• State Not Answered
  • Replies 5 replies
  • Subscribers 25 subscribers
  • Views 320 views
  • Users 0 members are here
Attachments (0)
Nordic Case Info
  • Case ID: 283983
Options

nRF52832 cannot programmed

user82895

Hi Nordic

When I use JLINK to burn the firmware to the nRF52832 chip in the device, some devices cannot be programmed or cannot be programmed again after burning once

The following experiments have been done:

  • The nRF52832 chip on a normal device can be programmed repeatedly
  • Solder the nRF52832 chip of the problematic device to the normal device, it cannot be programmed
  • Solder the nRF52832 chip of the normal device to the problematic device, and it can be programmed in the first few times, but it cannot be programmed after a long time of power-on, and the nRF52832 chip can not be programmed after soldering the it back to the normal device again.
  • Use the 3.3V power supply of the problem device to power the nRF52832 chip of the normal device, and it can be programmed normally
  • There is a faulty device that cannot be programmed again after being programmed for the first time, but the chip can work normally

Entering the 'nrfjprog --recover' command, and use an oscilloscope to test the SWDIO and SWDCLK pins of the abnormal device. The photos are as follows, yellow is SWDIO, green is SWKCLK

The SWDIO and SWDCLK pins of a normal device, the levels are as follows

The SWDIO and SWDCLK pins of a normal device, the levels are as follows

layout is as follow

schematic design is as follow

nRF52832 part

core board part

5V to 3.3V power design

What could be the reason for this problem? Please help.

Parents
  • 0

    Hellow Håkon

    thank you for your reply, the answers to your questions is as follows

    user2115 said:
    Please correct me if I am mistaken, but it seems as the swap-test is failing for both "good" and "bad" device in this scenario.

    Yes, a bad device seems to cause chip damage, even if solder a good nRF52832 chip to the bad device, it cannot be programmed (there is also a chip that can be programmed once, but cannot be programmed later).

    user2115 said:
    Q1: Has the board(s) been checked to see if voltages are in order?

    The voltage is normal, the ripple test and noise test are also ok.

    user2115 said:
    Q2: How many devices are failing in total? And how many are produced in total?

    sample total: 50 PCS

    bad sample: 13 PCS

    The nRF52832 on the device has several GPIOs connected to the core board. Could it be that the GPIOs of the core board output high current or high voltage and damage the nRF52832 chip? Because from the waveform of the SWD data line captured by the oscilloscope, the SWDIO signal on the bad device seems to be abnormal.

    thanks

  • 0 in reply to user82895

    Hi,

     

    ndc_bee said:
    The voltage is normal, the ripple test and noise test are also ok.

    I agree, this looks well within.

    ndc_bee said:
    The nRF52832 on the device has several GPIOs connected to the core board. Could it be that the GPIOs of the core board output high current or high voltage and damage the nRF52832 chip? Because from the waveform of the SWD data line captured by the oscilloscope, the SWDIO signal on the bad device seems to be abnormal.

    Potentially, if there are two outputs (ie. nRF low, core board high), you can potentially damage the device over time. However, this should show up if you check the current of the bad board vs. a good board. I also see that the nRESET pin is routed, is this controlled by the external core board? If so, you could try to check if this pin is ever pulled low externally.

    When you mention the waveform, I assume it is this section in red that you are worried about:

    It is hard to say if this is noise, or if the nRF keeps the line low. Do you have a log from the programming/recovery sequence?

     

    Kind regards,

    Håkon

Reply
  • 0 in reply to user82895

    Hi,

     

    ndc_bee said:
    The voltage is normal, the ripple test and noise test are also ok.

    I agree, this looks well within.

    ndc_bee said:
    The nRF52832 on the device has several GPIOs connected to the core board. Could it be that the GPIOs of the core board output high current or high voltage and damage the nRF52832 chip? Because from the waveform of the SWD data line captured by the oscilloscope, the SWDIO signal on the bad device seems to be abnormal.

    Potentially, if there are two outputs (ie. nRF low, core board high), you can potentially damage the device over time. However, this should show up if you check the current of the bad board vs. a good board. I also see that the nRESET pin is routed, is this controlled by the external core board? If so, you could try to check if this pin is ever pulled low externally.

    When you mention the waveform, I assume it is this section in red that you are worried about:

    It is hard to say if this is noise, or if the nRF keeps the line low. Do you have a log from the programming/recovery sequence?

     

    Kind regards,

    Håkon

Children
No Data
Related
Terms of service