Hopper-WJ (William J. Hooper): Science Contrarian review (Step 2)

Reviewer role: Science Contrarian. Judges by experimental evidence; builds the strongest honest case; examines how claims were handled; applies the same rigour to the inventor. Source tags: [read] from the documents (including Investigator copies in research/files/), [web] with URL, [memory] with confidence. Calculations are mine (SymPy/NumPy venv).

Documents read in full by me: US 3,610,971 text, the Hooper typescript deep read (deep/hooper-9b2d23.md), Hooper's "Similarities of the Motional Electric & Gravitational Forces" (research/files/hooper-similarities.pdf), Gibson 1983 (research/files/rexresearch-gibson.htm), and NASA TM-106963 (research/files/nasa-TM-106963-fulltext.txt). Not read in depth: the 137-page New Horizons PDF (the WWII shielding experiments, Chapter 3), US 3,656,013, Experimental Notes (not online). Conclusions about the shielding claim therefore rest on Hooper's summaries of those experiments, not on the experiments themselves.

Terms for the reader

• B = magnetic flux density (tesla). V = electron drift velocity (m/s), the slow average speed of conduction electrons along a wire, v = I/(n e A), with I current, n conduction-electron density, e = 1.602e-19 C, A wire cross-section.
• "Motional electric field" E = B x V: Hooper's claim that the magnetic loops around a current move along the wire at V, producing an electric field.
• P.D. = potential difference read on a Keithley 640 vibrating-capacitor electrometer (an instrument that reads tiny DC voltages by modulating a capacitor).
• Seebeck (thermoelectric) effect: a junction of two different metals at different temperatures produces a voltage, typically microvolts per kelvin.

Corrections to Step 1

1. claims.md says "the project has no measured data" / "neither document contains numbers". True of the two folder documents, but the Investigator found the quantitative papers, and they change the review. Claims 2 and 3 now have data (Similarities paper; Gibson 1983). I used these. [read]
2. Claim 3's temperature statement needs splitting. The patent says drift velocity "approaches 10^8 cm/s" in a superconductor, 10^10 times the room-temperature value (3.6e-? cm/s printed garbled; the 1972 correction). The 10^8 cm/s is the random (thermal/Fermi) electron speed that the patent itself quotes, not a drift speed [read: patent col. 3]. My check: copper at 480 A/cm2 gives v = j/(ne) = 3.5e-2 cm/s (matches the corrected 3.6e-2) [calculation]. Step 1 correctly flags the misprint, but the claim is a conflation of two different velocities, not just a typo.
3. Claim 6 (rotor). Step 1 is right that Hooper reports the sound-absorber result against himself. I add: this is the best control in the whole record, and it is Hooper's. Nothing in claims.md should read as support for a rotor-gravity link.
4. Claim 5 / patent claims. Verified: the five patent claims cover hardware only [read: patent claims]. The NASA test used the flat pancake coil of the weight claim (patent Fig. 4/6), not the 4020-wire cylinder with which Hooper's capacitor data were taken [read: TM-106963]. Step 1 and sites.md describe NASA as "the only independent controlled test"; accurate, but it tested the weight claim, not the capacitor observable.
5. Minor: Hooper's Similarities paper has internal slips Step 1 did not catch (density "9.96" for 8.96 g/cm3; "1.78e-2 m/s = 1.78 cm/s" compared with a classical "1.762e-2 cm/s"; Hooper's own eq. 3 gives 1.81e-2 m/s, not 1.78). These do not change anything but show the paper was not carefully proofed. The 1970 APS Bulletin page is given as p.209 in the essay and p.207 in the Similarities paper [read].
6. The folder name "Hopper" is a misspelling; the inventor is Hooper [read].

Evidence inventory (what actually exists)

Claim-by-claim

Claim 1. Magnetic flux around a current moves along the conductor with the drift velocity, giving E = B x V.

