RankingsPrimerArbiterPhysicistMathematicianContrarianQESites logClaimsSourcesResearch notes

Physicist review (Step 2): Frank Znidarsic

Reviewer role: PhD physicist. Sources: facts.json, claims.md, the two deep reads, digests, and the Investigator's findings (sites.md "Leads for review", research/ notes, research/files/). Tags: [read] from a project document, [web] with URL, [memory] with confidence, [calc] my own arithmetic (project venv, CODATA constants). The Step 1b deep reads supplied page-image transcriptions of the equations; I re-ran their key numbers and found them correct (see "Corrections to Step 1" for what I changed).

No source gap: both papers have full text layers. Claim 6 (pp. 9-15 of "The Control of the Natural Forces") rests on the digest only; the paper itself gives no calculation there, so nothing was lost.

0. The idea in plain terms, and the one fact that decides most of this review

Znidarsic (an electrical engineer, Johnstown PA, later Alstom Power; see sites.md) proposes a "transitional velocity" Vt = 1.094 x 10^6 m/s. He says it is the speed of sound inside atomic nuclei and that quantum behaviour (Planck's constant h, hydrogen orbit sizes, the fine-structure constant, de Broglie waves) appears when the speed of light "in the electron shell" equals this nuclear sound speed.

Fact that decides the matter: [calc] Vt as he computes it equals e^2/(4 eps0 h) = alpha c / 2 = 1,093,846 m/s, where e is the electron charge, eps0 the vacuum permittivity, alpha = 1/137.036 the fine-structure constant and c the speed of light. That is exactly half the speed of the electron in the Bohr ground state of hydrogen (alpha c = 2.19 x 10^6 m/s) [memory, high; the arithmetic reproduced by me]. He builds it from three tabulated numbers: fc = 1.236e20 Hz (the electron Compton frequency m_e c^2/h, which contains h), rp = 1.409e-15 m (= half the classical electron radius, no h), and Fmax = 29.05 N (the Coulomb force between two electron charges one classical radius apart, no h) [read: deep/Control p.14; calc]. Once h is in fc, every "derivation" of h, of the Bohr radius and of alpha from Vt returns what went in. That is the central finding; the claim-by-claim verdicts below follow from it.

The Investigator independently found the same identity stated by the Natural Philosophy Wiki (Chappell Natural Philosophy Society; a dissident-science society, a third party with no stake in Znidarsic's work) [web: https://wiki.naturalphilosophy.org/index.php?title=Frank_J_Znidarsic]. That wiki also credits Znidarsic with being quantitative and with reporting null results openly; I agree with the second point, and it is recorded in the evidence section.

