Meyers-Stan (Stanley A. Meyer, "Water Fuel Cell"): Arbiter (Step 3)

Arbiter, 2026-10-09. Inputs: facts.json, claims.md, physicist.md, mathematician.md, contrarian.md, deep/, sites.md, research/. Tags: [read] source text (via the reviewer or myself); [calc] my own arithmetic (python, .venv); [web] URL; [memory] with confidence. Verdict scale from prompt.txt.

Source gap (repeated from claims.md)

• No digests exist for this project. (Corrected in the QE revision, section 14: Step 0 was re-run at 02:20 on 2026-10-09 and a manifest.json now exists, 66 documents and 84 chunks.) Triage used the log, 19 deep reads and the fact-finder's reading.
• Every Meyer document in the folder (patents, newsletters, the 222-page "Birth of New Technology" brief) is Meyer's own characterization. His stake: patent holder, seller of dealerships and "profit-sharing certificates". The folder holds NO measurement, test report, lab report or independent witness statement about any device.
• (Superseded by the QE revision, section 14: the New Zealand video transcript, the 39 photographs, newsletters No. 4 and No. 8, EP 0 111 573 A1, Birth pp.126-222, US 4,389,981 and Puharich book pp.176-204 have now been deep-read.) As first written: not read by anyone: the 2 h 38 min 1989 New Zealand video (not transcribed), the 40 photographs, newsletters No. 4 and No. 8 beyond a page-image read, Puharich's 344-page book beyond selected pages. Birth pp.57-222 were read selectively by the three reviewers (Memo 424 pp.117-125 in full by all three; Memo 426 and Appendix A by two). Memos 425, 427-430 and the appendices are not read in depth.
• The strongest numbers in the whole review (Meyer's 1995 energy table, Perkins, Nagypal, the court docket) come from stanslegacy.com, an advocacy archive with a stake, in machine transcription ("unverified transcription") of Meyer's own binder. Where a verdict leans on them, I say so. The scan images should be checked. (Note: the energy table is not in any numbered arbiter section by itself; see sections 2 and 3 and the physicist and contrarian files, binder pp.59-60. [corrected 2026-10-09 after QE: an earlier note pointed to "section 11"].)
• Conclusions that rest on [memory] alone: the O-H bond-scale field (about 10^9-10^10 V/m, medium confidence), tap-water conductivity as a typical range (now web-checked), barrel of oil = 6.1 GJ (high, not rechecked). No verdict below depends on the first of these by less than two orders of magnitude.
• The NIST Office of Energy-Related Inventions (ERIP) evaluation, the DTI/Queen Mary report, the Fayette County decision transcript and the probate valuation were NOT obtained. Where they matter I say "second-hand".

1. How the reviewers line up

2. Numbers where reviewers differ, recomputed [calc]

Constants: F = 96,485 C/mol; molar gas volume 24.5 L at 25 C (24.0 L at about 16 C); water enthalpy of combustion 285.8 kJ/mol (HHV), Gibbs energy 237 kJ/mol; 1 US gallon of water = 3.785 kg = 210.1 mol; barrel of oil = 6.1 GJ.

3. Claim 1. Water capacitor splitting ("resonant charging choke", pulsed unipolar high voltage, "without electrolysis")

Where they agree. - The series-LC formulas (resonant frequency, V_L, V_C) are textbook-correct in form. The printed Eq 9 adds the dielectric constant (78.54, dimensionless) to ohms; Memo 426 repeats "78.54 ohms"; Eq 17 gives watts as A^2; Eq 14 has a typesetting exponent error. [read: Birth pp.8-15, 142-143, per all three] - "At resonance voltage is theoretically infinite" holds only at zero resistance. Meyer builds the chokes from resistive wire. - At kHz, tap water is a resistor, not a capacitor. Cell conductance divided by cell susceptance is sigma/(omega eps0 er), independent of geometry. With the numbers in section 2, the cell Q is below 4 even for theoretically pure water. - No neutral test with simultaneous input energy, net charge and gas volume exists. What exists (Hindley/Laughton/Griffin 1990, Perkins 1992, Nagypal 1991, Lawton builds) is qualitative or lacks the energy figure. - Meyer's own table (binder pp.59-60), once the pressure unit and the 10 cc/Ah baseline are corrected, matches ordinary electrolysis at about 40 A (28.6 vs 27.4 L/h).

Where they disagree, and why. 1. How to label the strong form. The physicist says "Refuted by calculation" (field about 10^3-10^5 below bond scale). The contrarian says "Never properly tested". The mathematician splits it. Stronger argument: the physicist and mathematician are right about the bulk field mechanism and about the 20-40 kV resonant gain, because both rest on arithmetic with accepted laws and neither needs an experiment. The contrarian is right that the weaker statement "this circuit gets more gas per joule than a plain DC cell" has never been measured. These are different claims and Meyer's documents slide between them, so I give each its own verdict. 2. Field versus bond scale. Applied bulk fields are 2e5 V/m (650 V over 3.2 mm) to 1.3e7 V/m (20 kV over 1.5 mm). Field-enhanced water dissociation (second Wien effect) appears at about 10^7 V/m for classic weak electrolytes and above 10^8 V/m for water itself [web: https://arxiv.org/pdf/2405.05882 ; https://arxiv.org/pdf/2208.10828]. Bond-scale fields are about 10^9-10^10 V/m [memory, medium; not web-checked]. Even with the smaller contrarian figure the bulk field is 100 times too small. Local field enhancement at electrode points and breakdown streamers could be larger; neither reviewer excluded it, and Electronics World's own description ("the water breaks down and a momentary high current flows") is a repetitive breakdown regime. That is plasma/discharge chemistry in a spark channel, which is still charge transport and still energy-accounted. 3. Nagypal. Physicist: unresolved factor 2-4 in a crude measurement; "300%" is against a 10-12 W reference, i.e. about 3% of the thermodynamic minimum. Mathematician and contrarian: same numbers. I accept all of it. Nagypal's report is a single published source (letter of 4 Jun 1991, transcribed by stanslegacy). It stays in the record as the one number in the file that exceeds a Faraday estimate for the stated current. It is weakened by (i) the current figure is not tied to a waveform-aware meter, (ii) gas read to about 1 cc by eye, (iii) dissolved-air release and heating also give gas, (iv) he is a prospective distributor, and (v) he writes that he cannot reproduce Meyer's waveforms and that "it works in another way". Even if real, 1 cc/min is 0.13 W of gas energy, and 600 V x 20 mA = 12 W if that is the supply side. His own statement of "20 mA" does not say whether that is cell current or supply current. 4. Perkins "less than 1 mA" flame. The contrarian's best case: a sustained flame at a milliampere reading would be a violation of Faraday's law by a factor of 2,600 (a 0.5 mL/s flow needs about 2.6 A of net charge). I agree the arithmetic is correct. But the flow rate, duration and meter location were not recorded, and a thin stainless wire melts in a very small oxy-hydrogen flow. It is an unquantified observation, not a measurement of anomaly. 5. Hindley's "independent confirmation by patents". The physicist and contrarian both note it is Hindley's own characterisation; USPTO practice presumes utility unless doubted [web: https://www.uspto.gov/web/offices/pac/mpep/s2107.html, as the physicist checked]. I accept that and keep Hindley's eyewitness report as an observation of a cold cell and a melted steel bar, which an ordinary small electrolyser also produces. 6. The NIST ERIP evaluation. All three treat "net consumer of energy, not an energy producer" (Spy, Dec 1991; Heard 2000, citing a NIST letter of 28 Jun 1991) as second-hand. I repeated the web check; it failed again (a search found no NIST/NBS page on Meyer; one item found only Wikipedia and a fan-archive account) [web: search 2026-10-09, no source]. NIST/NBS as author is a government agency characterisation, no known stake, method unseen. It is not decisive, and I do not use it to reach any verdict above.

What the strong form needs from physics. Either a charge-free bond-breaking channel (not found) or a thermodynamic bound broken. The only quantitative claim in the physicist's and contrarian's work that stays inside accepted physics: a pulsed charge-pump cell may beat a badly built DC cell, up to the reversible 120% of the Gibbs-based figure (dH/dG = 1.207) when heat comes from the surroundings. The mainstream literature does not exclude "pulsing helps a bad cell" (Shimizu-type reports; a recent review argues pulsing cannot beat DC for a standard convex cell model: Puteanus et al., IJHE 2025, read from the saved PDF by the physicist and mathematician).

