I had been using the xiao_ble board with the xiao_expansion shield with vanilla zephyr and so far things worked as they should. I could do the basic stuff – GPIO, OLED, Buzzer, etc.,
Yesterday, I enabled support for mcuboot bootloader and USB DFU support for firmware upgrades and the board stopped working a while after I flashed the firmware with USB device support (I verified that the firmware indeed exposed a USB DFU class). When I say not working, I mean: LED stoped blinking and jlink can no longer detect the board.
At this time, I found that the metallic shield cover on the Xiao board was heating up too. So I quickly removed the board from the expansion board.
So the question is, is this a known issue? has anyone else faced any issues with Xiao BLE board and the Xiao expansion board? have these two boards been qualified/designed to work with each other?
So , Nothing you could do in Software/Firmware is going to Self destruct the MCU, Only plugging it into the Dev board backwards or off by one pin.
From your description, sounds as if you overwrote the BootLoader and code segment that was for Adafruit/Bootloader (stock) , your addressing may be off, if using the MCU boot what options in the K_file do you have set?
I wiped the adafruit bootloader on day one as it doesn’t fit my taste/needs; I was developing my application without a bootloader using jlink to directly flash the board using the swd interface exposed by the expansion board.
After my app matured a bit, I enabled mcuboot (another bootloader) and verified that it was indeed stage 1 bootloader and it booted my application correctly. Before the failure last night, I was enabling USB DFU to ease flashing new applications using the now fully working bootloader. I suspect that enabling the USB driver is somehow related to this issue as they happened in close succession.
The context above should also rule out the possibility of the dev board plugged in the wrong way as I was using the swd interface to flash the chip – which would certainly not work if the pogo pins don’t align pefectly with the pads.
I agree with PJ nothing can be done in software to damage the device
with this device it is easy to install one pin forward or one pin backward
especially if the unit is powered or the expansion board has a battery installed and this will short either the 5v or the 3v3 to ground… that may let the smoke out… I have done this when connected to a usb to computer and you immediatly hear the disconect sound play as the USB on the computer shorts out so be careful
I’m referring to this board: Expansion Board Base for XIAO | Seeed Studio Wiki and yes it’s possible to have one pin forward or backward even with this board but then the SWD interface just won’t work. Like I said, I was able to flash the board just a few seconds before I noticed misbehavior and then the heat up. Also, I have double and triple checked that it’s not something as trivial as a mispositioned xiao board.
Since then, I’m now inspecting the schematic of the expansion board and it appears the 3.3v from the internal regulator is connected to the one on the expansion board. This could cause problems if one of the regulator was shutdown briefly and other was supplying no? Need to take a closer look at the components they are used.
A short or excessive power consumption might happen on the board after flashing.
Can you please elaborate? In what situations can this happen? I have been flashing this board dozens of time a day a few weeks without any issues.
But to be fair, I haven’t used the USB port on the Xiao board other than to power the board until very recently when I started working on the bootloader.
i think you are not understanding that a short bad enough to cause heat could do any number of instant damage either to the XIAO or the Board
I would say test the device off of the board… my gues is it is fried or a power regulator on the Expansion board is fried— I would say get a new shipment of gear and try not to do that again IMO FWIW
I already verified that the Xiao module is fine (I pulled it out as soon as I noticed the heat-up). Now I found that the 5v and 3.3v output of the expansion board when powered by a battery is < 1v. So the power circuit on the expansion board is definitely toast.
That also explains why the Xiao was heating up: since two 3.3v regulators outputs are tied together, when the regulator on expansion board failed, it started drawing too much current from the one of the xiao.
Guess I have to write to Seeed studio support about this issue.
Be sure the Battery Polarity on the Xiao Expansion board is wired properly (double check ) if your using the JST plug on there.
Definitely a strange occurrence though , I know I had something similar a long time ago XIAO got HOT , I pulled it fast, and ended up finding when the xiao had gotten hot like that , The Battery CHARGER was fried , other than that everything else on the chip appeared to work ok. I could still flash it even. So just beware.
So you have the Jlink connected to the SWD pins ,how is it powered , what’s the Vref? , USB from XIAO or JTAG * SWD pins. I have pictures on the thread here where I have the JLinK connected directly and powering the XAIO, and also the Nrf52840 -DK board connected to it. Each method has it’s own set of caveats and bewares, FYI
Using the USB_c on the XIAO at the same time is a NO-NO…btw.
I do have a 3d printed case on the expansion board. And I was careful. This is not my first embedded project
you broke it you bought it… IMO
Are you even reading what I wrote before responding? I told you the xiao was plugged in right, and that I flashed with SWD interface (xiao powered by the USB port) with a debugger and that caused this failure. What part of that triggered your “you broke it” response? Is operating things the way they are documented called “breaking it” to you?
Please, try to keep the discussion objective and to the points I’ve mentioned here instead of asking me to be more “careful” or casting your opinions. This is getting us nowhere. If you are totally unsure what could have gone wrong, please allow others who might know something to respond.
It will be pretty darn stupid of me to buy another expansion board without knowing exactly why the first one failed now wouldn’t? For all I know the new one will also fail as soon as I flash the USB DFU firmware.
The battery polarity is correct, as I said, I have been using this board for quite some time. The 3d printed bottom case doubles as a battery compartment too (pulled the battery out for this picture). The board is powered via USB-C and jlink uses the 3.3v pin as Vtref to detect target voltage only (not supply it).
Hi,
Thank you so much for your feedback.
We’re delighted to tell you that your issue report has been received by the technical support team! We will respond to you by email.We’ll get back to you as soon as we can, so please be patient.
Hi,
We apologize for the inconvenience. Our technical support team is already following up on the issue and will continue to communicate with you via email. If you encounter any problems, please feel free to contact me directly.
PJ 
