Frequncy measurement at 78MHz using R series FPGA

[tintin_99]

Hi 

I can compile my FPGA code at 140MHz and I want the measure frequncy of signal around 77.8Mhz using 2 counters by counting rising edges in a given time , let’s say if I set my gate to 1s gate time then I should be able to count 77.8Mhz +- 1 Hz . Is my calculation correct ? Can I measure 77.8 MHz using this methods with 1 hz error using 140 MHz clock ? 

[GerdW]

Hi tintin,

---

@tintin_99 wrote:

Can I measure 77.8 MHz using this methods with 1 hz error using 140 MHz clock ? 

---

Have you heard about Mr. Nyquist?

When you run your FPGA loop at 140MHz you can detect changes in a digital signal with a frequency of 70MHz: first iteration you get a FALSE, the next iteration gives a TRUE. How do you want to detect faster signal changes?

Which hardware do you use? How do you measure the digital input? Does your hardware even allow detecting/measuring digital signals at this samplerate?

[Henrik_Volkers]

Isn't there the 'trick' that parts of the FPGA can run faster than the clock?

How about a prescaler dividing the input by 4/8/16/...

create a fast clocked timer , fetch the timer value each prescaler slope 

now you have timestamped prescaler ticks and you can apply a slope -> frequency

depending on the timer clock you get your 1 Hz resolution much faster 🙂

and you can use more than 2 timestamps and fit the slope ...

.. make an adjucated guess of the uncertainty ...

later you wonder about stuff like phase noise 😉