RankingsPrimerArbiterPhysicistMathematicianContrarianQESites logClaimsSourcesResearch notes

Stiffler: Mathematician review (Step 2)

Reviewer: PhD mathematician. Scope: derivations, notation, units, hidden assumptions. Physical verdicts belong to the physicist; where a verdict here depends on physics I say so. Method: I read facts.json, claims.md, every deep/ file, the sites.md "Leads for review", and the research/ notes on Energetic Forum, Scribd and YouTube. No downloaded documents exist in research/files/ (the folder holds only saved web pages under pages/). I re-ran the arithmetic myself (Python, sympy/numpy venv; script at /tmp/stiffler_math/calc.py). Tags: [read] source text (via the deep reads); [calc] my arithmetic; [web] URL; [memory] with confidence.

Source gap (repeated from claims.md): all nine documents are Stiffler's own pages and white papers (an interested party who offered SEC boards and licences through Stiffler Scientific). The ECAT scope photos (ecattp001-005), the SEC Special Coil figures 11, 13 and 16, and the companion "Measurements for SEC Exciters" data were not seen. Several statements below therefore turn on numbers that are absent, and I say so each time.

Corrections to Step 1

  1. deep/AA-WHITEPAPERS-17695261 (Wireless guide), "Coil data": the deep read says 143 turns in 2 7/8 in "implies fine wire, roughly #30-#32 or finer". That is wrong. 73 mm / 143 turns = 0.51 mm pitch, which is almost exactly the diameter of #24 AWG (0.511 mm bare) [calc]. The winding is consistent with #24 enamel wire (the gauge used for L2 and the 12 ft link). Not a contradiction in the guide; the guide just omits the gauge.
  2. deep/ExHeatExp section 4 says leaving out glass "would make Q smaller than the true heat produced, not larger". True, but it is the wrong half of the risk. The error that can inflate the ratio is on the temperature side (non-uniform water temperature, below) or the input side (how V and I were integrated), not the omitted glass. Adding an ~8 g Pyrex tube (0.84 J/g K, my assumed values [memory, medium]) would make CEC 1.40, not lower [calc].
  3. claims.md claim 1 says the arithmetic "does not close: 12 W + 6 W is 18 W, and 8.25 W at 12 V is 0.69 A versus 90 mA quoted". Fair, but incomplete. 18.11/8.25 = 2.195 (page says 2.194) [calc]. And the 90 mA is described as a "quiescent current", so the page may be comparing different operating points; this is an ambiguity, not a proven inconsistency. If input really were 12 V x 0.09 A = 1.08 W, the claimed gain would be 16.8, not 2.2 [calc]. Either way the page is not self-consistent as written.
  4. claims.md claim 3 and deep/SECExciter: confirmed. Both printed energies are a factor 10 too small (0.18225 J and 0.3645 J, not 1.8225E-2 and 3.6450E-2) [calc]. The ratio 2 is unaffected.
  5. claims.md claim 6 says the papers "concede pickup errors"; correct. Add: the 12.0 MHz "amplification peaks" on SEC Special Coil are numerically the +6 dBV high-side points of the secondary tests (12,002,053 and 11,963,991 Hz), so they are band edges of the response, not independently identified peaks (deep read section 5, confirmed against the table).
  6. From the Investigator (sites.md "Corrections"): the ECAT work dates from 1990 by Stiffler's own forum post (not only 1993-96). Accepted; it does not affect any calculation.

Mathematical points that cut across claims

A. CEC is a ratio of two integrals, and every document leaves the integrals undefined. Eq 1 of the single-wire paper: CEC% = (Uout/Uin) x 100 with "U expressed in Joule Seconds" [read: p.2]. The reference note on p.15 instead lists Watt hours, kWh, BTU and calories. A joule-second is a unit of action (energy x time), not of energy. A ratio of two quantities in the same unit is dimensionless whichever unit is used, so the slip does not by itself change the ratio, but it shows U was never defined. Eq 3 of the measurement paper writes Uinput = Avg(dVin x dIin): an average of an instantaneous product (power, in watts), with "d" unexplained, and the paper itself says its energy equations "do not include a value for Time" [read: p.14]. The statement "UCEC > 1 implies PCEC > 1" holds only if both energies are accumulated over the same interval; the paper does not say so. In ExHeatExp the definition is finally operational (Fig 15): CEC = Q / Sum(V x I x t), with Q = c m dT [read: Fig 15 image], and in that form it is a proper energy ratio. [calc] I reproduced Fig 15 exactly: Q = 4.187 x 25 x 15.10 = 1580.6 J; 1580.593/1198.759 = 1.3185.

