Arbiter: NRGFromTheVac ("Cold Electricity")

STEP 3. Inputs weighed: claims.md, facts.json, physicist.md, mathematician.md, contrarian.md, sites.md and the research/ notes. Tags: [read] project documents or the reviews' quotations of them; [web] with URL; [memory] with confidence; [calc] my own arithmetic.

Source gap (repeated from claims.md)

The project holds one 22-page blog printout (text readable) and one hand-annotated circuit image (no text layer). There is no primary document by the inventor, who posted as "NRGFromTheVacuum" (real name unknown). His videos are deleted. The demonstration is known through (a) the blogger's description, (b) commenters, (c) forum posts relayed by third parties, and (d) one Wayback capture of his 2008 video description [web: https://web.archive.org/web/20120423162527/http://www.youtube.com/watch?v=2cUS03yNl40, via the Investigator]. Nobody in this review watched the surviving re-upload (Mo6gViPywRI) or the XenomorphLabs relay replication (5otnGu9iZUM). The Genesis and EVGRAY PDFs were not downloaded or read. No oscilloscope trace, current measurement or temperature measurement exists in the record. Conclusions about what the circuit did rest on secondhand description; the physics verdicts rest on [calc] and [memory: high] standard results, not on any measurement of this device.

Who said what (important for weighing)

• The inventor's own principle, in his words (26 Aug 2008): apply 24 V briefly to a ferrite-cored coil "before the current saturates", disconnect, let the "counter-charged or negative potential ... move through the coil and the battery into the capacitor"; "its all about the timing" [web, Wayback]. He puts the battery in the discharge path. He never mentions positrons, negative inductance or Genesis in anything found.
• "No closed current loop", "positrons", "s-field", "negative inductance" and "atmospheric tap" are the blog author's (chadj2 = the blog author per the Investigator) and commenters' (Andreas, Warren) theories [read; web]. The blog author's first reaction was that an inductor "can step up voltage" so it was "not too interesting" [read: p.1]; by 2021 he conceded the inductor point and revised his theory to "not electromagnetic" [web: sites.md / semresearch notes]. These later views are kept in the record.
• Interested parties: none with a commercial or institutional stake appears in the record; all voices are hobbyists or the proponent himself. No institution tested or rejected the work.

Numbers recomputed where reviewers differed

Capacitor energy E = (1/2) C V^2 with C = 30 uF: 170 V gives 0.434 J, 200 V gives 0.60 J, 600 V gives 5.40 J. All three reviewers agree [calc].

Minimum inductance L for the coil to hold that energy at its full steady current I = 24 V / R (E = (1/2) L I^2):

Agreement: mathematician and contrarian figures match mine; the physicist's 34 mH matches the first row. The physicist's phrase that 5.4 J "would need 5 A and tens of mH only if the coil is large" is loose: at 5 A it needs 0.41 H, not tens of mH. Corrected.