Final verdicts for claim 1 (split): - Pulsed high-voltage drive across a water cell releases H2 and O2 (weak form): Established (as electrolysis and discharge chemistry; the precipitate in Electronics World is direct sign of Faradaic reaction at the electrodes). - Resonant LC voltage gain to 20-40 kV across water-filled concentric tubes, as described: Refuted by calculation (cell Q at most 4; holding 20 kV costs 2.6 kW to MW; 40 kV at 1 mA implies a 40 Mohm cell, 260 times above pure-water resistivity limits). One escape is open and recorded: microsecond transient voltages shorter than the water relaxation time. Meyer's pulse lengths are unstated, so this cannot be tested from the documents. - Bulk-field, charge-free bond breaking ("covalent bond ceases to exist"): Refuted by calculation (bulk field 100 to 10^5 times too small). - Gas yield per joule above ordinary DC electrolysis, or any gas energy above input: Never properly tested. Not dismissed without testing, but the testers whose reports exist (Hindley, Perkins, Nagypal) measured current or gas, never both with input energy; the negative government evaluation and the DTI/QMW report are second-hand. Any result above 286 kJ/mol of gas out for each mole in would break energy conservation. - The printed equations (Eq 9, "78.54 ohms", Eq 14, Eq 17): Internally inconsistent.

What would settle it. One bench run, 100 seconds long: a Meyer-patent cell and drive (Lawton/Nagypal geometry), tap and deionised water; a power analyser giving the integral of v(t) i(t) dt at the supply, a fast scope at the cell, net DC charge through the cell, gas collected over water in a burette, a plain-DC control in the same cell, and a flow or ice calorimeter. Compare gas moles with Faraday (0.686 L per A h of net charge) and the gas's 285.8 kJ/mol with the total input. Records: the USPTO file wrappers of US 4,936,961 and 5,149,407 (the patent examiner's file would show whether a demonstration or affidavit was filed); the NIST ERIP file; the DTI/Queen Mary report; the Laughton 13 Apr 1992 report.

4. Claim 2. "Hydrogen fracturing": energy beyond combustion (44,000 to 2.5 million barrels per gallon of water)

Agreement. The three agree completely. Memo 424, "Atomic Energy Balance of Water" (the memo whose title promises the balance), contains no calculation: the figure "exceeds 2.5 million barrels" is repeated, E = h nu and p = h nu / c appear with Planck's constant printed as 6.547e-27 erg s (accepted 6.626e-27), and E = M_d c^2 is printed on the cover without evaluating M_d [read: Birth pp.117-124; Birth p.22]. The earlier claims.md request for a deep read of "the full energy calculation" is closed: none exists in Memo 424.

Calculation (section 2). 2.5 million barrels is 2.5e8 times the chemical value of the gallon, 4.5% of its entire rest-mass energy, 57 times the ceiling for fusing every hydrogen nucleus to helium (43,700 bbl), and 6e5 times the energy of fusing every deuterium nucleus. The older figures (44,000-108,000 barrels, 1987-88) are 10^6-10^7 times the chemical energy; the lowest quoted figure happens to coincide with the hydrogen-to-helium ceiling (a coincidence: nothing in the documents says Meyer derived it so). Meyer's own injector figure (7.4 uL of water per cycle runs a 50 hp VW) is at the chemical scale: 117 J per injection, 3 to 13 times too little. The two figures are separated by eight orders of magnitude and are never reconciled.

Self-contradiction. Meyer writes that the process "does not create energy" and "complies with the Laws of Thermodynamics" while also tapping "Universal or Zero Point energy" through an "energy aperture" [read: Newsletter 11a pp.2, 9; Birth]. He calls the effect nuclear ("nuclei decay") in one place and chemical in another. The one rider the contrarian adds, and I keep: the claim that the burned gas is hotter than an ordinary hydrogen flame (beyond 5,000 F) is not anomalous, because stoichiometric H2/O2 reaches about 2,800 C (5,000 F) [memory, high; physicist rates the adiabatic figure medium].

Evidence for. None in the folder. The electrically exploded water reports Meyer cites (Graneau et al., Phys. Lett. A 1986; Johnson, Kansas State) were not read by any reviewer; the contrarian and mathematician note they describe electrical discharge energy, not nuclear release. Kucherov US 5,632,870 (cites Meyer) draws ambient heat, a different claim capped by the 120% bound. The reports are kept in the record.

Verdict: Refuted by calculation (the quantitative barrel figures). The "complies with thermodynamics" and "Zero Point energy" statements together: Internally inconsistent; the thermodynamics statement alone: Not testable as stated.

What would settle it. A bomb calorimeter on the burned gas of a measured water volume with the electrical input logged; the pass line is more than 285.8 kJ/mol out. The decisive calculation is already done: no process described could cross the eight-order gap.

5. Claim 3. Water-fuel injection and vehicles

Agreement. Meyer's Appendix A [read: Birth pp.220-221]: 111 ml/min of gasoline replaced by 44.4 ml/min of water ("3.3 amp/hr", which should be a current; the "740 ml/sec" slip first recorded here was an OCR loss, the page prints ".740 ml/sec", see section 14); 100 injection cycles/s; 7.4 uL per cycle; "40,000 volts @ 1 ma = 40 watts" (arithmetically right). Water at best carries 11.7 kW of combustible energy (2.46 mol/min x 285.8 kJ), about one fifth of the gasoline it replaces (59 kW); splitting it costs at least 9.7 kW (Gibbs) against the stated 40 W (ratio 240-290). "Water is 2.5 times more powerful than gasoline" holds only for pure hydrogen per kilogram (about 120 vs 44 MJ/kg); per kilogram of water it is about 16 MJ/kg, 0.36 of gasoline [memory, high]. The 25 miles per litre statement (quoted by Griffin) needs 0.8 to 1.5 times the whole chemical energy of a litre of water at 100% engine efficiency (contrarian's assumption: 8-15 kW road load, medium). WO 92/07861's "about 5 gallons of water per hour" at 0.65 W drive power is a factor of about 10^5 from the thermodynamic minimum of 83 kW [read: p.13-14; calc].

Disagreement. Only on the size of the shortfall (section 2: depends on the engine load assumed). It does not affect the verdict.

Tests. No dynamometer, fuel-flow or gas-supply log. The folder's own newsletter (1994) says the buggy was still being retrofitted "to match the acceleration performance of gasoline-driven cars". The Perkins note of 28 Feb 1993: "On none of the 3 occasions, October 1990, August 1992 and October 1992, when UK experts visited Ohio, was there a practical demonstration of a running internal combustion engine fuelled by water" (a party that wanted to proceed; reported as its characterization). Meyer's explanation for refusing a test in 1990 is the "national security review"; the two foreign-filing licences on Birth pp.1 and 18 are routine permission to file abroad [memory, high; contrarian], and the two pages print different years (15 Nov 1988 vs 1989), which needs checking against the scan. In favour: a 1985 ABC news item on the dune buggy (claimed, not in the folder; about 9 of an alleged 3 hours survive per a secondary source); one first-hand recollection (Charles Hughes, Columbus Dispatch 2007: the tractor ran 15 minutes on well water, "no fumes whatsoever"), recalled 20 years later, with no measurement and no second witness found. I keep it in the record as a witness account.

Verdict (split): - Water as the net energy source of a road vehicle, using Meyer's own flow and power numbers: Refuted by calculation. - The specific buggy or tractor running on water: Never properly tested (no neutral, instrumented run exists; UK experts report not being shown one). - An engine supplemented by hydrogen from an on-board electrolyser: Plausible, untested (not Meyer's claim; no free energy).

What would settle it. A fuel-flow plus electrical-power log from one engine run (44.4 ml/min, 40 W, 50 hp per Appendix A); a dynamometer run with the battery and alternator on the books. Records: the Perkins TACO outcome, the 1985 ABC/Columbus TV footage, the Laughton refusal story.

6. Claim 4. Electrical Particle Generator ("gas battery", "voltage amplification")

Agreement. The patents give no voltage, current, turns, velocity or material [read: deep read CA 1,213,671]. Faraday induction and transformer action explain a voltage step-up without a power gain. The estimates of the gas's magnetic moment agree in order of magnitude with a negligible secondary signal: oxygen susceptibility about 2e-6 SI [memory, medium] gives M of about 1.5 A/m in 1 T and a self-field about 10^-6 of a magnet's; ferromagnetic powder or spark-gap vapour (claim 16 of the patent) would induce more, but is "not net generation". Newsletters later call it a "gas battery" that "produces electric energy directly from the fuel cell gases" with no mechanism.

Verdict (split): voltage step-up by induction: Established; net power gain: Not testable as stated (no quantity is given; any quoted output above input would be refuted by energy conservation); the device as built and measured: Never properly tested (no test or measurement exists in any source).

What would settle it. One measured prototype with input and output power on a power analyser; any numbers at all in the patents.

