Konstantin Meyl's potential vortices, scalar waves and "neutrino power": a primer

Project folder: Researchers/Meyl-Konstantin. It holds 24 documents (2000-2016), nearly all by Prof. Dr.-Ing. Konstantin Meyl, an electrical engineer who taught at Furtwangen University (Hochschule Furtwangen) in the Black Forest and ran a technology-transfer centre ("1st TZS") and the publisher and kit seller INDEL GmbH [read: claims.md; research/2026-10-08-northdata.md]. The core works are a 657-page textbook, Scalar Waves: first Tesla physics textbook for engineers (2003), the documentation of his two-coil experimental kit (2000), the book-length interview Neutrino Power (2000), a 2001 paper in the Journal of Scientific Exploration (JSE) and short papers on vortex losses, Earth growth and the "theory of objectivity" [read: facts.json].

Source gap (repeated from claims.md and arbiter.md, updated after the QE remediation of 2026-10-08). The 2003 textbook has no equations in its text layer; the deep readers read page images for chapters 5, 7, 11-13, 21-24 and 26-30. The first review left chapters 8.2-10 and 14-20 (PDF pp.183-241 and 295-455: Tesla history, other inventors' free-energy devices, Meyl's own 1989-1999 experiments and the 1999 eclipse) at digest level, because the deep reads of those pages were lost. The quality check (QE) flagged this, and those pages have since been deep-read from the extracted text, together with the rest of Neutrino Power (pp.1-60, 133-222 and 223-302, so the book is now deep-read in full), the Wiesbaden medicine lecture (2001, pp.1-14) and the FGF newsletter (pp.4-7, 13-14) [read: arbiter.md, Source gap and QE revision]. The figure pages of ch.19-20 were read as images; those of ch.14-18 were not. Chapters 1-4, 6, 8.1 and 25 still rest on digests (ch.6 repeats the argument of the two Objektivitätstheorie papers, which were deep-read). The devices of other inventors (Testatika, Searl, Bedini/Ecklin, Patterson, Stan Meyer, Podkletnov, the railgun and others) are known here only from Meyl's second-hand account, and no verdict is given on any device (section 4.5). Records named in the documents but not obtained by anyone in this review: T. Treskatis, bachelor thesis, University of Konstanz, 2010 (the only measured data behind the capacitor-loss claim); Meyl, PIERS Moscow 2012; the TU Clausthal "control survey" of 6 July 2000; Prof. Runckel's opinion (Ulm, 2003); A. Waser, NET-Journal 12/2000; T. Eibert, FGF Newsletter 1/2004; Meyl's eclipse articles (NET-Journal 7/8-1999; Magazin 2000plus nr.148, 150, 151); the Tamarack mine plumb-line records. Tesla's 1919 Electrical Experimenter article, listed as missing in the first version, has now been read on the web by the arbiter (section 4.4). Where a conclusion rests on [memory] alone, it is marked.

Source tags: [read] = from the project documents, the deep reads (deep/) or the Investigator's notes (research/); [web] = Internet source, URL given (most checked by the reviewers or the arbiter); [memory] = recalled, with confidence; [computed] = calculated for this primer (run example.py). Verdicts are the arbiter's (arbiter.md); this primer adds none.

---

1. The idea in one paragraph

Meyl believes that the four Maxwell equations, the foundation of all electrical engineering, are an approximation. He starts from two older formulas that describe what a moving observer sees (a moving magnet seems to carry an electric field, a moving charge a magnetic one), and argues that if you derive Maxwell's laws from them, one term drops out that Maxwell's followers threw away. He keeps it and calls it the "potential vortex": a whirl of the electric field inside insulators and empty space, the partner of the familiar eddy current in metals. These vortices, he says, explain why capacitors lose energy, where antenna noise comes from, and what happens in the region close to an antenna. Moving vortices form a "scalar wave", a wave that pushes along its direction of travel (like sound) instead of sideways (like light), and whose speed is not fixed: in his calculation it can reach 1.618 times the speed of light, the golden ratio. He identifies this wave with the "radiations" Nikola Tesla described around 1900 and with neutrinos, the ghostly particles that stream from the Sun. If neutrinos are vortices that can be caught in resonance, they are an energy source ("neutrino power"), and the Earth, which absorbs some of them, should grow: he computes about 18 cm more equator every year. He sells a suitcase kit of two flat coils with metal spheres that lights LEDs at a distance, and presents it as a demonstration of scalar waves, faster-than-light signalling, over-unity energy and transmission through a Faraday cage. A private test institute reproduced the kit's effects in 2001 and explained them as an ordinary pair of tuned circuits with a wire between them. This primer shows the real physics under each of Meyl's starting points, then where each inference holds or fails, with the Earth-growth calculation done in full.

---

2. What you need to know first

Vector calculus identities. For any vector field F: curl(curl F) = grad(div F) - Laplacian(F), written in Meyl's German notation rot rot F = grad div F - Delta F. The Laplacian Delta (also written nabla squared) applied to a vector acts on each Cartesian component. A field that is all "grad" (curl zero) is called longitudinal or irrotational; one that is all "curl" (divergence zero) is transverse or solenoidal. Every reasonable field splits into these two parts (the Helmholtz decomposition). Read: Griffiths, Introduction to Electrodynamics, ch.1 and Appendix B [memory, high].

Maxwell's equations and the wave equation. The four equations, the Lorentz force law, and the step from them to the wave equation for light (section 3.1). Read: Griffiths ch.7 (Maxwell's equations) and ch.9 (electromagnetic waves) [memory, high].

Waves in a lossy medium (the telegraph equation). In a conductor or a lossy insulator the wave equation gains a first-derivative "damping" term, E_t (subscript t means partial derivative with respect to time). The solutions are waves whose amplitude decays. A complex permittivity (a permittivity with an imaginary part) is the standard bookkeeping for this loss. Read: Griffiths ch.9 (section on waves in conductors); Jackson, Classical Electrodynamics, ch.7 [memory, high].

Phase, group and signal velocity. The phase velocity of a wave is how fast a crest moves; the group velocity is how fast an envelope moves; the signal (front) velocity is how fast a sudden switch-on travels. The first two can exceed c in special media without any information going faster than light; the front never does. Read: Griffiths ch.9 (dispersion); Jackson ch.7 [memory, high].

Resonant circuits and transmission lines. An inductor L and capacitor C resonate at f = 1/(2 pi sqrt(LC)). A coil wound from a long wire also has standing-wave resonances along the wire, like a transmission line (a "Lecher line" is a pair of parallel wires used to show standing waves). Two tuned circuits near each other share energy back and forth; a tuned load draws power from the source and lowers its voltage. Read: any circuits text; Pozar, Microwave Engineering (transmission lines) [memory, high].

Special relativity and moving fields. An observer moving with velocity v through a magnetic field B sees an electric field E' = E + v x B, to first order in v/c. The exact formula has the Lorentz factor gamma = 1/sqrt(1 - v^2/c^2). Read: Griffiths ch.12 [memory, high].

Rotation and angular momentum. A spinning body has angular momentum J = I omega, with I the moment of inertia and omega = 2 pi / t the angular speed (t the rotation period). If no outside torque acts, J is constant, so a body that grows (larger I) spins more slowly: the ice-skater effect. For a uniform ball I = (2/5) M R^2. Read: Taylor, Classical Mechanics, ch.10 [memory, high]; Kleppner and Kolenkow, An Introduction to Mechanics [memory, high].

Neutrinos. Neutrinos are electrically neutral particles that interact only through the weak nuclear force (and gravity). About 6.5e10 (that is, 6.5 x 10^10) solar neutrinos cross each square centimetre each second, and almost all of them pass through the whole Earth without touching anything. They have small but non-zero masses, shown by "flavour oscillation" (Super-Kamiokande 1998, SNO 2001) [memory, high]. Read: Griffiths, Introduction to Elementary Particles, the neutrino chapter; Bahcall, Neutrino Astrophysics (1989) [memory, high].

Dielectric loss. An insulator in an alternating field absorbs a little energy each cycle as its molecules' dipoles lag behind the field. The loss factor tan delta = (energy lost per cycle) / (2 pi x energy stored). Read: Jackson ch.7 (dispersion), or a materials text such as Jonscher, Dielectric Relaxation in Solids (1983) [memory, high].

---

3. The textbook view

3.1 Maxwell's equations and why light is transverse

In SI units, in vacuum or with free charges and currents:

Symbols: E electric field (V/m); B magnetic flux density (T); rho charge density (C/m^3); J current density (A/m^2); eps0 = 8.854e-12 F/m and mu0 = 4 pi e-7 H/m the vacuum constants, with c = 1/sqrt(mu0 eps0) = 2.998e8 m/s. In a material we also use D = eps E (electric displacement), H = B/mu (magnetic field), and Ohm's law J = sigma E (sigma conductivity, S/m). The force on a charge q moving with velocity v is F = q(E + v x B) (the Lorentz force).

Take the curl of Faraday's law and use the identity of section 2:

curl curl E = grad(div E) - Laplacian(E) = -d/dt (curl B) = -mu0 dJ/dt - (1/c^2) d^2E/dt^2.

With Gauss's law inserted, the exact result is

Laplacian(E) - (1/c^2) d^2E/dt^2 = (1/eps0) grad rho + mu0 dJ/dt.

In empty space (rho = 0, J = 0) the right side vanishes and E obeys the wave equation with speed c. The "grad div E" term is zero there not because anyone forgot it but because Gauss's law says div E = rho/eps0 = 0 when there is no charge. For a plane wave E = E0 cos(kx - omega t), div E = 0 forces E0 to be perpendicular to the direction of travel x: light is transverse. Where there is charge, longitudinal electric waves are standard physics: the plasma (Langmuir) wave in an ionised gas obeys omega^2 = omega_p^2 + 3 k^2 v_th^2, with omega_p the plasma frequency and v_th the electrons' thermal speed; sound in air is longitudinal too [memory, high].

