Sylvia Else wrote:
On 6/07/2011 2:46 PM, D wrote:
I am a watch collector & repairer. I also have a fair, though somewhat
dated, electronics background. I have several early '60's vintage Bulova
Accutron wris****ches. These watches operate using a very basic
oscillator circuit which energizes a tuning fork via a pair of coils.
The tuning fork then drives the mechanical movement of the timepiece. A
basic description can be found he
http://www.timezone.com/library/horo...72882451976629

These are very cool watches to the collector. They are stylistically
very much of their era, and because of the tuning forks, they hum,
rather than tick. The problem is, these watches were originally designed
to run on a 1.35 volt mercury oxide cell (343) which is no longer
available. There is a 1.55v silver cell which will fit, but it can cause
some Accutrons to run very fast, and possibly even damage them. At least
one supplier sells modified cells with a component to drop the cell's
voltage to 1.35 volts. I think this is accomplished with a surface mount
type diode. Here's a thread on these cells in a watch discussion, with
pictures of the component added
http://bdwf.net/forum/showthread.php?t=75397 I would like to simply
insert an appropriate diode into the circuit of the watches as a one
time modification, after which I can use the much cheaper/readily
available silver cells. But, as I said, my electronic component
knowledge is late 70's vintage, so I'm not sure what component to add. I
know in general silicon diodes have a forward voltage drop on .7V,
germaniums, .3V, so I'm not sure how to produce .2V. A zener? A
Schottky? Anyone have any suggestions on what would lower the output of
the 1.55 volt cells to 1.35 volts, over the life of the cell? It would
have to be fairly small, about 1/2 the size of a 1/4 watt resistor, or
less.

TIA

The forward voltage drop is more a rule of thumb approximation than a
physical reality. A diode's current versus voltage graph is a continuous
exponential curve, with the actual forward voltage drop being a function
of current.

I'd have thought these watches would draw a pretty much constant
current, knowing which would allow a resistor to be used.

Sylvia.




Might be interesting to know the startup current. Resistor may run it,
but starting may...or may not...be an issue. Don't forget to check it
over temperature. Yes, the arm is a quite stable environment, but
some people take their watch off occasionally.

For fun I stuck an old HP 5082 Schottky diode on the curve tracer.
200mV at 10uA.
I'd look at the spec sheets for modern SMT diodes and find one that
has the right voltage at the required current.