Central connection attempt times out the peripheral connection

RE: Central connection attempt times out the peripheral connection

umr_engnr — Mon, 19 Oct 2026 18:42:45 GMT

Hi Edvin, we will work on getting you the sniffer traces.

RE: Central connection attempt times out the peripheral connection

Edvin — Sun, 18 Oct 2026 19:40:27 GMT

Hello,

I am sorry for the late reply.

Is it possible to share some sniffer trace of the connection that is dropping out? Also, if possible, also a sniffer trace of the other connection at the same time (two different traces).

Best regards,

Edvin

RE: Central connection attempt times out the peripheral connection

umr_engnr — Wed, 14 Oct 2026 14:48:59 GMT

We were hopeful the PPM change fixed the issue, but one of the teams was able to reproduce this issue even with the change. We are attempting find a better way to reproduce the issue and try to isolate what condition causes it to occur. We are open to other suggestions why the connection would appear to just go away.

RE: Central connection attempt times out the peripheral connection

Mariano Goluboff — Wed, 30 Sep 2026 16:32:44 GMT

Only the net core one is used for controlling the window widening. I can't find anywhere in the code of the app core where the value is used.

Mariano

RE: Central connection attempt times out the peripheral connection

umr_engnr — Tue, 29 Sep 2026 03:47:51 GMT

I noticed that the app core code was set to 20PPM and the net core code was set to 50PPM. Are both of these values used or just the one in the net core code?

I bumped them both to 150PPM, which matches what we expect over our full operating temperature range, and initial testing looks promising, but more testing is needed.

RE: Central connection attempt times out the peripheral connection

Mariano Goluboff — Wed, 23 Sep 2026 12:36:20 GMT

An incorrect CLOCK_CONTROL_NRF_ACCURACY_PPM can definitely lead to the behavior of the peripheral connection being disconnected due to timeout. This is if you miss enough connection intervals. With an incorrect PPM, you can miss the connection intervals because the window was widened enough.

Setting the accuracy to the expected values over your use case should reduce the missed connection intervals, but it does have some trade offs. See this section of the DevAcademy:

https://academy.nordicsemi.com/courses/designing-low-power-bluetooth-le-products/lessons/lesson-5-soc-specific-power-optimization-i/topic/clock-sources/#accuracy

Mariano

RE: Central connection attempt times out the peripheral connection

umr_engnr — Wed, 23 Sep 2026 01:04:11 GMT

Hi Edvin,

These are two distinct connections. First the nRF5340 in question forms acts as the peripheral in a connection to another nRF5340. This peripheral connection can be active for 30 minutes or more without problems and with the same device making and unmaking multiple new central connections. Then later it attempts again to create a connection to an nRF52840 but fails.

We have not tried to reproduce this on dev kits yet. To me this seems very timing dependent because generally it takes quite a while to reproduce but sometimes it happens almost immediately. I will look at setting up a test that just continually makes central connections while a peripheral connection is active to see if that aids in reproduction.

We are using NCS 2.6.6.

RE: Central connection attempt times out the peripheral connection

Edvin — Tue, 22 Sep 2026 12:28:37 GMT

Hello,

Just to clarify:

[quote user=""]We ran into a strange issue where this device attempted to form a central connection, but after 1 second (CONFIG_BT_CREATE_CONN_TIMEOUT) a connection failure was reported with reason 0x02 (BT_HCI_ERR_UNKNOWN_CONN_ID). Three seconds later (4 seconds after the initial connection attempt which is our supervisor timeout), we got notified that our peripheral connection disconnected with reason 0x08 (BT_HCI_ERR_CONN_TIMEOUT).[/quote]

These two connections, when you say central and peripheral. Are these two different connections from the same device? First, the central connection fails to establish with BT_HCI_ERR_UNKNOWN_CONN_ID, and then 3 seconds later, another connection from that same device, where the nRF53 is the peripheral times out? Or is the peripheral connection in this context the other device that the central tried to connect to?

I know you said that this was not easy to reproduce. Is it possible for me to reproduce it on a DK?

Perhaps it reproduces easier if you turn down the TX power of the peripheral device that the central tries to connect to in the first place?

What NCS version are you using when you are seeing this?

Best regards,

Edvin