B. A gain in voltage, in light, or in "brightness" is not a gain in power. In the single-wire paper the scope shows 4.32 V p-p at points A and B and 26.5 V p-p at C: a voltage ratio of 6.1 [calc]. That is what any resonant step-up (a high-Q coil, or a transformer with turns ratio) does. Power is voltage times current times phase factor. The paper gives no current at any node, so no power ratio exists. Likewise 130-168 V at the free end of the secondary is an open-circuit style figure at a very high impedance node; "168/2.7 = 62 LEDs" [read: p.14] [calc: 62.2, or 56 at 3.0 V per LED] only says the voltage would forward-bias 62 LEDs, not that they can be driven at any useful current.

C. The calibration of output by LED light (Eq 2) is undefined. POut = Ptest x LED's x n [read: p.12]. n is never defined, Ptest comes from an unshown calibration curve, and the calibration compares DC-driven LED light with light from an LED driven by a rectified 12 MHz waveform of unknown peak and duty. Light vs. average electrical power is not linear across such waveforms in general (LED efficacy depends on current density). No POut value is ever computed. So Eq 2 is a formula with an undefined symbol and no data.

D. "Coherent summation" is redistribution, not creation (ECAT analogy). The ECAT page's street-corner analogy says phasing ambient sounds "properly" gives pressure "greater than any one of the individual sounds". For N equal sources of amplitude A, the in-phase point has intensity proportional to (NA)^2 = N^2 A^2, while incoherent sources give N A^2 (for N = 10: 100 versus 10) [calc]. But the extra intensity at the constructive points is exactly paid for by destructive points elsewhere; integrated over space and time the energy is the sum of the sources' energies. The analogy therefore shows concentration of energy that was already supplied, and it says nothing about adding energy. The page never says what the ambient source in the ECAT circuit is.

E. If the output is electrical power, "COP > 1 does not imply self-running" needs an argument that the page does not give. ECAT claims 18.11 W "produced" for 8.25 W in, the output being lamp plus transformer power. Close the loop: if a fraction eta of the output power can be fed back to the input rail, the loop sustains itself when eta x COP >= 1, that is eta >= 1/2.194 = 0.456 [calc]. A rectifier and DC converter at 90% would leave 16 W for an 8.25 W demand. The "door in the wall" analogy argues that you need continuing external input, but it does not show why an electrical output of 2.2 times the input could not be fed back. Either the 18 W is not real output power (measurement issue) or the device would be self-running with a trivial add-on. Stiffler's own later posts accept this logic in the other direction: "energy cannot be manufactured" and CEC above 1 does not permit looping (Energetic Forum post #2122, 2011-05-30, per the Investigator [web] https://www.energeticforum.com/forum/energetic-forum-discussion/renewable-energy/4067-high-voltage-from-thin-air). The two positions (a gain of 2.2 yet no loop possible) are mutually incompatible unless the gain is a measurement artefact or the output is not usable electrical power. The inventor also wrote that "the math could not support the (conversion) concept" and therefore renamed "Conversion" to "Coherence" (post #369, 2009-05-01, [web] .../page25, via the Investigator). That is the inventor's own report that an earlier mathematical model failed; it should be kept in the record.

Claim-by-claim (questions a-d, then verdict)

Claim 1: ECAT air-core MOSFET circuit gives 8.25 W in, 18.11 W out (gain 2.194) by "Poly-Phasing" capture of external energy

