Discuss going into closed loop on cammed car?
05-21-2009, 02:30 AM
#1
11 Second Club
Thread Starter
iTrader: (14)
Join Date: Nov 2004
Location: Michigan (Macomb or Lansing)
Posts: 1,653
Likes: 0
Received 0 Likes
on
0 Posts
Discuss going into closed loop on cammed car? My VE and MAF are dialed in pretty perfectly. I have torquer v3 (231/234 .644/.598 111lsa) ls6 intake, longtubes, full boltons.
Discuss closed loop. Im debating going to it. Also discuss leaving ltfts off.
What are pros and cons?
I always seem to get bad gas mileage, partially because lots of gassing and stop and go but I am wondering if this can help.
Discuss closed loop. Im debating going to it. Also discuss leaving ltfts off.
What are pros and cons?
I always seem to get bad gas mileage, partially because lots of gassing and stop and go but I am wondering if this can help.
05-21-2009, 10:58 AM
#2
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
The only question is, can you trust the narrowband
O2 sensors to give you good feedback for trimming.
Long tube headers are a problem, you may be short
on heat by the time the gas gets to them. Also the
delay to get from cylinder to sensor can be long
and possibly destabilizing. But I have run closed
loop with a more modest cam and coated LTs in
a much more consistently warm climate, OK.
The cam can also put excess air into the exhaust
and warp both the NB and wideband readings to
the lean side, which would cause rich trimming
(though the wideband may agree in error). The
closed loop idle might take a double hit from O2
heat (read lean, when cool) and shot-through air
(read lean).
Your "handle" for fixing some of this is the O2
mV vs Airflow Mode table. You can bump this lower
to lean out the idle (or any) closed loop target
mixture. Begin with looking at the wideband
reading and bump the target mV until it closes
in on a suitable (~stoich) indicated mixture. On
a big cam blower motor 500mV was making ~ 13:1
at idle and going down to ~350mV gave us ~15:1
closed loop. 350mV is what the cars use, stock.
But come winter, those long tubes may just put
your closed loop low pedal operation out to lunch.
All you can do is try, and keep at it.
LTFTs on if you can eliminate all false trimming.
LTFTs off, STFT-only if your base tune (OL) is
stable and centered and weather-independent
(yeah, good luck) or you just have a mistrimming
/ history effect that you can't live with.
Trimming gets a lot better if you change the FTC
boundaries to fit how you drive and how the motor
behaves. Like, you want all cells "in play" (not half
assigned over 6500RPM), lose one to idle (Cell 0)
and one to PE mode (15). Your RPM boundaries
something like 1000, 2500, 4000 and your MAP
lined up to min(idle, cruise) MAP less a hair, PE MAP
threshold less a hair and split the difference for the
middle boundary. Idea is, to subdivide the closed-
loop operating space as finely as possible and to
favor resolution where the closed loop has the
hardest time (i.e. where VE is changing the most
vs RPM and MAP, where exhaust gas temp varies
the most, etc. - low flows). Like, not much difference
between 3000RPM and 4000RPM in how the motor
runs, but big difference 1000RPM to 2000RPM.
Binning it all into 0-2500 and 2500-6500 is lame
and makes LTFTs and STFTs slosh around instead
of being useful memory relevant to the present
operating point.
O2 sensors to give you good feedback for trimming.
Long tube headers are a problem, you may be short
on heat by the time the gas gets to them. Also the
delay to get from cylinder to sensor can be long
and possibly destabilizing. But I have run closed
loop with a more modest cam and coated LTs in
a much more consistently warm climate, OK.
The cam can also put excess air into the exhaust
and warp both the NB and wideband readings to
the lean side, which would cause rich trimming
(though the wideband may agree in error). The
closed loop idle might take a double hit from O2
heat (read lean, when cool) and shot-through air
(read lean).
Your "handle" for fixing some of this is the O2
mV vs Airflow Mode table. You can bump this lower
to lean out the idle (or any) closed loop target
mixture. Begin with looking at the wideband
reading and bump the target mV until it closes
in on a suitable (~stoich) indicated mixture. On
a big cam blower motor 500mV was making ~ 13:1
at idle and going down to ~350mV gave us ~15:1
closed loop. 350mV is what the cars use, stock.
