On Tue, 9 Dec 97 18:24:56 -0800 Rich Measures <measures@vc.net> writes:
>>
>>On Mon, 8 Dec 97 20:00:24 -0800 Rich Measures <measures@vc.net>
>writes:
>>SNIPS
>>I AM suggesting that 1500W of 28MHz RF from any amp will seriously
>heat
>>up a #16 CU wire from the plate cap thru to the top of the Tune cap.
>I
>>suspect that you could understand why if you stopped your hyperbole
>long
>>enough to study it.
>>Dick, Tom, Ian and others I believe have covered this ages ago.
>RG58/u is rated at roughly 600w at 28MHz -- which works out to 2.3A.
The actual ratings of RG58A/U is 650W at 10 MHz not 10M. At 50 MHz it is
250W. But that has nothing to do with the discussion.
You must have been taking dancing lessons from another.....
>As
>I recall, the center conductor of RG58/u is quite a bit smaller than
>#16
>-- like maybe #19. If 1A will "seriously heat up #16 gauge Cu", why
>doesn't 2.3A seriously heat up #19 (or so) gauge copper?
>- At 25A/60Hz, #16 gauge Cu gets seriously warm in free air. Does it
>
>make sense that 1A at 28MHz could do essentially the same thing?
Why dont you just try it and find out. Replace all the plate wiring and
strap in your SB-220 and key down for several minutes.
>>>... ...
>>>>Rich...how many people have a hi-pot tester available? You, Tom, me
>and
>>>>who else on this reflector???
>>>
>>>Beats me, Carl, but next to a DMM a high-pot. tester is the guddest
>>>amplifier troubleshooting tool around.
>>
>>Way down on my list.
>
>Without a high-pot. tester I can not find:
>1. a bad vacuum relay
>2. gas in tubes
>3. grid - filament shorts
>4. the actual piv in rectifiers
>5. flaky insulation in a transformer
>6. gas in a vacuum capacitor
>7. bad insulation in an air variable capacitor.
Like I said...way down on my list since those are not the types of
problems that people send me amplifiers to repair.
>>>
>>>>I have not been lucky enough to be present during a Big Bang since
>I
>>>>built the tester so I cant make assumptions based on zero data.
>>>rotten luck
>>>>Nor am I a tube expert that can unequivably state what happens
>inside
>>>a
>>>>tube during and after a gas arc. Are you???
>>
>>>I know that gas does not vanish whilst one is hooking a tube up to
>the
>>>high-pot. The gas arc theory does not wash because:
>>
>>Proof? Your reasoning just does not convince me. I asked you for a
>>spectrum picture....you have one to back up any claims???
>One SB-220 owner told me that he took the 220 to work and put a probe
>from a spectrum analyzer through the perforated cover of the output
>compartment. The signal was 50wpm CW dits. He said he could see the
>damped-wave ringing at c. 110MHz at the beginning and ending of each
>dit,
>and whenever the amp. was keyed or unkeyed with zero signal drive.
And replacing the possibly overheated and overvalue 47 Ohm 2W carbon
cured it ??
There should have also been grid current flickering when it was zero
drive keyed if the parasitic was that pronounced as to always be there.
It would not suprise me that a slight change in the L spacing on the
resistor would have altered the analyzer picture. Did your owner take his
tests thru to a cure or conclusion?
>However, 110MHz regeneration did not occur during the test. I was
>not
>present during the test and no pictures were taken.
Too bad; but that is still just a minor parasite....I asked about a Big
Bang analyzer picture. Or do you find those dancing shoes moving again?
>>
>>>1. an anode to grid arc sends no current through the bias zener,
>>>which,
>>>as you well know, often shorts during a big bang.
>>
>>No DC current.....
>>
>There is the normal DC cathode current. However the current from an
>anode/grid arc would not pass through the bias zener. The return path
>is
>through the meter shunts to the negative lead of the HV filter C.
The Big Bang will generate tremendous amounts of RF, at that instant, of
unknown frequencies. Zeners and the like do not handle that very well. It
does not have to be in the direct path if the wiring is saturated with it
and in the SB-220 at least the Zener protrudes into the RF cavity. The
SB-220 zener is much more prone to shorting in an event than the TL-922
as an example.
>>>2. an anode to grid arc is quite unlikely to destroy Rs.
>>
>>prove it.
>>
>Rs is typically paralleled with a 3.5 inch length of coiled #16ga.
>copper
>buswire -- the resistance of which is around 342 micro-ohms per inch,
>for
>a grand total of 1200 micro-ohms. Assuming a 200a peak fault current,
>
>the voltage drop across this amount of R would be 2.4v peak -- which
>equals 0.06w peak in a 100 ohm Rs.
That assumes a DC path only. If the L in the suppressor is acting as a
RFC then the R takes all the gaff or in actuality the load is somewhat
shared assuming an imperfect RFC.
>
>>>3. gas does not magically vanish.
>>
>>Neither do parasites
>>
>agreed, but you can make them less common and add glitch protection.
No argument there at all.
73 Carl KM1H
>Rich...
>
>R. L. Measures, 805-386-3734, AG6K
>
>
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|