StClairJohn: Arbiter (Step 3)

Project: John Quincy St. Clair, "Hyperspace Research Institute", San Juan, Puerto Rico. Corpus: five US patent applications filed 2002-2004 (published 2003-2006).

Tags: [read] from the project documents or the Step 1 deep reads; [calc] my own arithmetic in the project venv (mpmath) this session; [web] checked on the web (URL); [memory] recalled, with confidence.

Source gap note

There is no text gap: all five applications were readable, and the figure-sheet equations were read from page images by the deep readers. There is an evidence gap, and every later step must repeat it:

• Every conclusion rests on the inventor's own filings. No other person is on record as testing, building, endorsing or criticising any device [read; sites.md].
• The only experimental material in the five documents is one sentence and one unscaled photograph in US 2004/0164824 [read].
• The inventor's fuller experimental narratives are in sister applications that are NOT in the folder (US 2004/0200925, 2004/0102810, 2007/0285325). They were reached only through the Investigator's notes from FreePatentsOnline [web: https://www.freepatentsonline.com/y2004/0200925.html and sister pages]. No reviewer re-fetched them, so what they say about the experiments is second-hand through the Investigator.
• Unread: S. D. Horsburgh's thesis (the one third-party observation the inventor leans on), the copyright deposit copies of the tetrahedron diagram (1996-2008), and all USPTO file wrappers. Where a point turns on these, I say so.
• No [memory] fact below carries a verdict alone, except the standard physics noted as high-confidence.

Where the three reviewers stand

They agree on the main outline. Differences are in labels, not substance. I recomputed every number on which they differ.

Smaller points checked: 1/9 - 1 = -8/9 = cos^2(asin(1/3)) [calc]; this is built in by the 1/3 radius ratio. 8 pi G/c^4 x eps0 E^2/2 at E = 3e6 V/m is 8.3e-42 per m^2 [calc]. eps0 E^2 = 79.7 N/m^2 and /4pi = 6.34 [calc]. m_p/m_e = 1836.15 [calc]. c^2/G = 1.34659e27 kg/m for G = 6.6743e-11; the inventor's Omega = 1.346812891e27 corresponds to G = 6.67320e-11 [calc], a slightly older or rounded value, a 1.6e-4 difference. The Einstein-tensor results (G_tt for FIG. 9; G_zz for FIG. 15) were recomputed independently by the mathematician and the contrarian (G_tt) and the mathematician (G_zz). I did not re-run them myself, but three reviewers and the deep reader agree, so I treat them as reliable.

Web checks done (budget used up: the search tool quota was already exhausted, so only fetches worked): - Hau et al. slow light: Wikipedia "Slow light" reports about 17 m/s in 1999 work [web: https://en.wikipedia.org/wiki/Slow_light]. The fetch summary named the medium as strontium atoms; I recall a sodium Bose-Einstein condensate [memory, medium]. Not load-bearing; it only shows that slow light in a dispersive medium is real and has nothing to do with an extra dimension. - Kohnstamm phenomenon: after a prolonged voluntary contraction (pressing arms against a doorframe for about 30 s), the arms rise involuntarily [web: https://en.wikipedia.org/wiki/Kohnstamm%27s_phenomenon]. This is a competing explanation for the arm-stays-up report in US 2004/0200925, not a finding. The inventor's report is a spinning ring, not an isometric push, so the match is loose and I do not know whether it applies. - Planck constant: the exact SI value is 6.62607015e-34 [web: https://en.wikipedia.org/wiki/Planck_constant]. The page does not list older CODATA values, so the claim that 6.626195e-34 matches an old recommended value stays [memory, low]. Not load-bearing. - Not verifiable this session (search quota): the Horsburgh thesis content; the Witten attribution; EM drive and lifter test results. The wormhole exotic-matter requirement and 26-dimension bosonic string facts were checked by the mathematician on Wikipedia.

Corrected claim list (final verdicts)

Scale: Established; Plausible, untested; Never properly tested; Tested, not supported; Refuted by calculation; Internally inconsistent; Not testable as stated; Historically superseded but instructive.

Claim 1. Opposed coils: negative "spacetime spring constant", negative mass, wormhole to "hyperspace" (US 2003/0197093)

Agreement: all three agree. The arithmetic (-8/9) is right. The minus sign on the inner coil is assigned by hand. The unit check to N/m works only with curvature K = 1/r^2 instead of the standard 1/r. The metric of FIG. 9 contains none of the coil quantities and puts tesla into a dimensionless slot. G_tt of that metric is never positive, so the "negative mass" is a property of the chosen ansatz. Negative K with positive M gives exponential runaway, not "therefore M is negative".

