lm-1 wide band smoothing the graph
#1
TECH Enthusiast
Thread Starter
Join Date: Jan 2005
Location: wisconsin
Posts: 658
Likes: 0
Received 0 Likes
on
0 Posts
lm-1 wide band smoothing the graph
why when i smooth the graph then my afr changes alot, can someone tell
me how to look at this in a diffrent fashion. when i look at the graph
it shows 12.5 then smooth and its says 14.3. maby im just plain
stupid, i get these senior moments sometimes. thanks
me how to look at this in a diffrent fashion. when i look at the graph
it shows 12.5 then smooth and its says 14.3. maby im just plain
stupid, i get these senior moments sometimes. thanks
#2
Moderator
iTrader: (11)
Join Date: Mar 2002
Location: East Central Florida
Posts: 12,605
Likes: 0
Received 6 Likes
on
6 Posts
If you have zero "smoothing" then the LM-1 puts out data
in-the-moment at a 12/sec frame rate. The scan tool is
reading meanwhile at 10/sec (<=24 PID) or 5/sec (48). So
you may have a bit of "beat action" between the two data
rates, like it could be bouncing between 12:1 and 16:1 in
a sort of cyclic pattern but you only happen to catch the
12:1 samples. Next time you might catch all the 16:1s
instead.
Smoothing makes the LM-1 data slow and averaged so
the scan tool is now well faster and always seeing the
same data anyway.
You might want to pull a LM-1 log and a scan tool log
from the same run, and compare then (different viewer
but the idea is, see how much data you might be
missing.
Now if your wideband data is extremely choppy, despite
what the smoothed result might be, you may want to
have a look at things like cylinder balance (injectors);
what accounts for any large uneveness in the exhaust
gas content cylinder to cylinder?
You might also like to cut the PID list to <12 and see
how the scan tool data looks relative to a LM-1 log.
If you start to catch all the data (maybe a bit of time
skew as the sample strobe slides past the data output
transition, around the beat-circle) then you can place
more faith in the logged data, that you aren't dropping
useful info etc.
in-the-moment at a 12/sec frame rate. The scan tool is
reading meanwhile at 10/sec (<=24 PID) or 5/sec (48). So
you may have a bit of "beat action" between the two data
rates, like it could be bouncing between 12:1 and 16:1 in
a sort of cyclic pattern but you only happen to catch the
12:1 samples. Next time you might catch all the 16:1s
instead.
Smoothing makes the LM-1 data slow and averaged so
the scan tool is now well faster and always seeing the
same data anyway.
You might want to pull a LM-1 log and a scan tool log
from the same run, and compare then (different viewer
but the idea is, see how much data you might be
missing.
Now if your wideband data is extremely choppy, despite
what the smoothed result might be, you may want to
have a look at things like cylinder balance (injectors);
what accounts for any large uneveness in the exhaust
gas content cylinder to cylinder?
You might also like to cut the PID list to <12 and see
how the scan tool data looks relative to a LM-1 log.
If you start to catch all the data (maybe a bit of time
skew as the sample strobe slides past the data output
transition, around the beat-circle) then you can place
more faith in the logged data, that you aren't dropping
useful info etc.
#3
TECH Enthusiast
Thread Starter
Join Date: Jan 2005
Location: wisconsin
Posts: 658
Likes: 0
Received 0 Likes
on
0 Posts
thanks jimmy but im not hook to hp tuners im trying to tell
what im running afr wise, one way it looks right and the other
way smoothing dont... ya ill have to see if i can line the scan up
with the lm1 .thanks
what im running afr wise, one way it looks right and the other
way smoothing dont... ya ill have to see if i can line the scan up
with the lm1 .thanks