[Prev][Next][Index][Thread]
Re: Beginners Tesla Coils
Tesla List wrote:
>
>
>
> Air capacitors don't get very hot and really shouldn't need air cooling
> untill you get to multi-kilowatt power levels.
Shouldn't ever get hot at all, except from ohmic heating of the plates..
> The reason is that the
> dielectric air will just rise away from the plates if it gets hot.
> Here are the main disadvantage: Air has a dielectric constant of 1.
> Therefore an air capacitor will be 4 times larger than a glass one or
> about twice as large as a PE cap (probably more since you can't roll it.
A problem, but not the big one...
> I'm not really sure about the losses of air in the RF range, but I'm
> pretty sure they are higher than PE.
To the contrary, air has a lower loss than just about anything, except a
vacuum. This is why they use things like waveguide and air insulated
coax for the ultimate in low loss.
Plus, making sure that the plates
> are exactly at the same distance all around is quite difficult and you'd
> need hard plates (steel) that would be much heavier than the aluminium
> flashing all of us use (not to mention more expensive and harder to work
> with).
A mere mechanical detail..
Actually the real reason not to use air is that it's breakdown is only
71 volts/mil, as compared to typical values of 200-500 volts/mil for
plastic. You want your caps to have a breakdown voltage of something
like 20-30 kV, which works out to an inch of air gap, and only 60 mils
of plastic. To get a decent amount of capacitance with an air insulated
cap would require positively huge plates (many, many square feet), and,
while one could actually build such a cap, the large size of the plates
would make them have a lot of parasitic inductance, as well as series
resistance (unless they were superconducting), which causes all sorts of
other problems.
Now, if you put the air at a substantial pressure, like 7 atmospheres
(around 100 psi), then its breakdown would be comparable to that of
plastic. This is how you make a very low loss capacitor with reasonable
capacitance and high voltage (used as such to measure very high
voltages). The problems of containing the air, and the fact that you
still need a substantial plate area to get a decent capacitance, mean
that rolled up polyethylene and alumininum is the best all around
solution.
Using a dielectric that is solid (even if porous) helps maintain an even
spacing of the plates, which is important for reliability (the closer
spots would be higher electric stress). The classic capacitor is made
with paper and aluminum, then impregnated under vacuum with oil. The oil
is the dielectric, the paper is just there to space the aluminum apart.
The vacuum makes sure there isn't any air in the gaps.
Good ideas though.....
--
Jim Lux Jet Propulsion Laboratory
ofc: 818/354-2075 114-B16 Mail Stop 161-213
lab: 818/354-2954 161-110 4800 Oak Grove Drive
fax: 818/393-6875 Pasadena CA 91109