radial cantilever inward

[b-betoni]

What would you like

the standard examples of cantilevers expanding outwards, away from the main body work like a charm. I tried to print this simple spool guard design for cardboard spools, where the cantilever is actually growing to the center (attached to the outter walls) and the slicer try to print in mid air, starting with a supporting circle and later filling the space with standard zigzag path (totally ignoring the wave overhang alg). screenshots attached from the slices. i can provide the stl file if required but i believe you can check that with simpler toy examples. At any rate im super impressed!!

Anything else

No response

[dennisklappe]

Thanks for the report and the screenshots.

I traced it to the bridgeability check that decides whether a region gets wave treatment. Outward-growing overhangs are anchored on one side only, so they correctly get waves. Your inward case is anchored on all sides, so the check concludes it bridges fine and hands it to the normal bridge pipeline. That is the mid-air circle plus zigzag you're seeing.

Workaround to try now: enable Wave overhangs instead of bridges. That bypasses the check and should apply waves with no other changes.

Could you attach the STL, or better the 3MF with your settings? I have two candidate fixes and can't tell which applies without seeing whether the unsupported area appears gradually or all at once at one height.

[FeNi64a]

Same issue here. The phone case has an inward-facing lip to hold the phone in, which the current version (0.4.0) prints the inside edge in mid-air. No change with or without "Use wave overhangs instead of bridges" and/or "support unfilled..."

Pixel10a-case03.zip

[dennisklappe]

Fixed in #92. Root cause: your lip is anchored on all sides with only a thin supported band around it, and the wave seeder inset its anchors by more than that band's width, so it found no place to plant seeds and silently dropped the region (that's also why instead of bridges made no difference; the region did reach the wave generator, seeding just failed). With the fix your Pixel10a-case03.3mf goes from 0 to 10 wave regions on the lip layer. Will land in the next release.