Innovate smoothing option?
01-09-2006, 05:50 PM
#1
TECH Enthusiast
Thread Starter
iTrader: (15)
Join Date: May 2005
Location: Valdosta, georgia
Posts: 549
Likes: 0
Received 0 Likes
on
0 Posts
Innovate smoothing option? I am logging AFR against MAP and RPM. I was wondering if I set the smoothing to the max amount of 4.18 secs. It will take more of an average right? And this is what I want correct, since it will show the very last value taken if you don't smooth.
01-09-2006, 06:03 PM
#2
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
4 seconds is a pretty long time. That's like a whole
gear's worth of RPM traversal at WOT. I'd pull it in
and only average as much as needed to get a clean
appearing line. More than that you'd start to obscure
data, especially if your mixture turns out to have a
strong RPM dependence.
Start with filtering minimized, see if the data looks
short-term-stable, and only bump up the time constant
if you can't deal with the jitter is my recommendation.
gear's worth of RPM traversal at WOT. I'd pull it in
and only average as much as needed to get a clean
appearing line. More than that you'd start to obscure
data, especially if your mixture turns out to have a
strong RPM dependence.
Start with filtering minimized, see if the data looks
short-term-stable, and only bump up the time constant
if you can't deal with the jitter is my recommendation.
01-09-2006, 07:20 PM
#3
TECH Enthusiast
Thread Starter
iTrader: (15)
Join Date: May 2005
Location: Valdosta, georgia
Posts: 549
Likes: 0
Received 0 Likes
on
0 Posts
The problem i'm having is most of my points seem good and seem to be almost what i'm commanding. But there is always a few points that will read like 15.05 on one run and in 2 hours later will read like 13.05. I mean how can I tune that? Don't know which one is accurate.
01-09-2006, 07:28 PM
#4
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
I'd go for "eyeball smoothing" and ignore the outliers.
I work almost entirely in Excel, plot the WBO2 AFR
and draw lines through the center of the "cloud".
You will always have to deal with bad data somehow,
but averaging it into the good data corrupts it all.
The "sports" are probably just electrical noise or a
misfire coughing out a little loogey. Just ignore the
"one percenters" and keep your eyes on the prize.
I work almost entirely in Excel, plot the WBO2 AFR
and draw lines through the center of the "cloud".
You will always have to deal with bad data somehow,
but averaging it into the good data corrupts it all.
The "sports" are probably just electrical noise or a
misfire coughing out a little loogey. Just ignore the
"one percenters" and keep your eyes on the prize.