found this in the hardcore vw forum. didnt realize there was such a relation between the actual A/F and the one you want.



I have a map that is not perfectly tuned that I want to refine. I have a wideband that I can datalog so I can relate specific cells to specific a/f ratios and specific pulse widths. Let us say, theoretically, that load point x has an air/fuel ratio of 14:1 air/fuel at an injector pulse width of 10 milliseconds and I want to change it to 13:1 air/fuel

If I convert air/fuel to fuel air I get:

14:1 a/f = 0.071 f/a
13:1 a/f=0.077 f/a

To calculate the change in pulsewidth needed to get to my desired 13:1 air/fuel I would:

Divide target f/a by actual f/a:
.077/.071 = 1.08

108% difference between my target values and my actual values.

My thinking is that all I need to do is multiply my current pulse width of 10 milliseconds by 1.08 to get the target pulse width that will result in my desired a/f ratio of 13:1 at that load point.

Of course this assumes that there is a linear relationship between pulse width and air/fuel ratio. I guess that is the big question.

Basically, I am looking for the lazy way to calculate pulse widths without having to go through that huge formula that Ben posted a while back.

If this is true it means that I can do a couple of datalogged runs and then make the changes from the comfort of my dinner table. It also means that is would be possible to tune the car without having to rope someone else to ride along with me.


after this was a bunch of posts basically saying that yeah, this works

would be cool for people that have to burn new chips (viet) so that you can make a bunch of changes as once.