1. Worked numerical checks (reader-facing)

  1. Vt from fc and rp: 2 pi x 1.236e20 Hz x 1.409e-15 m = 1.0942e6 m/s [calc]. Using exact constants: alpha c/2 = 1.093846e6 m/s. The two differ only through rounding of fc and rp.
  2. "Planck's constant emerges": e^2/(4 eps0 Vt) = (1.602e-19)^2 / (4 x 8.854e-12 x 1.094e6) = 6.625e-34 J s [calc]. But because Vt was built using h, this is h returned, not derived.
  3. The Bohr radius from Eq. 19: Fmax rp^2 / (Vt^2 m_e) = 29.05 x (1.409e-15)^2 / ((1.094e6)^2 x 9.109e-31) = 5.29e-11 m [calc]. Substituting the identities above, this is exactly e^2/(4 pi eps0 m_e alpha^2 c^2) = a0. Again the input was alpha and h.
  4. The only step that is not an algebraic rearrangement (Eq. 6 of Control, Eq. 5 of Flow): Vt = (1/pi) sqrt(Fmax rn / (2 m_nucleon)). To reproduce alpha c/2 it needs rn = 1.3597e-15 m (proton mass) [calc]; he uses 1.36e-15 m and gets 1.0948e6 (proton or "1.67e-27 kg") or 1.0932e6 (neutron mass, Flow). The two papers use different nucleon masses and so differ by 0.15%, which is as large as the agreement itself. Sensitivity [calc]: Vt scales as the square root of rn. rn = 0.84 fm (proton charge radius) gives 0.86e6 m/s; rn = 1.2 fm gives 1.03e6; rn = 1.36 fm gives 1.094e6. So any plausible nuclear length lands at "about 1e6 m/s". Only the third significant figure needs rn = 1.36, and rn is a free choice (the paper gives no calculation; see Claim 1).
  5. Nuclear sound speed: from the standard incompressibility K ~ 230 MeV the sound speed in nuclear matter is c sqrt(K/(9 m c^2)) = c sqrt(230/(9 x 938.9)) = 0.165 c = 4.9e7 m/s [calc; K = 230 +/- 40 MeV and sound speed c_s of order 0.2 c at saturation density, roughly half that in finite nuclei: Nuclear incompressibility and sound speed in uniform matter and finite nuclei, https://arxiv.org/pdf/2207.01884 and https://journals.aps.org/prc/abstract/10.1103/PhysRevC.106.044305, from my web search; I read only the search summary, not the papers]. So the real nuclear sound speed is roughly 30 to 50 times larger than Vt.
  6. Fine structure: the claim that (longer line - shorter line)/longer line equals alpha. Hydrogen n=2 fine-structure splitting is about alpha^2 x 13.6 eV / 16 = 4.5e-5 eV against the 1.89 eV H-alpha photon, a fractional spacing of 2.4e-5, about alpha^2 (5.3e-5), not alpha (7.3e-3) [calc; physics: memory, high].
  7. Josephson gap (Flow Eq. 27): 2 x (4 pi hbar/e)^2 (eps0/m_e) with e = 1.6e-19 C gives 1.333e-9 m; with the stated e = 3.2e-19 C it gives 3.33e-10 m [calc, reproduces the deep read]. And 1.33e-9 m = 8 pi a0, a multiple of the Bohr radius [calc].

2. Claim-by-claim review

Claim 1: Vt ~ 1.094e6 m/s was observed in LENR and Podkletnov-type experiments, and is the speed of sound in the nucleus