3.2 Lossy media: the telegraph equation

In a conductor, J = sigma E adds a damping term:

c_m^2 Laplacian(E) = d^2E/dt^2 + (1/tau1) dE/dt, with tau1 = eps/sigma (the charge relaxation time) and c_m = 1/sqrt(mu eps) the speed in the medium.

This "telegraph equation" has damped wave solutions. Dielectric loss in an insulator is handled the same way by writing the permittivity as complex, eps = eps' - i eps''. A complex permittivity makes the wavenumber k = omega sqrt(mu eps) complex: the real part sets the wavelength, the imaginary part the attenuation. The speed of light c (the vacuum constant) is not affected. Worked number: polypropylene with eps_r about 2.2 and tan delta about 2e-4 has refractive index n = sqrt(2.2) = 1.483 and extinction coefficient about n tan delta / 2 = 1.5e-4: light travels at c/1.483 and fades slowly [read: physicist.md, claim 2; material values memory, medium]. Dielectric loss in polymer film capacitors is modelled as dipole relaxation with a spread of relaxation times (Debye, Cole-Cole, Havriliak-Negami) and Jonscher's near-constant "universal" loss, plus the series resistance of the electrodes, which adds tan delta = omega R_s C (R_s series resistance, C capacitance) [memory, high]. The microscopic origin of Jonscher's near-flat loss is empirical and has no single accepted first-principles derivation [read: contrarian.md, claim 2; memory, medium].

Noise and loss are tied together by the fluctuation-dissipation theorem (Nyquist 1928; Callen and Welton 1951): any element that absorbs energy at temperature T also emits noise of power kT per hertz of bandwidth (k = 1.381e-23 J/K, Boltzmann's constant) [memory, high]. So a capacitor's loss and its noise are expected to go together without any new mechanism.

3.3 Near field of an antenna

A short oscillating dipole has field terms that fall as 1/r^3 and 1/r^2 (the reactive "near field") and as 1/r (the radiated far field). They are equal at kr = 1, i.e. at r = lambda/(2 pi), where lambda is the wavelength and k = 2 pi/lambda. The near field has a component along r, the direction away from the antenna. The textbook picture of closed field-line loops pinching off from a dipole is an exact solution of Maxwell's equations [memory, high]. Worked number used throughout this primer: Meyl's kit runs near 7 MHz, so lambda = c/f = 3.0e8/7e6 = 42.8 m and lambda/(2 pi) = 6.8 m [read: physicist.md section 0].

3.4 Moving observers

The exact transformation of the fields to a frame moving with velocity v is E'_perp = gamma (E + v x B)_perp (perp = the part perpendicular to v), and similarly for B. To first order in v/c this is E' = E + v x B and H' = H - v x D. Transverse fields of a fast charge grow by gamma: at v = 0.99 c, gamma = 7.09. Beam-position monitors in accelerators measure this routinely [memory, high].

3.5 Coupled resonators and wireless power

Two circuits tuned to the same frequency, coupled through a shared capacitance or a mutual inductance, exchange energy efficiently at short range. Resonant wireless power transfer over 2 m at about 40% efficiency was shown in 2007 (A. Kurs et al., Science 317, 83) [memory, high]. If one end of each circuit is joined by a wire or by ground, the pair forms one circuit with a return path. A tuned receiver loads the transmitter (back-reaction). A nonlinear load such as an LED clips the waveform and creates harmonics (multiples of the drive frequency). Energy out cannot exceed energy in.

3.6 Earth's rotation, leap seconds and the growing-Earth question

The Earth's day lengthens slowly because the Moon's tides brake its spin; the angular momentum lost by the Earth goes into the Moon's orbit, which grows by about 3.8 cm/yr (lunar laser ranging) [web: arbiter.md, Moon recession]. The tidal prediction is about 2.3 ms per century of day-length increase; eclipse records give a net 1.7-1.8 ms per century (post-glacial rebound of the crust offsets part of it) [web: https://theconversation.com/temporal-flux-why-we-need-to-keep-adding-leap-seconds-44077; https://arxiv.org/pdf/2110.09037, via the contrarian and mathematician]. Leap seconds are a different quantity: the SI second was matched to the length of the day of a century or two ago, today's day is about 2 ms longer than 86,400 SI seconds, and those 2 ms accumulate to about 0.7 s a year, which is why a leap second is added every year or two [memory, high on the mechanism; medium on the date the second was matched]. Space geodesy (very-long-baseline radio interferometry, satellite laser ranging, GPS, DORIS) measures the Earth's mean radius change as 0.1 +/- 0.2 mm/yr (1 sigma; Wu et al., Geophys. Res. Lett. 38, L13304, 2011) [web: https://www.sciencedaily.com/releases/2011/08/110817120527.htm].

The expanding-Earth hypothesis (O. C. Hilgenberg 1933, later S. W. Carey and P. Jordan) was a respectable minority position before plate tectonics: it explained why the continents seem to fit together on a smaller globe. Plate tectonics explains the same fit by sea-floor spreading balanced by subduction; the oldest ocean floor is about 180-200 million years old, as subduction predicts [memory, high].

3.7 Neutrino numbers

Solar neutrino flux at Earth: about 6.5e10 per cm^2 per second [memory, high]; Meyl's patent uses 66e9 [read: research/2026-10-08-google-patents-espacenet.md]. Mean energy about 0.26 MeV (the low-energy pp neutrinos dominate). Interaction cross-section with an electron at around 1 MeV: about 1e-45 cm^2 [memory, medium; the arbiter notes it may be nearer 1e-44 at 1 MeV, which changes nothing below]. Neutrino speed: the neutrinos from supernova SN 1987A arrived within hours of the light after 168,000 years of travel, which bounds |v/c - 1| below about 2e-9 for 10-40 MeV neutrinos [web: arbiter.md, web check 4].

---

4. Where Meyl departs

The fair way to read Meyl is to see that almost every one of his starting points is a real piece of physics, and that the departure comes at one specific step. The table lists the steps; the subsections show the algebra.

4.1 The "derivation" of Maxwell from E = v x B

Meyl takes Faraday's motional law E = v x B (a conductor moving through a magnetic field feels an electric field) and the "convection" law H = -v x D, takes their curls, and with standard vector identities gets terms like (v . grad) B, v div B and B div v [read: deep/Faraday-or-Maxwell s.4.2]. He then (a) identifies (v . grad) B with dB/dt, (b) identifies -v div D with the current density j (Gauss's law used as an input), and (c) is left with b = -v div B, which he keeps and writes b = B/tau2. Faraday's and Ampere's laws then read curl E = -dB/dt - B/tau2 and curl H = dD/dt + D/tau1, and their combination is his "fundamental field equation"

c^2 Laplacian(E) = d^2E/dt^2 + (1/tau1 + 1/tau2) dE/dt + E/(tau1 tau2)

(5.7 and 27.26 in the 2003 book). The algebra from the extended laws to this equation is correct, units included; three reviewers re-derived it [read: deep/Meyl-Scalar-Waves-...-603e47 s.2; arbiter.md claim 1]. What a beginner should credit: Faraday's motional emf is a legitimate starting point (some German textbooks, e.g. Kuepfmueller, start there [read: physicist.md claim 1]), Maxwell's own 1865 equations did contain the v x B term (the four-equation form is due to Heaviside and Hertz) [memory, high], and a magnetic-current term is a legitimate formal extension that has been studied for a century as the "dual" of electric current [read: contrarian.md claim 1].

Where it fails:

• (v . grad) B is the change of B seen by an observer moving through a field that is still; dB/dt in Faraday's law is the change at a fixed point. For a pattern that moves, B(r - vt), the fixed-point derivative is dB/dt = -(v . grad) B: the opposite sign [read: physicist.md claim 1; deep/Objektivitaetstheorie-1 s.3].
• Gauss's laws are put in, not derived; b is exactly what is left when div B = 0 is not imposed. So the claim reduces to "let div B be non-zero" [read: deep/Faraday-or-Maxwell s.4.2(3)].
• Put Meyl's two starting laws into each other: E = v x B = mu v x H = -mu eps v x (v x E). For E perpendicular to v, v x (v x E) = -v^2 E, so E = (v^2/c^2) E. This holds only if v = c. Meyl prints this himself as "result v = c" (book 28.5, p.583) [read: deep read; arbiter.md claim 1]. The same point was made in words by the mathematician G. W. Bruhn (TU Darmstadt), an organised critic (the sceptics' society GWUP), and so an interested party on the critics' side, in the critique the JSE printed beside Meyl's 2001 paper [read: research/2026-10-08-tu-darmstadt-bruhn.md]. The calculation agrees with him.
• No value of tau2 appears anywhere, and its dependence on conductivity is given three incompatible ways in the book [read: physicist.md claim 1]. The 2012-2013 papers concede "the potential vortex has only been postulated and not mathematically derived" [read: deep/About-Vortex s.3.4].
• If tau2 is short in vacuum (Meyl's own choice tau2 = r/c with r a vortex radius; at 7 MHz r about 7 m gives tau2 about 2e-8 s), radio waves would fade to 1/e in 40 ns, i.e. 12 m [read: physicist.md claim 1]. Radio astronomy and spacecraft telemetry rule that out.
• The 2009 spin-ice "magnetic monopole" experiment Meyl cites (Morris, Tennant et al., Science 326, 411) concerns emergent defects inside a crystal; the authors' abstract states that div B = 0 remains intact [web: https://archive.aps.org/mar/2010/w36/5, abstract only].

