You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Record the 2026-09-05 UM FeatherS3 rewire sheet in docs/HARDWARE.md #282
Copy the UM FeatherS3 rewire record from the 2026-09-05 session into docs/HARDWARE.md, verbatim:
the bench sheet (Adafruit hole → UM hole) the breadboard was rewired from (05:54, 06:02),
the header position-by-position tables with GPIO numbers for both boards,
the resulting ten board-specific firmware values (Adafruit vs UM), and the three that resolve from the variant without a source change (I²C via Wire.begin(), GPS RX/TX, SPI 36/35/37),
the caveats carried from that exchange (GPIO 0 in the second 3V position, GPIO 2 not on a header, 33/38 to confirm, an unanswered question about the A2 wire),
the owner's ruling: wire to match the UM silkscreen so the board can be traced later; keep the Adafruit 5477 assignments so the project can move back.
The Adafruit 5477 tables stay unchanged.
Why
The sheet was drafted on 2026-09-05, offered for docs/HARDWARE.md twice, and never written. On 2026-09-07 a session re-derived the UM pins from the variant header's A0–A5 aliases (the exact mistake retracted at 05:59 two days earlier), produced a wrong "TFT on the wrong pins" table, and stopped a flash the owner had ordered. Three /violation corrections followed. The record exists only in a chat transcript; this issue makes it durable.
Firmware status this records (main @ 1be96cb, 2026-09-07)
platformio.ini has one env, feather_s3 (board = adafruit_feather_esp32s3); every pin define is the Adafruit value. A feather_s3 image on the UM board drives the wrong GPIOs for TFT CS/DC/RST, FRAM_CS, SD_CS, CTP_INT, TFT_BL, the buttons, I²C and the GPS UART. The UM env and board-profile work is separate (under #151); this issue is documentation only.
Acceptance
docs/HARDWARE.md carries the rewire section with all tables verbatim from the 2026-09-05 record, sourced and dated
Revision history row added
No firmware change
Owner has walked the bench sheet against the physical board (the UM column was read from the FeatherS3D pinout PDF; unverified against hardware as of filing)
In PR #288. The record grew past this issue's scope during the 2026-09-08 bench session: docs/HARDWARE.md carries the rewire record, the display module wire-by-wire table, the FeatherS3D pinout-card corrections (two I2C buses, LDO2 hole, MAX17048 present) and a warning that its Adafruit tables are not UM holes; docs/um-bench-wiring.md is the LOCKED UM wire list (owner, 02:20 EDT), silkscreen labels only, every peripheral verified up at 02:16. Owner's A2 ruling recorded and confirmed on hardware (RESET on hole 14). Closes on merge.
Activity
Strycher commentedon Sep 8, 2026
In PR #288. The record grew past this issue's scope during the 2026-09-08 bench session: docs/HARDWARE.md carries the rewire record, the display module wire-by-wire table, the FeatherS3D pinout-card corrections (two I2C buses, LDO2 hole, MAX17048 present) and a warning that its Adafruit tables are not UM holes; docs/um-bench-wiring.md is the LOCKED UM wire list (owner, 02:20 EDT), silkscreen labels only, every peripheral verified up at 02:16. Owner's A2 ruling recorded and confirmed on hardware (RESET on hole 14). Closes on merge.
docs(#282): record the 2026-09-05 UM FeatherS3 rewire sheet in HARDWA…
docs(#282): record the owner's A2 ruling -- A2 is TFT_RST until prove…
feat(#283): UM FeatherS3 build env and board pin profile (um_feathers3)
docs(#282): record the 2026-09-05 UM FeatherS3 rewire sheet in HARDWA…
docs(#282): record the owner's A2 ruling -- A2 is TFT_RST until prove…
feat(#283): UM FeatherS3 build env and board pin profile (um_feathers3)