a. Internal consistency: Partly. [calc] 12 W + 6 W = 18 W, not 18.11 W. 18.11/8.25 = 2.195. The logic "the transformer wastes 6 W of 8.25 W input yet the lamp gets 12 W" is only meaningful if 18 W is the power at the transformer input and was measured; the page gives no instrument or method [read: p.1]. 8.25 W at 12 V is 0.69 A, versus a "quiescent" 90 mA (1.08 W). "Poly-Phasing", "cross phasing" and "virtual energy pump" have no equations, frequencies, phases or budget. A 120 V, 12 W lamp rating is a nominal rating at 60 Hz sine, not a measured power at 4.3 kHz pulses with "arcing at the base of the bulb". On the input side the measurement is in principle easy and sound: a 12 V battery or supply behind a 2000 uF capacitor gives mean power = mean V x mean I, if the supply sags little [calc/reasoning]. The hard side is the output waveform (non-sinusoidal, 4.3 kHz, HV pulses), which needs true V x I integration, which the page does not claim to have done. b. Experimental consistency: Conservation of energy in a closed circuit forbids it; the claim is that the circuit is open (external energy captured). No identified external source, no measured quantity of it. The inventor's own statement that replacing air-core coils with ferrite or metal-core parts removes the gain is an observation that a mainstream engineer would first read as a change in loading, resonance and loss rather than loss of a hidden source; no data are given either way. c. Tested? By the inventor only, once, no repeat, no error bars. Stiffler says Internet copies "failed" because builders changed the design [read: p.1]; no copy is named or linked. Naudin's Time Energy Pump is described by Stiffler as having "fell short of sustained usable excess energy" [read: p.1]; the JLN Labs index is live and does not mention Stiffler (sites.md). No independent test on record. One forum poster estimated "roughly 4 W" extra from an ECAT video but said proper measurement was needed (Investigator, thread 9344 post #3, [web] energeticforum.com). d. If true: conservation of energy would stand only if an external source is identified; otherwise energy conservation in circuit theory (Poynting's theorem for a closed system) would fail.

Verdict: split. - Numerical claim (COP 2.194): Never properly tested (one unrepeated figure from the inventor; the page's numbers do not close; not dismissed without testing, it was simply never independently measured). - Mechanism ("Poly-Phasing", "virtual energy pump", capture from the space between coils): Not testable as stated (no equations, no source named).

Claim 2: Spatial Energy Coherence (single-wire coil with Avramenko diode plug) draws "cohered" energy from an "Energy Lattice"; CEC = Uout/Uin

a. Mathematically, there is no SEC theory: SEC is defined only in a reference note, "a theory developed by Dr. Ronald Stiffler on the Coherence of energy from the Energy Lattice" [read: p.15], with no derivation. CEC is defined (Eq 1, units inconsistent; see A) and never evaluated. Eq 2 has an undefined n (C). The measured data are voltages (4.32, 26.5, 130+, 168 V), frequencies and photographs. My own resonance checks: C1 (400 pF) with L1 (8.8 uH) resonates at 2.68 MHz; with L1 plus the primary (14.98 uH), 2.06 MHz; the secondary alone (686 uH with 23 pF), 1.27 MHz [calc]. None is near the 10-12 MHz peak; the paper offers no model, and stray capacitance and coupling are unmeasured, so this is not a refutation, only an unexplained gap. The frequency-response chart (Fig 18) has swapped axes, "relative power" ticks of exactly 0-4 at the five points (looks like index labels, not measurements) and a point at (4, 0) that is unexplained [read: page image p.14]. The "amplification" on SEC Special Coil is undefined (no ratio, units or power); the stated 12.0 MHz peaks coincide with the +6 dB band edges (see Corrections). Further: on SEC Special Coil, changing C1 by 17.5% moved the null by only 0.6-0.7% where a pure LC would move 10.1% (sqrt(400/330) = 1.101) [calc]. That means the nulls tabulated are coil self-resonances, not C1-L1 resonances, which is ordinary; the page does not say it. The impedance tables Tbl 1-6 on SEC Special Coil use a wrong formula. I checked the identification: tabulated Z = sqrt(|Xl - Xc| + R), not sqrt(R^2 + X^2). Example, Tbl 1 at 4 MHz: X = 79.37 ohm, true |Z| = 79.37 ohm, table gives 8.96 (sqrt(80.37) = 8.965) [calc]. Minima fall at the right frequencies (the function is monotonic in |X|) but all tabulated impedances are wrong, by a factor of about 9 to 20. b. Experiment: A single-wire resonant link with a diode plug is established engineering (see claim 4). A lit LED and a neon bulb lit by body capacitance are what a high-impedance 130-170 V RF node produces; none of the documents shows more power out than in. The inventor himself says SEC does not create energy ("conversion ... with a 'mathematical' gain", Energetic Forum 2009-03-12 [web], via the Investigator). c. Tests: by the inventor and by replicators; the replicators' reports (Lidmotor videos 2009-2011; Raui, Inquorate, zeos71 on Energetic Forum) show voltages, ringing and brightness, and none gives an input-output energy balance (Investigator notes). The late measurement relayed in 2018 (CREE board draws 35 mA at 18 V, 630 mW) gives an input only. One replicator (2019-12-01) says that at certain frequencies V and I are out of phase so that "the power calculation is always negative" [web]; this is a replicator's own reading and bears directly on how power is computed from V and I (it needs the phase angle; real power = Vrms Irms cos(phi)). The Stiffler Loop series (2017-18, 163 videos on his channel) was not reviewed by anyone in this project. d. If true: a new reservoir of extractable energy (the "Lattice"), contradicting energy conservation unless identified.