4.2 Dielectric losses and noise

Meyl's claims: a complex permittivity "implies a complex speed of light" and is "physically wrong"; capacitor loss is the decay of potential vortices; antenna and resistor noise are the same vortices [read: claims.md claim 2]. Section 3.2 shows why the first is mistaken. The evidence offered is one qualitative plot of tan delta against frequency for a polyester (MKT) capacitor from the 2010 Treskatis thesis: three straight lines, no multiplier on the tan delta axis, no capacitance, temperature, instrument or formula for the "Meyl model", captioned "qualitative comparison"; the "orthodox" comparison curve is the tail of a single resonance that nobody uses for polymer film [read: deep/About-Vortex s.6]. Relaxation plus electrode series resistance reproduces the shape (for C = 1 microfarad and R_s = 10 milliohm, omega R_s C runs from 6.3e-6 at 100 Hz to 6.3e-3 at 100 kHz) [read: physicist.md claim 2]. The noise curve Meyl matches, hf/(1 - exp(-hf/kT)) (h Planck's constant, f frequency), rises at kT/h = 6.1e12 Hz at room temperature; the capacitor loss rise he compares it with is at 0.1-100 kHz, 8 to 10 decades lower [read: arbiter.md table]. The circular marks on capacitor electrodes he shows as "visible proof" come from papers (Yializis et al. 1980; Taylor 1984) whose titles call them electrode corrosion [read: deep/About-Vortex s.7]. Papers call Treskatis "a colleague at the university for theoretical physics"; Meyl's own note found by the Investigator says he supervised that bachelor thesis in summer term 2010 [read: research/2026-10-08-google-search.md]. No second laboratory has measured it, and the thesis was not obtained.

Meyl's own tests of claim 2 (added after the QE check). The first version of this primer, like the first arbiter, said the Treskatis plot was the only evidence. That was incomplete: chapter 19 of the 2003 book reports several informal experiments of Meyl's own [read: deep/textbook ch.19 s.1.1-1.4; arbiter.md claim 2]. - The 1990 capacitor-Q test (HF laboratory, FH Furtwangen, 11.06.1990). Q, the "quality factor", is roughly 1/tan delta for a capacitor: a higher Q means lower loss. Meyl's idea was that a dielectric "sheeted" in layers (a silver-coated stacked ceramic capacitor) should lose less than the same ceramic with the silver removed. In a resonant circuit near 400 MHz on a network analyser he measured Q = 48 with the silver on (mean of 5 runs), Q = 28 with the silver polished off, and Q = 9 with the silver etched off in dilute nitric acid (he concedes the acid may have damaged the dielectric). Then, in his own words: repeats showed "hardly a difference", the results "fluctuated extremely", and "at a critical consideration of the errors the results seemed to 'drown' in the mean variation"; the colleague who had proposed a joint paper "withdrew" [read: textbook PDF pp.410-412]. No uncertainties are given, and removing the silver also changes the capacitor itself (electrode area, contact resistance, stray capacitance of the socket), so even the first-run difference would not isolate a "vortex" loss. Arbiter: Never properly tested. It is worth knowing because it is the only dated, numbered test of claim 2 in the documents, and Meyl reports its fading honestly. - Kirlian "phantom leaf" (c.1990, plexiglas plate, "20000 Volt") and microwave "vortex decay" (practical of 22.10.1996, microwaved water against heated water, curves 1-2 degrees C apart): single uncontrolled runs with no cleaning of the plate, no blind comparison and no masses or power levels. Meyl also reports a negative follow-up: a student's spark-gap map of "vortex distribution in space" showed no dependence on place, which he explains by the spark "chasing away" the vortices. Arbiter: Never properly tested. - Charge amplifier (1989-90): a pointer that "danced" and a measuring point that "moved away" were taken to show that E is not a gradient field; no readings, controls or drift checks are reported. Arbiter: Not testable as stated.

4.3 Scalar waves and the golden ratio

Meyl's starting point is correct: Laplacian(E) = grad div E - curl curl E, and the grad div part is a longitudinal (along the direction of travel) component [read: deep/...603e47 s.3.3]. His methodological point is also sound: v = lambda f has three quantities, and a claim about a speed needs two of them measured, not one plus an assumption [read: deep/Neutrino-Power; contrarian.md claim 3]. The departure is reading the grad div term as a wave Maxwell's theory "neglects": in empty space Gauss's law sets it to zero (section 3.1).

The 1.618 c comes from the 2016 Fluid Mechanics paper (and book ch.29). He takes his field equation in vacuum with only the vortex damping kept, c^2 E_xx = E_tt + (1/tau) E_t (subscripts are partial derivatives), sets tau = r/c (a field oscillation that "runs with c, but in circles" of radius r), and tries E = E(t) exp(-x/r). With his derivatives dE/dx = -E/r and dE/dt = v dE/dx = -(v/r) E, substitution gives c^2/r^2 = v^2/r^2 - (v/r)(c/r), i.e.

v^2 - v c - c^2 = 0, so v = c (1 + sqrt 5)/2 = 1.618 c [read: deep/Meyl-K-Fluid-Mechanics s.3, page image p.2].

The algebra is right. Three things go wrong [read: deep/Meyl-K-Fluid-Mechanics s.6; arbiter.md claim 3; computed: example.py Part B]:

1. Direction. Meyl's derivatives are exactly those of E0 exp(-(x + vt)/r), a pattern that moves toward -x, i.e. inward, toward the source. Meyl says the vortex moves radially outward. For the outward pattern E0 exp(-(x - vt)/r) the same steps give v^2 + v c - c^2 = 0 and v = 0.618 c, below c. example.py Part B checks both with SymPy.
2. The golden ratio is chosen, not derived. With tau left general, the positive root is v = r/(2 tau) + sqrt(r^2/(4 tau^2) + c^2): any value above c, and 1.618 c only when tau = r/c exactly.
3. What moves is a non-oscillating, decaying exponential profile. Its "speed" is a ratio of its time-decay rate to its space-decay rate, like a phase velocity; it carries no signal. The paper also writes div E = 0 while using a field E(x) along x that varies with x, for which div E = dE/dx is not zero. No equation in the paper gives the near-field extent as a function of v, so "the near field reaches to infinity at 1.618 c" is asserted, not calculated.

Faster-than-light "tunnelling" experiments (Nimtz in Cologne, Chiao in Berkeley) that Meyl cites are real; the mainstream reading (e.g. H. Winful, arXiv quant-ph/0601085) is that they show short group delays with the signal front still at c, while Nimtz himself reads them as superluminal signals [read: arbiter.md claim 3; memory, medium on the details]. Neither reading involves a vortex mechanism.

4.4 The two-coil kit

The kit (800 EUR demonstration set, 1400 EUR experimentation kit; Meyl is the seller and says the manual omits winding data on purpose) has a transmitter and a receiver, each a flat spiral coil with a metal sphere on top, joined by a ground or one-wire link; LEDs at the receiver light up [read: deep/Scalar-Wave-Technology; arbiter.md, pertinent material]. Manual and papers claim: wireless energy transfer, receiver "feedback" on the transmitter, efficiency 500-1000% (the figure varies between Meyl's texts; see the table below), a speed of 1.5 c, and transmission out of a closed Faraday cage [read: claims.md claim 4].

What is real: the effects in the manual occur and were reproduced by the Institut fuer Gravitationsforschung (IGF), Waldaschaff, a private institute funded through the Goede foundation, in two documents of 2001: a companion report signed by Herbert Weidner (10 August 2001) and a six-author report (September 2001; oscilloscope screens dated 10 July 2001), with instruments listed (Tektronix TDS 3054 oscilloscope, HP 8594EM spectrum analyser, Fluke meters) [read: deep/Experimente; research/2026-10-08-wayback-gravitation-org.md; arbiter.md correction 12]. Their explanation: the coil plus sphere is a capacitively shortened, coiled Lecher line with a fundamental at 5.35 MHz and a node-mapped mode at 29.5 MHz; 3.4 mA rms flows in the link wire; the LEDs clip at 1.7 V and generate odd harmonics (the 21.6 MHz third harmonic was 15 dB above the 7.2 MHz drive at the sender coupling coil); the "feedback" is a resonance of the 18.8 m wire at 21.6 MHz; with a low-harmonic oscillator and the LEDs removed, "the effects described by Meyl no longer exist at all"; efficiency about 45% (by eye-matched LED brightness); no over-unity [read: research/2026-10-08-wayback-gravitation-org.md; deep/Experimente]. The IGF invited Meyl to repeat his tests in its laboratory; no reply is on record (absence of a record is not proof of no reply) [read: arbiter.md]. Meyl's reply calls the IGF a "questionable private institute" that made only conventional measurements [read: deep/Kreuzfeuer]. His own granted patent DE 10348862 B4 (filed 2003, granted 2009) describes the link in circuit terms: "a return conductor is required. It can be a single-core cable or a ground connection", forming one L-C circuit at self-resonance, with no over-unity claim [web: https://patents.google.com/patent/DE10348862B4/en].

The IGF record has its own gaps, kept here: efficiency by eye-matched LEDs; a wavelength of 56.07 m on p.6 (right for 5.35 MHz) and 44.4 m on p.10 (which belongs to 6.75 MHz); "28% of oscillator current" which implies 12 mA while the power figure uses 16 mA; no measured Faraday-cage test (it says the cage was checked but prints no data); no wire-removed control; and its power-versus-spacing curve (about 8 mW at 17 cm, 3.3 mW at 37 cm, then about 3.1 mW out to 67 cm) fits conducted transfer as well as Meyl's "no fall-off with distance" [read: arbiter.md claim 4]. The IGF also states that the kit is far too small and low-powered to count as a replica of Tesla's apparatus, and that Tesla's experiments "have not yet been exactly replicated" for lack of money [read: contrarian.md claim 3, IGF report p.13].

