Frolov (Frollov-Alexander): Science Contrarian review

Role: Science Contrarian (judges by experiment, not consensus; same rigour applied to the researcher and to his critics). Sources: both PDFs read directly (single wire power transmission.pdf in full; FreeEnergy.pdf text layer, body read, patent appendices skimmed). Tags: [read], [web url], [memory confidence].

Source gap note: only two documents by Frolov exist in the folder, and no primary document for any device he cites (Brown, Sweet, Searle, Avramenko, Strebkov, Correa). Everything about those devices is Frolov's second-hand report [read], supplemented by the web checks below.

Corrections to Step 1

1. facts.json says FreeEnergy.pdf was "text layer sampled"; I extracted the whole text layer (about 1,600 lines) and read the body, so "full text (appendices skimmed)" is more accurate. Minor.
2. claims.md claim 5 says Brown's capacitors are real examples of reactionless thrust. Frolov is more careful than that: he writes "We have to exclude ideas on electrokinetic apparatus since it is just a reactive method and propulsion force is result of ionization flow" [read, FreeEnergy.pdf p.14-15]. He concedes ion flow for the electrokinetic apparatus and claims reactionless force only for Brown's asymmetric-surface patent (US 3,187,206) and the 1927 GB 300,311 capacitor. Yet two paragraphs earlier he writes "Ionization is not considered here!". The text is self-contradictory on this point. The facts.json key claim "produce propulsion without ionization" should be read in that light.
3. A claim Step 1 omitted that is experimental and Frolov's own: "the author of this article tested device" in which loading a transformer secondary that contains an arc made input power fall, with input measured "in DC battery circuit" to rule out phase-shift errors, at about 300 mA [read, FreeEnergy.pdf]. No numbers, no waveforms. Add as a sub-claim of claim 1/3.
4. Step 1 says the 3 W input is "30 V, 0.1 A" and 0.1 W output "about 3%": correct. I reproduced 0.5 x 1e-8 x (2000)^2 = 0.02 J, x 5 Hz = 0.1 W [read + own calculation]. Confirmed.
5. Step 1's quote that Strebkov's team "built and tested lines from 20 to 100 kW" is accurate [read], but it is Frolov's statement with no citation. Web: Strebkov's group published systems of 1 kW at 230 V / 10 kV / 100 kV (see below). The 20-100 kW figure is not confirmed by what I found.

Claim-by-claim review

Claim 1. The potential (scalar) field can do net work around a closed cycle, without fuel, if the negative-work half is removed (pulsing, gradients, polarity change, screening).

Frolov's text: the "work of potential field on closed trajectory ... is equal to zero. Yes, it is right for only case: if one part of work is positive ... and another part is negative. But one part ... can be excluded by ... spatial superposition, pulsing mode and change of interaction polarity or by the screening" [read].

a. Mathematics. No derivation is given at all; the argument is verbal. Internally the statement conflates two situations. In a static field E = -grad(phi), the work around any closed loop is the line integral of E, which is zero (curl E = 0, Stokes). Screening or a gradient does not change this: a screen that removes the deceleration half still must be inserted and removed, and that act costs work. If the field is pulsed in time, E is no longer a pure gradient on the loop and net work is possible, but then the energy is supplied by whatever switches the field. The article never accounts for the pulser's energy. So the claim is vague, not shown inconsistent.

b. Experiments. No measurement is reported for his tube with grid under the cathode or the unequal capacitor. Established results: every measured electrostatic or gravitational machine obeys the zero-loop-work rule; pulsed devices (a cyclotron, a Van de Graaff, a switched-mode converter) do deliver work, but always with input power accounted for.

c. Tested? By Frolov himself only, anecdotally. The DC-measured arc-loading result (sub-claim above) is the one place he says he excluded a measurement artefact, which is to his credit, but there are no data. No replication on record. Dismissed without testing? The article lacks any quantitative claim to test, so mainstream could not have tested it. I found nothing on the web showing anyone tried.

d. If true: a static conservative field would no longer be conservative (Maxwell's curl E = 0 for statics); energy conservation (first law) would fail for closed cycles, not only in the vacuum-source sense.

Strongest honest case: the weak form is true and used industrially: time-varying fields do net work on charges. A contrarian can say Frolov is pointing at the real fact that "no energy cost to keep a field in existence" (a charged capacitor holds a field without power) and that the cost is paid at switching. That is correct but gives no free energy, because the energy of the field itself comes from somewhere.

