secondary resonance vs working frequency of the secondary coil / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

secondary resonance vs working frequency of the secondary coil

1 2 next

Move Thread

LAN_403

IamSmooth

Sat Nov 11 2006, 01:03AM

Registered Member #190 Joined: Fri Feb 17 2006, 12:00AM
Location:
Posts: 1567

I was reading through a Tesla Coil design paper and it distinguishes between self-resonant frequency and working frequency. The Fres is the natural frequency for the L and C values of the secondary. It describes the working frequency as being based on considering that the secondary coil's wire length is 1/4 of the wavelength (wavelength = c/f; wirelength*4 = wavelength).

My question is to which frequency does one tune the primary LC circuit; or, should the working frequency and self-resonant frequency be the same?

Marko

Sat Nov 11 2006, 01:15AM

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

We simply forget 1/4 wave thing since TC is not an antenna.

Atleast I do so, although I see some people hunting for their secondary wire length to be closely 1/4 of operating wavelength for reasons I can't understand.

Since wire is coiled up and it's business is not to radiate EM from all I figured out it doesn't have any effect on TC operation.

Hazmatt_(The Underdog)

Sat Nov 11 2006, 03:33AM

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

It's really hard to get the 1/4 wave to match the particular secondary because of self capacitence. Then you're changing the system again with the topload.

What you're looking for is basically:

LpCp = LsCs

I knew where the secondary performed best from a few previous runs, so I had a ballpark figure. I had to change some of the wiring on the primary side. I measured the primary side wiring and coil, then adjusted so the total was Lp. Afterward I varied the inductance slightly and the performance dropped off, so I knew I was in the right area.

So basically, if you know your sum Cs, Ls, Cp, you can measure your connections with a meter, adjust, and you get really close ;)

Matt

***edit**

You know, there is one thing I forgot to mention, and its exactly what we're talking about in Phyisics right now. Your wavelength calculation is wrong actually, because there is a transmission impedance of the wave. What this means to you is that c, the speed of light, is not actually the speed of light. The wave actually bounces back and forth inside the conductor as it propagates.

conductor
--------------------------- --
/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ --------> wave propagation
------------------------------

Now what you would have expected if your wave speed were actually c would be:

conductor
----------------------------
------ --------------------------> wave propagation
----------------------------

The permitivity of the conductor 'slows' the wave speed down, and if you look this up in a Belden catalog its something like .66 * speed of light, and that also varies between cables too.

so f = c/v; yea, for light and ideal cases, but not so much here.

WaveRider

Sat Nov 11 2006, 10:11AM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

The TC resonator has a lot in common with the 1/4 wave helical resonator. However, being a close wound helix with dimensions much smaller than a wavelength at the operating frequency, it does not radiate very much. Another consequence of the close windings is that the physical length of the wire in the coil is longer than 1/4 wavelength, even though the magnification effect at the open end is alot like the magnification one gets at the open end of a 1/4 wave transmission line. This is because the turns are closely inductively/capacitively coupled. Likewise, if the turns are not so close (loosely wound helix), the electrical length of the helix winding wire approaches 1/4 wave at the fundamental resonance...and if the physical dimensions are a significant fraction of a wavelength, you will get radiation in an open structure!

Tuning the primary gives you a coupled tuned circuit. The coupling gives you a so called frequency-splitting effect. There are two resonances that move around as you tune the primary and vary the coupling. The maths is straightforward, but can be a little tedious.. Do you have a good circuits book? I also believe the ARRL handbook talks a bit about doubly-tuned circuits...

Sulaiman

Sat Nov 11 2006, 10:21AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

TC programs often give the frequency at which the inductor alone is self resonant
'Wheeler' Inductance in parallel with 'Medhurst' Capacitance
This frequency will be well above the 1/4-wavelength frequency for normal coils (H/D = 2 to 5)
Adding a topload capacitance reduces this resonant frequency
A toroid of similar dimensions to the inductor bent into a circle
brings the resonant frequency close to 1/4 wave
An even larger topload can be used for operation below 1/4 wavelength
with a lot of charge available for streamer growth.

Operating without a topload produces the highest voltages part way down the coil from the top
(causing corona/sparks to come from the winding)
operating at 1/4 wave or lower frequency ensures the highest voltage is at the topload

Marko

Sat Nov 11 2006, 12:57PM

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

Operating without a topload produces the highest voltages part way down the coil from the top
(causing corona/sparks to come from the winding)

No, it simply can't be true.

In any LC, peak voltage is always going to appear between 'ends' of the coil itself, and I think it's pretty obivous thing.

Some people obivously aren't even sure if it's good or bad to have 'omg 1/4 wavelength' secondary and nobody explained yet how it really affects coils in any way.

Does it really *mysteriously* teleport peak of voltage to somewhere else?

Or causes secndary to emit more magnetic component like antenna, wastng power?
Or soemthing else?

All that about 'omg 1/4 wl' seems like having no physical basis and seem to be just placebo statisfaction for some coilers, while others seem to spark very well without even mentioning it.

I feared to open such a topic before but now it seems that this needs to be solved right now for sake of 4hv.

Due to my ignorance I need something to be step-by step explained, and it doesn't seem possible when opinions are divided like such...

So, again, why to tune a secondary coil like it's some kind of 75 ohm antenna (how is it important, and why?) while lots of other coilers (steve ward, conner, TDU, EVR ...) don't mention a thing about it and are still fine?

WaveRider

Sat Nov 11 2006, 01:28PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

So, again, why to tune a secondary coil like it's some kind of 75 ohm antenna (how is it important, and why?) while lots of other coilers (steve ward, conner, TDU, EVR ...) don't mention a thing about it and are still fine?

