I am curious, what is your use case for measuring P-P with a spectrum
analyzer?
Just like you mentioned yourself:
P-P is easy to measure and not dependent on waveform purity but does
not translate easily to power in a scope if not a pure sine wave.
With a spectrum analyzer you always have a sine wave at the
fundamental frequency of a complex wave.
Main reason: Vpp is easy to measure (and see) on scopes.
A simple use case is looking at the output of a generated signal at a
desired frequency and seeing if the different instruments agree and get
a feeling how they correlate, visually. Some simple signal generator
chips specify their output in terms of Vpeak/Vpp. Yes, you could convert
it to dBm, but sometimes Vpp is just more "direct".
Similarly, you might want to look at an oscillator and see quickly how
close its Vpp is compared to the fundamental frequency shown on the
spectrum analyzer. You can then dig deeper to find out the causes, such
as overlayed noise/distortions etc, and see what other frequency
components exist. But first you can have a first high level impression
how much your signal reaches its expected output level.
Or you might have a sine wave that is split up with a directional
coupler, and compare outputs easily on the scope and on the spectrum
analyzer (and their respective power levels). I don't need to
characterize the full frequency spectrum/harmonics/other frequency
components always. Often simple sine waves are enough for many experiments.