Disagreement: none of substance. The mathematician adds two things the others lack, which I accept after reading them. First, I x Area x B = I x flux = L I^2 = twice the stored magnetic energy, which is positive for either winding direction. So the assigned minus sign cannot come from reversing the winding. Second, since g_t,theta depends only on theta, a change of time coordinate t' = t - B F(theta) makes the metric static. The geometry is a chosen curved surface, and the coil frequency and phase play no role. The contrarian's small correction stands: G_tt is exactly zero at theta = pi/4 for r > 0; the spike sits near that angle at B^2(1 - sin 2theta) = r^2, depth -1/(4r^2).

Stronger argument: all three, and the mathematician's are strongest because they are algebraic.

Final verdict (split): - The derivation (spring constant -> negative mass -> wormhole): Internally inconsistent. - The physical effect (a coil-made wormhole or negative mass): Never properly tested. It was not dismissed after a test. No current, field, turns or mass is ever given, so there is nothing a test could check. Mainstream rejection rests on theory (a real coil's field energy density B^2/(2 mu0) is positive [memory, high]; wormhole throats need exotic matter, Morris-Thorne, per Wikipedia [web: https://en.wikipedia.org/wiki/Wormhole], a requirement on a wormhole, not a recipe).

Settles it: derive the g_t,theta term from the coil's stress-energy tensor with the 8 pi G/c^4 coupling and real numbers (current, turns, radius). The expected result is a metric perturbation of order 1e-40 or smaller. Experimentally: any measurement of inertial mass or gravity at a coil whose current is known, with a powered/unpowered control and blind operation.

Claim 2. Braided gold coaxial cable (8.34 and 7.56 micron) generates "hyperspace energy"; photograph shows "white mist" (US 2004/0164824)

Agreement: the only datum is one sentence plus FIG. 1, a small dark halftone with bright cloud-like patches, no visible braid, no scale, no instrument or voltage [read: deep read]. Diffuse bright halos are a known charging artefact in scanning electron microscopy [memory, high], offered by two reviewers as a competing explanation, not a finding about this image. The radii are reproduced only to 1.5e-4 (about 1 nm) and only by solving three of the four constraints with the third radius c left free (c = 0.3531 micron, not the printed 0.3596) [calc by the mathematician]. Constraint 2 gives 9.885 not 10; constraint 3 balances only if lengths are in metres. The logarithm of a dimensional quantity depends on the unit. Constraint 4 matches to 1e-4 by a coincidence chosen from a large menu (look-elsewhere effect).

Physics: the physical premise "higher permittivity, lower speed of light" is correct inside a dielectric (c = 1/sqrt(mu eps)). It does not mean an influx from a co-dimension. A medium with n = 33,418 (c/8971 [calc]) in a micron-scale vacuum coax is not a property of any known medium.

Disagreement: the physicist calls the mechanism "unphysical as stated"; the contrarian emphasises that the photograph is non-diagnostic, neither supporting nor excluding anything. Both are consistent. Final weighing: the physical claim has no test on record; calling it "plausible" would be wrong, and calling it "tested, not supported" would also be wrong.

Final verdict (split): - The physical claim (the cable produces hyperspace energy): Never properly tested (one uncontrolled, non-diagnostic photograph; the inventor's caption is the only identification of the object). - The geometric derivation of the radii: Internally inconsistent (not reproducible as printed; unit-dependent; ten printed digits unsupported by the inputs).

Settles it: a scale-calibrated micrograph of the actual cable, with drive conditions and instrument settings, plus a powered/unpowered control; and a measurement of the line's propagation velocity and impedance, which would show whether any "low c" exists beyond the ordinary dielectric value.

Claim 3. The tetrahedron diagram: derives the constants, shows 10 non-redundant dimensions, shows electron and proton are one particle (US 2004/0164824, 0197093)

Agreement: m x lambda_Compton = h/c is a definition true for every particle, so the "base constant" cannot distinguish electron from proton. ln(h/c) = -95.915482 [calc]; the patent's value is 1.9e-5 off. The nested radical (printed target 80.1104395) comes out 80.3027 (0.24% high) from the printed recipe; the "sum = 1" identity gives 1.0036; the constant 0.599547652 has no derivation. In a nested root, the i-th constant enters with weight 2^-i, so a 6-7 digit match cannot test later constants and one free constant could be tuned [calc, mathematician]. The target equals -ln(Planck length) = 80.11045 with the inventor's own Omega, i.e. it is a measured quantity re-expressed [calc]. Edge counts (24 = 2 x (4+4+4), 10 edges) are correct arithmetic; identifying them with spatial dimensions is asserted. The Witten attribution matches no result I can find; string theory uses 10 (superstring) or 26 (bosonic; 24 transverse) [web: https://en.wikipedia.org/wiki/Bosonic_string_theory].

