Bob :
I still don't know of a good test, but maybe I can clarify how it works a
little.
The 'electromagnet' you found is actually a heater, the wire is resistance
wire so running current through it produces heat. How much heat depends on
how much current, which in turn depends on the voltage. The contact it is
wound around is what's called a "bimetal strip", which is formed of two
dissimilar metals such that it will bend (and open the contacts) when it
gets hot.
So in operation, when the contacts are closed, the ignition circuit voltage
(what we usually call 12 volts, but can actually be anywhere from 11v to
15v) is applied to the heater. The heater warms the bimetal strip until the
contacts open, then cools off until the contacts close. The resultant
output is a voltage that swings from 12v to 0v and back again, with the
average being around 10v. (Some people have reported that the 10v average
is also dependant on ambient temperature. Makes sense, but I haven't tried
to measure it to be sure.) It should be relatively independent of how many
gauges are connected, or what they are reading.
The gauges that run on the 'stabilized' voltage are also 'hot wire'
movements, with a heater acting on a bimetal strip that moves the needle.
They have a very slow response time, so will filter out the effects of
gasoline sloshing, etc. as well as the 'chopped' voltage coming from the
voltage stabilizer. IMO the whole arrangement was just to be cheap, I much
prefer balanced gauge movements like my TR3 and Chevy have.
The best explanation I can think of for your problem is that the contacts
inside the stabilizer are burnt or dirty, so that they will make contact
once, but not repeatedly. I've found this to be common failure mode for
cheap relays. Your 'worn through wire' explanation would work too, since a
short to ground will cause the heater to draw too much current, and possibly
chop the voltage so low that the gauges wouldn't work at all. Another
possibility would be the bimetal strip being bent, thereby changing the
calibration.
Substitution of a known good unit is the usual way to troubleshoot such
things.
Randall
Bob Kramer wrote :
>
> No responses yet on the voltage stabilizer question I asked.
> I was looking
> for a simple way to determine if one is bad, or what else
> might cause the
> stabilizer to work when separated from the speedometer (not
> grounded) but
> fail when properly connected. I opened up one yesterday, and
> found what I
> think might be the problem. I think all 3 that I have are bad. The
> electrical engineers on this list can explain it better, but
> it looks like
> the unit basically has 2 sets of contacts that are connected
> only when the
> power is turned on. A wrap of the ground wire (connected via
> a tab to the VS
> case) around one of the contacts must form an electromagnet
> holding the
> contacts together. My guess is that this is done to prevent
> unwanted short
> term voltage disconnections when the sender drops its signal (sloshing
> gasoline?) to keep the gauge from registering bouncy
> readings. Therefore, I
> have determined that the ground coil must be bad, burned
> through the wire
> coating, causing a short to the case, and I need a new
> voltage stabilizer.
> Can anyone enlighten me as to wether my thinking on this is correct?
|