Date: Thu, 24 Jun 1999 17:05:33 EDT
Reply-To: Ssittservl@AOL.COM
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: S Sittservl <Ssittservl@AOL.COM>
Subject: Water tank level sensor problems
Content-Type: text/plain; charset="us-ascii"
I need some advice in regard to the water level sensor for my
'88 Westfalia's water tank. It's not working - the panel light
for "full" is always lit. It worked fine when I got the van,
but stopped working after I gave the water tank a thorough cleaning.
(Apparently that's happened to other people, too.)
I'm pretty sure that the rod with the floating donut that sits in
the tank has gone bad, and I'm willing to replace it (about $50 from
Euro-American, if I remember). However, I'd like to ensure first
that (1) it for sure is bad, and (2) the LED panel ISN'T bad.
I've followed the discussions of the Westfalia water tank level
sensor over the last couple of months, and I've searched the archives,
and read the Bentley manual, but I still can't quite figure out for
sure how to test this. If I understand correctly, the LED panel is
really just an ohm meter, and the float rod is really just a variable
resistor. There are magnetically-activated switches in the rod at the
top and bottom of the float travel, and the rod can thus produce three
different resistances - "float at the top", "float at the bottom", and
"float anywhere between the top and the bottom". These correspond to
the green, red, and yellow LED's on the panel. (My panel has 3 LED's;
I understand some have 4, and the fourth is "so empty that we'll
disable the pump".)
To test the float rod, I figured I could use an ohm meter to, in
effect, simulate a working LED panel. Here's what I get:
Float at top: 0 ohms (no resistance)
Float at bottom: a very wobbly reading that (with the float just
sitting there, not moving) varies from about 320K ohms to
about 380K ohms. Seems to settle at about 380K ohms after a
while.
Float in middle: same as float at bottom.
OK, looks like at least my bottom switch is no good. (I don't hear it
click, either, like I do the top switch.)
To test the panel, I figure I can hook up a 1M ohm potentiometer
($1.49 at Radio Shack), and I should be able to light up the lights
by varying the resistance. I haven't tried this yet, though - does
this make sense as a test?
The one test I did try is shorting together the green and white wires
from the panel that normally go to the float - I hope that wasn't a
stupid, panel-destroying thing to do - and the green LED remained lit.
So, at infinite ohms (float rod disconnected) and 0 ohms (wires shorted),
and 320K to 380K (what my float rod produces for a less-than-full tank),
the panel says "full tank". That seems surprising to me - I would have
expected at least that 0 and infinite would mean different things - so
that's why I'm wondering if the panel might also be bad.
Does anyone know what the "proper" resistance values for the float
rod are, and what effect different resistances ought to have on the
LED panel? Any other suggestions for testing, or getting things working?
Please P-mail me with any suggestions or information, and I'll summarize
them back to the list later. Thanks.
-Steven Sittser
|