Babcock-Paul: Arbiter (Step 3)

Project: US 8,860,273 B2, "External Field Interaction Motor" (P. M. Babcock, D. J. Babcock, P. N. Smith; Flyback Energy, Inc.). Reviews reconciled: physicist.md, mathematician.md, contrarian.md.

Source gap (repeated from Step 1)

The project has one document, a 17-page patent PDF with no text layer. Reviewers read page 1 as an image and pp. 10-17 by OCR. [Superseded by QE revision: all 17 pages were OCR'd, and the drawing sheets pp. 3-9 have since been read as page images (deep/ read).] The patent contains no measurements, efficiency figures, torque curves or equations [read; confirmed by Google Patents, [web] https://patents.google.com/patent/US8860273B2/en]. Every verdict below therefore rests on the patent's own statements plus textbook physics (standard E&M, E = 1/2 L I^2, P_mech = F v), not on any test. To obtain: dynamometer data, Flyback Energy test reports, provisional 61/290,259, publication US 2011/0156522 A1, and the incorporated US 7,602,157.

Context (now web-checked)

• Patent granted 14 Oct 2014; status Expired for non-payment of maintenance fees; security interest assigned to RPR Ventures LLC on 6 Dec 2016 [web: https://patents.google.com/patent/US8860273B2/en]. Reasons for lapse unknown; no evidence of suppression.
• Spokane Journal article: company projects 2-8% lighting-switch savings and 8-10% for cell-tower power conversion, no free-energy claim, and no independent testing is mentioned; claims come from company founders/CEO [web: https://www.spokanejournal.com/articles/9211-flyback-prepares-to-market-energy-capturing-technology]. Those figures concern the company's switch products, not this motor.
• Paul Babcock's 2012 conference talk (contrarian) remains [web, search snippet only, unverified]; no verdict depends on it.
• All other [memory] facts used are high confidence (Faraday, Lenz, E = 1/2 L I^2, USPTO examines novelty not function); no medium/low [memory] fact carries a verdict.

Recomputed numbers (reviewers differed in examples, all rechecked)

• Physicist: L = 1 mH, I = 10 A: W = 0.5 x 1e-3 x 100 = 50 mJ; at 1 kHz, 50 W. Correct.
• Contrarian: same W at 10 kHz = 500 W cycling through the clamp. Correct (circulating, not net, power).
• Mathematician: L = 10 mH, I = 5 A: W = 0.5 x 0.01 x 25 = 0.125 J; at 1 kHz 125 W; at 80% capture 0.1 J. Correct.
• Power budget: 50 N m at 3000 rpm: omega = 3000 x 2 pi / 60 = 314.2 rad/s, P = 15.7 kW. Correct. F = 50 N, v = 20 m/s: 1000 W; at 90% efficiency input = 1111 W, loss 111 W = 10% of input. Correct. F = 10 N, v = 20 m/s: 200 W. Correct.
• All agree: the maximum saving from any back-emf or flyback scheme is bounded by the loss terms, never by F v. No numerical disagreement among reviewers; the only disagreements are verdict labels.

Corrections to Step 1 (all three reviewers concur)

• "Greatly reduces" back-emf is Step 1's wording; the patent says "substantially reduced" and claims only a zero-current/open state while overlapped (claims 10, 11, 14).
• Claim 3 has two uses: supplement input to raise efficiency, OR run external devices (a load, not a gain).
• The patent does contain a strong qualitative sentence (little or no counter-motive force allows a given torque and speed "with much less electrical power") and loose wording "reduction in input impedance"; Step 1 said no efficiency statement. Correct to: no numbers, but qualitative efficiency assertions ("extremely efficient", "has produced", no data).
• Zero-current-while-overlapped is in all three independent claims (device claim 1, method claim 5, device claim 14; corrected by QE): it is the point of novelty (contrarian).
• Over-unity: never claimed in the patent; company disclaims it.

