sport-performance (mgiachetti) errors

4 recurring problems engineers report in sport-performance (mgiachetti), ranked by impact.

#139 by errors trackedECAD

Recurring sport-performance (mgiachetti) errors, by impact

  1. 1A KiCad schematic generated from a declarative netlist is then re-read from geometry to prove it matches, with the check now a CI step. A side detail is that the pin electrical types of all 10 symbols from easyeda2kicad came all unspecified and had to be filled in by hand.ECAD · Agent failure · 1 report66
  2. 2EasyEDA footprints checked pad by pad against datasheets were mostly off: the Quectel LC29H had 1.8 mm pads instead of 2.5, the Qorvo DWM3000 bottom row was 0.2 mm off with no antenna keep-out, and TI TPS22916C balls were 0.18 mm against TI's 0.23. Everything redrawn from datasheets is still marked not verified.ECAD · Agent failure · 1 report66
  3. 3The GNSS patch keep-out on L4 prohibits pour and vias but lets tracks through, so the patch power leaves through L4 from the 2.2 mm pad. A KiCad keep-out cannot filter by net.ECAD · Agent failure · 1 report66
  4. 4To quote a v0 board the author wrote bom_pcbway.py, which builds PCBWay and JLC BOMs, a DNP list, extras and per-part cost at 5, 10 and 100 boards from the netlist plus a parts CSV. Stock and lifecycle came from LCSC/JLCPCB APIs and DigiKey pages, TPS7A2032 has no LCSC stock, and the SOT-23 load-switch alternative needs a different footprint.ECAD · Agent failure · 1 report66

How it fails

  1. Agent failure4

Sourced from public community reports. Costs and delays on individual entries are modeled estimates, not measurements.