The 1.5 c. Meyl reads the ratio of two resonance frequencies, 7 MHz and about 4.7 MHz, as a ratio of speeds at a fixed wavelength: 7/4.7 = 1.49 [read: deep/Scalar-Wave-Technology]. The IGF's measured mode structure (non-harmonic resonances of a shortened line) shows that two modes of one coil do not have the same wavelength, so the ratio is not a speed. No time-of-flight or node-mapped wavelength measurement of the claimed fast wave exists. Meyl's second route is Tesla's number. Tesla's patent US 787,412 (1905) does say the disturbance travels "with a mean velocity of about four hundred and seventy-one thousand two hundred and forty kilometers per second"; it does not mention 12 Hz or Schumann resonances [web: https://patents.google.com/patent/US787412A/en]. 471,240 km/s = (pi/2) x 300,000 km/s to five digits; 471,240 km/s divided by the Earth's circumference 40,030 km is 11.8 Hz (Meyl's "12 Hz"), and c divided by the same circumference is 7.49 Hz [computed: arbiter.md table]. One reading (an inference by the physicist and arbiter, also offered by the contrarian from Tesla's 1927 wording as quoted by Gao Peng) is that pi/2 is the ratio of a path along the surface (half circumference, pi R) to the diameter (2R), i.e. a geometric factor, not a measured superluminal speed. Gao Peng's 2016 torsion-balance test of a dual Tesla coil (viXra 1607.0130) found resonances at 4.17 and 2.65 MHz, ratio 1.57, also close to pi/2 (coincidence not excluded) [read: research/2026-10-08-vixra-gaopeng.md].

Meyl reads the same patent a second way in the 2003 book (p.214): he takes Tesla's "0.08484 of a second" as half a period, giving 1/(2 x 0.08484 s) = 5.9 Hz, "about 6 Hz", and a speed of 2 pi x 6363 km / 0.08484 s = 471,239 km/s = 1.57 c using the polar radius 6363 km [read: deep/textbook ch.8-10; arbiter.md correction 26]. The two readings (12 Hz in the kit lectures, 6 Hz in the book) agree with each other under Meyl's own assumptions, but both treat 0.08484 s as a measured travel time around the Earth. The patent gives it as the length of a wave-train, "not less than one-twelfth or probably 0.08484 of a second" [web: https://patents.google.com/patent/US787412A/en].

The Kelvin-Tesla story. Every Meyl document supports the "two kinds of wave" idea with a meeting between Lord Kelvin and Tesla, ending "Both are right, Tesla as well as Hertz!". The arbiter read the original: Tesla, "Famous Scientific Illusions", part III, Electrical Experimenter, February 1919 [web: http://www.tfcbooks.com/tesla/1919-02-00.htm]. Tesla writes that in "the summer of 1897" Kelvin visited his laboratory, first "condemned my project ... 'an illusion and a snare'", then asked "Then you are not making use of Hertz waves?", and Tesla answered "Certainly not ... these are radiations ... In my system the process is one of true conduction". Kelvin then became "the warmest of supporters". The words "longitudinal" and "transverse" are not on the page, and neither is "Both are right". Those, and the claim that Kelvin called Tesla's wave longitudinal, are Meyl's additions [read: arbiter.md QE revision s.5]. So the visit is documented in one published source, Tesla's own retelling 22 years later (Tesla had a stake in his own priority); notice that Tesla himself calls his method "true conduction", which is what the IGF and Meyl's own patent describe for the kit.

Does the link need its wire? Meyl's own manual says yes. In Experiment 2 of the kit manual: "If the cable connection is carefully unplugged, the light emitting diodes installed onto the transmitter side shine again. The LED's, installed on the receiver side, extinguish" [read: _text kit manual, PDF p.19, quoted in qe.md]. The Fortschrittliche lecture says that if the earth lead is disconnected "the oscillation stops", and the Wiesbaden lecture calls the ground "indispensable as reference potential" [read: arbiter.md claim 4; deep/Skalarwellenstrahlung s.4]. Meyl reads the ground as a waveguide for the scalar wave; a circuit engineer reads it as the return conductor his own patent says is "required". Either way, by Meyl's own text, unplugging the wire stops the transfer.

The over-unity figure, read at source. Meyl's book gives the only worked efficiency [read: _text, book p.411, PDF p.422]:

"An output power of 49 mW resulted from the measurements of the current and tension voltage for a loading of the receiver with a 100 Ohm resistance. Simultaneously the power taken up by the transmitter amounted to 233 mW. If we however subtract the idle power consumed by the transmitter from this, and that was determined to be 223 mW for switched off receiver, then actually only 10 mW are available for the wireless energy transmission. The degree of effectiveness according to that would be 490%! If we here really have obtained an over-unity of 4.9 then the receiver must have collected along free space energy, or did some parts evade our power measurement?"

Output over input is 49/233 = 21%. The 490% is 49/(233 - 223) = 49/10: output divided by the increase in input. A 5% error in the 233 mW reading is 11.7 mW, larger than the 10 mW difference itself, so the 490% is inside the measurement noise [computed: arbiter.md table]. The book asks the right question itself ("did some parts evade our power measurement?").

The deep read of chapter 19 found that the real uncertainty is larger still [read: arbiter.md claim 4, QE revision]. The record is from December 1999 (6.7 MHz, about 2 m, measured by M. Andresen and M. Rehm on a LeCroy 9304C oscilloscope). The transmitter's voltage and current were almost 90 degrees out of phase: the power factor cos(phi) was about 0.055, phi near 87 degrees. Real power is P = S cos(phi), where S = V_rms I_rms is the "apparent power"; here S = 233 mW / 0.055 = 4.24 W. Near 90 degrees a small error in the measured phase changes P by dP = S sin(phi) dphi, about 4.24 W per radian. A timing mismatch (skew) of 1 ns between the voltage and current probes at 6.7 MHz is a phase error of 2 pi x 6.7e6 Hz x 1e-9 s = 0.042 rad, which moves P by about 178 mW. The whole 10 mW difference corresponds to 56 picoseconds of skew [computed: example.py Part D]. No deskew, no "receiver present but unloaded" control and no repeat of the pair are reported. Arbiter: over-unity Tested, not supported.

How big is the claimed over-unity? It depends on which Meyl text you read [read: arbiter.md QE revision s.4]:

The figure ranges from 1 to 10 within three years for the same kit. Meyl names his own uncontrolled variables: the result depends on set-up, grounding, time of day "and even the resonance capacity of the spectators" (Wiesbaden), and Neutrino Power p.19 says "considerable fluctuations" occur with the day-night rhythm and "cosmic constellations" [read: arbiter.md claim 4]. The Clausthal result is Meyl's report, as seller of the kit, of a third party's measurement with no document. The other confirmations are named only loosely: "over 50 universities worldwide (Edinburgh to Vienna, Stanford, TU Peking)" with no result or report; an "EMC laboratory" that "checked and confirmed" the cage test, with a CE declaration; "hundreds of measurements at more serious institutes"; "one German university bought three kits" [read: qe.md section 3 item 13; deep/Kreuzfeuer]. These are leads the Investigator has not yet traced, not evidence.

Flat versus cylindrical coils. Neutrino Power p.20 says that if the flat spiral coils are replaced by equivalent cylindrical coils, the efficiency "drops below 100%". No coil data or measurement is given. Meyl's own engineering reason for Tesla's flat coil is insulation along the winding. Arbiter: Never properly tested; this is one a reader with a signal generator could try (equal L, equal C, same circuit, power measured at both ends).

Meyl's own negative results. These sit beside the positive reports and bear on how much weight they carry [read: arbiter.md QE revision s.8]: - Neutrino Power p.240: an inventor phoned Meyl about a magnetic converter giving "three kilowatts output in continuous operation". Meyl saw it run, but "the effects could not be reproduced in my laboratory so far ... we could not yet reach an over-unity effect". As one possible cause he names "the absence of the inventor" (p.241: an inventor may act "as field modulator"). Read it both ways: an honest report of a failed reproduction, and a hypothesis that would make any replication without the inventor uninformative. The inventor designed the device after talking to Meyl, so Meyl is a participant, not a neutral witness. - Book p.346: "I judge these measurements rather sceptical, after I already in three cases had to break the message gently to the inventors that their device ... was nothing but an energy destroying machine and they merely had measured wrong." He recommends a closed loop (the output drives the input) as the only convincing demonstration. No device in the book's device chapters (PDF pp.295-400) is reported to have run closed-loop. - Neutrino Power p.238: "none" of his rebuilds of Seike's windings worked.

Stakes. Meyl sells the kit (800 and 1400 EUR in 2000) and, later, an SWG-A "scalar wave device" at 3950 EUR (section 4.7). He says telecom firms bought his books but declined funding and that his students are paid from a private donation fund (Neutrino Power p.295); he is president of a "Society for the Advantage of Physics" and sits on the advisory board of the egm "Wirbelwandler" (vortex converter) company [read: arbiter.md QE revision s.8; sites.md]. A seller can also be a sincere inventor; the stake is stated so the reader can weigh it.

The Faraday cage. The manual says transmission from a closed case works "with a gap in the door or a cable which leads out of the suitcase" and calls that "perfect coupling" [read: deep/Scalar-Wave-Technology Exp.5]. A cable through the wall carries the signal by ordinary conduction. Meyl concedes in Neutrino Power that the received signal cannot show what share was scalar [read: deep/Neutrino-Power, pp.108-109]. Neither side printed a cage measurement with no conductor leaving the enclosure.

4.5 Neutrinos as vortices; neutrino power; the growing Earth

Meyl pictures the neutrino as a ring vortex that oscillates between an electron and a positron, with zero average charge and mass but a non-zero root-mean-square (RMS) value, "like 50 Hz alternating current"; it travels faster than light and interacts only "in resonance", at the same frequency and opposite phase [read: deep/Teslas-Radiations s.2, introduced as "Stellen wir uns vor" ("let us imagine")]. No equation, cross-section or coupling strength is given. In his own analogy the RMS charge would be e/sqrt(2), while neutrino charge limits are near 1e-21 e [memory, medium]. His 1996 patent application DE 19614575 A1 is the one place with an energy budget: 66e9 neutrinos per cm^2 per second would give "about 27 W per m^2" if all were converted, which is the right solar value (0.26 MeV each) [web: https://patents.google.com/patent/DE19614575A1/en]. The growing-Earth calculation is section 5.

