RF conductor materials (fwd)

---------- Forwarded message ----------
Date: Wed, 29 Apr 1998 20:45:20 -0600
From: terryf-at-verinet-dot-com
To: tesla-at-pupman-dot-com
Subject: RF conductor materials

Hi Mike,
        I was so super busy at the time you originally sent this that I
didn't give a good response.  Sorry about that. :-(

Aluminum and copper are not used in high-power high-Q RF circuits due to the
problem of oxidation.  Bare aluminum and copper almost instantly begin
oxidize in air.  Over time this layer can become very thick.  Aluminum oxide
is an excellent insulator and is used as such in the common form of alumina.
Copper oxide easily forms with similar results.  Since RF currents travel on
the outside surface of conductors, this can have a very significant effect
over time as the oxide layer builds up.  Aluminum house wiring used years
ago keeps fire departments busy today :-(( 

There are four common solutions used in the industry:  

Coated copper - like magnet wire.  The coating protects against oxidation.
In high power RF circuits the coating may not be reliable however.   

Brass - does not oxidize badly and is a good cheap alternative.  Over time
it will give much better performance than copper.  Used extensively in the

Silver coated copper - At first this sounds like it would have the same
problem as aluminum and copper.  So why is it used as a coating on so many
high power RF components?  Nope!  Not because it is a better conductor.
That really has nothing to do with it.  Silver oxidizes to silver sulfide a
conductor (a neat trick! :-)).  Not a good conductor, but much better than
brass if you need the low losses.  At my company we pump 100 amps of
continuous RF through conductors of silver coated copper with good results.
We cheat a little by pumping cooling water through the conductors to keep
them cool too :-))
Anything else will melt down very quickly except...

Gold coated copper - Excellent conductor of heat and RF and immune to
oxidation.  Often used on RF power transistors and other "high value" places
where maximum performance is needed.  Cost is the killer for most normal

        In our relatively low frequency uses, copper is generally very good
(although my recent testing raises many questions about what frequencies we
really need to design for).  Litz wire is good at our frequencies but
expensive.  (We get custom Litz cables over 1 inch thick for 300KHz at
100KW, the cost is stunning!)  Aluminum should be avoided since the oxides
are so aggressive and thick.  Of course, everybody uses it all the
time.....just because :-))


If it looks simple, look closer!

At 10:38 PM 4/15/98 -0500, you wrote:
>From:  Hollmike [SMTP:Hollmike-at-aol-dot-com]
>Sent:  Wednesday, April 15, 1998 7:29 PM
>To:  tesla-at-pupman-dot-com
>Subject:  Re: 40MHz Spark Gap Behavior
>    I was just reading the specs on your cap and have one question:  How come
>you chose brass to use as the "bus" for the doorknobs.  Aluminum is less
>resistive than brass, and copper is far superior to either.   It may not be
>that significant, but it could lower the Q of your primary system.  
>Just Curious,
>Mike Hollingsworth