Date: Fri, 9 Nov 2001 11:47:02 -0500
Reply-To: David Beierl <dbeierl@ATTGLOBAL.NET>
Sender: Vanagon Mailing List <vanagon@gerry.vanagon.com>
From: David Beierl <dbeierl@ATTGLOBAL.NET>
Subject: Re: coolent level sensor
In-Reply-To: <2FAC6A580E76544C842BC944943D1B0C02638C84@xch-sw-03.ds.boei
ng.com>
Content-Type: text/plain; charset="us-ascii"; format=flowed
At 09:11 AM 11/9/2001, Greenamyer, William L wrote:
>there an interface/signal conditioning device in the instrument cluster that
>processes the low level signals from this sensor (basically measuring
>conductivity of the coolent) which then feeds the flasher circuitry?
'Zackly, except it's not in the cluster. It's a "relay" with #43
stencilled on it, and in the '84s it's mounted on the wall above and
forward of the fuse panel -- very hard to get at with the dash in. You may
be able to reach it by pulling the cluster, or by dropping the fuse panel
and coming from below. Make sure you get the right one, there are two there...
As you surmise, the "sender" is simply a pair of stainless-steel pins --
one of them is grounded and the controller puts a very high-impedance 100
Hz square wave on the other one. If there is coolant present the signal is
grounded through the coolant to the other pin; this triggers the controller
output. In fact any constant signal that inhibits the square wave will
cause triggering.
This output is connected to the gauge in parallel with the gauge sender --
it's an "open collector" output and floats with the temp signal until the
controller triggers. If the controller module is tall, it simply drags the
line down to a couple volts or so, which pins the gauge and makes the light
flash just as though the system had overheated. If the controller is
roughly cubical, it periodically spikes the line which triggers the light
without noticeably affecting the temperature reading.
The actual flasher circuit is in the gauge itself -- it starts flashing any
time the input signal corresponds to a full-scale reading, and continues
for a few seconds on its own after that. It also flashes for a few seconds
when power is first applied, so if it's not doing that you have a gauge
problem.
Note when you're testing: actually grounding the temp gauge input gives a
reading far off-scale -- you can do it for a few seconds but longer will
overheat the gauge, making a bad smell. I don't think the gauge will
actually fail if run this way for a minute or two (long enough to notice
the smell) but it will probably change the calibration.
Your problem isn't common. It's far more common for the flasher circuit in
the gauge to simply flash continuously, due to leakage in an electrolytic
timing cap inside the gauge.
You know it isn't an open in the input line, since that would trigger the
system. The line could be shorted, though, either high or low. I forget
exactly, but I think that even 100 K-ohms would be sufficient to inhibit
the trigger. Or it could be an open between the controller and the gauge
input line. If the gauge itself is working correctly that lets out the
panel wiring except possibly the ground connection on the back of the gauge
which is needed for the flasher to function -- but if the flasher test
works then that's ok too. Aside from that is the controller itself.
david
David Beierl - Providence, RI
http://pws.prserv.net/synergy/Vanagon/
'84 Westy "Dutiful Passage"
'85 GL "Poor Relation"
