Mini-HOW TO: Setup your High/Low Octane AFR MAP

[Clipse3GT]

Just a small how to setup the EVO X AFR map. I know some people just randomly increase and decrease these numbers with out really knowing what the number means.

In the EVO, these numbers are just values, they do not correspond to real world AFR values from a wideband. Therefore, if you try to use this technique your air fuel map will be a mess.

Only setup you AFR maps, after your injectors are scaled and latencies are dialed to be within +/- 5%.

If you are using "Closed Loop - LT TRIM CONTROL" mapping in ECUFLASH, please redo your AFR's since most likely they changed, since now the LTFT's do not influence WOT conditions.


TOOLS:

Wideband with Serial/USB Output
Datalogging Software - i.e. EVOSCAN/COBB AP & COBB ATR
ECUFLASH/COBB ATR

LEGEND:

New Map AFR = NM-AFR
Desired AFR = D-AFR
Current Map AFR = M-AFR
Actual WB AFR = A-AFR

FORMULA:

NM-AFR = (D-AFR x M-AFR) / A-AFR


Example.

NMAFR=11.5 X 10 / 12.0 = 9.583

OR 9.6 will be put into your High Octane AFR map for a particular load/rpm.

Just drop this formula in excel after a WOT pull and it will tell you were to set your AFR. Then just slowly smooth out the fuel map again.

FORMULA in this EXAMPLE:

=(M2*J2)/L2

[LAX]

Thanks for the write-up! One issue I struggle with is how to get a smooth AFR when the map AFR is smooth, but actual AFR fluctuates. In your example, at higher rpm's, you map AFR is 10.3 the entire time, but your actual AFR jumps between 11.8-12.5. Based on the formula for NEW map afr, you would end up with very odd jumps in AFR from cell to cell. Is that desirable? Or is it just unavoidable that your actual AFR may jump around a half point or so even if map AFR is smooth.

Also, I feel that it's very hard to adjust any particular cell because there is also a bit of lag in the wideband o2 readings. I feel the best way to tune is to get some target AFR's at particular loads/rpm and then smooth from there. Thanks for any insight you can share!

[Clipse3GT]

Originally Posted by LAX

Thanks for the write-up! One issue I struggle with is how to get a smooth AFR when the map AFR is smooth, but actual AFR fluctuates. In your example, at higher rpm's, you map AFR is 10.3 the entire time, but your actual AFR jumps between 11.8-12.5. Based on the formula for NEW map afr, you would end up with very odd jumps in AFR from cell to cell. Is that desirable? Or is it just unavoidable that your actual AFR may jump around a half point or so even if map AFR is smooth.

Also, I feel that it's very hard to adjust any particular cell because there is also a bit of lag in the wideband o2 readings. I feel the best way to tune is to get some target AFR's at particular loads/rpm and then smooth from there. Thanks for any insight you can share!

That is a old log/map from when I tuning my car. It was for example purposes. Yes, the AFR changes constantly depending, on weather, boost pressure, baro pressure, etc... You can only try to get it close to the targets you want. But day to day they will vary a little bit. Like 0.2 AFR points. Also if you are hitting lean in a spot in the map, typically you have to adjust the cells around that value, not only that particular value to have a desirable outcome.

[LAX]

Thanks, I agree with what you're saying. The only part that I probably wasn't clear about is that even if map AFR is the same value (e.g. 10.3), the actual AFR can jump around a lot at varying RPM's.

E.g. from your sample log (note I see this also in my own logs)
5000 rpm = 12.1
5400 rpm = 12.7
5600 rpm = 12.2

Would you then go and adjust afr in the maps based on the new map afr formula? If so, it seems like it would cause pretty big jumps in your afr tables and it wouldn't really be smooth. Main thing is, I can't seem to get my car have a very stable afr (it can jump around 0.5 afr or so even when I want it to be a steady 11.0 afr for example). Is this something that I should even attempt to fix, or just leave it alone?

[Rogue]

Kris I was gonna ask why you look at AFRMAP.. so much reading to do.

[Clipse3GT]

BTT

[imdrax]

Pretty old thread but very useful information, i used your formula to create a nice spreadsheet, let me know how it looks or if I missed something important.