On Nov 8, 2006, at 12:47 AM, pentalab wrote:
--- In ham_amplifiers@..., R L Measures <r@...> wrote:
RICH SEZ.... In my experiences, the G2DAF design is c. 11db
cleaner
with a 2-tone modulation test than it is with voice modulation.
With voice modulation, total IMD is c. 21db below pep.
### I have a 2 x 4-400 G2DAF linear. About 900w out with 75
w
of drive... and only 2700 V. Any more than 75 w of drive... and
it
won't do any more than 900w.
### The secret with the g2DAF design is selcting the PRECISE values
of screen caps to charge up. Too small... and u are outa luck....
too big... and all the energy gets diverted into charging em up. A
friend here in town built several 4 x hole 4CX-250B linear's..
and
also several 2 x 4-1000 linears... all in G2DAF. The 4CX-250B
design came straight from G2DAF himself... back in 1968/69. You
can't just take the original 813 design... and stuff any tetrode
into it. It has to be tweaked just right... for a specific tube.
### My 2 x 4-400 was built by the local RI here in town... uses a
4:1 un-un to step up the Z to 200 ohms... then terminated into
200
ohms.
### IMD on ssb is better than I thought. We ran several tests here
in town.... NOBODY bitched about IMD. I could switch it with a GG
linear in under 1 second... and nobody heard the difference. Not
one
of my favourite designs.... but done exactly right... it works
good. Done wrong... and it's not as clean as could be.
## It changes class as drive cycle increases. The whole thing
hinges on charge rate of the screen caps.... get em wrong.. and IMD
will go to hell.
### I have heard several 4-1000 G2DAF linears on ssb.... also
switched to 4-1000 GG [2nd linear] in mid sentence.... while a few
of us monitoring off freq [some up.. some down]. Damned if we could
hear any diff on or off freq. I have also heard some 4-400 G2DAF
linears that were nothing but problems. They are a trick to get
just right.... so u just can't right em all off as.. "bad".
### With no drive applied... no screen V is developed... and no
idle
current... tube cut off..... it saves a complicated, possibly
failure prone regulated screen supply.. + 2 x bias supplies, [RX +
TX].... + NO tuned input required... since it's terminated in a
globar. They look + sound very good on CW too..
On CW, Class C sounds delightsome.
.. so it DOES
have
some merits. Somebody In EU.. thought SM land... has a 4CX-10,000
in G2DAF.... who swears by it IMD wise.
Those who operate 5kHz from a G2DAF swear at the IMD. Here in the land of fruits and nuts (California) I have listened to three of them on 40m and 80m. All were ok with a 2-tone test, but with a human voice they definitely did not cause others to rejoice.
As I understand it, G2DAF did not know how to go about building a regulated screen supply, and that's why he came up with his cockamamie idea of a virtually Class C linear with Class C levels of efficiency.
Dunno what gain it
has.....
a lot less than if a real screen supply used..... and a bit better
than GG.
### My 1st choice would still be a hi MU triode in GG.
The 8160 high-Mu triode requires c. 1500w to drive it nicely. An 8171 or 8281 tetrode in AB1 can be driven with 170w . However the g- g triode is cleaner, but OTOH the input circuitry for AB1 grid-driven is easier to build.
Later.... Jim VE7RF
R L Measures, AG6K, 805.386.3734
r@..., , rlm@..., www.somis.org
Yahoo! Groups Links
R L Measures, AG6K, 805.386.3734
r@..., , rlm@..., www.somis.org
