"Knock diff" in megatune logs

[dnb]

What is this actually measuring?  The difference between the readings of the 2 windows (with some form of scaling) or something more interesting?  How do I relate it to actual magnitude of signal from the sensor?

Also, the spark log seems to always be an integer despite VEMS having 1/4 degree resolution.  Is this a known problem I've never noticed in the logging?

Thanks.

[rob@vems.co.uk]

Its the knock value / noise value.

Looking at the datalogging part of the ini file that configures megatune it says that spark is a float with one .1 degree precision.  I've not noticed the values in the logging TBH.

[dnb]

OK.  Thanks.  I think I understand the MT ini file now...
The sparks thing is a bit odd.  I'll have another look at it as I'd like to get to the bottom of it.

I've modified my ini file to log the raw knock noise and knock value signals.  This should mean I can get a better handle on the integrator. 

How is the thresholding done in the ECU?  "knock_diff" exceeding the threshold set in the knock activity settings?

I've also adjusted the injector size in the ini file so that I get an approximation for fuel use - should be a good indicator for testing lean burn strategies etc...

I wonder is it possible to get the MT logs to contain the RPM and MAP bins?  I know it's easy enough to do this in excel afterwards, but it would save a few minutes per analysis session if they were done already.

[rob@vems.co.uk]

The ini takes a bit of chewing on by makes sense eventually.

Threshold is a result of knock diff IIRC.

The RPM and MAP bins seem to be calculated in MegaTune.

[dnb]

Re threshold - mine's very much too high then... 

I wish Excel had a better 3D graph plotter.  It would save a lot of this messing about.
There's always SciLab (a free Matlab clone with added user hostility ) but it's a pain to get the data imported.