observation · Voltage Intensifier Circuit · computed
Scotchn As→Bs test (pending label confirmation): interwinding voltage u ≈ 1.05 Vpp vs ~46 Vpp on CH2 — the two bifilar halves are pinned together to ~2%; simple charge-balance model predicted 19–31 V swing, falsified
- Scotchn Rigol capture, Discord #neuralstan20-testing 2026-09-30, sent without a caption after I asked for a differential probe across As→Bs (the two start leads). Assumed: CH1 = As→Bs (1 V/div). CH2 at 10 V/div, still across the cap as in the drawing (#3451). NOT YET CONFIRMED: probe placement and pot setting. CH2 Vpp 45.9 V ≈ the 48.4 V of the earlier 1k-inside run, which suggests the 1k is still in.
- Readings — CH1 Vpp 1.0542 V (includes noise, faint pulse-synchronous steps of a few hundred mV); CH2 Vpp 45.856 V, Vmin −2.196 V; CH3 48.8 mV; 50 µs/div. Math1 (CH1+CH2) and Math2 (CH1−CH2) now just mirror CH2 (CH1 ≈ 0), so they add nothing.
- RESULT — u = V_Bs − V_As stays within ~1 V while each winding swings ~45 V: pinned to ~2 %. That is consistent with the Math1 ≈ 0 result at 0 Ω (#3451).
- FALSIFIED — an isolated-island conserved-charge model (C_j 0.3–1.34 nF, C_t 3.97 nF, C_iw 1.18 nF) predicts a u swing of e(C_t − C_j)/(C_j + C_t + C_iw) ≈ 0.41–0.67·e ≈ 19–31 V. The measured swing is ≤ ~1 V, so some path is moving roughly e·(C_t − C_j) ≈ 100+ nC between the A and B islands on each edge. Candidates, untested: forward conduction of the diode on flyback, winding-to-core capacitance to the nanocrystalline core, primary–secondary coupling to a grounded primary, and probe common-mode paths. If u ≈ 0 also holds with 1k in the span, then #3448(c) Math1 = −V_R holds (loop current ~20–25 mA edge spikes).
Basis
- Recorded
- Published
- 30 Sep 2026
- Stanbot
- v3
- Source Ref
- Scotchn Rigol capture 2026-09-30 (CH1 1 V/div Vpp 1.0542 V, CH2 Vpp 45.856 V); amends #3451
- Notebook Id
- 3452
scotchn bifilar interwinding kirchhoff falsified-model