Use lower frequency(32 KHz) to measure higher frequency(8 MHz) ticks/counts for 1 second

[SKV]

Hi all,

I want to measure higher frequency (8 MHz) count/ticks by using the lower frequency (32 KHz) is there is any method to do it or any examples.

I want to measure the count as shown in the figure.

Thanks in advance..

[SKV]

Thanks for your suggestion. I have edited the code as follows. The code runs in a while(1) loop.

void clock_init(void)
{ NRF_CLOCK->EVENTS_CTTO=0;
  NRF_CLOCK->EVENTS_DONE=0;
	NRF_CLOCK->CTIV=100;
	
	NRF_CLOCK->TASKS_CAL=1;
	printf("task cal %d\n \r", NRF_CLOCK->EVENTS_DONE);
	nrf_delay_ms(100); 
	printf("task cal %d\n \r", NRF_CLOCK->EVENTS_DONE);
//	while(NRF_CLOCK->EVENTS_DONE==0)
//	{
//		nrf_delay_ms(1);
//	}
	NRF_CLOCK->TASKS_CTSTART=1;
	int i=0;
	while(!(NRF_CLOCK->EVENTS_CTTO))
	{

		i++;

	}
	NRF_CLOCK->TASKS_CTSTOP;
	printf("Count for CTTO %d \n\r",i);

	
	
}

Wanted to cofirm if this is what you have suggested. Further i have a few questions:

1) I have given a delay of 100ms for the NRF_CLOCK-> EVENTS_DONE to become one after triggering TASKS_CAL. The code prints the value as 0 before the delay and 1 after the delay. However instead of giving the fixed delay, if i try to implement the same in a while loop (the commented lines in the code), the program stays in the loop forever, giving an impression that the EVENTS_DONE is never becoming 1. Anything wrong with this implementation.

2) I wanted a better clarity about EVENTS_DONE and EVENTS_CTTO. The time required for EVENTS_DONE=1 is the time required for calibration. What exactly the EVENTS_CTTO indicate (is it the period across which the calibration is triggered ??. Does it mean that calibration happens after every CTTO interval, however the actual time for calibration is indicated by EVENTS_DONE flag?).

3) How is the clock for EVENTS_CTTO getting counted. Is it using the low frequency clock (32kHz) or the high frequency clock (16MHz)?

Thanks,

[Jørgen Holmefjord]

1. The implementation of the while loop looks correct. Have you checked through debug if the issue is happening on the first iteration of the loop, or on second/later loop? You should not need to call the CTSTOP task: "The calibration timer is started by triggering the CTSTART task and stopped by triggering the CTSTOP task. The calibration timer will always start counting down from the value specified in CTIV and generate a CTTO timeout event when it reaches 0. The Calibration timer will stop by itself when it reaches 0." Also make sure you follow this warning from the doucmentation: "Due to limitations in the calibration timer, only one task related to calibration, that is, CAL, CTSTART and CTSTOP, can be triggered for every period of LFCLK."
2. EVENTS_CTTO indicate that the calibration timer timed out, and that it is time for a new calibration (based on what you set in CTIV register). You need to trigger the calibration task yourself when the timer times out.
3. This is an internal module in the clock peripheral. It is using the LFRC clock.