Is anybody knows about Timer1 accuracy

Posted on September 25, 2012 at 18:28

It depends.

It's granularity will be based on the TIMCLK used, presumably 168 MHz or 84 MHz.

It's accuracy is going to depend on the stability/value of the source clocks, and the stability of the clocks synthesized from that. With crystals you'd need to consider the aging and temperature.

The timer input has a synchronizer circuit in front of it, that is going to exhibit nyquist sampling issues, add delay (phase shift), and you're going to observe some amount of jitter, and perhaps hysteresis, depending on the phase relationship between the input signal and the sampling clock, and the setup/hold time of the flip-flops.

Posted on September 27, 2012 at 02:42

Hi Clive1

Actually timer is super accurate and it's accuracy is +/- 5.92 ns(1/168Mhz) 

But my fault was I capturing pulse with 5V amplitude and this is not correct because capturing unit and edge detector unit starting counting when it's reach the middle of edge(with VCC 3 volt it's 1.5 volt) and when the signal is more than 3 volt the trigger trig in 1.5 volt and 1.5/5 volt is  30% of signal and for that reason it's show the wrong number.

Thank you 

I M A N