Stiffler: Quality Engineering (Step 4b)

QE role, 2026-10-08. Inputs read: _text/manifest.json, every digest (9 files), every deep read (7 files), facts.json, claims.md, sites.md, all research/ notes (7), physicist.md, mathematician.md, contrarian.md, arbiter.md, primer.md, example.py (run). Spot checks were made against the extracted text, the raw HTML, the saved-page image folders and PDF page images (scratch renders in /tmp/stiffqe). For cross-checks I also read the claims.md and arbiter.md of the sister project review/Ainslie-Rosemary-Ann, because claim 5 is Stiffler's test of that circuit. Tags: [read] = source text or image, with file and page or figure; [web] = URL; [calc] = my arithmetic.

Verdict: MINOR

The review is complete in coverage and careful in what it says, and the headline verdicts are sound. The defects can all be fixed by updating facts, arbiter and primer: (1) figures with data were left unviewed for two pages whose images are in the source folder (ECAT, SEC Special Coil). As a result the "source gap" note is partly wrong: the SEC Special Coil charts hold no gain data at all, and the ECAT scope photos do hold voltage readings and a schematic. (2) Several pertinent items were missed, chiefly the link to the sister Ainslie review, which supplies the "Ainslie specification" the Stiffler files say is missing. (3) A few corrections listed by the arbiter have not yet been applied to claims.md and facts.json, and there are small accuracy slips. None of these is likely to move a central verdict. ECAT still has no current or power measurement, and the calorimetry still has one run and no blank. So Steps 2-4 need not be redone.

1. Coverage

Every manifest document was digested. Step 0 did not extract the five saved-page "_files" folders. facts.json has one entry for them, and that is the right treatment. But those folders are not browser junk: they hold the scope photos, schematics and charts that carry the pages' data (see 2).

2. Shortcuts

• ECAT figures not viewed (defect; claim 1 is central). The source folder holds ecatcir000.gif (the schematic), ecattp001-005.gif (test points TP-1 to TP-5) and ecat000-012.gif. I viewed the schematic and four scope photos for this check [read: images]:
• The schematic shows two sources: V1, a pulse generator driving the MOSFET gate through R2/R1, and V2, the DC supply. It also shows a source resistor R3 (test point TP-5), a "Polyphase transformer" L1-L4, a power coil L5 and a load coil L6 into R4 ("Load"), and a D2/C4 branch back to V2.
• Readings: TP-1 shows 106 V at 50 V/div, 100 us/div, about 4 pulses per 0.93 ms, i.e. about 4.3 kHz [calc]. TP-2 shows 172 V. TP-3 shows 94.5 V. TP-5, across R3, shows 0.31 V p-p at 500 mV/div. The drain photo (ecat010) shows 205 V at 500 us/div.
• What the page does not give: no current trace, and no value for R3. Whether V1's power is part of the 8.25 W "input" is not stated.
• So the review's claim that "scope numbers are unknown" is out of date. The figures do not supply a power balance, but they bear on the input basis (a second source; a current-sense resistor whose value is not given) and should be deep-read.
• SEC Special Coil figures not viewed (defect; this changes the source-gap note). I viewed Fig 10, 11, 13, 15 and 16 [read: images coilset08, coilset09, coilset10, therm001, temp001]:
• Fig 11 ("Amplification Points") and Fig 13 ("Primary & Secondary Power Peaks") are 3-D cone charts whose only axis is "Hertz". Each cone's height is the frequency listed in the text (about 15.2, 15.2 and 18.6 MHz; plus 12.0 and 12.0 MHz in Fig 13). They hold no amplitude, gain or power value.
• Fig 10 ("Coil(s) Response") plots the fsl/fsr/fsh frequencies only.
• Fig 16 is a photograph of the thermal bottle wrapped in cloth, with a temperature-probe display, an Extech meter and a timer. It holds no data.
• Fig 15 is a photograph of about 17 lit white LEDs on a breadboard under the black plate.
• Correction: "amplification" on this page is shown only by frequencies, in the text and in the charts alike. There is no hidden gain figure. The thermal-bottle result is absent from the images too.
• Digest-only coverage elsewhere (coils.htm, the kit manual, Measurement Methods pp.1-10) is acceptable. Those documents carry no central claim.
• Investigator time cap: the run stopped at 9.3 minutes, under the 9.5-minute floor, and the sites not reached are logged as "stopped early". That is a process slip, but it is honestly recorded.