a. Mathematics. Internally consistent as a hypothesis: if loops of B move at V, the Lorentz-type field E = -V x B follows. But it is not derived from anything; it is the premise. The step "flux moves" has no definition in standard electrodynamics for a steady current: the field is static, so there is nothing to move, and by Faraday's law (curl E = -dB/dt) a time-independent B yields no induced E. Hooper's own analogy (a toroid has induction with B = 0 outside) is correct but concerns time-varying flux. b. Established experiment. Steady currents in neutral wires produce no E field outside beyond the surface-charge fields due to the resistive voltage drop. [memory, high] Against Hooper's own number: with the classical copper drift speed his formula predicts a capacitor P.D. of 4.9e-8 V for his 4020-wire geometry (n=4020, I=10 A, r1=0.1 m, r2=0.1035 m, v=1.76e-4 m/s), against 5e-6 V measured: a factor 103 too small [calculation; inputs from Similarities paper]. Hooper acknowledged the factor ~100 and rescued it by saying "a decrease of 100-fold in n, as indicated by Fermi-Dirac statistics" [read]. That is wrong: free-electron/Fermi-Dirac theory gives n ~ 8.5e28 m^-3 for copper (one per atom), the same number Hooper used, and Hall-effect measurements agree [memory, high]. So the quantitative test of his own formula fails unless n is 100 times smaller, which contradicts Hall data. Cullwick's 1957 book is a real source for moving flux [read: Millis ref 9 confirms E. G. Cullwick, Electromagnetism and Relativity, p.245]; Cravens 1990 (excerpt via rexresearch) says there was "no critical experiment which can disprove whether a magnetic flux rotates with its source" [web] https://www.rexresearch.com/hooper/hooper1.htm. That statement concerns the old unipolar-induction debate (does B rotate with a magnet?), a genuinely historic dispute; but it is a different question from a moving-flux field around a conductor at 0.2 mm/s. c. Tested? By Hooper only (capacitor, rotor, FIG. 11 yoke experiment in US 3,656,013 per Investigator). Not replicated. NASA did not replicate the capacitor measurement. d. If true. Faraday-Maxwell electrodynamics would need an extra, non-Lorentz-covariant field around steady currents; relativistic transformations of fields would need an additional non-electrostatic electric field not sourced by charges or changing B (Hooper explicitly says this, citing Jeans, Winch, Bridgman). Verdict: Never properly tested (strong form: flux moves and generates a spatial E field); the quantitative version on Hooper's own numbers is Tested, not supported (his measurement is 100x his own prediction). The unipolar-induction sub-question (does flux rotate with its source) is Not testable as stated in the usual formulation: standard theory gives identical observable predictions either way [memory, medium].

Claim 2. A non-inductive winding of >4000 conductors has no net B, yet the B x V fields add, charge an external capacitor, and are immune to shielding.

