The Bounce Is Real

In He Chose the Losses the archive put Meyer's invented vocabulary next to the standard terms it replaced, and every entry had an equivalent except one. The Electron Bounce Phenomenon had nothing beside it.

This is that entry, worked out. It turns out to be two standard things at once.

What he says happens

From Electron Bounce Phenomenon:

Magnetic Field Coupling (71) … causes and produces copper ions (643a xxx 643n) (Positive Charged atoms having missing electrons) when moving external electromagnetic field strength is sufficient enough to dislodge electromagnetically charged electrons (641a xxx 641n) from copper atoms making up copper wire material (52).

And out loud, at an evening workshop, more precisely:

the electron of the copper atom, the farthest one away … the external magnetic field, when it's greater in intensity, then the electron will pluck it out, just like voltage will do so. And as a result, you now have created current flow in one direction.

The outermost electron is the conduction electron. Current in one direction is an EMF. That is Faraday's law with the nouns changed — and the same chapter finishes the job: “Once Secondary Coil-winding (52) is de-energized by the removal (collapsing magnetic field during pulse off-time T2) … the dislodged electrons return, terminating and switching off opposite voltage potential.” The voltage exists while the field changes and stops when it stops.

The wire is already what the written sentence asks for

Copper metal is a lattice of positive copper ions in a sea of free electrons. That is the metallic bond, and one choke's worth of wire holds ~5.5 × 10²³ free electrons before anything is switched on. The field cannot ionise it further either: 20 kV along 300 m is under 100 V/m inside the wire, so an electron gains ~2 × 10⁻⁶ eV between collisions against the 7.73 eV needed.

But it does not need to. It only has to push the ones already there.

How much charge clusters

Q = CV, and the only C is the coil's own — which has been measured. Don Gabel's 2009 readings of an estate VIC give each choke 158 and 180 pF on the core at 10 kHz. At 20 kV that is 3.2 µC: about one conduction electron in thirty billion.

They do bounce

In alternating current the electron sea does not travel. It is pushed, the drive reverses, and it comes back. At 1 mA and 10 kHz an electron moves 0.20 nanometres — about four fifths of one copper atom — and returns. Meyer's word is the right word; the distance is the surprise.

(In 29 AWG copper, which is what every winding of the measured transformer is, 1 mA drifts at 1.2 µm/s. An electron takes about seven years to cross one choke. The signal crosses it in under two microseconds.)

Where the voltage really comes from

The kickback is real — V = L·di/dt — but it is bounded, because a real coil rings into its own capacitance instead of climbing forever:

V_peak = I × √(L/C)

For the measured choke, √(1.22 H / 160 pF) = 89 kΩ. So 10 mA gives 890 V, not 25,000; you need ~225 mA to see 20 kV.

And Electron Bounce Phenomenon names a second source in its first paragraph, which almost nobody quotes: the flux from the closed core penetrates all the windings, and the potential equals “the sum of Voltage Potential (Vp) developed across each Pickup Coils (VpT + Vp1 + Vp2)”, with Eq 30 — Ltcc = L1 + L2 + 2M — under it. Three windings on one core, phased to aid.

Which wire are you winding?

source	the choke
memo 426 (Circuit Resistance)	430F/FR stainless, 36 AWG, 11.6 kΩ
memo 429 (Propagating Electrical Stress)	“two copper wires Bifilar wound”
memo 429, fig 10-4 (Voltage to Amp Differential Ratio)	copper stages and stainless stages, stainless nearest the gap
the estate transformer (VIC coil readings)	29 AWG, 76.7 Ω — copper

Figure 10-4 reconciles the memos: it was never copper or stainless, it was how many turns of each, "to obtain optimum Voltage to Amp Differential Ratio". And the only device anybody has metered has no stainless stage at all.

What the supply says, and how to read it off a meter

Bouncing is free; resistance is not. Both scale as I², so the ratio is fixed — and it has a name your LCR meter already prints:

per-cycle resistive loss ÷ peak stored energy = 2π / Q

coil	Q at 1 kHz	ratio	heat at 10 mA
memo 426's stainless (11.6 kΩ, 250 mH)	0.14	the wire burns 46×	2.3 W
the estate choke, on the core (Gabel)	69.5	the bounce carries 11×	0.022 W

Same claim, same arithmetic, opposite answers. The thing that separates them is which coil is in your hand — and you do not need this table, because you can read Q off your own.

One more thing the 2009 readings give you

The estate choke measures 1.22 H and 158 pF on the core, which puts its self-resonance at about 11 kHz — which is why Gabel's meter printed "n/a" for the inductance at 10 kHz. Meyer's stated band is 1 kHz up to and beyond 10 kHz. That is exactly where this coil's impedance peaks, and so where it inhibits amps best. His band is the coil's band, and it needs no theory about water to explain it.

See Q factor and voltage rise and VIC coil readings for the whole workbook.