WAS: The VTCC, NOW: Rotary Marx SGTC

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

Original poster: David Sharpe <sccr4us-at-erols-dot-com> 

Ken

Basic concept looks very intriguing.  One comment, the diode
doubler as shown, the lower diode should be mirrored
horzonitally (i.e. the cathode to line input in series with anode
of upper diode to complete a full wave voltage doubler).

Now imagine what one could do by scaling up to MOT's,
MOT rectifiers, MOT's (as resonant charging reactor).
Doubled output of MOT is ~5.0kV. Two 1.0uF caps
in series good for 5.0kVAC; x 12 stages in this case
would be a whopping 60kV peak on a 41.6nF cap
(~75J discharge energy).  100X /sec ~ 7.5kW power
processing.  This looks intriguing, now the engineering
issues to deal with...
1.  Finding low loss, light weight (since caps are on
      rotating disk), HV, and high volumetric efficiency
      capacitors that are low cost in quantity.
2.  Tuning system is limited to adjusting primary
      inductance versus adjusting tank capacitance.
      Not a big issue on small machines, could be
      a huge issue on larger machines (>15kW power
      processing throughput).
3.  Rotating mass of disk may require higher HP
      DC (or VFD 3ph AC) motors.  Plus side is
      rotating disk velocity is very slow (<500 rpm).
      The slow approach and away electrode
      velocities are effectively multiplied by number
      of active stages, concerned about quenching
      with slow approach/away velocities on
      individual electrodes.  May require magnetic
      assisted quenching.

Very novel idea!

Regards
Dave Sharpe, TCBOR/HEAS
Chesterfield, VA. USA

Tesla list wrote:

 > Original poster: "K. C. Herrick" <kchdlh-at-juno-dot-com>
 >
 > On Thu, 25 Dec 2003 11:07:03 -0700 "Tesla list" <tesla-at-pupman-dot-com>
 > writes:
 >  > Original poster: Ed Phillips <evp-at-pacbell-dot-net>
 >  >
 >  > Tesla list wrote:
 >  >  >
 >  >  > Original poster: "K. C. Herrick" <kchdlh-at-juno-dot-com>
 >  >  >
 >  >  > Hi Steve-
 >  >  >
 >  >  > I enjoyed seeing the photos of your VTCC.
 >
 > [snipped]
 >
 > ...a Marx-like primary--where a set of resonating
 >  > capacitors is
 >  >  > repetitively charged in parallel from the mains & then discharged
 >  > in
 >  >  > series thru the primary coil.  I've always been reluctant to work
 >  > with
 >  >  > 5-12 KV of mains-ac lurking around!
 >  >  >
 >  >  > Regards,
 >  >  >
 >  >  > Ken
 >  >
 >  >         That idea will take lots and lots of capacitors.  In place
 >  > of the spark
 >  > gaps, how about a motor-driven multi-contact mechanical switch?
 >  > Such
 >  > things have been used for a similar purpose. If designed right you
 >  > could
 >  > eliminate the resistors.
 >  >
 >  > Ed
 >  >
 >  >
 >  >
 >
 > Ed, Steve Ward (& all)-
 >
 > Funny you should suggest that, Ed...  I've been toying with that very
 > idea.  I'll ask Terry to put up http://hot-streamer-dot-com/temp/tspk16s2.pdf
 > which is a schematic of what I've had in mind.
 >
 > I show 12 capacitors (each one actually 4 in series-parallel for reduced
 > electrical stress) mounted circumferentially on a rotating disk.  Each
 > connects to both a pair of brushes & a pair of gap-contacts (each such
 > pair shown schematically only as arrows).  The disk assembly is to be
 > mounted directly under the secondary and, so mounted, incorporates not
 > only the capacitors of the primary LC resonator but also one of the coil
 > turns (of which the capacitors are a part).
 >
 > The mains connects at the left, charging a pair of storage capacitors
 > thru current-limiter L1.  L2 and the associated rectifier are for
 > resonantly-charging the capacitors.
 >
 > As the disk rotates, the capacitors are alternately charged in parallel
 > from the mains, thru the brushes, and discharged in series into the
 > primary coil thru the gaps.  Notice that every other charging event puts
 > opposite-polarity charge on the capacitors.  That is to minimize
 > material-transfer across the gaps due to net dc current flow.
 >
 > If the assembly were ~14" in diameter, to accomodate my existing 12"
 > secondary, then 3 1/2 turns of 3/4"-diameter primary "wire" spaced at
 > 1.5" would yield about 3.3 uH inductance.  To resonate at ~140 KHz (my
 > secondary's Fr), I would need about 0.36 uF.  For twelve
 > capacitors-groups in series, each one would then be about 0.36x12=4.4 uF.
 >   All of those in parallel, for charging, would constitute 52 uF.
 >
 > I've simulated the charge/discharge cycle with 160 V-pk mains.  At 100
 > "sparks"/sec, I get the C's charged to ~450 V each.  But with only 12,
 > 4.4 uF capacitor-groups, the rms mains current is a low 8.5 A.
 >
 > If, instead, I use 24 groups, then 0.36x24=~8.6uF and 8.6x24=~207 uF to
 > charge up.  Using 207 uF in the simulation, I draw 14 A rms from the 117
 > V mains, for a more respectable net charge to dump into the primary.
 >
 > I show a "main" gap at the right, if such is to be used while setting the
 > rotating gaps to essentially zero width.  A rotationally-synchronized
 > main gap could allow for more convenient adjustment of spark duration,
 > but it would add complexity.
 >
 > Tuning is to be accomplished using a flexible turn made from half-hard
 > copper sheet-stock.  It is to be bowed out, or not, using a threaded
 > nylon rod passing along its axis.  A traveling nut-arrangement would
 > connect to the end of the flexible turn, pulling it along & flexing the
 > turn.
 >
 > With 24 capacitances, the disk would turn at 300 rpm for a 100/s spark
 > rate.
 >
 > Will anyone take on this kind of design as a project, or at least
 > embellish this design with more or better ideas??
 >
 > Ken Herrick