si-eda (srinivasan-fh) errors

4 recurring problems engineers report in si-eda (srinivasan-fh), ranked by impact.

#138 by errors trackedECAD

Recurring si-eda (srinivasan-fh) errors, by impact

  1. 1Asked for relay switch boards with no MCU, the AI board compiler pulled the button up to the board's 12 or 24 V VIN and switched it to GND, even on an optocoupler-isolated board, which joined GND_IN to GND. The fix adds a VCC_IN pin for the controller's 3.3 or 5 V and returns the switch to GND_IN.ECAD · Agent failure · 1 report74
  2. 2Designing a full 230 VAC off-line supply through the MCP tools surfaced nine bugs: placement ignored mains and HV net-class clearances, the router double-counted class extras so mains nets could not route, a 2.5 mm track crossed a THT pad, DIP and header packages ignored pitch above 2.54 mm giving the wrong land pattern, and thermal analysis sent every watt into the board.ECAD · Agent failure · 1 report74
  3. 3Adding a second label silently joined two nets, so sch_wire and pcb_net now refuse a pin already on another named net.ECAD · Agent failure · 1 report66
  4. 4Asked for a DC-to-AC converter with an ESP32 and a 24 V battery, the AI board compiler wrote a correct spec but built a board with 2 open connections and a DRC error. The LM2596 leads pointed at the board edge, and three retries only grew the board to 139 x 80 mm.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.