This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

F28335_The processor becomes hot

Hi,

I was doing testing by using the F28335. All on a sudden, i saw the LED is blinking beside the processor and i checked it became so hot. I removed the power immediately. Does anybody have any idea what happened? I even dont know whether the module will work or not.

Arafat

  • Hello A K M,

    From my experience a controller that was working, then you did something, there was a brilliant flash of light or a puff of magical smoke is seen rising up and then it gets hot - usually means you fried the chip.  But there are exceptions.

    If you power it up after removing everything connected to the port pin (if possible) and it still gets hot enough to cause severe pain on your finger, then it may be NFG!

    But if it only get mildly hot and some of the peripheral ports still work, then you can use it until it eventually fails, which it will.

    Test for a short between Vcc and GND, shorted caps or components.  If there are none, remove the chip, test again for shorts and replace it with a new one.

    Thanks,

  • Hi Arafat,

    First of all, I would like you try flashing the controller with the same configuration. If you're able to do so, then your LDO is damaged and you'll have to replace it. If you're unable to flash ie you get some error while flashing, your MCU is bricked/damaged. Do share your observations.

    Also, most importantly before conducting this test please check the continuity between power and ground signals. Check for any shorts on the board. If no such shorts are observed, monitor the voltgae levels on the board and compare them with those mentioned in the schematics.

    Regards,

    Gautam 

  • How hot was the controller? if it was too hot, it can only because of a short circuit. 

  • On the older F281x devices, connecting anything over 3.3V to any GPIO pin kills the associated GPIO/peripheral block instantly. In this case the chip gets very hot, although the core sometimes survives and the JTAG interface will still work.

    It is easy for this to happen accidentally when probing around your board.

  • Hi,

    Thanks for your reply. This might be happened. But, the board is designed in such a way so that there is no way to get voltage more than 3.3v. I shall double check it. Thanks.

    Arafat

  • It is getting hot gradually and becomes severe. I have tested it disconnecting all the pins and connected only to vcc and ground. Still it gets hot.

    Arafat

  • Hi Gautam,

    I do not know how to flash F28335.

    Arafat

  • It is getting hot gradually and becomes severe. I have tested it disconnecting all the pins and connected only to vcc and ground. Still it gets hot.

    Feels like a short, Arafat! Device might be damaged.

    Regards,

    Gautam

  • Hi Greenja,

    Can you briefly explain what is NFG! I can easily replace with new one.But, if i do not know what happened exactly i may face this again which i do not want to.

    Arafat

  • LOL!

    Not Functioning Good.  That is the definition you would use in public.  

    The real definition is No (replace Function with curse word that starts with F) Good. 

    Once you have removed the chip, then it will be easier to determine if the problem is chip based.  A Microcontroller will only get hot if there is a problem internal to it or with a component in very close proximity to it.  

    A problem with the voltage regulator would appear at the regular unless it is providing greater than Vcc to the MCU.  In that case, you would have noticed the irregularity in the voltage.  A shorted component after the regulator would cause it to shutdown in most cases.  A problem before the regulator usually causes the regulator to overheat.

    When you applied power to it, did you still get 3.3V?  Was there an abnormal current reading?

    Once you remove the MCU, you can test both the board and MCU separately.  You should be able to find a Vcc and gorund pin far enough apart from each other to solder wires to them.  Apply 3.3V and see if it still gets hot.  Test the board to see that you are getting 3.3V and the current draw should be minimal.

    Thanks,

  • Hi Greenja,

    I tested the MCU independently. I applied 3.3v using power supply. The chip still gets hot. There is chance of having internal abnormalities within the DSP. I am not sure exactly. I also checked the power supply which is very stable not giving any spikes at starting.

    Arafat 

  • Sorry Arafat, your MCU might be damaged..

    Nishanth              

  • greenja said:

    LOL!

    Not Functioning Good.  That is the definition you would use in public.

    I didn't know that!

    Arafat said:
    I tested the MCU independently. I applied 3.3v using power supply. The chip still gets hot. There is chance of having internal abnormalities within the DSP. I am not sure exactly. I also checked the power supply which is very stable not giving any spikes at starting.

    Applying 3.3V from external power supply and your chip still getting hot itself denotes that the mcu is internally damaged. Also, check whether your power supply is going to cc limit. Please check your LDO output ie U2 of controlCard (if you're using one) or if its a custom one the whatever you're using for generating 3..3V. Is it stable?

    Such shorts happen while using DSO probes and you accidentally touch 2 pins together etc.

    Regards,

    Gautam