Re: Modulated VTTC (plasma tweeter) question (fwd)

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Dr. Duncan Cadd by way of Terry Fritz <twftesla-at-uswest-dot-net>" <dunckx-at-freeuk-dot-com>

Hi Carl!

<snip>

>I just built a "plasma tweeter," which is a resonant coil base-driven
by an
>modulated RF oscillator.  The arc discharge from the coil then makes
sound
>according to the modulation. My system has a 6146-B tube operating at
10
>MHz, which is screen grid modulated by a 6CL6 tube as a clamp
modulator.
>There is a 12AU7 AF amp preceding the 6CL6 to boost the very low AF
input.
>
>Anyway, here's the problem.  I have quite a bit of distortion in the
output
>at high volume, and I suspect it results from improper resistance
values in
>the clamp-tube and screen grid coupling.  I can find no information
about
>how to use this type of modulation, especially for this application.
So if
>there's any help to be found, I figure one of y'all might have a
comment.
>

Well, let's assume that the 6146 runs from 750V, takes 120mA plate
current, 160V on g2 and g2 current 11mA. (ICAS Class C Telegraphy
example, RCA manual.)  Unfortunately, I don't have any data at all on
the 6CL6, much less characteristic curves, so can't do a complete
trial calculation.  However, if we assume that 160V on the 6146 g2 is
as high as it ought to go, then take say 110V as minimum and 135V as
average.

(This may be a bit conservative, you might want to lower the minimum
and average accordingly.  Also, there should be no rf on the screen of
the 6146, so a small bypass cap of say 0,01uF rated >=500V to earth is
in order.)

What you want to do is set up the 6CL6 to give a quiescent current
(with no modulation) through the dropping resistor such that there is
(in this example) 135V on the 6CL6 plate/6146 screen. As modulation is
applied, that voltage ought to change to 135 +/- 25V.  I'd be wary of
too large excursions in g2 volts, as the modulation may be far from
linear at very low g2 volts and also at too high g2 volts; moreover at
high g2 volts the g2 current becomes excessive and the dissipation is
easily exceeded.  You are unlikely to get more than maybe 75-80%
modulation depth undistorted with g2 modulation - perhaps your
distortion problems are simply due to a too large modulation depth.
But this example above should give you the idea.

The grid bias of the 6CL6 may be what is determining the distortion
assuming the 6146 g2 volts excursion is not too much.  Unfortunately,
you can only determine the correct value for the cathode resistor
which sets the grid bias for the 6CL6 if you know what negative grid
volts it should be running and what current it passes at that bias.
(Use the 6CL6 Class A amplifier characteristics for this.) Let's
assume it passes 20mA at -8V as an example and that this is
appropriate for the required modulation.  (Yet again, without the
characteristics I can't say if it would be.)  8/0,02 = 400 ohms, the
nearest standard value being 370 ohms.  It will dissipate 370x 0,02^2
= 0,15W so a half watt resistor is ample.  This should be bypassed by
say a 25uF 25V capacitor and a 0,01uF in parallel for the rf
(electrolytics can be a bit too inductive for rf decoupling.)  The
grid of the 6CL6 could then be connected to earth via say a 47k
resistor.

This hypothetical 6CL6 current of 20mA plus the 6146 current of 11mA
gives 31mA, which would be the total current through the dropper.
Assume the dropper is fed from the 6146 plate supply of 750V.  It must
drop 615V to give the standing voltage of 135V on the 6146 g2 and 6CL6
plate under no modulation conditions i.e. 615/0,031 = 20k ohms, and
20k x 0,031^2 = 20 watt (use larger for headroom).  This assumes that
the modulation on the 6CL6 is such that the swing of plate current
will vary the voltage drop 615V +/- 25V, i.e. with 20k that's equal to
a mere +/- 1mA in the plate current of the 6CL6, so you can see this
calculation is unlikely to be realistic, but it should give you the
idea of what to do and how.  From this you can see you'd probably be a
lot better off feeding the 6CL6 and 6146 g2 from a separate, lower
voltage supply, e.g. 250V, which you may well be doing for all I know.
That will reduce the value of the dropper resistor and make the
excursions in 6CL6 plate current more "real world".

Hopefully someone else on the list will have 6CL6 characteristics to
hand and will be able to give you more useful information.

>About the "tesla coil" itself:  I am a newcomer to the idea of
"plasma
>tweeters", but the apparently standard RF circuit is very simple and
>effective

The "original" is at:

http://members.aol-dot-com/uhaumann2/plasma/eplasma.htm
and my version, without modulation:
http://home/freeuk-dot-net/dunckx/wireless/qdvttc/qdvttc.html

>(Modulating it as a speaker and making it sound good takes a while
>longer!)
>
>-Carl


Yes, I really ought to squirt some WD40 on that round tuit and maybe
use a bigger hammer ;-)

Dunckx