I've encountered this issue on one of my printers, even though it worked perfectly fine out of the box. Issue #40 may be related.
Either due to deformation or conflicting clearances between the pcb and the arm, the following pieces jammed and were stuck open (as "filament present"). This caused the filament sensor to not detect filament run out.
This was relatively easily rectified by just sanding the outer surface of the arm. However, why the issue appeared in the first place is a mystery. The printers are kept in a semi-enclosure, though I doubt ambient temperatures of ~30°C would cause some deformation due to stress being relieved. Maybe? I have tried a test piece printed in PLA and although I couldn't really see a dimensional difference, that one worked fine.
Making it slightly more pointy on the outside and straight on the inside (it interfaces with a flat surface anyway) may avoid this issue. Depending on the position of the sensor beam itself, top/bottom points of the wedge could also be chamfered slightly to avoid elephant foot related interference.
I've encountered this issue on one of my printers, even though it worked perfectly fine out of the box. Issue #40 may be related.
Either due to deformation or conflicting clearances between the pcb and the arm, the following pieces jammed and were stuck open (as "filament present"). This caused the filament sensor to not detect filament run out.
This was relatively easily rectified by just sanding the outer surface of the arm. However, why the issue appeared in the first place is a mystery. The printers are kept in a semi-enclosure, though I doubt ambient temperatures of ~30°C would cause some deformation due to stress being relieved. Maybe? I have tried a test piece printed in PLA and although I couldn't really see a dimensional difference, that one worked fine.
Making it slightly more pointy on the outside and straight on the inside (it interfaces with a flat surface anyway) may avoid this issue. Depending on the position of the sensor beam itself, top/bottom points of the wedge could also be chamfered slightly to avoid elephant foot related interference.