search

zephyrproject-rtos / zephyr

Primary Git Repository for the Zephyr Project. Zephyr is a new generation, scalable, optimized, secure RTOS for multiple hardware architectures.
https://docs.zephyrproject.org
Apache License 2.0
10.55k stars 6.46k forks source link

my nrf52 DK gets damaged partially after hundreds of erase and write? #12462

Closed summer20100514 closed 5 years ago

summer20100514 commented 5 years ago

I choose one NRF52 DK(nrf52840_pca10056) for my daily development. After hundreds of erase and write cycles, the DK won't behaves like it should do. The other DKs runs well while this one not. But some basic example runs well. How can I figure out if this DK is damaged? If so, why is it damaged partially?

what I have tested is, zephyr ibeacon sample works well, while peripheral sample wont work as expected. However, other DKs work well.

peripheral sample output

Booting Zephyr OS zephyr-v1.13.0-3079-g4e23d6c875 Bluetooth initialized [00:00:01.514

and it is stuck at static inline void cpu_sleep(void) { WFE(); __SEV(); // it stucks here WFE(); }

Environment (please complete the following information):

summer20100514 commented 5 years ago

the ibeacon log using the same DK is

Booting Zephyr OS zephyr-v1.13.0-3099-g35da3e3e0f Starting iBeacon Demo Bluetooth initialized [00iBeacon started :00:00.007,415] bt_hci_core: HW Platform: Nordic Semiconductor (0x0002) [00:00:00.007,415] bt_hci_core: HW Variant: nRF52x (0x0002) [00:00:00.007,415] bt_hci_core: Firmware: Standard Bluetooth controller (0x00) Version 1.13 Build 99 [00:00:00.007,843] bt_hci_core: Identity: e9:08:a3:ee:25:67 (random) [00:00:00.007,843] bt_hci_core: HCI: version 5.0 (0x09) revision 0x0000, manufacturer 0x05f1 [00:00:00.007,843] bt_hci_core: LMP: version 5.0 (0x09) subver 0xffff

and I can see the beacon using my phone.

summer20100514 commented 5 years ago

I also write a post on nordic devzone, but I am not sure whether if it is caused by hardware failure or by software damage.

carlescufi commented 5 years ago

This might be a hardware issue, but I've copied a couple of people to see if they can provide additional information

summer20100514 commented 5 years ago

@carlescufi Thank you. What still confused me is the chip shouldn't be damaged by just hundreds of erase cycles.

summer20100514 commented 5 years ago

The officials from nordic commented that it must be caused by ESD damage. It's a pity.