Spark Timing Adjustments Based on Data Logs
 

I'm new to tuning. I have a PV2 from @fuelmoto with their canned tune. I have run multiple Auto Tune sessions to improve the VE tables. Now I'm looking at timing. I know I can reach out to FM to get the timing dialed based on my logs. However, I like to learn and have an understanding of what is going on. I have the time to look at it myself and want to get some hands-on experience tweaking a tune.

I plotted my spark knock retard events from my data logs. Below is RPM vs MAP (kPa) x Spark Knock Retard (degrees). The plot contains over 15,000 data points from my log of over 30,000 data points (i.e., over half of my data points contain knock retardation).

Based on this info, from 50-85 kPa and 2000-3500 RPM, I am getting between 2-3 degrees of knock retardation. Therefore, should I retard my timing in that region by the corresponding degrees of knock retard (i.e., the timing advance number would be smaller)? Or perhaps, retard the timing by half of what is indicated in those regions?

Am I thinking about this correctly? Or should I just leave it alone and let the ECM sort it out?

This is with E10 91 octane. This graph gets off the charts (each retard event is about 2 degrees higher, max of around 6 degrees) on non-ethanol 87 (the only non-ethanol in my town).

https://cimg6.ibsrv.net/gimg/www.hdf...51be8a9fe9.png

FWIW, the rear cylinder has about half the degrees of retard as the front cylinder.

Pull out timing, 1/4 of a degree for each degree of retard, @ the RPM just before the knock retard occurs, and the RPM it occurs. Save, flash, and retest. This should be done with regular 91-93 octane, after you're sure your VE tables are calibrated. Inaccurate VE can induce knock activity.

Thank you very much!

As far as ensuring VE tables are calibrated, all I have done is "auto tuned" 4 sessions (~30 minutes each) in a row one afternoon. I did as many sessions as were necessary to achieve <5% max change. Out here in BFE, TX, in one session I'm able to get 99 hits on every conceivable cell I need to out on public roads and never see another vehicle. So, I got 99 hits on 99% of cells in a session, load that auto tuned tune, and repeat for a total of 4 times to achieve <5% change.

How do you know if pulled out too much timing? What are the consequences/effects off pulling out too much timing?

You will feel the bike become sluggish, and get very hot very fast. Hard to do when you're only pulling 1/4 of a degree at a time. :)

Perfect. Thanks again!

I made these changes to the front and will be doing a log tomorrow.
https://cimg7.ibsrv.net/gimg/www.hdf...a0e3967ba8.png

For the rear cylinder:

Spark Retard Events
https://cimg9.ibsrv.net/gimg/www.hdf...2d80763841.png

Adjustments
https://cimg0.ibsrv.net/gimg/www.hdf...26bd4d2c06.png

Have you been doing this with the adaptive knock retard disabled?

No. I suppose I’m missing something?

I haven’t loaded the tune. Just played around in Powercore.

I always check my fuel in the few frames before the knock event before pulling timing, making sure it's not a lean condition causing the knock