Skip to content

hw: v3.0 board, 330R ISP isolation, no 18650 holder, THT parts - #85

Merged
arcadien merged 8 commits into
masterfrom
fix/pcb_v3
Nov 18, 2026
Merged

arcadien merged 8 commits into
masterfrom
fix/pcb_v3

Conversation

@arcadien

Copy link
Copy Markdown
Owner

v3.0 board hand edits in KiCad (still stamped Rev 3.0, not ordered yet).

Board

  • R9/R10 330 Ω in series between the MCU (nets SCK/MOSI, ISP J6) and the sensor connectors J3/J14 (nets SCL/SDA): flashing with the BME280 plugged in. Bus pull-ups R2/R3 on the sensor side.
  • R5 10 kΩ RESET pull-up (was R9).
  • BT1 18650 holder removed; J13 RAW IN is a JST-XH (was a screw terminal); J10 CR2032 now behind the Q1 reverse-polarity MOSFET.
  • Capacitors checked: C4 100 nF ceramic at the radio, C1 bulk on a Ø6.3 footprint (10 µF, or 220 µF low-leakage for CR2032), C7 film for the PA1 divider (no tantalum).
  • THT footprints, 5 mm LED, Keystone 1060, three M3 holes; JP6 SDO delivered open, chosen at assembly.
  • Board 94.81 × 39.33 mm. DRC with parity: 0 unconnected, 0 parity issues; silkscreen warnings left.

Docs: BOM, WIRING + wiring.svg, BACKLOG (2026-11-18 section), README / docs/debugging.md flashing note, enclosing/README.md, PLAT-POWER-007 (pull-ups on the sensor side of R9/R10). SDO/CSB notes follow PORTA = 0.

Not done / to check

  • Silkscreen cleanup, 4 dangling wire ends in the schematic, Gerbers.
  • Flashing with the BME280 in place and I2C readings: bench check on a built board.
  • Enclosures still model v2.2; the mushroom foot needs a fit check (v3.0 cut at x = 147 ≈ 60.85 × 39.33 mm vs 60.02 × 38.97 mm).

🤖 Generated with Claude Code

arcadien and others added 8 commits November 18, 2026 14:46
WIRING.md and BACKLOG R11 said the firmware pulls PA5 up, giving address
0x77 with JP6 2-3; since TECH-HAL-004 PA5 has no pull-up, so SDO would
float. WIRING.md also states that SDO needs no resistor with JP6 1-2 (as
v2.2 pad 6 on GND) and that the working v2.2 board has 10 k on CSB.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…, THT parts

- R9/R10 330R in series between the MCU (PA4/PA6, ISP J6) and the sensor
  connectors J3/J14: the board can be flashed with the BME280 plugged in.
  Nets: SCK/MOSI on the MCU side, SCL/SDA on the sensor side; the bus
  pull-ups R3/R2 sit on the sensor side.
- R5 10k RESET pull-up (was R9).
- BT1 18650 holder removed; J13 RAW IN is a JST-XH B2B-XH-A (was a screw
  terminal); J10 CR2032 now on RAW_IN, behind the Q1 reverse-polarity MOSFET.
- C4 100n at the radio (replaces C8); C1 bulk on a 6.3 mm footprint
  (10u, or 220u low-leakage for CR2032); C7 film 100n for the PA1 divider.
- THT footprints for R1, R5, R7-R10, C6, C7; D1 5 mm; J10 Keystone 1060;
  three M3 holes MH1-MH3; JP6 SDO left open, chosen at assembly.
- Board 94.81 x 39.33 mm. DRC: 0 unconnected, 0 parity issues; only
  silkscreen warnings left.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…e R9/R10 330R

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
R9/R10 330R ISP isolation (SCK/MOSI vs SCL/SDA), R5 RESET pull-up, no BT1,
J13 JST-XH, J10 behind Q1, C4 at the radio, C1 value per supply
(220u low-leakage for CR2032), C7 film, JP6 open at assembly, M3 holes.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…impact

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
…eds it unplugged

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@arcadien
arcadien merged commit eece1c0 into master Nov 18, 2026
11 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant