Skip to main content
daviddawe1982
Associate II
May 11, 2013
Question

STM32F4 running hot

  • May 11, 2013
  • 3 replies
  • 1590 views
Posted on May 11, 2013 at 10:43

hello my sth32f4 discovery is mcu is running hot and u1 is hot also what is going on??

    This topic has been closed for replies.

    3 replies

    Tesla DeLorean
    Guru
    May 11, 2013
    Posted on May 11, 2013 at 14:42

    How hot would that be? In units of C or F, or how long your figure can dwell without burning/pain?

    What was happening prior to this condition being observed?

    High voltages, exceeding operating frequency, high load, shorts?

    Does it persist if you reset with BOOT[0] High?
    Tips, Buy me a coffee, or three.. PayPal VenmoUp vote any posts that you find helpful, it shows what's working..
    daviddawe1982
    Associate II
    May 12, 2013
    Posted on May 12, 2013 at 13:12

    i was using the adc and i think i shorted 12v to the adc there is a lot of noise on the adc now .

    boot did not fix the problem

    and i can run the board for about 30-40 seconds before i have to remove my finger

    jj2
    Associate
    May 12, 2013
    Posted on May 12, 2013 at 22:19

    Does not sound good - but damage may be confined just to that single ADC pin.  (we hope you limited the application of ''illegal'' excess voltage to just one pin)

    We've been guilty - many others too - but in our case - no more.  Our solution was to design a multi-channel, voltage-clamped, buffer/op-amp board - which is unable to output beyond 3V3 to our MCUs under test.  We have voltage clamps on both every input channel and redundantly (for safety) upon every output channel.  And - should our circuit detect an over-voltage condition - we immediately open a power FET which routes all voltages to our buffer/op-amp board.  Not quite to ''space station'' or ''human implantation'' standards - but ''good for Gov't work - in most cases.''

    Now in your case - check/double check that the 12V has been safely/properly divided down to ''MCU appropriate'' levels.  Suggest that you remove all signals from that suspect pin (recipient of your over-voltage) - does the MCU still over-heat?  And - ''finger method'' of heat measure may introduce ESD - along with not providing an ''overly exact/repeatable'' measurement.

    If you're lucky - you may be able to salvage many/most of the MCU's other pins.  A good measure of the MCU's health may be its current draw - prior to and after abuse.  A second M4 demo board enables such a, ''baseline'' measure...