7. Claim 5. Resonance in the electrode gap (1982 form: EP 0 103 656, CA 1,234,773, US 4,798,661)

Agreement. No dimensions, frequency, equation or data. The only quantity that links a gap to a frequency is an ion transit time. The reviewers' three results (section 2) are consistent: 4 Hz at 12 V/1 mm; about 23 Hz at 650 V/3.2 mm; 160 Hz at 1 kV/1.5 mm; 3 kHz only at 20 kV/1.5 mm. The contrarian read the 3 kHz figure as "a point in favour of a physical picture". I do not. Reaching that field needs 20 kV across the water, which the mathematician showed cannot be held in conducting water. In liquid the ion motion is viscous drift with no inertia, so "velocity increasing geometrically" has no meaning; the water's own dielectric relaxation is near 19-20 GHz [web: https://arxiv.org/pdf/1704.01667, as checked by the mathematician]. The ion mobility value I use is checked: the proton's 36.24e-8 m^2/(V s) at 298.1 K [web: https://download.uni-mainz.de/fb09-ak-roesch/Roesch_Puplikationen/113 Mauerhofer-2003-PCCP.pdf, found in search; the page was not opened].

US 4,798,661 gives a ladder of onset voltages (2.5, 4, 5.5, 7, 8.5 V) with distilled water and "negligible" current. This is the only quantitative data in the 1982 family; it is in kind consistent with the onset of ordinary electrolysis (1.23 V plus overpotentials), with no gas or current figures.

Meyer's later documents replace this mechanical resonance with the electrical LC resonance of claim 1; it was a stage of his thinking.

Verdict: Not testable as stated (no dimensions, frequency or data). A nearby testable form (does pulse frequency versus gap change gas yield per joule?): Never properly tested. Historically superseded within Meyer's own work, but not by a test.

What would settle it. Gas yield per joule against pulse frequency and gap in one cell with a power analyser and a DC control.

8. Claim 6. Puharich (US 4,394,230, 1983), resonant splitting, 80-100%, "114.9%" (not a Meyer claim; Meyer's US 4,798,661 cites it)

Disagreement and resolution. The contrarian calls the accounting "internally consistent" and says 114.9% is below the reversible 120% bound, so not over-unity. The physicist and mathematician say the arithmetic uses the wrong gas volume. I checked the deep read of the patent [read: deep/...Puharich..., lines 36-37, 56]: the patent divides 236,954 J/mol (Gibbs) by 22,400 cc to get 10.58 J/cc, and 286,021 J/mol (enthalpy) by 22,400 cc to get 12.77 J/cc. But 1 mole of liquid water gives 1.5 mol of gas (1 H2 + 0.5 O2), 33,600 cc at STP, and the patent itself writes "33,400" once. With 33,600 cc: 7.05 J/cc and 8.51 J/cc. Then the measured 16.2 cc holds 137.9 J of enthalpy = 76.6% of the 180 J input, and is 63.5% of the Gibbs ideal. The 114.9% and the 91.3% disappear. The contrarian's point that 114.9% is below 120% is true but beside the point, because the figure itself rests on the slip. The input is also rms V x rms I on a waveform the patent says is nearly 180 degrees reactive (no cos phi), so real power is unstated. The contrarian's Faraday check (49 mA mean net current needed for 10.8 cc H2 at 289 K; I get 48 mA at 24 L/mol) cannot be compared to a 25-38 mA rms reading of a modulated waveform.

Other points kept. Parameters in the patent differ in different places (resonance 600, 3,980, 1,272-6,474 Hz). GC/MS in "two laboratories", unnamed, +/-10%, no replication on record. The "tetrahedral" 109.5 degree resonance has no support in molecular spectroscopy (water modes are in the THz-infrared range) [memory, high]. Author's own patent text is the source (stake).

Verdict: the 91.3% and 114.9% figures: Internally inconsistent (a unit slip; corrected to 63% and 77%). The underlying claim (resonant splitting at 80-100% efficiency by audio-frequency drive): Never properly tested by neutral parties.

What would settle it. Repeat the 30-minute, 100 mW run with a power analyser (true power), gas burette and a plain-DC control.

9. Meyer's claims of outside confirmation and finance (not numbered claims)

• "Laboratories around the world confirm": the third-party items found are Hindley/Laughton/Griffin 1990, Perkins 1992, Nagypal 1991 and an unseen Sussex "breadboard" report. None reports input energy against gas energy; Griffin's own paper says he never got more than a third of the electrical energy back from another cell, "about the norm for conventional electrolysis" (contrarian), and states electrolysis needs "roughly 3 times the energy of the hydrogen released", which is wrong for practical electrolysers at 60-80% [memory, high]. Griffin, Nagypal and Perkins each had a stake in proceeding. The contrarian's case for the claims also names five British professionals who travelled to Ohio and declined to dismiss it; a DTI-sponsored Queen Mary and Westfield report "left open the possibility" while being "unable to reconcile" the descriptions with established doctrine (per Perkins; unseen).
• The Einsiedeln 1989 talk: "once you hit resonance we can excite it for 5 seconds and you are producing gas for 94 seconds" [read: deep read]. The Electronics World witnesses said gas stopped instantly when the drive was switched off. Bubbles from dissolved gas can explain post-drive gas. Meyer's version is not a measurement.
• Finance: dealerships, profit-sharing certificates ($100-$1,000,000; 8% of net profits per year up to 300%); costs quoted vary between issues; patent counts vary (about 40; 42; 25; 26; the brief lists 13 US patents). The court dispute was about the dealers' purchases (below).
• The court: Fayette County CVH19930292, Willis and Harper v. Meyer, judgment 24 Sep 1996 for the plaintiffs, $40,515.72 garnished; Meyer's own account says the judge ruled the cell "only electrolysis" and that the court audio was switched off (his allegation, interested party; no second source found). The Sunday Times (1996) says three court experts found "nothing revolutionary". Neither is the decision. The decision transcript (11 Sep 1996) and the Final Judgment Entry are not obtained; exhibits and depositions were destroyed 18 Jun 2014. A fraud judgment on a sales contract is not a physics test.

10. Absence-of-evidence analysis: where the record goes dark

There are four silences.

A. No energy-per-gas measurement, 1982-1998. For 16 years engineers, an admiral, a professor and a company visited the cell. No document reports a gas-volume rate with input energy. The magazine editor asked exactly that question in Jan 1991. Competing explanations and evidence: 1. Nothing remarkable existed (ordinary electrolysis). For: the only worked numbers reduce to electrolysis once units are corrected; the precipitate is chemistry at the electrodes; UK experts regarded WFC as breaching the first and second laws (Perkins); NIST reportedly found "a net consumer". Against: not an explanation of why Meyer's sympathisers did not measure. 2. Trade secrecy. For: Meyer "declines to release details" (Electronics World); Nagypal could not reproduce the waveforms. Against: a patent applicant withholding the details that would make the claim testable invites exactly this suspicion; secrecy does not explain why a gas burette and a wattmeter were not allowed. 3. Measurement was made and was unfavourable (NIST, DTI, the court's three experts). For: second-hand quotations. Against: none of the reports is in hand. 4. The instruments could not read pulsed power. For: Hindley's "mean current in milliamps" with tens of thousands of volts pulsed; Puteanus et al. say gains in some earlier studies came from computing power as mean V times mean I. Against: not a reason no better meter was used in 16 years. 5. Suppression. For: Meyer's and the family's accounts only (buy-out offers of $250 million to a billion; "They poisoned me" recollected by a relative; court recorder off). Against: no document of any blocking of a test; the only blocked test is Meyer's own refusal (1990-92); the coroner's finding was cerebral aneurysm, toxicology only lidocaine and phenytoin [read: Dispatch 2007 notes]. The absence of taped interviews of the two Belgian investors in the police file is unexplained and is the best-defined missing record. What decides: the NIST ERIP file, the DTI/QMW report (probably the same document as "Laughton Report on SM's patents", 13 Apr 1992, Perkins Annex J1; my inference, as the contrarian notes), the USPTO file wrappers, the Perkins TACO outcome.

B. Laughton. A favourable observer in 1990 (Electronics World), probable author of a 1992 report that could not reconcile the claims with established science, named as the 1996 court expert in the Sunday Times but not shown in the docket transcription. If the same person, his assessment changed, and the reports are unpublished. The decision transcript and the DTI report would settle it. Unresolved.

C. The post-1998 quiet. US 4,936,961, 5,149,407 and 5,293,857 were assigned from the estate to Marilyn Meyer on 22 Mar 2001, fees lapsed 2004-2006, no licensee or buyer found, her death 29 Jul 2003; the 1999 probate valuation of "Water Fuel Cell: sole proprietorship" is in the court file, not online. Explanations: (1) the offers (family account) were not real or serious, or the technology could not be reproduced by the buyers or the estate; (2) the estate lacked the means or knowledge to commercialise (documented: small estate of about $306,608, litigation by creditors); (3) suppression (family accounts only). No replication on record has shown more gas energy than input in 28 years of builders (Lawton: 1-2 A, no gas rates; forum builders report amps, i.e. tens of watts). The simplest reading needs no secrecy, but suppression cannot be excluded without the missing records. This is inference, not a test.

