Custom Board Configuration

[ISN_B]

Hello,

My group has developed a board using the nRF52840 SoC and we are in the process of transitioning our firmware from the development kit to the custom board. Our firmware code is based on examples meant for the pca10056 board configuration. We have changed the board in the preprocessor definitions to BOARD_CUSTOM and created the appropriate .h file with the specifications of the custom board based on the pca10056.h file. When testing out the bluetooth capabilities of the custom board, we are using the ble_blinky example in the nRF5 SDK to see if the board is capable of advertising.

The debugger gets caught in a while loop in some assembly code occurring after this line:

ble_stack_init();

SVCALL(SD_SOFTDEVICE_ENABLE, uint32_t, sd_softdevice_enable(nrf_clock_lf_cfg_t const * p_clock_lf_cfg, nrf_fault_handler_t fault_handler));

Is there more that needs to be done besides changing the preprocessor definition to the custom board? We are unsure about how to begin debugging this issue and if it is related to software or hardware.

Thank you.

[Einar Thorsrud]

Hi,

Does your custom board have the optional 32.768 kHz crystal or not? If it does not, you need to change the SoftDevice clock configuration in sdk_config.h to use the LFRC (see this post). If not, the sd_softdevice_enable() will never return as it waits for a non-existing clock to start. 

[ISN_B]

Hello,

The board does not have that crystal. Changing the config allowed the code to progress.

However, even though the code seems to be running smoothly, I cannot locate the device on nRF Connect or any LE scanner. We are looking into if this is a hardware issue, but could something on the firmware side be causing this?

Thank you

[Einar Thorsrud]

Does everything look good from the firmware perspective? Code runs and starts to advertise etc? If not, you should look into that. If not (or in any case really) you should consider your HW, and in particular your layout and antenna tuning. If you have not done so before, I suggest you open a new ticket where you attach your schematics and layout and ask for it to be reviewed.

[ISN_B]

The firmware appears to be running and makes it past the advertising_start() function. We have already done the HW review with Nordic actually and designed the antenna according to their specifications.

Where would be a good starting point to test the hardware? If this is an antenna issue, how can we go about debugging it?

[Einar Thorsrud]

ISN_B said:

The firmware appears to be running and makes it past the advertising_start() function.

I see. If it does, and there are no errors later, then that should be good.

ISN_B said:

We have already done the HW review with Nordic actually and designed the antenna according to their specifications.

I see. But it is easy to make mistakes, so I really suggest you get your HW reviewed by us (in a separate case) if you have not doen so already. This can potentially save you a lot of time, and I strongly encourage it (we see many designs every week, and often spot issues right away that the customer may never though about).

ISN_B said:

Where would be a good starting point to test the hardware? If this is an antenna issue, how can we go about debugging it?

First of all, start by measuring the radiated power, typically using a spectrum analyzer, and see that it is within the expected range. You can also refer to this white paper on tuning.