question about knock retard and timing
05-25-2006, 12:25 AM
#1
11 Second Club
Thread Starter
iTrader: (4)
Join Date: Mar 2005
Posts: 661
Likes: 0
Received 0 Likes
on
0 Posts
question about knock retard and timing I just got my efi live to work and i did a few runs and at WOT the car goes to 22 degrees timing and it shows 6 knock retard...
I guess my question is, what should my timing be at during WOT and what is happening when it is showing 6 knock retard, how much knock is actualy happening?
I guess my question is, what should my timing be at during WOT and what is happening when it is showing 6 knock retard, how much knock is actualy happening?
05-25-2006, 09:13 AM
#2
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
All you know right now is, you would've had 28 (22+6).
Knock is individual events (or a series). Knock retard is
the PCM response to detected knock, applied for a long
timescale, to get you through it without hassling GM's
warranty claims department. The KR settings are all
about safety and not at all about information. The KR
jumps up fast and goes away slow. This suppresses
ping all right, but you can't see the true timing tolerance
through all that foam.
If you reduce the attack rate (say, 1/4 the stock value)
and increase the decay rate (like, 4X) you will have KR
that increments finely per knock event, and bleeds down
in a few tenths of a second (stock KR can persist all the
way down the track). I don't know what units EFILive
likes to display for these, so have to talk relative.
This takes away safety but lets you see the real deal.
Every time KR jumps up, you had a ping. If you pull steady
at RPM*load you would see the KR ratchet up to the point
of just barely pinging, and motorboat along at that level;
this would show you the ping threshold.
Best torque lies a couple of degrees to the left.
22 degrees is low tolerance for a stockish / external bolt-
ons motor. High compression, big cam might need lower
than this; Nitrous, forced induction, anything that raises
the pre-spark cylinder pressure needs lower still. All of
this follows the increasing burn speed with gas pressure,
needing less "lead" to position the pressure pulse peak
after TDC in actuality. If you have rising burn pressure
plus you are still on the compression stroke, that's the
devil's double whammy.
Once you have better visibility into the coming and going
of KR, you can back into the causes; you can look at the
changes (up-KR) and see what precedes them. Look in
particular for hard dips in the O2 voltage on opening
throttle, which show a lean transient; there are "holes"
in the stock setup, where you fall out of the combined
MAF/SD normal airflow metering mode and into straight
speed density; speed density error is changed by any
mods aft of the intake manifold, and any mod worth
doing biases the error toward the lean side. Change in
MAP, like from opening throttle, is the trigger here.
If your O2s jack lean for a while after you mash it, you
know you have to fix the VE table. If the dip is very
brief then maybe you want to play with transient fueling
or just patch it over with burst knock retard, but more
than a couple "ticks" worth of lean dip is saying VE is off,
in the MAP*RPM region you just left.
Other than the usual mods-induced VE error, there are
things like coolant temps, MAF error, fuel-side errors
(low rail pressure, crusty fuel filter, poor fuel pump
supply voltage, injectors that you've outgrown) that
can produce lean conditions. The region & character of
knock will point somewhat to the source (filter/pump/
injectors will show up only at the top end while low RPM
KR points more to the VE airflow model; coolant temps
are a source of ping that comes and goes at "random"
(coolant temps on stock setup F-bodies can swing from
195 to 230+, just from rolling airflow and engine output;
not truly random, but following a variable not observed).
Knock is individual events (or a series). Knock retard is
the PCM response to detected knock, applied for a long
timescale, to get you through it without hassling GM's
warranty claims department. The KR settings are all
about safety and not at all about information. The KR
jumps up fast and goes away slow. This suppresses
ping all right, but you can't see the true timing tolerance
through all that foam.
If you reduce the attack rate (say, 1/4 the stock value)
and increase the decay rate (like, 4X) you will have KR
that increments finely per knock event, and bleeds down
in a few tenths of a second (stock KR can persist all the
way down the track). I don't know what units EFILive
likes to display for these, so have to talk relative.
This takes away safety but lets you see the real deal.
Every time KR jumps up, you had a ping. If you pull steady
at RPM*load you would see the KR ratchet up to the point
of just barely pinging, and motorboat along at that level;
this would show you the ping threshold.
Best torque lies a couple of degrees to the left.
22 degrees is low tolerance for a stockish / external bolt-
ons motor. High compression, big cam might need lower
than this; Nitrous, forced induction, anything that raises
the pre-spark cylinder pressure needs lower still. All of
this follows the increasing burn speed with gas pressure,
needing less "lead" to position the pressure pulse peak
after TDC in actuality. If you have rising burn pressure
plus you are still on the compression stroke, that's the
devil's double whammy.
Once you have better visibility into the coming and going
of KR, you can back into the causes; you can look at the
changes (up-KR) and see what precedes them. Look in
particular for hard dips in the O2 voltage on opening
throttle, which show a lean transient; there are "holes"
in the stock setup, where you fall out of the combined
MAF/SD normal airflow metering mode and into straight
speed density; speed density error is changed by any
mods aft of the intake manifold, and any mod worth
doing biases the error toward the lean side. Change in
MAP, like from opening throttle, is the trigger here.
If your O2s jack lean for a while after you mash it, you
know you have to fix the VE table. If the dip is very
brief then maybe you want to play with transient fueling
or just patch it over with burst knock retard, but more
than a couple "ticks" worth of lean dip is saying VE is off,
in the MAP*RPM region you just left.
Other than the usual mods-induced VE error, there are
things like coolant temps, MAF error, fuel-side errors
(low rail pressure, crusty fuel filter, poor fuel pump
supply voltage, injectors that you've outgrown) that
can produce lean conditions. The region & character of
knock will point somewhat to the source (filter/pump/
injectors will show up only at the top end while low RPM
KR points more to the VE airflow model; coolant temps
are a source of ping that comes and goes at "random"
(coolant temps on stock setup F-bodies can swing from
195 to 230+, just from rolling airflow and engine output;
not truly random, but following a variable not observed).
