User-agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:24.0) Gecko/20100101 Thunderbird/24.1.1
On 12/6/13 12:53 PM, Steve Ward wrote:
Hi Jim,
Just a few comments/questions.
bear in mind that tesla coils run at very low frequencies compared to HF
radio. Folks who come into this from the ham radio side get all excited
about "that long ground wire", because of their experience with HF antennas
.. but that long wire is a tiny fraction of a wavelength at 100kHz (lambda
= 3km).
Most experiences with EMI and TCs (at least for solid-state driven types)
suggests that the fundamental frequency is not where the trouble is. That
is, when the TC is just sparking into free air there usually isnt a
problem, but when the sparks find a large charge-accepting body (not even
necessarily ground, could be a large isolated object) then higher
frequencies are generated. I measure big spikes in the 5-10MHZ range, but
also the entire spectrum above that is elevated, so 10s of MHz easily...
Any ideas on dealing with that?
I think that's more like the EMI from lightning.. you've got a pulse
with a rise time in the "few tens of ns" range and that has strong
spectral components there.
The question would be: is the EMI getting to the victim by radiation or
conduction? If it's by conduction, then filtering "wires" that lead
to/from the coil would help, and for 1 MHz on up, conventional powdered
metal/ferrite cores are probably the right way to go. I use a lot of
2.4" diameter toroids in the 31 mix from Fair-rite. It has good
resistance and inductance over a pretty broad band in HF, unlike the
ones you see used on computer cables, which tend to be optimized for
higher frequencies.
Terry Fritz found significant emissions in the VHF range for spark gap
coils. For that, it's probably radiating from the wires going to the gap.
And what you really don't want to do is have your coil inside, and ONLY
grounded to a stake in the ground outside. Now the RF path *is* going
through the building.
I believe that anything near a TC becomes part of its "RF" circuit just
because of mutual capacitance with the TC itself. So being near the very
high dv/dt of a ground-sparking TC can induce currents in un-intended
paths, and your grounding system may not make enough difference to avoid
this problem. An electrostatic shield (not truly a faraday cage) often
fixed this problem. Sometimes your ground plane should go vertical so as
to capture more of the TCs field.