I've got an unusual thing going on. For some reason, the ECM is advancing the timing by as much as 36° (total) at WOT -- not a good thing on a 10:1 motor with pump gas! There is audible detonation as the timing surpasses about 32°.

I've gone through all of the spark/timing tables ($6E, ARAP) and there are no entries with that high of a value at the LV8 loads experienced at WOT. Thinking that PE spark advance might be the culprit, I zeroed the entire table and went for a test run, taking two passes.

The first pass went fine; aside from being slow, I did not hear any spark knock. My logs show about 20° total timing, which corresponds to the main spark advance table.

On the second run, the timing shot up again and the detonation returned. The logs show upwards of 36° total timing.

My knock sensor also seems to be malfunctioning, because it picked up no knock counts and did not retard timing when all of this was occurring.

What would cause the ECM to advance the timing like this? Is it triggering some "hidden" table that I'm not aware of? There are virtually no differences in the sensor data, between the two passes, that would justify the timing to increase the way it does.

The following is a link to a portion of my datalog. The first section is the first pass, and the second section is the second pass.

http://home.alltel.net/usaaf/etc/wot_det.xls

 

Whats in the main spark table? Is the 36 degrees the commanded advance (cant open XLS as I dont have excel on this machine)? Keep in mind that if you didnt set the base SA, this timing will be added to whatever base timing you have at the dist.

 

$6E has that stupid forced knock test. Looks like the routine works

Set EPROM location $237 to 255, that is the minimum required coolant temperature to run the test. Can also zero location $239, that is the max added SA for the test.

RBob.

 

Shoot! dimented24x7, the base advance in the chip was set correctly, but the timing just kept increasing out of nowhere. RBob, you hit the nail on the head. I read about the knock test in $8D, but I didn't make the connection that it might be in $6E, too. I also ran the AUJM bin during my troubleshooting and the timing went nuts... makes sense.

I replaced the knock sensor, too -- whaddaya know, it pulls out timing now! With the new sensor, and with the knock test enabled, I reach about 31° before retardation begins. Without the knock test, the timing corresponds exactly to the tables (~28°), and there are far less knock counts. Thanks!

 

Originally posted by RBob

B]$6E has that stupid forced knock test. [/B]

What is this, please, and what determinations can it help make?[

 

Test the ecm runs to verify that the KS works.
Basically advances timing till knock occurs.

 

Learn something everyday...
Thanks