In VNA help，SA option，how to understand the words below ?
"It ensures the Nyquist images of the signal tones in the IF bandwidth are not falling back on top of real signal frequencies. To be able to enable Nyquist protection, the tone spacing of the coherent signal cannot be an integer divider of the ADC sampling clock (100MHz for a PNA). "
From my previous knowledge, is this setting guaranteed by anti-aliasing filters?
tks for help!
It's mostly true the anti aliasing filters prevent this, but they are not perfect. In the "wideband ADC filter" mode of the PNA, the filter rolloff is not fast enough to completely eliminate the nyquist images (they are down 60 dBc or so). So when we do coherence mode, we know exactly where they will be located and set the LO frequency for the SA in such a way that they do not land on top of the real signal. Since we can place them in-between tones, they don't get measured as part of an ACPR or NPR error and are eliminated. It is a little complicated. This doesn't have a lot of effect on normal SA functions, but when you are a testing a modulated waveform, it can have wideband energy that shows up due to this aliasing effect. You almost never see it for normal SA measurements where the drives are not wideband modulated. If you use the 11 MHz filter (the narrowband ADC fllter) the alias images are down more than 100 dB.
It's mostly true the anti aliasing filters prevent this, but they are not perfect. In the "wideband ADC filter" mode of the PNA, the filter rolloff is not fast enough to completely eliminate the nyquist images (they are down 60 dBc or so). So when we do coherence mode, we know exactly where they will be located and set the LO frequency for the SA in such a way that they do not land on top of the real signal. Since we can place them in-between tones, they don't get measured as part of an ACPR or NPR error and are eliminated. It is a little complicated. This doesn't have a lot of effect on normal SA functions, but when you are a testing a modulated waveform, it can have wideband energy that shows up due to this aliasing effect. You almost never see it for normal SA measurements where the drives are not wideband modulated. If you use the 11 MHz filter (the narrowband ADC fllter) the alias images are down more than 100 dB.