[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: DRSSTC thoughts...
Original poster: "Steve Conner" <steve.conner-at-optosci-dot-com>
>C. can comment
>on the models he's been using.
Wow. It seems DRSSTC fever has finally hit Pupman :P
First of all, here is the PSpice simulation I use.
Maybe Terry and Dan Mc would like to play with it. There are two
simulations, each has a preset list of displays and commands for Probe. If
you run the .cmd file in Probe, several plots should appear showing useful
It uses the parametric simulation feature to simulate the coil dozens of
times with different streamer loadings, and see the worst-case currents and
voltages. The expected spark length can also be predicted. There is another
simulation file that does the same but with different primary inductances,
to help you find the correct tuning point.
Qualitatively this sim seems to work quite well. It predicts the doubling of
primary current for extreme impedance mismatches (like heavy ground arcs)
and so on. And it also does a reasonable reproduction of the bizarre
waveform from my Tesla-2 run as a DRSSTC. But quantitatively it still has a
way to go!
To try and refine the model I am starting a series of streamer loading
experiments (these are a joint venture with the high voltage technologies
group at Strathclyde university)
I'm using a similar method to what Terry did but over a wide range of
breakrates and bang energies (which can be accurately controlled on the OLTC
II) I have hardly any data yet, the little I do have suggests that a 36"
streamer at 100bps, 66kHz has a shunt resistance of about 400k.
So far, I think that we can adequately model the streamer load as either a
series RC or a shunt RC circuit. This is a "rough approximation" that only
holds at a single frequency. But since the DRSSTC's output only contains a
narrow band of frequencies I don't see a problem.
I assume that the R and C vary in some way as a function of streamer length.
At the moment I believe that R is proportional to (1/Length) and C is
proportional to Length. But I wouldn't be surprised if there turned out to
be a squared or square root in there. The constant of proportionality will
probably depend on many things too, for instance the breakrate and the