Jerryg

I have not been able to get a knock reading I believe since I installed the MS3 Pro PNP. (stock ECU before that). I've installed a new knock sensor, I've check continuity of the wiring from the sensor to the MS3 and everything looks good, So either I have the settings wrong or the MS3 is bad. Before I pull the MS3 out and send it back to DIYautotune I would like to ask for someone to check my settings for me.

Specs
2002 NB
fully built engine
84mm pistons
8.8 CR
stock knock sensor

I've attached my tune and a log from a 3rd gear pull ----- Thanks for the help

O R X O N

Hi. Stock 2005 LS with MS3x and OE knock sensor as well. Just installed two weeks ago, in love with it so far!
Checked your log; there is reasonable fluctuations hinting that it's installed properly and functional.
Checking your tune I noticed your bandpass frequency is 13kHz. This is almost higher than human hearing range.

For best results, try setting it to 6.49kHz. This seems to be agreed upon by most for the "resonance" of our blocks. YMMV because you have built internals but still.

I then tweaked gains until my overrun noise on each cylinder looked pretty close. Then, using MegaLogViewer (HD version) scatter plot to get a feel for what your threshold should be. And note any knock you may find.
Also here's a peak at my windowing settings. It's pretty huge but I prefer it, but I'm N/A. The Reverant way is "take your spark advance at 70ish KPA, and use that as the window start. Then put that into the window duration but add 20 to it." What this does is starts listening a few degrees before spark (since 70ish KPA is more advance than 100+), and stops listening 20 degrees after TDC; since the "toque sweetspot" is 17* ATDC.


For my specific application; this results in the following in my logs. As soon as threshold hits, it solves the problem immediately with no fuss. Some folks may use different "count" values. I'd like to use "1" but I have some tip-in/transient knock because my AE needs work.


Good luck!

Jerryg

My bad, the log was ran at 6.94 kHz. I read an article that stated that 2X resonance frequency would generate a cleaner signal so I was just giving it a try and didn't set it back. Thanks for the info, my main concern is that I do not see any change in the knock signal as RPM's increase, I would think just the background noise would cause the Knock In signal to go up.

O R X O N

It would. You can see from my scatter plot, that my WOT and full decel values work out to be roughly the same.

Since you've replaced the sensor + ensured continuity, I'd send it back to them. This isn't the first time.