(a) Mathematics. The statement "empirical" in Eq. 1 is not supported by anything given. The numbers that make 1.094e6 to four figures are fc and rp, not measurements [read: deep/Control sec.2]. The "observations" are order-of-magnitude products: LENR 50 nm x omega_N (1e13 to 1e14) = 5e5 to 5e6 m/s; Podkletnov 3 MHz x 1/3 m = 1e6 m/s, which uses frequency x diameter, whereas Eq. 1 uses angular frequency x radius, a factor of 2 pi difference [calc; read]. In Flow the superconductor product 1e14 x 1e-9 = 1e5, not 1e6 [calc]. Also nx/ni ~ 1e-24 in Flow is not reproducible (Eq. 1 gives 2.8e-8) [read: deep/Flow sec.4]. Internally the "sound speed" and the "speed" in the other two anchors are different kinds of quantity: omega x r is the rim speed of something rotating at 1e13 rad/s, a sound speed is omega/k. Hence the three numbers are not shown to be the same physical speed. (b) Experiments. Real nuclear matter sound speed is about 0.1 to 0.2 c (3e7 to 5e7 m/s), see calc 5. Ordinary phonon speeds in metals and ceramics, including Pd and YBCO, are a few km/s [memory, high], which also contradicts "the speed of a longitudinal sound wave across atomic distances within the dissolved deuterium" (Control p.2). 1e6 m/s is an electron velocity scale: Fermi velocities of metals are of order 1e6 m/s [memory, high], and Bohr velocity is 2.2e6 m/s. Znidarsic's own book cites Charles P. Poole's Superconductivity (2007, p.4) as saying 1e6 m/s is "the speed of light within a superconductor" [read: research/files book; Investigator]. I did not check Poole's text. My expectation [memory, low] is that this is a Fermi velocity or a similar electronic speed, which would put Vt in the electron system, the opposite of "nuclear". The Eq. 6 agreement uses Fmax, an electron-electron Coulomb force (29 N), as the nucleon "spring" force, with the claim that it equals the strong-force spring constant where the forces balance [read: Control p.3]. Nucleon-nucleon forces at 1 fm are of order 10^3 N or more [memory, medium], and no calculation shows 29 N is the relevant value. The one numerical coincidence (Eq. 6, 0.1%) rests on rn = 1.36 fm, tagged "the nuclear Fermi momentum spacing" and sourced to Modarres, Europhys. Lett. 3, 1083 (1987) [read: Control p.15]. I could not read Modarres. My check: Fermi momentum of symmetric nuclear matter at density 0.16 fm^-3 is kF = (3 pi^2 x 0.16 / 2)^(1/3) = 1.33 fm^-1 [calc]; 1.33 to 1.36 fm^-1 is the standard figure [memory, medium-high]. That is an inverse length. A nuclear length of 1.36 fm also exists independently (r0 ~ 1.2 fm is the usual nuclear radius parameter; 1/kF = 0.75 fm) [memory, high for r0]. So the coincidence may be a unit mix-up (kF in fm^-1 read as a length in fm), or just a chosen length. This is a hypothesis flagged by the deep reader and not confirmed by me. (c) Tests. Podkletnov's gravity-shielding results (Podkletnov and Nieminen, Physica C 203, 441, 1992; the Control paper lists the second author as "Levi" [read: Control p.15], which I think is a citation error: [web: https://researchportal.tuni.fi/en/publications/a-possibility-of-gravitation-force-shielding-by-bulk-yba2cu3o7-x-/ lists Podkletnov and Nieminen]) have not been replicated by others in the sense of a confirmed effect. Tests found: NASA MSFC, Noever and Koczor 1999, gravimeter change below 2 parts in 1e8 of g with AC fields on levitated YBCO [read: research/files NTRS-19990023209; Investigator]; Hathaway, Cleveland and Bao, Physica C 385, 488 (2003): a rotating superconducting disk with radio-frequency fields, "no evidence of a gravity-like force to the limits of the apparatus sensitivity" [web: https://www.osti.gov/etdeweb/biblio/20682562; ADS https://ui.adsabs.harvard.edu/abs/2003PhyC..385..488H/abstract, summary from a search result; I did not read the full paper or its RF frequencies]. Znidarsic's own bench record: a single anomaly "just above the noise level" at 11 MHz with a 3.5-inch disk, later "We cannot now repeat this" (NEN 5(9), Jan. 1998) [web: https://web.archive.org/web/20080105120935/http://padrak.com/ine/NEN_5_9_7.html]; and a spinning-superconductor/RF run to ~1,000 rpm, 40 MHz: "No anomalous energy was observed" (NEN Mar 1999) [read: research/files/NEN-1999-03.pdf]. The "3 MHz" in Podkletnov's disk: I recall the RF excitation is described in Podkletnov's later reports (1997 arXiv and 2001-2003 papers), not in the 1992 paper that Znidarsic cites for it [memory, medium; not checked]. LENR: the 50 nm figure comes from a Jed Rothwell remark (Infinite Energy 29, 1999) [read: Control p.15], a popular source. No LENR experiment is analysed and none of its numbers enters an equation. LENR itself (Fleischmann-Pons-type excess heat) is a contested field, outside this review. (d) If true. The nucleus would need a sound speed 30 to 50 times below the one inferred from nuclear incompressibility, or "sound speed" would mean something other than the bulk-modulus quantity. Nuclear incompressibility, giant monopole resonance energies and equation-of-state fits would all have to change. Much more simply, nothing needs to change if Vt is read as an atomic electron scale, which it is.

