I have a log from a scan sesssion with TunerproRT with my 8D.
I reported problems with error counts (there is another post for this).
The exported .csv file has 4 spark related colums:
Spark Adv Rel. to TDC (Degrees)
Spark Adv Rel. to Ref Pulse (Degrees)
Knock Count
Knock Retard (Degrees)
When there is NOT Knock counts there is NOT Knock Retard (of course)
In this case (no Knock Retard ) if I do this calculation:
(Spark Adv Rel. to TDC) - ( Spark Adv Rel. to Ref Pulse ) = 5.98 deg
5.98 deg is the constant "Spark advance initial (Distributor Setting)"
Example:
Spark Adv Rel. to TDC = 23.91
Spark Adv Rel. to Ref Pulse = 17.93
23.91 - 17.93 = 5.98

When I have KNock retard I see that the Spark Adv Rel. to Ref Pulse is much more LOWER than expected.
From the eqaution above:
(Spark Adv Rel. to TDC) - ( Spark Adv Rel. to Ref Pulse ) = 5.98 deg
Spark Adv Rel. to Ref Pulse = Spark Adv Rel. to TDC - 5.98.

Example where I have knock retard.
Spark Adv Rel. to TDC = 31.99
Spark Adv Rel. to Ref Pulse = 22.15
Knock retard = 2.3 deg
31.99 - 22.15 = 9.84
Even if I subtract the 2.3 deg of the knock retard, I have
9.84 - 2.30 = 7.54 > 5.98
Could someone explaine me this?
(I like to know what I'm dealing with...)
Thanks
-Beppe-

 

Only thing I can think of is that the ALDL data is captured one byte at a time. This causes the data to not be real time with one another. Between the time the SA is output and the knock retard is output the ECM has changed the knock retard (decay or added more).

RBob.