On 16 Nov, 22:06, Alejandro Lieber wrote:
In figu
http://1fcr.com.ar/classB.jpg
you can see a very interesting diagram of the audio modulator of an old AM
Gonset transmiter.
A 12AT7 as voltage amplifier. one 12AU7 with both triodes in parallel as
power amplifier and two 6DQ6A grounded cathode, grounded control grid with
audio input to the screen grids working push pull as pure class B.
R211 and R212 are only used for measurement purpose.
Why would they choose to use the screen instead of the control grid as input
? . Any suggestion ?
--
Alejandro Lieber *LU1FCR
Rosario Argentina
Real-Time F2-Layer Critical Frequency Map:http://1fcr.com.ar
With a tube like 6DQ6 and other TV line output tubes the Eg2 needed
for many purposes is much lower than Ea than for say EL34. If you
examine the Ea vs Ia Ra curves for normal tetrode operation you'd find
the Eg2 used for the curves, and you should see that for where Eg1 =
0V, you don't need a very high Eg2 to give a peak anode current of
maybe 0.4A at low Ea, so the screens in such tubes have relatively
high gm and thus don't require a huge screen to screen drive signal if
the Eg1 is fixed at 0V. Examination of the screen characteristics
confirm this. Hi-fidelity isn't needed and distortion doesn't worry
some users of radio transmitters. Although screens draw current when
they go above 0V, the Ig2 isn't huge, and the paralleled 12AU7 looks
set up so that it can make the input power needed for the 6DQ6. My
guess is that the IST has a fair old step up ratio, and of course wide
AF BW don't matter much either, and there is no NFB used to make thing
more linear. I guess the load of the RF section of transmitter has a
constant input resistance so the high Rout of the modulator does not
matter.
In effect the 6DQ6 are working as triodes, but because of relative
electrode spacing the Ra in this mode is about the same as tetrode Ra
when "normal" g1 drive is used, but the gm is very much lower than gm
for g1, hence the g2 µ is low.
There are some who may connect g2 to g1, and drive both, a bit wild,
because its like tetrode mode positive FB, but its doable. For hi-fi,
one could also drive g2, but apply NFB to g1, also crazy because not
much NFB can be applied because Vg2 has then to be quite high, and its
high already as it is. Most modulators I've seen are mostly 807s with
UL OPTs and all done like a normal Williamson or similar, but OPT is
say 1:1 and sec does the business at RF amp
Other tubes are 6CD6, 6CM5 etc. One may find the screen drive
remarkably linear. OK, its in screen-triode mode, and there must be
some internal NFB as there is in any triode, and because Ra is so
airey-fairey and high ohm value, slight internal triode NFB gives
linearity, but the Ra value changes little. That may sound queer, but
from the curves it must be about right. I guess g1 acts to screen off
space charge between cathode and g1. Anyway, the screen drive would be
acceptable for hi-fi if just the right value of Eg2 at idle is chosen
when Eg1 is connected to cathodes, so that a reasonable amount of Ia
flows at idle, say 50mA, to allow class AB2 screen drive.
I won't try the Gonset schema any time soon, but you can, so have fun.
Patrick Turner.