ALF prototype work, RTC tests

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

Original poster: "Greg Leyh" <lod-at-pacbell-dot-net> 

Hi All,

I'm resuming work once again on the 1:12 scale ALF prototypes.
Assembled one primary/secondary/toroid set, measured Fres and set the 
coupling.  Put some HV to the finished driver [only 2kV to start] and 
measured the intra-coil dynamics, and tested the gate control circuits.
Rep-rate 4Hz max, as the test charging supply is small.  Pulse waveforms 
appear normal, looks quite expectedly like two coupled resonators.  No top 
breakout at 2kV drive, though there's lots of secondary electric field; 
open scope probes see ~100V.
However, the most satisfying bit came from testing the 'secondary energy 
recovery' concept, that I mentioned briefly a while back in this post:
http://www.pupman-dot-com/listarchives/1998/january/msg00059.html
For this test, the energy spent two full amplitude cycles in the secondary, 
before returning to the primary capacitor.  The IGBT then commutated the 
primary current to the diode and turned off in about 1uS, trapping the 
energy in the primary.  Approximately 1650V of the original 2000V was 
reclaimed, or about 68% of the original energy.  This percentage will of 
course be lower once there's significant streamer action, but the general 
concept appears to work.

Why would one wish to recover the secondary energy after only a few 
cycles?  Current measurements on Electrum indicate that only the first 2 or 
3 cycles of full secondary voltage contribute significantly to dart leader 
production; once the amplitude decays by a relatively small amount, the 
spiky dart-leader currents disappear, leaving only the normal reactive 
capacitive currents.  In a standard TC, this remaining unusable energy 
would remain trapped in the secondary, eventually ringing down to zero -- 
lost to the secondary copper and the surrounding air.  But now that primary 
switches are available that can fully quench in less than a half cycle, 
this energy can be successfully recycled.

Historically, traditional TC designs haven't needed to bother with this, 
for two reasons:
A)  Overall efficiency is not a major issue for most coil applications.
B)  Primary switches with suitable quench performance were not readily 
available.

However, the heavy AC power requirements for the *full-scale* version of 
the ALF [about 5.5MW] will absolutely require some form of energy recovery 
for economically viable operation.  The ALF 1:12 scale prototype will 
feature this regenerative energy control scheme as well, for further study 
and refinement.  At the risk of saturating the list with yet another 
acronym, I'll propose that this recovery scheme be referred to simply as a 
Regenerative Tesla Coil, or RTC.

Most of the key components [secondaries, toroid parts, primary drive units] 
for the 1:12 scale prototype are now complete, awaiting final integration 
and testing.  Depending on workload, I'll try to get some waveforms and 
pics up on the lod-dot-org site around Thanksgiving.


-GL