Knock during spool/load up
Knock voltage has nothing to do with the knock being real or not. The whole ECU knock routine and algorithms have been explained rather nicely by John Banks, if you have any questions.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
Knock voltage has nothing to do with the knock being real or not. The whole ECU knock routine and algorithms have been explained rather nicely by John Banks, if you have any questions.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
Knock voltage has nothing to do with the knock being real or not. The whole ECU knock routine and algorithms have been explained rather nicely by John Banks, if you have any questions.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
But, the voltage won't tell you if it's real knock or not, nor will the knock counts. The ECU simply registers knock counts when the voltage is above a certain level. Phantom and real knock will create the same knock voltages or knock counts as a result.
The only way to really tell has already been mentioned:
1. Adjust timing
2. Adjust fuel
3. Use higher octane fuel
If the knock stays the same, it's most likely phantom knock and you can then adjust the knock thresholds to not register the knock counts.
I have to disagreeable with you here.
Please, in your expansive knowledge, explain then the meaning of the knock multipliers and their ability (if raised) to filter out low level "noise" from real knock if, in fact, the voltage is the same regardless of if the knock is real or not.
Plotting knock voltage and knock count one can determine noise from real knock.
We can all agree that valve train noise and even metal to metal rubbing can be picked up as knock. By simply raising the knock thresholds in the multiplier tables we can eliminate this noise as seen as knock as it will require a harder knock (higher voltage reading) for the knock to register a count.
Applaud, I don't think you understand what I am trying to explain. I don't really feel like writing a book here...
But, whether the knock is real or not, it will register a voltage. If that voltage is over the threshold, then the ECU registers knock counts. The amount of knock counts depends on how much and how longer over the threshold. It is these knock counts that the ECU uses to pull timing (1* for every 3 counts), as well as a long term trim using the octane number to begin interpolation to the low octane maps for sustained knock count over 5.
There is no way, just by seeing the voltage, to tell whether that registered knock is phantom knock or real knock. The ECU sees both EXACTLY the same way. All knock is to the ECU is a voltage spike over a threshold value.
You have to determine whether it is real or phantom by the above mentioned methods...then you can adjust the threshold safely to eliminate the phantom knock. Of course, this will also eliminate any real knock in that range as well up to the new threshold level that you set.
But, whether the knock is real or not, it will register a voltage. If that voltage is over the threshold, then the ECU registers knock counts. The amount of knock counts depends on how much and how longer over the threshold. It is these knock counts that the ECU uses to pull timing (1* for every 3 counts), as well as a long term trim using the octane number to begin interpolation to the low octane maps for sustained knock count over 5.
There is no way, just by seeing the voltage, to tell whether that registered knock is phantom knock or real knock. The ECU sees both EXACTLY the same way. All knock is to the ECU is a voltage spike over a threshold value.
You have to determine whether it is real or phantom by the above mentioned methods...then you can adjust the threshold safely to eliminate the phantom knock. Of course, this will also eliminate any real knock in that range as well up to the new threshold level that you set.
Last edited by l2r99gst; May 1, 2010 at 01:40 PM.
Applaud, I don't think you understand what I am trying to explain. I don't really feel like writing a book here...
But, whether the knock is real or not, it will register a voltage. If that voltage is over the threshold, then the ECU registers knock counts. The amount of knock counts depends on how much and how longer over the threshold. It is these knock counts that the ECU uses to pull timing (1* for every 3 counts), as well as a long term trim using the octane number to begin interpolation to the low octane maps for sustained knock count over 5.
There is no way, just by seeing the voltage, to tell whether that registered knock is phantom knock or real knock. The ECU sees both EXACTLY the same way. All knock is to the ECU is a voltage spike over a threshold value.
You have to determine whether it is real or phantom by the above mentioned methods...then you can adjust the threshold safely to eliminate the phantom knock. Of course, this will also eliminate any real knock in that range as well up to the new threshold level that you set.
But, whether the knock is real or not, it will register a voltage. If that voltage is over the threshold, then the ECU registers knock counts. The amount of knock counts depends on how much and how longer over the threshold. It is these knock counts that the ECU uses to pull timing (1* for every 3 counts), as well as a long term trim using the octane number to begin interpolation to the low octane maps for sustained knock count over 5.
There is no way, just by seeing the voltage, to tell whether that registered knock is phantom knock or real knock. The ECU sees both EXACTLY the same way. All knock is to the ECU is a voltage spike over a threshold value.
You have to determine whether it is real or phantom by the above mentioned methods...then you can adjust the threshold safely to eliminate the phantom knock. Of course, this will also eliminate any real knock in that range as well up to the new threshold level that you set.
Your "cuts out" issue could be:
Too rich
Too lean
Needs fresh plugs
Hitting overboost fuel cut
And a whole lot more...
Can you post a data log of an incident?
I am getting something weird happening as well...when I do pulls from 3K the first pull seems to have 3-5 counts of knock from 3K-5K the clean to 7500. The next pull almost always is knock free. I have richened up spoolup as well as decreased timing and in several different maps this first pull knock usually happens like the motors cold. Now if I do several half throttle pulls before that first wot pull it pulls clean. I have FP4R's on stock springs.
Evolving Member
Joined: May 2008
Posts: 346
Likes: 0
From: Dallas, TX
I am getting something weird happening as well...when I do pulls from 3K the first pull seems to have 3-5 counts of knock from 3K-5K the clean to 7500. The next pull almost always is knock free. I have richened up spoolup as well as decreased timing and in several different maps this first pull knock usually happens like the motors cold. Now if I do several half throttle pulls before that first wot pull it pulls clean. I have FP4R's on stock springs.
Evolving Member
Joined: May 2008
Posts: 346
Likes: 0
From: Dallas, TX
i have always had "random" spool up knock (read : noise ) with my setup. I have filtered it out on knock settings, however do extensive logging to ensure its noise and not real before doing that.
I am getting something weird happening as well...when I do pulls from 3K the first pull seems to have 3-5 counts of knock from 3K-5K the clean to 7500. The next pull almost always is knock free. I have richened up spoolup as well as decreased timing and in several different maps this first pull knock usually happens like the motors cold. Now if I do several half throttle pulls before that first wot pull it pulls clean. I have FP4R's on stock springs.
post up your log file.. chances are it has to much timing at spool area.. im only running 10-11* max at spool and dropping it off fast until i reach my peak torque area then ramping it up from there.. if you paste your log i will see from the timing curve if its right or wrong.. there shouldnt be any pulls that knock and ones that dont if everythings setup the way it should be.
