, that we can flash the 551K/Q software over the 551M without problems.
If the different partnumbers will be checked by the internal OBD flash routine, it must be flashed in bootmode.
I believe also that we can use the original ECU coding of the 551K/Q, if we have successful flashed.

We don`t have answers to this case, so it must be tested...

@newt
Do you have more detailed informations for the coding tables?

Interesting thread... may I ask why the shift to the Euro flash and moving away from the M-box flash?

16-bit load, k04 pid etc. etc.

You're right, I forgot about that...

Sorry I can't be of more help on this platform :(

What is the map limit?

2559mbar

Is anyone aware off the fact that you change other important modules like knock recognition and intake model that will not fit the hardware by the software change?

I worry about this, but I have a guinea pig...  he wants a "new tune" that is "cutting edge".
Frankly I'm excited by this project.  I guess we'd have to compare the modules for knock recognition and the intake setup, but we'd run into problems if the modelling dimension of "parameter-space" grew too large.  The FR has enough explicit detail if the board's commentaries are joined with the FR.  In fact, that's a OT-thread unto itself:  "pages in the FR which have threads associated", which could be quite useful.

Totally agree with you on the impact of RS-altered hardware not present on the S-model... can this be fixed?

there are builds with significantly different motors and intakes that work fine on M-box... and are arguably much more different from stock than RS4 is to S4.

thats not to say they could be calibrated better by properly taking into account those differences, but there doesn't seem any serious issues that come as a result.

What sort of changes do you think most important and require attention?

You are definately right that it will work somehow but it also depends on how the car is used. If you use it only as a drag car on a sunny day you won't be bothered as much by some cut corners as you will use it as a DD in winter.

I have no first hand experience with a RS4 file on a S4 but from my general calibration experience I'd say that the main topic will be the intake model as you have a different throttle body and intake manifold. This will influence cold start, idle speed control and dynamic fueling behaviour mostly. So taking over the S4 values in that deparment will most probably help. Unfortunaltely this can't be done that easily in some cases as the RS4 uses an other structure for throttle angle precontrol in the start phase for example. Also taking over the knock recognition stuff can't hurt if you are on stock engine hardware as the noise behaviour of the RS4 engine with different pistons and cams should not perfectly fit the S4.

Maybe I'm just a bit hypersensitive on stuff like this because of my working background and all of this is no problem and works fine for most needs, but I wanted to make aware of it.  ;)   

Just in case someone is comparing the files and is wondering why start throttle angle is 0 on the RS4:

The posted .ols is missing the KFWDKTHO map.

(http://img204.imageshack.us/img204/6684/kfwdktho.jpg) (http://imageshack.us/photo/my-images/204/kfwdktho.jpg/)

thanks for the heads!  In reality many of us with the serious builds are running the rs4 intake pieces and or heads so it should greatly improve those same areas.  I have attempted to make changes for it yet still to this day i do not know if they worked.

Hi -

I took a risk and tried starting the car with a basic q-box setup.  Here's what I did:  TVUB, KRKTE, MLHFM, MLOFS, KFMLDMX, KFMLDMN, KFKHFM, TABGSS, TABGSS2, TEMIN modified to have the same values as my running m-box interpolated correctly.  Please recall I'm running a modified M-box on a originally A-box car and have K04 turbos.

The car started up immediately four times in a row, sounded perfect and started very smoothly.  It then immediately shut-off.
This must be pedal related.  It shut off within 2, maybe even 1 second of startup.  However the RPM actually went upto about 1200 rpm and it wasn't just sputtering, it was 100% running for sure...   The car did not go into any kind of bad mode, except that it shut the hell off...  LOL.

What can be the solution to this?  I know this must be pedal related.  I was quite surprised this did not brick my ECU.  I reflashed the old m-box tune, did two startups to check, and its again like nothing happened.  I think the risk is low here, we just need to figure out the problem to convert the pedal maps.  That's my guess.  Somethings tells me snowtrooper is required.  LOL

What makes you thinks it is pedal related? Did it shut off when you did a tip in or just like that?
Using a bigger throttle body with a S4 file is not that big of a deal because the start and afterstart phase is calibrated with the small throttle body so worst that can happen is your rpm overshoot is too high. The other way round it's not that easy.

