Announcement

Collapse
No announcement yet.

Debugging the CSL software on non-CSL DME

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    Debugging the CSL software on non-CSL DME

    For those that have attempted this... Can I pick your brain?

    I have a 2005 M3 with the MSS54HP. I did a conversion to a CSL Cone using a CSL style airbox, 288/280 cams, stepped headers, open exhaust, GM Map sensor, 93 octane, etc.

    The main differences are: I am using the MSS54HP with no H-Bridge. Also, the engine was built with slightly higher 12.0:1 forged pistons. Other internals are stock.

    I had been running a custom Alpha-N tune. But I have time on my hands and want to try for something better.

    I am now running the modified 0401 CSL software for the non-CSL bootloader. I got it working with a combination of the BimmerGeeks Tools, MSSflasher, Binary Modification Tool, TunerPro, etc. Amazing tools.

    Today I went for the first drive. It feels really good. But it has a pretty violent misfire under high load around 3000 and below. At 4000 up it pulls hard and feels smooth. On regular streets under partial load, it's fine. But on the highway attempting a pull in a higher gear, it misfires.

    I interpret this as the DME cutting the ignition because it senses something wrong. If it were knock, I would expect it would just retard the timing. It seems to do that well elsewhere. If it were fuel, I would not expect it to feel like that. I know there are about 50 compensation maps in there. But I need an idea where to investigate.

    I have an Innovate LM2 logger. But I need to replace the O2 sensor before I can do some logging.

    Given I am not the first person to try this, can anyone who has tried give me hint were to look? Has anyone encountered this?

    Thanks

    #2
    It sounds like the famous hesitate/bucking that occurs when you run the CSL software out of the box with non CSL cams.

    you would see this on the Dyno more often kn the street as Dyno operators out the car in 4th gear and then start at 2-3k Rpms and give full throttle and see the hesitation/bucking occur.

    Some people have been able to modify the vanos offsets in the CSL software with non CSL cam values I believe.

    Comment


      #3
      I'd use testo logger to check what's going on, log rpm/load and lambda integrators (assuming you still have front o2 sensors). I usually use MegalogviewerHD to view logs, it's well worth the money IMO. After you have a log, just open it in MLVHD and set the histogram table up to match the alpha-N table. The logged parameters should be for RPM, Relative Opening, Lambda integrator. This will give you a table view of the fuel trims in each alpha-n cell, it's probably a good place to start looking (obviously doesn't work for WOT). IT's probably on the rich side (you'll see lambda under 1.0).

      You can use the log method to get everything pretty close if you want to, there's a bunch of other things I'd change to give you more accurate data during tuning, this will give you a quick overview though.

      You can do the same thing for the ignition tables too, the logging rate on testo is pretty awful so I'd recommend doing them separately. There's also a 'standalone' version of testo if you don't have INPA installed.

      Comment

      Working...
      X