Case against the researcher: the article treats "no power to maintain a potential field" as if it implied "no power to switch it". No energy ledger anywhere.

Verdict (split): - Strong form (net work from a static conservative field, closed loop): Refuted by calculation (curl E = 0; line integral over a closed loop is zero). - Weak form (pulsed/gradient devices yielding output beyond input): Never properly tested; no neutral controlled test on record, and it was not dismissed after testing but never put in testable form (no input/output measurements).

Claim 2. A single-wire line with Avramenko plug carries wattless displacement current yet delivers real power "for free"; home test: 2 kV, 10 nF, 5 Hz gives 0.1 W vs 3 W input.

a. Mathematics. The arithmetic is right: 0.02 J x 5 = 0.1 W [read + own calculation]. But the conclusion does not follow from his own numbers. Output (0.1 W) is 3% of input (3 W DC). To claim "free", he would need to show input power unchanged with and without load; he gives no such measurement. A 3 W DC supply drives a 30 V, 10 kHz generator, a transformer, and diodes with core, winding and switch losses of order watts, so the load's 0.1 W could easily be hidden inside the unloaded draw, or fed by a 3% share of it. "Free since not connected with conduction current in the source circuit" is a definitional move: displacement current through a stray capacitance is real current in the source winding.

Worked check of plausibility: charge per spark Q = CV = 1e-8 x 2000 = 2e-5 C; at 5 Hz the mean current to the load is I = 100 microamps. If the single wire couples to the world through a stray capacitance C_s with the supply at f = 10 kHz and V about 2 kV peak, the displacement current is about 2 pi f C_s V. Setting this equal to 1e-4 A gives C_s = 1e-4 / (6.28e4 x 2e3) = about 0.8 pF. Less than a picofarad of capacitance to ground or surroundings (a wire and a person's hand are several pF) suffices. So the observation fits ordinary capacitive coupling and diode rectification, and the small spark rate (1-10 Hz) shows the supply is current-limited by exactly this tiny coupling. [own calculation from [read] numbers]

b. Consistency with experiment. Capacitive coupling and rectification are standard. Tesla's single-terminal lamp demonstrations (Tesla's lectures; [memory] medium; not web-checked, no verdict depends on exact date or place) are consistent with ordinary high-frequency electrostatic coupling. Avramenko's own figure, 60 mA x 50 V = 3 W at 100 kHz, gives no input power [read], so it cannot show over-unity either.

c. Tested? Frolov's own bench test and Avramenko's (reported via Russian-language reference [4], not available). Web: Strebkov and Nekrasov (VIESH, Moscow) built resonant single-wire systems of 1 kW at 230 V, 10 kV and 100 kV, at 3-30 kHz; related reported transmission efficiencies are modest, e.g. 66.7% over 2 m at 40 W and 10% over 30 m at 100 W [web https://link.springer.com/article/10.1007/s43236-022-00383-4 and https://viesh.ru/pre/development-of-resonance-methods-of-electric-power-transmission/ (as returned by search summary; I did not read the full texts)]. These are efficiencies below 100%, so they support transmission, not free energy. Not replicated as over-unity by anyone I found.

d. If true (free power from the load): conservation of energy, plus the notion that source power equals load plus loss (Poynting theorem) would fail. If only the weak form (single-wire capacitive power transfer works): nothing needs to change; it is a known, modest-efficiency technology under active engineering study.

Strongest honest case: single-wire resonant transmission is real and works, which mainstream textbooks underplay; Frolov and Strebkov were right that a return wire is not needed at high frequency. The sparse reporting is partly a Samizdat-publication problem (Russian journals, a magazine for the community), not proof of error.

Case against the researcher: no input-with-load vs input-without-load measurement, no waveform, no calorimetry, and he himself reports 3% efficiency, then calls it free by redefining what counts as the source circuit. Tesla "in London in 1892" is cited to a Russian paper on conduction current (ref. 4), not to Tesla.

Verdict (split): - Single-wire capacitive/resonant power transfer: Established (standard capacitive coupling; engineered at kW scale by Strebkov per the web sources). - "Free" energy from it: Never properly tested (no input/output controlled test; his own numbers show 3% output/input and cannot distinguish free from hidden input). Not dismissed after testing; it was never put in a testable form.

