Date: Sat, 28 Jul 2001 10:43:48 -0400
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: Re: Coolant resistance thru coolant level probe
In-Reply-To: <3B62BAAB.B27ED23C@enteract.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 09:14 AM 7/28/2001, Larry Alofs wrote:
>Right. Although I hope that some other people will try to check the trigger
>point because my van has been flakey in the past and it is very possible that
>this value is not typical.
Should be about right -- you're swamping a signal supplied through 100
k-ohms, in the old modules at least.
> > Why then some people had
> > it blinking with tanks full of coolant?
>
>That is the main question. In the past mine has blinked for weeks at a time
>with the tank completely full of a reasonable mixture, and then healed itself
>again. David's statement about a marginal capacitor on the circuit in the
>dash
>being affected by humidity seems to make sense. I guess I should go into that
>or learn to ignore the flashing or replace level sensor part with a different
>circuit.
You can easily test that -- unplug and jumper across the sender leads. If
it doesn't blink, could be dirty probes. If it *does* blink, bad gauge or
open wiring or bad control module.
> >
> > Perhaps the sensor is capacitive? Being more sensitive than resistance?
> Just
> > guessing.
>
>I don't think it could be capacitive. The capacitive between probes that
>small
>and that far apart even with liquid between them would be so small that it
>would
>be comparable to the capacitance between the two wires leading to the probe or
>between the wires and any nearby metal. The change in resistance would be
>much
>easier to measure.
The active terminal of the probe will trigger the flasher if it gets pulled
either high or low -- the circuit is a multivibrator (makes a square wave
at ?100? Hz -- I forget) coupled through 100 kohms and a pair of 1000 pF
caps to an arrangement that grounds the gauge terminal (thus simulating an
overtemp and triggering the light) if the square wave stops. The sender
terminal is between the two 1000 pF caps and will swamp the square wave if
it's tied to *any* fixed voltage of low enough impedance.
The other terminal is ground, thus supplying a fixed voltage to tie the
active terminal to.
That's based on the older discrete-components control module; I haven't
checked out the other one yet. It has an 8-pin DIP (custom VW) and not
much else, and triggers the light without visibly affecting the gauge -- I
presume that it pulses the gauge line instead of giving it a steady
ground. I imagine its detector works pretty much the same way as the other
one.
david
David Beierl - Providence, RI
http://pws.prserv.net/synergy/Vanagon/
'84 Westy "Dutiful Passage"
'85 GL "Poor Relation"