3. Missed material

1. The sister Ainslie review supplies the "Ainslie specification" (cross-project; claim 5). review/Ainslie-Rosemary-Ann/claims.md says the 2009 Ainslie paper "still quotes 2.4 kHz and 3.7% duty", and that with the gate resistor tuned the circuit self-oscillates at 143-200 kHz (2002) or over 500 kHz, "and this is what makes the gain appear" [read: ../Ainslie-Rosemary-Ann/claims.md item 3]. Stiffler's Fig 11 shows 2.38 kHz at 3.80% duty [read: deep/ExHeatExp]. So he set his 555 to the 2009 Ainslie-group specification. That dates ExHeatExp to about 2009, which matches the Scribd snippet that names "Dr Stiffler" in the Ainslie et al. Dec 2009 paper. His "MOSFET oscillations ... orders of magnitude higher than described" is the same self-oscillation regime the Ainslie review identifies as the source of her apparent gain. The arbiter and primer say "Ainslie's own documentation" is missing and that "no independent critique or replication [was] found". The sister project has the Quantum 2002 article and patent, deep-read, plus a forum analyst's (Poynt99) artefact report and advocacy reports of 2009 replications (Lettenmaier) [read: ../Ainslie-Rosemary-Ann/arbiter.md]. The sister review also says the circuit has a flyback diode. That contradicts the mathematician's [web, medium] note that the Ainslie circuit uses the MOSFET body-diode path.
2. ECAT transformer described as "cold". "Under normal operation the MOSFET and its heat sink is cool (not cold like the transformer) but cool" [read: ECAT page, extracted text line ~175]. A 60 Hz transformer said to "waste some 6 watts" is described by the author as cold. Either the 6 W is not a measured heat loss, or the author is reporting a cold-running transformer, which is itself an unusual observation in free-energy literature. No review noticed this, and it bears directly on the 12 + 6 = 18 W accounting.
3. The unlabelled "Measured (MHz)" table on SEC Special Coil is the single-wire paper's Fig 18 data. The SEC table reads 3.9866 / 10.4230 / 12.3340 MHz [read: SEC Special Coil text lines 73-75]. Fig 18 "Energy Peaks" plots 3.9066 (as the low-resolution label reads; the point sits on the 4 MHz gridline) / 10.42302 / 12.334 MHz [read: single-wire PDF p.14, embedded image]. The SEC deep read says the table's identity is unknown. It is very probably the same three "energy peaks", with one transcription slip (3.9066 versus 3.9866). facts.json should link them.
4. NILS "Self Charging" / "Eternal Light" claims (Investigator, Scribd). The Panacea course says "Self Charging" NILS (Near Infinity Light System) units were never released [web: https://www.scribd.com/doc/35609446]. A self-charging light sits in tension with Stiffler's 2011 "Urban Myth" post, which says looping output to input "would require that we were manufacturing energy" [web: thread 4067 p.142, quote verified in research/pages/ef_4067_p142.txt]. The arbiter's ruling 5 (self-running logic) and the primer should keep both statements side by side. Neither file mentions NILS.
5. Forum estimate of a possible ECAT gain (supporting lead, dropped). Kokomoj0 (thread 9344 post #3) estimated from an ECAT video "an addictional 4 watts roughly", and said proper measurement needs DC conversion and a load [web; quote verified in research/pages/ef_9344.txt]. Only the mathematician kept it. The arbiter and primer list no third-party estimate for claim 1.
6. Inquorate's observation that the white-noise bump at the self-resonant frequency appeared only after the coil had been pulsed [web: thread 4067 p.145, Investigator]. This is a small anomaly report. The arbiter and primer give only his name.
7. Kit tuning rule: the SEC 15-3 is tuned for "maximum neon glow and minimum supply current" [read: kit manual p.6]. That is the textbook signature of a parallel-resonant tank at resonance (impedance at its maximum, so supply current at its minimum). It is a teaching point the primer could use. Only facts.json mentions it.
8. Stiffler's "Spatial Gate" challenge (thread 4067 post #1): over 300 V from 18 V input "without a step-up transformer or multiplier", by "pinging the Energy Lattice" [web, Investigator]. This is the inventor's own short statement of the SEC principle. It is not in the arbiter or primer.
9. Smaller items: the dominant 13.6 MHz label in Measurement Methods Fig 19 is not the tallest peak on the screen [read: deep read, p.17]; the contrarian found a 2019 IEEE paper on measuring single-wire efficiency (Olkhovskiy, Muellerova, Martinek) [web: https://dspace.zcu.cz/items/ecbbaecc-629a-4d7d-8f19-0a21b92e631e/full], which belongs under "Further reading" for claim 4.

