please please help me idle is swinging bad...
01-07-2008, 08:04 AM
#1
Teching In
Thread Starter
Join Date: Apr 2006
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
please please help me idle is swinging bad... 
i have had this problem since cam instal.i orderd efi live and got a free tune and have called the guy but he keeps saying vuc. leak but i have spraid carb cleaner everywhere whith no luck. my idle will go from 990 to 430 in about 1.5 sec. up and down up and down.. its driving me crazy!!!!!! i have no idea where to even start,could not tuning for stall cause this or does everyone think vac leak? my mods are comp cam 224-228 .534 .537 112lsa. bbk tb and intake, hs roller rockers, ss lt headers,3"x pipe dumped before axel,preformabuilt tranny w/3800 stall (which i love)from someone on this site. Last edited by SSMIKE319; 01-07-2008 at 08:22 AM. Reason: firgot to add mods
01-07-2008, 08:41 AM
#4
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
There are three things that play in idle surging. The usual three
(fuel, air, spark).
Begin with the low end of the VE table since you put a cam in.
Try to command a stoich open loop AFR and dial in the VE to get
NBO2 switching without the closed loop / prop fuel doing it. Too
rich an idle mixture makes the motor slow to react to air and
unstable. When you find values for your idle-RPM VE (at the
MAP it likes) propagate the same scaling to the rest of that RPM
column. Then try commanding IAC to get lower and higher RPM
and true those up as well. You can get trouble from the 400RPM
column, you can't tune this by histogram or street data because
you spend no time there but you swing into it once and start
the yo-yo if it's too fat.
Check that your IAC is not pegged to either end. Any case where
you command more and don't get more, or command less and
don't get less, also destabilizes the loop big time. "Windup".
Things like an IAC port epoxied in the TB and not redrilled big
enough, or a huge vacuum leak. See this by the IAC count
vs airflow g/sec. If IAC has a free hand this will plot as a
straight clean line. If it's bound up you will see the line bend
over at one end or the other (pic, from my epoxied TB before
(1/4") and after redrilling to 3/8").
Fixing running airflow will help by not putting such a large
"initial stimulus" to the airflow loop. However you have to look
elsewhere for the instability. The PID coefficients could probably
fix the instability if they were retweaked but I haven't found any
good info on that, played some without much result and ended
up going back to stock settings there.
You have an adaptive spark which should settle down RPM swing
in itself, but this needs to be working about a point that makes
sense. Too little advance and another couple degrees doesn't help
the way it should, same for over-advanced. With the cam giving
you "poor man's EGR" in the low RPM you can stand another 10
degrees or so of idle advance. Use bidirectional controls to find
what advance makes idle happy and how much stops increasing
idle RPM at a fixed IAC count. Then give the idle advance about
"happy" degrees and stay well short of "helps none" so that the
adaptive spark has room to contribute meaningfully. Open up
the adaptive spark maximum and make sure it's a rising straight
line (underspeed) and falling straight line (overspeed), watch the
sign because you may be allowed to command more spark rather
than less for overspeed, etc. which would not help. On some
models the adaptive underspeed spark actually "quits" on you and
high-underspeed suddenly gives zero advance adder while lesser
error was getting help; this kind of "cliff" is a good way to get
a hysteretic instability.
While messing about with bidirectional controls you may find it
useful to lock down the "Big 3" (open loop IAC, open loop spark,
open loop AFR) and perturb one individually to gain knowledge
of the motor's response without the others trying to meddle.
Whichever one seems to help (raise RPM) the most, is probably
the short leg of the stool.
(fuel, air, spark).
Begin with the low end of the VE table since you put a cam in.
Try to command a stoich open loop AFR and dial in the VE to get
NBO2 switching without the closed loop / prop fuel doing it. Too
rich an idle mixture makes the motor slow to react to air and
unstable. When you find values for your idle-RPM VE (at the
MAP it likes) propagate the same scaling to the rest of that RPM
column. Then try commanding IAC to get lower and higher RPM
and true those up as well. You can get trouble from the 400RPM
column, you can't tune this by histogram or street data because
you spend no time there but you swing into it once and start
the yo-yo if it's too fat.
Check that your IAC is not pegged to either end. Any case where
you command more and don't get more, or command less and
don't get less, also destabilizes the loop big time. "Windup".
Things like an IAC port epoxied in the TB and not redrilled big
enough, or a huge vacuum leak. See this by the IAC count
vs airflow g/sec. If IAC has a free hand this will plot as a
straight clean line. If it's bound up you will see the line bend
over at one end or the other (pic, from my epoxied TB before
(1/4") and after redrilling to 3/8").
Fixing running airflow will help by not putting such a large
"initial stimulus" to the airflow loop. However you have to look
elsewhere for the instability. The PID coefficients could probably
fix the instability if they were retweaked but I haven't found any
good info on that, played some without much result and ended
up going back to stock settings there.
You have an adaptive spark which should settle down RPM swing
in itself, but this needs to be working about a point that makes
sense. Too little advance and another couple degrees doesn't help
the way it should, same for over-advanced. With the cam giving
you "poor man's EGR" in the low RPM you can stand another 10
degrees or so of idle advance. Use bidirectional controls to find
what advance makes idle happy and how much stops increasing
idle RPM at a fixed IAC count. Then give the idle advance about
"happy" degrees and stay well short of "helps none" so that the
adaptive spark has room to contribute meaningfully. Open up
the adaptive spark maximum and make sure it's a rising straight
line (underspeed) and falling straight line (overspeed), watch the
sign because you may be allowed to command more spark rather
than less for overspeed, etc. which would not help. On some
models the adaptive underspeed spark actually "quits" on you and
high-underspeed suddenly gives zero advance adder while lesser
error was getting help; this kind of "cliff" is a good way to get
a hysteretic instability.
While messing about with bidirectional controls you may find it
useful to lock down the "Big 3" (open loop IAC, open loop spark,
open loop AFR) and perturb one individually to gain knowledge
of the motor's response without the others trying to meddle.
Whichever one seems to help (raise RPM) the most, is probably
the short leg of the stool.
01-07-2008, 09:01 AM
#6
Teching In
Thread Starter
Join Date: Apr 2006
Posts: 9
Likes: 0
Received 0 Likes
on
0 Posts
thank you very very much but i honestly did not understand. how do i comand open loop? im just learning (or i should say trying to) which table do i change for spark?
Last edited by SSMIKE319; 01-07-2008 at 09:08 AM.
01-07-2008, 10:54 AM
#7
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
I'm not familiar with later EFILive, but my old EFILiveV6 had
bidirectional controls in the scanner that would let you push
things around.
You have idle spark in gear, P/N (on autos) tables. One thing
to keep in mind is, you want these and the main spark table
to be coherent where they overlap (so as not to have crazy
changes in spark, as you enter / exit idle (TPS threshold).
You can do that blending once you have the idle situation
sorted.
bidirectional controls in the scanner that would let you push
things around.
You have idle spark in gear, P/N (on autos) tables. One thing
to keep in mind is, you want these and the main spark table
to be coherent where they overlap (so as not to have crazy
changes in spark, as you enter / exit idle (TPS threshold).
You can do that blending once you have the idle situation
sorted.
