From: spam_vanishes_here@yahoo.de

Newsgroups: sci.physics.plasma

Subject: Polarisation of noise from gas discharge

Date: 2 Jan 2005 08:03:08 -0800

Organization: http://groups.google.com

Hi,

I am not an expert in plasma physics, so sorry if this is a trivial

question for some of you.

While experimenting with the LNB of a satellite receiver and a "sat

finder" which indicates the received signal strength, I found that the

noise that is emitted by fluorescent tubes ("neon lights") in the 10

GHz microwave region is strongly polarised perpendicular to the

direction of AC current flow through the tube.

Naively, I would have expected that the radiation is either unpolarised

or, arguing by symmetry, that the electric field component of the

radiation is parallel to the electric field that drives the current

through the gas discharge. To my surprise, experiment shows that the

opposite is true.

I made sure that the LNB is set to the correct polarisation. The

polarisation of these devices is normally controlled through the

voltage supplied by the sat receiver to the LNB. There is a standard

which specifies that vertical polarisation is used when the voltage is

below 15.5 V, and the LNB has to switch to horizontal polarisation at

voltages above that (typically 14 and 18 V are used). I used both

settings of the LNB with the same result (i.e. polarisation of

radiation is perpendicular to the AC current which is along the axis of

the tube). I also checked the effect of wires (a single wire and a wire

grid) that are brought between the source and the LNB. The signal is

attenuated when the wires are parallel to the nominal polarisation of

the LNB, while they have little effect when oriented perpendicular to

it. This independantly confirms the assignment of polarisation

directions.

I performed the measurements on conventional fluorescent tubes as found

in many households as well as one of the newer "energy saving lamps"

which fit into the socket of a regular light bulb but actually use a

gas discharge as well. The signal level read from the meter of the "sat

finder" is not really a linear measure, but the difference in signal

strength seen when rotating the LNB (or the tube relative to it) is

very significant and may even be an order of magnitude.

So my question is: What is the theoretical explanation for the strong

polarisation of the emitted microwave noise, and why is it polarised

perpendicular to the current flow of the gas discharge?

Joachim Schueth