[Milsurplus] Milsurplus Digest, Vol 160, Issue 25 Capacitor losses

[Kmec at aol.com]

Hi Ray & Mike (et.al.)
 
Mike: Thanks for the interesting and flattering words about RF engineers as 
 "I are one".....
 
Ray: Why any capacitor gets hot has to do with dielectric losses in the  
material used. And this varies greatly. A simple test of a ceramic or plastic  
quality as a high frequency dielectric is to put a chunk of it into a 
microwave  oven and see if it gets hot. If it does, it is lousy at 2.45 GHz and 
so may be  quite lossy at HF even!
 
Material science for RF stuff is key: Even though the circuits look very  
simple and should be easily computed and work as predicted, the materials can 
 kill you. Plastics, ceramics, metals, solders joints, connections with 
bolts,  dissimilar metal contacts and so on can kill you. One of the reasons 
good  quality RF caps are expensive: the materials used to make them "good" 
aint  cheap.
 
Skin effect, restricting current to thin layers is an reason why silver  
plating is used in RF parts, for its conductivity and the fact that the 
surface  oxidation is still relatively a good conductor. Also explains why center  
conductor of large heliax is hollow, as current only flows on the outside 
skin  of the inner conductor, not center.
 
Choice of proper capacitors in (higher) power RF circuits ALWAYS has been a 
 big deal. American Technical Ceramics (ATC) makes multilayer porcelain 
chip caps  and gets about 5.00 each for them. They work, and have been in use 
since the  '60's. Only recently has materials science been able to provide 
low cost  substitutes. Tiny modern 5 cent chip caps work amazing well!!
 
The tiny leaded cap that works so well in the original test? Probably a  
modern one with a low loss dielectric so no heating. Heating only occurs in an 
 RF part if there are RESISTIVE losses, ideal L & C have no loss. Resistive 
 losses in a cap occur primarily in two ways: metal losses due to 
conductivity  and dielectric losses due to loss tangent of the materials used for the 
 dielectric at the particular frequency. It is not the  current flowing  
through the C part of the capacitor but the RF current flowing in the lossy R  
parts of it (the parasitics in any "real" component)!
 
So, really no mystery here: All caps are NOT created equal.
 
VBR
Jeff Kruth
WA3ZKR
 
 
 
In a message dated 8/7/2017 4:10:35 P.M. Eastern Daylight Time,  
milsurplus-request at mailman.qth.net writes:

Just  want an idea of what type of RF current is across the capacitor, 
don?t need to  be complex. But I am going to assume that if the radio is 
developing   around 0.9 amperes of RF current and P=(I * I) * R  that would be 
around  40 watts. And at about ? of an Amp that would be just under thirty 
watts. This  is all assuming 50 ohms with an ideal load. So is it that you are 
developing  significant heat with currents as low as or under one amp?
Unless you?re  dealing with some weird power factor thing from the L52 and 
new capacitor  combination the voltages are below one hundred or so volts so 
back to the  original question, why dose the capacitor get hot? I know 
engineering can get  complex but this is about the simplest output tank out 
there.

Ray  F/KA3EKH


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/milsurplus/attachments/20170808/1b7a33cf/attachment.html>