hi David,
yeah, that's a  good thought - the signal wire for the temp guage shorting
to ground between the engine and guage.....
not likely I don't think........but that could do it all right.  Never saw
100 % defection though - more like 80 % max.
Thanks for suggesting it as a possiblity.

I thought I would just try another temp guage as an easy thing to try first.
heck, for that matter, I can just pop on another 84 instument cluster
quickly too.

scott

----- Original Message -----
From: "David Beierl" <dbeierl@attglobal.net>
To: "Scott Daniel - Turbovans" <scottdaniel@turbovans.com>
Cc: <vanagon@GERRY.VANAGON.COM>
Sent: Tuesday, July 22, 2008 2:27 PM
Subject: Re: coolant level control unit location

> At 04:56 PM 7/22/2008, Scott Daniel  - Turbovans wrote:
>>gas guage seems normal so far.
>
> Suggesting it's probably not the voltage regulator.
>
>>re "intermittent ground in the sender wiring, '
>>It's a 1.9 wbxr, 1984..........single contact guage temp sender. Ground
>>for the sender is  the sender  screwed into the engine.
>
> Yes, I mean the sender wiring shorting to ground between the sender and
> the gauge.  A full short would very rapidly peg the gauge and likely
> damage it eventually, but a partial short would give a high reading.  It
> could be in the sender I suppose, but I'd expect the sender to fail open,
> not short.
>
>>But I havn't even started investigating grounds, which are an obvious
>>place to start looking.
>>those push-on spade ground connectors above the fuse box, they just can't
>>work after 20+ years in my opinion.
>
> If they had used the correct brass terminals instead of steel they'd
> probably be fine.  But of course at some point you have to make contact
> with the steel body...
>
> The flasher circuit needs a ground at the gauge, but the gauge itself is
> only grounded through the sender.
>
>>I removed the level control unit from an 86......and turned on the key -
>>LED still flashes as normal.
>>the LED function is totally in the guage I would think.
>>and it can be triggered, or course, by an outside input, like the level
>>sensor system.
>
> That's correct.  The flasher has a three-second self-test at power-on, and
> is supposed to trigger any time the gauge receives a signal that would
> indicate overheat.  Once triggered it runs for three seconds regardless.
> The level controller simply simulates an overheat condition (i.e. low
> resistance to ground) when it detects low coolant.
>
> If you get no flash at power-on, the *strong* presumption is the flasher
> circuit in the gauge is toast, or the gauge ground is bad.  Again, this
> should not affect the actual gauge operation, only the flasher.
>
>>the key......and the guage needle seemed to have a mind of it's own.  I
>>didn't ohm-check the sender for the guage yet.............should do that,
>>but can't say I've ever especially seen a bad one for intermittlant
>>erratic temp needle deflections.
>
> A high gauge reading, assuming the +10v from the regulator is correct, is
> absolutely guaranteed to be a problem of too little resistance to ground
> somewhere, passing too much current through the gauge.  I would *expect*
> that an internal short in the gauge would make it read low rather than
> high, put perhaps not -- the gauge reads power, and P=I^2xR.  I'd need
> some real resistances to figure it for sure.  But to be safe, put in a
> different gauge before doing anything else.
>
> --
> David Beierl - Providence RI USA -- http://pws.prserv.net/synergy/Vanagon/
> '84 Westy "Dutiful Passage," '85 GL "Poor Relation"