Cold-switch relay voltage ratings

[Bruce Toback]

The voltage and current ratings of relays are normally given with
the assumption that the relay will be called upon to switch a circuit
at that voltage while the rated current is flowing. Obviously, if a
relay is switched "cold" it can handle higher voltages and currents.

Can someone give me an idea of how to calculate the maximum voltage
rating of a relay that is cold-switched? It should be a function of the
contact separation distance, assuming that the rest of the insulation
on the relay is adequate to handle the voltage. I'm looking for a way
to electronically bandswitch a high-power amplifier, and I'd rather not
spend lots of money on vacuum relays if I can avoid it. Are mercury-
displacement relays good for this kind of service? (How do you calculate
the inductance of a pool of mercury, anyway?!)

Another related question: Does anyone have ideas for a safety interlock
that would discharge the HV power supply capacitor faster than the
bleeder alone (without causing an explosion!)? The time constant for
a 35 uF capacitor and a 200K bleeder resistor is such that lethal
voltages remain in the HV circuit for about 30 seconds after power is
removed. I'd like to shorten this to a second or two if possible.

Thanks,
-- Bruce Toback
KN6MN

[Brian]

bto...@netcom.com (Bruce Toback) writes:

> Can someone give me an idea of how to calculate the maximum voltage
> rating of a relay that is cold-switched? It should be a function of the
> contact separation distance, assuming that the rest of the insulation
> on the relay is adequate to handle the voltage. I'm looking for a way

Are you familar with CCInk: The Computer Applications Journal (and the
associated BBS)? This discussion was hashed out a few months back.
E-mail me if you need contact info.

---
Brian Mork UUCP bm...@opus-ovh.spk.wa.us / ARO ka9snf@ka7fvv.#ewa.wa.usa
.... .... USMail 6006-B Eaker, Fairchild, WA 99011
.. .. .. .V:509-244-3764 D:509-244-9260