2026.2_PI1_Grupo1_Diogo (fcte-pi1) errors

3 recurring problems engineers report in 2026.2_PI1_Grupo1_Diogo (fcte-pi1), ranked by impact.

#145 by errors trackedFirmwareECAD

Recurring 2026.2_PI1_Grupo1_Diogo (fcte-pi1) errors, by impact

  1. 1Firmware still used a provisional 32 mm wheel and 280 counts per turn after hardware measured 34 mm and 360; odometry came out about 21% short and cell counts and turns were wrong. Open question left in the PR: the hardware doc's 360 may count both quadrature channels while firmware counts one, which would make it 180.Firmware · Agent failure · 1 report74
  2. 2On the built robot, the motor wires were soldered swapped relative to schematic 4.3, and the team decided not to resolder. The hardware test sketch already swapped sides by hand. The main firmware's BoardPins.hpp followed the schematic, so the Navigator would command a left turn and the robot would turn right. The fix is a build option, on by default for the ESP32-S3, that moves the left motor from pins 6/5 to 16/15 and swaps the encoders. Turn it off when a board built to the schematic shows up. The truth about the robot now lives in a Kconfig default. PR is in Portuguese.ECAD · Agent failure · 1 report74
  3. 3Firmware buzzer tones were 2.5 to 3.8 kHz while hardware calibrated the transducer at a 1.2 kHz resonance, so alerts could play much quieter than intended.Firmware · Agent failure · 1 report66

How it fails

  1. Agent failure3

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