Power Logger knock counts vs. Caspers gauge

darkred87T

Active Member
Joined
Oct 27, 2003
Hi,
I thought I was getting crazy amounts of knock because my Caspers gauge kept shooting into the red evertime I got on it. Well, I got my Power Logger working for the first time today and on two launches, it was only showing about 3 counts of knock while my Caspers knock gauge climbed into the red. That doesn't seem like enough knock to constitute sending the gauge into the red. :confused: Has anyone had this kind of experience?
 
yea i had that problem with the caspers guage everytime i would floor it the guage would go all the way to the red and the scanmaster would show 0.0 knock. no matter how long i was in it for, the guage would show knock and the SM would just stay at 0.0 it was confusing so i just took the guage out
 
Thanks, I feel like I should have something there to watch. I'm going to double check the wiring and then I guess I'll take it out and replace it.
 
There's nothing wrong with your knock gauge, it's doing exactly what it's suppose to do and that is translate what the knock sensor/ESC module is hearing to a visual display. See, the knock gauge is "seeing" what the knock sensor is hearing and the scanmaster/powerlogger is showing how the ECM is reacting to that information in the form of KR or knock retard.And don't confuse knock retard with knock counts,two different pieces of information.
Now the powerlogger has knock counts as part of it's available data but knock retard is there also and is what you should be addressing if it's high.Not the knock counts.

I swear the Casper's knock gauge is the most misunderstood tools in the turbo buick world,and consequently blamed as "faulty" all because it's not being used properly.

Used properly it can aide in finding false,transitional knock in real time.
 
There's nothing wrong with your knock gauge, it's doing exactly what it's suppose to do and that is translate what the knock sensor/ESC module is hearing to a visual display. See, the knock gauge is "seeing" what the knock sensor is hearing and the scanmaster/powerlogger is showing how the ECM is reacting to that information in the form of KR or knock retard.And don't confuse knock retard with knock counts,two different pieces of information.
Now the powerlogger has knock counts as part of it's available data but knock retard is there also and is what you should be addressing if it's high.Not the knock counts.

I swear the Casper's knock gauge is the most misunderstood tools in the turbo buick world,and consequently blamed as "faulty" all because it's not being used properly.

Used properly it can aide in finding false,transitional knock in real time.

not trying to hijack but if the knock guage is doing its job can you explain why my scanmaster wasnt showing any kr when the guage was in the red?
 
not trying to hijack but if the knock guage is doing its job can you explain why my scanmaster wasnt showing any kr when the guage was in the red?

Exactly. On a hard launch the Knock count on the Power Logger was only 3 and my gauge was in the red. There was hardly any knock retard (maybe .2 degrees for a fraction of a second.)while the knock gauge was stuck in the red.
 
not trying to hijack but if the knock guage is doing its job can you explain why my scanmaster wasnt showing any kr when the guage was in the red?


Sure...the scanmaster is reading the ECM's reaction ( or what it has done to combat knock) to the ESC/knock sensor. Which is a calculation using knock counts and this calculation takes time,albeit milliseconds of time, but it appears on the knock gauge before the ECM reacts to knock. I'm sure the Casper's also locks those lit L.E.D.s for a fraction of a second so as to allow the user to see them and decide whether it's false or real.

Here's a cut and paste from an older tech article on the turbo tweak website,it explains knock and knock count and how the ECM reacts to it. I highlighted in bold why the ECM is slower than the Casper's in reacting:

The ESC module’s job is to discriminate between normal engine noise and knock. It contains specialized filters that pick out the signals related to knock then produce a signal with a duration proportional to
the knock intensity. Each knock has to occur for a minimum of 5 msec before the ECM is notified. The ESC module has a single output to the ECM, which is normally at 8 – 10 VDC (called a “high”).
When knocking vibrations are detected, this line goes “low” (0 V) and stays low for a period of time determined by the knock intensity.
ECM Knock Processing:The ECM converts the ESC signal into a number that is determined by the length of time the ESC signal is held low (the duration of the knock). The way it works is, a counter is allowed to count up only while the ESC input is low. This counter is called the ESC PA
(pulse accumulator). The longer the knocking occurs, the higher the count gets. The higher the count, then the greater the retard to the timing. When knocking stops the counter stops, but the count
remains at the last number. This is the starting number for the next knock event. The ESC PA can count from 0 to 65,535, although it will never count that high. The amount of detonation it would take for the maximum count would cause extreme engine damage.

The rate at which knock retard is applied is called the attack rate. The rate at which this retard is removed is called the recovery rate. The attack rate is much faster then the recovery rate. The retard amount (attack rate) can possibly increment every 12.5 msec, whereas the recovery rate is calculated
only once every 200 msec. From this, you can see the amount retard can accumulate (at almost a 20:1 ratio) before any attempt at recovery.

END; clear as mud,right??



....
 
So, basically the Casper's gauge should effectively be called a "Knock Count Gauge" since it displays/registers knock counts and real timing retard (whether real or false)? If so, I'm glad I never purchased one of those because I'd be chasing ghosts all over the place. IMO, it should have been designed to only display real timing retard a la the Scanmaster.