Try it with KLAF copied from a S4 file or better all variables of BGMSZS.
If this doesn't help you have to log the start with the varibles I have told you plus wdksom_w

I've never heard of those modules, so I'll try to look them up tomorrow.  Also, I had two DTC's.
P0600 and P0562.  P0562 was just the typical voltage too low, I always get something like this that I clear off after flashing.

However P0600 did not go away after clearing nor after multi-starts.  (Start-stops, or whatever they were...)

16984 P0600 Serial Comm. Link (Data Bus) Message Missing

That's what I found for this DTC.  What could this mean?

Ok,
only 1 real DTC is not bad for the first try.

DTC 16984 - P0600 means "Datenbus Antrieb fehlende Botschaft" => "CAN data bus malfunction / CAN message missing / CAN time out"

I believe that`s a problem with an expected CAN devices which is not there.

Can it be that you have no ASR/ESP in your car?

The "7" in the ECU coding of the 551K (coding 10711) and the 551Q (coding 5711), means that the car has ASR/ESP over CAN bus.

So it can be a coding failure.

ASR/ESP is a critical device.
If there is any failure, the ECU runs in limp mode and the car has no power.

Ok.  Are you suggesting to reflash but this time try 05611 sans 05711?
What variables should i log?  I read silent bobs post, should i just try that?

Does your car have ESP?

I have never had an issue with an esp file on a box cars that lack it physically.

No, car does not have ESP.  My car is currently coded to 06611.  I'll reflash and then read the code that is on there.

Ok, then we know what`s going on with this DTC.
The engine ECU is waiting for a CAN communication with the ESP control unit and gets no answer.

Now we have a problem, because the 551K and Q are fix coded and can`t be recoded by a workshop diagnostic tool.
I have tested this on my RS4 last week.

The original Euro RS4 (551K) was ever build with ESP.
10 = Euro III (with CAT control and with SAS), 7 = ASR/ESP with CAN, 1 = manual gearbox, 1 = vehicle type B-Class.

Also the japanese RS4 551Q version.
5 = EU3-D (without CAT control and without SAS),  7 = ASR/ESP with CAN, 1 = manual gearbox, 1 = vehicle type B-Class.

Now we must have a look to the coding mechanism of "VARDEF", or both RS4 versions can only be used with cars which have ASR/ESP over CAN.


What`s with the problem-solving approach of "newt"?

Shouldn't "10711" and "05711" be stored in the VARDEF tables of both the K-box and Q-box respectively?  Can we search the assembly code for this value?  It should be a 5-digit integer, or?

VARDEF is located in a 16bit LoHi value at hex 155D6.
D7 29 for dec 10711 in the 551K
4F 16 for dec 5711 in the 551Q

(http://www7.pic-upload.de/20.08.12/b5q2glmz5idz.jpg) (http://www.pic-upload.de/view-15689429/VARDEF.jpg.html)

So just change this to 06611 and retry?  What will occur because ASR has been disabled by VARDEF?

For the Tuner-Pro users out there:

Yes, you are right.
Please change VARDEF in your file to 6611 and flash it back to the ECU.
With this doing I believe that we can switch off the expected ESP signal.
Nothing happens with your car, because you have no ASR, but I hope the DTC will disappear.
We know now that we do not break anything by flashing a 551K/Q into a 551M or similar.

The second possibility is to change only the ASR/ESP digit in the VARDEF value of the original RS4 file.

That means 10611 for the 551K or 5611 for the 551Q.

btw
the_crazy_test.bin is the right name...  ;D ;D ;D

Coded VARDEF to 06611.  Checksummed, flashed.  Cleared codes.  Started three times, same-story.
Read the codes.  Here are the DTC's:
      <DTC>1536  <Status>100     P1536:   Intake Camshaft Contr.Circ.,Bank2 Open
      <DTC>4200  <Status>100     P4200:   Can't find anything.  (?)
      <DTC>16     <Status>100     P0016:   Can't find anything. (?)
      <DTC>1048  <Status>100     P1048:   Can't find anything. (?)
      <DTC>1042  <Status>100     P1042:   P1042 Injector 2, supply voltage - circuit malfunction
      <DTC>1299  <Status>104     P1299:   Fuel metering solenoid - circuit malfunction

Douglas Adams would be proud of me!  Now to lookup what this is and how to fix this.

