OK, it took more than a couple of days, but here is the design for the apparatus that could be used to test the tornado hypothesis:

 

Tornado Vortex Exp.pdf

 

My biggest questions concern how to develop the degree of ionization necessary to simulate the real environment. The target should be 5 kV between the positively-charged air at the bottom and the negatively-charged electrodes at the top. Where the charges meet in the middle, everything should get totally neutralized. My understanding of Cockroft-Walton multiplier circuits in action is that after an hour or so, in an average sized room, you're likely to see 1 part per million that's charged, assuming that you've got a fan blowing over the electrodes. But this apparatus can't recycle partially-charged air to increase the charge, since the charge is supposed to get totally neutralized inside the vortex. So it needs to develop 5 kV in the air, and get the charge well distributed within the air, so that an even body force develops due to the potentials, and it has to develop that kind of charge in just one pass through the duct(s). So if my suspicions are correct, it might take a more elaborate apparatus just to get highly-charged air out of the bottom platform.

 

The negatively-charged electrodes at the top of the apparatus have to develop the same potential, but it doesn't have to be distributed as a space charge — it just has to create a supply of electrons that can fly off the electrodes and follow the low pressure, high conductivity path through the vortex to the bottom. Still, I don't know how to develop 5 kV at those electrodes, unless I can just solder leads onto the same kind of multiplier circuits that are proposed for the bottom platform that develops the positive ions.

 

Anyway, have a look, and don't scratch your head too much — if something isn't clearly explained, just ask.