Verdict, split: - "Vt is the speed of sound in the nucleus": Refuted by calculation (nuclear matter sound speed ~0.1-0.2 c, Vt ~ 0.004 c). - "Vt was observed at about 1e6 m/s in LENR and Podkletnov-type experiments": Never properly tested for LENR (the number is an order-of-magnitude product from uncited inputs; no measurement of the claimed speed was made); Tested, not supported for the Podkletnov-type gravity effect that is supposed to anchor it (Noever-Koczor 1999, Hathaway et al. 2003, the inventor's own two null runs). One marginal 1997 anomaly is on record in the inventor's own letter.

Claim 2: Quantum transitions occur where the speed of light in the electron shell equals the nuclear sound speed (impedance match)

(a) Mathematics. Eqs. 7-15 are valid algebra but rest on asserted steps: the photon "capacitor" has plate area = lambda^2, separation = lambda (Control) or lambda/2 (Flow, a different choice, which changes the answer by a factor of 2 unless compensated), a factor 2 for two polarisations, and wavelength = Vt/f instead of c/f (so lambda_t is not the photon's wavelength, which is c/f) [read: deep/Control sec.5; deep/Flow Eqs.7-9]. The two papers use different geometries for the same step and both claim to get h. Eq. 3 does not reproduce Coulomb's energy (Eq. 2): it gives one quarter, a factor 4 that is hidden later by a matching factor 4 in Vt^2 [calc; deep/Control sec.3]. Control uses 1/(2 pi) in Eq. 5-6 and omega in Eqs. 1, 16, 17 without justification. Symbols rp, M-e, nB are undefined in Flow. "Impedance match" is never expressed as an equation with numbers: Eqs. 21-22 are the standard transmission-line formulas (wave speed 1/sqrt(LC), impedance sqrt(L/C)) with no L or C assigned [read: deep/Control sec.7]. (b) Experiments. Atomic transitions are accurately described by quantum mechanics and QED (hydrogen levels, Lamb shift, g-2 of the electron to ten significant figures) [memory, high]. The Znidarsic scheme reproduces only the Bohr-model radii and levels, which the Schroedinger/Dirac theory replaced (hydrogen ground state has zero orbital angular momentum, not Bohr's one unit; the radius is an expectation value, not an orbit) [memory, high]. His model offers no prediction of anything outside the Bohr level scheme: no selection rules, no spin, no fine-structure splitting calculation, no Lamb shift, no transition rates (he says the intensity "requires no probability waves" but only shows a classical amplitude-squared sketch, Control pp.10-12 [read: digest]). Also the "transitional state frequency" ft is stated not to be the frequency of any stationary state, but photon frequencies in spectra are differences of stationary-state energies, as observed. (c) Tests. None that could distinguish it from the Bohr model; the model has not been put to any measurement beyond the radius/energy numbers that are inputs. Nobody has replicated or independently tested the impedance-match condition. The paper itself notes that LENR and gravity results it relies on are "assumed speculative by mainstream science" [read: Flow p.1]. (d) If true. Quantum mechanics would have to be replaced by a classical-wave theory that nonetheless reproduces its detailed successes (spin statistics, Bell-inequality violations), and the electron's "elastic constant" and "discontinuity rp" would need independent evidence (e.g. electron scattering structure: the electron is point-like down to below 1e-18 m [memory, high], while rp is 1.4e-15 m). The paper's non-local "wavefunction collapse with superluminal speed" remarks are interpretation, not measurable.

Verdict: Not testable as stated in its strong form (the "impedance match" has no equation and no prediction outside the Bohr numbers); the Bohr-radius content is Historically superseded but instructive (the Bohr/Sommerfeld model); where it is stated precisely (Eq. 3, the capacitor geometry) it is Internally inconsistent (factor 4 in Eq. 3 vs Eq. 2; two different capacitor geometries between the two papers).