But come winter, those long tubes may just put
your closed loop low pedal operation out to lunch.
All you can do is try, and keep at it.
LTFTs on if you can eliminate all false trimming.
LTFTs off, STFT-only if your base tune (OL) is
stable and centered and weather-independent
(yeah, good luck) or you just have a mistrimming
/ history effect that you can't live with.
Trimming gets a lot better if you change the FTC
boundaries to fit how you drive and how the motor
behaves. Like, you want all cells "in play" (not half
assigned over 6500RPM), lose one to idle (Cell 0)
and one to PE mode (15). Your RPM boundaries
something like 1000, 2500, 4000 and your MAP
lined up to min(idle, cruise) MAP less a hair, PE MAP
threshold less a hair and split the difference for the
middle boundary. Idea is, to subdivide the closed-
loop operating space as finely as possible and to
favor resolution where the closed loop has the
hardest time (i.e. where VE is changing the most
vs RPM and MAP, where exhaust gas temp varies
the most, etc. - low flows). Like, not much difference
between 3000RPM and 4000RPM in how the motor
runs, but big difference 1000RPM to 2000RPM.
Binning it all into 0-2500 and 2500-6500 is lame
and makes LTFTs and STFTs slosh around instead
of being useful memory relevant to the present
operating point.
05-21-2009, 11:44 AM
#3
TECH Enthusiast
iTrader: (1)
Join Date: Dec 2005
Location: Denver International Airport, Colorado USA
Posts: 549
Likes: 0
Received 0 Likes
on
0 Posts
Well Jimmyblue gave one of the most comprehensive and educated responses concerning the pitfalls of closed-loop and header installed narrowbands O2's readings.
The only thing I will add it is that my experience with using open-loop exclusively has its own share of problems. Unless you are successful in having a VE Table nailed down, I and others have experienced variations in AFR readings from a morning run vs an evening run. I am still convinced that some of us attempting to tune open-loop have a difficult time sampling the same RPM/MAP value as atmospheric conditions change.
Others do not have that problem, but just giving you a valid experience some of us have had.
Jimmyblue is correct in that the narrowbands give 'wacky' readings with bigger cams especially under 2000 Rpm's. But, I still think having a trim function is adventageous. Yes, the trimming lags reality, but as stated getting your FTC boundaries set-up properly can really help.
Personally I set-up my FTC boundaries a little 'lower' than JimmyB's. I use 1000/1600/2400, but that is where I cruise around at.
My cam is ~10 degrees of overlap, so my issues are not as complicated as those using bigger cams.
Oher options are to run closed-loop SD if you don't like MAF. While it took some effort to lay out the FTC cells, my subjective feeling is that closed-loop can work. My ride is smooth and logging the wideband and FTC cells, everything still switches at ~14.7 with no apparent negative results.
Hopefully the additional information helps. Each has advantages and disadvantages, and require work to get it right.
Best of luck.
..WeathermanShawn.
The only thing I will add it is that my experience with using open-loop exclusively has its own share of problems. Unless you are successful in having a VE Table nailed down, I and others have experienced variations in AFR readings from a morning run vs an evening run. I am still convinced that some of us attempting to tune open-loop have a difficult time sampling the same RPM/MAP value as atmospheric conditions change.
Others do not have that problem, but just giving you a valid experience some of us have had.
Jimmyblue is correct in that the narrowbands give 'wacky' readings with bigger cams especially under 2000 Rpm's. But, I still think having a trim function is adventageous. Yes, the trimming lags reality, but as stated getting your FTC boundaries set-up properly can really help.
Personally I set-up my FTC boundaries a little 'lower' than JimmyB's. I use 1000/1600/2400, but that is where I cruise around at.
My cam is ~10 degrees of overlap, so my issues are not as complicated as those using bigger cams.
Oher options are to run closed-loop SD if you don't like MAF. While it took some effort to lay out the FTC cells, my subjective feeling is that closed-loop can work. My ride is smooth and logging the wideband and FTC cells, everything still switches at ~14.7 with no apparent negative results.
Hopefully the additional information helps. Each has advantages and disadvantages, and require work to get it right.
Best of luck.
..WeathermanShawn.