D. Perkins Project TACO (VW engine, then the Thames ferry Brunel; deposit $125,000; demonstration planned for July 1993). The record stops at the meeting of 28 Feb 1993: nothing shows it was signed, run or abandoned. Explanations: Perkins's own reservations ("no redress" clause; UK experts' view of the first and second laws); proof of concept failed; Meyer's secrecy; loss of files. The Perkins file at Peterborough or Livett Launches would settle it.

11. Corrected claim list (final verdicts)

1. Water capacitor splitting. - Pulsed HV across water yields H2 and O2: Established (as electrolysis and discharge chemistry). - Resonant gain to 20-40 kV in conducting water: Refuted by calculation. - Bulk-field, charge-free bond breaking: Refuted by calculation. - Less energy per mole than ordinary electrolysis, or any gas energy above input: Never properly tested. (QE revision: Meyer's own 1983 patent data are those of an ordinary electrolytic cell; no change.) - The 1983 text claim that pulsing "enhances considerably" the splitting: Internally inconsistent with the patent's own Fig. 16 (QE revision, section 14). - Printed equations (Eq 9, 14, 17, "78.54 ohms"): Internally inconsistent.
2. Hydrogen fracturing, 44,000 to 2.5 million barrels per gallon. Refuted by calculation; with the "complies with thermodynamics / Zero Point energy" statements, Internally inconsistent.
3. Water-fuel injection and vehicles. Water as net fuel using Meyer's own numbers: Refuted by calculation. The buggy or tractor itself: Never properly tested. Battery-electrolyser hydrogen supplement: Plausible, untested (no free energy).
4. Electrical Particle Generator. Voltage step-up: Established. Power gain: Not testable as stated. The device: Never properly tested. QE revision: laser "compounding" (10 W striking a thousand times gives a 10-100 kW equivalent): Refuted by calculation; "permanently magnetized gas": Not testable as stated (field bound in section 14).
5. Gap resonance (1982). Not testable as stated; testable version Never properly tested; superseded in Meyer's own later patents (instructive history).
6. Puharich's resonant splitting (prior art, not Meyer's). Figures 91.3% and 114.9%: Internally inconsistent (corrected 63% and 77%). Underlying claim: Never properly tested.
7. Self-sustaining loop / electron extraction (added by QE revision). Electricity from "plucked" electrons fed back through the VIC so the system "goes until there's no more water": Refuted by calculation. "Not perpetual motion because water is the energy source like gasoline" together with "does not create energy": Internally inconsistent.
8. Peripheral Birth-memo claims (added by QE revision). Atmospheric O2 "less than 20% now compared to 30%": Tested, not supported (ice-core measurements by others, section 14). Exhaust-air reclaimer and Steam Resonator: Never properly tested; any output above input would be Refuted by calculation.

12. What would settle it (consolidated)

• One calibrated bench test as in section 3. It has been possible since 1990 and costs little. Its absence is the central fact of this project.
• A calorimeter on the burned gas for claim 2 (the calculation already refutes the yield).
• A fuel-flow and electrical log of one engine run for claim 3.
• Records, in order of value: USPTO file wrappers of US 4,936,961 and 5,149,407; NIST ERIP file (Lewett 28 Jun 1991); the DTI/Queen Mary report and Laughton's 13 Apr 1992 report; the Fayette County decision transcript and Final Judgment Entry; the stanslegacy scans of binder pp.59-60, 66-68 (to confirm "7 lbs/min" and Nagypal's numbers); the probate valuation; the Grove City police and coroner files; the Perkins TACO file; the 1985 ABC footage.

13. Open questions

1. Was Nagypal's 2-4x excess over Faraday a real effect, or a current measured by the wrong meter? A repeat with net charge measured would answer it.
2. What were the pulse lengths and the actual waveform at the cell? Only microsecond transients shorter than water's relaxation time (14 ns tap water, 0.13 ms pure water) could keep the water capacitive. The documents give none.
3. Does "7 lbs/min" in the stanslegacy transcription read that way on the scan, and is it gauge psi? My conclusion that the 500 W test matches ordinary electrolysis depends on this.
4. Was Laughton the 1996 expert, and did his view change between 1990 and 1996?
5. What did the court's three experts examine, and what did they find?
6. Why is there no tape of the Belgians' police interviews?
7. What was the NIST evaluation's method?
8. Were Memos 425, 427-430 and the appendices of the brief concealing a quantitative result? They are unread in depth; a deep read is still outstanding (Opus), as is a transcript of the 1989 New Zealand video and a check of newsletters No. 4 and No. 8.

14. QE revision (2026-10-09)

Trigger: qe.md verdict MINOR (three gaps: the self-sustaining claim, the unread New Zealand transcript, Meyer's own 1983 graphs). Inputs: qe.md; the new deep reads in deep/ (EP 0 111 573 A1; Video 1989 New Zealand house meeting; Birth pp.126-222; photos; newsletters No. 4 and No. 8; US 4,389,981; Puharich ELF book pp.176-204); my own arithmetic [calc] (python, .venv). Web checks used: 4 more (listed at the end of this section). All Meyer material below is Meyer's own characterization (patent holder, dealership seller); the New Zealand talk is a whisper transcript with no speaker labels, so figures can be mis-transcribed (the deep reader found several).

Result. No existing verdict changes. One claim is added (claim 7), two sub-verdicts are added to claim 4, one to claim 1, a peripheral-claims row (8) is added, and several statements in this file are corrected (end of section). The three reviewers have not weighed claims 7 and 8 or the new sub-verdicts; these are my verdicts from the energy accounting and arithmetic below, and a later review update should have the physicist and contrarian confirm them.

14.1 Claim 1, what Meyer's own 1983 patent data show (EP 0 111 573 A1, filed 14 Dec 1982; CA 1,234,774)

• Operating point: "the source 30 supplied 12 volts at 1 ampere" [read: p.12]. Faraday gives 0.685 L/h of mixed gas for 1 A [calc], and 12 W cannot make more than 12 percent of its energy appear as hydrogen heat of combustion (1.48/12) [calc].
• Graphs (all page images): Figs. 8, 9, 11, 12, 13, 15, 16 have no gas scale. Only Fig. 10 (and Fig. 14, the same numbers per kWh) carries gas figures: about 3 ft tube, 84 W, 456 cubic inches/h; 10 ft tube, 120 W, 2700 cubic inches/h. No voltage or current is given for them.
• My check [calc]: 456 in^3 = 7.47 L/h. As gas heat of combustion that is 16 W, 19 percent of 84 W. 2700 in^3 = 44.2 L/h = 96 W of combustion heat, 80 percent of 120 W. That is inside the energy limit (thermoneutral 1.48 V; reversible 1.23 V). It equals Faraday's law at 7.7 V (3 ft) and 1.86 V (10 ft) cell voltage. If the unstated voltage was the 12 V of the text, the 10 ft point would be 6.5 times Faraday (the deep reader's figure; I get the same). The patent cannot say which. So these are figures for a good electrolyser, or for a cell whose voltage was not recorded. They are not an over-unity number.
• Fig. 11: for distilled water, voltage against amperes is a straight line, about 87 V at 1 A (an ohmic cell of about 85-90 ohms). The text says gas is "a linear function to magnitude of voltage applied" (p.12). Together, gas proportional to current: Faraday behaviour. Fig. 9: water temperature rises faster than linearly and the water boils at about 107 V; Meyer calls the heating "a serious limitation". Figs. 8 and 15: current ("amp load") rises 1, 1.2, 8, 10, 15, 20, 23 from distilled to ocean water, ordinary ionic conduction.
• One departure, drawn but not measured: in Figs. 8 and 15 the gas curve is nearly flat while the amp load rises 23-fold, annotated "contaminates aid gas generation without ionic interaction". There is no gas scale, so it is a sketch of a claim. Ordinary effects (chlorine from sea water, vapour in warm gas, recombination) point the same way.
• Self-contradiction in the 1983 patent: the text says pulsing "enhances considerably the splitting" (p.8), but Fig. 16 (gas against pulse on-time) falls in a straight line as on-time decreases, so continuous DC gives the most gas. Internally inconsistent (own figure against own text). Also the 37 percent figure the QE quoted as an electrolysis efficiency is about photosynthesis (p.6; the deep reader's correction; qe.md is wrong on this point).
• CA 1,234,774 (granted 1988) drops the words "non-ionic" and "sub-atomic" from claim 1 and adds "a variable D.C. voltage source ... restricting the current thereof to a negligible value relative to the magnitude of the voltage" and "less than 1 ampere" (claim 5). The "restricted current" idea that drives the later water-capacitor story entered the claims after the 1982 filing; the 1982 text itself uses 12 V at 1 A and 84-120 W. Fig. 8's 23 "amp load" and Fig. 10's 120 W do not fit a "less than 1 ampere" claim. The European search report (9 Sep 1983) cites only category "A" documents (Wolf DE 2 852 712; Inoue Japax JP 54-89978); nothing was cited as destroying novelty. That is the examiner's search result and says nothing about whether the cell works.
• Conclusion for claim 1: the 1983 numbers support the "weak form" (a pulsed or DC cell releases gas) and, read honestly, are the numbers of an ordinary electrolytic cell. No new test of the strong form exists. Verdicts in section 3 stand.

