pd1180-epr (techmannih) errors
5 recurring problems engineers report in pd1180-epr (techmannih), ranked by impact.
#101 by errors trackedECAD
Recurring pd1180-epr (techmannih) errors, by impact
- 1After a move to RP2040 and WAGO terminals, the assembler-facing release still had STM32 programming steps, JST harness instructions, obsolete mounting notes and the old via count. The fitted design is 237 top-side parts and 27 filled and capped via-in-pad spots. The author's own line is the finding: "Integrity hashes alone could not catch these internally consistent but obsolete instructions." They added semantic release checks for stale counts, programming targets, harness steps and mistyped JLC codes. The KiCad board itself didn't change.ECAD · Agent failure · 1 report74
- 2Failing repro PR: a TI TPD4S480 import gave connector-side pins 4/5 the same CC1/CC2 aliases as protected-side pins 12/11, merging four CC nets into two and creating false internal connections, and the KiCad exporter then emitted seven extra vias on top of existing ones, 650 vias where 643 are real.ECAD · Agent failure · 1 report74
- 3Fix PR for #16, renaming the pins after checking the datasheet table to undo the alias merge and the extra vias.ECAD · Agent failure · 1 report74
- 4J1/J10 had eight genuine copper-to-NPTH violations at the 0.20 mm rule already present in Circuit JSON, but native generation masked them by overriding the rule to 0.13 mm; the fix keeps 0.20 mm and trims the pads, taking website converter hole errors from 8 to 0 and KiCad DRC to 0 errors.ECAD · Agent failure · 1 report74
- 5Replacing JST terminals with WAGO clamps meant updating footprints, routing, BOM/CPL, strip lengths and the assembly harness models. Adding raw PD_VBUS sensing pulled firmware in too: PD_READY now needs a fresh conversion, config readback and a valid 48 V/5 A contract. Mounting holes had to stay aligned with the motor maker's STEP file. A small part change ripples across every discipline on the board.ECAD · Agent failure · 1 report66
How it fails
Sourced from public community reports. Costs and delays on individual entries are modeled estimates, not measurements.