Speaking from the viewpoint of a CW coil: The arc at the top of the coil behaves like a resistive load(with a bit of capacitive loading too). If you want to get the best transfer of power into that arc, you need to "match" the driving amplifier to the coil.

Before the arc is struck (before breakout), the coil behaves as a resonator loaded only by its internal losses and the driving circuit. Resonator Q is quite high (several hundred or so). Power from the driving amplifier (or switch) goes into building up the resonant oscillations of the resonator and some is dissipated in the circuit resistances. When the oscillations build up enough, the air around the breakout point breaks down. This appears like extra loss in the resonant circuit. Since the plasma is conductive almost like a metal, extra capacitance appears...lowering the resonant frequency a little (this is easily verified with a CW coil). Since we have this plasma sucking power out of our resonator, we need the amplifier to supply power in the most efficient way to the resonator....this means an attemt at matching. This is difficult because the plasma is not a constant impedance...it varies with power level and time (it "flickers" like a flame).. The best hope is to pick a power level and design a matching network around the expected feedpoint impedance for that power level (arc size). This is what I did with my 250 W CW coil and I can say with confidence that it works!

With pulsed coils, the design criteria are somewhat different. The idea seems to me (and the Steves, EVR, et al can elaborate on this) to dump as much energy into the coil resonator as quickly as possible to get the big sparks. Here the idea of matching in the CW sense does not work, because you are talking about dumping energy in a short pulse with as little pulse spreading as possible so you get very high voltages with sudden breakout at extreme peak powers. In short, the design goals are different...

Operating without a topload produces the highest voltages part way down the coil from the top
(causing corona/sparks to come from the winding)
operating at 1/4 wave or lower frequency ensures the highest voltage is at the topload

As Firkragg adamantly and correctly states, this cannot be true. The fundamental resonance of the TC without the topload is the 1/4-wave like (even tho' we all know the winding is longer than 1/4 wave) mode with (nearly) max current at the base (ground-plane) and max voltage at the top. Adding a topload reduces the frequency of resonance. The max voltage will still be at the top of the coil, altho' the "magnification" effect will be somewhat less. (consider the voltage magnification in a resonator formed using a less-than-1/4-wave length of transmission line with a capacitor at the end compared to the pure 1/4 wave transmission line.) Theoretically, a large number of modes can be excited on a helical resonator. We all shoot for the fundamental mode which has a voltage peak at the top. . Only if you excite a "higher-order" mode on the helix will you get voltage peaks "in the middle" somewhere. This happens if you choose an operating frequency higher than the fundamental (first) resonant frequency. However, sparks can appear in the middle of you resonator for a variety of reasons (e.g. insulation breakdown, etc.)

Marko

Sat Nov 11 2006, 02:17PM

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

The first part of your text (about loaded and unloaded impedances) is familiar enough to me - that's etc. why class E coils are so difficult anyway (very narrow frequency band and sensitivity to load impedance changes).
I wasn't asking about secondary and spark impedances anyway, but what tuning to 1/4 wl frequency helps system in any mean!

It has been explained on richie burnett's site niely long ago (site is now down, sadly

)

On his site, the thing aboul 1/4 wl also wasn't mentioned anywhere IIRC.

The fundamental resonance of the TC without the topload is the 1/4-wave like (even tho' we all know the winding is longer than 1/4 wave) mode with (nearly) max current at the base (ground-plane) and max voltage at the top. Adding a topload reduces the frequency of resonance. The max voltage will still be at the top of the coil, altho' the "magnification" effect will be somewhat less. (consider the voltage magnification in a resonator formed using a less-than-1/4-wave length of transmission line with a capacitor at the end compared to the pure 1/4 wave transmission line.) Theoretically, a large number of modes can be excited on a helical resonator. We all shoot for the fundamental mode which has a voltage peak at the top. . Only if you excite a "higher-order" mode on the helix will you get voltage peaks "in the middle" somewhere. This happens if you choose an operating frequency higher than the fundamental (first) resonant frequency. However, sparks can appear in the middle of you resonator for a variety of reasons (e.g. insulation breakdown, etc.)

Yeah, I agree that.

And from all I know, tuning the coil's 1/4 wl frequency to it's LC resonant frequency practically has no effect on TC operation.

WaveRider

Sat Nov 11 2006, 02:41PM

Registered Member #29 Joined: Fri Feb 03 2006, 09:00AM
Location: Hasselt, Belgium
Posts: 500

Firkragg, I think the confusion comes about because the first resonance the TC resonator behaves a bit like a 1/4-wave transmission-line resonator, even tho' it is not really the frequency corresponding to the "1/4 wavelength" frequency, as you put it.. .

In fact, I've done the simulations which clearly show the current distribution along the coil windings. Qualitatively, they look conspicuously like the current distribution on a 1/4-wave transmission-line resonator... Unfortunately, no one seemed to be too interested in the results to discuss them much....

Terry Fritz

Sat Nov 11 2006, 08:11PM

Registered Member #393 Joined: Tue Apr 18 2006, 12:30AM
Location:
Posts: 297

Hi,

In fact, I've done the simulations which clearly show the current distribution along the coil windings. Qualitatively, they look conspicuously like the current distribution on a 1/4-wave transmission-line resonator... Unfortunately, no one seemed to be too interested in the results to discuss them much....

Paul Nicholson has done many studies of secondary resonators too:

I have done it the hard way

I trust all these simulations by now independent studies are giving the same answers?

Programs like JAVATC can calculate everything for typical coil use to very high accuracy now:

Nobody bases anything on "wire length" these days. It does not matter. The secondaries coil's inductance and self capacitance and the added capacitance of the top load rule the resonant frequency and that can be calculated by computer.

Cheers,

Terry

1 2 next

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