Meyer's own water analysis (Birth Fig. 7-15, p.163). "R & D Laboratory, A Certified Biological Testing Laboratory, Columbus, Ohio 43204, C. Thomas Bell, Ph.D., Lab Director, Test Date: January 10, 1983, Spectrophotometer Analysis, contaminants Na and K": distilled under 2 ppm, tap 11, rain 16.1, lake 20.2, drilled 24, river 45, ocean 26.1K ppm [read: page image p.163, QE and deep read]. The laboratory is credited with the sodium and potassium assay only, not with any energy test. Newsletters 10 and 11a reprint the chart; their deep reads recorded the test date as "1963", the Birth page shows 1983 (rechecked 2026-10-09 against the scans: both newsletters print "January 10, 1983"; No. 10 PDF p.6 and No. 11a PDF p.17, each printed "Page Six"). Using this source-based water instead of my generic tap-water range [calc; molar conductivities Na+Cl- 126.4 and K+Cl- 149.9 S cm^2/mol, memory, high]: 11 ppm Na as NaCl is about 60 microsiemens/cm (0.006 S/m), relaxation time 117 ns, cell Q at 5 kHz 0.004; as KCl 42 microsiemens/cm, Q 0.005; 2 ppm Na (distilled) 11 microsiemens/cm, Q 0.02. Hardness ions were not measured, so real conductivity is higher. The central point stands more strongly: at kHz the cell is a resistor, not a resonant capacitor, even for Meyer's own water. A new contradiction inside Meyer's record: the 1983 patent says contaminants "aid gas generation" (Figs. 8, 15), while Birth Memo 426 (p.155) says "less water contaminants nets even higher energy-yield" and ranks the chart that way. These describe different devices (a DC/pulsed plate cell in 1983, the voltage-driven VIC cell in the 1990s memos); the second reading fits a capacitor picture, the first fits ordinary electrolysis. Neither is a measurement.

The "Patent Validation Report dated January 14, 1983" is cited once (Birth p.190, Memo 429) as "WFC Test-Results, Mode of Operability" for 35 USC 101. It has no author, laboratory, witness, instrument or number in the PDF; it is four days after the Bell analysis. It joins the records list: it may be the "Patent Validation" document in the 1995 binder at stanslegacy (not checked) or in the 1983 Canadian/European file wrappers. Birth pp.126-222 name no other test, laboratory or witness. Other bare citations (Appendix A): "on-road tested" 111 ml/min of gasoline at 65 mph, a "WFC International Independent Test-Evaluation Report", and the Northeastern University "Powerful Water-Plasma Explosion" and Kansas State "Electrically Induced Explosion in Water" reports (their titles and contents are not checked; nobody has read them).

Open question 2 (pulse lengths), answered in part. No pulse width in seconds exists in Birth pp.126-219. "Within a millisecond or less" (p.133) is a voltage rise time to "up to and beyond 90,000 volts" starting from "typically 20,000 input volts", not a pulse length; it is a design statement with no instrument or witness. In the New Zealand talk Meyer gives drive "from zero up to 10 kilohertz ... 20 kilohertz" and "zero to 2,000 volts", with 20,000 to 90,000 V for the gas lattice. At 10-20 kHz the 14 ns (tap) to 0.13 ms (pure) relaxation times of section 2 still leave the cell resistive for anything but the sub-microsecond edges. Question 2 remains open for the actual waveform at the cell.

14.2 Claim 7 (new): the self-sustaining loop, "electron extraction circuit"

What Meyer says, verbatim as transcribed [read: NZ transcript l.636-690; Einsiedeln deep read 5.11]: at voltages high enough to "pluck off the electrons" of the water molecules, a light bulb on the circuit draws them; "100 volts times 100 amps gives me ... 10 kilo[watts] of electrical energy ... you don't even need any mechanical rotating device"; feeding that "back through the voltage intensifier circuit ... so I now have a self-sustained oscillation system ... it goes until ... there's no more water". Einsiedeln: "a self-sustaining oscillation system that once you excited from an external source it maintains itself as long as there is water in the fuel cell"; "I'm now producing electrical energy without consuming the fuel source". At the Pentagon, Meyer says, "General Abramson" called it a perpetual motion device; he answered that it is not, because "I'm using water now as the energy source just like ... gasoline" and perpetual motion "does not define power input versus power output". No input or output measurement, only the 100 V x 100 A illustration, is given; the transcript conflates 6.25 x 10^18 electrons (one coulomb) with "one volt". The general's identity is unconfirmed: James A. Abrahamson headed the Strategic Defense Initiative Organization 1984-1989 and had run NASA's shuttle programme [web: https://www.af.mil/About-Us/Biographies/Display/Article/107876/lieutenant-general-james-a-abrahamson/ ; https://en.wikipedia.org/wiki/James_A._Abrahamson], which fits "Pentagon and NASA", but nothing in the folder ties him to this meeting and the only source is Meyer's report. No independent record of the meeting has been found.

Calculation [calc]: 1. The 100 A illustration. 100 A is 100 C/s = 1.04 mmol of electrons per second. Taking electrons from water means splitting it (2 electrons per H2O), so 0.518 mmol/s = 33.6 g of water per hour. Burning all its hydrogen returns 148 W (285.8 kJ/mol). Meyer's 10 kW is 68 times that. Oxidising water to supply those electrons costs at least 1.23 V x 100 A = 123 W, and the electrons must return to the cathode or charge builds up, so the water is consumed in ordinary electrolysis. The "plucking electrons like a neon tube" picture is conduction in a plasma or electrolyte, which is paid for by the supply. 2. The loop. Splitting a mole of water takes at least the Gibbs energy 237 kJ; burning or recombining the gas returns at most the enthalpy 286 kJ as heat (heat engines recover a fraction), or at most 237 kJ as electricity in an ideal fuel cell. So the best loop is break-even at 100 percent in both directions. With ordinary electrolyser (about 70 percent) and fuel cell (about 55 percent) efficiencies the loop returns about 38 percent (0.70 x 0.55 [calc; memory, medium for the two efficiencies]). A loop that "goes until there's no more water" cannot sustain itself; it stops from losses long before the water is used up, and burning the gas gives the water back anyway. 3. "Not perpetual motion because water is the energy source like gasoline." Water is the ash of burning hydrogen; it can be an energy source only if it is first split, and the splitting cost is the whole energy balance. Meyer himself writes that the process "does not create energy" and "complies with the Laws of Thermodynamics" (section 4) and in Newsletter 4 that "the nucleus remains intact" (1987, 44,000 to 108,000 barrels per gallon), while in 1989 he calls the same process "a nuclear reaction" in which oxygen and hydrogen "both disintegrate ... totally consumed into pure energy" and cites "another time dimensional zone" and an "aperture". A system cannot be at once energy-conserving, chemical, nuclear and fed from a zero-point source.

Verdict, claim 7: the loop (electricity from the water's electrons, fed back to make more gas, net output) Refuted by calculation; "not perpetual motion because water is the fuel" with "complies with thermodynamics" and "Zero Point energy": Internally inconsistent. Not tested: no device, input or output is reported. Evidence for: none beyond Meyer's statements (an interested party). The Ford six-cylinder run on condensed water, said in the room by an audience voice (not Meyer; "started to continue on with less energy input"), is a single unquantified recollection kept in the record. It is the combustion product recycled; condensing and re-injecting water does not make fuel.

What would settle it: close the loop on a bench with the input and output power logged; the pass line is net output above 237 kJ per mole of water cycled, which nothing in physics permits. Any unit that runs "until there's no more water" with a battery-free power meter would be the test.