The eclipse "lens". For the total eclipse of 11.8.1999 Meyl forecast that the Moon would focus "slow" solar neutrinos onto the shadow axis like "a glass ball", raising the flux 27 to 68 times along Ulm-Augsburg-Freising (his peak figure is printed as 27, 28, about 2800% and about 3200% in different places for the same calculation) [read: deep/textbook ch.20 s.2.1; arbiter.md QE revision s.2]. The first review refuted this with the Moon's gravitational lens, which is not Meyl's mechanism; the QE check asked for the calculation that answers his own model, refraction. Here it is as a short worked number.

Neutrinos do have a refractive index in matter. A neutrino moving through electrons feels a tiny potential energy V from coherent forward scattering (the effect behind the Mikheyev-Smirnov-Wolfenstein, MSW, effect in the Sun):

V = sqrt(2) G_F N_e, and n - 1 = -V/E,

where G_F = 1.166e-5 GeV^-2 is the Fermi constant (the strength of the weak force), N_e the number of electrons per unit volume, and E the neutrino energy [memory, high]. (G_F is in "natural units"; multiplying by (hbar c)^3, with hbar c = 1.973e-14 GeV cm, turns G_F N_e into an energy.) For lunar rock, density 3.34 g/cm^3 with about one electron per two nucleons, N_e is about 1.0e24 per cm^3, so V = 1.27e-13 eV. At 1 MeV, |n - 1| = 1.3e-19; at 0.26 MeV, 4.9e-19. A glass-ball lens of radius R focuses at about R / (2|n - 1|) beyond its centre; with R = 1.74e6 m that is 7e24 m, some 700 million light-years. To focus on the Earth, about 3.6e8 m away, Meyl's Moon would need |n - 1| of about 2.4e-3, which is 2e16 times what the weak force gives [computed: example.py Part C; arbiter.md QE revision s.2]. For electron neutrinos the sign of V even makes the Moon a weak diverging lens [memory, medium]. Meyl's escape is that his "slow neutrinos" couple far more strongly; he gives no energy, cross-section or index for them. Arbiter: the lens is Refuted by calculation; the slow-neutrino premise is Not testable as stated.

Dated forecasts: check it yourself. Meyl invites this: "anyone with my indications and an eclipse of the sun-CD can analyse at home all further events" [read: textbook p.448]. His claims, after the fact and in advance [read: deep/textbook ch.20; qe.md section 3 item 4]: - After 11.8.1999: the Izmit earthquake (17.8.1999) and aftershocks "in a rhythm of 6 to 7 days"; "a proven temporary acceleration of the rotation of the earth"; Foucault-pendulum anomalies; a Salt Lake City tornado; the 26.2.1998 Montserrat eruption. - Forecasts printed in the 2003 book: after the eclipse of 29.3.2006 the Earth's rotation slows and there is "danger of earthquakes (approx. 4.4.2006) for the Island of Crete and Asia Minor"; "no good" from 13.11.2012; 21.8.2017, "The thankless role of Turkey ... this time takes over South Carolina".

The public records are the USGS earthquake catalogue (ComCat, https://earthquake.usgs.gov/earthquakes/search/) and the IERS length-of-day series (https://hpiers.obspm.fr/iers/eop/eopc04/). The arbiter's queries [web: arbiter.md QE revision s.3]: 27 earthquakes of magnitude 6 or more worldwide in the two months after 11.8.1999, against about 25 expected from the 1995-2004 background; length of day rose after the 1999 eclipse (0.45 to 0.59 ms above 86,400 s, the Earth slowing, not accelerating) and fell after the 2006 one (the Earth speeding up, not slowing); nothing at Crete or in Anatolia in 2006, though an M5.5 earthquake did strike off Zakynthos, western Greece, about 400 km from Crete, on 4.4.2006, the day Meyl named, and opened a swarm; nothing at the named places after 13.11.2012 or 21.8.2017. Both length-of-day windows are dominated by the regular 13.66-day lunar tide, which moves the day by about +/-0.5-0.7 ms [memory, high]. Weigh it both ways: the 2006 date-match is real and a fair point for Meyl, but the place was wrong and one dated event per eclipse is not a statistical test. A proper test fixes regions and windows in advance for every eclipse. Arbiter: Tested, not supported at the places and quantities named; Montserrat (a volcano erupting since 1995), Salt Lake City and the pendulum claims are Not testable as stated (no data or sources given). A useful number for the pendulum: a 1 ms change in the day is a fractional change of 1e-3 / 86,400 = 1.2e-8, far too small for any Foucault pendulum to show, so a pendulum anomaly, if real, would need some other cause (the "Allais effect" question).

Third-party eclipse anomalies. Meyl cites Shu-wen Zhou (21st Century Science & Technology 12(3), 1999), who reported caesium-clock and spectral anomalies during eclipses (1987, 1992, 1995; a US Naval Observatory clock pair, 1989-91). This is an anomaly report by others of the Allais type; the eclipse-pendulum and gravimeter literature is mixed and disputed [memory, medium], and a few microseconds of drift between two caesium clocks over months is within ordinary clock noise [memory, medium]. The magazine is linked to the LaRouche movement [memory, medium]. Arbiter: Never properly tested by neutral parties; the record should be obtained.

Galaxies and red giants. Chapter 10 says "a still unknown resonating interaction" of synchronously oscillating neutrinos holds galaxies together, and that stars grow by collecting neutrinos, so the Sun will become a red giant. There is no equation, force, flux or rotation-curve data [read: deep/textbook ch.8-10]. Arbiter: Not testable as stated.

Other inventors' devices. Chapters 14-18 relay devices by others (railgun, Testatika, Kromrey, Ecklin, Bedini, RQM, Hooper, Seike, Coler, Patterson, Meyer, Hutchison, Searl, Podkletnov). Every figure is second-hand (Schneider's compilations, NET-Journal, Raum & Zeit, the Internet, company pages, Gruber's conference report), and no primary test report is in the folder, so no device gets a verdict here (arbiter: Never properly tested within these documents) [read: arbiter.md claim 5]. Examples, as Meyl's report of others' claims: Coler's device, in which "Professors Kloss and Franke of the Technical University of Berlin found a degree of effectiveness of 450%", with the documents said to have been confiscated by British intelligence and partly declassified in 1962 (one compilation, Schneider 1994; a declassified British report is the record that would decide it); Patterson cells with "up to 80%" reduction of radioactivity, reported by Gruber after a 1997 visit [read: arbiter.md QE revision s.8]. Meyl's own judgments sit beside them: Testatika "may rather be given as a counterexample ... There thus can't be talked of free energy"; permanent-magnet machines cannot be free-energy converters by his theory; Podkletnov's 2-4% "is not exactly much". One device figure can be checked with a calculator: the railgun said to give "399 GJ" (24,000-fold) from 16.7 MJ input. Its projectile, 0.317 kg at 4200 m/s, carries (1/2) m v^2 = 0.5 x 0.317 x 4200^2 = 2.8 MJ, 17% of the input; 399 GJ in 0.317 kg would need 1.6e6 m/s. Arbiter: Refuted by calculation as an energy budget [computed: arbiter.md]. Meyl adds that small replicas show no avalanche effect and that he has built no model.

Further inputs a reader can check (noted by the QE check; not weighed by the reviewers): - Kamiokande's "3 billion" against Gallex's "66 billion" neutrinos per cm^2 per s is read as "4% positrons, 96% electrons" [read: digest JGG p.2]. The two detectors have different energy thresholds and see different parts of the solar spectrum (Kamiokande only the rare high-energy boron-8 neutrinos, Gallex down to the low-energy pp neutrinos) [memory, high], so the ratio is not a charge split. - Aristarchus's year of 365.25062 days, compared with "today's 365.25637 days", gives 497 s in 2300 years, "0.22 s/a" [read: digest JGG p.3-4]. Today's 365.256 days is the sidereal year (one orbit against the stars); the year of the seasons (tropical year) is 365.2422 days [memory, high]. Which year Aristarchus measured, and how precisely, has to be settled before the difference means anything. - The free-neutron lifetime is given as 918 s (book p.168) [read: digest]; the measured value is about 880 s [memory, high]. (918 s was a commonly quoted value in older tables [memory, medium].) - Day-night: Meyl writes that "a Japanese team at Kamiokande reported ... at night only half as many solar neutrinos" (Neutrino Power p.28). For solar neutrinos the measured effect is a small night excess (section 5, step 8). The arbiter's reading is that Meyl's "half" most probably garbles the 1998 Super-Kamiokande atmospheric result, a deficit of muon neutrinos coming up through the Earth [web: https://link.aps.org/doi/10.1103/PhysRevLett.81.1562, abstract]; that identification is the arbiter's inference. Either way the explanation is flavour oscillation with the total number conserved, not absorption in the core.

