First Totem build — right-half XIAO RP2040 won't enter bootloader or enumerate after soldering. Salvageable without hot air?
Howdy! First Totem build here, and I've hit a wall with the right-half RP2040 that I can't troubleshoot my way out of. Hoping someone's seen this.
Setup
- Totem split, XIAO RP2040 on each half, wired (TRS between halves)
- Firmware: Pre-Compiled VIAL
- XIAO mounted pad-to-pad, castellated vias filled
- Left/master half works perfectly. Issue is isolated to the right half
The problem The right XIAO won't enter bootloader anymore. Holding B + tapping R, shorting RST, nothing gets RPI-RP2 to mount, and the computer doesn't recognize the board at all. I flashed this exact chip fine before soldering it on, so something broke or shorted during/after mounting.
It briefly connected once out of nowhere on a reset, but now it won't, and it doesn't register keys over the TRS cable so the firmware may also be in a bad state (or it never got a clean flash in the first place).
What I've checked (multimeter + testing)
- No pads shorted to each other. RST not shorted to GND and it reads ~3.3V while powered, so it's not being held in reset.
- Flux cleaned with IPA (twice).
- All joints look well soldered. no visible cold joints or bridges.
- Not battery powered. No heat, no buzzing, no sign of a short.
- Cable is good as the same cable enumerates the working left half fine.
- Red power LED stays solid, so power is reaching the board. It's the data/connection side that's failing.
- Wiggling the USB-C cable may be changing the behavior as certain angles seem briefly connect or hold longer which makes me suspect the connector but there is no angle that "solves" it.
LED behavior (video attached)
- Originally: USER LED flashed white to green sporadically, seemed to track USB-C wiggle.
- After a second IPA + toothbrush clean: now it's a sporadic white to off flash (video). Current state.
My guesses (newbie, grain of salt)
- The USB-C data pins are cracked/damaged. not sure how to verify this directly.
- The USB-C anchor tabs may be cracked, letting the connector wobble. no visible proof.
- RST is shorted in some way I can't catch with the meter, causing constant resets.
- The chip is just broken from some unforeseen issue.
Constraints / what I'm hoping for
- No hot air station, so I can't easily reflow/replace the connector or remove the chip.
- This is the right/slave half, so it only needs USB for flashing. it should run off the TRS cable in normal use. So if I can land even one successful flash, it should work.
- I looked at SWD, but the debug pads are on the underside facing the PCB (how it's mounted), so they're buried/inaccessible without desoldering.
- Worst case I cut it off with flush cutters, wick the pads, and start over but that's a 2-3 week wait for a new chip, which I'd love to avoid if this is fixable.
Any way to verify the data pins, or revive the connection without hot air? Thanks!