Histogram plot filter assistance

is there a way to filter the afr error% histogram in such a way that it only displays the lean values and alternatively only the rich values?

Meaning that I want to filter the resultant averaged values after all of the data has been loaded into the histogram....

I was playing around with adding some formulas to the filter referencing sensor values, but the results were undesirable because it appears to filter each individual sensor reading as it builds the histogram.

I only want to filter on the final averaged values.

If you're talking about AFR error I'm assuming you're using a wideband. So all you need to so is filter on the SENS.120 sensor (air fuel ratio). Your filter would be "[SENS.120] < 14.7" to display cell hits richer than stoich. You can use boolean statments to join multiple conditions like "[SENS.120] < 16 AND [SENS.120] > 12" to throw out any excessively lean or rich data. Keep in mind you can apply filters to already logged data so you don't need to go for a drive everytime you want to test a filter. just change the filter, commit the changes and load all data again.

You cannot filter on the averaged values. What the filters will do is look at every frame for the parameters the histogram and filter look at and throw out frames that don't meet the filter criteria. So the average display only averages the frames that meet the filter criteria.

Quote:

---

Originally Posted by my_bd

is there a way to filter the afr error% histogram in such a way that it only displays the lean values and alternatively only the rich values?

Meaning that I want to filter the resultant averaged values after all of the data has been loaded into the histogram....

I was playing around with adding some formulas to the filter referencing sensor values, but the results were undesirable because it appears to filter each individual sensor reading as it builds the histogram.

I only want to filter on the final averaged values.

---

learn how to use excel and you can do that..LOL

its a little trickier...but it can be done and I have done it..
takes using a pivot table and filtered data and a fiarly good amount of averageif and match and offset statements to make it happen....
probably not worth your time...and you could easily just do it by eye much quicker anyways..LOL

you would just be shooting yourself in the foot trying to do it that way
the point of filtering is to keep out the bad data so that you donthave a wierd value when its all averaged...

as an example
doing something like filtering out all AFR above 15.5
is a bad thing...what if you really do have a lean spot and then you wont see that because you have filtered out the end data...


its smarter to use the cell hits filtering as then you know you have enough data to be accurate..
try setting cell hits to something like 5 or 10 or more
you will find that the cells you are trying to filter out in the end...will probably be taken care of by the histogram cell hit filtering

if you filtering smartly it will be what you should use in the end anyways

Quote:

---

Originally Posted by my_bd

yeah I totally understand what you are saying and that's not what I was trying to do, I don't want to exclude the lean data during the scan, that would be just plain ol nutty.

I already use cell hit counts too.

I only want to filter or change the final displayed results in the histogram, of all data collected and averaged out.
I just like to break things down a bit.
I find it easier to examine and analyze where the lean spots are (temporarily remove the rich cell clutter), and then I can apply changes to my VE addressing the lean spots and review how it affects on my ve.
Then I do likewise with the rich cell data from the same histogram..

In the end I still apply both lean and rich cell data changes and then do some smoothing, nothing important gets left out.

---

personaly I still think you are crazy for trying to do what you just explained...

youe best method is to do them all at once...using AFR error or LTFT/STFT stuff
using "Multiply by %" or "Multiply by % - Half"
then go back and do a smoothing of it all by hand
or use some of the interpolation functions for a small amount of cells
no need to split it out for lean or rich...