Disagreement: on the electron = proton part. The mathematician says Not testable as stated, because "the same particle entering from hyperspace at two places" predicts nothing different from the standard picture. The physicist and contrarian say Tested, not supported, because the proton's measured structure (deep inelastic scattering [memory, high]) and the separate conservation of lepton and baryon number contradict identity. I side with a split. The plain-language claim (one particle, not two) contradicts existing data: proton/electron mass 1836.15 [calc], a proton that scatters electrons like a composite object while the electron does not. The mathematician is right that a rescue version ("two different aspects of one hyperspace object") is not testable. I record both.

Final verdict (split): - m x lambda = h/c and the Planck-box rewriting: Established, but trivially (definitions). - Numerical agreement with CODATA constants: Established (they are the inputs). - "Derives the constants geometrically": Internally inconsistent where it can be checked (radical and sum do not reproduce; unit dependence), and elsewhere Not testable as stated; no unused constant is predicted. - "10 non-redundant dimensions": Not testable as stated. - "Electron = proton", strong form: Tested, not supported (by decades of scattering and quantum-number data, though not by experiments aimed at the inventor's idea). Weak, reinterpreted form: Not testable as stated.

Settles it: the copyright deposits (VAu000372532, 1996) would show the intended derivation; then ask the diagram to output one quantity not used as an input (for example the muon or neutron mass) before comparing, and redo it in cgs units: if the relations survive a unit change, they are geometry; if not, they are numerology.

Claim 4. Electromagnetic stress-energy gives net lift or propulsion with no reaction mass (US 2003/0209637, 2003/0230675, 2006/0145019)

Agreement: three devices, three independent failures, all confirmed by at least two reviewers. - 0209637: the force formula (FIG. 9) is plain electrostatics times area; curvature and Omega never enter it. SI epsilon_0 and Gaussian 4 pi are mixed (a spurious 1/4 pi, a pure number that a unit check cannot catch). The "time lever arm" makes the force depend on where t = 0 is set. "Imaginary newtons" have no meaning. T^zr contracted with a radial normal gives a vertical force, not the claimed horizontal one. The inventor concedes a continuous radial field gives zero net force. The segmented pairs are internal forces. Real GR coupling for E = 3e6 V/m is 8.3e-42 per m^2 [calc], which the inventor himself calls too small to notice (para 0022). - 0230675: the "imaginary angular momentum" is a complex position coordinate subtracted for two counter-rotating charges; the real parts cancel. With a real value G_zz is positive everywhere and no negative spike exists. The rotor energy -L K J^2 A^2 is postulated "by taking a hint", and drops the 1/2 of (1/2) L I^2. With the patent's own definition of turns (one turn per groove height, n = 1/h) the sign of dL/dr reverses; and a radial gradient is a radial force on a part that does not move radially. - 0145019: the printed Faraday tensor omits the wave's magnetic field. With it included, the divergence of the stress-energy tensor is exactly zero and the static-field-times-wave cross term averages to zero. What remains is radiation recoil P/c = 3.3 micro-N per kW [calc]. The force "formula" the Step 1 reader was told to look for is not in the patent (and E . E_W = 0 in the patent's own geometry). - Credit due, stated by the contrarian: the wedge electrostatics in 0145019 (charge density going as rho^(-2/5) for a 300 degree opening) is correct textbook work. The mathematician adds that the printed line potential is a factor 2 too large.

Disagreement: only on the label. The contrarian preferred a split (Refuted by calculation for the static-plus-wave and closed electrostatic cases; Internally inconsistent for 0230675; devices Never properly tested). The physicist used Refuted by calculation (strong form), plus Internally inconsistent. The mathematician pointed out the honest scope: what is refuted is the derivations as written, not every conceivable field-based propulsion scheme, and no device was ever built or measured. I adopt the contrarian's split because it is the most exact. The momentum-conservation argument for a closed system is the strongest single argument, and independent of any detail of the patents.

Final verdict (split): - 0145019 (static field plus radio wave gives net force) and 0209637 (closed-system electrostatic thrust): Refuted by calculation. Only radiation recoil P/c survives, and that is Established but not what is claimed. - 0230675 (imaginary angular momentum metric and rotor lift): Internally inconsistent. - Built devices: Never properly tested; none were built or measured on record, so "tested and failed" would be wrong. - Not covered by these verdicts: the related experimental families (asymmetric capacitor "lifters", EM-drive tests), which no reviewer read for this project; I do not assert their outcome here (reviewers recalled ion-wind explanations and failed controlled EM-drive tests at [memory] medium, not checked).

Settles it: a thrust-balance measurement in vacuum of any one of the three designs, with the power supply on board or shielded, and a momentum-conservation accounting of the field momentum; the calculation alternative is the full Maxwell stress integral over a closed surface for the 0145019 geometry with the wave's B field included (already done: zero).

