Cambridge 5ft spark

Once the public demos were over at Cambridge 2005, Paul Benham, Nick Field, Mike Harrison, Bob Golding, me, and probably some other people that I forgot, stayed back to try and break the Mjöllnir DRSSTC. We put an extra toroid on it, doubled the tank capacitance to 0.1uF @ 10kV, and turned everything up as far as it would go: the breakrate to 200Hz, the burst length to 300us, the current limiter to 450A, and the supply voltage to 270v using a large variac. ;-)

Not only did we fail to break it, but we got an amazing 5ft arc to one of the overhead light fittings. That's 4.4 times the length of the secondary coil. Before anyone gets all excited, it took 3.3kW (as measured by a Kill-A-Watt digital wattmeter) to do this.

We found that the gate drive had to be started first and the voltage turned up gradually. If the gate drive was turned on last there were disastrous flashovers. Steve Ward predicted this effect. His explanation is that you get an abnormally high output voltage if you switch on instantly at full power. If you apply power gradually, the air heats up and ionises and forms a heavier load that drags the output voltage down. This effect persists between bursts.

We also found it quite difficult to find the proper tapping point on the primary. Our first few tries gave huge primary current and no sparks at all. :-<


Warming up for the giant spark (pic by Mike Harrison)

Movie of the spark, shot on a camera phone by Bob Golding. 1.5MB MPG format.

Back to DRSSTC page