How I fixed a dead USB port on my Black Pill with a bodge wire

Nov 01, 2026 Last reply: 1 day ago 3 Replies

First reply

swalsh

Before you start chasing traces, try flipping the cable over. I burned a whole evening on a similar F411 board assuming the connector was faulty, and it turned out the USB-C socket only had one set of data pins wired to the pads. Cheap two-sided connectors like that are common…

Read the full reply

Ran into the exact same thing last month with one of those cheap F411 boards. USB-C plug did nothing, no enumeration, but the board ran happily off a separate supply. Traced it back the same way you did and found the data line was open between the chip and the socket.



Here's what worked for me. First put your meter on the small passives sitting between the chip and the connector. On some of these clone boards there's a pair of tiny series resistors in the data lines, and on mine one of them had a cracked joint you could not see without a loupe. Touching it up with fresh solder and a bit of flux fixed one board entirely. If the resistors check out, follow the trace with continuity beeps untill you find where it dies. On mine it was a via right under the connector footprint, probably cracked from the factory or damaged when the socket was hand-soldered.



Once you find the break, scrape a little of the green mask off each side with a craft knife, tin it, and lay a thin wire across. Ugly but it has survived months of plugging and unplugging. Honestly if the board was cheap I'd just do the fix, returns on these are a coin flip anyway.


Before you start chasing traces, try flipping the cable over. I burned a whole evening on a similar F411 board assuming the connector was faulty, and it turned out the USB-C socket only had one set of data pins wired to the pads. Cheap two-sided connectors like that are common on these boards, and plenty of cables only make contact through one orientation. Plugged in one way, nothing. Flipped it 180 degrees and it enumerated instantly on my Linux box. I only found it because I had two cables on the desk and one of them was built differently and worked either way. Now I check that first on every new board before touching a meter. If flipping does nothing on both cables and both orientations, then yeah, go after the small parts near the connector like the earlier reply said. For what it costs, I keep a few known good cables in a drawer just for board bring-up so I am never debugging my own wiring at the same time as the board.

I was having trouble with a diffrent setup where the chip just wouldn't boot at all when plugged in. The data lines seemed fine on my end, but the power connection felt iffy. Sometimes the board whould draw a tiny current and then die right away. I wonder if the issue is that the power delivery isn't properly handshake-ing with the host before it even tries to talk to the data lines? If the chip isn't getting stable voltage, does it skip the entire initialization sequence for USB?

Always assume the physical connection is bad before you start soldering up the PCB. Even if the microfracture is tiny, the mechanical stress of pushing and pulling that connector often causes intermittent failure, and that's enough to kill enumeration. I’ve seen cheap sockets lose thier grip after just a couple of weeks of use and then fail completley when you try to plug it in. It isn't always about the trace itself.

Try shaking the connector slightly while plugging it in. If the connection feels loose, that's your culprit. These F411 clones are often cut loose with the cheapest components available, and they just aren't built for repeated heavy use.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required