Claim-by-claim

Claim 1: thrust from fields interacting in arc-shaped regions outside the armature/coil rather than across a narrow gap

• Agree: all three say it is ordinary magnetostatics (Maxwell stress, F = -dW/dx), no equations in patent, no test, no new physics.
• Disagree: none on verdict (all Plausible, untested). Mathematician separates existence of thrust (trivially established) from competitiveness (never tested).
• Stronger argument: the split is right. Contrarian adds that fringe-field interaction generally gives lower force density than a small-gap circuit; this is [memory, high] engineering knowledge, and it sits against the patent's assertion that "increased area compensates for loss of force", which is unquantified.
• Settle: finite-element force-vs-position calculation, then dynamometer comparison with a conventional machine of equal mass.
• Never tested vs dismissed: never reported publicly; not dismissed.
• Final: Plausible, untested. (Thrust exists: Established. Advantage over gap motors: never properly tested.)

Claim 2: zero current in overlapped coils removes back-emf, allows high thrust at speed / "much less electrical power"

• Agree: weak form true and standard (an open coil still has induced voltage, e = -N dPhi/dt, but no current, so no opposing force; relaxes voltage-limited top speed; same family as phase sequencing, field weakening, switched reluctance). Strong form cannot beat P_in >= F v + losses.
• Disagree: strong-form label. Physicist: Never properly tested; mathematician and contrarian: Refuted by calculation if read below F v, contrarian adds Not testable as stated.
• Stronger argument: the calculation is decisive for the literal reading; the reviewers are describing different readings. Literal (input below mechanical output): Refuted by calculation. Charitable (less loss than some baseline): not testable as stated, no baseline or number; it is bounded to the loss fraction (about 10% in the 90%-efficient example). The physicist's "Never properly tested" is correct for the charitable reading only. Note also that an idle coil produces no force, so the benefit has a cost (lost thrust the patent says area must offset).
• Settle: dynamometer input power vs torque x speed over a speed range, against a conventional motor of equal mass.
• Final: split. Weak form: Established principle (untested in this geometry). Strong form, literal: Refuted by calculation. Strong form, charitable (lower losses): Not testable as stated, and never tested.

Claim 3: flyback energy captured in a capacitor and reused as secondary source, raising efficiency

• Agree: regenerative snubbers are standard; recovery bounded by 1/2 L I^2 already supplied by the source; no over-unity claim; "reduction in input impedance" is imprecise (reactance 2 pi f L is not changed by a switching clamp; a higher drive voltage speeds current rise, V = L dI/dt).
• Disagree: none on verdict. Mathematician alone notes the patent text on voltage direction is internally loose.
• Evidence: company's own figures (2-8%, 8-10%) are modest and untested by third parties [web above].
• Settle: measure input and output with and without the capture path on a dynamometer; per-event bookkeeping 1/2 L I^2 against the measured saving.
• Final: technique Established; efficiency gain in this motor Plausible, untested (a few percent at best); reading as net source or "extra" energy: Refuted by calculation (not claimed by the patent).

Claim 4: lighter, cheaper motor (non-metallic housing, little shielding, modular coils), high thrust-to-weight

• Agree: no data; physics unproblematic; eddy-current point correct; internal tension that thrust uses strong external fields while stray flux is said to be low (needs a field plot); the "ironless" wording in the reviews is wrong (see QE revision: the stator core and armature ring are ferrous).
• Disagree: physicist Plausible, untested; mathematician and contrarian Never properly tested.
• Stronger argument: the claim is a measurable ratio with no controlled test on record, so the scale's "Never properly tested" is the precise label. "Plausible, untested" reserves for claims with no test yet but consistent with known physics and no specific testable number; both fit, but because the claim is a measurable ratio (thrust per kg, stray field) with no number and no test on record, I apply Never properly tested. (The original rationale cited an "ironless = lower force density" trade-off; that premise was wrong, see QE revision. The label stands on the measurable-ratio ground alone.) Not dismissed; never reported.
• Settle: thrust per kg against a conventional machine, plus a stray-field map (FEM or gaussmeter).
• Final: Never properly tested (not dismissed; no suppression).

Tested / never tested / dismissed (contrarian's distinction preserved)

• Tested and failed: none. No experiment of any kind is on record, so "Tested, not supported" applies to nothing.
• Never properly tested: claims 1 (advantage), 2 (strong, charitable), 3 (size of gain), 4. The only numbers are company statements for its switch products, with no independent testing [web].
• Dismissed without testing: none documented; no critics on record. Lapse of the patent for unpaid fees is not evidence of dismissal.
• Anomalous result: none. There is no data to be anomalous. Contrarian's framing is endorsed: this is an honest engineering disclosure, a control case, not a suppressed-science case. "Flyback"/"reverse EMF" vocabulary invites over-unity misreading that the patent does not support.