a. Mathematics. Internally consistent as a counting argument (B and V both change sign in opposed conductors, so the product keeps sign). But one thing is mixed: the formula P.D. = (mu0 n I V / 2 pi) ln(r2/r1) uses n I as a single lumped current (B1 of 2010 conductors, doubled) as if every wire's B acted at full strength at the capacitor; the real wires have B contributions that cancel at distance, and the product B x V is taken per wire. That is Hooper's hypothesis, not an error in algebra. Predicted scaling P.D. proportional to I^2. b. Established experiment, and the data the Investigator found. My strongest honest case for Hooper: - He identified and tested several artefacts (Similarities paper, "Crucial Experimental Evidence") [read]: rapid up/down current stepping (30 -> 10 -> 30 -> 10 A, readings "substantially identical"); a second, larger outer capacitor with the inner pair grounded; a gaussmeter (max ~0.5 G) and a five-wire pickup loop on a microvoltmeter (no emf); balanced +/- supply to stop lead wires carrying charge into the housing (a special supply was built); an unnamed "disinterested expert" confirming with an oscilloscope. - Readings were steady for as long as current was held, not a swing and return (Gibson) [read]. - Gibson's report is unusually candid and records problems; these records are exactly what an honest experimenter's notes look like. The strongest case against, from the same documents: - The signal is small and the nulls are not. 5 uV at 10 A. The artefacts Hooper and Gibson themselves recorded are as large: an unbalanced supply changed readings by 11 uV of a 28 uV deflection at 26 A; polarity depended on plate thickness (thick versus thin plate reversed the charge); readings "varied from day to day"; 8-10 degree room-temperature swings changed readings, "the higher the temperature, the higher the readings"; and readings rose as the generator warmed [read: Gibson]. Millis and Williamson also found the capacitor readings "too erratic" [read]. - Added series resistance outside the generator raised P.D. at the same current. Gibson: a carbon-pile rheostat (3.7 to 5.3, 7.5, 12.3 ohm) and a 1500 W photoflood bank each raised P.D., "the form of that resistance... didn't make any difference"; at 13 A with 12.3 ohm P.D. was substantially above the 15 A reading with no added resistance [read]. Under Hooper's theory P.D. depends on B and V in the winding, set by the current alone, so an external resistor cannot change it. What changes with an external resistor at fixed current is the supply voltage and the potentials of the leads and housing (and, for the photoflood, heat and light in the room). That points at electrostatic/charging coupling or temperature, the very artefact class Hooper said he had excluded. Gibson and Hooper instead read it as a change in conduction-electron number n. I treat this as the single most important internal evidence against the motional-field reading. The leads were balanced about ground in the later special supply, but the pre-balance data (all the tables Gibson uses) were not redone [read: Gibson, "we did not redo them"]. - Even-in-I does not distinguish. P.D. positive for both current directions fits B x V, but it also fits any I^2 effect, notably Joule heating (370 W at 10 A, 3.3 kW at 30 A, using R = 3.7 ohm [calculation; read: R and 3300 W]). The "I^2 parabola" is therefore not a signature of the theory. Further, Gibson's dry-ice data fit P.D. = K I^4, not I^2 (resistance 3.7 to 2.3 ohm) [read]; the theory has no I^4 term. And AC gave a straight line "beginning at less than zero" [read]; the theory (B x V with both proportional to I) predicts I^2 for AC as well, and the vibrating-capacitor electrometer is a DC instrument, so what it indicates for 60 Hz current is unclear. - Thermoelectric magnitude. 5 uV corresponds to under about 1 K difference if the junction thermopower is a few uV/K (metal pairs typically a few uV/K [web] https://en.wikipedia.org/wiki/Seebeck_coefficient; the specific copper-to-stainless value I did not verify). Hooper's rapid-stepping control argues against a slow thermal lag; I take that seriously, because the interior heat from 370 W reaches the surface on a time scale of tens of seconds to minutes by my estimate [calculation, rough], and readings that track current within seconds are better explained by something fast. That leaves electrostatic pickup (potentials of ~37 V across the winding at 10 A) as the unexcluded fast explanation, supported by the resistance and plate-thickness observations above. - No dummy. No control winding in the same geometry in which the B x V sum would not occur (same wire, same current, but the same-direction arrangement; or the same geometry with nichrome in place of copper, same n but 70 times the heat) was reported. B x V predicts the same P.D. for same-n metals; the artefacts predict a large change. This is cheap and decisive. - Unit/number issue. The Similarities paper asserts "good agreement with accepted values" while its own numbers differ by 100 (above). - Shielding. The claim that the field passes through grounded 1/8 in stainless is observed only as the capacitor charging; but the capacitor was inside the shield, and Hooper's own "inner capacitor cylinder... forms a completely grounded electrostatic shield" (Similarities) means a non-shielded signal path (the connecting wires, the epoxy, the power leads) is not excluded. The WWII shielding experiments are described nowhere in the folder documents; the claim "immune to every kind of shielding, magnetic or electrostatic" [read: essay p.2] has no data in any document I read. Hooper's reading of the WWII device is also odd: the device failed because of cancellation, yet later "no diminution by shielding" is said to have been established. How a failed instrument demonstrated this is not stated [read: deep read]. c. Tested by whom? Hooper and Gibson (an interested party, her husband financed the work); one unnamed outside expert confirmed the instrument reading, not the interpretation. No independent replication of the capacitor measurement exists. NASA attempted the surface-voltage measurement with a pancake coil and got erratic data and did not pursue it; its authors wrote "Hooper may have misinterpreted thermal effects" [read], an agency's characterization from a propulsion screen that selected only the weight claim. After Hooper's death in Feb 1971 [web, single source: Burridge snippet via Scribd, per Investigator] and Warren Gibson's funding ended, work stopped; Gibson says she burned out the tube of a vacuum-tube generator and stopped, and in 1983 "All the original equipment is still available" [read]. Fits death of the sole investigator and the end of private funding; no evidence of suppression found, none looked for in archives yet (USPTO file wrappers, Ohio filings remain unchecked). d. If true. A non-electrostatic, non-magnetic, unshieldable electric field would exist around ordinary resistive coils, implying a gap in Maxwell's equations; shielding is a consequence of Gauss's law and conductor boundary conditions, so unshieldable E would violate them. Verdict: Never properly tested (no neutral replication; no dummy winding; the key artefact classes not excluded). Weighted by the record: internal evidence (series-resistance dependence, plate-thickness polarity reversal, day-to-day drift, warm-up enhancement) favours artefact, but this is internal evidence, not a controlled test. Not dismissed without testing: Hooper did run controls, and NASA made an attempt that did not reach the capacitor observable. The shielding half: Not testable as stated from these documents (experiments undescribed); should be re-assessed after the New Horizons chapter is read.

