[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

microwave ovens Re: Ball lightning - Terry's thoughts....

Original poster: Jim Lux <jimlux@xxxxxxxxxxxxx>

At 10:18 PM 8/11/2005, Tesla list wrote: 
Original poster: Scott Stephens <scottxs@xxxxxxxxxxx>

Tesla list wrote:

some physics student claimed that microwave ovens couldn't produce corona,
since 500 watts RF doesn't give a wave with a high enough voltage to
trigger sparks.

To get corona, all you need is a local field > 30kV/cm, which is pretty easy to achieve with conductive stuff in the cavity. Crumple some aluminum foil and throw it in the microwave, and you'll see plenty of little sparks where the local field exceeds the limit.


 Ah, but what if we aren't cooking a roast? What if we
have a high-Q resonant waveguide? In that case we might only inject
500watts... but inside the waveguide we'd have 100.5 kilowatts going out,
and 100 kilowatts bouncing back again, creating an immense standing wave
and huge electric currents in the ground connection, but with only
500watts being absorbed in the conductors.


Two things to consider before you discount what your physics student:

1. The amount of resonant voltage increase depends on the loss (often derivable from the Q) of the cavity.

2. The impedance, moreover the capacitance of the waveguide will determine the impedance and voltage at a given power level.

So:

1. What is the impedance of a microwave oven cavity?
2. What is the Q of that cavity?
The Q is quite high (given that it's a metal box, and doesn't get hot in normal operation, so the losses aren't all that high.. dissipating 10W would make the walls get noticeably warm), but the mode structure tends to be quite complex.

There's more to peak E-field than just Q though. You need to know the mode distribution.


MO frequencies are 2 to 3 orders of magnitude higher than TC frequencies.

Try more like 4-5 orders of magnitude... TC at 250kHz, (2.5E5) and MW oven at 2.45GHz (2.5E9)


For comparable Q's, (IIRC single-mode waveguide filters have Q's in the 100's) I'll SWAG the resonant voltage increase might be 10 to 30 times the nominal value, up to perhaps 1000 volts. Not enough for corona in air. Superconductive filters for 800 MHz cell phones might have Q's 10,000 - 100,000, IIRC.




Since MO plasmas usually require a conductive carbon or metal, or ionization from the short UV from an external arc source, I suspect your physics student is right about kWatt-class MO's inability to ignite a plasma in air, even if argon. Perhaps somebody should do some math?

It's well known that you can ionize in Ar as well as other gases (at lower than atmospheric pressure) in a conventional microwave oven. There's an article in a Rev Sci Inst about 20 years back describing just such a modification. There's also a commercial plasma etcher (the PlasmaPreen) which looks for all the world like an upside down pyrex baking dish inside a microwave oven.

But even at atmospheric pressure, it's remarkably easy to ionize air with fairly low RF powers. All you need is sharp points to help the process along. The voltage doesn't need to be all that high (i.e. you don't need to be above the minimum sparking voltage for the gas), just the field has to be high enough.

For what it's worth, in microwave waveguide breakdown testing, you use a waveguide that has a small height (while keeping the width basically the same).