Should I perhaps try recoding the VARDEF to 05611 for the q-box?
I would try this but isn't there a specialty procedure to login and set this?
<looking stuff up>   yes found it:   http://s4wiki.com/wiki/Engine_recode

I'll go test 05611 and use what's above.  Let me first check the DTC error overlap probabilities above.

Recoded VARDEF to be 05611.  Flashed ROM.  Cleared codes.  Logged in and changed coding to 05611.
Started 3 times, same story.  BEWARE:  Must use VAGCOM to clear codes after this.  Car can't be recoded to 06611 after it's been set to 05611 and failed.  If you don't use VAGCOM to clear codes, when you return to Nefmoto you will get a security write access error.  If you clear the codes using VAGCOM, then you can reflash to old binary, then after you clear the codes again, you can recode to 06611 like original.  BE CAREFUL OF THIS.

Anyways, I made the mistake of not taking VAGCOM output for my 06611 VARDEF previous post earlier, but the codes were right.  Here's my 05611 VARDEF attempt, and of course the warning from the previous paragraph!

Control Module Part Number: 8D0 907 551 Q
  Component and/or Version: 2.7l V6/5VT     G   0001
           Software Coding: 05611
            Work Shop Code: WSC 87119 
           Additional Info: XXXXXXXXXXXXXXXXX     XXXXXXXXXXXXXX
                      VCID: 3871D1E91848AA3
6 Faults Found:

18058 - Powertrain Data Bus
            P1650 - 35-00 - Missing Message from Instrument Cluster
17476 - Intake Air Switch-Over Valve (N335)
            P1068 - 35-00 - Open Circuit
17933 - Camshaft Timing Adj. Bank1 (N205)
            P1525 - 35-00 - Electrical Malfunction
17841 - Secondary Air Pump Relay (J299)
            P1433 - 35-00 - Open Circuit
17840 - Secondary Air Injection Solenoid Valve (N112)
            P1432 - 35-00 - Open Circuit
17978 - Engine Start Blocked by Immobilizer
            P1570 - 35-00 -

Readiness: 0110 0000

The reason your car dies is immo!

De-immo your EEPROM.

N205 worries me. Everything else can be safely coded out.

How could my car be IMMO-ed?  I have a running M-box on my A-box car - which had no IMMO to begin with...

Because neither the M or A box has an immo routine in flash, but RS4 files do.

The eeprom of both boxes will be set the same as far as immo control is concerned.

Ok.   That's the end of the line for me then, i don't have a bench flasher yet...

What tool are you using to flash the ECU?

Nefmoto standard with China blue.  Also have a vagcom and also galletto, but haven't yet built a bench harness.  That's fine, I'll work on building one so i can 95040 me7 write, this is what i need to bypass immo right?

Does this mean my car has a cam sensor problem?   Def dont get this on my regular setup ever....

I don't know. Log it and find out.
Seems more like it can't access the bank 1 tensioner for some reason.

Maybe a good idea to study the pinouts and part numbers very carefully.

You had a good comment about this on your "beginner's guide" response.  I found it humorous. OK, point taken.  There's noticeably more work to be done here.
[tags @JibberJive]

However, where can i find the actual RS4 DTC codes?  I don't think I have any cam-tensioner issues at all.  The car runs fine, and from what I've read, all VAG V6 30V motors use the exact same 4-tensioners for each motor, albeit a passat upto a rs4.

Also, P1650, this is a code I also get on my A-box car, pre-01.5+ S4 with aluminum uprights.  It has something to do with my instrument cluster not recognizing something.  This was the first thing I learned how to do when I was new on this forum:  code out this particular DTC.  No suprise to me that it's back, but it's effectively harmless.  It doesn't even show a CEL, just sits in the DTC-memory-banks.

I was gonna say, the euro/asia ECU's have IMMO programming.

That said,  are you guys sure the ECU's are 100% the same?  I have a hard time believing that for some reason.  Has anyone checked the pinouts between a euro RS4 and a US S4 on schematics? Or are you all just going on assumptions they're identical?

I haven't, but it'd be great to have assistance here.  Jibberjive is going to resume trying this sometime this week I believe, and Snowtrooper too.

Won't a simple cluster recode take care of that P1650?

