Dollard-Eric: Quality Engineering (Step 4b)

Inputs checked: _text/manifest.json, all 7 digests in digest/, facts.json, claims.md, physicist.md, mathematician.md, contrarian.md, arbiter.md, primer.md, example.py. I also spot-checked the extracted full text in _text/. There is no deep/ folder, because this project was reviewed before digests and deep reads existed.

Verdict: MINOR. None of the six central verdicts is likely to reverse. One sub-verdict (claim 5, "counter-velocity is undefined") was based on a sampled reading of EPDPaper.pdf, and the extracted text contradicts it. Two open questions in the arbiter can now be answered from the extracted text. One key document (EPDPaper.pdf) was only sampled, and the prompt counts that as a defect. All of this can be fixed by deep reads plus updates to facts, the arbiter and the primer. Steps 2-4 do not need to be redone.

---

1. Coverage

Every manifest document has a digest and a facts.json entry, and no files were skipped. Step 1's statement that the Notes PDFs have "no text layer" was true of the source PDFs. Step 0 has since OCR'd all 29 pages of each, so later steps can now quote them as text.

Defect in a digest (does not affect later steps): both Notes digests give Dollard's propagation factor as "√2" (Free-Energy digest: "√2 (1.57× light velocity)"; 73155 digest: "√2 times light speed"). The OCR text reads "/2, or 1.57 times the velocity of light" and "7? /2" (the "π" was lost in OCR) [read: Free-Energy _text p.18]. The reviewers read the page images and correctly used π/2. Any future reuse of these digests should correct this.

2. Shortcuts

1. EPDPaper.pdf sampled (defect). Claim 5 rests on this document alone. facts.json says "text layer sampled", and each reviewer read a different subset: - physicist: "pp.1-30 of that layer", read lightly; - mathematician: "key passages"; - contrarian: TOC, typescript pp.1-2, 11-12, and file pp.23-30.

No one engaged with the "General Theory" section (file pp.13-22). The result is a substantive miss (section 3, item 1). 2. The 1987 interview was never re-read by any reviewer (minor defect). Claim 6 and the "many amperes of dielectric current" part of claim 2 rest on Step 1's page-image summary plus one page quoted by the mathematician. I checked the key quotations against the OCR text and they hold (section 4). Claim 6 is peripheral physics, so this is acceptable once confirmed. The interview still contains claims nobody evaluated (section 3). 3. "Transmission of Electricity" Parts I and II, unit question left open. The arbiter could not decide whether Part I and Part II contradict each other on C = 1/c², or what the factor "p" in (4π·10⁻⁹ p) means. It explicitly asks for file pp.14-18 to be re-read. That re-read was not done before the primer. It is a small matter, but it is a stated open item. 4. The rest of the coverage is acceptable: - digest-only coverage of the Rotating Magnetic Field article (mostly standard two-phase motor theory, peripheral); - digest-only coverage of the interview's social and political commentary.

3. Missed material

Items found in the digests or extracted text that facts.json, claims.md, the reviews, the arbiter and the primer leave out:

1. Dollard's definition of "counter-velocity" and the longitudinal magneto-dielectric (L.M.D.) wave [read: EPDPaper _text pp.21-22, OCR garbled but legible]. Every reviewer and the arbiter say "counter-velocity" is undefined. The primer lists it as "genuinely open". The typescript does define it: - The L.M.D. wave has its magnetic and dielectric fields "co-linear or longitudinal to the direction of electric wave propagation". - Its "proportion of magnetic induction to dielectric induction is not numerically equivalent to the velocity of light, nor is the dimension of propagation a velocity". - Its dimension is "per unit length per unit time", hence "counter-velocity". - The full telluric wave is complex: T.E.M. (transverse electromagnetic) is the real part and L.M.D. the imaginary part.

This needs proper review: - Is "per unit length per unit time" dimensionally coherent as a propagation speed? - Do co-linear E and B fields in a source-free region satisfy Maxwell's equations? (A plane wave in vacuum cannot be longitudinal.) - The definition probably moves this sub-claim from "Not testable as stated (undefined)" towards "Internally inconsistent" or "Refuted by calculation". 2. Harmonic-structure claim [read: EPDPaper p.21-22]. Transverse waves have harmonics Fo, 3Fo, 5Fo with amplitudes Ao, Ao/3, Ao/5. Longitudinal waves have a "degressive" series Fo, Fo/3, Fo/5 (sub-harmonics) with the same amplitudes. The generalised wave is the product of the two series. This is a concrete, checkable statement. A passive linear resonator does not generate sub-harmonics at Fo/3 without nonlinearity or parametric pumping. It is not mentioned anywhere. 3. The history and lineage argument in EPDPaper's General Theory [read: digest EPDPaper.01 pp.14-20]. Not evaluated: - Tesla's oscillation transformer (1900) as a "uni-polar" phase transformer. - Marconi's "Flat Top" as an enlarged-capacity workaround of Tesla's patent that lowered Q and made the system "ineffective for both electromagnetic and telluric waves". - Alexanderson's 1919 multiple-tuned (loaded) aerial for GE and the US Navy as an advance. - The assertion that "Alexanderson and Marconi engineers understood" their flat-tops were electrostatic, not electromagnetic, antennas.