Claim 5. Enormous amplification by "hyperspace permittivity" (about a trillion) or by an inductance gradient on grooved rotors (US 2003/0209637, 2003/0230675)

Agreement: the factor from the inventor's own speeds is (c/8971)^2 = 1.117e9 [calc], not 1e12. The "enclosed reference calculation" is not in the publication. 0164824 separately gives "as low as 1 m/s", a different slow-light value. The tetrahedron-diagram source for 8971 m/s could not be tied to any printed edge. Whether more permittivity gives more force depends on what is held fixed: at fixed voltage stress goes as eps E^2 and rises; at fixed charge it falls as 1/eps [standard; both reviewers]. The patent states neither condition. A uniform increase of eps would also slow the fields that produce the force. The force formula contains no curvature, so amplifying curvature cannot amplify it. Size check: 6.3 N/m^2 x 1e9 would be about 7e9 N from a bench device with no stated power source [calc]. If the real number were true, atomic spectra and the fine-structure constant (alpha = e^2/(4 pi eps0 hbar c)) would change in the region [memory, high].

Disagreement: physicist chose Refuted by calculation for the stated numbers, mathematician and contrarian Internally inconsistent. I choose Internally inconsistent for the number (an arithmetic mismatch in the document's own inputs is a self-contradiction, not an outside law refuting it) and keep Refuted by calculation only for the claim that a curvature amplifier boosts a force formula that has no curvature in it. The rotor part is the same as Claim 4's 0230675 verdict.

Final verdict (split): - The stated amplification factor and the rotor gradient chain: Internally inconsistent (10^12 vs 1.1e9; two incompatible slow-light speeds; sign reversal under n = 1/h; unspecified fixed V or Q). - Existence of "hyperspace permittivity" as a local, magnet-induced change in vacuum properties: Never properly tested; no measurement of vacuum permittivity or c change was ever attempted, and if present it conflicts with precision tests of c and alpha [memory, high].

Settles it: measure c (or the vacuum capacitance) inside the magnet assembly with a standard cavity or time-of-flight test; a factor-of-1e9 change is not subtle.

Other stated claims in US 2003/0197093 (added in the QE revision; see that section for the evidence)

Verdicts here come from the new deep read of 0197093 pp.8-12 and my own checks; no reviewer gave a verdict on these. - Hot-surface cooling by negative mass (para 0001, claim 8 refrigeration): Never properly tested. It is testable as stated, but no temperature, power or sample is ever given. The only calculated quantity in the patent is a G_tt "mass" term for a chosen metric; it is not linked to any thermal quantity. - Carbon-electrode "spacetime curvature battery" (claim 9, para 0029): mechanism Not testable as stated (curvature, in 1/m^2, is said to "appear as voltages" with no conversion); device Never properly tested (no voltage, current or electrode data). Ordinary induction in a changing field is the obvious competing explanation for any voltage seen. - Monopole lift via G_zz (claim 13): Never properly tested. No monopole field is measured, and G_zz is named but never calculated for this geometry. A magnetic monopole has never been observed [memory, high]. - Surgery-less tables, cold-welded crystals, folding-space waveguide and teleportation, inertia-less craft at "hundreds of thousands of g" (claims 6, 10-12): Never properly tested. Each is a one-sentence consequence of the wormhole claim with no sample, distance or measured mass deficit. They inherit Claim 1's derivation verdict (Internally inconsistent). - The 19.47 degree "volcanoes and vortices" statement (para 0028, claim 16): split. The arithmetic (sin 19.4712 deg = 1/3, cos^2 = 8/9) is Established, and is a consequence of the 1/3 radius ratio, so it is not independent evidence. The planetary statement in its stated form ("all the large volcanoes and vortices on Earth, Mars, Jupiter, Uranus and Neptune") is Tested, not supported by published locations: see the QE revision for the quick check. The patent cites no survey and names no feature. Whether a null-model survey would show any excess near that latitude was never done by anyone: Never properly tested in the weak form. - Vector potential as "more important" than B (para 0024; 0230675 paras 0037-0038): the Aharonov-Bohm effect is Established [memory, high]; the inference that A reaches a rotor and, with its surface charge, makes physical angular momentum or lift is Internally inconsistent as written (see QE revision).

Investigator's material kept in the record (not claims, not verdicts)

These stay whatever the verdicts. Each is the inventor's report or a commentator's characterisation, tagged to its source.

