 Forum » Understanding AFR » AFR formula understanding

# AFR formula understanding

### Tech Articles

Discussion and questions related to the course Understanding AFR

Page 1

could someone please explain the have over want formula in respect to adding and subtracting fuel

i havent practiced this for over 2 year since doing the efi fundamental now i have access to a dyno 24/7 and i am practicing just need a little help thanks .it my be simple but i just need clarification

You simply take the measured air fuel ratio as shown by your wideband. Let's say this is 12.5:1. If however your actual AFR target at this point is 13.0:1 you can calculate a correction factor to get to your target. Divide 12.5 by 13.0 and you get 0.96. Multiply the value in your fuel table at that particular pint by 0.96 and you will find your AFR now reads on target.

thank you

thank you

A quick point for a couple of specific application, namely reflashing Evos and Subarus, maybe the same for other OEM applictions also. The fuel tables are displayed in AFR. Most tuners ignore the actual values in the table when tuning the cars, i.e. they don't try to make the actual afr equal to the afr in the table (it never corresponds on a stock car anyway). If you are going to tune it this way the formula needs to be flipped, i.e. the correction factor needs to be want afr/have afr as larger numbers in the table equate to leaner mixtures and smaller numbers are richer mixtures.

(This is the quick way of tuning these cars. The right way is to re-scale the MAF so the AFR table matches the actual AFR and then you can just punch in the AFR values you want into the table. When rescaling the MAF, the formula would be have/want, i.e. smaller numbers = leaner, larger = richer).

You're correct HS and this is covered in detail in the Practical Reflashing course. Provided you understand the concept (ie to richen the air fuel ratio you need to ask for a smaller AFR target) you can manipulate the formula to suit.

(first sorry for english level) (Tune with ecutek)

I actually have a problem with the understanding of AFR/fuel map for subaru. When I put a widebande in the exhaust of one subaru, the real AFR is not the same as in the afr/fuel map target. And I don't understand how the subaru ECU could show us (in log) an AFR/fuel map in AFR with a standard lambda.

So if I understand the explaination here, we need to rescale the MAF until the real AFR and AFR / fuel map target are equal. So it is a decoy for ecu? And once my real afr (see on my widebande) are equal with afr/fuel map, I could directly choose the afr i would like in afr/fuel map, is it the right way to tune injection for subaru?

Does this foruma also apply for the VE Table ? ...example I am street tuning my engine and my actual was .98 but my target is .91 can I divide that and enter that in my VE Table ?

Thank you

As Tony asked, does this formula approach only work with ms pulse based fuel maps and AFR represented maps or would it also work on VE based maps like Link's traditional fuel equation in conjunction with the open loop target table?

I've had mixed results trying to apply it to the latter (possibly depending the Master Fuel value or maybe I just didn't have dead time values spot on for the injectors).

It depends on the accuracy of all your sensor calibrations, your injector calibrations and whether or not you fuel pressure differential is measured or estimated/calculated

If all the above are accurate you can apply the changes to a VE table in some ECUs.

In Tony's example, he has a target was .91 and his actual is .98.

Provided there is no transient fuel at play at that time, and you take into consideration any closed loop fuelling that is at play at that point in time you can calculate it.

So Actual LA was .98 = 14.406AFR

Target LA was .91 = 13.377AFR

Provided no fuel trims :

Say his VE at that point was 72.8%, 72.8 * (14.406 / 13.377) = 78.39 = 78.4% VE

You could make the calculated change to your table, and see how it responds. You may find it tracks perfectly, or there is a margin of error. If you had an active close loop fuel trim, this work a little differently:

So Actual LA was .98 = 14.406AFR

Target LA was .91 = 13.377AFR

Say there was a +5% fuel trim, we would say our actual lambda at that point may would be 1.03LA = 15.141.

Say his VE at that point was 72.8%, 72.8 * (15.141 / 13.377) = 82.40 = 82.4% VE

So in other words, all else being equal, yes it should work.