nRF52840 stuck after programming

[Syl20D]

Hello everyone,

I'm currently prototyping a custom board with nrf52840, here the schematics:

I use the config 3 of the reference schematics.

I misunderstood the  PSELRESET register, believing I can change the gpio which have the nReset function but I can't so I don't have any reset Pin 

Now  I want to test firmware written on dev board on the my custom board, for this I use a nrf9160DK as JLINK probe. First programming is successfull and I can recover/reflash/read register as long I didn't unplug/plug the board ( Because I don't have reset pin I assume my firmware don't run at this stage)

After reboot I encounter a -21 error from JLink probe and I can't recover or read register or flash anything. The firmware doesn't seems to launch

First question, do you think I can recover these board without reset or there are lost forever? Maybe with the power on reset timing? 

Because I have a limited number of prototype I can't test many things to resolve this issue so with few researches I have found:

- The DC/DC enable config can lock the board but I have disabled it in the conf file of the project. I use the mcuboot and merged binary, so I'm wondering if the configuration for the mcuboot can be different?

I have measured the DEC pin, I have 0.5V on DEC4, 1.8V on DEC3, 0V on DEC5 and DECUSB

- The nrfjprog set the uicr register to use the nRST function, and because I haven't any pullup on the reset pin, it can be stuck on reset.

Do you know if there is internal pullup if nRST is activate? I can remove the line of the runner script for resolve and remove the uicr configuration from dts file but maybe there is another program or configuration that set the nRST function

Do you have any other idea of what can be the issue? Did you see some hardware issues?

Thanks you in advance

[Kenneth]

Hello,

The following information should ONLY be used to get the board up and running again, and not in a product. It's very important that you don't exceed the voltages mentioned, since this will very likely DAMAGE the device:

You can directly power Dec4 with 1.3V. When you have wrong setting for DCDC, then the failing DCDC regulator fails to output 1.3V to DEC4 which is the main supply to the nRF chip. If you supply 1.3V directly to DEC4 then it should bring the chip out of reset mode and you can reflash the firmware with DCDC disabled. 

It's possible to connect the reset pin directly to VDD if you have by accident enabled it and driven it low internally or externally. This should bring the chip out of reset mode and you can reflash the firmware with pin reset disabled. 

Kenneth