Disagreement: how large can L be? Mathematician: 0.2 to a few henries (effective mu_r 50 to 1000 for a 3.5 in x 1 in ferrite). Contrarian: "a few tenths of a henry at most". My calculation: for an air-filled solenoid geometry mu0 N^2 A / l with N = 800, core diameter 1 in (A = 5.07e-4 m^2), length 3.5 in (l = 0.0889 m) gives 4.58 mH per unit relative permeability [calc]. So L = 0.23 H needs an effective mu_r of 50 and 0.92 H needs 200. But a ferrite rod's effective permeability is capped by its shape: mu_eff = mu_r / (1 + N_d (mu_r - 1)) with demagnetizing factor N_d [web: https://www.researchgate.net/publication/313400238_Investigation_on_an_Effective_Magnetic_Permeability_of_the_Rod-Shaped_Ferrites, formula as quoted in search summary]. For a length/diameter ratio of 3.5 a prolate-ellipsoid estimate gives N_d = 0.090, so mu_eff cannot exceed about 11 whatever the ferrite grade [calc]. That caps L at about 0.05 H (0.2 H if the "800 turns" are two series-aiding 800-turn windings, since series-aiding bifilar windings give about 4x one winding [memory: high]). Whether the core is an open rod is not stated (relayed spec "3.5 in x 1 in core", single source). Conclusions: - With R = 4.7 ohm, 170 V (0.43 J) needs L >= 0.033 H: attainable but near the shape-limited ceiling, so it is a tight, not a slack, fit. 200 V needs 0.046 H: at the ceiling. 600 V needs 0.41 H: not reachable by one inductive transfer from this coil. The inventor's source variant (a 40 V, 47,000 uF capacitor as a stiff source) or a pre-charged capacitor could account for more [web, relayed]; the record cannot tell. - With R = 30 ohm the coil cannot supply 0.43 J (needs 1.35 H). With R = 300 ohm, impossible (135 H). So 30 and 300 ohm cannot be the working coil's resistance if the stored energy came from the coil alone. - Plausibility check on 4.7 ohm: 800 turns of #23 AWG (about 0.056 ohm/m [memory: medium]) at a mean turn length of 8-11 cm is 3.6-4.9 ohm [calc]. That matches 4.7 ohm for a single 800-turn #23 winding, and #26 (0.13 ohm/m) would give 8.6-12 ohm. A 30 ohm figure would imply wire near 0.4 ohm/m (about #30 AWG). I therefore take 4.7 ohm as the better-supported coil resistance, while noting the wire-gauge tables are [memory: medium]. - Energy ceiling from ferrite saturation: with mu_eff about 11 and saturation flux density about 0.3 T [memory: low-medium], current at saturation is roughly 3 A, so the coil energy ceiling is of order 0.2 to 0.4 J for L = 0.05 H. This is within a factor of about 2 of the 0.43 J reported, which means the reported voltage is near the edge of what a single clean inductive transfer allows. This is the one place where a reviewer's claim of "nothing missing from the energy budget" (mathematician) is stronger than the evidence supports, and where the contrarian's caution ("a reason to want the real coil data") is better. It is not evidence of excess energy: hand switching, contact arcs, a capacitor not fully discharged beforehand, the 24 V battery staying in series during the transfer (inventor's own description: the battery adds its 24 V and can keep supplying current during the kick), an LC ring-up over multiple cycles, and a different core geometry are all unmeasured alternatives. - Lossless LC transfer check: V_peak = I0 sqrt(L/C). With L = 1 H, I0 = 0.8 A: 146 V (mathematician) [calc, confirmed]. With L = 0.033 H, I0 = 5.1 A: 169 V [calc]. Both give the observed order of magnitude.

Atmospheric tap (claim 5): fair-weather field about 100 V/m and air-earth current density about 2 pA/m^2 are now confirmed [web: https://space.fmi.fi/~makelaa/fairw.html and other results in the same search]. Physicist and mathematician framed it differently (2e-12 A x 1000 V = 2e-9 W per m^2 versus 100 V/m x 2e-12 A/m^2 = 2e-10 W per m^2) [calc, both confirmed]; both are of order nanowatts or less per square metre, so the conclusion is the same. Positron annihilation: 1 mA of positrons annihilating at 1.022 MeV per pair is 1.02 kW of gamma power [calc, confirmed].

Nottingham effect (cooling of a field-emitting surface): confirmed that field emission can cool or heat the emitter depending on field, work function and temperature [web: https://en.wikipedia.org/wiki/Nottingham_effect and https://journals.aps.org/pr/abstract/10.1103/PhysRev.151.327]. The effect is real but demonstrated for tiny emitting areas, not shown at a macroscopic spark gap. No verdict depends on it.

Ferrite and winding facts used above that I could only partly confirm are tagged accordingly.

Final verdicts per claim

Claim 1. 24 V battery charges 30 uF to about 200 V (170 V seen) via an 800-turn bifilar ferrite coil, "no closed loop / no battery current"

