Errata

Known faults in built boards, how to spot them, and how to rework them.

Known problems with boards as built, listed by the revision printed on the silkscreen. Each entry gives the symptom first, since that is what you will see on the bench.

FemtoLCC v1

E1 — R5 not fitted: the module never leaves reset

Symptom. The board powers up but the ESP32-C6 does nothing. Plugged into a computer over USB-C at P1, nothing enumerates at all — no 303a:1001 device, no /dev/ttyACM* port, not even a failed enumeration. The UART header at J1 is silent too, and holding BOOT while tapping RESET makes no difference.

Cause. R5 is the 47 k pull-up from +3.3V to the module’s EN (CHIP_PU, U14 pin 3). The C6 has no internal pull-up on that pin, so with R5 absent EN floats low and the chip is held in reset permanently. The schematic and PCB are correct; the fault is in the fab pipeline. tools/do-not-populate.txt listed R5 as “unpopulated by design”, so it was dropped from the assembly BOM and CPL and JLC never placed it.

Affected. Every v1 board assembled from the repository’s fab outputs.

Rework. Fit a 47 k 0402 resistor at R5. Anything from 10 k to 47 k works; a leaded resistor from U14 pin 3 to +3.3V does as a bodge if 0402 is awkward.

A solder bridge across the R5 pads also brings the board up, and it is safe — the reset button pulls EN to ground through R24 (2.2 k), so pressing it only draws about 1.5 mA. But it costs two things:

• The RESET button stops working. EN is held at 3.3 V however hard SW1 pulls.
• The power-on delay is lost. R5 and C46 (1 µF) normally hold EN low for around 50 ms while the 3.3 V rail settles. Tied straight to the rail, EN rises with it, which Espressif advises against and which can show up as the odd failed cold boot.

Flashing is unaffected either way: esptool resets the C6 through its USB Serial/JTAG peripheral, not through EN.

Fix for future builds. The R5 line was removed from tools/do-not-populate.txt and the assembly files regenerated. R5 now goes out in the BOM and CPL with its lcsc-parts.csv match, C25792, the same part as R14.

Status. Fixed in the fab files for future builds. The DNP entry was removed on 2026-09-14. Boards already built still need the rework above.