Has micro:bit got a over-hot chip issue?

Mike Meakin got in touch with EW to describe an over-heating problem he found with his BBC micro:bit.

BBC Micro:bit - one million are to be manufactured for October

It looks like that one of the chips on some micro:bits is getting very hot.

It happened with his, and he has found two bloggers that refer to similar events:


Andre Boulton

The second of which has a video of a drip of water boiling away on the chip.

In this case, the chip is the NXP Kinetis ARM Cortex-M0 chip that handles USB communication for the board – the chip just below the white connector and push button in the picture.

Mike is thinking that maybe is is the result of an ESD incident while the board is powered.

Or maybe it is getting briefly over-volted?

Does anyone have a proper circuit diagram, asks Mike.





  1. Thanks Malcolm

    That makes a lot of sense.

    Because of its insight, I am going to put your message up as a separate blog.

    Thanks again

  2. The issue is likely latchup due to a deviation in design from the the orignal Nordic development board from which the micro:bit is derived. The micro:bit has two processors, the main Nordic and a second (MKL26Z) used for USB programming and comms. To reduce power consumption, only the Nordic is powered from the battery supply whilst the USB supplies both. A potential issue is when powered by battery only, the Nordic can place voltage on some of the GPIO pins of the MKL26Z whilst its unpowered. So far not a issue, but as soon as the USB is plugged whilst the battery is still connected, the MKL26Z powers up. Powering up a chip when voltage is already present on input pins is very bad and usually results in ‘latchup’ and a very hot blown processor.
    micro:bit schematic can be found at:

  3. Thanks Jim
    I didn’t know that CodeBug even existed. I see the family resemblance. It looks just like the early micro:bit prototype we were sent a photo of.
    From the schematics, it appears that codebug has PIC18, while the chip getting hot on at least one micro:bit is a Kinetis from NXP.
    It looks like the CodeBug polyfuse is in the USB power line – is that the one you are referring to?
    I am not sure fuse there would stop the problem on micro:bit starting, but it would help control the aftermath.
    Anyone know where the micro:bit schematics are?

  4. Looks like cost cutting. Interesting that the CodeBug http://www.codebug.org.uk, which micro:bit derived from has a polyfuse to protect it.

Leave a Reply

Your email address will not be published. Required fields are marked *