4.6 The theory of objectivity

Meyl feeds E = v x B into H = -v x D, gets a field reduced by a factor (1 - v^2/c^2), and sets that equal to (l/l0)^2 with an equation marked "!", concluding that fields set lengths (E, H proportional to 1/l^2) and the speed of light (c proportional to l) [read: deep/Objektivitaetstheorie-1 s.3]. With the vectors he specifies (v along x, H along y, E along z) the product gives E(1 + v^2/c^2), not E(1 - v^2/c^2); at v = 0.99 c his formula gives 0.02, the sign-corrected form 1.98, and special relativity 7.09 [read: physicist.md claim 6]. The "!" step is a postulate. His Moon anecdote uses a factor 36 (in Neutrino Power, 2000, p.191, and in the Objektivitätstheorie papers) while his own proportionality (and his 2003 book, eq. 13.29) gives sqrt 6 = 2.45 [read: arbiter.md claim 6 and QE revision 2]. The difference matters for the anecdote itself: star brightness falls as 1/distance^2, so 36 times farther means 1296 times fainter (7.8 magnitudes; on the astronomers' magnitude scale 5 magnitudes is a factor of 100 in brightness, larger numbers are fainter, and the naked-eye limit is about +6), and Sirius at -1.46 would sink to +6.3, at the naked-eye limit; 2.45 times farther means 6 times fainter (1.9 magnitudes), Sirius +0.5, still one of the brightest points in the sky [calc]. On his own law the astronauts should have seen stars. Meyl also says, in the same book, that the law cannot be checked from inside: proving it in a laboratory "dürfte ziemlich schwierig bis unmöglich werden" (would be rather difficult to impossible), because every ruler and the observer change with the field; if c were not constant "we would in no way be able to grasp it, neither to observe nor to measure" [read: Neutrino Power pp.193-195]. Hold the two facts together: by its author's account the law can only be seen from outside, and its one outside number contradicts it. The same book adds particle-structure statements that can be checked against measured values (rows 6g'-6i in the table) [read: deep/Meyl-Neutrino-Power-2000-5d7650.md Part 3; arbiter.md QE revision 2]. The particle-mass tables use one anchor (m_p = 1836 m_e), take the muon as input (9 x 207), and allow free choice of shell structure ("an infinite lot of combinations are conceivable"); the one derived ratio, m_p/m_mu = 9, is 1.35% above the measured 8.88; the neutron-proton mass difference comes out 39 times too large [read: arbiter.md claim 6]. The "derivation" of the Schroedinger equation ends with a mass term of the wrong sign and a constant potential of 255.5 keV, i.e. a free wave, not an atom [read: deep/...603e47 s.6-9]. Meyl's Hafele-Keating figure of 214 ns is 273 - 59 with a sign dropped; the published west-east difference is 273 - (-59) = 332 ns [web: https://hyperphysics.gsu.edu/hbase/Relativ/airtim.html].

One clean prediction deserves credit: in a 2014 NET-Journal paper Meyl predicted that the proton radius measured with muons is smaller than with electrons by sqrt((m_p + m_e)/(m_p + m_mu)) = 0.9483, against the then-measured 0.84087/0.886 = 0.949 [read: research/2026-10-08-k-meyl-de.md; contrarian.md claim 6]. It was written after the 2010 "proton radius puzzle" was known, with an electron value at the high end. Later electron measurements (PRad 2019: 0.831 fm; CODATA 2018: 0.8414 fm) agree with the muonic value, so the ratio is now about 1.00 and the predicted 5% difference is not there [web: https://pdglive.lbl.gov/view/S016CR]. It is the programme's only specific, falsifiable number, and it failed on later data.

4.7 Biology and electrosmog

Meyl proposes that nerve conduction is a longitudinal standing wave with nodes at the nodes of Ranvier (1 mm spacing, 120 m/s gives 60 kHz), that mobile-phone frequencies "converted" with the speed of sound give tinnitus (32 cm wavelength / 343 m/s gives 1.07 kHz), that bioresonance and zappers work through the "scalar share" of fields, and that cancer is a neutrino-capture deficiency [read: physicist.md claim 7, digest]. The nerve arithmetic is right; the premise is contradicted by direct measurement of ionic currents through voltage-gated channels (Hodgkin and Huxley 1952), a single travelling pulse, not a standing pattern [memory, high]. Dividing a different speed by a radio wavelength does not change a radio wave's frequency. Meyl himself says the scalar share cannot be measured with present instruments, that frequency therapy "should not work" by present physics, and that the zapper also kills beneficial bacteria [read: claims.md]. Two of his biology papers (2012) were retracted in 2013 for fourfold duplicate publication, after a complaint by the president of the Austrian sceptics' society (an organised critic and so an interested party); no experiment was refuted, and one reviewer comment quoted by Retraction Watch found the work "way out there" but possibly right [read: research/2026-10-08-retractionwatch.md].

The Wiesbaden lecture (2001), now deep-read, offers no trial, patient data or measurement series for nerve conduction, tinnitus or cancer; the nerve figure is a worked number (node spacing 1 mm taken as half a wavelength, 120 m/s taken from physiology, so f = 120 / 0.002 = 60 kHz), and no 60 kHz signal on a nerve is offered. Meyl hedges in his own words: "wahrscheinlich" (probably), "möglicherweise" (possibly); bioresonance has "only a subordinate role" and can make patients worse; the zapper is a "shotgun into the forest"; there is an "enormous research need" [read: deep/Skalarwellenstrahlung s.1, s.4; arbiter.md claim 7]. The first version of the arbiter went further and said there were "no measurement data or trials in the documents", and the first version of this primer left out the same two items. That was incomplete: two items in the documents are tests of a kind.

Meyl's grounding-detection session (14.06.1997). At a weekend seminar a self-described "electrosensitive" woman was asked, at "now" prompts every 5 s, whether the receiver's grounding was clamped or not; she could see neither the audience nor the oscilloscope. "After a training round her proportion of hits was lying at 100 percent!" A second subject with a pendulum at the back of the room also answered correctly. Meyl adds: "I already have repeated this experiment several times with different success" [read: _text textbook PDF p.418; deep/textbook ch.19 s.1.6]. What is not given: the number of trials, how the on/off order was chosen, and the results of the later sessions. The experimenter, who knew the state, was in the room, the audience could see the oscilloscope, and clamping a lead can make a sound, so the test was not blinded on the experimenter's side [read: arbiter.md claim 7]. Single-source (Meyl) and unblinded as described. Arbiter: Never properly tested; what would settle it is a randomised, double-blind on/off sequence with a fixed number of trials set in advance.

The Bonn blood-rouleaux study (2000). Rouleaux ("Geldrollen", money rolls) are red blood cells stacked like coins. Live dark-field "vital blood" demonstrations at citizens' meetings against mobile-phone masts showed rouleaux after a phone call, and Meyl cites the "blood roll phenomenon" with mobile-phone fields [read: digest About-Vortex pp.6-7]. Frank Gollnick and Gabi Conrad of the Physiological Institute, University of Bonn, repeated the demonstration in their laboratory (phase contrast, "a multitude of test preparations", with and without a call, and a no-call control from the same person). They found rouleaux "entirely independent of the test conditions", different in each preparation and even between regions of one slide; "in no case" was any effect of a call found; and the subject with the strongest rouleaux had never used a mobile phone [read: deep/Newsletter Part A, pp.5-7]. Limits, stated plainly: the note gives no numbers of subjects or preparations, no call length or phone type, no blinding and no statistics; it tests the demonstration method, not whether radio fields act on blood (the authors say proper studies "are missing to date"). Stake: it was printed in the newsletter of the Forschungsgemeinschaft Funk (FGF), a Bonn association founded through the federal post ministry in 1992 whose members include network operators and manufacturers; the FGF says it commissions independent university work, and its own PR group records the accusation of lobbying. The funding of the Bonn work is not stated [read: deep/Newsletter A1-A8, B1-B7]. Meyl's own demonstration of the effect is not in the folder. Arbiter: the demonstration as proof of an effect is Tested, not supported (for the method only).

Stakes in claim 7. Meyl sells an SWG-A "scalar wave device" at 3950 EUR, used to place substances with "biological information" in the coil field [read: digest kit manual p.34, via qe.md]. His patent application DE 10 2014 109 660 A1 (filed 2014, withdrawn) describes information transfer by a modulated carrier in a self-resonant LC circuit with organic fluids in the coils, the patent form of remote "drug information" transfer [web: https://patents.google.com/patent/DE102014109660A1/de, via research/2026-10-08-google-patents-espacenet.md]. The Investigator also found plant ATP and plant-growth papers (2015-2016) [read: sites.md]; they were not reviewed here.

---

5. Worked example: does the Earth grow 18 cm a year by swallowing neutrinos?

This is Meyl's most concrete quantitative claim, and every step can be checked with a calculator. Source: "Erdwachstum durch Neutrino-Absorption", a paper for a colloquium on Earth expansion held 24-25 May 2003 at Schloss Theuern, co-organised by TU Berlin [read: deep/Meyl-Erdwachstum s.Document]. example.py Part A reproduces every number below.

Step 1. Meyl's model, in his words and symbols

The real physics: a body that grows while nothing twists it from outside keeps its angular momentum and spins more slowly. Meyl writes [read: _text p.7, checked against the page image by the deep reader]:

"Zur Drehimpulserhaltung muss J omega = (2/5) M R^2 (2 pi / t) = konstant sein. Aus der konstanten Relation R^n/t = (R+dR)^n/(t+dt) folgt dR = R[(1+dt/t)^(1/n) - 1]." ("For conservation of angular momentum, J omega = (2/5) M R^2 (2 pi/t) must be constant. From the constant ratio R^n/t = (R+dR)^n/(t+dt) it follows that dR = R[(1+dt/t)^(1/n) - 1].")

Symbols: J angular momentum (he writes "J omega"); M Earth's mass = 5.972e24 kg; R radius = 6.371e6 m; t the rotation period; dt its change; dR the change of radius; n = 2 if the mass stays constant (J ~ R^2/t), n = 5 if the density stays constant (M ~ R^3, so J ~ R^5/t). Check the formula: from R^n/t = (R + dR)^n/(t + dt), (1 + dR/R)^n = 1 + dt/t, which gives his dR. For small changes, dR/R = (dt/t)/n. The algebra is right. (Treating the Earth as uniform, I = 0.4 M R^2 instead of the real 0.33 M R^2 [memory, high], cancels out of the ratio.)

His input [read: _text p.6]: the PTB (Germany's national metrology institute) reported that since 1 January 1958 "a time difference of 32 seconds has accumulated" between atomic time TAI and UTC, which follows the Earth's rotation; "averaged over 45 years we have 0.71 s per year". A press note says "a day lengthens by 1/500 s, which corresponds to 0.73 s per year".

Step 2. Reproduce Meyl's numbers

Meyl does not say what t is. Only t = one year (3.156e7 s) reproduces his results [read: deep/Meyl-Erdwachstum s.4; computed]:

dt/t = 0.71 s / 3.156e7 s = 2.25e-8.

• n = 2: dR = 6.371e6 m x 2.25e-8 / 2 = 7.17 cm/yr; circumference 2 pi dR = 45.0 cm/yr. Meyl: 45 cm, "clearly too much".
• n = 5: dR = 6.371e6 m x 2.25e-8 / 5 = 2.87 cm/yr; circumference 18.0 cm/yr (18.5 with 0.73 s). Meyl: 18-19 cm.