4. Accuracy spot checks

Accuracy slips found: - arbiter.md, Absence-of-evidence A, and primer section 7 say "white papers and videos continued to 2018". The latest white paper on record is the Series 18 documentation, Rel 2.8, September 2011 (studylib, not read). What continues to 2018 are videos and a circuit shared through a replicator (Sun0008, 2018-04-30) [web, sites.md]. Restate as "white papers to at least 2011; videos and circuits to 2018". The "eight years" without an energy-balance paper still stands. - arbiter.md D says "the same words [Patent Pending] in the 2012 course would need a later filing". The "Patent Pending: 61254675" text was seen in Scribd document 35609446. The 2012 Panacea PDF at quanthomme.free.fr was not read (sites.md). Nothing on record shows the 2012 document repeats it. Remove or qualify the sentence. - mathematician.md, claim 5: the [web, medium] note that sources describing the Ainslie circuit do not clearly show a freewheel diode conflicts with the sister review ("air-core inductive 'resistor' with flyback diode") [read: ../Ainslie-Rosemary-Ann/claims.md item 1]. The arbiter should note the conflict. - The physicist's section 0 web check on Ainslie "failed to retrieve primary sources". The sister project holds them. The arbiter should point to it rather than leave the item [memory]/failed.

5. Retention check

• The arbiter and primer keep the pertinent negatives: the black-plate null result, the cup runs at 0.83, the inventor's "math could not support" admission and "Is the Loop really doing anything?". They also keep the pertinent positives: the replicators, the patent literature, the Ainslie preprint, the inventor's disclaimers of over-unity, and his own proposed tests. No item was found dropped because a reviewer judged it negatively. Single-source items (the obituary identity, the Lumen notices, the provisional number) are described, not labelled away.
• Minor retention defects: supporting or anomalous items present in sites.md or a review but absent from both the arbiter and the primer. These are Kokomoj0's ~4 W estimate (item 5 above), Inquorate's "only after pulsing" observation (item 6), the NILS/self-charging claims (item 4), and the contrarian's single-wire efficiency paper (item 9). They seem to have been lost to condensation, not to a negative judgement. Restore them briefly.
• Interested parties are marked consistently (Stiffler Scientific, Panacea-BOCAF as promoter, the Ainslie group as originators, patentees).

6. Process

• Source tags: present and honest throughout. [calc] is used as an extra tag, which is reasonable and is explained in each file. The deep reads mark inference separately from what was read.
• Web checks: physicist 2 of 6, contrarian 2 of 6, arbiter 3 of 8. The medium-confidence facts a verdict touches were checked or explicitly set aside: the provisional date (checked through a neighbouring number), the Ainslie preprint (checked), Eq 9 (checked). The glass specific heat and the "lower still at VLF" items are medium-confidence, but each is used only in a sensitivity or a supporting argument. That is acceptable.
• Verdict scale: applied with the exact names; split verdicts are explained. The arbiter settled the five disagreements with reasons and stated a settling test for each.
• Primer: follows STEP 4 in order (1-9 plus answers); the worked example is real and reproduced. Reader-profile slips:
• acronyms used without expansion: LED (light-emitting diode), RF (radio frequency), AC/DC, NIST (National Institute of Standards and Technology) and SEC. The section 1 paragraph says "SEC exciters" without "(SEC = Spatial Energy Coherence)".
• the source-gap paragraph repeats the now-incorrect "SEC Special Coil Figs 11, 13 and 16 ... numbers unknown".
• example.py: runs cleanly with the project venv (exit 0, 121 lines, ends "All steps completed."). Its numbers match the primer.
• Pending corrections not yet applied (orchestrator):
• claims.md item 1 still says "arithmetic does not close".
• claims.md item 5 states "returned energy ... is insignificant" as fact.
• claims.md item 6 still says "excludes standard grounded instruments by construction".
• facts.json ECAT summary still carries the "logic problem" reading. These are all listed in arbiter "Corrections to earlier documents". The 1990/1985 dates and the mass/gradient sensitivity are not yet in facts.json either.

