On Thu, Jun 18, 2009 at 1:23 PM, Kyle Mcallister <[email protected]>wrote:
> > --- On Tue, 6/16/09, John Berry <[email protected]> wrote: > > > > Yes, though Gray has air and hence no inner tube between > > the plasma and the metal terminal. > > What is he using as plasma? Corona? No, A channel of plasma formed by electrical current. > > > > > The outside of the envelope is metal > > > foil. > > That should work however everything indicates that the > > results will be better if wire is used as it's > > capacitance will not be greatly diminished but the contact > > with the escaping charges will be reduced. > > Aluminum window screen then, if foil doesn't work. > > > This part is not a mystery, if the gas is not in a plasma > > state, not a conductor then we have a single plate. > > This raises significantly the voltage and energy of the > > remaining plate. > > I understand that it will raise the voltage, more or less the same as if > you charged a two plate capacitor and dragged the plates apart. As > capacitance diminishes, the voltage on the plates rises. > > But in a situation where one plate more or less disappears, not sure. Have > to think about this. If something unusual *does* happen, well, then things > get very interesting. > > > Please note we want smooth, Stiffler used a medical HV > > source as noisier options with ripples failed. > > Is it necessary to have a smooth DC exciter? Or just a smooth DC charging > supply? A smooth DC sources for establishing the plasma arc current for sure. It may also be required for the capacitor charging source to be well filtered also, the following test could possibly determine this... Establish an arc, measure the capacitance of the capacitor in the plasma solid on state either with a meter or in a more manual way by charging and discharging the capacitor. Then recharge it, disconnect the source that charged it, wait a second, then see if it still retained useful % of it's charge, if it does then we know that everything should be working. > > Suppose we use a tube like a blacklight, a Wood's glass tube with a mercury > vapor discharge inside. No phosphor usually there, AFAIK, so nothing to > cause a problem. I have doubts that the plasma of a fluro is suitible for a swift collapse. The plasma does not die instantly and it can remain semi-energized/ionized. Also I am not sure that the type of glass and whatever else may cut out the more dangerious part of the UV spectrum will not impeed the effect. Which is not to discourage you but don't rule the effect out if that option fails to give satisfactory results as Hiddink said no Fluros. May I suggest Xenons or just custom fill a tube/jar as you were previously thinking? We wrap foil or screen around the glass envelope of the tube, Yes, although think of this like a phansworth fusor, you want the speeding particles to not hit the grid, you would therefore want far more gap that grid, a fine mesh used on windows however is more grid than gap. I would not except it to work on a Fusor and would not think it ideal here. I would personally get some magnet wire and make a grid that more closely resembles the photos of Edwin gray's tubes, do you need a link to one? > this is plate A. Plate B is made when the gas inside is excited to a > plasma. Yes, more specifically a DC arc channel. > A DC potential is applied across this odd capacitor, plasma becoming the > negatively charged "plate", foil/screen becoming the positively charged > plate. But to excite the plasma into conduction, what if we used a separate > 'exciter' supply? A brief HV pulse is used to ionize the gas in the tube, > then a 'keep alive' current flows from one electrode (those built into the > tube from the factory) to the other Sounds good so far. > , keeping the gas ionized. The DC potential is placed then across the > "capacitor", charging it to an energy of 1/2CV^2. At that point, the > keep-alive supply is switched hard-off, maybe via a fast MOSFET. The plasma > is gone, and the effect, if any, happens. We > then repeat the whole cycle. Closer? Yes. > Note that the keep-alive and exciter supply can be made battery operated, > and isolated/floated at HV. Good except if electrons are lost into the enviroment and not replenished the arc will stop being negative. > > > Matter of fact, it doesn't even have to be a Wood's glass blacklight tube, > I reckon a neon tube or bulb would work. That might well be the case. > > > > One thing though, Hiddink said that bulbs with a coating > > would not work probably because the coating acts as a > > capacitor plate? > > Not all phosphors are conductive. Some possibly are, but I'm not certain. > To be safe, I'd start with a blacklight bulb. Behind glass, in case of > excessive UV release. > > > Personally I favor the plasma having an absolute negative > > potential while charged as I imagine electrons being easier > > to have "fly away" than a positive charge, but for > > completeness both should be tried. > > Can do it either way. I tend to make mirror images of most supplies I > build, so if I have something making +10kV, I probably have a reverse supply > somewhere. > > > The metal "plate" will function better if it is a > > mesh or wire grid, we want capacitance but with as little > > coverage as possible. (Stiffler confirmed mesh worked > > better) > > > > Stiffler initially failed to get the effect when using a > > regular HV source and only when he switched to a medical HV > > supply for it's low ripple did he get any results. > > Low ripple is not hard to get, we just need plenty of smoothing caps. They > can be pricey...but in a pinch, can be made at home. > > > Finally we want a very swift disconnection and Stiffler > > seemed to agree with Hiddinks patent that disconnection at > > both ends is desirible to de-excite the plasma fast. > > Too late tonight to go get a Wood's glass tube, but I can grab one on the > way home from work tomorrow, do some tests. Matter of fact, I might try some > things with my neon bulb now... Good luck, a variety of tubes would make sense. There are some large 1KW Xenons to be found online for plausible prices. > > > --Kyle > > > > >