His arithmetic is right. (If t were the day, 86,400 s, the n = 5 case would give 66 m/yr of circumference growth (10.5 m/yr of radius), which is absurd, so a year it must be.)

Step 3. What the 0.71 s actually measures

The formula needs dt/t = (change of the day's length) / (length of the day), for the same span of time. Leap seconds measure something else. Today's day is about 2 ms longer than 86,400 SI seconds; atomic clocks therefore gain about 2 ms on the Earth each day, and over a year that adds up to 0.71 s:

0.71 s / 365.25 days = 1.94 ms per day; 1.94 ms / 86,400 s = 2.25e-8.

That is exactly Meyl's dt/t. It is a level (how much longer the day is than a fixed standard), not a rate (how much longer the day gets each year). A day that were 2 ms long and never changed would still produce 0.71 leap seconds every year, and by Meyl's formula zero growth. The press note "a day lengthens by 1/500 s" (2 ms) is the same level read as a change [computed; read: arbiter.md correction 6].

The rate is measured separately: the day lengthens by about 2.3 ms per century (tidal prediction) or 1.7 ms per century (eclipse records) [web: section 3.6]. So

dt/t = 1.7e-3 s / 100 yr / 86,400 s = 1.97e-10 per year (2.66e-10 for 2.3 ms).

Meyl's input is 114 times (or 85 times) too large; "about 100" [computed]. Put the other way: Meyl's dt/t would lengthen the day by 86,400 s x 2.25e-8 = 1.94 ms every year, against about 0.02 ms per year observed.

Step 4. Growth from the correct input

With n = 5: dR = 6.371e6 m x 1.97e-10 / 5 = 0.25 mm/yr in radius, 1.6 mm/yr in circumference [computed]. Even this assumes that all of the slowing comes from growth. It does not: tidal friction from the Moon removes the Earth's spin angular momentum at about the observed rate, and the Moon's orbit grows by 3.8 cm/yr to carry it [web: section 3.6]. Meyl's angular-momentum budget leaves the Moon out (he dismisses tidal friction as a cause of the extra mass, not as a sink of angular momentum) and assumes that the mass arriving from outside brings no angular momentum of its own [read: deep/Meyl-Erdwachstum s.4]. Meyl did respond to the tidal point in a 2005 note, accepting 1.7 ms/century as "negligible against the measured fluctuation of 7.3 s per year"; the 7.3 s/yr is again an accumulated quantity, not a yearly change of the day [read: research/2026-10-08-k-meyl-de.md].

Step 5. The mass and energy 18 cm a year would need

Constant density means M is proportional to R^3, so dM/M = 3 dR/R:

dM = 3 x (0.0287 m / 6.371e6 m) x 5.972e24 kg = 8.06e16 kg per year.

Meyl: "8x10^16 kg bzw. 72x10^32 Nm Energie" (8e16 kg, or 7.2e33 J) [read: _text p.7]. His energy is E = dM c^2 = 8.06e16 x (2.998e8)^2 = 7.25e33 J per year. Meyl stops here. Turn it into a power: 7.25e33 J / 3.156e7 s = 2.30e26 W. The Sun's whole output is 3.83e26 W [memory, high], so the Earth would have to absorb 60% of the Sun's luminosity, continuously [computed].

Step 6. What all the solar neutrinos can supply

Meyl's 1996 patent: 66e9 neutrinos per cm^2 per second, "about 27 W per m^2" if every one were converted (27 W/m^2 / 6.6e14 m^-2 s^-1 = 4.1e-14 J = 0.26 MeV each). The Earth intercepts the flux over its cross-section, pi R^2 = 1.275e14 m^2:

P_supply = 27 W/m^2 x 1.275e14 m^2 = 3.44e15 W.

Even if the Earth swallowed every solar neutrino and turned all of its energy into mass, it would fall short by 2.30e26 / 3.44e15 = 6.7e10, i.e. about seventy billion times [computed; the arbiter's 6.6e10 differs only in rounding]. Meyl writes that "entirely different orders of magnitude are collected" here, without a number [read: _text p.7].

Step 7. How many neutrinos actually stop

The Earth's mean density is 5,514 kg/m^3. A path through the centre crosses 5,514 x 2 x 6.371e6 = 7.0e10 kg/m^2 of rock; with about one electron per two nucleons (6.0e26 nucleons per kg), that is 2.1e37 electrons per m^2, or 2.1e33 per cm^2. With a cross-section of about 1e-45 cm^2 [memory, medium], the chance that one neutrino interacts on the way through is 2.1e33 x 1e-45 = 2e-12 (2e-11 if the cross-section is nearer 1e-44) [computed]. The power actually deposited is then of order 3.44e15 W x 2e-12, about 7 kW for the whole planet, an upper bound since a scattered neutrino keeps part of its energy [computed]. Meyl's answer is that resonant coupling is far stronger than the weak interaction; he never states how much stronger. On this budget it would have to be about 5e11 times stronger just to stop every neutrino (1 / 2e-12), and even then the supply falls short by 7e10 (step 6); for Earth growth the coupling would have to beat the weak interaction by about 3e22 [computed].

Step 8. Compare with measurement

Meyl's radius growth, 28.7 mm/yr, against space geodesy's 0.1 +/- 0.2 mm/yr (Wu et al. 2011): 290 times the central value and about 140 standard deviations from it [computed; web: section 3.6]. The Super-Kamiokande day-night measurement, which Meyl cites as showing "only half as many solar neutrinos at night" (absorbed in the core), in fact shows a night flux 3.6-4.2% higher than the day flux (about 2.8 sigma), explained by flavour regeneration through the Earth [web: arXiv 1606.07538, via contrarian.md]. (Meyl's "half" is most probably a garbled report of the 1998 atmospheric-neutrino result, also explained by oscillation, not absorption; see the day-night note in section 4.5.)

What the example teaches. The angular-momentum reasoning and the algebra are right. The error is one misread input: a level (how much longer the day is than a standard) used as a rate (how fast the day is getting longer). That one slip makes the input about 100 times too large, and the conclusion then needs 60% of the Sun's power, which the neutrinos cannot supply by ten orders of magnitude or more. The arbiter's verdict: Earth growth of 18-19 cm/yr by neutrino absorption is Refuted by calculation and Tested, not supported (space geodesy, lunar laser ranging, solar-neutrino day-night data). The historical expanding-Earth hypothesis of Hilgenberg and Carey, which Meyl credits as "the pioneer" while rejecting Hilgenberg's aether mechanism, is Historically superseded but instructive [read: arbiter.md claim 5].

Run it yourself: .venv/bin/python review/Meyl-Konstantin/example.py. (Parts C and D of the same file reproduce the eclipse-lens numbers of section 4.5 and the probe-skew numbers of section 4.4.) Try changing true_ratio or the cross-section sigma to see how little the conclusion depends on the exact inputs.

---

6. Evidence table

Verdicts are the arbiter's final verdicts (arbiter.md), on the STEP 2 scale. "IGF" = Institut fuer Gravitationsforschung, Waldaschaff.

---

7. What a learner should take away

What is solid. - The identity Laplacian(E) = grad div E - curl curl E, and the transverse/longitudinal split, are correct. Longitudinal electric waves exist wherever there is charge to carry them (plasmas); near fields of antennas have components along r. - The algebra from Meyl's extended laws to his "fundamental field equation" is correct. It is the telegraph equation with an extra postulated term. - Faraday's motional emf is a legitimate starting point, and a magnetic-current term is a legitimate formal extension; both need a measured constant to mean anything. - "Measure two of v, lambda and f" is a sound rule, and it is the rule the 1.5 c claim breaks. - Resonant capacitive power transfer with a return wire is real, works well at short range, and is what Meyl's own granted patent describes. The kit's effects are real and reproducible. - Meyl's 1996 patent puts the right number on the solar-neutrino energy flux (27 W/m^2), and his 2014 proton-radius ratio was a genuine, falsifiable prediction. His eclipse forecasts were also falsifiable, and he invited readers to check them. - Meyl reports some of his own negative results plainly: the 1990 capacitor-Q difference that "drowned" in the scatter, the 3 kW converter he could not reproduce, the three inventors he told had "measured wrong", the Seike rebuilds that did not work, and the manual's own "unplug the cable and the receiver goes out". These are honest data points and belong in any fair weighing.

What is wrong, and why the mistakes are instructive. - Two meanings of "d/dt" (moving observer vs fixed point) silently swapped: the most common error in motional-emf arguments. - Treating a constraint (Gauss's law) as an oversight. When a term is zero in a derivation, ask which law made it zero. - A sign that sets a direction: the 1.618 c pattern moves inward; outward gives 0.618 c. Always check which way your trial solution moves. - A result fixed by a choice: the golden ratio follows from setting tau = r/c, not from Maxwell. - A level used as a rate (leap seconds), worth a factor of 100 (section 5). Check the units and the meaning of every input, not only the algebra. - Efficiency from a small difference of two large numbers (49/(233-223)): the error of the difference can exceed the difference. - A ratio of two resonance frequencies read as a ratio of speeds: true only if the wavelengths are shown to be equal. - Not carrying a calculation to the end: Meyl computed 7.2e33 J per year himself; turning it into watts and comparing with the Sun would have shown the problem. - Measuring real power at a power factor near zero: when voltage and current are almost 90 degrees apart, picoseconds of probe skew swamp the answer (section 4.4). Measure the phase error before trusting a small difference. - Answering the wrong mechanism: the first review refuted the eclipse "lens" with gravity, which Meyl never proposed. A refutation has to address the model actually claimed (section 4.5). - An unblinded on/off test (1997) and one dated event per eclipse cannot settle anything either way; the remedy is a blinded, randomised design and a forecast rule fixed in advance for every case.

