nrf52DK higher power consumption after transmission

[cpc5038]

Hey guys,

I have been fighting with this for a couple of days and I am hoping someone can give me some insights as to what might be going on.

We currently use the nRF24L01P, and I am trying to evaluate the nRF52 as our potential replacement because it is backwards compatible with our current hardware via ESB.  Our applications are very low power, so I need to prove that it doesn't break our power budget, but I am running into a road block.

I have a relatively simple application that basically does the following:

1) Setup and run app_timer for 1 second intervals

2) After 4 seconds elapses, call esb_init(), then setup a payload and call nrf_esb_write_payload()

3) In nrf_esb_event_handler(), call nrf_esb_disable() 

What I am finding is that prior to the transmission I get about 2uA of power consumption (on average).  Then after the transmission, I get 7-8uA average.  Looking on a scope, the average goes up because the spikes caused by Refresh Mode of the DCDC/LDO are about 4 times more frequent.

My question is why is this the case?  After I disable the radio, shouldn't it go back to the lower power mode it was in prior to the transmission?  From what I have read about Refresh Mode, it sounds like it is periodically charging the caps.  Why does it need to charge them more frequency after a single transmission has occurred?  This rate does not change after the transmissions (in other words, it doesn't settle back to the original rate).  I also only see this problem if i call the nrf_esb_write_payload() function.  If i just call esb_init() but don't transmit anything, the refresh rate does not change (it stays at the lower rate).

Thanks!

-Chris

[Stian Røed Hafskjold]

Hi, sorry for the late response. I think you're seeing the same issue as in this thread: https://devzone.nordicsemi.com/f/nordic-q-a/31058/higher-current-draw-in-system-on-low-power-mode-after-esb-ptx-than-after-esb-prx/123074#123074

Please see Torbjørn's answer about shutting down the timer before going to idle.

[cpc5038]

That was the problem! I found that I had to insert the following two lines:


NRF_ESB_SYS_TIMER->TASKS_SHUTDOWN = 1;
NRF_ESB_BUGFIX_TIMER->TASKS_SHUTDOWN = 1;

I had to also shutdown the ESB_BUGFIX_TIMER in order to get the power consumption back down.  I am using SDK 15.0.0 so I'm guessing that bugfix timer was not in the SDK in previous versions?

Thanks again!

-Chris

[Stian Røed Hafskjold]

The bugfix timer is a workaround for this errata: http://infocenter.nordicsemi.com/topic/com.nordic.infocenter.nrf52832.Rev1.errata/anomaly_832_102.html

The workaround was introduced in SDK 14.0.0.

The errata is fixed in rev.2 of the nRF52832.

Thanks for letting us know that you need to shutdown this timer as well to get lower consumption. I will add this to the answer in the other thread.