Verdict: split. - "Energy Lattice" theory: Not testable as stated (undefined entity; no equation). - Weak measurable form (an SEC exciter delivers more electrical power to its load than it draws): Never properly tested (no input and output power measured in any document; replicators report no balance; the inventor's own statement is that it is a conversion).

Claim 3: Stacking AV Plugs multiplies "excess energy" (two double it, three triple it)

a. Arithmetic: [calc] 0.5 x 20 uF x 135^2 = 0.18225 J; two in series, 10 uF at 270 V = 0.3645 J; the page prints both a factor 10 too small. The "doubling" is exactly two 20 uF capacitors each at 135 V (2 x 0.18225 J): charging two capacitors stores twice the energy of charging one. The example has no input measurement, so it cannot show an "excess". The 135 V is an assumption ("Assume a Vo of 135 V"), not a measurement. "If one plug can provide 1/3 Joule then three provide 1 Joule" is hypothetical and matches no number on the page (the example is 0.18 J). A voltage doubler or tripler cannot give more energy out than it takes in (the ideal output energy is the input energy minus losses); the scaling n x is additivity, true of any n independent supplies. The "empirical formula" for the limit is announced and absent; the raw HTML contains no formula and none of the 15 images is one (deep read; I accept it). b-c. No test; no measurements. d. If true as an "excess" per plug: same as claim 2.

Verdict: split. - Stored energy of two capacitors is twice one: Established (trivial arithmetic; note the exponent misprint). - Multiplication of "excess energy": Not testable as stated (excess never defined or measured; formula missing).

Claim 4: Energy is transmitted through one 12 ft wire, no return conductor, 24 V in, 48 LEDs lit

a. Internally consistent as a circuit: a single-transistor oscillator, an inductor L3 (143 turns, tap at 37), top hat (two stainless cups), a single wire to an identical L4 with a top hat. The guide states no frequency, no L3/L4 inductance and no top-hat capacitance. Estimates [calc, my assumptions]: Wheeler's formula for L3 (r = 6.25 mm, length 73 mm, N = 143) gives about 40 uH; a total capacitance of about 6 pF (top hat of roughly 2.5-3 pF, treating it as a 25-30 mm sphere in free space, 4 pi eps0 r, plus winding capacitance) gives a self-resonance near 10 MHz, which is the range of the other SEC papers (12.3 and 13.5 MHz). The loop closes by displacement current through the top hats and stray capacitance to the surroundings: at 10-13.6 MHz, 3 pF and 100 V rms carry 19-26 mA [calc], and 48 LEDs at a few mA and 2.7 V each is 0.13-0.4 W [calc]. A supply of 24 V at 50 mA is 1.2 W. So the demonstration needs no unusual physics to close its energy budget; the numbers are an order-of-magnitude illustration and not measurements. The guide admits a hand near C2 detunes the oscillator, that metallic objects near the wire "degrade the operational efficiency", and that the fields cause "physiological effects" and interfere with communication equipment [read: p.8-10], i.e. substantial near fields beyond the wire. b. Experiment: Single-wire resonant transmission with top hats is established in the literature; the Investigator lists patents using the same Avramenko-plug arrangement (Strebkov, RU2340064C1 and family; Bank, US9608441B2; Egely, US20140126679A1) [web] (see sites.md). Those are by others with their own stakes (patentees and a state research institute); I record them as related work. c. Tests: Lidmotor reproduced the visible effect at 10 ft and 25 ft around a wall (YouTube, 2009; descriptions only, no power balance), reported by the Investigator [web]. The cited demo video y9aJohml130 is gone. No efficiency has been measured by anyone on record. d. If true as an ordinary capacitively returned resonant link: nothing. If it implies energy through the wire by a means other than displacement current: new field theory; no document claims a mechanism.

