nrf51822 died after download firmware

[aaa]

My six nrf51822 modules died after I erase flash and download "ble bcast mesh" (with sd110 8.0.0) github.com/.../nRF51-ble-bcast-mesh

At first time, I download firmware into chip, and enter debug mode, everything is good. When I leave debug mode, add some function to firmware and try to enter debug mode, I find that jlink can't find the chip anymore.

the following message is from jlink commander:

=================================================================== SEGGER J-Link Commander V5.10d (Compiled Dec 9 2015 12:49:51) DLL version V5.10d, compiled Dec 9 2015 12:49:14

Connecting to J-Link via USB...O.K. Firmware: J-Link V9 compiled Oct 9 2015 20:34:47 Hardware version: V9.30 S/N: 59304981 License(s): GDB Emulator has Trace capability VTref = 3.409V

Type "connect" to establish a target connection, '?' for help J-Link>device nrf51822_xxac Please specify target interface: J) JTAG (Default) S) SWD TIF>S Specify target interface speed [kHz]. : 4000 kHz Speed>100 Device "NRF51822_XXAC" selected.

Found SWD-DP with ID 0x0BB11477 Found SWD-DP with ID 0x0BB11477

****** Error: DAP error while reading DP-Ctrl-Stat register. Found SWD-DP with ID 0x0BB11477 Found SWD-DP with ID 0x0BB11477 Can not connect to target. Found SWD-DP with ID 0x0BB11477 Found SWD-DP with ID 0x0BB11477

=================================================================

I try "nrfjprog.exe -e", "nrfjprog.exe --recover" and nRfgo, but don't work.

rom: 0x18000 size 0x28000 ram: 0x20002000 size 0x2000

[aaa]

After several try, another BLE module died. But I find that the problem is the first line in the main loop, "NRF_POWER->RESET = 1;"

If I mark this line, than no BLE will died. But, what is the reason? Why this single command in application region will make this chip die?

====================================================================

int main(void) { NRF_POWER->RESET = 1; /* init leds and pins / gpio_init(); NRF_GPIO->OUTSET = (1 << 4); / Enable Softdevice (including sd_ble before framework */ SOFTDEVICE_HANDLER_INIT(MESH_CLOCK_SOURCE, NULL);

[Susheel Nuguru]

When your first line in main is power reset, then i guess the debugger is not getting enough time to lock to your device. What is the output of --recover from nrfjprog?

[aaa]

maybe you are right. but I never see this problem on other cortex M series MCU.

nrfjprog result:

C:\Program Files (x86)\Nordic Semiconductor\nrf5x\bin>nrfjprog.exe -e ERROR: Cannot connect to any nRF device. Please make sure a device is ERROR: connected to the debugger and supplied.

C:\Program Files (x86)\Nordic Semiconductor\nrf5x\bin>nrfjprog.exe --recover Recovering device. this operation might take 30s. ERROR: Cannot connect to any nRF device. Please make sure a device is ERROR: connected to the debugger and supplied.

C:\Program Files (x86)\Nordic Semiconductor\nrf5x\bin>

[aaa]

finally, I know the reason. the module I used has an external reset circuit: one 10K res and one 10nF in series. I connect swclk and swdio to oscilloscope and see that the rising edge of swdio is odd. After removing the cap, I can use nrfjprog to recover those module.

[Susheel Nuguru]

Thanks for coming back to this thread and updating it with the answer. Great job.