Magnifier first light

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

Original poster: "Antonio Carlos M. de Queiroz by way of Terry Fritz <teslalist-at-qwest-dot-net>" <acmq-at-compuland-dot-com.br>

Hi:

Today I powered up my experimental 6th-order magnifier and made it work
at full power.
It was designed for mode 3:4:5, with these calculated parameters:
C1=    5.0800000000 nF
L1=   53.7909448819 µH
C2=   69.0793650794 pF
L2=    3.9480000000 mH
C3=    8.5000000000 pF
L3=   28.2000000000 mH
k12=     0.3504383220
Design theory: http://www.coe.ufrj.br/~acmq/tesla/magnifier.html
The structure was essentially this, but with a better C2:
http://www.coe.ufrj.br/~acmq/tesla/mag345t.jpg
The tests were quick, and there are some construction details
and tuning that can be improved, but the results were:
It works, with good performance, but not better than a similar
two-coils system using the same C1, L3, C3
(http://www.coe.ufrj.br/~acmq/tesla/tefp.html).
The performance is better than what I obtained with a slower
directly coupled version, with an even more lossy C2
(http://www.coe.ufrj.br/~acmq/tesla/mres6.html).
The C2 capacitor was implemented with two Leyden jars made with
plastic cups in series, for 45 pF of capacitance, and many tens
of kV of maximum voltage. The capacitor was adjusted slightly below
the correct value, antecipating some increase in capacitance due
to corona. The influence of the exact value of C2 in the
waveforms is quite high.
This C2 capacitor shows significant of corona around the
plates, and becomes quite warm quickly. It's evidently dissipating
significant energy.
The system works better -without- C2, producing longer streamers.
The distributed capacitances may be enough for some higher mode,
or are not having significant influence, or the losses in C2 are
too high.
An oscilloscope with an antenna close to the system shows the
correct waveforms, but no quenching when C2 is connected. The
energy disappears quickly in small streamers in all directions
that cover the entire top load, accompained by a weird blue haze.
The oscilloscope shows essentially only a single burst with 4 cycles
per bang.
The addition of another turn to the primary (allowing greater C3)
improved the performance a bit. With greater C3, longer streamers
are possible without significant detuning, and with some more
it would be possible to use the half-sphere top load that worked
so well in my two-coils system.
Some corona was observed in the connection between L2 and L3,
and at the base of L3, possibly damaging the insulation there.
Pictures, only after a slight reworking and retuning, maybe by next
week.

Regards,

Antonio Carlos M. de Queiroz