Date:         Sat, 4 Nov 2006 13:08:04 -0800
Reply-To:     jon <jon@KENNEKE.COM>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         jon <jon@KENNEKE.COM>
Subject:      Re: AFM Voltage Ratios
Comments: To: Geza Polony <gezapolony@SBCGLOBAL.NET>
In-Reply-To:  <vanagon%2006110415003866@GERRY.VANAGON.COM>
Content-Type: TEXT/PLAIN; charset=US-ASCII

The mathmatical importance of an AFM (or TPS/MAP) is somewhat easily
understood by reading the writeup on www.megasquirt.info.

In the incomplete understanding of the fueling system that I had
yesterday, I thought you could replace the AFM with a MAP (and some
scaling electronics). But, the real answer is that you could replace the
AFM with a MAP *and* a TPS. Of course, you would still need an IAT, and
the one in the AFM seems to last forever.

The MAP/TPS route would require a microcontroller that would translate
that data into something that would simulate the AFM. At that point, you
might as well swap out the ECU for a megasquirt.

My whole mission was to find an easy replacement for the AFM. Mission not
accomplished.

Jon

On Sat, 4 Nov 2006, Geza Polony wrote:

> True, the AFM doesn't produce voltage. The question is really, what signal,
> measured in millivolts, does it return to the ECU?
>
> This is important because the signal returned to the ECU, along with other
> data, tells the ECU what air/fuel mixture to send into the cylinders.
>
> It seems to me that for any given set of conditions (say, trying to
> accelerate  up a 3% grade at 3100 RPM in fourth gear) the signal returned to
> the ECU should be constant.
>
> Otherwise, the engine will be running too rich or too lean.
>
> If the millivolt signal didn't really matter, it seems the VW engineers
> could have designed a device much less complex than the AFM, or eliminated
> it altogether.
>