• State Resolved
  • Locked Locked
  • Replies 10 replies
  • Subscribers 52 subscribers
  • Views 726 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

CC2650: CC2650 dead after a while

Jose Vicente
Prodigy 240 points
Part Number: CC2650
Other Parts Discussed in Thread: BLE-STACK, CC2640

Hi,

I designed an own device with a CC2650. Some devices with this design were made. After few months some of these devices stopped working.

The CC2650 is powered from a 3V3 regulator and works with internal DC-DC regulator and external 24MHz and 32.768kHz crystals with the recommended layout. In addition, the device works near a NFC transponder (maybe it's important to know).

I have tried to access them for reprogramming but it is not possible, the microcontroller is dead.

When turn on the power, the device advertising for milliseconds sometimes and other times not.

Also I have measure the voltage on internal DC-DC output (pin 17). At startup the value is correct but immediately go down.

Any idea what the problem is?

Thanks in advance.

Regards,

Jose.

  • 0
    TI__Guru* 81261 points
    Hi Jose,

    Are you working on a custom board or a TI development board?

    What version of the BLE-Stack are you using?
  • 0 in reply to Marie H
    Prodigy 240 points
    It's a custom board with the 2.1.1 BLE-Stack because code was developed more than 2 years ago.

    Thanks for the reply,
    Jose.
  • 0 in reply to Jose Vicente
    TI__Guru 56590 points
    Hello Jose,
    Have you checked the power source/battery?
    Have you measured both VDDR and DCOUPL?
    Have you tried to connect with SmartRF Flash Programmer 2 and do a mass erase?
    If you are not able to connect over JTAG you can try to desolder the chip over to a TI CC2650 Launchpad and try to connect the debugger again.
  • 0 in reply to Eirik V
    Prodigy 240 points

    Eirik V said:
    Hello Jose,
    Have you checked the power source/battery?
    Have you measured both VDDR and DCOUPL?
    Have you tried to connect with SmartRF Flash Programmer 2 and do a mass erase?
    If you are not able to connect over JTAG you can try to desolder the chip over to a TI CC2650Launchpad and try to connect the debugger again.

    Hi Eirik,

    Thanks for the response.

    - Battery and 3V3 voltage regulator have correct outputs.

    - Yes, when turn on the device they are correct for a few moments but immediately go down with internal DC-DC output (pin 17).

    - Yes. It has specifics contact points for re/programming CC2650. Also I have tried to desolder and insert it in a 32-pins socket but doesn't work any way.

    Do you think is possible that the generated electromagnetic field from the NFC transponder may cause damage to CC2650 after some time?

    Regards,

    Jose.

  • 0 in reply to Jose Vicente
    TI__Guru 56590 points
    Hello Jose,
    I don't know if this field will be strong enough to induce ESD damage, altghoug it seems like the device might be damaged in some way.
    Can the transponder affect the CC2640 power supply or voltage on any IO pin to go above max ratings?
  • 0 in reply to Eirik V
    Prodigy 240 points

    Eirik V said:
    Hello Jose,
    I don't know if this field will be strong enough to induce ESD damage, altghoug it seems like the device might be damaged in some way.
    Can the transponder affect the CC2640power supply or voltage on any IO pin to go above max ratings?

    Hello,

    I attach the measure of VDDR (from the DCDC output) in a functional device, other IO signals are normal. The peaks are induced every reading of the NFC transponder (every 5 minutes in normal use) probably helped by the required coil. The only idea is that after some time it may have caused damage.

    What is your opinion? Any idea is welcome. Thanks again.

    Best regards,

    Jose.

  • 0 in reply to Jose Vicente
    TI__Guru 56590 points
    Hello Jose,
    Is the device running any BLE connection or other processing or does the peaks simply correspond to NFC reads (induced field)?
    How does VDDR look without any NFC communication?
    According the the plot VDDR peak is 2.92 Volts, correct?
  • 0 in reply to Eirik V
    Prodigy 240 points
    Hello Eirik,

    The device is running a BLE connection with a smartphone but peaks appear only when an NFC reading occurs. For the remaining time VDDR has a stable signal without interferences.

    Yes, peaks are around 2.95V in VDDR.

    Thanks for the help.
    Best Regards,
    Jose.
  • 0 in reply to Jose Vicente
    TI__Guru 56590 points

    Hello Jose,

    That is far above max ratings for VDDR and might cause damage to the internal circuits that are powered from VDDR. For DCDC regulator mode VDDR should remain stable on 1.68 V all the time.

    Have you connected anything else to VDDR (to draw power from VDDR externally)?

    You need to figure out if this comes from the transponder or is due to the NFC field directly.

    Transponder:

    Review if there are any connections from the transponder that might cause voltage above max ratings on the CC2650 Supply voltage or on any other pin. In that case you might find a way to redesign the board. Try to measure voltages on the other CC2650 pins as well.

    NFC Field:

    You can try to check if the NFC field induced on the CC2650 antenna cause the spikes by disconnecting the antenna and measure. The max ratings for RF input is given in the data sheet. Maybe additional ESD protection on the antenna or a redesign can mitigate the issue.

  • 0 in reply to Eirik V
    Prodigy 240 points
    Hi Eirik,

    VDDR is not used to power external circuit. I think problem is NFC field. NFC antenna is too close to VDDR strip. It will be necessary to modify this in future.

    Many thanks for your help.
    Best Regards,
    Jose.