BTW, I see knock counts on the Powerlogger register quite often and never see knock on that particular run.
 
So, basically the Casper's gauge should effectively be called a "Knock Count Gauge" since it displays/registers knock counts and real timing retard (whether real or false)? If so, I'm glad I never purchased one of those because I'd be chasing ghosts all over the place. IMO, it should have been designed to only display real timing retard a la the Scanmaster.

It really doesn't show timing retard just intensity of knock and frequency.

BTW, I see knock counts on the Powerlogger register quite often and never see knock on that particular run.

Yep, because the knock counts are "accumulated" before the ECM reacts. And I ssume you mean you never see "knock retard" on those runs.


Go to the turbo tweak website,under the downloads link there is the rest of this article. It's labeled "knock". A good read...


And I'm not saying the knock gauge is the end all,be all for knock problems. I'm saying it's a valuable tool,especially for false knock identifying..


...
 
It really doesn't show timing retard just intensity of knock and frequency.

Yeah, I guess poor choice of words on my part. What I was meaning to say was that it should have been designed to show knock only and not knock counts as well...that is if I'm understanding how the guage works. :confused: Disregard this statement if it only shows knock.

Yep, because the knock counts are "accumulated" before the ECM reacts. And I ssume you mean you never see "knock retard" on those runs.

Correct. Yes, I understand the difference in counts vs. actual timing retard and how they coorelate with each other.
 
I can't say for certain if the Casper's accumulates knock counts and displays them as a level of knock or not. But it is my understanding that it only registers knock incidents. And if I'm right the increased LED lights lighting up show how severe the knock is as it rezonates(sp). I guess we could test one using the old socket extension and hammer test to duplicate a knock sound. Say,One solid rap and see if it lights up a geen LED or so,or if it lights up a full scale of LEDs because the duration of resonating is long enough to accumulate.

But the long and the short of it is that gauge is waaay upstream,per say, of the scanmaster or powerlogger. It hears the knock and just displays it on a scale.

There was thread on here where someone had corrolated the LED scale to KR numbers,but my own testing showed that to be an average and not cold and hard numbers. Like sometimes I'll get a full scale YELLOW to RED display and show no KR on the scanmaster. But one thing for sure whenever I have the ECM pulling timing it always shows up on the knock gauge first..which is how I know to glance at the scanmaster for more info as to whether to lift or not.
 
There are conditions whereby the knock wont be shown on your scanmaster. Example at idle the knock guage may be showing lights.. but the SM wont becuase the chip ignores knock under 1k. Another example is some chips have a patch to dis-reguard knock upon certain conditions like downshifts.

The audible knock sensor, and the lights hooked up to the ESC output/ecm input will tell at real time the deal. Up to the individual to ascertain what they should pay attention to. GM disreguarded some of it already in the chip programming.
 
I've got the audible knock detector from caspers...and sometimes, but not all the the time, it will start chirping while rolling down the road in 4th gear at a constant cruise speed...say 60 MPH. I still haven't watched a scantool when it happens...I'm not sure if it wigging out or just doing what it supposed to do.
 
If the gauge measures knock counts that makes sense. The powerlogger was showing 3 counts on a hard launch and pulled about 1.2 degrees of timing. What I really don't get is why I have any knock when my 02 was between 800 and 850 on these pulls. :confused:
 
Take a look at the underside of your downpipe near the upper control arm. Bet there are dings in it. ;)
 
Hard to say as to why, but maybe rich knock,false knock. Attach one of your powerlogger runs here and we can all take a stab at it. Better yet, start a new thread with your file attached.

Like I was saying, I'm not certain the gauge accumulates knock counts or if it just displays knock as the ESC/knock sensor hears it. But I can tell you that I use it to track down false knock on the TTA. Occasionally I would get a couple GREEN LED's when I went over this one bridge in town. It would happen at light throttle cruise speed at around 45 MPH. Turns out when I rolled over th expansion joints the sub frame connector would contact the floor board where it was a real tight fit and must have resonated enough to trigger the knock sensor. I got no KR on the scanmaster. A little bar stock and a welder shored it up now.

To me, that's mostly how I use it, and in conjuction with the scanmaster. I sometimes get crazy KR on a hard launch, but my new 6.0 chip is burned with knock ignore under 45 MPH. And I have yet to get it from storage to test out if that was a wise choice.
 
You know another variable here is how the power logger displays knock counts, meaning does 1 represent 1 knock incident or a low signal lasting a period of time to a value of, let's say 1000. Since the ECM's pulse accumaltor counts up to some 65,000ish I would think the number displayed represents a group of knock counts. Just observing a tech 1 scanner on my '92 TPI car I can see knock counts accumulate up 60 or so and that's just around town driving. and I'm not sure how that scanner represents knock counts either.

I guess turbobob could answer the powerlogger knock counts question.??
 
The culprit could be the exhaust however the DP doesn't seem to be making contact with anything underneith. I do remember cranking the whole thing up pretty tight to tuck the mufflers up.

It's annoying not even being able to get on it just a little without the Knock gauge in the red. No fun.

BTW, how do you post Power Logeer data? Do I have to upload it as a pic attachment?
 
Top