start-up current nRF52810

[buja]

Dear Team,

chip- nrF52810

I notice a startup current sequence existing for 26ms and the average value is 600 uA. This sequence is present when waking up from systemOFF as well. 

I would like to know if this behavior is common ? Can we avoid this power up/wake up current sequence, so that we can save some energy ?

[Stian Røed Hafskjold]

Hi, I was able to reproduce the same sequence here. It occurs after the LFXO clock (NRF_SDH_CLOCK_LF_SRC 1) has ramped up, which takes about 250-400 ms, and is caused by the softdevice enable call. What happens is that the random number generator is running for about 26 ms generating 194 bytes of random data which is later used by the softdevice. This behavior is not configurable by the application.

[buja]

Hi Sitan,

Thank you for your reply. 

Is there any way we can get rid off this current consumption (by changing clock source or anything like that) ? Also , can I ask you that if it is necessary to enable Softdevice each time after waking up from systemOFF ?

[Stian Røed Hafskjold]

Hi, this is internal to the softdevice and not possible to change. The chip wakes up in reset state after system OFF, so if you are going to use the softdevice you will have to re-enable it each time. If you are not using BLE every time you wake up from system OFF, you can skip the softdevice enabling, and wait until you actually need to use BLE.

[buja]

Thank you for your reply

[Stian Røed Hafskjold]

FYI, if you are using NCS instead of the nRF5 SDK, random number generation is handled outside of the Softdevice controller, and the buffer sizes are configurable.

[buja]

Hi Sitan,

Thank you for the info.

I am new to nRF annd currently using nRF5 SDK. Could you please let me know how this is different NCS. Also I see from the link provided by you that  "This version of the nRF Connect SDK supports prototyping and evaluation, but should not be used for production and deployment in end products."

Could you please clarify this for me please ?