QE (Step 4b): Koen J. van Vlaenderen (folder "van Vlaenederen Koen")

Date 2026-10-04. QE agent (Opus). Inputs read: prompt.txt; _text/manifest.json and _textprep.log; all 4 digests; all 3 deep reads; facts.json; claims.md; sites.md; the list of research/ files; physicist.md, mathematician.md, contrarian.md; arbiter.md; primer.md; example.py (run). I also checked the extracted text (_text/*.txt) for the spot checks below. Tags: [read: doc p.N] extracted text; [check] my own arithmetic or algebra; [memory] with confidence.

Verdict: MINOR. The pipeline is thorough. All three documents were deep-read, the algebra was checked symbolically, example.py runs, and the Investigator's findings are carried into the arbiter and primer. The defects are: one sub-claim with no verdict (the "far-field inconsistency" of Lorenz-gauge Maxwell); one arbiter verdict whose reasoning I think is incomplete (the static-charge power term, "Internally inconsistent"); a few cited experiments that were dropped (Strebkov, Tesla "teleforce", Ignatiev's "unusually high" energy); two verdict labels that are not on the scale; and a small arithmetic slip. None of these changes a headline verdict on claims 1, 3, 4, 5 or 6. Fixing them means updating facts.json, claims.md, an arbiter "QE revision" section, and the primer.

---

1. Coverage

No document was skipped. There are no OCR pages and no source gap (the prep log shows 3 documents, 4 chunks, OCR 0). claims.md records "no source gap" correctly. The folder-name typo ("Vlaenederen" vs the printed "van Vlaenderen") is noted in every file.

The Investigator's downloads in research/files are not manifest documents, so they are not part of coverage. Three of them carry primary material that no role has read in full: the 2001 Hadronic Journal scan (the origin of S and theorem (33), and the basis of the priority claim; the sites.md note says it "Needs OCR or page-image reading"); the 2002 conference draft (wayback-raccoon-ELECTRODYNAMICS2.doc); and the viXra 1512.0297 / UJPA 2016 papers (the reviewers know these only through the Investigator's notes). See remediation R6.

2. Shortcuts

• 2015 paper pp.16-25 was covered by digest only in Step 1. This is the section that carries claims 4 and 5: the experiments review, Aharonov-Bohm (AB), Ignatiev-Leus, Monstein-Wesley (MW), Strebkov, Podkletnov, Moray, Correa and Chernetsky. A central claim resting on a digest alone is a defect under STEP 4b. It is mitigated: all three reviewers say they read pp.16-25 themselves from the text layer (physicist: "I re-read the 2015 paper's pp.16-25"; mathematician: "[read: pp.16-23]"; contrarian: "[read: 2015 paper pp.16-25]"), and their quotes check against the text (section 4). What is still missing is a page-tagged deep-read file for that section. As a result, facts.json's key_claims for the 2015 paper come from the digest, and some items in that section never reached claims.md, the arbiter or the primer (section 3). Remediation: deep read D1.
• The 2015 deep read covers pp.1-16, all the derivations. That is acceptable.
• Cited primary experimental papers were mostly not read: MW 2002, the Bray-Britton Comment, Rębilas 2008, Hively-Loebl 2019, the Fletcher affidavit, Rutherford's 4/3 pages, Cavalleri, McDonald's hairpin note, and the de Sangro replies. Every role says so plainly, and the arbiter says which verdicts rest on abstracts or search summaries. This is an honest limit, not a hidden shortcut. The one verdict it weakens is "Exploding wires: Tested, not supported (provisional)". That verdict rests on a search summary of Cavalleri's repeat, and the arbiter labels it provisional, which is correct.