These are historical claims that can be checked. The Alexanderson multiple-tuned antenna is real history [memory: high]. 4. Seismic-warning specifics [read: EPDPaper p.12]. Reception points are grouped in threes in a "Scott configuration", there is a network "discriminator" for a precursor waveform, and the "A.S.W." and "P.P.I." (plan-position indicator) terms are used. The contrarian mentions "Scott" briefly. Nobody notes that a Scott-T connection is a two-phase/three-phase transformer arrangement, not an array geometry. Dollard is borrowing a power-engineering term. 5. Farnsworth multipactor as "probably the only real free energy device that anyone ever demonstrated which can be reproduced" [read: Free-Energy _text p.2]. This is Dollard's one positive free-energy claim, and it bears directly on claim 6. The reviewers instead report only "all claimants are frauds but one unnamed person". The multipactor also appears in EPDPaper p.8 as a "negative conductance" amplifier element. A multipactor is a secondary-emission negative-resistance device that needs a DC supply [memory: high], so its "taking off" is oscillation powered from that supply. It belongs in claim 6. 6. "The velocity of light continually changes which changes the capacity of all capacitors and changes the effect of orgone" [read: Free-Energy _text p.7, interview]. The mathematician could not find this claim and the arbiter marked it "unconfirmed; not used". It is in the interview. Step 1 was right. It is a testable claim, contradicted by the fixed SI value of c and by the stability of capacitance standards [memory: high]. Same page: "slowing down the velocity of light" by the orgone accumulator's dielectric layers. 7. Interview claims with numbers or tests, not evaluated [read: digests]. Peripheral, but each should get one line in facts.json: - a four-year Sonoma State University study of planets, sun, weather and radio propagation; - RCA's use of planetary right-angle positions for radio-disturbance prediction (this matches the RCA engineer J.H. Nelson's 1951 work [memory: medium; needs a web check]); - naval sonar at about 750,000 W peak blamed for whale beachings; - the national grid "humming" at 60 cycles in phase; - a geomagnetic reversal turning Earth into a "dielectric" regime needing ceramic compasses. 8. Lightning energy figure [read: Dielectricity _text line 125]. "250 KWH for lightning stroke". This is a checkable number, about 0.9 GJ, and plausible against mainstream estimates of about 1-10 GJ per flash [memory: medium]. Not used. 9. Part I superconductor/pressure argument [read: Free-Energy _text p.16]. Persistent current would need "perpetual motion, an unlikely proposition". The contrarian caught this. The physicist's and arbiter's verdicts leave persistent-current experiments at [memory: medium] with no web check, even though the claim-1 strong-form verdict partly leans on it.

4. Accuracy spot checks (against _text)

Arithmetic: I re-ran example.py, and its numbers match primer section 5: - peak current 992.2 A; - heat 0.5000 J; - light-time ratio 4709; - 2.68 MA for 1 kWh at 20 kV.

5. Process

• Source tags. Present throughout, and honest about what was not opened. Weak points:
• The physicist cites a Tumblr post as [web] for Tesla's 471,240 km/s, and flags it as unverified, which is acceptable.
• The physicist's Brown-Lewis-Epstein efficiency figure came from a search summary. The arbiter's check failed and it correctly stopped relying on it.
• The contrarian's [web] Wikipedia citations were "not opened". The arbiter later opened Speed_of_electricity, Schumann_resonances and Monopole_antenna.
• Web checks. The arbiter used 6 page fetches (search quota exhausted) and listed what it did not recheck. Still at [memory: medium] or unchecked:
• persistent currents in superconductors (claim 1 strong form);
• the Loma Prieta and Parkfield reanalyses (claim 5d), which rest on the reviewers' secondhand [web];
• the 313 mV/m reference figure.

Only the first has any bearing on a verdict, and only alongside the stronger resistivity argument. Acceptable. - Verdict scale. Applied consistently by the arbiter, with split verdicts stated. The physicist used one off-scale label ("Established-as-stated-by-Steinmetz is a matter of emphasis"), and the arbiter did not carry it forward. The arbiter caught and corrected the contrarian's Schumann formula error (ideal n=1 is 10.6 Hz, not 7.5 Hz). - Primer. Follows the STEP 4 order 1-9, and every section is present: - prerequisites, each with a textbook; - symbols defined where introduced; - a quantitative worked example backed by example.py; - an evidence table; - 5 check-yourself questions with answers.

Defects against the reader profile: several acronyms are used without definition (OCR, GPS, RF, AM, AC, EM, ODE). BSRF is defined only in the last section. - example.py. Runs cleanly with .venv/bin/python (exit 0). SymPy solves the ideal LC case, and the closed-form damped peak agrees with RK4 integration to within 1.7e-11 A. The energy check closes.