• Sister-application experiments, inventor's own reports [web via FreePatentsOnline, second-hand through the Investigator]: a person spun on a palletizing ring 99 times a day for months, right arm said to stay raised without exertion (US 2004/0200925); a flat bifilar bucking coil at ten volts, energy "felt as it rises through the feet" and nine-second body rocking (US 2004/0102810); a sonoluminescence rig built from a February 1995 Scientific American kit; pendulum swings beside an 8.5 inch clay dome with a slot antenna: alternating three-quarter inch swings every 20 kHz across 20-120 kHz, and six inches at 2.78 MHz (US 2007/0285325). Conditions as reported: hand-held pendulum and an oscilloscope, no blinding, no second observer, no stated controls, no subject count. These are the only experimental narrative on record. They are uncontrolled, not negative tests, and a cheap shielded blind repeat is possible (see Open questions). Reported fairly: they are observations by a person who built the apparatus and believed the effect; they were never examined by anyone else.
• The 2.78 MHz result: the pendulum response is observed (as reported), but the "prediction" it is compared to is the reciprocal of the cable length scale relabelled hertz (see table). That weakens it as a confirmed prediction; it does not say the pendulum did not move. At 2.78 MHz the free-space wavelength is 107.8 m [calc], far larger than the dome, so ordinary antenna resonance is not an obvious explanation either. Electrostatic and induced forces on a pendulum bob near a kilovolt-level driven antenna are a competing explanation at any frequency [physicist, memory].
• Horsburgh thesis and the 2.83 Hz match: the inventor reports a third-party observation of a sonoluminescent bubble "beaconing" at about 3 Hz and matches it with a modified Thirring frame-dragging formula including a fine-structure factor. A real bubble oscillation observed by someone else is evidence of an interesting hydrodynamic effect; assigning it to a rotating wormhole is the inventor's reading. The standard frame-dragging term is of order 1e-26 Hz for a kilogram-scale object (see table), so the match needs a large undocumented factor. Thesis unread.
• Copyright registrations (tetrahedron diagram VAu000372532, 1996; later items to TXu001618682, 2008) [web: publicrecords.copyright.gov, per research/2026-10-04-copyright-office.md]: establishes that the diagram predates the patents by six years and was maintained for twelve. It establishes priority and persistence, not correctness. Deposit copies unread.
• Related work: Jefimenko's gravitoelectromagnetism (the "co-gravitational K field" uses his name and symbol) [web: https://en.wikipedia.org/wiki/Oleg_D._Jefimenko]; gravitoelectromagnetism is the weak-field limit of GR [memory, high] and frame dragging has been measured (Gravity Probe B, LAGEOS) [memory, high]. Examiner-cited prior art for 0197093 (Minovitch, Willis, Brotz, Retter) and the Pais Navy applications are the same field of claimed EM-based propulsion; none compared in detail here. These stay as the nearest legitimate and fringe neighbours.
• Commentary (IPWatchdog 2013, unmakeme 2021, Reddit and UFO-forum threads linking the triangular craft to "TR-3B" lore): commentators' characterisations, not tests.
• Lightning anecdotes in 0197093 (objects merged into wood or metal): reports with no place, date or witness. Not testable as stated; kept as the inventor's motivating observation.

Absence-of-evidence analysis: the record goes dark after 2008

What is expected: an inventor with 16+ US filings, a named institute and a copyright series from 1996 would normally leave one or more of: later patents, a company registration, talks, a website, a continuation, a reply to an office action, an obituary. What is found: last filing 2006-06-07; last copyright registration 2008-03-17; applications abandoned (0197093 reported abandoned 2003-11-06 for non-response to an office action, one-source summary from Google Patents, not an office-action read) [web: https://patents.google.com/patent/US20030197093A1/en]; no foreign family; no paper in Google Scholar; no assignee; no independent test.

Competing explanations, with evidence for and against:

1. One-person private effort that ended (funding, interest or money). For: self-filed, no assignee, "Not Applicable" for federal sponsorship, abandonment for non-response, an unsourced forum remark that the "institute" was his bedroom. Against: twelve years of copyright maintenance show sustained activity; an unsourced remark is not evidence of failure. This is the best fit but is a default, not a demonstration.
2. Continued privately or under another name or place. For: an unverified LinkedIn profile and EverybodyWiki entry placing him in Madison, Wisconsin; Minnesota clay named in a 2006 filing. Against: both weak and unsourced; nothing dated after 2008 from him. Would be decided by USPTO correspondence-address changes and Wisconsin and Puerto Rico business registries (the Wisconsin DFI search returned nothing per research/2026-10-04-wisconsin-dfi.md).
3. Death or incapacity. For: nothing about the inventor himself. Against: no obituary found; the Reddit snippet about a death is about his father, not him. Silence is not evidence either way.
4. Absorption into another firm, classification or suppression. For: nothing in the record. Against: nothing that excludes it either; but the work is public patent text that never had a funded owner, and no government interest, assignee change or secrecy order appears. Such mechanisms usually leave a trace (assignment records, secrecy-order notices, foreign filings), none found. A thin record is what a single self-filing applicant normally leaves, so I do not infer suppression from it.
5. Failure of the claims once checked privately (an experiment that did not work, reasons not published). For: the abandonment pattern is consistent with it. Against: no evidence; and the claims' own errors (shown above) are enough to explain an examiner's difficulties without any experiment. This one is conjecture and is stated as such.

Physics-relevant conclusion: the absence of a record is an absence of testing, not a negative test. Nothing in the record separates explanations 1-3; 4 and 5 are unsupported but not excluded.

