Those are all the same spot MAP-wise but not clustered
at all, by RPM. Tight little RPM bands are more the false-
KE signature. I think maybe this wants the PE enable MAP
threshold moved down, or the spark pulled out just in the
CylAir region that corresponds to this MAP level.

 

how do i 'convert' the CylAir to corresponding MAP values?

 

I would try and do a scatter-plot of your MAP and CylAir
(just within the RPM range of interest) and it should give
you a pretty good straight-line fit. CylAir falls away from
MAP as the VE deviates from 1.00, at the extremes.
Offhand I'd say that the 0.28-0.32 MAP corresponds to
about 0.24 g/cyl. Or, maybe the histogram is already on
the CylAir basis? I forget. That's where you have high spark
(and, for that matter, that looks like it might be in the area
where EGR is further pushing, or was, on the spark advance
table value; the EGR adder might want squashing especially
if EGR plumbing, er, fell off or something. Disable/Enable is
under Engine and the table is under Spark Advance, EGR
adder table. Are you seeing more advance, (logged+KR), in
those KR-prone cells than the main spark table appears to
command? If so suspect that the EGR is meddling. Use the
histogram in max mode and cross-reference the main table,
are they the same?

 

Your retard MAP looks to me like its getting ping at part throttle and light/medium loads. Its not WOT related. Furthermore, you may be seeing tip-in knock which is not actually ping but preventative retard to protect from lean transients.

How much timing do you have in those cells where knock occurs? I bet you have around 30*. Pull it back by 2* where you log the initial knock event and smooth it off for the surrounding cells. It wont affect WOT nor mid-range power as you are retarding there anyway.

Also double your Dynamic Cylinder Air Burst Knock limiters.

 

I have a feeling this is going to be a great thread. By the way thanks JimmyBlue for the screen shots of the settings I couldn't quiet grasp what spots to make what changes just going off of your earlier statements. I'm going to apply the settings to my car now and I'll be back on shortly to update on the results. Thanks again!

 

well, the car's in a shop now, so i'm not gonna be able to dick with it for few days, hopefully it will be fixed by retorquing all the nuts and bolts proper

JimmyBlue, as always, you seem to know yer ****. could you please enlight us peons how do the parameters affect the KS/timing in a more wholistic manner? It's hard to figure it out when fields in HPT are mislabeled and the help file is a bit on the useless side ('knock sensor senses knock' sort of explanations are a waste of time and energy)

one weird part i found already is that i downloaded a bunch of 'solid' bin's from Horist's site and the values for MAX KR table are all over the place, negative, positive, combinations, small values, large values...garden varity, doesn't even depend on type or year of the car. so i'm really not sure what's going on here, are there that many different types of knock sensors out there, or are people screwing with it a lot?

 

I would like to understand the why, when, and how of this as well. Please impart some of you words of wisdom upon us!

I changed it to exactly like your screen shots. Flashed it, warmed it up at Idle, drove a couple of miles at 35mph, and then took the opportunity to put some HP to some pavement.

No KR the first 2 times I ripped threw 1st going very sideways on into second, but on the third time I got a 1* KR blip.

I was actually happy to see the 1* blip, so I know I didn't filter out all the Knocks. I'll be able to tell better tomorrow when I can check it out with out calling all the police around.

Disclaimer: This story was fictional and was in no way intended to represent any person or group.

 

I've gone my own way with the trans TM and the
Max KR vs RPM (so-called) is my way of limiting the
spark pull, cleanly. I don't like the whole scheme of
calculating a torque, a precent reduction, a spark
retard and, oops, just slam it to -10 degrees anyway.
I had bad luck trying to educate it, so it's on a leash.
Better than shooting it.

That field only limits the lowest advance, that -any-
retard function (or combo) can pull back to. So there's
a compromise between protection and performance as
usual. But I see no need for more than 8 degrees of
spark retard (my 28 degrees or so of WOT timing vs
a clip limit of 20 for a floor).

I don't know anything much about the sensitivity params,
which are altogether different. The preceding stuff was
all about, how are you going to react to what you find.

At this point it doesn't look like false knock to me, it looks
like excess advance for the mixture at a certain range of
operation. The questions are whether it's spark that's
heavy or fuel that's short, under what conditions and
why.

I tend to ignore the histograms more or less, myself.
And statistics in general. Lies, damn lies and all that.
The pretty pictures, if they don't outright obscure the
detailed truth, at least distract you from pursuing it.
I've found better results from going line by line in the
Excel file, finding the KR blip, seeing what was going
on in the preceding frames and trying to dope out
cause & effect. Not the "big picture" but the event
holds the key.

Rising TPS -> rising MAP -> speed density mode ->
lean hole -> ping is always a good suspect on a car
with back-end breathing mods and no rework on the
VE table, midband big end.

I'd say that most of my KR that I chased down, was
from this. Transient mixture error, making KR that lasts
way longer than the problem. I "fixed" this by over-
compensating in the VE table. That is, every time I
found transient KR I fattened up the value at the MAP*RPM
where I saw the "lean hole" develop, and smoothed it in
by hand until it "looked right". I like smooth suraces.
Some don't but in my opinion the engine is a pretty orderly
air pump and the airflow response to speed and pressure
ought to be, not linear, but well behaved - smooth.
The stock table fades early and is all lumpy. Put a cam
and headers on there and it's downright unrealistic
anymore. Flash the throttle and you flip over to the SD
tune until the PCM thinks the MAF is more reliable. But
it doesn't know the SD side is still wearing bell bottoms.