6. Remediation actions

Deep reads (most important first):

1. EPDPaper-943421, pp.13-22, Opus. Read the General Theory section in full, rendering page images where the OCR is garbled. - Extract Dollard's exact definitions of the T.E.M. and L.M.D. waves, "counter-velocity" (dimension "per unit length per unit time"), the "degressive" sub-harmonic series (Fo, Fo/3, Fo/5) and the reciprocity departure (pp.13-14). - Check them for dimensional consistency and against Maxwell's equations in source-free and conducting media. - State whether they yield any testable prediction.
2. EPDPaper-943421, pp.1-12 and 23-30, Sonnet. Read the system description and Part Two claim-style enumerations completely. - List every concrete parameter (frequencies, loading values, spacings, lead times, the "Scott configuration", A.S.W., P.P.I.). - Confirm that Part Three is absent and that no drawings or test data exist. - Report the Tesla, Marconi and Alexanderson history claims (pp.14-20 overlap is fine).
3. Free-Energy-Research-of-Eric-Dollard-Notes-1986-1991-784bbd, pp.13-19, Sonnet (page images for equations). Transcribe the Part I equations exactly: C, L, the factor "p", t0² = -LC, Z0² = -L/C. Settle the arbiter's open question of whether Part II (p.17) contradicts Part I (p.14) on C = 1/c².
4. Free-Energy-Research-of-Eric-Dollard-Notes-1986-1991-784bbd, pp.2-12, Sonnet. Read the 1987 interview in full and extract every testable claim with page numbers. In particular: - the multipactor as a reproducible free-energy device; - the velocity of light changing capacitance; - "many amperes" of dielectric current; - the orgone accumulator "slowing light"; - Sonoma State and RCA radio-weather correlations; - geomagnetic reversal as a "dielectric" regime; - the frauds-but-one remark.

No deep read is needed for the 73155 scan (duplicate confirmed by OCR comparison above) or for the 1990 Dielectricity article (full text already read and spot-checked).

File changes after the deep reads:

• facts.json
• EPDPaper: set type to "other" (claim-style typescript); set how_read to "digest + deep read (pp.1-30)"; add the L.M.D. wave, counter-velocity and sub-harmonic claims.
• Free-Energy: set how_read to "page images + OCR text (Step 0) + digest"; add the multipactor claim; keep the "velocity of light changes" claim and cite p.7.
• 73155: record the page-by-page OCR comparison as the duplicate evidence. Name pp.26-27 (similarity 0.82 and 0.69) as checked only for gist, not one by one.
• Correct the digests' "√2" to "π/2" wherever it is reused.
• claims.md
• Claim 5: add that Dollard defines the telluric wave as T.E.M. (real) plus L.M.D. (imaginary, "counter-velocity", dimension per length per time), with a sub-harmonic series.
• Claim 6: add the multipactor as his one named example of a reproducible free-energy device.
• Update the source-gap note: OCR text now exists for both Notes scans.
• arbiter.md (add a "QE revision" section)
• Claim 5b: re-judge with the actual L.M.D. definition. A source-free longitudinal plane E/B wave is excluded by ∇·E = 0 and ∇·B = 0, so the likely verdict is Internally inconsistent or Refuted by calculation rather than "undefined".
• Assess the sub-harmonic series.
• Claim 6: add the multipactor (a DC-powered negative-resistance oscillator, not over-unity; web-check this).
• Resolve the open questions on "velocity of light changes" (present, interview p.7) and on Part I vs Part II from deep read 3.
• Web-check persistent-current decay bounds in superconducting rings and J.H. Nelson's RCA work if either is used.
• primer.md
• Section 4: add an L.M.D./counter-velocity row with a short Maxwell divergence argument (why a vacuum plane wave must be transverse).
• Section 6: update rows 5b and 6.
• Section 7: remove "what counter-velocity means" from "genuinely open", or rephrase it as "defined, but inconsistent with ...".
• Define OCR, GPS, RF, AM, AC, EM and ODE on first use, and define BSRF where JBR is first introduced.
• No change needed: example.py and the claim 1-4 verdicts.

---

Recheck

Checked against the current facts.json, claims.md, arbiter.md ("QE revision" section), primer.md, deep/ and example.py. example.py still runs (exit 0).

Defects

Remediation actions

Remaining issue (1): deep read 3 (page-image transcription of the Part I equations and the factor "p").

Final: OPEN (1 issues)

Recheck 2

Per the user's instruction, the remainder is paused pending their own quality review. No new deep reads were done. State rechecked against current facts.json, claims.md, arbiter.md, primer.md and deep/.

The lightning 250 kWh figure was marked "remains, acceptable" in Recheck. It is unchanged and stays outside the verdicts.

Final: OPEN (1 issues)