MS-II pulse width spikes

[renns]

MS-II running Rev 2.30 code.
6-cylinder BMW with EDIS ignition.

The car runs nicely, but has a random misfire/stumble that occurs throughout the operating range. Initially the stumble was thought to be ignition related, but plugs, wires, EDIS module, and coil pack have all been swapped with no change.

Datalogs with markers after each stumble haven't yielded much info except for an occasional spike in reported pulse width. These spikes always jump to 30.000 ms, and last for one log record only, and do not appear prior to each stumble marker in the log. I suspect this may be the cause of the misfire, and wonder if anyone has any thoughts. Unfortunately the LC-1 wideband is inoperative at the moment, so we have no reliable afr data.

In looking at the log, I see no reason for these spikes. Accel enrichment values are set quite low, and there is no accel condition logged at the point of the spikes. The car runs very cool at present, with temperatures cycling between 158-162, causing warmup enrichment to repeated trigger, but the warmup % is quite low at those temps.

Is there a revision log for MSII code? In looking in the MSII forum, I see nothing, although beta appears up to 2.6xx

Thanks for the help,
Roger.

[Bernard Fife]

Roger,

Use the 2.36 code, available here: http://www.megamanual.com/ms2/code.htm

The latest beta code is good for all ‘non-missing tooth’ applications.

There is a revision log at the top of the C source file, though I am in the process of putting this on a web page (thanks to Marc!)

Lance.

[renns]

Thanks Lance, looks like that site is up again.

Any thoughts on that 30ms spike? Has that issue been seen before? I'll help my friend re-flash to the new code, and see if it still exists.

Thanks,
Roger.

[Bernard Fife]

Roger,

I *believe* that happened in some earlier versions of the code (it's been a while since I ran anything less than 2.6xx, so my memory has faded!).

So upgrade, that may fix the 30ms issue.

Lance.

[Bernard Fife]

Roger,

I have put the code revision details here:

http://www.megamanual.com/ms2/revisions.htm

I'll be adding links to relevant sections of the manual, etc. as time permits.

Lance.

[renns]

Awesome, thanks for the info. We're planning to look at the car again early this week, so I'll do the re-flash process first thing.

Cheers,
Roger.

[renns]

Re-flashing with 2.36 eliminated the random miss problem. The owner is thrilled! Must have been a glitch in that code somewhere. Now to chase down LC-1 problems. Unfortunately the old Win95 laptop will not run the LC-1 setup software, and the new laptop has no serial port.