Repository navigation
Baseplates: at small pitches the socket's rim stands inside the corner pockets, so a magnet can't seat #83
Description
Found while fixing #75 (PR #82), and noted in the #70 work. It is on main (f9873f5), with no joint at all.
What happens. With corner pockets (Mounting base: corner pockets), a magnet pocket near the cell's edge runs under the foot of the socket's rim, and the rim is not cut out of the pocket. The solid stands inside the magnet's circle for the pocket's whole depth, so the magnet can't go in. The plate builds, Checks says nothing and Download stays on, since the magnet limit for corner pockets (inBoss in mountLimits) lets the pocket come within 1 mm of the cell's edge, inside the rim's 2.85 mm foot.
Repro (Baseplates page, default 6 x 2 mm magnets from beneath, no joint):
#pi=34.5&w=138&d=69&mm=custom&ml=0&mr=0&mf=0&mb=0&bw=85&bd=400&cn=none&bm=bosses&mg=1: the page takes the 6 mm magnet (its max there is 6.3).
Measured by reading each piece along vertical lines 0.1 mm apart inside the magnet's circle, every 0.05 mm up its 2 mm height:
| Pitch | Magnet | Solid inside each pocket | Pockets |
|---|---|---|---|
| 34.5 mm | 6 mm | 14.6 mm³ (41% of the columns) | 32 of 32 |
| 36.13 mm | 6 mm | 1.8 mm³ (17% of the columns) | 32 of 32 |
| 37.5 mm | 6 mm | 0.01 mm³ | 32 of 32 |
| 38 mm | 6 mm | none | |
| 40 mm | 8 mm | none | |
| 42 mm | 6 mm | none |
A solid floor is not affected, since its pockets are cut after the rim.
Possible fix. Cut the boss pockets after the rim (or cut the rim out of them), or cap the corner pocket's size so it stays clear of the rim's foot, and say so in Checks. Then add an audit row that measures the pocket's open room at 34.5 mm.
Activity