What is genuinely open. - No one has built Tesla's full-scale magnifying transmitter; the IGF itself called for it. - The raw Treskatis data and the vortex-loss formula (PIERS 2012) have never been fitted against standard relaxation plus series resistance; the microscopic origin of near-flat dielectric loss in polymers is itself not settled in mainstream physics. - No clean measurement of the kit exists with a low-harmonic source, no LEDs, calibrated power at both ends, the link wire removed and a sealed enclosure for the cage test. The arbiter calls it one day's work in an RF laboratory. - A tuned "neutrino receiver" has never been placed in a known neutrino flux (a reactor) with an on/off control; that test would turn the enhanced-coupling idea from "not testable as stated" into a testable claim. - The record goes quiet after 2001 (kit) and after about 2014 (Meyl's output). The arbiter sets out competing explanations without choosing: nothing more to find; institutional distance (Furtwangen's dean wrote in 2002 that independent opinions were negative "in Meyl's sense" and in 2008 the chancellor barred him from lecturing on scalar waves; the university is an interested party in its reputation); commercial incentive (kits and devices still on sale in 2026); personal circumstances (a 2014 stroke reported by one critic wiki, Psiram, with no second source; retirement 2018); and records never made public (the Clausthal lab notes of 6 July 2000, EMC laboratory reports, INDEL customer records). The records that would decide are listed in arbiter.md, "Absence-of-evidence analysis". One item of reception history, in Meyl's own account only (Neutrino Power p.214, an interview book; no second source found): his paper "Potentialwirbel in der Elektrodynamik" was submitted in April 1990 and rejected; the journal, Annalen der Physik, is named by the interviewer, and Meyl does not correct it. As Meyl reports the grounds, "it was argued" (no one named) that Maxwell's theory covers all phenomena, and "a referee writes" that cases it does not cover only need better approximation methods. No rejection date, referee report or second source is given [read: Neutrino Power pp.213-214; arbiter.md QE revision 2 s.2]. - Eclipse effects: a proper test of Meyl's forecasts (regions and windows fixed in advance, every eclipse checked against USGS and IERS) has not been run; nor has a neutral analysis of Zhou's clock series. - Leads found by the Investigator and not yet weighed in depth: Meyl's late patent DE 10 2021 127 321 A1 (2021, withdrawn; magnetised gas before compression in an expansion engine); a 2005 Hamburg lecture hosted by an industrial association and a 2008 Südwest-Messe show (interest, not tests); the Bearden "MEG" and Evans "AIAS" circle cited in support at a 2003 congress (DGEIM); a critique by Kühlke, a Furtwangen physicist (not online); H. J. Herrmann's statement, via Bruhn, that his Stuttgart institute never worked on scalar waves; the PIERS Prague 2015 proton-size paper; and the press items Meyl lists (Spiegel 21.5.2001; Südkurier "Gebäude der Physik wankt" 23.4.2003; a DVR note of 4.4.2003) [read: sites.md; arbiter.md QE revision s.8].

---

8. Check yourself

1. Polypropylene has eps_r = 2.2 and tan delta = 2e-4. Meyl says a complex permittivity implies a complex speed of light. What does it imply instead, and how fast does light travel in the material?
2. Using Meyl's own book figures (49 mW out, 233 mW in with the receiver on, 223 mW in with it off), compute the efficiency two ways. If each power reading is uncertain by 5%, how uncertain is the 10 mW difference?
3. Take Meyl's damped equation c^2 E_xx = E_tt + (c/r) E_t and the outward-moving profile E = E0 exp(-(x - vt)/r). Derive the condition on v and solve it.
4. Suppose the day were exactly 2 ms longer than 86,400 s and stayed that way for ever. How many leap seconds would be added per year, and what Earth growth would Meyl's formula then give if fed correctly?
5. Meyl's kit runs near 7 MHz. How far does the reactive near field reach? His manual tests distance-independence by lengthening the ground cable to a little over 6 m. Why does that not show transmission "far beyond the near field"?

---

9. Further reading

Mainstream. - D. J. Griffiths, Introduction to Electrodynamics (4th ed.), ch.7 (Maxwell's equations), ch.9 (waves, including conductors and dispersion), ch.11 (radiation), ch.12 (relativity and field transformations). - J. D. Jackson, Classical Electrodynamics (3rd ed.), ch.7 (waves in media, dispersion, signal velocity) and ch.11 (special relativity). - J. R. Taylor, Classical Mechanics, ch.10 (rotation of rigid bodies). - D. Kleppner and R. Kolenkow, An Introduction to Mechanics (rotation and angular momentum). - A. K. Jonscher, Dielectric Relaxation in Solids (1983), for dielectric loss. - A. Kurs et al., "Wireless power transfer via strongly coupled magnetic resonances", Science 317, 83 (2007). - J. N. Bahcall, Neutrino Astrophysics (1989); Super-Kamiokande day-night analysis, arXiv 1606.07538. - X. Wu et al., "Accuracy of the International Terrestrial Reference Frame origin and Earth expansion", Geophys. Res. Lett. 38, L13304 (2011) [title memory, medium]. - H. G. Winful, "Tunneling time, the Hartman effect, and superluminality", arXiv quant-ph/0601085, for the mainstream reading of the tunnelling experiments. - J. C. Hafele and R. E. Keating, Science 177, 166-170 (1972) [memory, high on journal and volume].

Meyl's own key works (in the project folder unless noted). - Scalar Waves: first Tesla physics textbook for engineers (INDEL, 2003): ch.5 (fundamental field equation), ch.19.11 (p.411, the efficiency measurement), ch.27-29 (field equation, 1.618 c). - "Scalar Waves: Theory and Experiments", J. Scientific Exploration 15(2), 2001, printed with G. W. Bruhn's critique. - Scalar Wave Technology: Documentation for the Experimental Kit (2000). - Neutrino Power (interview book, 2000). - "Erdwachstum durch Neutrino-Absorption" (2003) and "Earth Expansion by Neutrino-Power", J. Geol. Geophys. 4:5 (2015, OMICS). - "The Transition from Far to Near Field calculated by Vortex ...", Fluid Mechanics: Open Access 3(1):122 (2016), the 1.618 c paper. - Patents DE 19614575 A1 (1996, neutrino receiver), DE 10348862 B4 (granted 2009, energy and data transfer) and DE 10 2014 109 660 A1 (filed 2014, withdrawn; information transfer with organic fluids in self-resonant coils, the patent form of claim 7) [web: Google Patents, https://patents.google.com/patent/DE102014109660A1/de]. - Chapters 19-20 of the 2003 book (PDF pp.401-455): Meyl's own experiments of 1989-1999 and the 1999 eclipse forecasts. - Independent tests: IGF reports of 2001 (H. Weidner, "Analyse eines Teslaschwingkreises", 10 Aug 2001, Wayback copy in research/files/; six-author report, Sept 2001, in the folder as "Experimente zum Nachweis von Skalarwellen"). Gao Peng, viXra 1607.0130 (2016). - Critiques: G. W. Bruhn, TU Darmstadt web pages (2001-2004) [read: research/2026-10-08-tu-darmstadt-bruhn.md]. - Tests of claim 7: F. Gollnick and G. Conrad (Physiological Institute, University of Bonn), the rouleaux ("Geldrollenbildung") article in Newsletter 1/2000 (April 2000) of the Forschungsgemeinschaft Funk, Bonn, pp.4-7 (in the folder; an industry-member publisher).

Other primary sources used. - N. Tesla, "Famous Scientific Illusions", part III, Electrical Experimenter, February 1919 (the Kelvin visit) [web: http://www.tfcbooks.com/tesla/1919-02-00.htm]. - N. Tesla, US 787,412 (1905) [web: https://patents.google.com/patent/US787412A/en]. - Shu-wen Zhou, eclipse clock and spectral anomalies, 21st Century Science & Technology 12(3), 1999 (not obtained). - USGS earthquake catalogue (https://earthquake.usgs.gov/earthquakes/search/) and IERS length-of-day series (https://hpiers.obspm.fr/iers/eop/eopc04/), for checking the eclipse forecasts yourself.

---

Answers

1. A complex permittivity gives a complex wavenumber k = omega sqrt(mu eps): the real part sets the wavelength, the imaginary part an exponential attenuation. The vacuum constant c is untouched. Light travels at c/n with n = sqrt(2.2) = 1.483, i.e. 2.02e8 m/s, and its amplitude fades with an extinction coefficient of about n tan delta/2 = 1.5e-4.

2. Output over input: 49/233 = 21%. Output over the change in input: 49/(233 - 223) = 4.9, Meyl's "490%". A 5% error on 233 mW is 11.7 mW and on 223 mW is 11.2 mW; if the errors are independent, the difference 10 mW carries an uncertainty of sqrt(11.7^2 + 11.2^2) = 16 mW, larger than the difference itself. (If the same meter drifts identically in both readings, the errors partly cancel; that is why a calibrated measurement at both ends is needed.) The record itself has a bigger problem than a 5% meter error: at a power factor of 0.055 the 10 mW difference equals only 56 ps of skew between the voltage and current probes (section 4.4; example.py Part D).

3. For E = E0 exp(-(x - vt)/r): E_x = -E/r, E_xx = E/r^2, E_t = +(v/r) E, E_tt = (v^2/r^2) E. Substituting: c^2/r^2 = v^2/r^2 + (c/r)(v/r), so v^2 + vc - c^2 = 0 and v = c(-1 + sqrt 5)/2 = 0.618 c. The outward-moving pattern is slower than light. (example.py, Part B.)

4. 2 ms x 365.25 = 0.73 s per year, i.e. a leap second roughly every 16 months, for ever. But the day would not be changing, so dt/t = 0 and Meyl's formula, fed with the rate of change of the day, gives dR = 0. This is the level-versus-rate error in section 5.

5. lambda = 3.0e8/7e6 = 42.8 m and the near field reaches about lambda/(2 pi) = 6.8 m. A separation of a little over 6 m is about one near-field radius, not many, and in any case the cable itself is a conductor joining the two circuits, so power reaches the receiver by conduction whatever the distance through the air.