Resonators in enclosed spaces / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Resonators in enclosed spaces

1 2 3 next

Move Thread

LAN_403

ragnar

Fri Jul 13 2007, 10:56AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I'm hoping to gain some insight into the use of air-cored resonators (whether they be tesla coil secondaries, parts of impedance matching networks, chokes, whatever) in enclosed spaces like metal boxes and near metal sheets.

I understand that the Q of an inductor will get lower when in the proximity of a big conductive sheet; how then, would it be possible to shield a tesla coil, without severely inhibiting the resonant action?

Does the frequency of the coil affect the proximity needed for such a metal sheet to have a detrimental effect to its Q?

Will a tesla coil with a groundsheet directly underneath it perform worse than a tesla coil tied to a groundsheet a farther distance away? What effect does inductance between the base of the coil and the RF ground have on the coil?

Curious,
Matt

Steve Conner

Fri Jul 13 2007, 11:19AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

If the material that the shield is made from is a good conductor, like copper or aluminium, it hurts performance less than you'd think. It just lowers the inductance of your coil somewhat. This does decrease the Q a little because the inductance was lowered but the resistance didn't change.

Radio equipment uses inductors shielded in copper and aluminium cans all the time, and radio manuals recommend that the shield shouldn't fit the coil too snugly: if the metal gets close enough, bad things will eventually happen. (If you made a shield that fit perfectly, it might reduce the inductance to near zero!) IIRC they recommended a clearance equal to the coil radius on all sides in one text, but I could be wrong.

For a Tesla coil secondary, the capacitance between coil and shield will lower the resonant frequency somewhat, probably more than the decreased inductance raised it. This may lower output voltage. However, I remember WaveRider reporting that his Class-E coil gave bigger streamers when it was enclosed in a wire mesh wastebasket. I guess he got benefit from keeping his E-field in a cage and not letting it spread out into the room and incur "couch losses"

The logical conclusion of all this is the cavity resonator, where the single-turn coil, shielding box and tuning capacitor are all mixed up into each other. These work great up into the microwave region. But they're still not that far different from a Tesla coil in a box...

ragnar

Fri Jul 13 2007, 11:43AM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

I guess most of the inductors shielded in copper and aluminium cans I'm thinking of usually have ferrites in them, but you're right. In the case of a small coil, say, 1"dia x 2"high, having a square-foot groundplane directly underneat the coil kills ANY possible breakout.

If I wanted to shield a coil, should my casing be closer or further away depending on frequency? :P

Dr. Shark

Fri Jul 13 2007, 11:56AM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

Funny how this relates to the stuff I just posted in the EM guns section. I made some measurements on how the inductance of a coil is affected when I put copper plates, aluminium rings etc. on top of it, and in a pinch it reduces the inductance by about 1/3. However once the metal part is about as far away as the radius of the coil, the effect becomes neglible.

I'd also be interested to know how (if?) this effect depends on frequency, for some reason I expect high frequencys to couple better through a large distance of air, but that's just a feeling.

Marko

Fri Jul 13 2007, 01:17PM

Registered Member #89 Joined: Thu Feb 09 2006, 02:40PM
Location: Zadar, Croatia
Posts: 3145

Oi

I was also confused and misinformed until recently.

This is why I did these little experiments.

Royer oscillator with two IRFZ44's was running at about 1,5Mhz into a small air-coil quite well. I could wirelessly power a small 5W bulb to few centimeters of distance.

When I added a copper rear shield, the bulb glowed brighter. I was using no resonant cap at all at receiver side so rear shield must have *increased* coupling with it's presence. Shield itself shifted thefrequency only a little, for few tens of kHz.

So *DO* put a solid rear shield underneath your coils. It will increase overall primary-secondary coupling, and put more power into secondary (sort of 'good' decrease in Q). It also acts as an excellent low impedance ground point.

Using a side shield will confine the magnetic field and even further increase the coupling. But, I think that at that point you'l run into a problem Steve already mentioned, that will also greatly increase your secondary-ground capacitance and hurt Q much more than benefit it.

If you want side shields give them reasonable clearance from the secondary. If you think they are needed at all.

Dr. Shark

Fri Jul 13 2007, 01:54PM

Registered Member #75 Joined: Thu Feb 09 2006, 09:30AM
Location: Montana, USA
Posts: 711

No f**** way! I don't have much understanding of RF and magnetics, but a conductive shield increasing coupling is just counter to all my beliefs! While the shield will in a certain sense "reflect" the electromagnetic waves impinging on it, it will do it in such a way as to cancel out the impinging field. Unless there is a funny phaseshift somewhere, the field reaching the receiving coil should be made smaller, not bigger.

Are you sure that the frequency shift, even if it is small, does not affect the driver circuit? Both halfs of the center-tapped driver coil now have a different inductance, maybe that has some funny effect on the driver?

Hazmatt_(The Underdog)

Fri Jul 13 2007, 06:07PM

Registered Member #135 Joined: Sat Feb 11 2006, 12:06AM
Location: Anywhere is fine
Posts: 1735

Steve: If you place a target in proximity to the Tesla coil you tend to get longer connecting arcs. I noticed this when testing my systems. I was getting about 18" of weak arcs without a target, and connecting strikes at 25" with a pole tied to the coil return lead and ground plate. And of course the hottest spot was in the center.

I'm playing with the mutual inductance and coupling right now and what is interesting is if you avoid both extremes ... 0.1 and 1.0 and go closer to k = 0.3 -0.4 the resonant system will peak.

I will try to see if the frequency has a larger impact on tuned coupling in the system. Right now I can only go to about 10khz because of the crummy amplifier chip in the power amp in my test setup.
This setup is really for TC coupling and Q evaluation, but I'm going to cover what I can in time.

Carbon_Rod

Sat Jul 14 2007, 08:50AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

BlackPlasma,

There are numerous variables you are inquiring about.

Typically,
a. Silver plated RF boxes are designed to quench surface based noise.
b. Ferris metal shields work better for LF audio (toroids can be better.)
c. LF Mechanical oscillation prevention with epoxy etc.
d. Air cores can be unstable for obvious reasons...
e. UHF boards use copper/silver plumbing...
f. Uncommon grounding points have been breaking stuff for years.
g. If the box is attenuating it may not always be bad.... only 98% of the time...(the f**** is in bios Area?)

Any Hints? ( wavelength or band of interest could help...)

ragnar

Sat Jul 14 2007, 02:19PM

Registered Member #63 Joined: Thu Feb 09 2006, 06:18AM
Location:
Posts: 1425

Thanks for your contributions, guys.

I'm curious in a strictly tesla coil context where I am faced with three choices:

A) ground the base of the secondary
B) connect the base of the secondary to a metal sheet
C) connect the base of the secondary to a metal sheet, and ground the metal sheet

The frequency band of interest is 5MHz - 15MHz.

How about a bipolar coil? Should the center be grounded?

I guess I come across as a little too vague, because I do want to understand more from a shielding perspective, but also from a coil performance perspective -- I'm interested in power radiated vs power absorbed, vs power converted into plasma.

Dr. Drone

Sat Jul 14 2007, 03:13PM

Registered Member #290 Joined: Mon Mar 06 2006, 08:24PM
Location:
Posts: 1673

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob