I am curious to know if there is a way i can filter out values in a histogram which occur during large changes in throttle over a short time? (eg a quick lean condition when stomping on the throttle).
I thought i could maybe create a custom PID that showed me the rate of change in TPS and then use that to filter out values on my histogram, though i'm not sure how to go about creating this. Any ideas?
Cheers
Benno
You shouldn't go lean when stomping the throttle. If that's really happening, something is very wrong (VE?) and you're in danger of cooking the motor.
It is normal for them to go lean when you lift OFF the throttle due to Deceleration Fuel Cut Off. It will make your life easier if you turn DFCO off while tuning, then renable it after you're done for fuel economy.
This is happening to me. The problem, I think is the Acceleration Enrichment...something to do with the refcounts. I'm not really sure how to fix the problem. I created a thread earlier with screenshots:Originally Posted by sgarnett
http://www.hptuners.com/forum/showthread.php?t=4760
The last post has the screenshot from logworks (can't wait for March 1 to get the EIO upgrade)
Just seems that when i am tuning the VE i seem to get better results the smoother i am on the throttle. If i am jumping on and off the throttle quickly i tend to get some values that are way out popping up around the place, whereas when i hold the car at these points for a while they balance themselves back out to around 14.7afr or whatever i am commanding. Was just curious if there is a way to ignore all these values that happen when jumping on and off the throttle.
Cheers
Benno
When you stomp on the throttle you could be entering PE mode, check your 'PE enable TPS' for what throttle position gets you into PE. Also have you disabled Close Loop control as this is when you get AFR 14.7 ie smoother on the throttle.
LS1 LS2 LS7 tuner, L98, LT1 & Haltec
If its something you'd like to see I'd strongly recommend posting in the request section of this forum...it sounds like a very good idea though
It doesn't have to be perfect, it just needs to be done in two weeks...
A wise man once said "google it"
I know this type of thing can be done in flashscan, seems like it can't be done in hptuners (or maybe not yet anyway).
When tuning my VE table i have DFCO disabled, fuel trims/closed loop disabled, COT disabled etc.
Thanks for the help
Benno
If this is only happening while you are scanning in open-loop with PE, DFCO, MAF, etc, turned off, then the simple answer is DON'T stomp it. Make all your throttle changes very smooth. It isn't meant to be a final tune, just a means of collecting a specific type of data. Transient response is going to be totally wacked.
a derivative function is in the works..on the "to do" list...
at some point in the future you will eb able to filter out by amount of change in throttle..among other cool things that will happen with a derivative function...
no set timeframe for this as its on a large "to do" list
some complex filterng is on its way as well
-Scott -
good to hear. thanks for that soundengineer
how about detecting any delta among any number of cells backwards? this would be useful for more than just tps (ie boost when you hold a steady throttle but suddenly boost hits)
Reply With Quote