phase noise agilent dicriminator
anyone kowns the relationship between FM deviation and phase noise? What mean is FM deviation data during testing PLL using spectrum analyzer? If i have FM deviation data, how could i get phase noise data?
thanks a lot!
Phase Noise to Residual FM relation
They are related, but different.
If you have a frequency discriminator feeding a spectrum analyzer as a measuring device, and you modualate a tone onto the carrier with a constant FM deviation, then as you change that tone's frequency a straight horizontal line gets drawn on the spectrum anlyzer face.
On the other hand if you use the same measurement method, and instead modulate that tone onto the carrier by phase modulation, then as you sweep the tone frequency the spectrum analyzer shows a straight line with a 20 dB/Decade down slope as the tone frequency gets higher.
I do not remember the equations of motion here, but search agilent or Rhode and Shwartz for a delay line discriminator phase noise measurement system, and you will be satisfied with more than enough equations.
Remember, for a delay line discriminator, at high enough modulation frequencies the output goes to zero (like a sinc function in modulation frequency).
Thanks a lot!
to vfone, what mean is F1 and F2 in the equation? thanks
F1 is the lowest modulation frequency and F2 is the highest modulation frequency.
In this FM example the lowest modulation frequency is 50 Hz and the highest is 15 kHz.
hi, thanks for your reply
but I want to know why they give me the FM deviation data, what difference exits between FM deviation and phase noise? I mean in what situation we should test FM deviation, in what situation we could get phase noise directly! Could anyone introduce something about how to test FM deviation?
thanks! I have little experience with testing.
They are related in that FM noise is the integral of phase noise. One might be more important to you than the other depending on what type of modulation you are using. Also, since FM noise is an integral of phase noise, there are very many different shaped phase noise curves that can integrate into the same FM residual noise.
Another issue on PLL noise meas:
I have noticed that usually there is a small difference in phase noise plots of PLLs if I use a spectrum analizer (for Example an R&S with FSK4 option) or a phase noise meter based on discriminator method.
The last one usually gives a better phase noise (1-2 dB).
I'm not talking about 'high-end' noise meas (such as -155 dBc/Hz, for example) for which the difference is clear.
I'm talking about the 'in band' phase noise, for example in the range 1kHz-100kHz, -70dBc/Hz to -120dBc/Hz.
My opinion is that spectrum analizer measures the AM noise as well, while discriminator cuts away AM noise giving a better results.
Any opinion? Agree? Docs?
Thanks in advance.
Mazz
Is the difference more like 3 dB? If so, you are not programming one of the machines properly. Read the manual.
Close to the carrier, if the setups of both methods are well calibrated there should be no difference between measurements.
Discriminator method is useful in measuring lower noise at higher offset frequencies.
Biff: no 3dB difference. No SSB to DSB mistake.
The difference is always smaller, sometimes negligible, and it depends on DUT.
vfone: I agree. There should be no difference. But I have seen it.
Suppose that and AM noise is on a perfect pure CW signal (I mean with no phase noise).
The spectrum analizer is not able to distinguish this AM noise from PM noise.
The discriminator method, at mixer output, will cut the AM noise, showing a clean CW.
Agree?
Mazz