Claim 3: h, the hydrogen radii, alpha, the Compton frequency and the de Broglie wavelength all follow classically from Vt

(a) Mathematics. Each step is algebraically right and circular. E = h f (Eq. 15), r = n^2 a0 (Eq. 20), alpha = 2 Vt/c (Eq. 23) and fe = 1.6448e15 Hz (Flow Eq. 26, which is half the Rydberg frequency, 3.29e15 Hz [calc]) are each Vt = alpha c/2 rewritten. Flow's Eqs. 1, 13, 16 are the same number by identity, not three coincident measurements [read: deep/Flow sec.5]. The only number not obtained that way is Eq. 6/5 (see Claim 1). The identification of n with "the number of electrons" in r = ne^2 r+h (Flow Eq. 17-18) is wrong: in Bohr's formula n is the principal quantum number [read; memory, high]. The de Broglie step (Flow Eqs. 19-23, Control Eqs. 35-41): the algebra is right (with fc = mc^2/h) [calc], but fc already contains h, so de Broglie lambda = h/(mv) is again built from h. The attribution to de Broglie's 1924 thesis of a "Doppler-shifted reflection" is unchecked [memory, medium]. Eq. 19B gives the hydrogen-like n^2 a0/Z; the multi-electron generalisation ("fractional values of frequency") and the muonic-atom statement are asserted with no numbers [read: Control p.7]. (b) Experiments. The values of h, a0 and alpha are among the best-measured constants, but nothing here is a new measurement. As a "classical origin of alpha" the fine-structure argument fails against spectra: real splittings are of order alpha^2, not alpha (calc 6), and field-induced splitting is Zeeman, not fine structure [memory, high; Control p.8]. (c) Tests. None are possible, since all outputs are inputs; no one has replicated a derivation because it does not derive. (d) If true. If Planck's constant truly were fixed by a nuclear sound speed, then h, alpha and the nuclear incompressibility would be linked: a change of nuclear matter properties would shift atomic spectra. No such link is observed; atomic and nuclear scales are separated by the factor alpha-squared times m_e/m_p scale hierarchy that the standard theory explains.

Verdict: Internally inconsistent as a derivation (it assumes what it claims to derive; plus the n = number of electrons and the factor 4 in Eq. 3). The numerical Eq. 6 coincidence: Never properly tested as a physical claim; it is best described as an unmotivated numerical match (see "Look-elsewhere note" in sec. 3).

Claim 4: the Josephson-junction scale (~1.33 nm) and the LENR domain share the same ~1e6 m/s speed, so LENR releases are discrete bursts like electron-pair transitions

(a) Mathematics. 1.33e-9 m = 8 pi a0 follows from Vt = alpha c/2, so "ballpark of the junction gap" is a match to a number chosen to be the same atomic scale (calc 7). The stated input e = 3.2e-19 C (Cooper pair charge) does not give 1.33 nm; it gives 0.333 nm, and 1.33 nm requires e = 1.6e-19 C [calc]. The Robertson-type product omega_e rJ = 1e14 x 1e-9 = 1e5 m/s, an order of magnitude below the claimed 1e6 [calc]. "fe of the range of the LENR" is 100 to 1000 times higher than the LENR frequency given in the same paper; "about twice that of the gravitational anomaly experiments" has no frequency to support it [read: deep/Flow Eq. 26]. (b) Experiments. Josephson tunnel barriers are about 1 to 3 nm thick and high-Tc coherence lengths about 1 to 2 nm [memory, medium], so a 1 nm scale is common to nearly all nanoscale structures. A match to a nanometre is not specific. The Cooper-pair "fluctuation time ~1e-14 s" is taken from co-author G. A. Robertson's own then-unpublished chapter [read: Flow p.3, ref 7], published in 2012 in Bentham's "Gravity-Superconductors Interactions" [web via Investigator: research/2026-10-04-booksellers-bentham.md]. Peer review of Flow was "under responsibility of" the Institute for Advanced Studies in the Space, Propulsion and Energy Sciences, Robertson's own institute [read: Flow p.1]. I report this as a fact about the publication (Physics Procedia 20, 457, 2011, SPESIF 2011), not as a judgement. (c) Tests. None. No LENR measurement of burst energy or timing is given, and LENR excess heat is generally reported as continuous or with unknown time structure, which this paper does not address. (d) If true. LENR would have to show quantised, electron-pair-sized energy bursts at nanometre scale, an observable (time-resolved calorimetry) that nobody has reported.

