I am using 2.176 with Emu Black.
In older versions also, i was surprised by the large changes in some cells, only to be corrected with the next testrun.
The current situation:
Autotune suggests a change in the cell for 100 TPS and 3500 rpm by VE -9.5
However, the Lamda differene is 0.0
How can this be ?
The log also shows only minor corrections for short term trim, i do not thin that -9.5 VE is right.
The cell was hit about 100 times, no transients involved.
Older versions (2.175) show the same results.
Mike
I would say that any time you see a large variance like that, you should check the number of samples taken to get the value. If it is a low sample size then it’s best to ignore it as it usually indicates an anomaly in my experience with autotune. I tend to only use autotune with many samples being in the cells I’m changing. If it’s high boost/rpm runs where your samples will be low I like to use logging and targets to make sure the cell is where I want it. Preferably on a dyno of possible.
2 Likes
I do understand that more samples are always better.
But how can a Lambda difference of 0.00 propose a -9.5 VE correction ?
Just look in the surrounding cells. -9.5 is absolutely off base. Something I wouldn’t even consider as “good” data anyway. I take it you are just sitting there revving the engine at idle? If so that’s always going to provide random data.
