[AMPS] parasitics

[Jon Ogden]

> The main
>difference is that Rich tries to get by with #16 wire whereas National
>used 1/2" wide strap around a big 50 Ohm non-inductive resistor. 
>I do object to using that material in the complete plate to blocking cap
>path and prefer the traditional low inductance wide silver plated
>strap/tubing strap. If the suppressor is doing its proper job right at
>the tube then there is no reason to continue nichrome to the blocking
>cap.

Carl, I wondered about this as well.  The first thought in my mind was:  
Just 16 Ga. wire...will that work?  Then I realized the the coil in my 
original supressor was made with 20 Ga.wire!  If 20 Ga works, then 
certainly 16 will.  Additionally, the standard supressor might work fine 
today.  But what if it is marginal and as the tube ages, something 
changes to decrease its effectiveness.  Or what if you put in a tube that 
has better gain which again does the same thing.  I have always 
over-designed all my circuits and I am just trying to bullet-proof 
things.  Lowering the Q is never a bad thing.

 
>Layout is critical and in a sloppy version it may be necessary to use
>more nichrome or whatever. But in a good layout I would think that the
>lowest possible inductance in order to move resonance above the tubes
>ability to oscillate would be the "good engineering" approach.

Yes, layout even at HF is EXTREMELY critical and can solve many a 
problem.  Agreed.  However, a 4-1000 still has gain up at 150 MHz and 
many of the tubes today go much higher.  If the anode circuit is resonant 
at 75 or 80 MHz, please tell me how you are going to move it 70 MHz 
higher, let alone 20 MHz higher with a coil and resistor.  From what I 
can figure there is no way to move the resonant frequency that much (at 
least easily).  Sure, maybe you can move it from 80 to 90 MHz, but I 
doubt the tube gain is that much different there.  So what you are left 
with is having to reduce the Q.
>
>Secondly, I have harped for a year about VHF resonances in the tank
>circuit with nary a bit of comment from anyone. Now I see that Rich has
>finally added that possibility to his ever varying repertoire. National
>Radio solved that problem very neatly in the NCL-2000 and NCX-1000 by
>adding a SERIES cap...not a shunt as has been discussed about here the
>past few days. The National engineers added a 10pf capacitor across the
>20 and 40M  shorting bandswitch contacts which eliminated a particularly
>nasty proclivity to arcing.; perhaps you would wish to analyze that
>circuit.

It's an interesting idea.  Although, I do believe that you could put a 
cap in series or in shunt.  And series circuit can be turned into a 
parallel equivalent circuit and vice versa.  Although a series cap across 
the bandswitch might be easier.  Yes, filters can have resonances outside 
their passband and that could be a problem.  Adding the caps could help 
that.

>I use that idea in stock SB-220's and have had users report no more
>switch or Tune cap arcing. I've used it in various Alpha 8874 amps that
>had arcing problems...and suppressor R that was as new.
>
Very possible.  So since the cap removed the VHF resonance in the tank 
circuit, I would conclude that you do admit the arcing was caused by a 
VHF oscillation.  So oscillations can cause arcing.
>
>Thirdly, my dis-belief centers around the big bang and that is where I
>suggested ( or that was my intent anyway) an article. You are the
>engineer, show me how a VHF parasitic has enough energy to bend a 3-500Z
>filament. 
>I say it is a gas/plasma discharge and partially caused by barnacling AND
>the lack of proper PS surge protection. It is this area only that I
>refered to junk science.

Well, as I said earlier, this I cannot comment on since I do not have 
knowledge of tube structure.  My experience with parasitic oscillations 
has not been of the "big bang" type, but I do not dispute that it could 
happen.  So I would not want to write a paper on anything that I don't 
have knowledge of or understand.  The part of Rich's theories that I 
support are about lowering the Q of the resonances in the tank.  
>  
>The BIGGEST problem in many ham amps IMO is the save a buck attitude of
>the manufacturers. They use marginal components, leave out key safety
>features and would be laughed out of anything but the ham market.

I'd agree.  However, you'd be surprised how many people have the "we've 
always done it this way before attitude." 
>
>My own suggestions for an amp would include:
>
>1. HV surge suppressor resistor of a value and wattage that can limit the
>instantaneous PS discharge current to a safe non-destructive value.
>2. Use of a suitably placed series capacitor in the tank circuit to
>provide a low impedence VHF path to the load. 
>3. Use of a suitably sized parasitic suppressor resistor and phase out
>carbon composition.  A 5W metal oxide should be the minimum for 3-500Z
>size tubes; a pair of them for the 4-1000A.
>4. Use of a resistive suppressor L material only if necessary to tame a
>known problem tube such as the 3CX1200A7 or to compensate for layout
>problems.

Certainly not bad ideas at all.

To those who are tired or reading about parasitics:  I am sorry.  But it 
is an area some of us are interested in and I have been learning a lot 
about them.

73,

Jon
KE9NA


--------------------------------------------------------------------------
Jon Ogden

jono@webspun.com
www.qsl.net/ke9na

"A life lived in fear is a life half lived."




--
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