How to tune high load points on VE table???
Thread Starter
12 Second Club
Joined: Oct 2003
Posts: 117
Likes: 0
How to tune high load points on VE table??? Just curious how you guys who are doing Mafless tuning are adjusting the high load VE cells? Are you just hand smoothing a certain trend or what? How do you know what the cell at say 90 and 5000rpm is rich or lean? Hopefully most people would be in PE at that point and if your VE table is correct you should be locked at 0 which gives no actual indication of what that cell actually is. You could be very lean in that cell but still get a 0 because of the PE mode being active. I don't have a WB so is there a safe way to get in that high load area without being in PE to see what it is really at before the pcm locks it for PE? I am doing all my VE tuning off of LTFT averages and PE using the O2's and erring on the rich side (950mv). It would take a lot of WOT runs and some pretty accurate O2's to get a feel for what the VE cell should be in a PE state. I don't trust the O2's enough to try and adjust the VE according to them.
Thanks
Matt
Thanks
Matt
Thread Starter
12 Second Club
Joined: Oct 2003
Posts: 117
Likes: 0
Originally Posted by nbmls1ta
Just curious how you guys who are doing Mafless tuning are adjusting the high load VE cells? Are you just hand smoothing a certain trend or what? How do you know what the cell at say 90 and 5000rpm is rich or lean? Hopefully most people would be in PE at that point and if your VE table is correct you should be locked at 0 which gives no actual indication of what that cell actually is. You could be very lean in that cell but still get a 0 because of the PE mode being active. I don't have a WB so is there a safe way to get in that high load area without being in PE to see what it is really at before the pcm locks it for PE? I am doing all my VE tuning off of LTFT averages and PE using the O2's and erring on the rich side (950mv). It would take a lot of WOT runs and some pretty accurate O2's to get a feel for what the VE cell should be in a PE state. I don't trust the O2's enough to try and adjust the VE according to them.
Thanks
Matt
Thanks
Matt
If you had the benefit of some correlation data that
let you make sense of in-car O2s you could use those.
950mV is insanely rich, by the way. I used to fiddle
around looking for 910-920s but even that was well
into 11.5:1 territory. You could shoot for 900mV,
get that consistent, and then play about that point
trading odd fuel for spark until you figure out which
gives you the most out of the motor.
But buying or borrowing a wideband meter just
eliminates a whole lot of cut-and-try from the
process (presuming you aren't just doing it because
you enjoy it).
If you have a couple of different O2 mV readings
with corresponding wideband results you can pretty
well pick your location on or between the "standard"
narrowband sensor curves, interpolate your own
with a pencil, wotk out your best guess AFR, and
go correct by evaluating that against commanded.
If you think your actual AFR is leaner than what you
log commanded AFR at, increment VE where you have
data and stucco in around it until it's built up smooth.
But this is prone to be a little fuzzy, what with thermal
history pushing your readings around and so on.
But like they say, when all you have is a hammer,
the world looks like a nail - and you can just chip
away at it using the narrowbands until you get some
better feedback to "close the larger loop".
let you make sense of in-car O2s you could use those.
950mV is insanely rich, by the way. I used to fiddle
around looking for 910-920s but even that was well
into 11.5:1 territory. You could shoot for 900mV,
get that consistent, and then play about that point
trading odd fuel for spark until you figure out which
gives you the most out of the motor.
But buying or borrowing a wideband meter just
eliminates a whole lot of cut-and-try from the
process (presuming you aren't just doing it because
you enjoy it).
If you have a couple of different O2 mV readings
with corresponding wideband results you can pretty
well pick your location on or between the "standard"
narrowband sensor curves, interpolate your own
with a pencil, wotk out your best guess AFR, and
go correct by evaluating that against commanded.
If you think your actual AFR is leaner than what you
log commanded AFR at, increment VE where you have
data and stucco in around it until it's built up smooth.
But this is prone to be a little fuzzy, what with thermal
history pushing your readings around and so on.
But like they say, when all you have is a hammer,
the world looks like a nail - and you can just chip
away at it using the narrowbands until you get some
better feedback to "close the larger loop".