Verdict: split. - Lighting LEDs through a single wire with the return completed through capacitance: Established (accepted theory; qualitative reproduction by others; no efficiency measured). - Any claim of efficiency or of a "lattice" mechanism: Never properly tested (no power figures in the guide; "Measurements for SEC Exciters" is not in the folder).

Claim 5: Calorimetry of the Ainslie heater: heat in water 1580.593 J exceeded electrical input 1198.759 J (CEC 1.319); returned energy to the battery is insignificant

a. The arithmetic is correct [calc]. Q = 4.187 x 25 x (41.25 - 26.15) = 1580.6 J; CEC = 1.3185; mean input 2.00 W; heat into water 2.63 W. The excess is 381.8 J, or 31.9%; equivalently the temperature rise would have to be 11.45 K, not 15.10 K, for CEC to equal 1.0, an over-read of 3.65 K [calc]. Hidden assumptions that decide whether that is a real excess: (i) Uniform water temperature. The cell has no stirrer, the sensor sits in the top ~5 mL, and the coil heats from the outside of the tube; warm water rises. A top-layer reading above the volume-average of a few kelvin is not implausible for 25 mL heated at 2 W, and 3.65 K is the size of error needed. This is a possible explanation I cannot confirm or exclude from the page; it is the first thing a control run would settle. (ii) Mass. The text says 20-25 mL, the screenshot 25 g; with 20 g, CEC would be 1.055 [calc]. The ratio is proportional to the entered mass. (iii) Input integration. The footnote says AC-component energy was integrated at 40 MHz sampling, but Fig 15 shows a 1 Hz read interval of V and I labelled "PSP Charge Supply", with min volts 16.97 against 20.00 and current 0.092-0.108 A. Whether the 1.49 MHz ringing on the supply and the pulsed current enter Sum(V x I x t) as a true time-integrated product, or only as DC readings, is not stated. For a correct integral, ringing that is 90 degrees out of phase between V and I contributes zero mean power; any averaging that rectifies it (for example |V| x |I|) would inflate the input, in the other direction from the artefact needed here. Meter-based input errors on pulsed loads could go either way. (iv) Heat loss and glass reduce the apparent CEC if corrected for, so correcting for them would make the excess larger; any real source of error must then be on the temperature or input side. (v) One run, one screenshot; no repeat, no blank (the same cell on DC at the same mean power), no calibration constant. In the cup enclosure the ratio fell to ~0.83 above 42 C: consistent with growing heat loss, but also what any apparatus whose early readings were biased would show; the page has no time-resolved CEC to tell them apart. Decisive control: run the same cell, wire and sensor on a steady DC supply at 2 W for 600 s; the measured Q/E for DC is the apparatus's own CEC. If DC gives about 1.3, the 1.319 is a calorimeter artefact; if it gives about 1.0 the pulsed-run excess needs another explanation. Returned energy: with the Fig 16 schematic as drawn (D1 across the L-R pair, anode at the MOSFET drain end), the coil current at turn-off circulates L -> R -> D1 -> L, not through the battery (the loop is closed without it). So no energy returns to V1, and the stored 1/2 L I^2 is dissipated in R and D1, both inside the calorimeter. Stiffler's conclusion is correct for his schematic [calc/circuit analysis]. How big is that energy? Fig 11 shows ~3.8% duty at 2.38 kHz; at 0.096 A mean the peak current is ~2.5 A and the loop resistance ~8 ohm [calc, rectangular-pulse assumption]. For an assumed L = 5-50 uH the energy per cycle is 16-160 uJ, i.e. 38-380 mW, 2-19% of the 2 W input [calc, L assumed; not on the page]. So the spike energy is modest, in agreement with the claim "little energy"; "1.49 MHz ringing integrates with interest" is never quantified (no joules, window or current). The Fig 16 circuit has a freewheel diode that [web, medium] sources describing the Ainslie circuit do not clearly show (the Ainslie circuit uses the MOSFET body diode path per one secondary source; https://es.scribd.com/document/50427252/EIT-paper-1); whether the tested circuit is Ainslie's original is not established, because the page does not give her specification values. b. Experiment: energy conservation says Q cannot exceed energy drawn from the supply (plus any stored energy change). A ratio of 1.32 is therefore a flag for an error in T, mass or input, unless an external source exists; Stiffler himself denies energy creation. c. Tested by Stiffler once; reported by a related group: Ainslie et al. (Dec 2009), COP above 4.0 on the same circuit class [web] https://www.scribd.com/document/23455916 (originators of the circuit: interested parties; Stiffler's name may appear in acknowledgments, unconfirmed). No independent calorimetry of Stiffler's cell exists. d. If true: a nonconservation of energy in a battery-switch-resistor circuit, which is an elementary closed system; or an unidentified energy source.

