Hello All,

I am now simulating the IIP2/IIP3 of my front-end (LNA+Mixer) circuits.
Two situations:
1. w/o divider: simulation is simple and fast, just say set the LO frequency (5G), and input frequency (5.002G) with amplitude -50dBm, also set input PAC amplitude to -50dBm. Use QPSS and QPAC to get the fundamental and IM2 term. (Ref: " Accurate and rapid measurement of IP2 and IP3 " from this website)
2. But w/ divider, that is, my LO now is 10G, the setup for the input keeps the same. Then there is a problem when using QPSS and QPAC because I cannot manually set the beat frequency to what I need (here it is 5G), then the simulation will not converge.  I can still use PSS/PAC, but now the max. beat frequency is only 2MHz, in order to get the right LO freq. (5G), I have to put at least 2500 in the harmonics. The overall simulation is too long.

Is there any way to run the linearity simulation with QPSS/QPAC, also with divider in the circuit (case 2 above) ?

Thank you so much for the help.  

Yy

Thank you so much.

Yy