Claim 3. Capacitor voltage measures electron drift velocity and conduction-electron density; drift velocity is higher at low temperature, by up to ~10^10 in superconductors.

a. Mathematics. Eq. 3 is internally consistent (checked: V = 2 pi P.D./(mu0 n I ln(r2/r1)) = 0.0181 m/s for 5 uV). The patent's superconductor claim conflates drift velocity with random Fermi velocity (10^8 cm/s); a drift of 10^6 m/s would need current density j = n e v = 1.4e16 A/m^2 for copper, against practical critical current densities of order 1e10-1e11 A/m^2 [calculation; critical current figure memory, medium]. So the 10^10 enhancement is Refuted by calculation as drift velocity. At fixed current in a given wire drift velocity is I/(neA), independent of temperature (NASA says the same); Hooper's "greater at low temperature at large current density" reflects higher j at the same resistive voltage, not a velocity change at fixed I. b. Experiment. The measured "V" is 100x classical (above). Gibson's resulting deductions (electron number falling with current, ten times as many electrons at dry-ice temperature, number depending on external resistance) contradict Hall-effect and the Wiedemann-Franz-consistent electron density of copper [memory, high]. Gibson says these findings "appear more consistent with... gases, semiconductors, semi-metals"; Hooper's own earlier comparison with Fermi-Dirac gets the logic backwards. c. Tested only by Hooper/Gibson; no replication. d. If true, the free-electron picture of metals (Sommerfeld/Drude) would have to change; Hall coefficients and Fermi surface measurements already fix n. Verdict: Refuted by calculation (superconducting 10^10 drift enhancement; Fermi-Dirac "100-fold" rescue). The instrument-for-drift-velocity claim: Tested, not supported on Hooper's own data (100x discrepancy) because the signal is not shown to be B x V at all (see claim 2).

Claim 4. The motional field polarizes neutral atoms and attracts matter like gravity; gravity is an electric polarization effect.

a. Mathematics. The force on an induced dipole is F = (1/2) alpha grad(E^2), correct textbook electrostatics (Jeans p.125 as cited). Internally consistent, but it produces a force proportional to the atom's polarizability alpha, not its mass. b. Experiment. Gravity's acceleration is independent of composition to ~2e-13 (Eöt-Wash, eta(Be-Ti) = (0.3 +- 1.8)e-13) [web] https://ui.adsabs.harvard.edu/abs/2008PhRvL.100d1101S/abstract. A polarizability force would vary strongly with element (my estimate: alpha/mass differs by a factor ~13 between hydrogen and helium [calculation from atomic polarizabilities, memory, medium]). To get g on a carbon atom (m = 2.0e-26 kg, alpha = 1.96e-40 C m^2/V) requires grad(E^2) ~ 2e15 V^2/m^3, i.e. fields of order 4.5e7 V/m over a metre [calculation]; the real field near Earth's surface is ~100 V/m [memory, high]. Brauner's plant-polarization work (Endeavour, 1969, cited) concerns biology, not gravity's mechanism [read: deep read]; Burridge's article is in a popular magazine. Hooper himself says the attraction experiment was "planned" and never reports it [read]. c. Never tested by Hooper. No test I know of. d. If true, the equivalence principle (Eötvös) would fail by many orders of magnitude, and general relativity would be replaced. Verdict: Refuted by calculation (equivalence-principle / energy-scale argument). The vague kinship claim "unshieldable like gravity" is Not testable as stated.

Claim 5. A strong B x V field could cancel/reverse gravitational polarization (weight loss, antigravity).

a. Follows from claim 4; no numbers offered. Hooper's patent says "Energy from the atoms of the earth would supply the lift". b. No weight measurement by Hooper is reported in any document I read. c. Tested: NASA Lewis (Millis and Williamson 1995): four flat counter-wound coils, a 555 g acrylic disk on a balance beam, detection 0.02 g (about 4e-5 of the disk's weight), accelerometer threshold 40 ug, coils also in ice: null at all currents [read]. The test was on/off toggling, deliberately avoiding warm-up, whereas Hooper took data after warm-up [read]; the authors admit Hooper's weight claim in the patent rested on a pancake coil and the cylinder data are different, and that "a minuscule effect below their thresholds cannot be excluded." Magnetic leakage of the coils (not fully self-cancelling) affected their accelerometer [read]. No superconducting coil was tried (cost, ~$10,000 [read]), though the patent calls for cryogenic or superconducting operation as the way to reach large effect. Agency characterization: NASA selected the claim as a propulsion screen. Hartman's "witnessed" antigravity forces: one secondary review, no witness named, no source. d. A repulsive gravity coupling would be a new force. Verdict: Tested, not supported (weight change in room-temperature and iced coils, to 4e-5 of the disk weight); the superconducting form was Never properly tested; the underlying mechanism is refuted by claim 4. Not dismissed without testing, but the test reached only the weak, room-temperature form of the claim.