14.3 Claim 3 (vehicles), additions

• Appendix A, Tab 33-39 [read: Birth pp.217-222, deep read]. Meyer's hydrogen mass fraction of water is 20 percent (one "mass unit" per proton, neutrons ignored, Appendix B note 2); with standard atomic masses it is 11.2 percent [calc; memory, high]. Gasoline is counted the same way (carbon = 6 units, real 12). His energy figures: water 57,000 BTU/gal, gasoline 22,800 BTU/gal, "2.5 times". My check [calc]: the real hydrogen in a gallon of water (8.345 lb x 0.1119 = 0.934 lb) times hydrogen's higher heating value (61,100 BTU/lb; 142 MJ/kg, memory, high) is 57,060 BTU, so the 57,000 figure is correct as the heat of burning all the hydrogen in a gallon of water. The gasoline figure is wrong by a factor of about five: the EIA lists motor gasoline at about 120,214 BTU/gal [web: https://www.eia.gov/energyexplained/units-and-calculators/british-thermal-units.php, via search summary; I did not open the page]. The "2.5 times" ratio then becomes 57,060/120,214 = 0.47: water's hydrogen energy is about half a gallon of gasoline, before the cost of splitting. This extends the earlier note in section 5 (0.36 per kg) and is consistent with it. It also shows Meyer's own table is not self-consistent: his 1.4854 lb of hydrogen would be 90,800 BTU, not 57,000.
• Appendix A also puts 7.4 microliters per injection, 100 cycles/s, 40,000 V at 1 mA = 40 W, "3.3 amp/hr" (a current, the unit slip stands), and "148 microliter water droplet ... = 1000 Bhp I.C. engine power-yield / injection cycle" (Tab 39). The 1000 Bhp is 20 times the 50 hp VW's per-cycle figure, with no basis other than the cited Northeastern and Kansas State explosion reports. My check: 148 microliters is 8.2 mmol, 2.35 kJ of chemical energy at best; 1000 Bhp for 0.01 s (100 cycles/s) is 7.5 kJ at the shaft [calc]; so the figure is 3 times the chemical value at 100 percent efficiency, and about 10 times at engine efficiencies; it needs the energy "gtnt" that the 2.5-million-barrel claim requires.
• Newsletter 8 (Winter/Spring 91/92, reprint of Grove City Record 31 Jul 1991): of five retrofits listed (dune buggy, world-record Corvette, rotary-engine Alaskan Bushmaster plane, Jet Commander jet, a boat in Britain) only the dune buggy "is now installed"; none is reported running on water. Its claim "a patent for the Electrical Polarization Process was issued on June 25, 1991" does not match US 4,936,961 (26 Jun 1990), and "May 21, 1991" for the gas management system matches no number in its own table. "Patents: more than 25 US" against 11 US numbers in its own table. Single source (the company's own newsletter).
• Newsletter 4 (Winter/Spring 87/88): 44,000 to 108,000 barrels "predicted" from "preliminary tests" (body, p.2) and "exceeds 44,000" as fact (pp.4, 5); no derivation. "Nine out of forty-two patents ... issued in the United States". Military statements ("working with the military on a systems analysis") name no agency, contract, officer or date. Finance: a 351-acre Product Development Center, "five percent of all profits ... to financial investors", Meyer's spending "less than $1,000,000" and "$375,000" on hydrogen fracturing. These are the inventor's statements and are part of the financing record that section 9 describes.
• The New Zealand talk: dune buggy "pre-engineering", production "in 18 to 24 months"; an Australian desert-race plan was dropped because the sponsor's contract "boiled out to be nothing more than a deed of property transfer"; "we got the green light to ... demonstrate it like in the Indy 500" (nature of that demonstration not described); no apparatus shown. One audience member says a demonstration "doesn't need to be demonstrated it just needs to work"; Meyer insists a first model is needed "to generate interest". No dealership price is stated; dealership terms (licence, first-come numbering, no resale to a foreign entity) are described. The 1997 placard "Now, under WFC Industrial Contract: Europe" (photos p.16) is Meyer's own marketing text; no contract, party or country appears.

Verdicts for claim 3 are unchanged: net-fuel use of water Refuted by calculation; the buggy or tractor Never properly tested; battery-electrolyser supplement Plausible, untested.

14.4 Claim 4 (EPG), the 1989 figures

The earlier statement "no numbers" (claims.md, section 6 above) is wrong for the talks; there are numbers, but still no measurement. Meyer's New Zealand description [read: NZ deep read 1.6]: a non-magnetic tube wrapped by pickup coils, filled with a "permanently magnetized gas" (ionised argon and iron, or nickel, cobalt, neodymium-iron-boron, made at 2,000-90,000 V with laser energy), "closed loop", "only one moving part"; output = field strength x field velocity x turns x number of coils; a home unit "half the size of a hot water tank, about 18 inches in diameter" with a garbled output ("2,300" amps or watts); and laser amplification: a 10 W laser, a tube looped out "20 miles" and back with 1000 turns per coil, struck "a thousand times", "a 10,000 or 100,000 watt laser equivalent". An audience voice objects "you get it by back emf"; Meyer defers. He declines to say how the gas is made. Source: Meyer's talk; no demonstration shown.

Calculations [calc]: - Laser compounding. Energy cannot be multiplied by repetition. A 10 W laser delivers 10 J per second on average, however it is pulsed; striking it a thousand times delivers 1000 pulses of the energy per pulse, not 1000 times the power. A transient field can be stored and released, but only at the stored energy, less losses, so the average cannot exceed the supply. The "10,000 or 100,000 watt equivalent" is Refuted by calculation. (The transcript's "less than a nanosecond" for light to cross a 20-mile loop is mis-heard or wrong: light takes about 107 microseconds for 20 miles; a nanosecond is 0.3 m. I flag the transcript as garbled there.) - Magnetised gas. Even if every atom in 1 atm of gas at 298 K (2.46 x 10^25 per m^3) were an ion with 1 Bohr magneton, all aligned, the magnetisation is 228 A/m (0.29 mT); with 5 Bohr magnetons (Fe3+) it is 1.14 kA/m (1.4 mT) [calc; Bohr magneton and 5 mu_B for Fe3+, memory, high]. A permanent magnet gives about 1 T, 700 to 3,500 times more. Spontaneous alignment of spins in a gas needs exchange coupling, which is a solid-state property [memory, high]; no gas ferromagnet is known [memory, medium]. So "hundreds of thousands and millions of cubic feet of magnetized gas" cannot replace a magnet, and the claim has no density, field or spin-alignment figure to test: Not testable as stated. - Moving the field. If the gas moves the field past the coil and a load draws current, Lenz's law gives an opposing force; the power taken from the field's motion equals the electrical power out. "The only energy input ... to move the magnetic field not to create it" is correct, and the input is the output. The net gain claim stays Not testable as stated because no input figure is given; any claim of output above that input is Refuted by calculation (energy conservation).

Verdicts for claim 4: voltage step-up by induction Established; net power gain Not testable as stated; the device Never properly tested; laser compounding Refuted by calculation; permanently magnetised gas Not testable as stated.

14.5 Claim 2, additions

The New Zealand talk repeats "2.5 million barrels of oil per gallon of water" (an unidentified voice; Meyer lets it stand), attributes to Einstein a train around the earth on a gallon of water, and describes the process as "a triggering process" in one sentence and "a nuclear reaction" in another. The numerical refutation in section 4 stands (2.5 million barrels is 4.5 percent of the gallon's rest energy). The earlier request for the "energy calculation" is now fully closed: Memo 424, Memos 425-430 and Appendices A-B contain none, only the Tab 33-39 fuel arithmetic of section 14.3. Verdicts unchanged.

14.6 Peripheral Birth-memo claims (new claim 8)

• Atmospheric oxygen: Memo 428 says "less than 20% now as compared to 30% prior to the Industrial Age as so determined by polar ice-cap samples"; the New Zealand talk gives "38 percent or 28 percent" (garbled). No ice-core publication is named; the only cited source is the IMAX-NASA film "Blue Planet". Ice-core oxygen measurements by others show a relative decline of about 0.7 percent over 800,000 years and about 0.1 percent over the last century, from about 20.9 percent today [web: https://blogs.princeton.edu/research/2016/09/23/ice-cores-reveal-a-slow-decline-in-atmospheric-oxygen-over-the-last-800000-years-science (search summary; page not opened); the 20.9 percent is memory, high]. A 30 percent pre-industrial atmosphere is not supported. Tested, not supported (measured by others, not by Meyer; Meyer's figure has no source).
• Exhaust Air Reclaimer (Memo 428): pulsed polarisation to pull apart NO and other oxide molecules, "NO < 1 ppm when running on water" (source of figure not given). No measurement, instrument, EPA test or witness. Never properly tested.
• Steam Resonator (Memo 430): oscillating polar water molecules heats the water; no temperature, flow, power, pulse width or input-versus-output figure. Never properly tested; if read as heat output above electrical input, Refuted by calculation.
• Optical Thermal Lens / "Easer" (Memo 429): stimulated emission from hydrogen atoms in a quartz vacuum chamber under pulse voltage; no wavelength, power or experiment. Not testable as stated.
• Other Meyer-document statements kept in the record: Meyer's quenching tube ("holes 15 to 25 thousandths"), burn-rate adjustment of hydrogen with exhaust gas (flame speed 325 cm/s hydrogen against 41-43 cm/s for fuels, brought to "around 47"; US 4,389,981 Fig. 9 gives 265-325 against 35-45 cm/s, from the NBS Monograph 168 "Selected Properties of Hydrogen" that the patent cites, and shows the adjustment only as a drawn curve) and the crash "jar switch". These are engineering claims about hydrogen handling, not anomalies; hydrogen's flame speed is the textbook one [memory, high].
• US 4,389,981 (1983 injector) contains no test, flow, power or current figure; the hydrogen source is deferred to the co-pending generator. It is apparatus description. Confirmed: no engine data in the 1982-83 injector family.

