Baseplates: a corner-boss plate at 32.52 mm with 4.3 mm magnets leaves 3 open edges where two bosses meet on the back edge #94

Requested by Oliver · project thread

What happens

A found-by-sweep case, not reported by a user. One 2 x 2 piece with the Corner bosses base builds with 3 open edges (edges used an odd number of times), at the bed of the boss where two cells meet on the back edge. The piece should be watertight, and the page's manifold check should have refused Download or reported it.

The same 3 edges are open on main 1e1c48a and on the #86 fix (PR #90), which does not touch this boss, so this is a separate defect.

Settings

• Pitch 32.52 mm, drawer 66.462 x 66.698 mm, margins left 0.802, right 0.62, front 1.658, back 0 mm
• Base: Corner bosses; no connector; magnets on, 4.3 mm wide, 1 mm deep (the field's cap at this size)
• Tolerance tight; top cutoff 0.7 mm; outer corner radius 3 mm; bed 400 x 400 mm so the plate is one piece (2 x 2 cells)

What I know

• Of the 40 bad edges on the piece, 37 are four-use edges on the known corner-pocket lines (named in the plate audit); the 3 open edges are the new part.
• It looks like the same family as Baseplates: corner pockets with screws leave 3 open edges a cell at 34.32, 34.73 and 38.95 mm #91 and Baseplates: a screw bore of nearly the magnet pocket's width leaves 6 to 12 open edges a cell at 42 mm #92 (a cap fan leaving slivers a few microns across that the weld joins on one side only), but here no screw bore is involved and no clip is cut, so it may be a different cause.
• It comes from a sweep of 7,649 designs; this was the only design where the open edges were neither a rounded-corner boss nor a joint.

Suggested check

Add the settings above as a test/plate-audit.js row (it should fail with 3 bad edges today), then find which cut leaves the faces unpaired.