[Amps] 50 w stud mount zeners
Jim Thomson
jim.thom at telus.net
Fri Aug 13 17:40:07 PDT 2010
Date: Fri, 13 Aug 2010 12:33:12 -0500
From: "Roy" <royanjoy at ncn.net>
Subject: Re: [Amps] 50 w stud mount zeners
"6A10s are a good choice, but for a high bias tube like a GS-35b you
need a lot of them - about 40. (24 VDC x .6 V per diode)."
CW key up / key down regulation is not good with a string of diodes. With a
dozen I measured a droop of 1-volt. With a 50 w zener it was nil.
73, Roy K6XK
## I measured a 12% increase in V drop across each 6A10 diode.... and that's going
from 200ma [sorta typ zsac] to 2-3A. About 15 yrs ago, I tried putting an electrolytic across
the entire string of diodes. On ssb/cw... it smooths out the bias really good. But the cap has to be big
enough... like 5000-50,000 uf.. and have a high enough ripple current rating, to handle the full cathode
current [sum of grid + plate current] . This implies the ripple current rating of the electrolytic'S has to be
at least 2.5 X the cathode current. You can always parallel lytics to increase their ripple current rating.
## Having said that, I dunno if the 10,000 uf cap improves TX imd or not, probably, but I suspect it's
not a huge difference. It would be easy to open up the lytic with a spst toggle..for tx imd testing.
## On these amps that use electronic grid over current protection, the typ deal is..a resistor is wired in
series with the grid meter/shunt. The typ 1-3 volt drop across this resistor is applied to a pot, and the pot
is just a V divider. The lower voltage from the pot, [one side to wiper] is applied to the rest of the
grid over-current protection scheme. The point here is, why get focused on a lousy 1 volt of bias regulation
from a string of 6A10 diodes.. when you just ADDED 1-3 vdc of bias from the grid over-current protection ?
## On any SB-220/221... the stock grid chokes used have a 25 ohm dc resistance. When you pull loads of
grid current, you get a V drop across the chokeS... again more un wanted yo-yo'ing bias. Incidentally, the chokes
used on the L4 / L4B are only .915 ohm.
## I have toasted both 10 + 50 W zener's in the past. The problem with zener's is they require a big heatsink...
and the 2nd problem is they don't have much of a surge rating. A single 6A10 has a 400A surge rating. You can
also parallel 6A10's to increase the surge rating. I also noticed that the 'new' 10A10 looks identical to a 6A10..
with the same 400A surge rating. It appears all they have done is heat-sinked the ends with big square traces on the
PC board. I noticed the same effect when testing strings of 6A10 diodes in the test jig. The 1st + last diode's would
run cool... when the middle diodes were warm / hot [ 3-6A flowing] . My test clips were in effect, heat-sinking the input
of diode #1... and the output of diode # 40 ! Since I was buying 6A10's 250+ at a time..for B+ supplies, I had plenty
on hand to make bias diodes, and also meter protection, etc. The 6A10 is a one diode, does everything diode. Come
glitch time... you won't be destroying 6A10 bias diodes. You can always insert a FAST 3AGC cathode fuse in series
with the bias diodes. [ and wire a parallel pair of 100K 3 watt mof's across the fuse holder]
Later... Jim VE7RF
More information about the Amps
mailing list