The lowest Power Consumption with RTC running on NRF52840

RE: The lowest Power Consumption with RTC running on NRF52840

Stian Røed Hafskjold — Fri, 25 Jan 2019 12:48:18 GMT

Here is a post on how to power down the RAM sections: https://devzone.nordicsemi.com/f/nordic-q-a/39396/nrf52840-ram-retention-in-system-on-mode/152628?focus=true

Here is a post on how to adjust the RAM settings: https://devzone.nordicsemi.com/tutorials/b/getting-started/posts/adjustment-of-ram-and-flash-memory

You want to change the length/size/end variable to correspond to how much RAM you have enabled.

I think you should try to stop the RTC clock to see what the current consumption is. The RTC clock itself should just consume 0.1uA, so if you get a larger current difference than this when you turn of the RTC, then the extra current is probably because of CPU interrupts when the RTC clock is timing out.

RE: The lowest Power Consumption with RTC running on NRF52840

chapman — Thu, 17 Jan 2019 05:48:26 GMT

I measured the current on P22 and it is same.

I don't know how to set the linker script and the rtc example do not user linker script.

Can you tell me how to power down some of the RAM.

According to the specification :, the current consumption is only 3.16uA

ION_RAMON_RTC System ON, full 256 kB RAM retention, wake on RTC
(running from LFRC clock) 3.16 μA

How can I reach such level? is there any setting ?

The RTC example is already set the minimum TICKS. which is 8hz. It cannot be changed to larger since the prescaler is only 12bit.

Thank you

RE: The lowest Power Consumption with RTC running on NRF52840

Stian Røed Hafskjold — Wed, 16 Jan 2019 13:10:55 GMT

Can you double check the numbers by connecting a multimeter on P22 (and cutting SB40), and powering with an external supply on P21?

What is the power consumption if you do not start the LF clk an RTC?

You can also try to power down some of the RAM (remember to adjust the linker script)

I don't remember the wake up interval in the RTC example in the SDK, but if this is fairly often then the little CPU spike that is needed when the chip wakes up will be causing the average current to go up. Can you set the interval to maybe 1 second, and measure the current in between? thanks