Claim 6. The 1957-59 Fisher/Hooper rotor gravity-meter deflections show a link between the field and gravity.

a. n/a (no mathematics). b. A heavy sound absorber removed the deflections; a high-pitched sound or a buzzer produced them [read: essay p.3]. Hooper accepted "our joint project collapsed". Controls (stopped rotor, reversed rotation, non-magnetic dummy) are not described. Historian Kragh independently confirms Fisher (1891-1966) and the APS presentations [web, per Investigator] https://dokumen.pub/ (Kragh, Varying Gravity, 2016, pp. 68-69). c. Tested by the experimenters themselves; the test result was negative and honestly reported. Dismissal of this claim was not a mainstream act; the inventor ended it. Hooper then retained the belief on the basis of the (undocumented) shielding experiments. I flag as the part of his reasoning that does not follow: the absorber removing the signal is evidence that the effect was not unshieldable. d. Nothing new. Verdict: Tested, not supported (by the inventor's own control). Its role is historical: it shows how the belief survived its own refutation.

Claim 7. The field penetrates barriers (communication, power, detonation) and Earth's electrical energy can be tapped.

a. n/a. b. No transmission through a barrier is reported; the patent proposes it. Gibson mentions Moray's device as a tapped gravitational field (Hooper's belief) [read]; no measurement. c. Never tested. d. Free-energy source from gravity: energy conservation; a static gravitational field does no work on a fixed system. Verdict: Not testable as stated (the "energy everywhere on the earth" part is vague); the barrier-communication part is a Never properly tested prediction that depends entirely on claim 2.

How the claims were treated by critics

• No mainstream reply to the 1969-70 APS abstracts is on record; I found no rebuttal paper. The record is silent, as it usually is for work at a small college and a private company. Not dismissed by experiment, but ignored.
• The one independent experiment (NASA 1995) was not a replication of the main observable; its dismissive "thermal" remark is an inference from reading Hooper's papers, not from a measurement.
• Posthumous patent grants (died before 5 Oct 1971 grant; Certificate of Correction 16 May 1972) mean the drift-velocity error was fixed by others.
• The funded company went dark after 1971. Competing explanations: death of the sole investigator and end of private funding (evidence: Gibson's preface; Sunbiz dissolution 1991); secrecy/classification (no evidence found); commercial failure (consistent). USPTO file wrappers and Ohio filings are the records that would decide.

Investigator findings kept in the record

• US 3,656,013 FIG. 11 yoke/magnet unipolar-induction experiment: galvanometer follows the iron yoke's rotation, read by Hooper as flux rotating with the magnetic system (Investigator, not read by me in depth).
• Hooper's reported controls (quick stepping, outer capacitor, gaussmeter, five-wire loop, balanced supply) and the outside expert; listed above.
• Gibson's data: dry ice 28 Feb 1969, PD = K I^4; resistance dependence; plate-thickness reversal.
• French family member FR 2 039 172 A1 (filed 1969, withdrawn; reason unknown).
• Later patents citing Hooper (Banduric, PST Associates, WWK Trust, Lockheed Martin homopolar); these are citations, not tests.
• Klicker's theory extension; Gilstrap 2017 (not opened); Moray link.
• Cravens 1990 excerpt on unipolar induction.
• Fisher biographical record (Kragh); Hooper biography (NSF fellowships; Berkeley PhD thesis on critical ionization potentials).
• Hartman: one secondary source reports witnessed gravitational/antigravitational forces, no names.

Summary of what would settle things

1. A dummy winding: same geometry, same current, copper versus nichrome (same n, 70x the heat), and an inductively-wound version; B x V predicts the same P.D. for same-n metals.
2. Hold the current fixed and vary the series resistance, power-supply symmetry and lead potentials; B x V predicts no change.
3. Read the New Horizons WWII chapter and the 16 Dec 1969 shielding entry in Experimental Notes.
4. Measure the capacitor output at a known temperature uniformity (pre-equalised) with a calibrated thermocouple on both plates.

Own-rigour note

I cannot give Hooper the high-confidence "debunked" label for claim 2's shielding half, because the experiments are not available to me. For the capacitor signal, the weight of the in-record evidence is against the interpretation (resistance dependence, polarity dependence on plate thickness, 100x numerical discrepancy), and I have said so without treating that as a controlled refutation.