Verdict: Internally inconsistent in its numbers (stated e vs result; 1e5 vs 1e6; frequency ranges), with the match being an identity (8 pi a0); the physical statement "LENR releases are discrete bursts" is Never properly tested.

Claim 5: at transition speeds >= ~1e6 m/s new phenomena occur and "free" (vacuum/dark) energy can be pulled from ~1e-9 m interactions

(a) Mathematics. None: "no energy balance, power estimate or equation for energy extraction from the vacuum" [read: deep/Flow sec.7]. The authors hedge ("would seem to imply", "may lead to"). (b) Experiments. Established vacuum phenomena (Casimir force, Lamb shift) do not give extractable net energy; energy extraction from a ground state is forbidden by definition of ground state [memory, high]. "Dark energy" is a cosmological term, with a density of about 6e-10 J/m^3 [memory, high], so a nanometre-scale device sees a negligible amount. (c) Tests. None. The paper has no experiment. (d) If true. The first law, or the identification of the quantum vacuum as the lowest-energy state, would need a mechanism that allows net extraction.

Verdict: Not testable as stated (no mechanism, no quantity). A strong form (net energy gain from the vacuum) would be contradicted by energy conservation unless a mechanism is shown; the weak form ("new phenomena at 1e6 m/s") is too vague to test.

Claim 6: driving a macroscopic body (trillions of atoms) into one transitional state by Bose-condensate vibration at ~1.094 MHz-m would induce strong gravity and long-range nuclear forces (propulsion, waste reduction, energy)

(a) Mathematics. Digest only (Control pp.10-15): there are no equations for the induced force, its magnitude or the gravitational coupling. The "theorem" that "constants of motion tend toward electromagnetic in a Bose condensate stimulated at 1.094 MHz-meters" has no derivation in the pages seen [read: digest]. The rule f x L = 1.094 MHz-m versus 37 MHz-inch = 0.94 MHz-m in the 1997 letter [web] shows the constant is an empirical rule that moved by about 15% between the inventor's accounts. (b) Experiments. General relativity gives the gravitational field of an object from its mass-energy; a vibrating or rotating superconductor produces gravitomagnetic fields of the London-moment kind that are tiny, of order 1e-30 to 1e-20 of what a "strong gravitational anomaly" requires [memory, medium]; Tajmar and de Matos did measure an anomalous acceleration signal near rotating superconductors in 2006, not replicated (see Investigator note on the Frontiers 2022 reference below). I did not verify this. Long-range nuclear forces contradict the nuclear force's short range (about 1 to 2 fm, the Yukawa range set by the pion mass) [memory, high]. (c) Tests. By the inventor: the 1997 anomaly and its non-repetition, and the 1998-99 null (cited above). The RF configuration at NASA MSFC was reportedly never completed, per the inventor's book (his account, no second source) [read: research/files book; Investigator]. Hathaway et al. 2003 tested a rotating superconducting disk with RF [web, above] and found null. Noever and Koczor 1999 null. BAE Greenglow: the inventor reports Ronald Evans said the RF method was not attempted for funding reasons (book, inventor's account). So the strong claim has been tested in the form of rotating/vibrating superconductors with RF, by the inventor and by an independent group, without positive result; the specific "1.094 MHz-m" scaling on a larger scale has not been tested by anyone but the inventor. The Nature 516 (2014) paper (nonlinear lattice dynamics and enhanced superconductivity [memory, medium]) cited in the book is light-induced transient superconductivity in a lattice, which is an interesting but different effect and does not bear on gravity or nuclear forces. (d) If true. A new coupling of laboratory-scale condensate states to gravity (or to the strong interaction) would be needed; the equivalence principle and GR's weak-field limit would have to be supplemented.

