[ §1 · resonator ]

Why a QCW adds capacitance instead of removing it

Every other coil treats topload capacitance as a cost. This one goes the other way, and putting a bigger toroid on is a three-dimensional move, not a one-dimensional one.

CommonGnd4 August 20264 min read

Every other kind of coil treats topload capacitance as a cost. A QCW goes the other way and adds it, sometimes inside the secondary itself.

Mads Barnkob: this is why Steve Ward added a series of small capacitors inside his secondary coil. His first QCW versions simply needed more self-capacitance to withstand detuning from the long sparks.

The reason is the thing the whole department is organised around. A metre of arc detunes the coil catastrophically, and the larger the resonator's own capacitance, the smaller a fraction of it the arc's capacitance is.

Where each kind of coil sits

Slayer exciter, a few watts: nothing to detune. A point, and nothing else.
Burnett's HFSSTC, 500 W: the frequency is near enough stable. A breakout bolt, no toroid at all.
QCW throwing a metre: catastrophic detuning. A large toroid and capacitors inside the secondary.

The topload is a trade in every case. You pay in volts, and volts are what make the arc long. What differs is how much frequency stability you are buying with them, which is the general version of this.

"Just make the toroid bigger" is three-dimensional

It looks like a one-move fix. Run the numbers for doubling the topload on an unchanged secondary:

As built. The resonance runs on Ces of 14.031 pF, of which the toroid is 10.879: 426.5 kHz with no arc, 355.1 under a 1.5 m one, so 16.7 per cent of relative detuning.
Toroid doubled, which adds one more toroid's worth and takes Ces to 24.91 pF: 320.1 kHz with no arc, 286.4 under the same arc, so 10.5 per cent of detuning.

Both lines are at one arc length, 1.5 m, because that is where the arc's capacitance was calibrated — 4.14 pF per metre. Quoting a detuning figure without the arc beside it is how the number that used to be on this line went wrong. The trajectory and the withdrawal are on one page. And that calibration is one coil's, not a constant. Measurements on four put a metre of arc between 3.3 and 5.0 pF, with 4.14 in the middle, so both lines above move by a few per cent across the range — and any arc length read off them moves by half a metre. The spread is worked through on the page that owns the trajectory. What the spread does not touch: the arc adds capacitance, that capacitance pulls the secondary down, and the pole you are switching follows it. That holds at every capacitance in the range, and it is the whole argument of this page. Only the decimals are soft. And note which capacitance the resonance uses. It is Ces, not the toroid alone: L from 426.49 kHz and 14.031 pF is 9.9251 mH against JavaTC's printed Les of 9.925. Working from the toroid's 10.7 pF is the other half of how that number went wrong.

The sensitivity to the arc improves and the absolute frequency at the end of the ramp gets worse. And for straight swords it is the absolute frequency that matters, because frequency is what buys straightness.

Compensate with turns, then. To hold 426.5 kHz at a Ces of 24.91 pF the inductance has to come down to 0.563 of what it was, which is about 25 per cent fewer turns. But:

Z_secondary = sqrt(L_sec / Ces)
  as built           26.6 kOhm
  doubled, retuned   15.0 kOhm

and a low secondary impedance means a higher peak current and a shorter ontime, which for a QCW with a long ramp is a movement in exactly the wrong direction.

And still not by winding

Whatever a QCW needs, it does not get it from the secondary's winding. Layers in grooves put hundreds of volts per turn across thin varnish at the transitions. Layers back and forth end up with more interlayer capacitance than the topload has. Ferrite in the upper half conducts well enough to short the coil out by sixty degrees. And thicker insulation without potting gives corona, which eats the insulation.

The principle underneath all four is the same one as above: capacitance won in the winding comes with impedance lost, and low impedance needs more current for the same voltage.

Capacitors placed inside the secondary as discrete parts, which is what Ward did, are a different thing from winding for capacitance. They add capacitance without adding interlayer voltage stress.

The toroid on the QCW diagram is drawn wider than the other departments' for this reason, and its short secondary is the other half of the same decision: the ramp does the work a taller coil would otherwise have to.

toploaddetuningQCW