solar-business-card (devinhorowitz) errors
3 failures reported against solar-business-card (devinhorowitz), sorted by how they fail and ranked by impact.
#217 by errors trackedMechanicalElectronics & PCB
Recurring solar-business-card (devinhorowitz) errors, by impact
- 1An independent tool (SolidCraft) measured the committed enclosure solids and found six problems: every STEP and STL is the mirror image of the part, all four supercaps sit 0.17 to 0.375 mm inside their mount bosses, both braces are 0.20 mm wider than their cavities, and one shell has 804 sliver faces that SolidCraft refuses to open. Every gate the repo already had passed these files, because it compared fit_rules against itself, and the fix gates the emitted STL instead. The author says nothing in enclosure/ should go to a shop or printer from main, and the mirror question is still open.Mechanical · Wrong verdict · 1 report74
- 2DigiKey sold through 490 units of C11's Murata cap within hours of a part swap; Mouser had 17,858. Moving the supplier part number tripped the repo's own consistency check, because Mouser abbreviates long Murata numbers. Rather than loosen the matcher, which had caught two real wrong-part failures before, the author added a ledger of live-verified MPN/P/N pairs and self-tests that keep the old failures out of it.Electronics & PCB · Wrong verdict · 1 report66
- 3Same project as finding 1: the PR swapped four hand-typed enclosure notes for computed ones, and two turned out to describe geometry that had changed underneath them without anyone deciding it should. The edge ledger had an entry for C7, a part that is not on the board, and once the ledger audited itself, all seven of its entries were dead. The brace README said the rails back the four solder tabs, which was wrong.Mechanical · Silent failure · 1 report66
How it fails
Sourced from public community reports. Costs and delays on individual entries are modeled estimates, not measurements.