14.7 Claim 6 and related, additions (Puharich ELF book, pp.176-204)

The "PHASER" chapter adds: a 600 Hz resonance (plus or minus 5 Hz) held by a microprocessor on salinity (I/E = 0.01870, 0.1540 M, i.e. 0.9 percent saline) and gap (2.30-2.46 V rms); nuclear-spin timing restated as 0.25 s; a reversible electrolysis threshold of 0.16 mW; and the origin of Component I in the electrostimulation of hearing. None of it carries gas volumes, laboratory names or witnesses, and it does not mention Meyer; the only link is Meyer's US 4,798,661 citing US 4,394,230 (absence of a mutual link in one file is not evidence of independence). The harmonic list's 63,690 Hz is an OCR reading; 3980 x 16 = 63,680. Verdict unchanged: Never properly tested (unit slip as in section 8).

14.8 Photographs (39 images, Gries Petty Research / RWGresearch.com / AlexPetty.com marks)

The photographs are third-party (watermarked) pictures of hardware, with provenance and chain of custody unknown (not in the folder). They show: a toroid and diode-ladder "voltage intensifier" cylinders, a ten-cavity resonant cell, an 11-slot control rack (Steam Resonator, Resonant Cavity 10 to 2, Gas Processor), control boxes copyrighted 1981 ("Electrical Polarization Process", rotary pulse generator) and 1982 ("Hydrogen Fracturing Process"), a GMS rack, a "Laser Accelerator Assembly", the "WATER POWERED" dune buggy, and the 1997 "FlameLess Central Water Heating System ... WFC Heat Resonator" (project 427-DA; a sticky note dated June 13, 2009). A Voltage Intensifier housing label lists US 5,149,407; 4,826,581; 5,293,857; 4,936,961; 4,798,661; 4,389,981; 4,613,304 ("One or more U.S. Patents apply"). No image shows operation: no gas flow, temperature or waveform. In the instrument images the DC volts needle rests near zero and the amps charge-rate needle at about 10 on a 0-100 scale; this is a photographed state, not a measurement. Value for the record: surviving hardware exists (and was photographed), which could be examined and, if run on a calorimeter with logged power, would test claim 1; this belongs in the records list (section 12). The buggy was also transferred in probate on 5 May 2000 (sites.md).

14.9 Retention, restored to the record (the QE found these dropped between sites.md and this file)

All are leads or hypotheses from sources outside the folder, listed with their stake; none is verified by me. 1. Lammertink's scrap-steel alloy hypothesis (tuks.nl) for why Meyer's first cell "worked then stopped", linked to Meyer's 1982 steel-analysis affidavits (Robert L. Queen). It predicts an electrode-material effect and would be tested by running the cell with different electrodes. Never properly tested; it is an explanation, not a datum, and it is consistent with ordinary electrode chemistry (surface films, overpotential). Lammertink is also the secondary source for the ABC footage (about 9 minutes surviving of an alleged 3 hours). 2. Pollack's exclusion-zone (EZ) interpretation (also on tuks.nl): charge-separated layers of water next to hydrophilic surfaces, hundreds of micrometres thick [memory, medium], could matter in gaps of 0.25-1.6 mm. It is a mainstream-contested proposal by others, not Meyer's, with no energy-balance prediction for this cell. Not testable as stated for Meyer's cell. 3. Griffin's Warsash lecture (Sep 1993, Royal Institution of Naval Architects; as reported in Meyer's binder, unseen): "10 W in, gas at 10 psi within 10 s, vessel cold". Check [calc]: 10 W for 10 s is 100 J; at 1.48 V that yields at most 0.00053 mol of gas, 12.9 mL at 1 atm; raising a vessel by 10 psi (gauge, 69 kPa) needs a vessel no bigger than about 19 mL, and 100 J heats 100 mL of water by only 0.24 K, which a hand cannot feel. So "cold" is expected, and the claim is neutral until the vessel volume is known. The figures are a single reported observation by an interested admiral; keep it and ask for the volume. 4. Dublin Institute of Technology explanation (1995 binder), named in the Investigator's leads; not read. Rea O'Neill (Dublin) and Hans Nieper (14 Dec 1996, "the tested overunity effect") are named leads; both unseen. 5. Press and broadcast record (all unseen by any role): Channel 4, "It Runs on Water" (17 Dec 1995); Ship & Boat International / RINA (Jun 1991); Asian Defence Journal (1992); IECEC 1993 (Bailey and Grotz); Fortean Times (1997); Air & Space Smithsonian (Oct/Nov 1991, with the NIST verdict "clear"); Spy (Dec 1991). Only Spy and Air & Space are weighed in the reviews (as second-hand NIST statements). 6. Other recorded talks: the 1992 Global Sciences Congress "Voltage Disassociation of Water", a 1993 interview, lecture videos on the Internet Archive (81 items, none downloaded). 7. Earlier investor suits in Franklin County: 79CV004853 (Schuler/Leukart v. Meyer, 1979-81) and 93CV005477 (Schneider, Brooks and Gilvesy v. Meyer, settled 1994; Brooks invested over $300,000), also 99CV006575 and 99CV006793 (estate period), alongside Fayette County CVH19930292 (1996). A fifteen-year pattern of investor disputes is pertinent to the stake and finance record. The Newsletter 8 retrofit list names Rick Schneider (jet owner) and Bill Brooks (Corvette owner), so two of the 1993 plaintiffs also appear as project partners in 1991; this is a coincidence of names with the docket captions, unverified as the same people. 8. Related work the physicist and contrarian discussed but the primer omits (primer is not mine to edit): Kucherov US 5,632,870, Graneau et al. (Phys. Lett. A 1986), Gary L. Johnson (Kansas State), Shimizu pulsed-electrolysis papers. Meyer cites the Northeastern and Kansas State explosion reports in Appendix A. 9. Surviving hardware and footage as settling records: the "83 DuneBuggy" (probate transfer 5 May 2000), the photographed cells and control racks, the 1985 ABC and Columbus television footage.

14.10 Absence-of-evidence analysis, additions (section 10)

• 10A (no energy-per-gas measurement): the New Zealand talk adds that Meyer asked the audience for "small amounts of money" to set up manufacturing and said a demonstration unit was needed "to generate interest"; the stated reason for the missing demonstration was "pre-engineering", "latter stages of debugging", and Meyer's refusal to disclose two things (how to "restrict amps in a dead short condition" and the magnetised gas) because "when the power structure had come against me ... they couldn't answer that question". That is secrecy as a stated reason (explanation 2), from the inventor, and consistent with his patent strategy statement (he filed "a series" of patents rather than a master patent). It does not explain the lack of a logged bench measurement.
• 10C (post-1998 quiet): the 1997 placard "Now, under WFC Industrial Contract: Europe" is a claimed commercial status one year before Meyer's death (Meyer is the only source; no counter-party, date or country). It bears on the Belgian-investor story; the investor-contract records (company registers, the Grove City police file) would test it. It adds an item to explanation (1) and (3) of 10C: either the contract was real and lapsed, or it was a concept.

14.11 Web checks (budget 8 for the arbiter; 5 used in the first pass, 4 more in this revision, so one over budget; the extra was the QE's field-strength check)

1. Bond-scale field, the check the QE flagged: ab initio molecular-dynamics studies of liquid water report a dissociation threshold of about 0.35 V/A (3.5 x 10^9 V/m), with 15-20 percent of molecules dissociated at about 1 V/A (10^10 V/m); earlier experiments cited there put the onset at 0.32-0.44 V/A [web: https://arxiv.org/pdf/2506.11734 ; https://iramis.cea.fr/en/llb/evenement/ab-initio-molecular-dynamics-study-of-dissociation-of-water-under-an-electric-field/ ; via search summary, pages not opened]. The tag on section 3 point 2 changes from [memory, medium] to [web]: the bond-scale field is about 3.5 x 10^9 to 10^10 V/m. Applied bulk fields of 2 x 10^5 to 1.3 x 10^7 V/m are therefore 270 to 17,500 times too small, which is inside my earlier "100 to 10^5" range; the verdict stands.
2. Gasoline energy: EIA 120,214 BTU per gallon, section 14.3 [web: https://www.eia.gov/energyexplained/units-and-calculators/british-thermal-units.php, via search summary].
3. Abrahamson identity: section 14.2 [web: af.mil biography and Wikipedia]. Only identifies a plausible person; no link to Meyer.
4. Atmospheric oxygen history: section 14.6 [web: Princeton summary of Stolper et al., Science 2016; search summary]. Still unchecked: the Northeastern/Kansas State explosion reports; the NBS Monograph 168 contents; Pollack's EZ thickness (memory, medium); the NIST ERIP report (check failed again in section 3).

