Default coded PHY (S=2 vs S=8) on nRF52833 for advertising and connections under ncs 3.1.0

[longo92]

Hi,

I am working with an nRF52833 using the SoftDevice Controller in nRF Connect SDK 3.1.0, and I have a question about LE Coded PHY (long range).

I understand that the device supports coded PHY, but I found conflicting information regarding whether both S=2 (500 kbps) and S=8 (125 kbps) are supported.

In some DevZone posts I saw that S=2 is not supported on the nRF52833 (only S=8).

My questions are:

1. When I enable coded PHY advertising on the nRF52833, which coding (S=2 or S=8) is used by default?

2. When establishing a connection on coded PHY, which coding is negotiated by default?

(Supposing that the central does not specify a preference between S2/S8)

3. Is it possible to explicitly request S=2 on the nRF52833, or is only S=8 supported?

I want to make sure I understand the default behavior both for advertising packets and for connections.

Thanks in advance!

[Einar Thorsrud]

Hi,

1. When I enable coded PHY advertising on the nRF52833, which coding (S=2 or S=8) is used by default?

Using coded PHY, S=8 is the defaut.

2. When establishing a connection on coded PHY, which coding is negotiated by default?

There is no negotiation on this part. All devices supporting coded phy need to suppor both, and the header is always transmited as S=8, and this tells if the rest of the packet is S=8 or S=2 encoded.

3. Is it possible to explicitly request S=2 on the nRF52833, or is only S=8 supported?

for recent nRF Connect SDK versions, both S=8 and S=2 are supported. You can specify S=2 using the BT_CONN_LE_PHY_OPT_CODED_S2 option. For a sample, please take a look at this post. However, you gain very little range using S=2, so this has little practical use.

[longo92]

Thank you for the detailed answer.

I have one point I’d like to clarify: you mentioned “however, you gain very little range using S=2, so this has little practical use.”
From my understanding, when using S=2 there is a gain in bitrate but a reduction in maximum distance compared to S=8. Moreover, from a power consumption perspective, S=2 should consume less since the transmission time is shorter. Am I correct?

[Einar Thorsrud]

Hi,

You are correct. I was not clear, but what I wanted to say was that you gain very little range using S=2 compared to using the normal 1 Mbps PHY. You get the downside of reduced bitrate and increased power consupmtion per transfered unit of information, but not a significant gain in range. So in practice, the choise whoudl be between 1 (or 2) Mbps PHY and S=8 Coded PHY.