nRF54L15 BLE RX current comsumption

[loquat]

I measured the power consumption of the BLE RX using the nRF54L15.(nRF54L15-DK v1.0.0)
3.0V operation
apps: scan_adv (delete advertising)
west build -b nrf54l15dk/nrf54l15/cpuapp
proj.conf
CONFIG_BT=y
CONFIG_BT_BROADCASTER=y
CONFIG_BT_OBSERVER=y
CONFIG_LOG=n
CONFIG_SERIAL=n

Result: The average current consumption was 3.73mA.
This is 330uA higher than the 3.4mA stated in the datasheet.
Is this correct behavior? Is further processing required to reduce it to 3.4mA?

[Simonr]

Hi

To be clear. With just the radio running in RX, the current consumption is indeed drawing 3.4mA.

However, the SoftDevice controller, which is what controls the radio peripheral, enables constant latency mode when running, which draws the additional 3.7mA.

So, the datasheet is correct, but the datasheet doesn't cover how the SoftDevice controller operates, which is where you're confused I believe. 

The last snippet you added is an example of the system in idle mode (no radio or any other peripherals running), with the constant latency mode enabled.

Best regards,

Simon

[Emil Lenngren]

In my opinion it is kind of "false advertising" or whatever you call it to say 3.4 mA in RX mode for the full system consumption (compared to only the radio current usage isolated as in docs.nordicsemi.com/.../parameters.radio_rx_current.html) because such an operation is practically useless without having the MCU domain turned on (e.g. by activating constant latency) while reception is ongoing; otherwise you might receive corrupted packets.

Please see my post here: https://devzone.nordicsemi.com/f/nordic-q-a/117339/radio-tx-doesn-t-work-unless-mcu-domain-is-active-during-ramp-up/520046 where I came to those conclusions. I am still awaiting feedback in that thread regarding if the errata should be updated to state this fact (i.e. the workaround is necessary for not only TX but also for RX).