The control of executing radio notifications on nrf51822

RE: The control of executing radio notifications on nrf51822

Øyvind Karlsen — Tue, 14 Jun 2016 13:50:55 GMT

Do your I2C functions in the topmost code example use the HFCLK? I am having a hard time seeing how you would be able to time an event that is a maximum of 5500 µs ~= 5.5 ms long by eye. Either way if the measurements take that long to complete they cannot be finished wholly in one radio notification, could you provide more information about your sensor?

RE: The control of executing radio notifications on nrf51822

AndrzejB — Tue, 14 Jun 2016 13:27:47 GMT

I use radio notifications because additional timer will have a negative impact on the energy efficiency.
I timed the code only in organoleptic way, without any dedicated equipment (only stopwatch)
I2C in main is executing in similar time as in timer_timeout_handler, that is short blink of LED.

Sorry for non professional description of the time period, as soon as I have access to logic analyzer I will precise this. But even now it is obvious that execution times of code in radio notification are changing between next cycles and in some of them this times are longer than 1 second, even without i2c measurements

RE: The control of executing radio notifications on nrf51822

Øyvind Karlsen — Tue, 14 Jun 2016 12:52:31 GMT

Hi,

Have you timed the code by probing the LED with an oscilloscope or signal analyzer? How does the I2C measurement perform when run from main?

Is there any reason you have to use notifications, is regular measurements with a timer an alternative?