• Agreement: all three say the voltage-above-battery effect is ordinary switched-inductor (boost/flyback) behaviour; the blogger's "no closed current loop" is not the inventor's wording; the free-energy form was never measured; no energy-in versus energy-out test exists; the one bifilar replication (Andreas, about 500 turns) gave only battery voltage and the plain-wire/diode builds gave the voltage gain.
• Disagreement: (i) the mathematician and physicist say "Established" for the voltage and treat the energy budget as comfortably covered; the contrarian says the same verdict but flags that the 30 ohm figure and the 600 V figure fit awkwardly. (ii) The mathematician adds "Refuted by calculation as a net-gain claim only if the accounting is done", which is conditional and not a refutation of a claim nobody measured. (iii) The physicist and mathematician count Andreas's bifilar null as "Tested, not supported"; the contrarian agrees but calls it weak.
• Stronger argument: the physicist and contrarian on the split. A refutation by calculation needs an accounting that includes the inventor's actual coil, and that is not in the record. My recomputation shows 170 V fits a 4.7 ohm, tens-of-mH coil with little margin, and 600 V does not fit a single inductive transfer. That narrows the open question to "where did the energy come from in the 600 V case" (pre-charged capacitor, stiff source, repeated cycles) and does not by itself indicate a surplus.
• Final verdict (split):
• Capacitor voltage above battery voltage: Established as a phenomenon (ordinary inductive kick; informal replications with other coils). Revised by QE revision: whether the documented coil (30 ohm in both blog and image) reaches 170 V is unreconciled; see QE revision A.
• 600 V figure: single relayed statement, Never properly tested; not reachable by one transfer from the stated coil.
• "Without closed loop / without battery current": the inventor's own description contradicts it, and (QE revision B) so does the circuit drawing, which puts the battery in the charging path. As stated by the blogger for the whole process, Internally inconsistent with a coil needing current to hold energy (a coil with zero current stores no energy) and with a capacitor needing a return path. No measurement of battery current was made.
• Net energy gain / battery kept charged indefinitely (inventor, relayed): Never properly tested (no input-output energy balance by anyone; not dismissed after a test).
• Bifilar necessity: Tested, not supported at amateur level (one null with a non-matching coil; the only timing-controlled replication used a non-bifilar coil). Weak evidence. CitizenDC's successful build knotted the bifilar strands in parallel (QE revision D), so it is not a test of the opposing-field idea either.
• Settles it: same coil with L, R and turns measured; mechanical or MOSFET switch of known timing; oscilloscope on the capacitor voltage and a current probe in the battery lead; integrate battery energy (V x I dt) and compare to (1/2) C V^2 plus coil and contact losses; repeat with a plain-wire coil of equal inductance and with the bifilar coil, from an empty capacitor, 3 or more runs each.

Claim 2. Two spark types; an "endothermic" spark cools and frosts the electrodes

• Agreement: no temperature was measured; two spark appearances at a gap are routine; the cooling evidence is three accounts of the same videos (inventor via blog, AetherScientist 2016, chadj2 2021), not independent observations; the footage is lost except one third-party re-upload.
• Disagreement: the physicist gives "Plausible, untested" to the weak form (one electrode cooled while the other heats; Nottingham and Peltier-type transport, now confirmed as real mechanisms in principle) and nothing to the strong form; the mathematician notes the energy scale: the whole 0.43-0.6 J stored would only cool 0.1 g of copper by about 11 K, so frost at the limit of the budget unless evaporation or a heat pump effect is involved; the contrarian highlights it as the most testable item.
• Stronger argument: the contrarian and mathematician together. The claim is unmeasured, and the energy budget makes frost difficult but not impossible (humid air, tiny thermal mass, evaporation). "Endothermic" is not defined in the post's sense.
• Final verdict (verdict unchanged by QE revision C, which adds the lost videos and the 6 kV rig as context): Never properly tested (observed on video, not measured, not dismissed after a test). The post's wording ("energy radiated inward") is Not testable as stated; recast as a measured electrode temperature it becomes testable. The weak form (one electrode cooler than the other) is Plausible, untested.
• Settles it: thermocouple or thermal camera on both electrodes during each spark mode, with humidity recorded, in the same circuit; compare with a hot-arc control at the same stored energy.

