[Amps] HV rectifier strings - to bypass or not?

[Jim Thomson]

Date: Sat, 30 Oct 2010 10:46:26 -1000
From: Ken Brown <ken.d.brown at hawaiiantel.net>
Subject: Re: [Amps] HV rectifier strings - to bypass or not?


> I'm getting ready to lay out the HV rectifier board for my YC156 
> project, and it occurs to me to wonder if anyone has ever really heard 
> HV rectifier "white noise" in the receiver? It would be simpler to 
> forget about the per-diode bypass caps and just bypass the output as it 
> leaves the board. The supply will be housed in a separate cabinet from 
> the RF deck, but connected of course with bypassed leads for HV, (SHV 
> connectors) 110VAC and control signals.

I thought the reason for capacitors and resistors paralleled with the 
series diodes was to help equalize the voltage division across the 
diodes, and reduce the chances of an unequal voltage division, exceeding 
the voltage rating of one of the diodes. I did not think it was for 
noise reduction. I had also heard that it was more important with 
earlier manufactured diodes, due to lower voltage ratings and not quite 
as good matching from one diode to another. With more modern production 
diodes the voltage division tended to be more equal, reducing the need 
for the resistors and capacitors.

DE N6KB

##  The paralleled resistors  were supposed to be there to equalize the
V drops.   The purpose of the parallel caps was to eliminate switching transients
 when switch's  from conduction to non conduction.   Interesting enough, Ten-tec
still uses  the resistor's in parallel with the diodes  [no caps].....and   Ameritron
still uses  caps [ and no resistors]. 

##  KM1H  came up with a better method... and that's to install one HV disc ceramic
across each leg.....so 4 x caps in all.    The purpose of these 4 x caps  is to  bypass any
junk/transients coming in off the street. Of course the V rating has to be high if a single 
cap is used.  You can get disc ceramic caps in 4700 pf @  10/15 kv variety.  [ Ceramite brand,
sold by mouser+ henry radio still has em].   IF  .01 uf caps [10,000 pf]  installed across each diode,
and 8 x diodes used, you only have 10,000/8 =   1250 pf  effectively  across each leg.  If individual
caps used, .02uf @ 1 kv would be better.    IMO... 4 x big caps, one per leg, is more effective, and 
messier than 32 x  caps. 

##  On these YC-156/YU-148/3x3 amps, etc....with the HV supply in a separate box..... use 2-3 x 4700 pf @
10/15 kv bypass disc ceramic caps  for bypassing the B+ to chassis.   I would also install 1-3 x 500 pf @
10-20 kv TX doorknobs for bypass as well. [HEC/ITT-Jennings type].  The doorknobs make for better bypassing
on the higher bands. The above is all inside the outboard B+  supply box..[right at your SHV connector].

##   I then install the same 4700pf disc/500pf doorknobs at the base of the plate choke.  If the B+ cable
with the SHV connector on each end is also shielded, then ur really have all bases covered.  Bond the chassis of
the HV supply, directly to the  chassis of the RF deck. If shv connectors and coax used, that will be the case. I run
10 ga rw-90 between the amp and the mating RF deck anyway.  

##  That noise you hear on RX is shot noise usually, as described by Bill, Wa4lav...and very rare.  

Later... Jim  VE7RF