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

Re: Streamer V/I evolution - Topics Moderator note



Original poster: "Mike" <induction@xxxxxxxxxxx>

Hi Jim, Terry, List
If you are measuring for the frequency because of this loading Jim speaks of, are you planning to use a guided discharge such as fixed electrodes into glass tubing of length N? I was thinking in not the loading and signal from other discharges could be a problem. The loading is one variable, the other discharge path signal is another, such as mixing in the non-linear discharge.
Many people that use RF driven plasma etchers with automatic matchboxes from 50 Ohm amplifiers into the chamber could tell you of the problems and harmonic generation / mixing. Sometimes the harmonics are so strong the SWR circuit sees it and the auto tune starts doing the drunkards' walk as it tries to resolve via the motorized variable cap and roller inductor. End result is it often wanders so far it goes way out of tune.
Coax lines as stubs have to be placed then to reduce the harmonic signal the detector sees. Sometimes it comes down to placing it on manual and running the tuner until the best dip is seen on the reflected power.
Have a HP 8566B Spectrum Analyzer 100 Hz to 22 GHz; You would be amazed at the mixing in a plasma.
Just thought pointing the mixing issue out might help, as it looks like you are going to maybe measure down to tight scales.
You need to be sure you are reading only the desired signal.
Mike


----- Original Message ----- From: "Tesla list" <tesla@xxxxxxxxxx>
To: <tesla@xxxxxxxxxx>
Sent: Wednesday, July 20, 2005 8:21 PM
Subject: RE: Streamer V/I evolution - Topics Moderator note


Original poster: Jim Lux <jimlux@xxxxxxxxxxxxx>

At 02:16 PM 7/20/2005, you wrote:
Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>

A
I will try to play with this too. My top voltage is monitored remotely, unlike your "dream system", so I really don't have numbers there but the waveforms should "look" right. My tiny sparks are also little...


That's the advantage of looking at frequencies, rather than absolute levels. The calibration of the probe doesn't make as much difference, and the difference in frequency would be quite small, so the frequency dependent aspect won't be as big a problem.

Take a coil resonant at 300 kHz, with a 50 pF total capacitance. Adding 5 pF for a streamer will only pull the apparent resonant frequency down by sqrt(50/55) to 286 kHz. Well.. maybe we CAN detect it.. I was thinking we'd need to measure very small changes (hertz, not kHz)

There's still the effect of the coupling to the primary, and, for that matter the lossyness. A bare primary rings down with a straight line, not an exponential, as I recall, because the loss element (spark gap) is nonlinear.