Claim 3. "Cold electricity" is a separate electricity tied to positive charge (positrons or an "s-field")

• Agreement: no equations; "holes = positrons", "negative-mass positron", and the swapped signs on p.4 of the printout are errors in standard terms; no 511 keV gamma radiation is reported; neon glow and "cold to the touch" are explained by high-frequency RF, as the blog author himself reproduced [read: p.20]; the theory belongs to the blog thread, not the inventor.
• Disagreement: physicist "Refuted by calculation" (positron current of 1 mA would release about 1 kW of gamma power, confirmed [calc]); mathematician "Internally inconsistent"; contrarian "Not testable as stated", with the positron sub-claim "Never properly tested".
• Stronger argument: physicist for the positron-carrier form, since the annihilation budget is a calculation with accepted laws and requires no experiment. The mathematician is right about the inconsistent definitions. The contrarian is right that the vaguer "s-field/separate electricity" form has nothing measurable to refute. These are different forms of the claim and the verdict should be split.
• Final verdict (split): positrons as the current carriers: Refuted by calculation; "holes = positrons", "negative mass = positrons", sign swaps: Internally inconsistent; "s-field / time-reversed field / separate electricity" as stated: Not testable as stated. The experimental observations (purple neon, cold-feeling discharge) stand as ordinary RF/glow-discharge behaviour. Related work (Gray/CSET, Sweet, Lindemann) is kept as related lineage; nothing here confirms or excludes those devices.
• Settles it: gamma-ray spectrometer for 511 keV near the circuit; plain solenoid versus Rodin coil at the same drive with a field meter (the control chadj2 requested and no one reported).

Claim 4. Same-sign charges moving relative to each other in a bifilar coil give a negative field and negative inductance (Genesis)

• Agreement: commenters' paraphrase only; the Genesis PDF was not read by anyone; the commenters disagree among themselves on the mechanism; a bifilar coil gives about zero, never negative, inductance; passive negative inductance has no known example; Andreas's bifilar null.
• Disagreement: physicist and mathematician "Refuted by calculation" (passive form); contrarian declines to call it refuted because the source was not read.
• Stronger argument: for the passive form, the mathematician. Inductance matrix for passive windings is positive definite, so for equal windings L_series-opposing = 2L(1 - k) >= 0 [calc, e.g. k = 0.99 gives 0.02 L, k = 1 gives 0]. This applies to what the commenters state. The contrarian is right that the Genesis text itself was not examined, so the verdict is on the stated principle only.
• Final verdict (split): passive negative inductance supplying extra energy: Refuted by calculation; Genesis device as an experiment (+500 V charged bifilar coil, cooling): Never properly tested (no replication found; the "hoax" label is one compiler's characterization seen only as a search snippet, and is kept as such).
• Settles it: read the Genesis PDF (URL in sites.md); build the described coil, measure its L with an LCR meter at several frequencies and with and without +500 V bias; a negative reading with no active circuit would be the anomaly.

Claim 5. Rodin-type resonant coils tap atmospheric potential; ground electrons recombine with atmospheric positrons

• Agreement: asserted by one commenter (Warren), not the inventor; no input-output power; observations (lit neon, lamp underwater, broadband noise) match high-voltage RF coupling; requested solenoid control not reported.
• Disagreement: physicist "Refuted by calculation" for the atmospheric tap; mathematician "Never properly tested" with the mechanism refuted on the power bound; contrarian "Never properly tested".
• Stronger argument: the physicist/mathematician on the mechanism: a source of 2 pA/m^2 at 100 V/m delivers about 2e-10 W per m^2, confirmed [web]. This is a calculation, independent of the unproven positron story. The contrarian is correct that the coil effect itself (lit neon) was never measured with a control.
• Final verdict (split): atmospheric-positron power tap: Refuted by calculation; the observed neon glow: ordinary RF/displacement-current behaviour, Established as a phenomenon but its cause in this coil was Never properly tested (no control).
• Settles it: Rodin coil versus plain solenoid at identical drive; measure RF field strength near the coil, and input power; light the lamp from a shielded versus unshielded setup.

Claim 6. Replications are mixed; commenters give conventional explanations

• Agreement: the summary is faithful. Positive: Andreas (plain wire), Warren (diode-dependent 215-225 V vs 100-115 V), CitizenDC (125 V), Magnethos (no figures), XenomorphLabs relay run (timing dependence; declines to judge gain). Negative: gmeat; Andreas's bifilar build. No one measured energy. Warren's "75 percent" does not match the factor-of-2 ratio of his own numbers; minor.
• Final verdict: voltage gain replicated informally: Established (as boost action); anything beyond that: Never properly tested. Not dismissed without testing; several people tried informally. No funding or publication was blocked: the inventor was a YouTube hobbyist.
• Settles it: any instrumented replication with raw traces published.

