Transformerless magnifier

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

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

Hi:

Continuing with my experiments about Tesla-coil-like circuits
without magnetic coupling, I made a "magnifier-like" directly
coupled circuit. The schematic diagram is:

                            (=====)
                               | 
                               L3
                               |
   +  o------+--C1--+--L2--+---+
             o      |      |  
 NST        gap     L1     C2  
             o      |      |  
   -  o------+------+------+---o Ground

The objective of the addition of L2-C2 to the basic
transformerless circuit was to achieve faster energy
transfer, and so reduce losses.
I used the same primary capacitor and the same resonator
used in my previous transformerless Tesla coil:
http://www.coe.ufrj.br/~acmq/tesla/mres4.html
The best performance that I had obtained with that system
so far was with a 5.09 nF MMC capacitor, with complete
energy transfer in 11 cycles, 38 us, with maximum voltage 
gain of 21.8 (mode 21-22).
With a 6th-order system I can obtain complete energy
transfer in just 3 cycles, 12.5 us, with maximum voltage
gain of 18.1 (mode 5-6-7).
The ideal element values are:
C3 = 15.4805 pF
L3 = 28.2 mH
C2 = 280.5975 pF
L2 = 1555.1870 µH
C1 = 5.09 nF
L1 = 90.7721 µH
I made L1 and L2 calculated by Wheeler's formula with #18 wire, and 
got 87 uH and 1521 uH. Close enough.
For C2, I made an adjustable capacitance with an 8.8 cm PVC tube,
with a fixed cylinder of aluminum foil inside and a moveable
cylinder outside. C2 would have to sustain about 20 kV during
the energy transfer cycle.
Low-power tests showed the correct waveforms:
The ideal voltage waveforms over the capacitances:
http://www.coe.ufrj.br/~acmq/tesla/567v.gif
Measured voltages:
http://www.coe.ufrj.br/~acmq/tesla/567v1.jpg
http://www.coe.ufrj.br/~acmq/tesla/567v2.jpg
http://www.coe.ufrj.br/~acmq/tesla/567v3.jpg
It's easy to tune the system by changing the length of an antenna
at the top of L3 and changing C2, while observing directly the 
voltage over C1 or the voltage over C3, picked by a nearby antenna.
Tests at full power, with a 5 kV 30 mA NST showed that the circuit
works, with almost the same performance of the 4th-order system.
http://www.coe.ufrj.br/~acmq/tesla/567all.jpg
http://www.coe.ufrj.br/~acmq/tesla/567spk.jpg
The performance was not very impressive. The input power is low,
It was raining, humidity was high, and the capacitor C2 was not good. 
There was more corona around it that at the terminal...
http://www.coe.ufrj.br/~acmq/tesla/567c2.jpg
As the tests proceeded, I noticed that the system was becoming out
of tune. Verifying C2, I observed that its capacitance had increased
a lot, becoming almost independent on the position of the moveable
plate. The surface of the PVC, where the intense corona was occurring,
had become significantly conductive, and hot. Cleaning the tube 
restored the normal capacitance, but I see that I have to make C2 
in some other form, or at least with some other material. 

Next experiment, If I manage to make a reliable C2 (it will have to
be bigger and to sustain more voltage, but not by much), will be to 
increase C1 to extract all the power available from the NST. With 
this structure, I can have C1>>C3 and high voltage gain, without
the compromise of long energy transfer time that appears with the 
two-coils system.

I will set up a page with more details soon.

Antonio Carlos M. de Queiroz