RE: gazell anomaly 107 in nRF52840

ovrebekk — Tue, 28 May 2019 10:48:49 GMT

In general using an address consisting entirely of zeroes is not a good idea, and will lead to increased packet loss on any of our radios.

The nRF52 series behaves even worse in this regard, and devices affected by anomaly 107 will not get any packets through.

While anomaly 107 is fixed in later revisions of the nRF52832, I don't think this will help for an address with only zeroes, as this address is affected by a slightly different issue. Essentially 0x0000000000 is the reset value of the internal address compare register, and if you set the RF address to the same value then you will get an immediate address match once you enable the radio.

In the nRF24L documentation we used to have the following recommendation regarding addresses:

"Note: Addresses where the level shifts only one time (that is, 000FFFFFFF) can often be detected in
noise and can give a false detection, which may give a raised Packet Error Rate. Addresses
as a continuation of the preamble (hi-low toggling) also raises the Packet Error Rate."

I would strongly suggest using an address following this recommendation, to avoid issues with RF communication, and if you do you should also be able to configure the Gazell library accordingly for the nRF52840.

Best regards
Torbjørn