Is it possible that the pcm isn't actually sensing knock but applying a knock value based upon engine parameters?
I know you have an 0411 so you aren't using EEHack, but I noticed--and confirmed from reports from at least as far back as 2017--that when EEHack has a communication error, it tends to show knock counts where there may be none. Just throwing it out there that the knock counts may actually be intermittent communications errors with whatever scanner you're using? Unless the scanner is actually pulling knock counts from a specific location in memory where the PCM itself is saving them, it's possible that these are false counts.
It would not, unless for some ungodly reason the scanner is altering timing without you doing anything, which it should never ever do. The low timing is being done by the PCM itself. Is "Knock Retard" an available parameter for you to inspect on the 0411 PCM? On the LT1 PCM it's a separate value from Knock Count. That could potentially explain the sudden drops in timing, though at that point it would also indicate that the knock is a real thing (or rather, that the PCM thinks it is).
Hi
What does EEHack show when it records a bad frame apart from knock counts. How can you tell from the log?
Mitch
'95 Z28 M6 -Just the odd mod.
'80 350 A3 C3 Corvette - recent addition.
Hi Mitch, I would prefer if we didn't clutter this person's thread with unrelated topics, but to quickly answer your question, the log will give you data that makes zero sense in context. For example it may say that for a single frame of data you had a Check Engine Light for a random code, but then the frames before and after do not, which is impossible. It may say that you had the exact same data for several minutes without any parameter changing, which is impossible. It may say values like speed changed by massive swings within two frames, which is impossible. Now, the program itself while logging also keeps track of error counts, but these counts don't actually get logged with timestamps and the program doesn't discard data from bad frames, so while it should be possible to know with absolute certainty which frames are bad, instead you'll have to use your noggin' to sanity check what's there.
Any thoughts as to calculated knock as opposed to actual knock? Perhaps something that is based upon a correlation between map, maf and TPS to calculate engine load?
Bookmarks