observation · Misc · computed
Confirmed: Scotchn's latching bench ladder had NO chopper between SCRs and step-up; his Falstad burst gate is a MOSFET (not NPN as labelled / not Stan's bipolar Q9)
Scotchn, Discord #neuralstan20-testing 2026-11-04, answering whether the bench build had the Q9-style chopper: "Oh, no, it didn't. That's a mosfet!" So the bench SCRs fed the 200:1000 step-up primary (about 0.83 H) directly: no forced current zero, the bridge acted as a freewheel path, and continuous conduction is the expected result (#3518). The latching is explained by the missing chopper, not by the ladder concept. In the working Falstad model (#3519) the burst gate is a MOSFET, whereas FIG. 20YI specifies a bipolar Q9 (2N6676 or 2N6496). Amends #3517-#3519. Still untested on the bench: chopper present plus real SCR turn-off time against the off-time (about 36 µs at 14 kHz, 50 %), and saturation of the bridge-fed attenuation core.
Basis
- Recorded
- Published
- 4 Nov 2026
- Stanbot
- v3
- Source Ref
- Scotchn Discord 2026-11-04; notebook #3517, #3518, #3519
- Notebook Id
- 3520
- Device Slug Unmapped
- hgg
20YI SCR chopper Q9 scotchn