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

RE: Twin SSTC (fwd)



---------- Forwarded message ----------
Date: Sat, 7 Jul 2007 13:14:51 -0600
From: S&JY <youngs@xxxxxxxxx>
To: 'Tesla list' <tesla@xxxxxxxxxx>
Subject: RE: Twin SSTC (fwd)

Greg,

Your preliminary experiments and results are absolutely fascinating!  Thank
you for sharing them!  It will be interesting to hear of your further
measurements and explanations of the low freq phase shifting effects you are
observing.  Please keep us informed.
--Steve Y.

By the way, what are the specs on the IGBTs you are using?
-----Original Message-----
From: Tesla list [mailto:tesla@xxxxxxxxxx] 
Sent: Saturday, July 07, 2007 11:31 AM
To: tesla@xxxxxxxxxx
Subject: Re: Twin SSTC (fwd)


---------- Forwarded message ----------
Date: Sat, 07 Jul 2007 10:19:32 -0800
From: Greg Leyh <lod@xxxxxxxxxxx>
To: Tesla list <tesla@xxxxxxxxxx>
Subject: Re: Twin SSTC (fwd)

Hi Steve,

The NLL twin prototype coils use a classic Tesla Coil primary circuit, 
with an inductor, capacitor and a switch.  There are no master 
oscillators, or any type of feedback system as used in DRSSTC designs.  
The switch is a single IGBT, which receives an ON signal for the 
duration of the 1st envelope (exactly 2 cycles in this case.)

Later on, during the secondary ringdown and after the main arc activity 
trails off, the primary switch is turned ON again, to recover the energy 
remaining in the secondary.  The timing and duration of this action is 
not critical.  It appears that up to 5-25% of the energy can be 
recovered using this method.

The design of the NLL twin prototype coils attempts to incorporate and 
test every feature that might be used with the final NLL 120ft coils, 
including aspect ratios, coupling, primary drive topology and energy 
recovery.  The design is necessarily as simple as possible, in order to 
scale to the final size with minimal risk.

Yes, each tower must have it's own independent drive system, as a 600ft 
round-trip xmsn line operating at 50,000A pk would not be practical.  
The phasing between the two coils is then adjusted by setting the time 
delay between the two envelope command signals.

So far, we've had just two opportunities to run the coils; at the Maker 
Faire and the Integratron.  The coils displayed a wealth of unexpected 
and unexplainable behavior, particularly in terms of how the two 
secondaries interacted and coupled with each other.

For instance I had originally assumed that the streamers would repel if 
the two coils were in phase, and attract when the coils were 180deg out 
of phase.  Observed behavior at Maker Faire was completely different.  
The arcs would bridge as expected at 180deg, but adjusting the phase in 
either direction would cause one streamer to *grow* in length while the 
other streamer would almost disappear!  The first 10 seconds of this 
YouTube clip shows this effect, as I run the phasing from 0, thru 180 to 
360, and back:  http://www.youtube.com/watch?v=06hWFQSqJfI  

Scope waveforms of both coils' Isec indicated that the phase was indeed 
changing, and that the Isec amplitudes were remaining relatively 
constant.  However, Ansoft Simplorer simulations indicate that 
substantial energy can couple between the secondaries through the 
toroid-toroid capacitance.  In the simulation, a mere 6pF (my rough 
estimate for the actual Ctor-tor) will couple nearly all the energy from 
one sec to the other in about 8 cycles.  I plan to make a more refined 
measurement of the secondary interactions when the San Francisco lab is 
set up later this month.

One other odd bit of behavior occurred when the phasing was set around 
180, and the primary voltage was decreased so that the streamers barely 
touched.  One could see the interaction point physically move several 
feet back and forth between the coils, at a speed of about 1-2 Hz.  This 
movement was strongly reflected in the Isec waveforms, in the form of 
their *Fres* alternately changing, at this 1-2 Hz rate!  That is, the 
start of both Isec waveforms remain locked in phase, but about 6 cycles 
into the envelope the relative phases are shifting past each other by 
over 90 deg, at this 1-2 Hz rate.  I don't have any footage of this 
effect, but it's fairly easy to replicate, and  plan to do so as well 
when the labspace is ready.  My only theory at this point is arc loading 
of Fres.   GL