Power Window Switches (again)

[Richard Thomsen]

Jim Jordan wrote last month:
> Hi,
> 
> Years ago telephone companies used relays exclusively for switching and they
> experienced the same problems we do with window switches.  The phone
> companies developed a contact cleaner that I've always wanted to use but
> have never seen one.  Their cleaner hit each pair of contacts in turn with a
> very short duration high voltage.  I don't know the values of the voltage or
> the duration or how to make one, but it seems that this would be preferable
> to other means if it could be done practically.  The guy who told me about
> this is gone on to the great Central Office in the sky.

	I sent this to a friend of mine who used to work for a telephone 
company to see if he had heard anything about it. This was his response.

		Richard Thomsen

    I think this is an apocryphal tale.   The relays employed in 
telephone work are always in service as part of the central office 
switching system. It would be totally impractical to build a parallel 
switching system alongside the actual phone system just to isolate 
relays and steppers from their normal connections so that high voltage 
pulses could be introduced....and this would be the only way it could be 
done.... high voltage pulses introduced into an operating telephone 
exchange would cause unacceptable amounts of impulse noise on the 
subscriber circuits.

    I have seen kits of tools for manually cleaning and adjusting 
telephone relays... and that's how it was done with the relay out of 
service on the bench.   Visual inspection of contacts, replacement of 
defective contact leaves, cleaning with abrasive tools and adjusting of 
contact tension and spacing with special gauges are a lot more reliable 
than "magic" electrical pulse machines.

    Probably, what was being talked about was the tendency of all 
mechanical contacts to "self clean" as long as there is a bit of current 
present. Current through the contacts sufficient to cause small sparks 
on make or break breaks down oxidation and exposes new metal.   This 
process keeps the contacts clean and making reliably.   There is a 
delicate balance between contact design, too much current (excessive 
wear and pitting) and too little current (no cleaning action).   The 
Bell System actually did research on how to make relays last longer and 
what the optimal contact design should be for each application.   This 
is where we got special contact alloys, bifurcated contacts, and 
mercury-wetted relays.   (And you thought Bell Labs only did glamorous 
stuff like invent the transistor!)

    I would say that the 48 volt central office voltage did the 
"cleaning". That and the fact that Bell Labs spent a lot of time and 
money researching how to make relays reliable, and ran their own 
manufacturing business (Western Electric) to ensure that their relays 
were built to extremely high and rigid standards accounts for their 
reliability in telephone service.

.... and now I've told you more than I know..... ;^)