Verdict: Tested, not supported for the gravity/propulsion version (the inventor's own null runs; Noever-Koczor 1999; Hathaway et al. 2003), with the caveat that exact Znidarsic-specified parameters were not shown to have been used by others. The "nuclear waste reduction and new energy sources" parts are Never properly tested (no experiment shown).

3. Look-elsewhere note on the Eq. 6 coincidence

If rn is accepted as a free choice in 0.8 to 1.5 fm, the match to ±0.1% in Vt (±0.2% in rn) has probability roughly 0.4/70 ~ 0.5% by chance for any one fixed target [calc, rough]. That is low enough to be worth a check but not low enough to ignore that (i) Fmax, rp, the choice of the factor 2rn, the use of a cyclic frequency without 2 pi, and the use of a proton or neutron mass were all chosen by the author and each shifts the answer by factors of 2 or 2 pi, so the effective number of trials is much larger; (ii) the proton and neutron versions differ by the same 0.1% (see calc 4). I regard it as an unexplained numerical match with no physical derivation behind it. It is retained in the record, not dismissed.

4. Investigator's findings, weighed (kept in full)

Inventor's own work - 1997 letter (NEN 5(9), Jan. 1998): patent application filed 19 Oct 1997; rule f(MHz) = 37 / L(in); anomaly at 11 MHz with 3.5-inch disk, not repeatable; the inventor himself reported that "latest tests showed no anomalous energy" [web]. Weight: primary and candid; it shows Vt began as one marginal bench observation and then an order-of-magnitude product, which supports the Step 1 reading of Claim 1. - NEN Mar 1999 null result for rotating superconductor and RF [read]. Editor Hal Fox set it next to the "Aspden Effect" at 3500 rpm and an unpublished report of gravitational effects only above 10,000 rpm. That the disk reached only ~1,000 rpm is a limit on what the null excludes (a stated rotation-rate caveat from the editor, not a rebuttal). - Book "Energy, Cold Fusion, & Antigravity" (2012, rev. 2019) [read]: the inventor's account of 1992 (Puthoff, Austin), 1995 (Patterson and Miley, Anaheim), 1997 (Wright-Patterson), 1998 (NASA MSFC). These are his account only; I did not find second sources. The account of how Vt arose (Podkletnov's disk, Poole's remark, then LENR) differs from the 1997 letter (own 3.5-inch disk). Both accounts are his; the order of discovery differs. - Papers not yet obtained: Elementary Antigravity (1989), Zero-Point Interaction (JNE 1996), Force and Gravity (Infinite Energy 22, 1998), Constants of the Motion (2000), 2005 and 2006 General Science Journal papers, 2010 NPA lecture. Their content could change the Claim 1 or 6 assessment if they contain measured data; none was read.

Supporting information by others - Robertson, NASA MSFC 1999 (NTRS 19990046249) [read]: a Josephson-junction mass-change mechanism (from Woodward's transient mass shift) and a proposed experiment. It is a theoretical proposal, not a result. Cited by Gallerati et al., Frontiers in Physics 2022 (Investigator). - Names of contacts (Puthoff, Patterson, Miley, Evans): the inventor's account only; recorded, no corroboration found yet.

Closely related work - Noever and Koczor 1999; Li, Noever, Robertson, Koczor, Brantley, Physica C 281 (1997), which I did not open; Hathaway et al. 2003 (above). - Natural Philosophy Wiki assessment: Vt = alpha c/2; the empirical anchors are not secure; credits candour. - Poole's Superconductivity p.4: unchecked (see Claim 1).