7. Corrections to earlier documents

1. claims.md "Source gap", facts.json (_files entry and SEC Special Coil entry), arbiter.md "Source gap note" and primer.md "Source gap": SEC Special Coil Figs 11 and 13 were described as unread "amplification" charts with unknown numbers. In fact they plot only the frequencies already in the text (axis "Hertz"; no amplitude or gain). Fig 16 is a setup photograph with no data. Fig 15 is a photograph of lit LEDs under the black plate. Evidence: images coilset09.gif, coilset10.gif, temp001.gif, therm001.gif in "SEC Special Coil_files", viewed 2026-10-08. Consequence: "amplification" on this page has no measured magnitude anywhere in the document.
2. Same files: "ECAT scope numbers unknown". The schematic and test-point photos give voltages (TP-1 106 V, TP-2 172 V, TP-3 94.5 V, TP-5 0.31 V p-p across R3, drain 205 V) and show two sources (pulse generator V1 and supply V2). They give no current and no R3 value. Evidence: ecatcir000.gif, ecattp001/002/003/005.gif, ecat010.gif. The deep read below should replace this note.
3. deep/SEC-Special-Coil: "Measured (MHz)" table identity unknown. It matches the single-wire paper's Fig 18 "Energy Peaks" (3.9066 or 3.9866 / 10.42302 / 12.334 MHz).
4. arbiter.md claim 5 and primer section 6 row 5a: "No independent critique or replication found". Add the sister review's material (Poynt99's artefact report; the 2009 replication claims by an advocacy source), each with its stake and its low reliance as recorded there. Add that Stiffler's 2.38 kHz / 3.8% setting matches the Ainslie group's 2009 specification of 2.4 kHz / 3.7% [read: ../Ainslie-Rosemary-Ann/claims.md].
5. arbiter.md A and primer section 7: "white papers ... continued to 2018" should read "white papers to at least 2011; videos and circuits to 2018".
6. arbiter.md D: remove or qualify "the same words in the 2012 course"; the 2012 PDF was not read.

8. Remediation actions

Deep reads, most important first: 1. ECAT figures (Opus): map the circuit and read every scope value, to settle whether the page's input figure is complete. 2. SEC Special Coil figures (Sonnet): confirm what Figs 1-16 and RFF01-03 show, and reconcile the frequency table with single-wire Fig 18. 3. ExHeatExp against the sister Ainslie review (Sonnet): compare Stiffler's test with the Ainslie group's circuit and specification.

The exact pages and questions are in the deep-read list returned with this file.

File changes after the deep reads: - facts.json: - ECAT: add the schematic and test-point data; the "transformer cold" quote; the restated accounting (from the arbiter); and the 1990/1985 dates. - SEC Special Coil: Figs 10/11/13 hold frequencies only; Fig 16 is a photograph; the "Measured (MHz)" table is linked to single-wire Fig 18. - ExHeatExp: link to the Ainslie 2009 specification and date the page c.2009; add the mass-proportionality and unstirred-sensor sensitivity. - _files entry: say which images were viewed. - claims.md: - apply the arbiter's pending corrections to items 1, 5 and 6; - revise the Source gap (SEC Special Coil figures are not a gap; the ECAT figures are partly read); - add to item 7 the NILS "Self Charging" claim beside the "Urban Myth" post. - arbiter.md: add a "QE revision" section. It should: - cover the ECAT figure results: if V1's power or the R3 sense data change the input basis, restate the claim 1 discussion (the verdict is expected to stay Never properly tested); - add the "transformer cold" tension to ruling 1; - in claim 5, use the sister-review cross-reference (specification match, self-oscillation regime, flyback diode present, Poynt99 and the 2009 replication claims, with their stakes and reliance) and replace "Ainslie's own documentation missing" with a pointer to review/Ainslie-Rosemary-Ann; - in claim 2, state that "amplification" has no measured magnitude in the document or its figures; - fix the "white papers to 2018" and "2012 course" slips; - restore Kokomoj0's estimate (claim 1), Inquorate's after-pulsing observation (claim 2) and NILS/self-charging (claim 7). No verdict change is expected; record any that results. - primer.md: - correct the Source gap paragraph; - expand LED, RF, AC/DC, NIST and SEC on first use; - add the kit's "maximum glow at minimum supply current" as a short resonance illustration in section 3.2; - add the "transformer cold" sentence to the ECAT accounting paragraph in section 4; - in the evidence table row 5a and section 7, cross-reference the Ainslie primer and review for the specification and the critique record; - fix "white papers continued to 2018"; - add the Olkhovskiy et al. single-wire efficiency paper and the NILS claim (open questions) where they fit. - Investigator, next run (sites.md already lists most of these): - retry the Wayback Machine (stifflerscientific.com, drstiffler.com, 67.76.235.52/Ainslie/ and the SECExciter page, for the "empirical formula" and "Measurements for SEC Exciters"); - read the full Darst obituary; - search USPTO Patent Center for 61/254,675; - download the 2012 Panacea PDF and the 2010 LED-array paper into research/files/; - try the Heretical Builders thread; - sample the 2017-2018 "Stiffler Loop" videos, especially "Is the Loop really doing anything?".

Recheck

Recheck date 2026-10-08. Read: claims.md, facts.json, arbiter.md (including its QE revision), primer.md, and the three deep reads' appended sections. Checked by search of the current files against section 7 (corrections) and section 8 (remediation).

No verdict changed after the remediation. The open items (12, 13) are logged follow-ups and not defects in the written review.

Final: PASS