Oct 18, 2026 3:41 PM ET: Jig v4e (modifier clearances, deeper inserts, two carrier orientations)
- Folder cad/jig_v4e/. Files:
- jig_v4e.py, jig_v4e.step (base + carrier_facedown + ghost board), jig_v4e_carrier_faceup.step
- STLs: base, carrier_facedown, carrier_faceup
- 3MFs: jig_v4e_base.3mf, jig_v4e_carrier_facedown.3mf, jig_v4e_carrier_faceup.3mf
- make_3mf.py, verify.py, PRINT_NOTES.md
- Modifiers: the 100% modifiers spanned the full height and left visible artifacts on the top and bottom skins. They're now pulled back: start ≥ 1.8 above the bed face, end ≥ 2.1 below the top, inset 3.2 behind the outer walls.
- Base: M6 floors z 1.8–3.9; M4 z 6.0–15.9; post cores z 16.5–27.9.
- Carrier: z 1.8–9.9. Clamp modifiers shrank Ø10 → Ø7.6 to stay behind the wall.
- Heat-set inserts weren't seating flush. All bores are +0.5 mm: M3 7.0 (carrier clamps x4, carrier pulls x3, base post pushes x3), M4 10.5 (base x4). Thinnest remaining wall is 3.3 mm (the clamp floor beyond the cone); nothing went under 1.0.
- Carrier orientations: face-down (v4d style) and face-up (v4c style, teardrops mirrored to point up as printed). the designer picks after comparing.
- 3MF: project_settings.config is byte-identical to jig_v4c_carrier_safe.3mf in all three (nozzle Standard, Supertack, black); the designer sets High Flow.
- Validated:
- STEPs reimport as valid solids; STLs watertight; 33 probe points pass.
- BS CLI --info: manifold for all three. Slices clean.
- Estimates: base 8 h 17 m / 486 g, carrier face-down 3 h 42 m / 217 g, carrier face-up 3 h 41 m / 217 g.
2026-10-18 3:57 PM ET — Heat-set insert fit (Ruthex)
- the designer uses Ruthex RX inserts (has both standard and short). Insert pilot end would not start in printed holes drawn at Ruthex nominal (M3 Ø4.0, M4 Ø5.6); printed holes run ~0.1-0.2mm undersize.
- Pending for next jig version: insert holes +0.2mm (M3 Ø4.2, M4 Ø5.8); depths already exceed L+1mm (standard M3 5.7, M4 8.1). Workaround on current parts: hand-drill to 4.2/5.8.
- Reprint on hold until the designer provides laser alignment (galvo center vs grid, focal height); fold offsets in then.
Oct 18, 2026 7:07 PM ET: OSSO blank v3b STEP model
- Folder cad/blank_v3b/. Files:
- OSSO_blank_v3b.step: FR4 substrate, 140x140x1.6, board frame z 0–1.6, plus ENIG_top and ENIG_bottom pad bodies 0.035 thick.
- make_blank_v3b.py: parses the GKO and the Excellon PTH/NPTH files; no coordinates retyped.
- jig_v4e_with_blank_v3b.step: drop-in check, board at z 30–31.6 on the face-down carrier.
- Also: OSSO_blank_v3b_substrate.stl, OSSO_blank_v3b_preview_iso.png, verify_holes.py, verify_holes_output.txt, HOLE_COUNTS.md.
- Holes match the drill files exactly by diameter and XY: PTH Ø0.3 x71, Ø0.4 x2, Ø1.0 x48, Ø3.2 x4; NPTH Ø2.2 x4, Ø3.0 x1, plus one Ø2.0 G85 slot. Cutouts: window 27x74 R5, wiring slot 12x26 R4. Substrate valid; volume matches the analytic value.
- ENIG: 40 pogo Ø3 per side, 4 corner rings Ø7/Ø3.2 per side, fiducial rings and lands, 32 top lap pads. 132 bodies.
- Finding: the NPTH slot in the ordered drill file is X70Y135 G85 X70Y137 with a Ø2.0 tool. That makes a 2.0 x 4.0 obround (y 134–138), not the 2.0 x 3.0 in FAB_NOTES and base_v3b. The copper clearance in the Gerbers was drawn for 3.0 (centers 135.5/136.5), so a 4.0 route would nick about 0.2 mm of copper pour at each end, under the mask. The Ø2 slot dowel still works; it just gets 2 mm of free travel along y instead of 1. Measure the real boards when they arrive. For future fabs, fix gen_v3b_gerbers.py to centers (70,135.5)-(70,136.5).
