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

Solid State Spark Gaps



On Fri, 13 Sep 1996, Tesla List wrote:

> > I gotta' chime here, too.  The mosfet would be the ideal switch if it
> > could just handle 22,000 volts and 600 amps of current at the same time.
<snip>
> > When all the smoke clears and theoretical stuff is pushed asside, 
> > when the wishes, hopes, and dreams for the future are
> > relegated to that future,  FETs, IGBTs and all the other stuff of this
> > century is pretty pathetic when it comes to big power at high voltages.
 
> Well.. how abuot using many (fast) thyristors or mosfets in series?
> Series using *is* possible with very carefull design. Anyone given that
> a greater thought?
> Thyristors could easily handle high peak currents. For fast switching
> they are still not as good as mosfets. Then, of couse there probably
> is quite few high current high voltage (say 1-2kV) capable GTOs
> around somewhere..


Actually, in recent years solid state devices have improved dramatically in 
their capacity for switching large peak currents, at high voltages.
The Pacific Intertie, which IMHO should be considered one of the seven wonders
of the world, delivers over 1,500,000,000 continuous watts from the Washington 
State Hydroelectric Power Authority to the LA basin nonstop via a 1 megavolt DC 
transmission line.  How is it that the receiving station located just north of 
LA can convert this immense amount of HVDC power back into 60Hz 3 phase and 
apply it to the local grid?  Solid State Switches!!

The primary reason that high peak capacity devices such as SCRs and GTOs
have not been applied to fast pulsed power applications (such as Tesla Coils)
is due to their di/dt limitation, which for a good inverter grade GTO is
about 500 amps/uS.  A typical TC has primary current risetimes on the order
of 1000 to 2000 amps/uS.  

The 500 A/uS limitation is due to the finite rate of growth of the conduction
area away from the gate structure, when the GTO is fired. However, A new 
gate design called the 'interdigitated gate structure' promises to raise this
di/dt limit significantly.  At work we are testing such a device, which is
called the Modified GTO.  At present, the device will hold off 4200V, and 
switch pulsed currents of over 50,000 amps peak, at risetimes of 15,000 A/uS!!

People at work imagine replacing the 6000 amp thyratrons along the accelerator
with MGTOs, but I already have dreams of sticking one into my coil.  My present
rotary gap drops about 1200V at 2000A, and takes over 200uS to quench.  The 
MGTO at work (stacked appropriately) would drop about 10V, and 'quench' in 20uS.

Perhaps, without all the rotating inertia of a rotary gap, I'll finally be able
to play 'Claire de Lune' on my coil.

-GL