Dear Alireza Kiyaei

You measure as you define, this is the concept.

So you have to know your free running frequency which is here 30Ghz.

Then you want to lock it initially at 15 GHz.

At first your oscillator must operate at 30GHz.

Then you inject signal with the suitable waveform and frequency starting from say 10 to 20GHz or even you can start with smaller rage say 14 to16GHz.

you must divide divide the output frequency by 2 and compare it by the input frequency. The two must be exactly the same. If there is no effect, the output frequency remains at its free running value.

Best wishes

Dear Alireza Kiyaei

You measure as you define, this is the concept.

So you have to know your free running frequency which is here 30Ghz.

Then you want to lock it initially at 15 GHz.

At first your oscillator must operate at 30GHz.

Then you inject signal with the suitable waveform and frequency starting from say 10 to 20GHz or even you can start with smaller rage say 14 to16GHz.

you must divide divide the output frequency by 2 and compare it by the input frequency. The two must be exactly the same. If there is no effect, the output frequency remains at its free running value.

Best wishes

dear Abdelhalim abdelnaby Zekry thank you for your answer

I had the same idea in mind but the problem is as you can see in the figure above that there is a spur at approximately 23.9 GHz which shouldn`t be there and when I shift input frequency for example at 14GHz input this spur will occur again at a lower frequency relative to 28GHz output frequency

now I'm concluding that the problem is with Cadence DFT function and I should calculate DFT without any windowing so I should extract transient plot and calculate DFT in Matlab using fft or fftshift function

and here problem strikes again because at such a small step there will be millions of data points which seems I can't extract correctly from cadence.

Dear Alireza,

I would like to propose that you use a frequency divider at the output of the oscillator so if the the oscillator in the injection tracking mode this frequency and the injecting frequency will be the same. If you down mix the two signals by a multiplier then you will get a dc or zero output meaning tracking, otherwise you will get a mixing waveform.

Best wishes o remove any spur by filtering it. Here you can use notch filters in form of series resonance circuit shunting the output of the oscillator.

Best wishes