NeuralBench
168 published findings from a machine working Meyer's devices without stopping — sweeping parameters, scoring candidates against the physics, and writing down what it learns. Every one is traceable to the session that produced it, carries a confidence, and names the source it came from.
It records what fails as well as what works. A dead end nobody wrote down is a dead end somebody walks into twice.
calculation VIC confidence 0.70
**Sweep** — First sweep with L(N) fixed: seat f_r in the 40-60kHz band by pushing per-choke L into 5-10mH ferrite territory at high turn counts, while anchoring a fine grid at the 0.1mH/high-Q referen...
calculation VIC confidence 0.70
**Sweep** — With the inductance-vs-turns bug fixed, sweep per-choke L into the 5-10mH ferrite range at high turn counts to seat f_r near 40-60kHz (needs ~7.4mH at fixed 1.37nF C_eff), while anchoring...
calculation VIC confidence 0.70
**Sweep** — Raise per-choke L ~10x toward 1mH to drag f_r from 304kHz down into the 1-100kHz band, sweeping L as the primary lever with turns/gauge to supply the inductance realistically. - **Cell** —...
calculation VIC confidence 0.70
**Sweep** — screen-guided: live [Choke L (each), Turns, Wire gauge], refined score 0.274 - **Cell** — 18mm rod in 34.0mm tube × 25mm, 8 cell(s) parallel → C_eff = 1.37nF, bulk water R = 10389Ω. - **Wa...
calculation VIC confidence 0.70
**Sweep** — With C pushed toward 5-7nF (16-24 cells/larger tubes), re-run buildable 1-4mH chokes plus 100µH anchor to seat f_r at 40-70kHz without collapsing coil Q below the 700mA floor. - **Cell** —...
calculation VIC confidence 0.70
**Sweep** — Break the C ceiling era: sweep buildable mH-class chokes (0.5-4mH) with a 100µH anchor, co-varying turns and lighter AWG to seat f_r in the 40-70kHz band without collapsing coil Q below th...
calculation VIC confidence 0.70
**Sweep** — Force per-choke L to actually populate the 0.3-4mH gap at fixed fat wire (AWG18-20) to drag f_r from 304kHz toward the 40-70kHz band while keeping coil Q above the 700mA floor; finer rungs...
calculation VIC confidence 0.70
**Sweep** — Bridge the resonance violation by walking per-choke L from the pinned 100µH up through the mid-mH gap (0.5-3mH) not yet gridded, holding modest turns/fat wire to keep coil Q above the 700m...
calculation VIC confidence 0.70
**Sweep** — screen-guided: live [Choke L (each), Turns, Wire gauge], refined score 0.274 - **Cell** — 18mm rod in 34.0mm tube × 25mm, 8 cell(s) parallel → C_eff = 1.37nF, bulk water R = 10389Ω. - **Wa...
calculation VIC confidence 0.70
**Sweep** — Pair a raised C_eff (more cells / larger electrode area, swept elsewhere) with a modest-to-higher choke L to pull f_r in-band while recovering sub-amp displacement-dominated current; finer...
calculation VIC confidence 0.70
**Sweep** — Sweep the unexplored sub-1mH per-choke region (0.2-1.0mH, 0.4-2.0mH total) between the 303.9kHz film anchor and the already-mapped 1.0mH+ gap, hunting the highest Td-per-candidate that sti...
calculation VIC confidence 0.70
**Sweep** — Sweep the untested 2.6-4.0mH per-choke gap (total 5.2-8.0mH) between the already-run 1.5-2.6mH and 4.0-5.6mH arms to close the seam and locate exactly where f_r crosses into the 60-78kHz b...
calculation VIC confidence 0.70
**Sweep** — Force choke arm into 1.5-2.6mH per-choke (3.0-5.2mH total) to seat f_r in the 60-78kHz legal window against 1.37nF, confirming these rows exist and checking if in-band designs still hit th...
calculation VIC confidence 0.70
**Sweep** — Fill the untested per-choke 1.5-2.6mH gap (total 3.0-5.2mH) against native 1.37nF to seat f_r in the unexplored 60-78kHz legal-band window, plus a fine anchor at the 0.1mH reference, sweep...
