ESB Device address from hardware ID.

[Andre]

I have previously developed on the nRF24LU1+.  This device would have a CHIPID in the InfoPage, This ID would have guaranteed it does not violate the requirements for ESB addresses. On the rRF5x, I have NRF_FICR->DEVICEID[2] and NRF_FICR->DEVICEADDR[2]. Can those values be trusted to not violate the ESB address requirements as well, or do I have to check for and replace 0x55 or 0xAA at the beginning?

[Håkon Alseth]

Hi,

 

The DEVICEADDR has checks for 0, 0xFF, 0x55, 0xAA in the first byte going on-air, but it has it for little-endian. If you are to use big endian packet format, you'll have to check the corresponding byte for the four non-recommended values.

You can set the endianness of the RF payload in the nRF5-series devices (NRF_RADIO->PCNF1 register, field ENDIAN), so it is important that you check for valid address based on your configured endianness.

 

PS: Endianness used in bluetooth LE is little endian.

 

Best regards,

Håkon

[Andre]

Thanks for your quick reply,

Looking at nrf_esb.c, function update_rf_payload_format_esb_dpl, ESB is configured to use big endian.

So, if I were to use

uint32_t address = __REV(NRF_FICR->DEVICEADDR[0]);
nrf_esb_set_base_address_1(   (uint8_t*)&address );

it would always be valid?

[Håkon Alseth]

Hi Andre,

 

That is correct. This shall work for BE format.

 

Cheers,

Håkon