Records that would decide between them: USPTO file wrappers for application nos. 10/127,098; 10/142,582; 10/170,847; 10/372,556; 11/017,093 (office actions, replies, correspondence-address changes, abandonment notices, any assignments); Puerto Rico corporate registry for Hyperspace Research Institute; Copyright Office name search after 2008; obituary and probate records; the sister-application file wrappers (0200925, 0102810, 0285325 and family).

What would settle it (collected)

1. Claim 1: a derivation, with units and the 8 pi G/c^4 coupling, of the metric term from a stated coil; failing that, a mass or gravity measurement at a documented coil with blind control.
2. Claim 2: calibrated micrograph and driven/undriven propagation-velocity measurement of the cable.
3. Claim 3: the deposit copies; a cgs-units rerun; one prediction of a quantity not used as an input.
4. Claim 4: vacuum thrust-balance test of one design with field-momentum accounting; or a field-solver check of the sign of dL/dr with n = 1/h for the grooved rotor.
5. Claim 5: a direct measurement of c or vacuum capacitance inside the magnet assembly.
6. Sister-application effects: a shielded, blind pendulum test across the 20 kHz - 3 MHz sweep (air currents and electrostatic force excluded, frequencies set by another person); a blinded repeat of the arm-raise report with a matched control group; each is cheap, and none has been done by anyone but the inventor.

Open questions

• What the Horsburgh thesis actually reports about the roughly 3 Hz bubble oscillation, and which m, R and omega the inventor put into the frame-dragging formula.
• What the tetrahedron-diagram deposit copies (1996, 1998, 2008) contain, and whether the unit dependence is addressed there.
• Why each application was abandoned (file wrappers), and whether any examiner raised the sign, unit or momentum problems found here.
• Whether the sister-application observations (pendulum swings, arm raising, foot sensations) have any ordinary explanation (electrostatic force, air motion, Kohnstamm-type muscle responses, suggestion); this is a design question for a controlled test, not a settled matter.
• Whether the inventor is alive, active elsewhere, or continued under another name.
• Whether the FIG. 1 photograph is an electron micrograph at all; the image should be viewed directly by the reader.
• Whether later work by other people (a 2016 mention by Diaz and a 2024 mention by Egely were noted by the Investigator) engages the physics or only cites the patents.

Corrections to earlier steps

• Mathematician: the frame-dragging gap is about 1e26 for the stated example, not 1e22 (see table).
• Physicist: the SEM charging remark and Kohnstamm remark are competing explanations only; both now web-checked in part (Kohnstamm [web], SEM charging not).
• Contrarian: confirmed that the 2.78 MHz figure is the reciprocal length scale relabelled hertz; I verified it algebraically (exp(R) x exp(-R) = 1).
• Step 1: facts.json/claims.md should note that experimental narratives exist in the sister applications outside the folder, and that the Omega value (1.346812891e27 kg/m) is supplied by 0285325.
• The facts.json statement that G_tt "spikes at theta = pi/4, 5 pi/4" should read: spikes lie next to those angles, where G_tt itself is zero at r > 0.

Summary table

No verdict in rows 1-5 changed in the QE revision; the added row is new coverage.

QE revision

Date: 2026-10-04. Basis: qe.md (verdict MINOR) and three new deep reads, all by Sonnet: deep/US20030197093-09173a.md (added sections on the winding, the drive circuit, missing quantities, and every application claim), deep/US20030209637-f6bf67.md section 9 (cross-references, "enclosed reference calculation", 8971 m/s), deep/US20030230675-ed3d2e.md section 6 (cross-references and the vector-potential argument). Tags as above; [web] URLs are given where used. I did not read the sister applications or file wrappers first-hand, so those parts of the record are unchanged.

Result: no verdict in the Corrected claim list changed. What changed is coverage: the inventor's applied claims and the 19.47 degree planetary claim are now in the record with verdicts, and several facts were sharpened.