3. Missed material (from my reading of the digests and the text)

1. The "far-field inconsistency" claim has no verdict. The 2015 paper's conclusions name two results: the Maxwell-Lorentz force law is wrong with respect to N3LM (Newton's third law of motion), and "Maxwell's electrodynamics theory in the Lorenz gauge is far field inconsistent" [read: 2015 p.21; digest part 02]. A third point goes with them: the "simple Maxwell" theory (drop E_l) is the Coulomb gauge (deep read 2015, §3.2). facts.json lists the far-field claim as a key claim. claims.md does not mention it. The physicist (claim 3 a) and the contrarian (claim 3 a) argue against it on [memory, high] grounds and cite Jackson, Am. J. Phys. 70, 917 (2002) [memory, medium]. The arbiter gives it no verdict, and the primer does not mention it at all. A learner would miss one of the paper's two headline conclusions, and the instructive lesson in it: Coulomb-gauge and Lorenz-gauge fields are the same physical field, and the paper uses "longitudinal" in two senses. Remediation: D2, then add it to claims.md claim 3, an arbiter verdict, and a primer paragraph in section 4.3.
2. Strebkov's single-wire resonant energy transmission [read: 2015 p.19, ref 6, Nekrasov, Strebkov, Avramenko]. The author cites it as LEM-wave evidence and states its own discriminator: if the system has a ground return the signals are TEM waves, and if not they are "most likely" LEM waves with energy flow S = (1/mu0) B_l E_l. This is in facts.json only. It is missing from claims.md, all three reviews, the arbiter and the primer. It is a cited experiment with a stated test, so it should be in the claim-4 evidence list.
3. Tesla's "teleforce" [read: 2015 p.18-19]. The author links Tesla's reported "teleforce" from non-dispersive high-voltage discharges to Podkletnov's impulse. Missing everywhere after the digest. Peripheral, but it belongs in the primer's Tesla row.
4. Ignatiev's "transmitted energy was unusually high" [read: 2003 p.8]. This is a second reported anomaly besides the 1.12c phase velocity. It is in the deep read but not in the reviews, the arbiter or the primer. It is single-source and quoted without numbers, and it should stay in the record as such.
5. Historical open-circuit argument [read: 2015 p.16]. The paper says the magnetostatic force experiments of Ampère, Gauss and Weber all used open circuits (batteries, capacitor banks), and that "the only example of a truly closed circuit stationary current" is a superconductor current induced by a magnet. This is the author's reason why claim 3 matters for real experiments. It is missing from the arbiter and the primer.
6. The author's own named decisive test, induction law 3.2 [read: 2015 p.17]: "primary sinusoidal divergent (curl-free) currents induce secondary ... El ... 90 degrees out of phase". He says the first experimenter to verify it also verifies B_l. claims.md and the arbiter mention it, but the primer omits it, although the primer's purpose includes things "the reader could test". It should be added to primer section 7 ("Genuinely open").
7. The 1999 paper's related-work context (Hofer's Material Wave Theory; Haisch-Rueda-Puthoff stochastic electrodynamics (SED) inertia from the Zero Point Field (ZPF); Petkov's electrostatic mass; Collins's mu = qA) [read: 1999 pp.1, 6-7]. This is in facts.json and the deep read, but primer section 4.4 names none of it, and it does not explain the "two mass models" the abstract promises. Minor. It matters for cross-project synthesis because SED/ZPF inertia recurs in other projects.
8. Nikolaev (Tomsk 2004), who gave B_l the name "scalar magnetic field" [read: 2015 p.6, ref 27], and the Marinov motor / Phipps and Moyssides, Peoglos, Hillas longitudinal-force references. These appear in one or two reviews but not in the arbiter or primer. Related-work leads. Minor.

4. Accuracy spot checks (against _text/)

Recomputed numbers [check]: - Arbiter 1.1: c^2 rho S = 8.988e16 x 1e-6 x 1e-6 = 8.99e4 W/m^3. Correct. - LES flux c S0^2/mu0 = 3.0e8 x 1e-12 / 1.2566e-6 = 239 W/m^2. Correct, and the arbiter's correction of the mathematician's 0.24 W/m^2 is right. - Rutherford shell masses 4.554e-31 / 6.073e-31 / 3.036e-31 / 9.109e-31 kg. Correct (r_e is defined so that e^2/(4 pi eps0 r_e) = m_e c^2). - Arbiter 1.3 dipole radial phase velocity c(1+x^2)/x^2: correct. Phase of E_r is x - atan(x), so dphi/dR = k x^2/(1+x^2). At x = 1.257 it gives 1.633c. - Small slip, arbiter 1.3: "A 12% excess over 0.5 km is 0.20 microseconds, about 8.6 degrees". The time saved at 1.12c is (500/c)(1 - 1/1.12) = 0.179 µs, which is 7.7° at 120 kHz. The 0.20 µs figure is 0.12 x 1.668 µs, which treats the excess as a delay difference. The conclusion is unchanged (still small against near-zone phase corrections). Fix the numbers.

5. Retention check

Question: was pertinent material from the sources or the Investigator left out of the arbiter or primer because a reviewer judged it negatively?

