[Dean]

I'm working on intake calibration and have read a couple threads and the HSPN article and have got much of it working for me.

The formula in the article doesn't work for me. It just averages the entire range, not the matching lines only. Any clues? I've been selecting ranges by hand instead which is a pain with 20 minutes of data...

WolfPlayer also talks about picking good data. Rather than use AVERAGE(range) and hand picking data, try using TRIMMEAN(range;trim) which will eliminate extreme readings. The trim is a decimal percentage. So to dump the high and low 25%, use TrimMean(range;.25). Try different values. Oh, the ';' is a ',' in Excel. I'm using OpenOffice.

[Dean]

DOH... When using Array functions, you have to hit Ctrl-Shift-Enter for them to work. When they grabbed the screen shot, they must have not done so. An array function when correctly entered has braces around it. Mine looks like this: {=TrimMean(if((MAF value column range=MAF Voltage); A/Fsum column range);.25)}

[R4ND0M_AX3]

I used averageifs to look for data that only applies to the maf voltage and is beetween a two cell values I have to chop out the noise. I also use countif to tell me how many samples apply. I use this as a confidence indicator.

Then I use the new 'heat map' styles of conditional formating in excel 2007 to give make the larger errors stand out.

[R4ND0M_AX3]

Here's an example of a problem I was working on.
2-2.5 volts read way different than everything else.
I have one of the cells highlighted so you can see the formula.
Click to see a larger version.

[jaxscuby]

when i was using ecutek to calibrate the maf I would
use delta dash to read the st & lt values. I would try to
achieve a total of 7~10%, the factory value range various
from 7~14%.

so after you observe the st & lt values, you would make adjustments
either increase or decrease by a percentage.

repeat until you get your target st & lt values.

but please remember that you should change the entire
maf table and not just the idle rpm

[Dean]

Oh, more formulas to dig into. Thanks. Care to share the whole spreadsheet?

Did you resolve the issue?

It is tough to tell without seeing your actual table curve, but just eyeballing the deltas, it looks like smoothing from as you say 2 or a little later all the way to 3 should clean it up.

But what do I know. I'm new to this stuff.

[R4ND0M_AX3]

Originally Posted by Dean

Oh, more formulas to dig into. Thanks. Care to share the whole spreadsheet?

Did you resolve the issue?

It is tough to tell without seeing your actual table curve, but just eyeballing the deltas, it looks like smoothing from as you say 2 or a little later all the way to 3 should clean it up.

But what do I know. I'm new to this stuff.

I've got it closer. I need to tweek the tip-in table now.
I didn't even think to look at the data this way until I was driving around with ST running one day and say some weird corrections going on.
The sheet is in a new format. that function 'averageifs' is also new.