1. What the new deep reads add (inventor's own statements, [read])

• Application claims, with no number, sample, device or test behind any of them [read: 0197093 pp.8-12]: negative mass "projected onto a hot surface" cools it (para 0001; the QE note cited para 0003, which is the tire-rim and tree anecdote); refrigeration (claim 8); a power supply from "differential spacetime curvature on the ends of a carbon electrode" (claim 9, para 0029); a wormhole to a "low pressure region with a magnetic monopole" and G_zz lift (claim 13); surgery-less medical tables (claim 10); cold-welded crystals (claim 6); folding-space waveguide and teleportation over light-years (claim 12); craft accelerating at "hundreds of thousands of g's" (claim 11). Claim 1 itself calls general relativity "a tested and proven theory"; that describes the theory, not the device. Federal sponsorship: "Not Applicable".
• Missing build quantities [read: 0197093 pp.9-11]: no absolute coil size, turns, wire, inductance, winding-spacing function, source voltage or current, capacitor or transformer values, or operating frequency. Only "thousands of volts", "thousandth-inch thick SuperMalloy tape" and "under 20 MHz" (an upper bound, with no stated reason) are given. The drive is a saturable-choke "magnetic switch" that dumps energy into a capacitor and sends a voltage spike to the coils. The circuit is a conventional pulse arrangement; the physical content claimed for it is the inventor's.
• Reason for the non-linear winding, quoted in full [read: 0197093 p.9, para 0024]: "to preserve the vector potential of the coil", because the vector potential "can extend out past the windings of a long solenoid". The chain contains the sentence "The inductance of the coil times the current density is equal to the vector potential." No equation is given. By units, L times current density (H A/m^2) is not the unit of A (T m); L times current per metre (H A/m = T m) is. The text uses "current density" and "current per meter" in consecutive sentences, so the relation is stated loosely, not derived [my unit check, consistent with the deep reader's note]. What "preserve" means and how the spacing was chosen are not stated.
• Vector potential in 0230675 [read: paras 0037-0038, 0048; pp.10-12]: the text does not argue that A is the primary field; it assumes it. It cites Feynman's solenoid formula (textbook, for an ideal infinite solenoid) and one sentence on "Bohm and Aharanov". The Aharonov-Bohm effect is a quantum phase shift of an interfering charge; it does not by itself exert a force on a classical charged rotor [memory, high]. The final lift expression (claim 9) does not contain A at all; A enters only through the angular momentum fed into the metric.
• Magnetron origin of the energy term [read: 0230675 paras 0007, 0048]: the inventor read the Bell System Technical Journal articles on the magnetron, treated the cavity as a one-turn solenoid (n = 1/height), and "taking a hint" wrote an energy proportional to L K J^2 A^2. This is how he describes the idea's origin; it supports the earlier finding that the energy term is postulated, not derived.
• Cross-references [read: 0209637 p.1, 6-8; 0230675 p.10]:
• 0209637 para 0023 cites "another patent application of mine entitled Magnetic Vortex Generator" (a rotating cylinder with stacked bar magnets that "produces a negative mass and negative spacetime spring constant"), and claim 2 cites "Magnetic Vortex Wormhole Generator and Magnetic Vortex Generator". No filing number or date is given. The first title is not in the folder; the second is US 2003/0197093. Whether "Magnetic Vortex Generator" is a separate application is unknown to me.
• 0230675 para 0008 cites "my previous patent application Dual Rotor Propulsion System" (no number); para 0042 says "in most of my spacecraft designs" angular momentum becomes imaginary, naming no other filing.
• Neither title appears in the Investigator's family list (qe.md R2; sites.md). I did not search for them.
• The 8971 m/s hyperspace speed of light has no source in 0209637 other than "my tetrahedron physics diagram" [read: p.7, para 0023]. No figure of the patent shows a tetrahedron. The "enclosed reference calculation" for the trillion factor is not enclosed anywhere in pp.1-8: the word "enclosed" appears once. This sharpens Claim 5: the factor is stated without the calculation that would let anyone check it. My arithmetic stands: (299792458/8971)^2 = 1.12e9 [calc], not 1e12. The stated proportionality (permittivity as 1/c^2) gives the billion; any extra factor of about 900 is undocumented.

2. The 19.47 degree planetary claim: quick check

Claim as printed [read: 0197093 p.9 para 0028; claim 16, pp.11-12]: the tetrahedron's corners touch the circumscribing sphere at 19.47122063 degrees latitude, "where all the large volcanoes and vortices occur on Earth, Mars, Jupiter, Uranus and Neptune". No feature, survey or citation is given.

Arithmetic [calc]: sin(19.4712 deg) = 1/3, cos^2 = 8/9. This is the same 1/3 radius ratio the coil already builds in, so 8/9 appearing here is not a second coincidence. Which "latitude" the corners have depends on how the tetrahedron is oriented; with one vertex at a pole, the other three sit at about -19.47 degrees [deep reader's geometry note, memory medium-high; follows from sin = 1/3].

Published locations I checked on Wikipedia: - Olympus Mons, Mars: 18.650 N [web: https://en.wikipedia.org/wiki/Olympus_Mons]. - Mauna Loa, Hawaii: 19.479 N [web: https://en.wikipedia.org/wiki/Mauna_Loa]. - Great Red Spot, Jupiter: 22 degrees south of the equator [web: https://en.wikipedia.org/wiki/Great_Red_Spot]. - Neptune's Great Dark Spot: the page says only "southern hemisphere" and that it drifts north-south, no latitude [web: https://en.wikipedia.org/wiki/Great_Dark_Spot]. I could not check this one. - Hoagland: the Wikipedia article states he proposes "hyperdimensional physics", which mainstream physics rejects, and lists a podcast episode titled "19.5" [web: https://en.wikipedia.org/wiki/Richard_C._Hoagland]. The page does not detail the tetrahedron claim. That Hoagland made the 19.5 degree tetrahedral claim stays [memory, high]; whether St. Clair drew on it is unknown.