• Kept (good): Butterworth et al. 2013 (sceptics reproduced the MW pattern, "a yet-unidentified mechanism"). It is carried as "the cleanest anomaly in the file" in both the arbiter and the primer. The Fletcher affidavit is kept as evidence and its content is reported as unread. The Moray "never reproduced" [memory, medium] claim was correctly not asserted by the arbiter. Graneau's blocked-hairpin jets, Saumont and Johansson are kept beside the Cavalleri repeat. Hively-Loebl's copper-penetration report is kept, with the patent interest stated. The Ignatiev-Leus statement that conventional theory can deduce the effect is kept. Bruhn is kept as a critic with a stake, along with the author's reply and Bruhn's concession ("Maybe that some reasons can be found ..."). The priority dispute keeps both sides, the 13 Nov 2001 / 15 Aug 2002 dates, and the note that MW's own position is known only through an interested party. The Ethergy B.V. absence-of-evidence analysis has competing explanations and names the deciding records. The Google removal notice is recorded and not interpreted. The 99-citation trail and the Minotti-Modanese line are kept.
• Dropped, but not for negative judgement: Strebkov, teleforce, Ignatiev's "unusually high" energy, the Ampère/Gauss/Weber open-circuit argument, Nikolaev, Marinov/Phipps (section 3, items 2-5 and 8). These look like oversights, not suppression. All of them except Nikolaev are evidence the author offers for his claims. Under "keep pertinent information" they should be restored to the arbiter's claim-3 and claim-4 discussion and to the primer's evidence table, stated as the author's citations with their source status.
• Dropped sub-claim: far-field inconsistency (section 3, item 1). The reviewers judged it negatively on [memory] grounds, and the arbiter and primer then omitted it entirely. The claim and its rebuttal should both be in the record. This is the one retention defect.
• The physicist's canonical-energy sign argument (a negative point) is kept, and its status is honestly marked as not re-derived by the arbiter. Good.

6. Process

• Source tags: present and honest throughout. [memory] confidence levels are given. The arbiter separates [check] from reviewer claims.
• Web checks: the arbiter used 7 of 8 calls on load-bearing items (Butterworth wording, Minotti-Modanese status, Caprez et al., Hathaway/Woods, Tesla patent date). It records the two failed checks (McDonald, Moray) rather than asserting them. One [memory, medium] citation that a verdict leans on was not web-checked: Jackson, Am. J. Phys. 70, 917 (2002), the reference for "Lorenz and Coulomb gauge fields are identical, so no far-field inconsistency". Check it when the far-field verdict is added.
• Verdict scale: mostly consistent, with exceptions:
• Arbiter claim 2 and the primer evidence table use "Unresolved" / "not established on paper" for the S^2/(2 mu0) energy sign. That is not a scale verdict. Use "Not testable as stated" (on paper the sign depends on which conserved quantity is called energy) or "Internally inconsistent" if the canonical-energy argument is re-derived and accepted. State the choice.
• Arbiter claim 6, "fractal dimension 3/2: units artifact, not a finding": give a scale label. "Refuted by calculation" fits: the exponent follows from the unit convention (C = 1, charge density squared = mass density) and has the same value in Gaussian units for a different reason.
• Static-charge power term, "Internally inconsistent (energy accounting)" (arbiter 1.1; primer 4.2 and table). The arbiter's argument assumes that matter gains energy only at the rate f·v. The 2003 theorems (33)-(34) divided by mu0 give field power lost = J·E + c^2 rho S and force on matter = rho E + J x B + J S. The S-parts are exactly S J^mu = S(c rho, J) with S a Lorentz scalar [check, my reading of the equations; covariance argument memory, high]. So the pair is a covariant 4-force density whose S part points along the 4-current. Such a force conserves total energy-momentum if the charged matter's rest (internal) energy changes at rate c^2 rho S (for charge at rest: heating or cooling, or a rest-mass change of the charged body). It is not an unbalanced budget. On this reading the equations are consistent, and they make a definite (if unmagnitude-d) prediction: a static charged body in an S field gains or loses internal energy. The arbiter's own suggested test ("measure heat from a static charged body in a known S field") already points there. The physicist ("exchange term") and contrarian ("Not testable as stated") were closer. Recommend: the arbiter re-examines this in a QE revision. The likely verdict is "Not testable as stated (no S source or magnitude given; if S existed, the equations predict internal-energy change of charged matter, c^2 rho S)". Either way, keep the mathematician's imbalance argument and this rebuttal side by side. Deep read D3 settles it from the paper.
• Primer vs STEP 4 and the reader profile: all nine sections are present in order. Symbols are defined, and the prerequisites are introduced (the Lagrangian is glossed in one line). Textbook chapter numbers are correct [memory, high]: Griffiths 4th ed. ch. 5, 7, 8, 10, App. B; Feynman II ch. 27-28; Jackson ch. 6. The worked example has real numbers and a conceptual payoff (two bookkeepings, one measurable difference). Check-yourself has 5 questions with answers. Gaps: sections 3 items 1, 6 and 7 above.
• example.py: runs with the project venv. Output: Grassmann sum = +k ŷ; with J B_l the sum = 0; Grassmann + dP_field/dt = 0; k = 1.000e-5 N; Coulomb/magnetic ratio 2.31e7; closed loops agree to 5e-6 relative; "All checks passed." Exit 0. These match the primer's numbers.

