I used the ST-LINK to inject the firmware. It went normally as in the tutorials.
The master and the slave work with blue and green lights.
But on the repaired part (master), it is impossible to connect via USB, but it works via UART with METR PRO and via CAN)
I tried to upload the bootloaders via VESC-tool on the master and slave but like that I bricked both, and it was necessary to go back through the ST-LINK for both (with the small cables that are really annoying to solder).
I tried injecting the bootloader via STLink, but it didn’t fix the USB problem.
I tried putting some solder back on the STM legs, but it didn’t fix anything.
I can change the firmware on this VESC only by injecting it via STLink. I have no issue connecting to the twin VESC via USB.
Does anyone know where the problem could come from?
Is it possible I have to change a parameter like adress or something else in STLink to inject the bootloader properly?
A Bluetooth module might be worth trying, either connecting to mobile or pc. I have a flipsky 6.7 mini that got partly vaporised, I think I can desolder that chip from the good side and send it to you if it is what you need.
But actually I can use this dual VESC. I access the VESC that has the dead USB through the CAN or the METR PRO module.
And I can change the firmware at my convenience through the STLink.
Currently it has the ACKMANIAC firmware and the board is fully functional. Back then, with the SK8 192KV motors, I felt that this firmware offered more power and less heat, compared to the official firmware.
On the other hand, even through the CAN, it is impossible to change the firmware through VescTool on this VESC that is inaccessible through USB.
That’s why I have the impression that it has to do with the bootloader.
But injecting the bootloader through the STLink or through the VescTool doesn’t change anything.
Nice! Congrats on the successful chip replacement, not an easy thing to do
If both sides of the vesc works, and you can set vesc settings on the side with the bad usb via CAN forwarding, that sounds like a good enough workaround. If I were in your shoes I’d probably just mark the one usb bad and move onto another project, debugging this is probably a lot of effort haha. Super curious what the issue might be though.
Is there certain capability that requires the second usb to work?
In my experience, if you’re inexperienced at chip replacement then my guess is it’s probably either an open solder joint, or a short. I’ve cocked up more than I’ve got right when changing chips.
Do you have a magnifying glass? I’d be looking close AF especially around the relevant pins. Even how well you clean the PCB afterwards could matter, more so if you’ve used a flux that could leave a conductive residue. It doesn’t take a lot to have a short from tiny balls of stray solder or an open leg.
I’ve recently swapped an MCU on a bms and one of the temp sensors is intermittent as the board warms up, and open solder joint for sure that I still need to revisit.