[AMPS] ARCING MORE

[Rich Measures]

>This thread about arcing tune caps comes at an opertune time for me.
>Progress on the latest amp project here has slowed while I root around in
>the very bottom of the barrel for a tune cap. I was going to use a nice vac
>variable but have been unable to find a drive unit for it. Now the problem
>is selecting a suitable air variable.
>
>In the last five amps I built, there has been no tune C arcs in spite of
>more than my share of "cockpit trouble". Wrong band, wrong antenna, wrong
>drive, wrong plate voltage, nothing can zap the switch or the cap. This is
>probably because I have in the past always selected the output components
>based on the design criteria (AM?) in the older Handbooks. (ARRL and Orr's)
>
>In this project I don't have the luxury of using up-sized components.
>Looking at the various commercial amps out there, the spacing on the tune C
>is usually about 1/4 what mine is for the same power level. My question is:
>How does everyone else out there determine the spacing and/or breakdown
>voltage rating of the tune C? 
>
It seems to me that the actual piv across the Tune C is equal to the 
anode supply V minus the minimum anode/to/chassis-gnd V during the peak 
drive signal.  For a 3-500Z or an 8877, this is about 250v.  Thus, with a 
4000v supply, the Tune C would see about 3750v-p.  However, more 
piv-capability in the bandswitch is not necessarily good Because the Tune 
C needs to protect the bandswitch from arcing during  glitches.  In other 
words, a Tune C should have a lower piv-capability than the bandswitch.  
The TL-922 uses a (approx.) 6000v Tune C and a 6000v bandswitch.   
TL-922s tend to eat bandswitches.   {Under normal operating conditions, 
the actual piv across the bandswitch is 3100v minus 250v = 2850v.} 

>How about some hard facts and formulas? I really don't care about
>statements like "Very high voltages can occur during unloaded conditions".
>I have tried to derive the requirements and each time I come up with a max
>possible peak to peak RF voltage that's a little over twice the DC plate
>voltage under worst possible case. This seems like it's too little based on
>the recent discussions about arcing.

During the grate debate on vhf parasitics between yours truly and Mr. 
Rauch, someone measured the piv on a SB-220 Tune C during deliberate 
mistuning.  (the measurement was done with a voltage-divider and an 
oscilloscope).  .  The result was 3600v -- which reportedly did not cause 
an arc in the Tune C.   I tried the experiment in my SB-220, and I could 
not make the Tune C arc with the wrong antenna or wrong Tune C setting.  
-  In other words, the approx. mistune/misload piv on the Tune C is the 
anode supply V plus 20%.  
-+-   The $64 dollar question is how is it possible to arc the Tune C in 
a TL-922?  The anode supply is 3100v.  Adding 20% makes a bit under 
3800v, yet the Tune C withstanding ability is 6000v, yet some TL-922 
owners reportedly experience intermittent Tune C (and bandswitch) arcing 
when the right antenna is in use and the amplifier is not being retuned.  
.  .  The $128 dollar question is why do some folks get tight-jawed when 
an old-fart in Somis suggests measuring the resistance of Rs in the vhf 
suppressor of a Tune C-arcing amp.?  
.  .  .  "Everything is more complicated than it looks".  -  Murphy - 


cheers
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