Date:         Sat, 15 Jan 2000 19:54:11 PST
Reply-To:     Lance LeBaron <lebarle@HOTMAIL.COM>
Sender:       Vanagon Mailing List <vanagon@gerry.vanagon.com>
From:         Lance LeBaron <lebarle@HOTMAIL.COM>
Subject:      engine management theory
Content-Type: text/plain; format=flowed

Ken:

I agree the vanagon application is more difficult than the airplane because
speed and load are constantly changing as you stated.

I wonder if The O2 sensor fake out box would work if it added or subtracted
a voltage from the O2 voltage instead of putting out a constant voltage.
The ECU would strive to maintain 0.7 volts when the O2 sensor is putting out
something other than 0.7 volts.  This would force the engine to lean or
rich.

I am not positive this would work.  The O2 voltage swings back and forth
when the ECU is running correctly.  Is there any difference swinging around
0.7 volts compared to 0.5 volts or 0.9 volts?  The voltage swings in either
case.  I recall the output of the O2 sensor described as having a rapidly
changing output right around ideal fuel mixture and flatter as you move
away.  I seemed like the ECU is more interested in the fact that the O2
voltage is swinging than in what the actual value is.

Of course we would also need some input to monitor so we can tell whether we
want to run richer or leaner.  David Marshall suggested a CO meter.  The
airplane used exhaust gas temperature.  Would we monitor it all the time or
only at full load?  Won't work at full load because the ECU ignores the O2
sensor at full load (correct?)

Ken wrote:

>Date:    Sat, 15 Jan 2000 09:08:49 EST
>From:    KENWILFY@AOL.COM
>Subject: Re: Engine management (manual control of mixture)
>
>I don't know if this would work well unless you are cruising with cruise
>control set at one particular rpm on a straight, flat road.  Isn't the
>computer changing the mixture to get the best possible mixture for
>accelleration, and deceleration as well. (etc)
______________________________________________________
Get Your Private, Free Email at http://www.hotmail.com