On Tue, 13 Aug 2013 11:29:46 -0400, "Michael A. Terrell"
wrote:


Jeff Liebermann wrote:

On Tue, 13 Aug 2013 10:18:34 +1000, "Phil Allison"
wrote:

** Silver does not form an oxide under any normal condition.

If forms a sulphide layer (Ag2S ) due to air containing some H2S from
burning fossil fuels etc.

Yep. Even worse, the silver sulfide to silver junction makes a
tolerable diode (band gap = 1 eV). I was having serious problems at a
radio site with what is now known as PIM (passive intermodulation). I
had stupidly cleaned the silver plated plugs, but not the jacks. After
a week of bad guesses, I cleaned both connectors, and the problems
went away.

I had a Bird Wattmeter on the bench at one job. (QA on the PRC77) The
thing read about 30% difference from one direction to the other because
it was silver to silver in one direction, and silver to crap the other.
The Cal lab threw a real hissy fit when they heard that I cleaned it
with a piece of paper.

About 20 years ago, I was involved in a wattmeter calibration party.
Mostly, we were trying to product frequency versus indicated power
graphs. Should be simple enough. Right.

Because of erratic readings, it was decided that everything needed to
have the contacts cleaned and tarnish removed. Someone dug out some
silver polish, which made the slugs and couplers look nice and shiny,
but also coated them with a layer of wax. The readings stabilized
only after everything was wiped with alcohol patches to remove the
wax.

After that, we found that many of the slugs produced seriously
inaccurate readings. Dig out the Model 43 data sheet and it says
+/- 5% accuracy. Ok, 5% of what? Calling Bird support determined
that it was 5% of full scale. Therefore, if I use a 100 watt slug to
measure a 25 watt transmitter, and 5% of 100 watts = 5 watts, 25 watts
indicated could be anywhere between 20 and 30 watts. So, we ran the
graphs mostly at full scale with a few spot checks 1/2 and 1/4th of
full scale. A Telewave 44A (no slugs required) turned out to be quite
good (specified accuracy +/- 6% of FS) above about 200 Mhz, but
required using the included calibration chart for lower frequencies.
Also, my collection of battered Radio Shack VHF/UHF VSWR meters were
amazingly accurate (in the ham bands).

There was some discussion that a light bulb dummy load and light meter
was more accurate, but we didn't have time to verify the claim.



--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558