Corrected claim list

1. Thrust can be produced by interaction of armature and stator fields in regions outside and between armature and nearby stator coils; competitiveness with gap-geometry motors is unproven. Plausible, untested. 2a. Holding an overlapped stator coil open-circuit/zero-current removes its opposing current and relaxes voltage-limited speed. Established principle; untested in this geometry. 2b. This gives "much less electrical power" for given torque and speed. Refuted by calculation if below F v; otherwise unquantified, limited to loss terms.
2. Flyback energy (1/2 L I^2) captured in a capacitor can supplement input power; gain is bounded and unmeasured here. Technique Established; gain Plausible, untested; net-source reading Refuted by calculation.
3. Non-metallic housing (not core), little shielding and modular coils give high thrust-to-weight; the stator core and armature ring are still ferrous. Never properly tested (rationale revised by QE).

What would settle it

One dynamometer campaign: build the motor and a conventional brushless motor of equal mass; log input V and I waveforms (with and without the flyback path), shaft torque and speed, mass, and stray field. Predicted under accepted physics: efficiency below 100%, a few percent gain at best from recovery, claim 2b never below F v. A prior finite-element magnetostatic model of force vs armature position and of stray flux would settle claims 1 and 4 on paper.

Open questions

• Does the external-region geometry actually deliver competitive thrust per kg, given fringe fields usually yield lower force density?
• How much is lost thrust from idle (overlapped) coils versus saved loss?
• What did the FIG. 12 circuit and the incorporated US 7,602,157 achieve in measurement? (drawings not read)
• Why was the patent allowed to lapse; did any tested prototype underperform? Unknown.
• How is "reduction in input impedance" meant to be read, if not as a change in L?

Final verdicts (summary)

Sources: [read] patent pp. 1, 10-17 (via reviewers); [web] https://patents.google.com/patent/US8860273B2/en ; [web] https://www.spokanejournal.com/articles/9211-flyback-prepares-to-market-energy-capturing-technology ; [memory, high] standard electromagnetism.

QE revision (after qe.md, verdict MINOR)

No final verdict changes. Four corrections to facts and rationale, plus new evidence from the drawings (deep read of pp. 3-9, deep/External-Field-Interaction-Motor-US8860273-3aadb4.md, [read: page image pp.3-9]) and from p.2 and pp.15-17 [read].

Page tags in the reviews

physicist.md, mathematician.md and contrarian.md label the patent's printed column numbers (or guesses) as "p.". Use the PDF page tags in primer.md and in this section. Example: the "much less electrical power" sentence is on PDF p.15, not p.12.

Corrections

• Independent claims: claims 1 (device), 5 (method) and 14 (device) all recite the substantially zero-current state while overlapped [read: pp.16-17]. Earlier text saying "both independent claims (1 and 5)" is superseded. Claim 12 sets the flux threshold for disabling a coil as a "predetermined function of armature speed" [read: p.17].
• Source gap: Step 0 OCR'd all 17 pages (65,746 characters). The text is available; what was not studied before this revision was the drawings, now read. The patent still contains no measurements, dimensions, voltages or test data on any drawing or in the text [read: page image pp.3-9].
• Iron is present (changes claim 4 rationale only). The stator core is "a continuous ring of ferrous material" (FIG. 3, p.12) or a gapped ferrous core (FIG. 4), and each armature has a steel or ferrous high-permeability ring 510 (FIGS. 5-6, p.13) [read: page image pp.4-5]. Only the housing and shielding are non-metallic or low-permeability. The "ironless", "eliminating the iron yoke" wording in the contrarian review and in this arbiter's earlier claim-4 text is wrong. Claim 4 stays Never properly tested because it is a measurable thrust-per-kg and stray-field claim with no test on record; it is not Tested, not supported and not dismissed.
• Prior art: the patent cites Bedini US 6,392,370 B1 (5/2002) and US 2002/0097013 A1 (7/2002) [read: p.2]. This is the likely reason the patent sits in an alternative-energy collection. It does not make the patent a free-energy claim. The reference list also holds Babcock family items (US 2007/0146958, US 2009/0073726, US 2009/0195199 listed as Babcock et al. or Babcock) and mainstream power-electronics papers on energy-recovery snubbers and active clamps (Mao 2005, Liu 2005, He, Bell) [read: p.2]. US 2009/0073726 (Babcock et al., current waveform construction for low-harmonic AC) is incorporated by reference with US 7,602,157 [read: pp.14-15]; add it to the "to obtain" list.

