Maximum RF Power Level for S-Parameter Measurement

Most of the time I have seen S-Parameters measured at no greater than -20dBm power levels. Are there cases in which they are measured at higher power levels? If so, what are the reasons/justifications?

S parameters are defined as LINEAR parameters, so independent from power level.
Measurements have to be done at power levels that put your circuit in linear conditions.
So for passive circuits you can easily choose higher power levels.
An higher input power can be used if you need. for example to extend the dinamic range of your measurement (as noise floor is always the same).

For active circuits, for example a transistor, drive level has to be low enough not to bring the transistor out from its linear region.


I hope it can help.

Mazz

Turn the question around: As said before S-parameter are linear measurements. So the correct question is what is the minimum amound of RF power that still results in good measurements without contaminating noise. You could do this experimentally by playing with the VNA power setting and watching how the measurement evolves.

This test may fail if the VNA dynamic range fails with the calibration.