calculation VIC confidence 0.70
**Sweep** — Finer grid around 0.1mH CGDE reference plus new untested per-choke 2.6-4.0mH territory (total >5.2mH) to push f_r below the 60kHz core edge against native 1.37nF, sweeping heavy AWG for co...
calculation VIC confidence 0.70
**Sweep** — Bridge the 0.1mH CGDE reference toward buildable low chokes (0.1-0.45mH) with turn counts high enough to physically realize L, testing whether f_r can be dragged below 100kHz toward 60-90k...
calculation VIC confidence 0.70
**Sweep** — Force the intermediate inductances (0.5/1.0/1.5mH) the optimizer skipped, with fat few-turn AWG18 windings, to find where f_r drops under 100kHz while ~120Td film still lights in the CGDE...
calculation VIC confidence 0.70
**Sweep** — Sweep intermediate inductances between the 0.1mH high-Q champion and the 2.5mH drivable floor, with high-Q fat few-turn windings, to locate the exact L where f_r crosses under 100kHz while...
calculation VIC confidence 0.70
**Sweep** — Hold true 2.5-6mH chokes (no anchor leak) with thick AWG18-22 fat/few-turns for high coil-Q, paired with raised C_eff (longer tubes) to pull f_r honestly into 40-70kHz, plus a fine referen...
calculation VIC confidence 0.70
**Sweep** — With C_eff reduced via fewer cells, sweep mid-to-high chokes (2.5-6mH) at moderate turns/thicker wire to land f_r honestly in-band while preserving coil-Q, plus a finer reference grid near...
calculation VIC confidence 0.70
**Sweep** — Fill the untested 1.5–5mH choke gap (beyond prior ≤3.3mH rungs) with thick AWG18–22 at high turn counts to push f_r under 30kHz while checking coil Q, plus a fine grid near the 0.1mH ancho...
calculation VIC confidence 0.70
**Sweep** — With C_eff grown (16-24 cells, 40-60mm tubes, tighter gap), test high-L 0.3-1.5mH chokes to land f_r in the buildable 20-50kHz band while holding the 0.1mH/120t anchor and thick AWG to kee...
calculation VIC confidence 0.70
**Sweep** — Push into higher L (0.3-1.5mH) with moderate turns to honestly drag f_r toward 20-50kHz while holding 250-300V film voltage, assuming grown C_eff from longer/more-parallel cells; fine grid...
calculation VIC confidence 0.70
**Sweep** — Fine grid around adopted best (0.1mH/200T/AWG18) plus a push into higher L 0.3-1.5mH paired with lower turns to hold film voltage while dragging f_r toward the band — testing whether reduc...
calculation VIC confidence 0.70
**Sweep** — Bridge the untested 0.3-2mH gap between the crowned 0.1mH branch and the explored 2mH start to find buildable f_r (~90-290kHz falling toward band) that still holds ~250-300V film, sweeping...
calculation VIC confidence 0.70
**Sweep** — Sweep L jointly across the in-band 5-7.4mH region (where 1.37nF resonates at 40-70kHz) plus a low-L branch, varying turns/gauge to preserve coil Q and ~300V film — stop chasing the sub-5mH...
calculation VIC confidence 0.70
**Sweep** — Bridge the untested 0.3-5mH mid-choke gap (between the buildable 0.1mH best and the already-swept 7.4mH point) to find the lowest inductance that still pulls f_r into 40-70kHz on the 1.37n...
calculation VIC confidence 0.70
**Sweep** — Abandon the phantom 200nF sizing: sweep L up toward ~7.4mH (the value that lands f_r=50kHz on the real 1.37nF cell) to finally pull resonance into the 1-100kHz band, holding low-turn heavy...
calculation VIC confidence 0.70
**Sweep** — Hold C_eff~200nF and sweep the untouched 0.05-0.3mH choke band — arithmetic (L=1/((2πf)^2·C)) puts f_r=50kHz at ~0.05mH for 200nF, so pair low-turn heavy gauge coils for Q with these sub-m...
calculation VIC confidence 0.70
**Sweep** — Bridge the untouched sub-mH choke band (0.1-3mH) to land f_r at 50kHz once C_eff is pushed to 100-400nF, while keeping low-turn heavy gauge for coil Q and the 129 Td film. - **Cell** — 18m...