Corrected claim list (replaces claims.md for later steps)

1. Inventor's claim: quickly connecting a 24 V battery to a ferrite-cored coil and then breaking the circuit sends a "counter potential" through coil and battery into a capacitor, giving much more voltage than the source; timing is the key (his 2008 description) [web]. Extra claims relayed in his name: up to "just over 600 V"; a mechanical switch could keep the battery charged indefinitely [web, relayed].
2. Blogger's overlay: the capacitor charges with no closed current loop and no battery current, by simultaneous make/break of two switches. (The blogger later wrote the switches are "not supposed to be closed at the same time" [web, 2016], so the switching sequence is undetermined.)
3. Two spark types at one gap, one said to cool and frost the electrodes (inventor; witnesses by recollection of lost videos).
4. "Cold electricity", positrons, "s-field", negative inductance (Genesis) and Rodin-coil atmospheric tapping are commenters' and blog author's theories, not the inventor's.
5. Replications are informal and mixed; nobody measured energy.
6. Coil data conflict (corrected by QE revision A): 4.7 ohm (inventor, relayed) versus 30 ohm (blog and image; the image reads "30 Ohms", not 300, and the two are probably not independent). Capacitor "30 uF 4000 V" (blog) vs "250 V 30 uF" (inventor, relayed) vs "30 F" as written in the image (almost certainly 30 uF). Both resistances can be matched by a plausible winding (4.7 ohm by about #23 AWG, 30 ohm by about #30 AWG); only the energy arithmetic favours 4.7 ohm. Which is right is unknown.
7. Circuit as drawn (QE revision B): a rotor commutator with one moving contact and two fixed contacts, plus switches S1 and S2; the battery stays in series with coil and capacitor while S1 and S2 are closed; "no closed loop" holds only after both switches open.

What would settle it (summary)

One instrumented bench test settles the central question: battery energy drawn (voltage x current integrated) versus capacitor energy and losses, with a measured coil, a defined switch and a plain-wire control. A thermocouple on the spark-gap electrodes settles the frost claim. A gamma detector and a solenoid control settle claims 3 and 5. Reading the Genesis and EVGRAY PDFs and watching Mo6gViPywRI and 5otnGu9iZUM settles what was actually claimed and shown.

Absence-of-evidence analysis (the inventor and his channel go dark)

The record: video uploaded 26 Aug 2008; 49,218 views and the page live on 23 Apr 2012; forum activity Sep 2008 to Mar 2009; named on a Dec 2010 conference wish list with no sign of attendance; by 2 Dec 2017 the "entire channel" was gone (commenters JW, Duncan); one video recovered from a private hard drive on 6 Mar 2018 [web, Investigator]. Real name, other accounts and later writings: none found in the queries logged in sites.md.

The silence does not support any single explanation. It does not show suppression: no record exists of pressure, and the videos were openly viewed for at least four years. It does not show a failure of the claim either. The disappearance is also a loss to the evidence, since the lost videos are the only possible source on the voltages, the frost and the switching. The footage AetherScientist says he holds (2016) and the Duncan Pickthall re-upload are the best leads.

Open questions

1. What were the coil's actual inductance, resistance, core geometry and wiring (series aiding or opposing), and what did the capacitor's initial state look like? The corrected arithmetic leaves the 170 V case just feasible and the 600 V case not.
2. Did the battery current integral equal or exceed the capacitor energy? Nobody has measured it.
3. Is the frost real, and which spark mode produces it?
4. What does the Genesis PDF actually claim, and was any version ever built?
5. Who was NRGFromTheVacuum, and where did he go?
6. What do the surviving XenomorphLabs and re-uploaded videos show (not watched by any reviewer)?

Items not checked by me

Videos, Genesis and EVGRAY PDFs, circuit image details beyond the annotations reported in facts.json, and the wire-gauge and ferrite saturation figures ([memory: medium/low]; no verdict depends on them). Web checks made: 3 (atmospheric electricity values; Nottingham effect; rod permeability formula).