Claim 3. Resonance (Tesla transformers, standing waves, quarter-wave lines) can raise output until efficiency exceeds 100%.

a. Mathematics. For a driven series LC circuit, quality factor Q = (1/R) sqrt(L/C) gives voltage gain Q at resonance, but the power delivered by the source is still V_s^2 / R, and dissipated power equals the power in. Voltage gain is not power gain. Worked number: L = 100 microhenry, C = 1 nF, R = 5 ohm: Q = (1/5) x sqrt(1e-4/1e-9) = (1/5) x 316 = 63. A 10 V source gives about 630 V across C, but the source supplies 10^2/(2 x 5) = 10 W (rms-peak convention aside) and 10 W is dissipated in R. Efficiency equals output/input and is at most 100% for a passive network. Frolov says the Tesla coil "can be more than 100%" with no calculation [read].

b. Experiment. Nothing in the article: the text says only "Resonance mode of the free vibrations can be the real way to excess power output". All measured resonant converters are below unity (the web results above show 10-67%).

c. Tested? No test is cited in the article. Strebkov's published figures are below 100% efficiency [web, as above].

d. If true: passive linear circuits would need hidden gain; the energy-conservation theorem for linear circuits (Tellegen's theorem) would fail. Nothing in the article justifies such a revision.

Strongest honest case: apparent over-unity in resonant circuits often comes from comparing peak reactive power to real power, or mis-measuring with power factor. Frolov's DC measurement of input (in the arc test) shows he understood that issue in one case.

Verdict: Refuted by calculation for a passive resonant network (output cannot exceed input; the Q example above). Not tested for any active-source hidden mechanism; no data offered.

Claim 4. Energy conservation is "generalized" to a 4-D balance: load heat paired with an anti-entropy process in reversed time; vacuum/aether is an unlimited source; its density sets the rate of time (Kozyrev).

a. Mathematics. No equations at all, only "0 = A + (anti-A)" [read]. Terms (aether density, rate of time, anti-process) are undefined, so internal consistency cannot be assessed; the idea that a balance of zero resolves energy accounting is a bookkeeping device with no measurable content. Mild inconsistency: Frolov says the energy density of technical processes is small compared with "natural energy density of space (aether density)" [read] but gives no value, and also says the devices' output varies with season and day, which implies a modulated supply, not an "unlimited" one.

b. Experiment. Kozyrev's torsion-balance and "time" experiments are the base; I did not check their status on the web within budget. Mainstream sources I recall regard them as unreplicated or explained by thermal and air-current effects [memory low]. Casimir and zero-point energy are real measured effects but extractable energy from vacuum for continuous power has not been demonstrated [memory medium].

c. Tested? The one testable prediction is stated: output of such devices varies with season and time of day [read]. The only evidence is a remark that Sweet's VTA output "was variable in different time of day and night" [read]. No logs, no controls. Sweet's device was never independently tested (web: "Sweet never had the VTA certified by independent testing" per http://amasci.com/freenrg/sweetvta.html and the search summary; unreplicated, Sweet kept the magnet-conditioning process secret). So it is neither tested and failed nor tested and supported.

d. If true: causality (a reversed-time anti-process), the second law of thermodynamics, and the vacuum's energy-momentum structure would need revision.

Strongest honest case: a daily/seasonal modulation is at least a falsifiable signature. A neutral lab could log the output of any claimed device for a year against a sealed control. That is a cheap, real test the author himself proposes.

Case against the researcher: the theory is assembled from loosely related ideas (Whittaker potentials, Bearden, Naan, Kozyrev, Hasslberger) with no unified definition, and a negative theory step (renaming "density of time" as "density of aether" [read: ref [10]]) is a relabel, not new physics.

Verdict (split): - "Generalized conservation" / reversed-time anti-process: Not testable as stated (no definable measurable difference). - Seasonal/daily variation of free-energy output: Never properly tested (no neutral controlled test on record; not dismissed after a test).

Claim 5. Other people's anomalous devices (Brown capacitors, Sweet's VTA, Searle disks, Cooper, Hubbard, unbalanced wheels) are real examples, including reactionless thrust and gravity reduction.

a. Mathematics. Mostly none. Unbalanced wheels: the claim that lifting and lowering stages "are different physical systems" is not an energy argument; for a gravity wheel with a conservative potential, net work per cycle is zero [mechanics; Frolov himself states the zero-work rule]. The claim "efficiency can be million to one" for Brown is a quote [read]; efficiency is defined as output/input and has no meaning when input is not measured.

