Hacking the Motronic ECU (Audi RS2)

Collapse
X
 
  • Time
  • Show
Clear All
new posts

  • Grunf
    replied
    Hi,

    this is my first post. I have UrS4 with AAN engine equipped with SMS-Revo 303hp chip which I converted to E85 by using bigger injectors, 3-bar FPR and larger fuel pump. I have access to WBO2 and lambda values are perfect at WOT (with stock fuel pump, engine went lean above 5000 RPM).

    Right now, my issue are cold starts. E85 requires upwards triple amount of cold-start enrichment compared to gasoline. I have "hacked" it externally by connecting an additional NTC temp sensor in series with stock one, together with switch that shortens this additional sensor above 85 degrees.

    It works but it's a hack. I wonder if anyone knows where cold-start enrichment values are stored on AAN fuel chip and if it's possible to change them, either by burning new EPROM or (maybe) through adaptation and VAG-COM?

    Thanx!

    Leave a comment:


  • rs2spinelli
    replied
    Sup

    My friend how u put more boost on your s2 and how the s2 not cut off?

    Leave a comment:


  • prj
    replied
    Originally posted by Sir Buff
    How did you learn to set the rev limiter from (for example) 7200 to 7400 rpm?
    Is this only possible to find out with a eprom emulator and real time tracking?
    Personally I learned by comparing different binaries and doing a full disassemble in IDA Pro, because I do not have friends at Bosch.

    Leave a comment:


  • Sir Buff
    replied
    How did you learn to set the rev limiter from (for example) 7200 to 7400 rpm?
    Is this only possible to find out with a eprom emulator and real time tracking?

    Leave a comment:


  • prj
    replied
    Yeah... or a friend at Bosch.
    Like a certain tuner.

    Leave a comment:


  • Dan
    replied
    Dear Bosch,

    I can has original sauce plz?

    Kthxbai,
    S2 Forum

    Leave a comment:


  • prj
    replied
    Originally posted by Sir Buff
    good to know, thank you dmitri.
    so, is the rev limiter only on one adress (1x1)?
    No...
    If you want to find it (or "them") you need to either disassemble the binary in IDA or be good at number crunching, data structures and hex binary comparison.
    The entire constants area in the ECU is a big mystery until you disassemble the code and see where things are accessed and how they are used.

    Maps you can tell using trace and shape, but the constants are a different thing...

    Leave a comment:


  • Sir Buff
    replied
    good to know, thank you dmitri.
    so, is the rev limiter only on one adress (1x1)?

    Leave a comment:


  • prj
    replied
    The info on that site is incorrect.
    Not a single RS2 has a such limiter.

    Leave a comment:


  • Sir Buff
    replied
    Hi

    Is the calculation of the rev limit correct for our cars as shown on this website?:



    Rev Limit = ScaleFactor / RL1
    Rev Limit = 983600 / RL1

    Leave a comment:


  • prj
    replied
    I run 1.6 bar boost with standard injectors.
    Injector duty cycle depends on RPM. The lower the RPM, the more time the injectors have to inject fuel...
    You can log injector opening time in milliseconds from vag-com, and then use that information to calculate the duty percentage.

    Leave a comment:


  • Dybro
    replied
    Originally posted by prj
    In your case your injectors will not inject more fuel. Because your car is not mapped to do so.
    This means, you won't max them out, but you will run leaner all the time. And yes, most noticeable will be when accelerating from light-load cruise, because motronic is mapped to have over 1.02 lambda during these conditions in stock form. With ethanol you will be getting even leaner. Torque starts to rapidly drop after approximately 1.09 lambda on these engines.
    With the current chip it can deliver 1.3 bar boost. But I think it's way too risky with std. injectors

    Leave a comment:


  • prj
    replied
    Originally posted by Dybro

    It's much the same I have in mind... I don't dare to rev it high with 0.8+bar boost while having too much ethanol in it.
    In your case your injectors will not inject more fuel. Because your car is not mapped to do so.
    This means, you won't max them out, but you will run leaner all the time. And yes, most noticeable will be when accelerating from light-load cruise, because motronic is mapped to have over 1.02 lambda during these conditions in stock form. With ethanol you will be getting even leaner. Torque starts to rapidly drop after approximately 1.09 lambda on these engines.

    Leave a comment:


  • Dybro
    replied
    Originally posted by prj
    In a chipped car with stock head on petrol, at 6000 rpm and 1.1 bar boost injector duty is close to 90%.
    Adding 20% ethanol to that will max injectors. 30% and you will get a lean spot, which is not a very good idea.

    It's much the same I have in mind... I don't dare to rev it high with 0.8+bar boost while having too much ethanol in it.

    No, it doesn't. average/high load conditions are not affected at all by the presence of the lambda sensor.
    On my car I have mapped it to only use lambda at idle/revving without load, so I can run 15.6-16 during cruise and still pass emissions.
    OK Thanks

    Leave a comment:


  • prj
    replied
    Originally posted by Dybro
    For good sake. Understand what I'm writing. I DON'T run it on E85 because it requires much more fuel! That's why I maximum use 20% ethanol.
    In a chipped car with stock head on petrol, at 6000 rpm and 1.1 bar boost injector duty is close to 90%.
    Adding 20% ethanol to that (and remapping properly) will max injectors. 30% and you will get a lean spot, which is not a very good idea.
    Do you mean that that it doesn't have an overall lambda-fuel-correction?
    No, it doesn't. average/high load conditions are not affected at all by the presence of the lambda sensor.
    On my car I have mapped it to only use lambda at idle/revving without load, so I can run 15.6-16 during cruise and still pass emissions.
    Last edited by prj; 2 December 2010, 11:12.

    Leave a comment:

Working...