You'd think so, but it didn't work for my car.  I had to code it out, something about the code "CLACINS".
You are referring to this, right? http://s4wiki.com/wiki/Cluster_recode

Trying to help you find the coding tables. Do some 16bit decimal searching for the codings you know do work on a given ECU, you'll very quickly find both the tables I mentioned. The table with the associated flags will show you all possible coding combos for the flags. Whether or not they all work as expected depends on the ECU, once you unlock the coding access table you're free to recode this thing all over the place and try 'em out.

Hint 2: I think you have half the access table defined inside of a map in the ols from here. At least one of the random self defined ols' I found did.

There is a hint in the Audi/VW dokument, repair and info system called ELSA, that the 551Kand Q are fix coded for the Euro or the japanese market,
and that they can`t be recoded by a workshop diagnostic tool.
So I can`t search for an other possible ECU coding in both files, because I don`t know an other coding variant.

Can you show us an example in an other file?
Then we can try to compare the dump and search for the coding table part in the 551K.

Sorry, but I don`t understand your "hint 2".

Wiring is different between RS4 and other ECUs.

Rick

I kinda recall that now, that's why we have to swap entire harness to federalize an RS4 and run s4 everything.

Can we pin swap?

thats correct, but no need to swap harness. its only one important wire (cam switching valve bank two, driver side).
so there are 3 options.

-change complete rs4 engine harness
-just fit this one missing wire (pin 115 from ECU)
-code out the fault (CLAENWSE2) in this case bank two is controlled together with bank1, like it is at all S4's

BR

So this doesn't even matter?   What does independent cam switching offer us?
If a wire could be plumbed without a harness mod, that' be ideal.
Honestly minus this error and the previous error, i think it seems easiest to code these two things out.
Is there any serious risk at all to coding it out?

Thoughts on the secondary air pump and relay?  Does the RS not have these and hence the code?
If so what will occur to the S equipment without the engine ecu communicating to it?
I was under the impression that only TIP models had SAI stuff.   Help me understand please.

hi,
there is no risk. you can code out all errors, it will work 100%. (except CLAENWSE2 you have to zero out all errors anyway, all the errors of sensors/valves you got are not needet!!)
if you want to have independent cam switching, just put in the missing wire to bank 2. zero it out too if you dont care about it. it will work just fine anyway.
BR

Let me find the quick link here for these DTC's.  I'm going to add this to my XDF.  I believe that I'm a boot-flash away from being good to go.

awesome. I look forward to a digestible summary :)

Does anyone know the updated formula given the RS4's relative DTC table to find the addresses we need to code out?

It sounds like we are supposed to just 0 out the airflap matrix.  I plan on using this as a meth controller after that's all understood...

Here are the codes I've seen and all of the Hexadecimal locations in the Q-box.  This list shows the memory address converted to decimal:

P1068    P1432   P1433    P1525    P1650    
78592   65920   78784   12576   77888   
79936   68000   89536   13696   80000   
117696   74176   89568   13984   264256   
290208   78816   106752   78080   371040   
291008   90208              78304   566208   
319712   92672              207776      
320256   106688              602816      
416512   106880              603008      
416544   123360              605120      
416576                         605728      
428448                         608032      
428608               
429376               
508480               
508608               


Using simple logical intersection, I'm convinced that the DTC table is in the range of  000-13xxx.
There's another ongoing post on this matter which hasn't be solved:
http://nefariousmotorsports.com/forum/index.php?topic=2365.0

Hi guys,

here are the original Audi engine circuit diagrams of the german S4 (265HP/195KW, AGB & AZB, 8D0907551D), and the german RS4 (380HP/280KW, ASJ 8D0907551F & AZR 8D0907551K/Q).
Sorry, but it`s only in german.

Sorry again, but there is no circuit diagram for the US S4 with APB engine found in the german ELSA docu.

Here is the US s4 if it helps.

I can confirm this. I put RS4 ECU in my S4, and had the cam sensor fault code. I used S4 wiring harness, and it uses one same wire to the both sensors. RS4 uses one to each sensors. The car worked ok even with the fault code, it was just that the ECU thinks that the other sensor can't get power because the wire is not connected. But actually it is, and you can see the camshaft timing values from block 091-093.

We have the DTC fault-code base address.  What we don't have is the DTC fault-code table location.  Until these values are resolved, these problems won't be coded out and the community can't move forward.

Just wondering, is it important that the second wire is added into the harness?  Are there any details or a write-up on adding this wire?