Historical record - US patent application of 19 Oct 1997: no grant or publication found. Competing explanations: (1) abandonment or rejection (the inventor says the patent office would not grant cold-fusion patents; applications filed before 29 Nov 2000 were unpublished unless granted [memory, high]); (2) no commercial interest after the null results he himself reported in 1998-99; (3) secrecy or suppression: no evidence for it found, and none expected for an unpublished application by a private inventor. Decisive record: the USPTO file wrapper (not reached by the Investigator). Physics note: the physics in the claimed process (RF vibration converting reactive power to real power) violates energy conservation unless a mechanism supplies net energy; the 1997 letter itself reports no net energy. - NASA MSFC programme 1997-1999 and DIA 2010 survey ("The Role of Superconductors in Gravity Research", https://www.dia.mil/FOIA/FOIA-Electronic-Reading-Room/FileId/237631/): appeared in my search results, not read. A DIA survey is an institutional characterisation by a government agency (an interested party only in the sense of being a defence-intelligence reader of the field). - The research ran as spare-time work next to a career, without institutional affiliation (the wiki's biography).

5. Corrections to Step 1

  1. claims.md item 3 and the deep reads are right on the identities; I reproduced Vt = alpha c/2 = 1,093,846 m/s, rn_required = 1.3597e-15 m, the 0.333 nm result at e = 3.2e-19 C, and the fine-structure ratio. No numerical error found.
  2. claims.md item 1 says Vt "is the speed of sound in the atomic nucleus" and the deep read called the Eq. 6 match "genuinely non-circular". I qualify: it is non-circular only in the sense of not being a rearrangement of Eq. 1; it relies on a free nuclear length (rn) and a force (29 N) not shown to be the strong-force spring constant, and the two papers differ by 0.15% depending on which nucleon mass is used. It does not show that Vt is a nuclear sound speed, and the real nuclear sound speed is 30 to 50 times larger (calc 5).
  3. The Control paper cites the 1992 Physica C 203 paper as "Podkletnov & Levi"; the 1992 authors are Podkletnov and Nieminen (the Flow paper has it right) [web: https://researchportal.tuni.fi/en/publications/a-possibility-of-gravitation-force-shielding-by-bulk-yba2cu3o7-x-/]. Step 1 recorded the citation as printed without comment.
  4. sites.md says the wiki "cites a Hathaway, Cleveland and Bao null result" and treats it as unchecked. I confirmed that the paper exists and that it used a rotating superconducting disk with RF fields, which is the closest independent test of the configuration that Znidarsic proposes [web: OSTI link above, search-result summary only]. I did not read its frequencies, so I cannot say whether it covered 1.094 MHz-m; that remains open.
  5. facts.json/claims.md say the Podkletnov 3 MHz x 1/3 m is Znidarsic's "observation". It is Podkletnov's reported value as relayed by Znidarsic, who cites the 1992 paper; I believe the RF details come from later Podkletnov reports [memory, medium; unchecked]. Znidarsic's "I had previously observed this speed within Podkletnov's disk" (book) is his claim and unconfirmed.
  6. The deep read of Flow lists the Josephson input as "ne = 2, charge 3.2e-19 C" which I checked (Claim 4). Step 1 should list the mismatch under claim 4 rather than only in the deep read. claims.md item 4 states the 1.33 nm figure without noting it fails with the stated charge.
  7. claims.md item 6 rests on the digest only (acknowledged). I add that the "theorem" and the 1.094 MHz-m value are stated without equation in the pages digested, so I could not check them against the page images; this is a limitation on Claim 6, not a reason to credit it.

6. What a learner should take from this review (short)

  • A "velocity" built as frequency x length is not a wave speed unless the frequency and length belong to the same wave.
  • When a constant such as h is hidden inside an input (here fc = m c^2/h), "deriving" h from it is circular. Always ask which numbers contain what.
  • A match to 0.1% means little when the inputs include a freely chosen length and several factors of 2 or 2 pi.
  • The inventor's honest null results (1997-1999) are part of the record and should be weighed beside the large claims.