USB port brownout fix: 220 µF tantalum at C513
When a USB stick drops the keyboard into 'Maple DFU' mode the mainboard's 1 A USB switch is sagging; a 220 µF tantalum at C513 cured it for one builder.
What it takes
- Disassembly
- Soldering
- Purchase required
Difficulty 3 of 3 — advanced.
Details
SilentByte's thread started with the symptom: plug a device into the USB-A port and the keyboard disconnects and re-enumerates as Leaflabs Maple DFU (1eaf:0003) until reboot, with hissing from the speakers at low battery — on a CM5, though YoelFievelBenAvram notes the keyboard drop happens on a CM4 too. QuantumKraken found the cause in the TPS2553 USB power switch, whose 24.9 kΩ set resistor gives a nominal 1 A limit, and fixed it by replacing C513 on the back of the mainboard with a 220 µF tantalum capacitor; vxi lists the part as a 3528 (B-case) 220 µF, ideally 10 V, 6.3 V being what was affordable. neocska92's less invasive advice for CM5 owners is 10 A-rated cells such as the Sony/Murata VTC5A and a downclock.
One confirmed repair and a parts list, so treat it as experimental.
Links
Open at Forum threadRelated mods
Remove R120 and bridge the compute module's 3.3 V to GND_VREF, and the power button's long press force-powers a CM5 off again — no battery pull needed.
bored99's one-piece replacement back leaves the CM5 heatsink open to the air and covers the cells with a magnet-held plastic lid for quick swaps.
UJW's deeper printed back fits two 10,000 mAh 146074 pouches and a Waveshare CM5 active cooler over the HackerGadgets boards — designed in CAD, not yet built.
Any genuine branded 3,000–3,500 mAh 18650 works, flat-top or button-top; listings claiming much more than 3,500 mAh are fakes.
The 21700 version of HG's NVMe battery board fits the stock backplate and roughly doubles runtime: 4 h 31 min of 1080p playback against 2 h 27 min on 18650s.
paragonnov's Debian package remaps the power key to a real screen-off sleep: backlight and LCD off, keyboard unpowered, CPU pinned to its minimum clock.