MEGAsquirt A place to collectively sort out this megasquirt gizmo

MS3x Knock Module Settings Thread

Thread Tools
 
Search this Thread
 
Old 08-25-2016, 12:11 AM
  #61  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Looking more at the data and knock per cylinder it looks like I need to turn the gain on 3 down a decent amount and gain 2 down some too



cyotani is offline  
Old 08-25-2016, 03:02 AM
  #62  
Senior Member
iTrader: (6)
 
yossi126's Avatar
 
Join Date: Sep 2013
Posts: 854
Total Cats: -15
Default

I am about to configure my knock window and your post was very helpful. Catted!
yossi126 is offline  
Old 08-25-2016, 12:27 PM
  #63  
SADFab Destructive Testing Engineer
iTrader: (5)
 
aidandj's Avatar
 
Join Date: Apr 2014
Location: Beaverton, USA
Posts: 18,642
Total Cats: 1,866
Default

Is there a reason you aren't looking at knock above 80kpa and abover 4000rpm?
aidandj is offline  
Old 08-25-2016, 01:08 PM
  #64  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

Originally Posted by aidandj
Is there a reason you aren't looking at knock above 80kpa and abover 4000rpm?
nope, no reason. Those were the default settings. The first batch of runs I was mostly trying to determine the appropriate knock gain per cylinder and knock threshold vs RPM.
cyotani is offline  
Old 08-25-2016, 11:06 PM
  #65  
Senior Member
iTrader: (8)
 
cyotani's Avatar
 
Join Date: Jan 2012
Location: Azusa, CA
Posts: 1,407
Total Cats: 116
Default

So this is probably obvious to most of you guys...
But for those who are lost on cylinder gain tuning, here is one way to do it:

1. Set all knock cylinder gains to 1.000
2. Data log a couple of WOT pulls using the entire RPM range. I did a couple of 0-60 on the street.
3. Plot the 4 knock channels using excel or any spreadsheet program
5. Create a math channel with each knock cylinder channel and multiply it by a constant gain and plot the results. The goal is to have the 4 cylinder stack on top of each other as much as possible (assuming the log did not have any knock)
- in my example cylinders 1 and 4 were lower than 2 and 3. This is due to the location of the knock sensor being closer to these cylinders. I tried a couple different multipliers and graphed them in excel until they stacked as much as possible (compare left vs right graphs).
6. update the gains in tunerstudio and burn the new tune.
7. datalog a couple more pulls and view the data to verify that they stack as best possible.



cyotani is offline  
Old 03-02-2017, 07:48 AM
  #66  
Junior Member
 
Barton's Avatar
 
Join Date: Jun 2015
Location: VIC, Australia
Posts: 182
Total Cats: 57
Default

Sorry to bump an old thread but it's relevant and I didn't want to make a new one just for this. I recently setup the knock control and sensor on my vvt engine however my signal doesn't seem to behave normally and doesn't increase with rpm. Values are highest at around idle and seem fairly consistent above that. Since the wiring loom is still na6 I ran a separate shielded wire back to the ecu for the sensor. Still appears to spike when it knocks when too much timing is added but just seems strange that the signals aren't increasing with rpm like others.

Name:  qEvrER8.png
Views: 78
Size:  23.7 KB

Name:  xEDb0pq.png
Views: 78
Size:  24.4 KB
Name:  pSlduai.png
Views: 78
Size:  22.6 KB
Name:  pEgQueG.png
Views: 78
Size:  106.6 KB
Barton is offline  
Old 03-02-2017, 06:41 PM
  #67  
Newb
 
Ivan's Avatar
 
Join Date: Oct 2014
Location: Vancouver, Canada
Posts: 36
Total Cats: -2
Default

Originally Posted by Barton
Sorry to bump an old thread but it's relevant and I didn't want to make a new one just for this. I recently setup the knock control and sensor on my vvt engine however my signal doesn't seem to behave normally and doesn't increase with rpm. Values are highest at around idle and seem fairly consistent above that. Since the wiring loom is still na6 I ran a separate shielded wire back to the ecu for the sensor. Still appears to spike when it knocks when too much timing is added but just seems strange that the signals aren't increasing with rpm like others.
I'm running a MS Labs MS 3 Basic on a 2004. My knock signal looks similar to yours.

Ivan
Ivan is offline  
Related Topics
Thread
Thread Starter
Forum
Replies
Last Post
StratoBlue1109
Miata parts for sale/trade
21
09-30-2018 01:09 PM
90 Turbo
MEGAsquirt
19
10-19-2015 03:23 PM
Trent
WTB
2
10-01-2015 12:15 PM
FrankB
Miata parts for sale/trade
6
09-30-2015 11:48 AM



Quick Reply: MS3x Knock Module Settings Thread



All times are GMT -4. The time now is 10:31 PM.