Something simple

RE: Something simple

Håkon Alseth — Tue, 10 Jun 2014 15:06:14 GMT

The app_gpiote.c only uses the PORT event of the GPIOTE peripheral, so it will not cause any high current consumption. It is correct that it is the IN/OUT registers (set in NRF_GPIOTE->CONFIG[] registers) that cause higher current consumption (due to them requiring the 16M clocktree, which gives you ~1mA added current consumption)

RE: Something simple

paul-b — Sun, 08 Jun 2014 03:14:17 GMT

Hi Hakon, thanks for the quick response. I'm still a little unsure about what particular setting causes the extra power to be consumed as in PAN-39. Judging by this thread (devzone.nordicsemi.com/.../), it appears that the problematic power consumption during Sleep mode is specific to setting the NRF_GPIOTE->CONFIG register. Is that correct? Would any of the APIs in app_gpiote.c,h trigger the problem? Thanks!

RE: Something simple

Håkon Alseth — Fri, 06 Jun 2014 09:35:24 GMT

Hi Paul,

We're still working on restoring older links, sorry for the inconvenience. The GPIOTE port event is available here: github.com/.../system-on-wakeup-on-gpio

And the RTC trigger ADC example is available here: github.com/.../rtc0-triggering-adc-sample

RE: Something simple

paul-b — Thu, 05 Jun 2014 18:20:00 GMT

Hi Hakon, Nordic, I know this is a very old thread, but this example code could be very useful to workaround the GPIO power issue documented in the PAN. Can you please restore the C and Zip file referenced here? Thanks!

RE: Something simple

sd — Wed, 24 Jul 2013 13:23:09 GMT

OK, got it to work. Thanks for the help!

RE: Something simple

sd — Wed, 24 Jul 2013 08:34:38 GMT

I suspect that with the framework in the HRS example doing what I want should be very straight froward, but I'm not able to make it work yet.

RE: Something simple

sd — Wed, 24 Jul 2013 08:25:01 GMT

Thanks! I think you're right (looking at your example). Since I'm trying to integrate this into the heart rate monitor example, I'm now looking into getting this to work with the existing GPIOTE IRQHandler... it's not that easy tracing all the events and calls ;)

RE: Something simple

Håkon Alseth — Wed, 24 Jul 2013 07:57:01 GMT

Hi,

I think you are over-engineering your problem a bit. Its not possible to have the GPIOTE/PPI to start a function call in such a "direct way". From your question, it seems like you want to generate an interrupt based on pin-change, and then run a function in code. This can be done using the GPIOTE PORT event. I am attaching an example on how to do this.

However, if you'd like to trigger a "chain reaction", for instance that pin change starts the ADC or similar, you can do this with PPI/GPIOTE. I am also attaching a example that triggers ADC each RTC TICK.

BR Håkon

rtc_trigger_adc.c

GPIOTE_Port_event.zip