[R-390] Tube heat and glass to metal seals

Sun, 21 Mar 2004 14:43:02 -0600

Yes, the metal and glass have the same coefficients of thermal
expansion, but they don't have the same coefficients of thermal
conduction. The metal is not necessarily at the same temperature
as the glass. Consider the heatsinking property of the socket and
its wires.

The largest difference probably happens at startup.

Steel chemical reactors are sometimes lined with glass to prevent
rapid corrosion of the steel. The glass must not crack. Temperature
control schemes for an external heat exchanger jacket must not
allow the difference between the jacket and the contents to be
more than about 30-50 degrees C.

Bill Hawkins

-----Original Message-----
From: [email protected] [mailto:[email protected]]On
Behalf Of Chuck Ochs
Sent: Sunday, March 21, 2004 1:53 PM
To: Kenneth G. Gordon; [email protected]
Subject: Re: [R-390] Tube heat and glass to metal seals

Yes, this is true. How else can one explain how vacuum tubes that were made
80 years ago still function perfectly.
This is an obvious problem in the  design of vacuum tubes, and was worked
out a long time ago, most likely by the people at Western Electric, but
maybe much earlier. After all, the mechanical characteristics of vacuum
tubes are based on the design for the lightbulb. Suspects like Tom Edison
come to mind.
Chuck N1LNH
----- Original Message -----
From: "Kenneth G. Gordon" <[email protected]>
To: <[email protected]>
Sent: Sunday, March 21, 2004 2:00 PM
Subject: [R-390] Tube heat and glass to metal seals

> I was under the (properly informed) impression that the metal used for
tube
> glass to metal seals was specifically chosen because its coefficient of
> expansion was identical to that of glass, so that leakage from this source
> was essentially non-existent.
>
> I have forgotten the trade name of this metal.
>
> Comments from the peanut gallery?
>
> Ken W7EKB