QE revision (2026-10-04)

Inputs: qe.md (verdict MINOR) and its two deep reads: deep/ancilary-circuit-0790f9.md (Opus, image enlarged 3x to 8x) and deep/Cold-Electricity-7c-Scalar-EM-Research-6c603e.md section "Post body and comments of 20 Sep - 13 Oct 2012" (Sonnet, pp.1-5). Tags as above; [calc] is my arithmetic (Python, project venv). I made no new web checks; the new numbers are [calc] or [memory] as tagged.

Summary of verdict changes. No claim moves to a different scale label, but two sub-verdicts are tightened and one is split: 1. Claim 1, "voltage above the battery": the phenomenon stays Established; the specific statement that the documented coil reaches 170 V in one transfer is now Never properly tested / unreconciled (A below). My earlier phrase "arithmetic fits the 170 V case with little margin" is withdrawn for the 30 ohm coil. 2. Claim 1, "no closed loop / no battery current": still Internally inconsistent for the whole process, now with the circuit drawing as a second witness; but split: the loop is open only after both switches open (B below). 3. All other verdicts stand.

A. Coil resistance: the image says 30 Ohms, not 300 [read: deep/ancilary-circuit; page image p.1]

The deep read at 5x to 6x finds "30" then a capital O starting "Ohms", which reads as "300hms" at low resolution. The blog says "about 30 ohms" [read: Cold-Electricity-7c p.2]. So the image and the blog agree. They describe the same video and are probably not independent. The "300 ohm" row in my table is therefore an illustration only (the coil would need 135 to 1690 H), and my statement that "30 and 300 ohm cannot be the working coil's resistance" now applies to 30 ohm alone.

Consequences, recomputed. I also correct my earlier plausibility check on the winding: - Wire resistance: #23 AWG is about 20.4 ohm per 1000 ft, i.e. 0.067 ohm/m, not 0.056 [memory: high; QE spot check 7 agrees]. 800 turns at 8 to 11 cm mean turn gives 4.3 to 5.9 ohm [calc]. That still matches the inventor's relayed 4.7 ohm. - But #30 AWG is about 103 ohm per 1000 ft, i.e. 0.34 ohm/m [memory: high], and 800 turns of it at the same turn length gives 22 to 30 ohm [calc]. So the 30 ohm figure is also a plausible 800-turn winding (just thinner wire). My earlier statement that 4.7 ohm was "the only one consistent with both the winding length and the energy arithmetic" is withdrawn: winding length does not choose between them, because wire gauge is unknown. Only the energy arithmetic favours 4.7 ohm, and that is circular if the aim is to test the claim. Which resistance is right is unknown; the inventor's relayed 4.7 ohm and the blog and image 30 ohm remain side by side. - Energy arithmetic with the battery kept in series during the transfer (the topology in B). Using v_peak = V_b + sqrt(V_b^2 + (I_0 Z)^2) with Z = sqrt(L/C), I_0 = 24 V / R, C = 30 uF [calc; formula as derived in primer.md]:

With the battery in series, the 4.7 ohm requirement drops from 0.033 H (my earlier coil-only figure) to 0.024 H, which is comfortably inside the 0.05 H rod-core ceiling. The 30 ohm requirement drops from 1.35 H to 0.97 H, which is still well above it: with L = 0.05 H the 30 ohm coil peaks at 65 V, and with L = 0.2 H (series-aiding double winding) at 94 V [calc]. The rod ceiling itself rests on one geometry assumption (open 3.5 in by 1 in rod, from a relayed spec, single source) and on a demagnetizing-factor estimate; it is a bound on a stated geometry, not a measurement. A closed or nearly closed ferrite path (toroid, pot core, two rods joined by yokes) would raise L several-fold, and nothing in the record excludes that.

Net effect on the verdict: for a 4.7 ohm coil, 170 V is feasible (and 200 V nearly so) from one transfer; for the 30 ohm coil that the blog and image both document, 170 V is not reachable from an open-rod core by one transfer. The record cannot say which coil produced the 170 V. This is a reason to want the real coil data. It is not evidence of excess energy: the alternatives in my original text (multiple cycles, pre-charged capacitor, contact arcs, different core geometry, battery staying in series) are all still unmeasured. Sub-verdict: "170 V from the documented coil" = Never properly tested; the 600 V figure is unchanged (Never properly tested, not reachable at either resistance by one transfer from a rod).