New evidence, claim 3 (FIG. 12 and FIG. 10) [read: page image pp.8-9]

• FIG. 12 has two switch modules (1210a, 1210b), not four; "1210d" in the OCR is an error. The "input rectifier" 1205 is a single blocking diode. The circuit is an H-bridge with a recovery switch (path D) returning coil current to capacitor 1215 above the supply voltage, a dissipative switched RC clamp (path C), and the capacitor-first then source sequence (paths A then B). FIG. 10 uses a gas tube clamp. All are textbook energy-recovery drive elements (asymmetric half-bridge and C-dump ideas from switched-reluctance drives) [memory, high].
• Energy ledger: recovery moves at most 1/2 L I^2 per event, and the clamp path C burns some of it. Worked check (illustrative, not from the patent): L = 1 mH, I = 10 A gives 0.050 J; with C = 100 uF at 48 V (0.1152 J) the capacitor rises to about 57.5 V. The faster turn-on (dI/dt = V/L) is the real content of "reduction in input impedance"; impedance is not changed. No path supplies extra energy.
• Verdict for claim 3 unchanged: technique Established; gain Plausible, untested; net-source reading Refuted by calculation (not claimed). Open question "what did FIG. 12 achieve in measurement" is now "nothing measured is disclosed". The deep read also notes a numbering slip (1035 used for two things) and "MC 110" as a typo for 1105; no verdict impact.

New evidence, claim 2 (strong form) [read]

• Generator mode: the patent says the machine "may be operated as a generator by allowing current to flow in a coil segment while armature flux has substantial coupling to the coil", recovering kinetic energy into a battery [read: p.16]. The patent itself therefore treats current in a flux-coupled coil as the energy-conversion channel (power = back-emf x current). This supports the strong-form refutation: opening a coil removes its drag and its thrust together. The deep read adds that the continuous iron ring couples magnet flux to the energised neighbour coils, so back-emf does not vanish from the coils that produce thrust [read: page image pp.4-6; reasoning, memory, high].
• Frequency inconsistency: the summary says switching "may be performed at low (e.g., synchronous) frequency using flyback capture" [read: p.11], while later text uses a "high frequency stream of pulsed D.C. current" [read: pp.14-15]. No quantity is given for either. Treat "frequency" as unspecified.
• Control: speed is set open-loop by the logic-generator clock rate and torque by master-pulse width (stepper-like) [read: pp.14-15]; a "320 degrees of open space", "vibration dampening" and "minimal hysteresis and core loss" are asserted without numbers [read: p.15].
• Verdicts unchanged: weak form Established principle; strong form literal Refuted by calculation; strong form charitable Not testable as stated, never tested.

New evidence, claims 1 and 4 [read: page image pp.4-7]

• "Low stray flux" is defensible only as far-field falloff. Thrust depends on fringing (radial) field outside the iron, which is stray flux by machine-design definition [memory, high]. With a continuous ferrous ring, the radial field concentrates where bucking excitation (S-S or N-N around each armature, FIG. 7) forces flux out of the iron, so the stator acts like a consequent pole, close to an ordinary permanent-magnet machine and unlike the "separate interaction planes" the drawings imply. The gapped core (FIG. 4) lets segments make local end-fields but raises reluctance. Cogging between magnets and stator iron is expected.
• The earlier remark that fringe-field force density is lower rests on [memory, high] general machine knowledge, not on a test; it is not tied to an "ironless" design.
• Verdicts unchanged: claim 1 Plausible, untested (thrust exists: Established; advantage over gap motors: Never properly tested); claim 4 Never properly tested. Settling experiment sharpened: a 2-D finite-element model (e.g. FEMM) of FIG. 7 with continuous then gapped core, then dynamometer thrust per kg.

Verdict changes

None. Table and corrected claim list above were edited only for the claim-3 note and claim-4 rationale.

Web checks: none added; no medium or low confidence [memory] fact carries a verdict. Sources for this section: [read] patent pp.2, 11-17 and page images pp.3-9 (via deep/ read).