Date: Sun, 20 Jan 2002 09:11:48 -0700
Reply-To: Jeff and Tamara Nelson <ramjeff@EARTHLINK.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: Jeff and Tamara Nelson <ramjeff@EARTHLINK.NET>
Subject: Re: AFM upgrade using 16 volt tantalum capacitor
Content-Type: text/plain; charset=iso-8859-2
I think some of the confusion may be because there are 2 or more
problems with the AFM. Each requires a different fix.
1) The resistive track that the wiper runs on can get worn or dirty.
This causes bad signal levels to be sent to the ECU at certain spots.
2) This is just conjecture, but from what I've heard, there is the
possiblity that static can build up in the air intake and then discharge
through the AFM. This could cause a sharp spike in the signal to the
ECU.
In either case, I think our ECU is a fairly simple device. It just
takes all the inputs (AFM, rpm, T2, O2, etc) and generates an output
(fuel amount). As far as I can tell, there is no checking of the input
signals to see if they are "realistic". I would guess that almost all
or all more modern sophisticated control systems perform a check on the
inputs to help identify fault sensor readings. For example, say the RPM
signal changes from 2000 rpm to 4000 in one crank revolution. That is
physically impossible due to inertia etc. I would guess our 20 year old
digi system would just spit out a fuel amount for each rpm without
hesitation. A more modern controller would determine that that change
is impossible and would do something more intelligent. The options
include: use the previous value (ignore 1 or more input readings),
interpolate between the 2, only change the output signal by a maximum
amount each time, etc.
By putting the cap on the AFM output, I believe the result is that the
AFM output signal is slew rate limited. I.e. its rate of change is
limited so that its output signal can't change any quicker than the
engine load can physically change. A cap will smooth out the output
signal and prevent sudden large changes in signal value. If you have a
static pulse or cross a worn spot on the track, the cap will make the
output signal change more smoothly and slowly vs. the (impossible)
instantaneous change that can occur otherwise.
So, in my mind, if you have worn out AFM track, you need to get a new
AFM or move the board so the wiper runs on a new track. Independent of
that, I think the cap is a good idea that can help prevent your engine
from being "jerked around" by faulty input signals to the ECU.
-jeff
Larry Alofs wrote:
>
> Someone ought to reverse-engineer one of these things to see what the VW
> solution really consists of. The service bulletin says something about
> "active components" (transistors).
>
|