Possible solution to reduce power consumption for nrf52832

[hoàng khánh]

Hello all,

I am currently working on a project to control a door lock and use ble_app_multirole_lesc as reference.

As measurement the current consumption is around 6.5mA when connected to another nrf52832 and advertising.When both connect to android phone and another nrf52832, the current consumption is around 5mA

Is this a normal consumption? Because it is written inside the example that this example is not optimized for power consumption

And if it is possible to reduce power consumption, how much can I reduce and how can I do it?

My device runnning on 18650 battery so power consumption is a big problem with me

Can anyone show me a already optimized example or give me a link so that I can do it by myself. I have found useful link but for nrf51 (https://devzone.nordicsemi.com/tutorials/b/hardware-and-layout/posts/nrf51-current-consumption-guide), is there any link for nrf52?

Thank you so much for your help 

[Hung Bui]

Hi Khanh, 

Constant current consumption of 5-6mA is very high. 

Which tool did you use to do measurement ? Have you tried to do measurement on our nRF5 DK or it was on your own custom board ? 

I would suggest you to test the current consumption with the ble_app_pwr_profiling (or even ble_app_hrs would also have current consumption in the range of <100uA)  to make sure you don't have other components on the board that draw exceed current. 

You can try to put the MCU to system off mode and check the current consumption, it should be ~1uA. 

[hoàng khánh]

Thanks for reply me

Since I'm not familiar with electronic and my board is a custom board, I used a VOM to measure current consumption

Now I can get the current consumption to about 2.8mA as measure by VOM.I dont really know if the way I measured was right or wrong but if compare to online power profile for nrf52832, its really a huge number

I have spent a day following the instruction in the link for nrf51 but still cant get the current consumption to uA

[hoàng khánh]

I have compiled and flash the rtc example in to my board and the result as measured by VOM is still 2.59mA (with only power led on), which is a much more higher result than expected 

Maybe the problem is my board, its a custom board using EBYTE E73 module from china but have the same schematic as nordic dev board so that I can reuse all the examples without modifying. In the specification of the module the Receiving current is 5mA and Transmitting current is 14mA, I dont really know if this is the main cause but till now that is the only thing I can think about

[Hung Bui]

I think you can try to test with a program to send the MCU to system OFF when press a button for example. You should be able to see the current drop about 4mA when the CPU goes to sleep. If you see 6.59 mA and then 2.59 mA after CPU goes to sleep, then it must be something else on the board that draws current. 

[hoàng khánh]

Thanks for suggeting me this solution

I have tried it with NRF_POWER->SYSTEMOFF = 1 without using softdevice and the current consumption is still at 2.56mA with only power led on, this maybe my board issue

[Hung Bui]

Do you see that it's higher before you call that function ? If you just let the CPU running in a loop, which current do you see ? 

[hoàng khánh]

I have 2 buttons to enable and disable system off mode, and the difference is 3mA

If I put only an empty loop inside main() the current consumption is quite high, over 12mA

Then if I put these 

__WFE();
__SEV();
__WFE();  

inside that loop then current consumption back to 2.59mA

[Hung Bui]

Hmm, the CPU current consumption should be around 4mA, there must be something really wrong if you see 12mA, unless you turn on the radio. 

I would suggest to  try testing on other board, or on a nRF52 DK for comparison.