User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.9; rv:45.0) Gecko/20100101 Thunderbird/45.7.1
St. Elmo's Fire isn't really a corona phenomena, at least how we all
think of corona as discharges from a body at a high voltage.
If you read descriptions of St. Elmo's Fire, it's more like a glowing
sheath around wooden masts and spars during a thunderstorm.
A colleague of mine, Stephen Fuerstenau, explains that St. Elmo's is
actually a much more interesting phenomenon.. It's a combination of
electrospray and corona breakdown on the droplets being sprayed.
So here's what's likely going on - the surface of the object is covered
with water (often salt water, but water that has some conductivity in
any case). The water is charged, along with the object it's on. Small
irregularities in the surface of the water causes droplets to be "spit"
away (think of the demo where you put a bucket of foam peanuts on the
topload of a Van deGraaff generator). As a bump develops in the surface
of the water, the surface charge is repelled by the surface charge on
the surrounding water, which causes the bump to get bigger, stretch, and
eventually form a separate droplet, which is then repelled, just like
the foam peanuts on the VDG.
As it happens, the surface charge in this situation creates a field that
is right at the limit for breakdown (the charging of the big object
ensures this). As the droplet moves away, the surface field actually
increases because of two reasons: 1) the potential difference between
the droplet and the surroundings increases... if the object from which
the droplet left was at 20kV, and the droplet starts out at 20kV, the
potential difference is small, and as the droplet moves away, the
potential difference increases; 2) the droplet evaporates, making it
smaller in diameter, but the charge remains the same, so the field goes
up. Both cause a surface breakdown (corona, if you will), and the
droplet is now glowing.
This phenomenon can occur over a wide area, and on things that aren't
particularly "pointy".. the small radius of curvature is in the
droplets, not the big charged object.
An experiment YOU can do if you have a suitable HV DC supply (it might
work with a NST's AC output, but I've not tried it).
Set up a metal pan with a thin layer of water in it - that's your ground
plane. in the middle of the pan, put a cork or rough block of wood that
just sticks up a bit above the level of the water, making sure the
cork/wood is "wet". A nice dense sponge might also work.
Have some insulating supports, and put another flat electrode (another
cookie sheet or pan?) above the bottom and put a voltage between the
two. If the field is high enough, you'll see the water come up the wood
and form tiny droplets (a bright light from the side makes it easy to
see them). get the electrospray working first.
Then, turn out the lights and behold your St. Elmo's fire generator.
There are some fiddly bits here - if there's much air movement it blows
the droplets away - you'll likely get some flashover arcs when you're
setting up - it *is* HV - resist the temptation to reach in and move
things when the power is on.
I used a 50kV DC supply with a big resistor in series to limit the current.
I think it might work real well with a Van de Graaff generator with a
suitable setup.
I've not fooled with changing the salt content of the water or a variety
of other factors - this would be a "most excellent" science fair
project, by the way.
The wikipedia article is kind of ok, but has some incorrect data (air
does not break down in a field of 100kV/m, more like 3MV/m. near small
radius of curvature objects, the field *is* >3MV/m, which is why it
breaks down)
However, the description from Dr. Braid about the glowing horse ears and
hat brim is a good one.