Weighing. Three prominent features sit within about 3 degrees of 19.47, which is why the claim reads as striking: Olympus Mons and Mauna Loa are close, the Great Red Spot is 22 degrees south (the sign is ambiguous in the patent, and planetographic and planetocentric latitudes differ by a few degrees at Jupiter, which I did not compute). Against the "all" form: Earth's volcanoes are spread over many latitudes (for example Mount St. Helens near 46 N and Etna near 38 N [memory, high]); Mars' other big Tharsis volcanoes lie much nearer the equator than Olympus Mons [memory, medium, not web-checked]; and Earth's features are tied to plate boundaries and hotspots, which is the mainstream account. The patent names no features, so a fair test needs a stated list of "large volcanoes and vortices" per planet and a null model (what fraction fall within a few degrees of any chosen latitude by chance). Nobody has done that, to my knowledge. Verdict recorded above: the "all" form is Tested, not supported by published locations; the weak form (some prominent features near that latitude) is Never properly tested and is easy to over-read, since any chosen latitude will have some features near it. The mainstream explanations (plate tectonics, mantle plumes, atmospheric dynamics) I state from [memory, high]; I did not weigh them further.

3. Corrections and process notes

• Hau slow light: the fetch summary in the first pass said strontium. Hau et al. used a sodium Bose-Einstein condensate, Nature 397, 594 (1999) [memory, high]; my earlier [memory, medium] recollection was correct. The sentence stays, now at high confidence. The Wikipedia fetch is the source for the 17 m/s figure only.
• Abandonment of 0197093: the "abandoned 2003-11-06 for non-response to an office action" statement rests on one fetch-tool summary of the Google Patents page and no office action was read. Treat it as one-source and unverified until the file wrapper (USPTO Patent Center) is read (qe.md R6). Explanation 1 in the absence analysis rests partly on it.
• Digest errors that QE found (tetrahedron sides are sqrt 1, sqrt 2, sqrt 3, not sqrt 5; no force formula in 0145019; "Sal (John)" is OCR garble) do not appear in any verdict here; the deep reads and my table used the corrected values.
• The earlier note that deep reads were "searched in full" for 0197093 is now supplemented by the applied-claims addendum, so no central claim rests on a digest alone.
• Corrections to the Investigator-side record: sister-application narratives remain second-hand; the Omega value (1.346812891e27 kg/m) is supplied by 0285325 as noted above. Both are already reflected in claims.md.

4. Additions to the record (lead items, no verdict)

• Copyright items [web per research/2026-10-04-copyright-office.md]: "Mars Experimental Planetary Institute--MEPI" (VAu000616124, 2004-02-23) and "Hyperspace Research Institute news: planck box corner at the cabibbo angle" (VAu000666486, 2005-04-27). They show the inventor's interest in Mars and a further tetrahedron-diagram claim in 2004-2005; they are leads for the absence analysis (a name to search) and the content is unread.
• The inventor's pointer "diagram tet0565 stored in the Library of Congress" (sister applications, second-hand) names the deposit that would show the intended derivation (Claim 3, settles-it list).
• One of the six examiner citations on 0197093 is a Paramount Pictures "Toy spaceship" design patent (USD260789S) [web, per research/2026-10-04-google-patents.md]. It is part of how the application was handled; I draw no conclusion from it.
• Later mentions by Diaz (2016) and Egely (2024) are Investigator-noted and unread (see Open questions).
• Records to add to the "would decide" list in the absence analysis: the application behind "Dual Rotor Propulsion System", and whether "Magnetic Vortex Generator" is a separate filing.

5. What would settle the new items

1. Hot-surface cooling and refrigeration: a heated plate near the running coil, cooling curve compared with a dummy coil of equal dissipation; a calorimetric heat-flow measurement per watt of input. Both are cheap and need no theory.
2. Carbon-electrode voltage: open-circuit voltage and load power on a carbon rod in the coil field, against the same rod in an ordinary AC magnetic field of equal amplitude (induction is the confounder).
3. Monopole lift and inertia-less claims: the coil on a balance or torsion pendulum with drive on, off and with a dummy load; a magnetometer map for non-zero divergence of B.
4. 19.47 degree claim: a fixed list of large volcanoes and vortex features per planet from published catalogues, fixed in advance, with a randomised-latitude null model.
5. Vector-potential claim: derive the force or torque on the rotor charge from A with the correct classical coupling (the Lorentz force uses E and B, and A enters through its time derivative and curl); show whether a static A at the rotors gives any force.

6. Residual uncertainty

Everything above rests on the inventor's filings plus small web checks. The 19.47 degree check used three Wikipedia pages and cannot stand for a survey. The Hoagland link is a lead. The existence and content of "Magnetic Vortex Generator" and "Dual Rotor Propulsion System" are unknown.