thanks! This helps a lot. I’m in Tampa and use 94 Octane. So this gives me a good starting point to develop a map for when I go to Sturgis where it will be mostly 91.

So one thing about your signals descriptions. I am wondering if the Spark Advanced field in the log is the long term retard. The reason I say this is, I wrote a program to go through the logs and normalize the rpm, map, and tp fields to correspond to the tables in the PV. Then I was able to compare Spark advance in the log with the spark advance in the spark tables. They are off. It’s definitely not copying from the tables. 🤔🤔. I know don’t really need to get this far into the weeds, but I’m working on developing a log analyzer for fun and to learn along the way.

anyway, thanks!

steve

 

How much were they off? You have to remember that the ECM does some rounding and interpolation because the cells in our spark tables are only in 10kpa intervals, and it's doing this on two axes that are changing very quickly as we ride. Also don't forget that there are some other settings that we might not even be able to see that can bias the timing advance, such as engine cooling strategies. I have no idea what might be going on behind the scenes in the M8 ECU. When I read my logs and compare them to the tables, I find that generally they are pretty close and the slight amount they seem off I chalk up to interpolation.

 

Its interesting. I need to look at this a bit more. Pure average in a 20 min log it’s off less than 2 degrees. But a lot of skew. You may be right. I’ll look through this more and let you know.

steve

 

When you're looking in your log, try to find a spot where it's not in between cells. For example, 60kpa @ 3500rpm. You can directly compare to that intersection of the spark table and I'll bet you it's right on. Unless of course you're showing spark knock retard in that spot, in which case it should be whatever is at that spot in your spark table minus the degrees of retard.

 

Well it is true. On your bike you have one table you can zero. On the M8's there's 2. They are available in the PV.

 

Well it is true. On your bike you have one table you can zero. On the M8's there's 2. They are available in the PV.

 

You should read what it says in your own screenshot: "Learned values will be reduced towards zero after each key-on cycle."

I can assure you after many many logging runs, adaptive retard does not persist from previous runs on my bike.

 

Yes but it may take 50 key on cycles to reach zero. Regardless of tuning device that's how the ecm works.

 

Well, I don't know what to tell you...none of my logs have ever showed spark knock retard from previous runs.

Also, that table you posted just shows the maximum amount of retard it will do, which you have set to 4. Mine is 6.

You should also look at the spark retard removal rate and percentage. On mine for example, the removal rate is 0.5 seconds and the percentage is 14.8 percent. Which means that every half a second the spark retard is reduced by 14.8% until it reaches zero. That's pretty quick.

 

I think you might be looking at the knock retard, not the adaptive knock retard. That table just sets the limit that the adaptive can learn.
I'm tuning today but will come back and post some data to show what I'm trying to explain.