This post is older than 2 years and might not be relevant anymore
More Info: Consider searching for newer posts

Achieving low power consumption for BLE advertising

We are developing a new product that requires extremely low power consumption with a target to run advertising intervals of between 2 to 10 seconds.

We are using the nRF52805 with S112 softdevice

The power-profiler states estimate consumption of < 10uA at 2 second interval - we are using the nRF52810 as a proxy for the 805 estimates.

Our testing:

1. All external circuitry connected, no BLE, all peripheries disabled (except necessary GPIO), RAM on and waiting for GPIO interrupt  approx = 10 uA

2. App_timer initialized without BLE = 250uA

3. App_timer + BLE = 400uA 

Questions:

1. Does the power-profiler estimate take into account the timer needed to time the advertising interval?

          - Is this additional or included in the power profiler estimate?

2. Do we need to use the app_timer for timing advertising? 

3. How do we achieve the estimated power consumption?

Parents
  • Hello,

    1. The power profiler estimates include the LF clock source pluss RTC0 used internally by the Softdevice, but not RTC1 used by the app timer. However, the run current for an RTC instance is typically only around 0.1 uA so it doesn't explain why you measure 250 uA on average.

    2. No, it shouldn't be neccesary. The Softdevice will run advertising in the background at the interval you have specified in the application before you called "advertising start"

    3. The measurements results should have been close to the estimates given by the online power profiler. Are you using the Nordic power profiler kit for measurement? It would be interesting to know if the floor current between advertising and timer events match the typical sleep current given in the product spec: Sleep.

    Would you be able share the code you are using for this test? Maybe I will be able to see what the problem is then.

    Best regards,

    Vidar

  • Thank you for your quick response and support - the issue was an incorrectly set clock source - using LFSYNT instead of LFRC.  We are now getting 1.5uA between advertising events.

Reply Children
Related