B. Circuit topology: the battery is in the charging loop [read: deep/ancilary-circuit; page image p.1]

What the deep read finds (it corrects QE's tentative reading, which had the coil ending at the capacitor directly): - The drawing is a whiteboard frame captured from video, titled "what was he trying to show us?". Its author is unknown. It may be someone else's explanation of the video, not the inventor's own drawing (the deep read's inference). Treat it as a third-party reconstruction. - Capacitor "+" ties to battery "+" through switch S2 (and carries the label "200+V"). The coil's bottom lead goes to battery "-" through S1. The coil's top lead goes to a rotor wiper, which has two fixed contacts: one to the capacitor "+" side, one to the capacitor "-". The battery is "x2", 24 V, "24V DC 10AH" (red). - Sequence, as the deep read traces it: wiper on the "+" contact energises the coil straight from the battery (capacitor out of circuit); wiper moves to the "-" contact; coil current continues through battery + S2 + capacitor, charging it with "+" up. The battery is in series with the coil during that transfer, which is the battery-in-series flyback that primer.md derives. With the wiper on "-" from the start and no stored coil current, a lossless LC loop would reach at most 2 x 24 = 48 V [calc, deep read]. So 200 V needs the energise-then-transfer order. The order is inferred from the polarity labels; the drawing has no arrows. - If S1 and S2 are both opened while the wiper is on "-" (the blog's "closed and opened simultaneously" [read: p.2]), no metallic loop remains, coil and capacitor hang in series between two open switch gaps, and the stored energy would have to cross as arcs or stray capacitance. That is the situation behind the blogger's question "How does a voltage appear across C when there is no closed current loop?" [read: p.2], and it has a conventional answer: the loop is open only after the energy has been put into the coil's field by battery current. - Rotor timing, dwell and break-before-make are not shown.

Effect on claim 1: the blogger's "no closed loop / no battery current" is contradicted by (i) the inventor's own 2008 words (battery in the discharge path), (ii) the drawing. The verdict is split rather than changed: "no battery current at all" Internally inconsistent (a coil with zero current stores no energy; the drawing has the battery in the loop); "no loop once both switches are open" is trivially true and not evidence of a new mechanism. The earlier remark that the switching sequence is "undetermined" is narrowed: it is a rotor commutator plus S1 and S2, with sequence 2-then-capacitor-minus inferred, not shown.

C. Inventor's other material and the record that goes dark (added to claim 2 and the absence analysis)

From sites.md and research/ notes (QE items 7, 8, 10), previously condensed out of this file: - Three further videos by the inventor are lost: "Free Electrical Energy 2" (Y9irQJ6mivs; a spark gap run from the circuit; 16,959 views by 2012), "Cold Electricity Light Bulb" ("Negative end of a modified laser drive") and a frost video [web: sites.md Leads 1]. Nobody has watched any of them. These are priority-1 material, and their loss widens the gap for claim 2. - A forum poster, Reiyuki, 14 Apr 2009 (Energetic Forum thread 3317), wrote of "CSET replication as well as a simple 2-capacitor discharge concept ... tesla coil secondary ... Operating on at least 6kv" [web: sites.md; research/2026-10-04-energeticforum.md]. Reiyuki names "NRGFromTheVacuum" in a list of replicators and describes this high-voltage set-up as his; it is one third-party forum post, with no second source found yet. If the inventor also ran a 6 kV rig, the "cold" or frosting spark in the videos may have come from that rig and not from the 24 V circuit. That is an open possibility, not a finding. It bears on claim 2 and on the absence-of-evidence table: it adds a reason that records under a real name might exist (a high-voltage hobbyist community), but no name has surfaced. - Later alternative accounts on the live blog, kept as those commenters' characterizations [web: research/2026-10-04-semresearch.md]: Ivo (2022 to 2023): high-side versus low-side switching and a quenched discharge of about 40 ns as the source of "cold" current; Bertilo (2016): the annihilation-gamma objection to positrons (the same calculation as my claim 3 figure of about 1 kW per mA). - Floyd Sweet recollections, all by chadj2 and none with a cited source [read: deep read section 2; pp.15, 19, 21]: barium-ferrite bifilar coils; shocks from "cold current" that are "very painful and can leave an ache for over a week"; bifilar windings wound so their fields cancel. Also Warren, 21 Sep 2012 [read: p.4] and the post's p.1 framing: "Floyd Sweet with his VTA and John Searl who noticed cold components in his device" (assertion only). VTA is Sweet's Vacuum Triode Amplifier [memory: high]. These stay in the record as the blog's named precedents. They are unmeasured assertions here, and I make no verdict on the Sweet or Searl devices from this project.