No it is not important. Both sensors will get the power anyway. But as someone said earlier, it is easy to connect wore to the pin 115 of you want.

This is funny, I used RS4 ECU, and my tuner put M-box file in to it :) Just the opposite as wanted here :)

To delete the N335 code set 0x106F4 to 0.

DTC table address is 0x12FA4
DTC error class table address is 0x10689

thank you so much!  this thread literally stalled out prior to your post.

For others that don't know, the N335 post above by prj is referring to the airbox-flap actuator which no S4 owners will have.

My car has been down for a while, fixing my AC compressor, but I plan on getting it together in the next week-ish, so I'll be able to do some experimenting.


Here's an excerpt from an email I just sent Julex, explaining my thoughts and intended plan of action on this:

"  So, it seems like things for us have to go like this.
1. Get a developed .xdf.  RS4boost posted a huge .kp, and if I'm not mistaken nyet has a program that turns the .kp into a nice .xdf.  Maybe that's something one of us should talk to him about.  Otherwise, prj has posted on the forums an .xdf that he autogenerated.  Also, nehalem sounds like he has since done more work on it, so maybe we should get his updated version before trying to do any more work on that.  We need to get unified on the .xdf development front so that effort isn't wasted.  But as-is, we have enough info to get an M-box car tuned with a K-box.

2. So, prepared with a K-box tune with the basic maps switched in the K-box with my values for fueling, maf etc, I think I'm going to flash an immo defeated K-box 95040 EEPROM with ArgDub's flasher, flash the K-box .bin (via Galletto boot mode if NefMoto won't), and I think everything should work.  Nehalem's issues with the VARDEF and all that I think will be cured by flashing the K-box 95040, which he hasn't done yet.  Eventually I actually plan on getting an immo ignition coil, RFID chip for my key, and adapting my cluster to the K-box, having immo active.

3.  With a running box, I think the pressing (not sure how pressing) issues are coding out the RS4 intake air flap (prj posted solution here http://nefariousmotorsports.com/forum/index.php?topic=2342.msg24639#msg24639 ), coding out the 2nd bank camshaft adjuster DTC (which seems like we should be close to finding, with the DTC table prj listed in the post below the last one linked), and tuning out the oil temp based rev limiter (referenced here http://nefariousmotorsports.com/forum/index.php?topic=2169.msg20878#msg20878 ).

4.  Once those are good, I think then we worry about tweaking the throttle body/plate and knock recognition stuff from the APB.

5.  After that, I'd love to figure out the map locations for all of the pressure conversion maps, and throw a 4-bar MAP sensor in there.  And also LC, NLS, and map switching.


Those are my thoughts on the whole thing at least.  I think someone could have an M-box car running on a K-box in one night if they just flash the immo defeated EEPROM and K-box bin modified for their hardware."

Unfortunately, i can't parse the .kp, but prj seems like he can...

Did anyone ever get a 551K bin working on a 551M ecu? it seems like all the information is here...

Nothing of importance is even touched upon here. All this thread has is error class deletes and bad assumptions. I can tell you the ESKONF changes to properly code out the cam sensor diagnosis, as well as the N335, and flashing an immo-off file to the eeprom along with softcoding the ecu to 10711 for ESP, or 10611 for non-esp will get you up and running (do not change the coding stored in vardef, or you'll get a DTC for improper coding), but that's the very least of the issues running the rs4 file on the s4 hardware.

In short, noone worth listening to is going to walk you through it. There will be a LOT of trial and error.

I don't know, I haven't had too many issues - it all has been manageable so far :)

What's the ESKONF required?  What are the more meaningful issues of which we haven't spoke?
I have a bench flasher finally, I just also have a fantastic running stage 3 with no issues, thus no motivation.

Oh, I know you've done it. I've even glanced at a few logs of your work and it's obvious you have it well sorted.

That being said, you aren't going to hold anyones hand are you? If you are, I'll retract my statement of "Noone worth listening to is going to walk you through it". If not, My point stands.


1. Copy ESKONF from the M-box to remove the Cam adjuster and N335 diagnosis. After that, you can do the standard deletes as outlined in the wiki (although the addresses will be shifted.
2. Meaningful issues? Knock calibration, hysterisis, throttle maps, fueling, the base boost maps that prevent tuning DIMX, and the pressure model.