14.12 Corrections to earlier documents (from this revision, for the orchestrator)

1. This file's source-gap note and the other files that say the New Zealand video was "not transcribed" or that Step 0 left no manifest.json (claims.md, facts.json, physicist/mathematician/contrarian, primer): both statements are now false (qe.md, correction 3).
2. Section 5 and Correction 4 of this file, and mathematician.md: the "740 ml/sec" slip is withdrawn; page 221 prints ".740 ml/sec" (page image; the OCR dropped the decimal point). "3.3 amp/hr" stands. Edited in place above.
3. Correction 9 of this file and contrarian.md Corrections item 6: the 1989 licence date is on p.1 of the viXra copy, not the folder copy (edited in place above). sites.md Leads should name the viXra edition.
4. qe.md (its "Missed material" B): the "37%" is photosynthesis in EP 0 111 573 A1, not electrolysis; "Figs. 8 and 15" are the water-type graphs and the plate-spacing graph is Fig. 13; there is no Fig. 19 (an OCR reading of "Figure 10" on p.13). Those pages are in the 1983 deep read.
5. claims.md claim 1 ("No gas rate, current, input power or energy balance is given anywhere"): wrong for the 1983 patent (Fig. 10/14 gas against watts; 12 V at 1 A), the 1995 binder (energy table pp.59-60; no arbiter section covers it beyond section 2 and the claim 1 discussion in section 3, where it is cited via physicist.md and contrarian.md) [corrected 2026-10-09 after QE: was "section 11", which is the corrected claim list and does not cover those figures] and the 1987 newsletters; add the sentence that no figure reports voltage and current with gas, so no energy balance can be made.
6. claims.md claim 4 "no numbers": there are numbers in the New Zealand talk (section 14.4); the correct statement is "no measurement".
7. facts.json EP 0 111 573 A1: EP claims differ from CA 1,234,774 (EP: "non-ionic ... sub-atomic"; CA: "current restricted to a negligible value", "0 to above 110 V, less than 1 A"); add the Fig. 10/14 numbers, the 12 V/1 A point, representative A. Wasmeier, the search report of 9 Sep 1983 (examiner I. Van Bellingen) citing DE 2 852 712 and JP 54-89978. Newsletters 4 and 8 have no text layer and were read from page images; the "extracted text" statement in facts.json for them is wrong.
8. Newsletters 10 and 11a deep reads: Fig. 7-15's test date reads 1983 on Birth p.163 and on the newsletter scans (No. 10 PDF p.6, No. 11a PDF p.17; checked 2026-10-09); the deep reads' "1963" was an OCR error and is corrected there.
9. Photos OCR corrections: "WEC" is WFC, "43128" is 43123, "1961" is 1981, and the patent list on the Voltage Intensifier label has seven numbers (OCR dropped three).
10. Appendix A Tab 34: the gasoline figure of 22,800 BTU/gal is about five times too low (EIA about 120,000); the water figure of 57,000 BTU/gal is right for the hydrogen in a gallon of water; the "2.5 times" is therefore an error in the gasoline figure (earlier notes accepted the 2.5 only as a per-kilogram hydrogen figure).

Corrections to earlier documents (for the orchestrator)

1. claims.md claim 1, "with current minimised at resonance": in a series LC circuit, current is maximal at resonance (mathematician). Replace by "with current deliberately restricted (Meyer: 'without consuming amps')". The Electronics World description (diode charge pump, "the water breaks down and a momentary high current flows") shows a repetitive-breakdown regime (contrarian).
2. claims.md claim 1 "No gas rate, current, input power or energy balance is given anywhere" is wrong for the wider record: Meyer's 1995 binder pp.59-60 gives 12.5 V at 40 A, 46,666 cc/h per tube and a "Differential Efficiency Factor" of 1,696 (physicist, contrarian). Also claims.md claim 3 "No engine test data" should add that Appendix A gives engine arithmetic (not measurement).
3. Memo 424 is no longer "unread": it contains no energy calculation (all three). The requested deep read now only needs Memos 425, 427-430 and appendices; the claims.md table row should be updated.
4. Appendix A slips: WITHDRAWN in part. The page prints ".740 ml/sec" [read: page image p.221, QE and deep read of Birth pp.126-222]; the leading decimal point was lost in OCR, so there is no slip there. "3.3 amp/hr" is 3.3 A and stands (mathematician). Planck's constant in Birth p.123 is 6.547e-27, not 6.626e-27.
5. Puharich: 114.9% becomes about 77% and 91.3% about 63% when gas per mole is 33,600 cc; the contrarian's "internally consistent" note is withdrawn (this document).
6. Electronics World: "dielectric constant of about 5" for pure water is wrong at these frequencies (about 78-80, relaxation near 19-20 GHz); and Meyer's 78.54 "at 20 C" (Birth p.52) should be about 80.1 [memory, high]. Griffin's "roughly 3 times the energy of the hydrogen released" for electrolysis is wrong for practical cells.
7. "Your theory works" (Nagypal, newsletter): quote only with his numbers (20 mA, 0.5-1 cc/min = 2-4x the Faraday value for 20 mA), "can't reproduce" Meyer's waveforms, and his distributor offer.
8. The 1996 case: Fayette County CVH19930292, Willis and Harper v. Meyer, $40,515.72 garnished (not $25,000; that figure is from the Sunday Times retelling). The Franklin County 93CV005477 case is a different, 1994 settlement. Death date: Dispatch 20 Mar 1998, probate 21 Mar 1998; both stand.
9. Birth p.18 of the folder copy prints foreign-licence #490606 as "Nov. 15, 1988"; the "1989" is on p.1 of the viXra 233-page copy (research/files/vixra-0811.0001), not on the folder copy, whose p.1 is a cover (QE page-image check). Two Meyer editions disagree; neither is an OCR error. Also claims.md corrections 1-8 (8XA board is Tony Woodside's, EP0103656 date 24 Sep 1982, newsletter No. 8 patent date 26 Jun 1990, patent counts, mislabels of US 4,613,304 / 4,613,779 / 4,394,230, duplicate newsletter pages, Stephen Foster Meyer patent not attributable to Stanley, 9 Apr vs 4 Apr 1982) stand and are not contradicted by the reviewers.
10. Investigator lead list: the "DTI-sponsored QMW report" and "Laughton Report on SM's patents" are probably one document (not confirmed).

Web checks done (budget 8; 5 used by me, plus those logged by the reviewers)

1. Proton mobility 36.24e-8 m^2/(V s), 298.1 K: found in the search summary of Mauerhofer et al. 2003 PCCP, https://download.uni-mainz.de/fb09-ak-roesch/Roesch_Puplikationen/113 Mauerhofer-2003-PCCP.pdf (page not opened). Retagged [web]; previously [memory, high/medium].
2. Water enthalpy of formation, liquid: -285.8 kJ/mol appears in secondary mirror pages; the NIST WebBook page itself did not show the liquid value in my search (the physicist confirmed the vapour value -241.8). Retained as [memory, high] with a [web] corroboration from secondary pages only (https://physicsforums.com/threads/standard-enthalpy-of-water.855095).
3. NIST ERIP "net consumer of energy": check failed again (WebSearch, no NIST/NBS document on Meyer found). Stays second-hand (Spy Dec 1991; Heard 2000).
4. Field-driven water dissociation (second Wien effect): about 10^7 V/m for classic weak electrolytes and above 10^8 V/m for water itself [web: https://arxiv.org/pdf/2405.05882 ; https://arxiv.org/pdf/2208.10828]. This supports the physicist's "greater than 10^7" claim; the bond-scale figure (10^9-10^10 V/m) remains [memory, medium], not checked on the web; the verdict does not turn on it, since the applied field is far below even the Wien-effect range.
5. Tap water conductivity: 50-800 uS/cm is the commonly cited range (up to about 1,500 for some supplies) [web: https://atlas-scientific.com/blog/water-conductivity-range/ via search summary]. My Q-table uses 0.01-0.05 S/m (100-500 uS/cm), inside that range; the result is a Q below 0.0022, so the conclusion is not sensitive to this. Not rechecked: barrel of oil 6.1 GJ [memory, high]; Planck's constant; the reviewers' own checks (USPTO MPEP 2107; the 19-20 GHz water relaxation arxiv 1704.01667; the Puteanus abstract) are cited from them.