b. Experiment. - Brown capacitors: a force toward the smaller electrode at about 30 kV is real: Bahder and Fazi (US Army Research Laboratory) built four asymmetric capacitors and confirmed the net force. They estimated that ballistic ionic wind is at least three orders of magnitude too small to explain it and that ionic drift gives the right order, and said the physics is not understood [web https://apps.dtic.mil/sti/tr/pdf/ADA416740.pdf; per search summary]. This is a real anomaly in a mainstream lab, which the contrarian reading must record. But Tajmar (AIAA Journal 42(2), 2004) found no thrust in vacuum within accuracy, with any non-corona Biefeld-Brown force at least five orders of magnitude below corona-wind effects [web https://www.researchgate.net/publication/245425484_Biefeld-Brown_Effect_Misinterpretation_of_Corona_Wind_Phenomena, per search summary]. So air-based thrust is real and likely ionic (drift) in origin; reactionless thrust in vacuum is not seen. - Sweet: reported by Bearden (tested April 30, 1987 per web sources), never independently certified, not replicated successfully; process kept secret [web http://amasci.com/freenrg/sweetvta.html and search summary]. - Searle/Godin-Roshin ("7 kW, 100 kg axial force"): no source in the document; nothing found on the web in budget; unknown. - Cooper, Hubbard, unbalanced wheels: no data given [read].

c. Tested? Brown: yes (Bahder-Fazi 2002-3; Tajmar 2004; others), not supported for reactionless/vacuum thrust, with open questions on the air force mechanism. Sweet: never independently tested. Others: no data.

d. If true: Newton's third law and momentum conservation for isolated systems (reactionless thrust); the equivalence of gravitational and inertial effects for weight reduction; energy conservation for the over-unity ones.

Strongest honest case: Brown's air-thrust is a robust measured effect that a government laboratory called not understood, and it was dismissed by many for decades as "ion wind" without quantitative proof; the Bahder-Fazi estimate that simple ballistic ionic wind is too small is exactly the sort of result that mainstream reviewers ignore. Frolov is right that the claim needs a quantitative test. Sweet's secrecy and Bearden's unreplicated testimony are not data.

Case against the researcher: Frolov lists a dozen devices with a single sentence each, no numbers, treats weight reduction and self-cooling as confirmation, and builds a theory (negative time, Kozyrev) on a device that has no independent measurement. He also contradicts himself on ionization (Corrections, item 2).

Verdict (split by device): - Brown thrust in air with asymmetric capacitors: Established as a force, mechanism incompletely understood (ionic drift favoured). Reactionless/vacuum thrust: Tested, not supported (Tajmar 2004; one neutral test within stated accuracy; I did not find others in budget). - Sweet VTA: Never properly tested (no independent test on record; not dismissed after a test; replication blocked by secret process). - Searle/Godin-Roshin, Cooper, Hubbard, unbalanced wheels: Not testable as stated in the article (no measurements given). Gravity-wheel machines: Refuted by calculation (zero net work of a conservative gravitational field around a closed cycle).

How the claims were received (separate from the physics)

I found no evidence in the folder or in the web checks that Frolov's own bench work was ever tested by an outside laboratory and rejected, and no sign that it was dismissed on the basis of suppressed data. The pattern is different: the work was never put in a form a lab could test (no power-in vs power-out, no waveforms, no controls), so there was nothing for a critic to reanalyse or omit. That is the main weakness of the claim set, and it is the researcher's, not the critics', responsibility. The strongest legitimate experiments to run (cheap and decisive): (1) log true input power (RMS V and I, with phase, or a DC supply with calorimetry) with and without the spark load at the same frequency, and (2) repeat with the diodes reversed or removed.

Summary table

Web checks done (5 of about 6 budget)

• Bahder and Fazi, ARL report: https://apps.dtic.mil/sti/tr/pdf/ADA416740.pdf (search summary only; [memory] retagged [web]).
• Strebkov/VIESH single-wire: https://link.springer.com/article/10.1007/s43236-022-00383-4 and https://viesh.ru/pre/development-of-resonance-methods-of-electric-power-transmission/ (search summary only; 20-100 kW claim not confirmed).
• Sweet VTA independent testing: http://amasci.com/freenrg/sweetvta.html (search summary).
• Tajmar 2004: https://www.researchgate.net/publication/245425484_Biefeld-Brown_Effect_Misinterpretation_of_Corona_Wind_Phenomena (search summary).
• Not checked: Kozyrev experiments status [memory low]; Tesla 1892 London demonstration [memory medium]; Searle/Godin-Roshin. No verdict above depends on them beyond what is stated.