[TowerTalk] Sudden shifts in shunt matching network?

[Red]

Hello to Bill and All;

I'll share a professional experience from the 1960s, in which tuned 
modules used in a system that the firm I worked for produced exhibited 
erratic frequency shifts.  I found that either small temperature changes 
or slight mechanical shocks applied to silvered mica capacitors in the 
circuits would cause the shifts.  With the aid of the manufacturer of 
the capacitors, it was confirmed that individual segments of the 
capacitors would lose their electrical contact with the stack of 
segments making up the capacitors.  As I recall, these capacitors were 
constructed of stacks of four segments, and the frequency shifts we 
observed were accounted for by the loss of 1/4 of the total capacity.  
When advised of that observation, the manufacturer quickly identified a 
manufacturing step that was out of control.  The manufacturer corrected 
the process and the problem was solved.  The percentage of defective 
capacitors was small and we discovered the problem in thermal cycling 
and mechanical stimulii used to screen potentially defective assemblies 
for a high reliability application.

73 de WOØW

Bill Coleman wrote:

>I use a shunt-fed 15m high tower on 80m and 160m. The matching  
>network for 160m consists of an L-network. The series inductor is  
>tapped about 45 turns on two stacked T200-2 cores. Probably about 70  
>uH max. Shunt C is on the rig side, consists of a 560 pf 1kV silver  
>mica and an 150 pf air variable. This network allowed me to tune for  
>a 1:1 match at 1825 kHz. I use an automatic antenna tuner in the  
>shack to bend the match to other frequencies in the band. I only run  
>100 watts
>
>During the ARRL 160m the first night, I was calling CQ for about 5  
>minutes around 1820 kHz. Suddenly, the tuner showed a mis-match.  
>Hitting the tune button couldn't restore the match.
>
>Whipping out the antenna analyzer, I found that the match frequency  
>had jumped up to 1875 kHz. I eventually moved up into that part of  
>the band, and had a lot of fun for about 3 hours calling CQ around  
>1870 kHz.
>
>Second night, I found a pretty good match at around 1825 kHz. Called  
>CQ for about an hour when the SWR jumped out again. This time, the  
>antenna analyzer showed a resonant frequency of 1980 kHz. I went  
>outside in the cold and attempted to adjust the matching network. I  
>had to add inductance, but I ran out of capacitance. Couldn't move  
>the match further down than about 1935 kHz. I had to QRT early that  
>night.
>
>I'm trying to figure out what might have caused these sudden shifts  
>in the match. When I touched the components of the network the second  
>night, none of them were hot, so I don't think there were excessive  
>losses involved.
>
>I'm wondering if either component might have been affected by the  
>constant transmitting of calling CQ. Could the circulating currents  
>be too high for the Silver Mica and caused it to lose some of its value?
>
>Does anyone have any suggestions for solving this mystery?  
>Suggestions for beefing up the matching network? What kind of fixed  
>capacitors should I be using in this circuit?
>
>I have really been impressed with the capabilities of the shunt-fed  
>tower. I was able to call CQ and run about 70-100/hr during the  
>contest. I worked 44 states, 5 provinces, and a couple of DX  
>stations, including the Czech Republic, in just a few hours of  
>operating.
>
>Bill Coleman, AA4LR, PP-ASEL        Mail: aa4lr at arrl.net
>Quote: "Not within a thousand years will man ever fly!"
>             -- Wilbur Wright, 1901
>
>_______________________________________________
>
>See: http://www.mscomputer.com  for "Self Supporting Towers", "Wireless Weather Stations", and lot's more.  Call Toll Free, 1-800-333-9041 with any questions and ask for Sherman, W2FLA.
>
>_______________________________________________
>TowerTalk mailing list
>TowerTalk at contesting.com
>http://lists.contesting.com/mailman/listinfo/towertalk
>
>
>  
>