Re: The Quick and Dirty VTTC

["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 again, Folks!


Update time.  Without the arc lit, the EL37 plate current Ia
was 85mA, Ig2 38mA.  With the arc lit, Ia becomes 128mA and
Ig2 25mA, total 153mA, which tends to explain why a nominal
420V 120mA supply drops down to 325V ;-)  The plate input is
thus 28W without the arc and 42W with the arc, suggesting
that at least 14W rf goes into the arc.  It is of course
possible that some of the circulating power under no-arc
conditions also goes into the arc once it has been struck.

The arc emits significant UV.  White paper lights up
(fluoresces) in the usual fashion, but fluorite
(naturally-occurring calcium fluoride) does not, so there's
a key to the wavelength if anyone can turn up the raw data
regarding excitation.

Now here's a surprise, well it was for me.  This thing emits
*narrow*band* which implies a very high value of loaded Q
and relatively little damping of oscillations due to the
arc.  My grid dip meter located it in the 29Mc/s band, not
bad for an audio power pentode, and my Racal RA17
communications receiver (Brit boatanchor from 50s/60s)
pinned it down to 29,38 +/- 0,05 Mc/s; the signal has a
rather "spitty" quality to the sound which is hardly
surprising.  This is to the limits of audibility.  In other
words, the loaded Q is at least 300 and probably more, which
frankly amazes me.  The unloaded Q I guesstimated at around
1500, so the efficiency of the coil under load is maybe 80%
or so.  Coil inductance I calculate to be about 10,1uH,
implying a capacitance of 2,9pF, including strays.  A little
more measuring/thinking should yield a ballpark figure for
the load impedance of the arc.

Comments and suggestions welcome.

Dunckx