Verdict: split. - CEC 1.319 as a measurement of excess heat: Never properly tested (one run; no DC control, calibration or repeat; the figure may be an apparatus artefact). - Returned energy to the battery is insignificant: Established for the schematic drawn (circuit analysis); whether the schematic matches the original circuit is unverified.

Claim 6: SEC exciters can be measured only with battery-powered, ungrounded instruments behind RF filters; grounding "will detune the AV Plug to where it no longer will interface with the Spatial Lattice"

a. The measurement-paper arithmetic mostly checks: lambda = 300/13.6 = 22.06 m; 1/3 lambda = 7.353 m; lambda/2 pi = 3.511 m; ratio 2.09 ("a factor of two") [calc]. Eq 9 is dimensionally inconsistent as printed: r = 0.62 sqrt(D^3/pi) has units m^1.5 [calc/units]. The standard radiating-near-field boundary is 0.62 sqrt(D^3/lambda), with lambda not pi [web] https://commit-web.commit.tu-berlin.de/en/book/rfi/chapter/ch08/section/s04 and https://www.itecspec.com/3gpp/38.810/s/7.2.1.2 (the 3GPP test text states R > 0.62 sqrt(D^3/lambda)). The "pi" is probably a transcription of the symbol lambda. Also: "0.001 V across 1 ohm = 0.001 mA" is a unit slip (it is 1 mA; the paper's own 0.024 V = 24 mA example is right). Eq 1 (Pload = (Vr/R)^2 R) is just Vr^2/R and presumes a DC/RMS value. Eq 2 (PLED = VLED n Is + DAV Is) is fine for DC-equivalent current but ignores that the current is rectified RF. Detuning is quantifiable with ordinary physics: for a node of total capacitance C0, adding probe capacitance dC shifts f by sqrt(C0/(C0+dC)). With C0 about 6 pF (my estimate for the guide's coil) a 10-15 pF scope probe reduces the frequency to 61% [calc]; a 3 pF stray reduces it to 82% [calc]. So the statement that attaching a grounded instrument detunes a high-impedance resonator is expected on conventional grounds, no "Lattice" needed. A battery-powered meter still has capacitance to its surroundings; floating reduces ground-loop current but does not remove loading, and the page offers no test (e.g. a dummy load with a known power) that shows the claimed method works. The inventor's own method concedes errors: "meaningless relative readings" from coupled field (p.16); moving the meter changes reading, then the reading is wrong (p.11); supply-unit meters are disrupted (p.13); instruments loaded the output so "measurements were meaningless" (single-wire p.12). Input measurement via 1 ohm and integration (Eq 3) is sound in principle. No uncertainty budget and no worked example of any measurement is given. b. Experiment: consistent with general RF measurement practice (use of ferrite chokes, short leads, differential probes); no experiment shows that grounding specifically destroys an SEC "interface". c. Not tested as stated: no document describes an A/B test (grounded vs floating) with numbers. d. If true in the strong form (the circuit works only when not earth-referenced because of a Lattice): standard circuit theory would not apply to the device in measurement, which would make it not falsifiable by any grounded measurement. Calorimetry (the thermal-bottle test on the SEC Special Coil page) avoids the electrical probes altogether; its result is not stated on the page.

Verdict: split. - Instruments load and detune high-impedance RF nodes, so measurement needs care: Established (ordinary RF practice; quantified above). - Grounding destroys the "Spatial Lattice interface": Not testable as stated (it excludes standard instruments by construction and no A/B test is reported). Note for the reader: the quick LED/black-plate test (no temperature rise) and the thermal-bottle test (no result stated) are the only calorimetric attempts on the SEC page; the first is the inventor's own negative result and is kept in the record.

