[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: SISG Testing

Original poster: Vardan <vardan01@xxxxxxxxxxxxxxxxxxxxxxx>

Hi,

At 10:50 AM 6/12/2006, you wrote:


All:

I assembled (4) SISG boards and tested them on a very small coil driven
by a flyback based power supply.  Worked fine though I have not yet
re-tuned and duplicated spark length.  Tried it with from (4) to (16)
Sidac circuits(1 to 4 boards - 3600 VDC to 14400 VDC).  Could clearly
hear the bps frequency increasing as reduced the # of SISG circuits.  At
17 SISG circuits the capacitor charged up, but the Sidacs never fired -
the SISG threshold voltage was higher than the power supply voltage.

Then moved to my DC resonant MOT based coil.  Here is the schematic of
this system:

http://www.teslaboys.com/SISG/TeslaCoilv1.bmp

The MOT transformers produce about 6000 VAC which when rectified gives
8500 VDC at the filter caps.  The DC resonant circuit then pushes things
up to about 17000 Volts.

I replaced the RSG with the SISG - trivial concept, but here is a
schematic:

http://www.teslaboys.com/SISG/TeslaCoilSiSGv1.bmp

When operated the SISG fires one bang and then stays on due to the DC
current supplied by the filter caps thru the charging reactor.  The
filter caps have reduced the DC ripple to around 1000 vdc and this is
not enough to allow the sidacs to shutdown.  What I need is a pulsed DC
or unfiltered DC to make it work.

My filter caps were integral with the rectifiers on a board so I cobbled
together a rectifier and tried this:

http://www.teslaboys.com/SISG/TeslaCoilSiSGv2.bmp


This looks much like mine but I have a blocking inductor after the bridge.



I could not get this to fire.  At the time I believed it was because my
voltage was now too low, but in hindsight I think I may have blown a
rectifier.
Beware that the bridge will see "twice" the maximum voltage. If you fire at 17kV "each leg" of the bridge must be able to take 34kV!! Better count on 40kV... This is due the the "ring" in the resonant circuit driving the load negative in the first 1/2 cycle while the power source is still positive in the opposite direction. That puts twice the voltage across the bridge. 



So I cobbled up a voltage doubler arrangement like this:

http://www.teslaboys.com/SISG/TeslaCoilSiSGv3.bmp

To get my peak voltage back and have a pulsed DC charging system... But
after it did not fire I discoverd the 15 kv rectifiers were dead short.
They were the same rectifiers from the previous test so I am not sure
when they blew.
See above. The Cd caps seem in parallel with Ctank. Would this not detune everything in the primary tank without a blocking inductor or resistors. 



Now I planning to set up the last arrangement properly with correct
components, but am confused as to what value to chooses for the caps(Cd)
to get voltage doubling, but not smooth the DC so much that the Sidacs
won't turn off.
Could the Cd caps be used as a tank cap. There is no reason to have a separate tank cap if each of the Cd caps is 2 X the tank cap value. 



I am also starting to feel uncomfortable with the center tapped MOT's
tied to the voltage doubler and am getting confused as to what the DC+
and DC- legs will be out of the voltage doubler.  Comments please?  I
don't see a current path for the DC side with this set-up.
Positive is on top. The caps and all are floating all over the place riding on the AC.
I think this will take 1/60th of a second to charge though so the BPS will be 60 BPS. 1/2 the cycle charges the top cap and the other 1/2 cycle charges the lower cap. 



Assume the input legs to the doubler are rms 3000 VAC phase-to-ground
and 6000 VAC phase-to-phase.  Thus 4200 V peak-to-ground and 8400 V
peak-to-peak.
Most MOTs get to about 2500V++ peak so that "might be" 10kV++ volts and doubled to like 25kVpeak!!! It depends on how high you turn it up and where the gap fires, but beware of that. The diodes have to be rated for 50kV!!! There needs to be a blocking resistor or inductor since the diodes will conduct hard once the gap fires spoiling the "Q". Once the circuit oscillates and the voltage across the cap is in the opposite direction, the diodes become a dead short (guess how I know this :o)))
Beware that the blocking inductor will have to withstand very high voltage across it too!! Might just us a secondary coil for it ;-) 



So I originally thought the output from the doubler would be 8400 DC+
and 8400 DC-, but now I am having reservations.  Normally, the DC- side
would be grounded, but I can't do that because I am feeding from a
center tapped MOT series(center tapped to minimize stress on the
windings).


The relative voltage on the doubler is 16800VDC floating on 8400VAC.
At the voltage your are working at, I wonder if an NST is a better choice than the MOTs...
DC charging coils or sort of messy anyway and the SISG adds to the messyness... It is not all perfectly figured out yet how best to do it.... Perhaps the voltage source should be across the gap somehow like we do for conventional NST systems... I am not sure at all... Have to work on it... The SISG is like a new toy that we have not learned completely "how" to play with yet ;-)) 

Cheers,

        Terry



Thanks.

Mark