Knock correction - interpretation
#1
Knock correction - interpretation
Having finally got the correct PSi3 unit after the "shall remain nameless" shop sold me one for a Mini, I would appreciate some words of wisdom regarding the knock correction readings.
Generally it sits at zero then displays readings up to 11 -12 deg. depending upon throttle. I haven't noted readings when giving it some as I tend not to look down then! Are these readings the amount of advance or retard from the base timing figure? (there being no - in front of the figure) As there is no Knocklink activity I assumed it was the ECU advancing the timing.
I know this has been discussed some time ago, but if anyone can confirm and maybe compare I would appreciate it.
JohnD
Generally it sits at zero then displays readings up to 11 -12 deg. depending upon throttle. I haven't noted readings when giving it some as I tend not to look down then! Are these readings the amount of advance or retard from the base timing figure? (there being no - in front of the figure) As there is no Knocklink activity I assumed it was the ECU advancing the timing.
I know this has been discussed some time ago, but if anyone can confirm and maybe compare I would appreciate it.
JohnD
#2
Scooby Regular
Join Date: Feb 2003
Location: CHIPP'N HAM
Posts: 3,931
Likes: 0
Received 0 Likes
on
0 Posts
John,
I am no expert, and I am quoting (from memory, so may be slightly fuzzy) what Bob Rawle told me when I had Knock Link activity, and I was comparing it to my KC value on my SECS.
The KC value displayed is the correction of timing being applied by the ec to avoid DET occuring, this figure is not the total value, as there is further correction within the ecu's "inteligent learning" software that is not read by the SECS. This value could be + or - 6? degrees of the figure displayed...
My KC usually hovers around 0, until you plant your foot, then rises to 13 or so.. (but this could be more, just not shown).
I am no expert, and I am quoting (from memory, so may be slightly fuzzy) what Bob Rawle told me when I had Knock Link activity, and I was comparing it to my KC value on my SECS.
The KC value displayed is the correction of timing being applied by the ec to avoid DET occuring, this figure is not the total value, as there is further correction within the ecu's "inteligent learning" software that is not read by the SECS. This value could be + or - 6? degrees of the figure displayed...
My KC usually hovers around 0, until you plant your foot, then rises to 13 or so.. (but this could be more, just not shown).
#3
Scooby Regular
Join Date: Nov 2000
Location: 32 cylinders and many cats
Posts: 18,658
Likes: 0
Received 1 Like
on
1 Post
System is complex, but from memory the system is as follows:
On a new age, positive (unsigned) knock correction is additive to the base map and is the item in [] below.
Total timing = base map + [(advance multiplier/16*correction map) + fine learning]
Negative knock correction is instantaneous retard which will be applied against the fine learning map for next time. So at that instant the ECU has perceived knock to have occurred.
Total timing = base map + ((advance multiplier/16*correction map) + fine learning + knock correction)
....here the displayed knock correction has a completely different meaning when negative.
The fine learning can be negative, or positive if the advance multiplier is under 16.
The fine learning is saved in a dynamic map in the ECU and is always changing. There are trip points that will then cause the advance multiplier to adjust if the fine learning map hits the end of its authority.
With SECS I don't think you can see the advance multiplier.
I may be slightly off on the exact priorities and bracket positioning, but it gives you an idea, and I'm sure some kind soul will correct me if wrong
I am more used to the old fashioned ECUs which have work similarly but report knock correction as the present fine learning map value, ish.
With reverse engineered ECU tools it is sometimes difficult to fully follow what is intended compared with a standalone.
On a new age, positive (unsigned) knock correction is additive to the base map and is the item in [] below.
Total timing = base map + [(advance multiplier/16*correction map) + fine learning]
Negative knock correction is instantaneous retard which will be applied against the fine learning map for next time. So at that instant the ECU has perceived knock to have occurred.
Total timing = base map + ((advance multiplier/16*correction map) + fine learning + knock correction)
....here the displayed knock correction has a completely different meaning when negative.
The fine learning can be negative, or positive if the advance multiplier is under 16.
The fine learning is saved in a dynamic map in the ECU and is always changing. There are trip points that will then cause the advance multiplier to adjust if the fine learning map hits the end of its authority.
With SECS I don't think you can see the advance multiplier.
I may be slightly off on the exact priorities and bracket positioning, but it gives you an idea, and I'm sure some kind soul will correct me if wrong
I am more used to the old fashioned ECUs which have work similarly but report knock correction as the present fine learning map value, ish.
With reverse engineered ECU tools it is sometimes difficult to fully follow what is intended compared with a standalone.
Last edited by john banks; 10 September 2004 at 01:12 PM.
#6
Former Sponsor
iTrader: (3)
note that on 06 onwards ecu's the advance multiplier does not go on a scale of 1-16. It is maximum of 1 and goes up in 0.06 steps.
Therefore it is a true 'multiplier' ie:
Total timing = base map + ((advance multiplier*correction map) + fine learning + knock correction)
also there are 'other' timing corrections added/removed.
If you are wanting to monitor knock in PSI3 then monitor the Knock Retard not the Knock Correction. Knock correction may well still show positive even if there is knock.
Therefore it is a true 'multiplier' ie:
Total timing = base map + ((advance multiplier*correction map) + fine learning + knock correction)
also there are 'other' timing corrections added/removed.
If you are wanting to monitor knock in PSI3 then monitor the Knock Retard not the Knock Correction. Knock correction may well still show positive even if there is knock.
Thread
Thread Starter
Forum
Replies
Last Post
Mattybr5@MB Developments
Full Cars Breaking For Spares
28
28 December 2015 11:07 PM
Mattybr5@MB Developments
Full Cars Breaking For Spares
12
18 November 2015 07:03 AM