Hello Nordic Team,
I'm developing a ranging application on a u-blox NORA-B1 module (which contains an nRF5340) and have encountered a stability issue with the Nordic Distance Measurement (NDM) library when using the synthesized low-frequency clock (LFSYNT).
When the LFCLK is configured to use the synthesized source (LFSYNT), the distance measurement example starts correctly, but the peer device stops measuring after a random period of 1 to 5 minutes. There are no explicit error logs or faults when this occurs; the ranging process simply halts.
It's important to note that our NORA-B1 module uses a 32 MHz external crystal to drive the high-frequency clock (HFCLK), so the LFSYNT source should be highly accurate (±30 ppm as per the documentation).
If we change the configuration to source the LFCLK from an external 32.768 kHz crystal (LFXO), the distance measurement application runs perfectly and remains stable indefinitely.
Is an external LFXO a mandatory requirement for the NDM library due to strict timing needs that LFSYNT cannot fulfill, even when derived from a crystal-based HFCLK? Or are we potentially missing a configuration step that is causing this instability?
Below are the Kconfig settings we are using to enable the synthesized clock:
# Clock Configuration
CONFIG_CLOCK_CONTROL_NRF_K32SRC_RC=n
CONFIG_CLOCK_CONTROL_NRF_K32SRC_SYNTH=y
CONFIG_CLOCK_CONTROL_NRF_K32SRC_XTAL=n
CONFIG_SOC_ENABLE_LFXO=n
Any insight into this behavior would be greatly appreciated.
Thanks in advance for your support!
