[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Micro Solid State Spark Gap Results
Original poster: "David Sharpe by way of Terry Fritz <twftesla-at-qwest-dot-net>" <sccr4us-at-erols-dot-com>
> Hi Dave,
> I am very interested in your research. Is a schematic available??
It's really almost too simple. DC power supply through a charging
resistor or resonant reactor to a low loss AC switch. Switch is
turned on and off using a Toshiba TLP250 optocoupled IGBT gate
amplifier module. Circuit is virtually the same a Terry's OLTC circuit,
except I used two N Channel Power MOSFET's connected G to G,
S to S. One drain goes to node between charging current limiter and
tank capacitor, opposite drain goes to ground. (See IGBT / FET
C-E-E-C ASCII art below, replace C with D, E with S for
Power N channel MOSFET's). Opposite side of
storage capacitor goes to primary inductor, then opposite end of
primary inductor is connected to ground. Keep in mind maximum
voltage on my lab top charging power supply is 16V, FET's are good
for 30V. Terry's solution of using IGBT's to conduct on positive
alternation and let negative alternation freewheel through diode is
an eloquent solution to reverse current flow. I elected to use two
gated switches to maintain symmetry, since that was one of the
parameters under measurement.
Also the previously posted idea of using an SCR / high power diode as
a freewheeling high power switch is in fact done with CDI systems. For
example I suggest reading the following App Note at SGS Thomson,
concerning capacitive discharge ignitions; paying particular attention
to topology 1.
I also suggest you visit SGS Thomson Application note area and review
those app notes pertaining to CDI using damped sine wave driving.
The ring wave in the primary of a CDI coil is EXACTLY the same
response desired for a solid state spark gap (SSSG) in a TC primary
circuit. The only difference is SCR's are not suitable; they will NOT
handle RF, in fact the highest frequency I've ever heard SCR's being
pushed is <50kHz, with 20kHz being used quite a bit is X-Ray PS.
More often in first and second generation power electronics, the
switching frequencies were less then 10kHz.
To develop the equivalent of this circuit to operate at appropriate Tesla
Coil RF, you have to use either:
1. A group of paralleled IGBT's or Power MOSFET's copack with
Schottky freewheel diodes (Terry's approach);
2. Use two IGBT's / FET's in what is termed in matrix converters a
'C-E-E-C' configuration, or multiples of this configuration wired in
The C-E-E-C configuration has two IGBT's or Power MOSFET's with
copack diodes connected with emitters connected together in series.
The gates are connected together, the power flow is between the
collectors: Current steering is performed by the copack fast recovery
diodes. The basic topology will also work with Power N Channel
MOSFET's, see fixed width ascii art below:
Copack D Copack D
| | |
HV --+-C E---+--+-E C--+-- HV (GND)
GE Drive In
'CEEC' Bidirectional Power Switch using IGBT's
Copack D Copack D
| | |
HV --+-D S---+--+-S D--+-- HV (GND)
GS Drive In
'CEEC' Bidirectional Power Switch using Power MOSFET's
You simply turn GE (GS) on for what ever period you want
your switch on and to allow under damped wave to disperse
(or whatever current "notch" you so desire)..
The only negative with this design is you need high
voltage devices for any real high power OR a number of them
wired in parallel. From what research I've done, it appears
that IGBT's in parallel are more problematic then Power
MOSFET's in parallel, but I have '0' experience with IGBT's at
Also a note for the astute; this switch COULD be used as a
high frequency "chopper" to PWM the utility AC wave to
control power into a transformer (a Solid State Variac).
The only difficulty with that approach is the line currents are
interrupted and filtering must be used on line and load side
to maintain line currents to prevent excessive kickback, and
control EMI/EMC generation. Unfortunately at this point, a
C-E-E-C "totem pole" module doesn't exist as a commercial
part, but once matrix converters become mainstream, there
will be economic justification for suppliers to develop
modules of this configuration.
So in closing; a SSSG system would basically be a
capacitive discharge ignition (CDI) system enlarged beyond all
sanity to handle bi-directional currents approaching 1kA or more,
input voltages of > 300V, and resonant frequencies in the
30kHz-500kHz range. There is enough CDI information on the
web, that a SSSG can be built; if the appropriate solid state
switch components could be obtained; and appropriate
precautions in regards to shielding, grounding, HV isolation etc.
Dave Sharpe, TCBOR
Chesterfield, VA. USA