None of this changes the claim 2 verdict, Never properly tested; it adds that three more of the inventor's own videos exist only in memory.

D. Replication detail for claim 1 (bifilar necessity)

CitizenDC (Energetic Forum 2711) knotted the bifilar strands in parallel ("combine the four bifilar wires into 2 wires") and wrote: "If you get it wrong ... you will see a nice spark. If you get it right, that big spark lands up in the capacitor" [web: research/2026-10-04-energeticforum.md]. His reading of the capacitor is "4kv 20uf" (a rating, not a measurement). A parallel connection gives an ordinary, positive inductance, so his success supports "boost action with a coil", not the opposing-field bifilar mechanism. This is consistent with my existing verdict for 1e.

E. Thread corrections and items now carried (pp.1-5 deep read)

• The comment thread starts 20 Sep 2012 (Warren, 12:52 PM, p.3), not October, and the page's own count is 30 comments at capture on 18 Sep 2013 [read: pp.3 and header]. Dates in my text and the reviews that say "Oct 2012" for the start should read 20 Sep 2012.
• Warren, 21 Sep 2012 [read: p.4]: "I understand cold electricity acts longitudinally and hot acts transverse"; "cold is a negative mass electron"; Avramenko plug and Dr. Stiffler's diode electrolysis ("completely backwards"); Moray valve by alpha emitters ionising germanium "into beta decay resulting in a flow of electrons ... I have yet to test this". All conjecture, no measurement. The longitudinal/transverse and negative-mass statements sit in the same family as claim 3, with the same verdict: Not testable as stated for "longitudinal vs transverse electricity" (no defined measurable difference), and negative-mass electrons as stated are Internally inconsistent with the electron's measured positive inertial mass and charge [memory: high]. The Moray valve idea is the commenter's own untested proposal.
• chadj2, 20 Sep 2012 [read: pp.3-4]: "mass is an electromagnetic effect", and a proposed test: "if you drop 2 identical objects from a tower except that one is magnetized will they both fall at the same speed?" It is posed as a question, with no result. Verdict on the question: it is a testable form of a free-fall (equivalence principle) comparison; tests of different materials falling together have been made for over a century (Eotvos-type torsion balances and successors) [memory: high], but I do not know of a test that specifically compares a magnetised and an unmagnetised sample of the same object, so that form is Never properly tested (unknown, not dismissed). "Mass is an electromagnetic effect" as a general statement is Not testable as stated. Learner exercise noted for the primer.
• chadj2, 13 Oct 2012 [read: p.5]: in his Stiffler device work an LED lit "to a lesser extent even at audio frequencies as low as 100Hz", and only the 1N4148 and old RF crystal diodes worked. This is an experimental observation by the blog author with no data in the printout; it parallels Warren's diode dependence (215 to 225 V with antique diodes versus 100 to 115 V with 1N4148, p.5-6). Diode dependence is expected from rectifier speed and recovery in a switched circuit; kept as related work, Never properly tested for any claim beyond that.
• Warren's patent link US 7,800,286 (earthionenergy.com) and U-235 "stable" remark [read: p.17] are related work cited in the thread; I did not check the patent.

F. Corrections of record

• 23 AWG resistance 0.056 ohm/m was low; correct 0.067 ohm/m (A above).

• "Coil-alone" L requirements are superseded by the battery-in-series table in A.
• Wording slip in the original claim 1 text, "tens of mH" for 5.4 J at 5 A, was already corrected above; not repeated.

What is still not done

Genesis and EVGRAY PDFs not downloaded or read; videos Mo6gViPywRI and 5otnGu9iZUM not watched; the image's top-right cropped word ("...TOR") undecidable; the rotor contact order inferred, not shown. Verdicts on claim 4 and on the rotor timing remain provisional on those.