Claim 7: Stiffler's reports on testing: Internet copies of ECAT failed because builders changed the design; Naudin's TEP "fell short"; the self-running "acid test" is a diversion by suppressors

a. Mathematically the key issue is E above: an electrical COP of 2.2 with usable electrical output makes self-running a trivial add-on (eta >= 0.456). The "door" analogy does not argue otherwise. The statement that "the acid test is a diversion proposed by those wishing to suppress research" is unsupported by any named person, document or event. b. The assertion that copies failed is itself evidence the circuit has not been reproduced; absence of named attempts means no way to check. Silence is not contradiction: Stiffler's disclosures omit coil dimensions, MOSFET part number and schematic values in text (the schematic is an image), which would make faithful copies hard. c. No tests documented. The Investigator's finding that Stiffler stated later he does not believe in over-unity ("I do not believe in OU"; CEC means conversion, no creation; thread 4067 posts #1 and #4, 2009-03-12 [web]) changes the framing: by his own later account the 1.319 or 2.194 figures are gains in a conversion of "energy forms" in an open system. d. If suppression were real: records of who asked Google to remove Stiffler results (the Lumen notices 75850580 and 95550170 [web], not accessible to the Investigator) would be the place to look; they are an unanalysed anomaly rather than evidence, because the requester is unknown.

Verdict: Not testable as stated (the suppression assertion and the characterisation of unnamed copies cannot be examined). Self-running implication: Internally inconsistent (a stated electrical COP of 2.2 together with a claim that loop-back cannot work, absent an explanation of the output form).

Investigator findings I weigh (kept whatever the verdict)

  • Inventor's own words that CEC is conversion not creation, and that "the math could not support the (conversion) concept": relevant to claims 1, 2, 7; it narrows what the inventor claims (no free energy) while leaving the measured ratios unexplained.
  • Stiffler's 2010 LED-array paper and the Panacea course (88 pp.), with "Patent Pending 61254675" and an assertion that an SEC exciter is a "Negative Resistance": a negative resistance could be a mathematically definite claim (V and I in anti-phase, so the real power is negative), but neither document is in the folder; I did not read them. The replicator's remark that "power calculation is always negative" at certain frequencies is the same issue seen from the other side. Settling it needs measured V, I and phase at the source terminals.
  • Supporting (replicators): Lidmotor's builds (visual); Raui (some effects); Inquorate (SRF of a tower; noise bump only after pulsing); zeos71 (130 V p-p on L1, 300 V p-p on L2); late 35 mA at 18 V (630 mW) figure. None gives output power.
  • Related work: the Strebkov, Bank and Egely patents on single-wire transmission with the Avramenko plug (claim 4); Ainslie et al. 2009 (claim 5; interested party).
  • Historical record: the inventor died 2018-11-12 (obituary notice, Investigator); the record of the work ends in 2018. The Wayback and tuks.nl mirror are blocked, and the Stiffler Scientific site is dead. The absence of the promised SEC paper is explained by the death on the evidence so far; the "empirical formula" on SECExciter.htm may be on a page not saved in the folder.
  • ZPEnergy (2010): SEC "patent pending"; no published patent found; the claimed provisional 61/254,675 has the form of a US provisional number [memory, medium, per the Investigator]. Mathematically, a patent filing would have to state an equation for SEC; its text would be the first place to look for the missing formula.

What would settle the main questions (calculations and measurements)

  1. DC control run on the ExHeatExp cell at equal mean power with stirring or a sensor at the volume centre (claim 5).
  2. A time-integrated V x I at the exciter input with a 1 ohm shunt, and at the load, with phase, plus a thermal calorimeter on the load (claim 2, 6).
  3. For ECAT: scope-integrated input and output power at 4.3 kHz with a resistive dummy load in place of the lamp; the page's schematic image and the scope photos (ecattp001-005) would be needed (claim 1).
  4. The missing formula on SECExciter.htm and the "Measurements for SEC Exciters" data (claims 3, 4).

Summary of verdicts

  1. ECAT gain: Never properly tested (number); Not testable as stated (mechanism).
  2. SEC / CEC: Not testable as stated (Lattice theory); Never properly tested (output above input).
  3. Stacking: Established (stored energy of two capacitors, with exponent misprint); Not testable as stated (multiplication of "excess").
  4. One-wire transmission: Established (capacitive return, qualitative); Never properly tested (efficiency or special mechanism).
  5. Calorimetry: Never properly tested (CEC 1.319); Established for the drawn schematic (returned energy insignificant).
  6. Measurement claim: Established (instrument loading and detuning); Not testable as stated (grounding kills the Lattice interface).
  7. Stiffler's testing reports: Not testable as stated; self-running logic Internally inconsistent.