7. Remediation actions

Deep reads (most important first):

• D1 (Sonnet). 2015 paper, pp.16-21. Make a page-tagged catalogue of every experimental, historical and device claim in Section 5 and the conclusions, with the author's numbers and wording: Ampère/Gauss/Weber open circuits and the superconductor remark (p.16); AB and Caprez (p.17); induction law 3.2 (p.17); MW Φ vs LEM and the charge-conservation remark (pp.17-18); macroscopic tunnelling (p.18); Tesla and teleforce (pp.18-19); Ignatiev-Leus (p.19); Strebkov and the ground-return test (p.19); Podkletnov (pp.19-20); Moray and Fletcher (p.20); Correa and Chernetsky (p.21); conclusions (p.21).
• D2 (Opus). 2015 paper, pp.8-10 and 21. The "far-field inconsistency" of Lorenz-gauge Maxwell (eq. 3.18 vs 3.16): state the exact argument and its definitional premise ("far fields are defined as two mutually inducing fields"). Check whether the Coulomb-gauge field (3.16) and the Jefimenko field (3.18) are the same physical field for a localized conserved source, with an explicit worked case (e.g. an oscillating dipole: show the 1/r longitudinal pieces cancel) so a verdict can rest on [check] rather than [memory].
• D3 (Opus). 2003 paper, pp.5-7 and 10. Do the generalised power and force theorems (33)-(34) form a covariant 4-force with an S J^mu term? If so, is energy conserved once the rest or internal energy of charged matter changes at c^2 rho S, so the static-charge power term is an exchange with matter's internal energy rather than an imbalance? Report what the paper itself says about where the energy goes.

Other actions (no deep read):

• R1 facts.json: add Strebkov (ground-return discriminator), teleforce, and Ignatiev's "unusually high" energy to the 2015 and 2003 entries. Update the 2015 how_read after D1.
• R2 claims.md: in claim 3, add "Maxwell's theory in the Lorenz gauge is far-field inconsistent; the 'simple Maxwell' theory (without E_l) needs no gauge". In claim 4, add Strebkov and the Ignatiev energy report.
• R3 arbiter.md, "QE revision" section: (a) a verdict on the far-field inconsistency claim from D2 (expected: "Refuted by calculation", with the Coulomb-gauge equivalence shown), with the Jackson 2002 citation web-checked; (b) re-examine 1.1, the static-charge power verdict, in light of D3, keeping both arguments; (c) replace "unresolved" (S^2 energy sign) and "units artifact" (fractal 3/2) with scale labels; (d) correct 1.3's "0.20 µs, 8.6°" to 0.18 µs, 7.7°; (e) add the Strebkov, teleforce, Ignatiev-energy and open-circuit-history items to the claim-3 and claim-4 discussion with their verdicts (likely "Never properly tested" or "Not testable as stated", as the author's citations).
• R4 primer.md: add a short paragraph and an evidence-table row on the far-field claim and the Coulomb-gauge point (section 4.3); add Strebkov, teleforce and the Ignatiev energy report to the evidence table; add induction law 3.2 to "Genuinely open" as the author's own named test; in 4.4, one or two sentences on Hofer's MWT and the SED/ZPF and electrostatic mass models; update section 4.2 and the table row if the static-charge verdict changes; replace "Unresolved" with the scale label the arbiter picks.
• R5: no example.py change is needed unless R3(b) changes the static-charge verdict. In that case, optionally add a short SymPy check that the S terms of (33)-(34) equal S(c rho, J).
• R6 (Investigator, next run, optional): read the 2001 Hadronic Journal scan (page images) to settle what theorem (33) looked like in 2001 for the priority question; read the 2002 conference draft (ELECTRODYNAMICS2.doc); retry thmoray.org via Wayback for the Fletcher affidavit text; run the Espacenet and KvK searches not reached on the first run.

---

Recheck

Date 2026-10-04. Re-read the current arbiter.md (new section 8, "QE revision"), primer.md, claims.md, facts.json (parses as valid JSON) and example.py (run with the project .venv; exit 0, "All checks passed", including new Part F).

Residual notes, not blocking: the arbiter did not rerun the D2 SymPy/NumPy calculation and says so; the Jackson citation rests on an abstract. Both are stated honestly in the text.

Final: PASS