I suggest to simulate the noise with cadence SPECTRE RF, instead of HSPICE.
Because PLL consists of VCO, which is a non-linear component. Frequency divider is also non-linear. For non-linear circuit, SPICE doesn't simulate well. Instead, you should run SPECTRE .pnoise analyse.
I suggest to simulate the noise with cadence SPECTRE RF, instead of HSPICE.
Because PLL consists of VCO, which is a non-linear component. Frequency divider is also non-linear. For non-linear circuit, SPICE doesn't simulate well. Instead, you should run SPECTRE .pnoise analyse.
I know that SPICE cannot predict noise directly in circuits having a quiescent operating point, like the PLL, but that is what I want to know wether there is some other way of doing the same using HSPICE.
I have a constraint of not having SPECTRE RF, which can directly run the PSS analysis, and am stuck to HSPICE only, so wanted to know if HSPICE also has some feature like PSS analysis.
Firstly, PLL is a large mixed system. Hspice itself cannot afford such large amount of computation. You may wait several days to see the actual transient results.
You can use spectre (not RF)'s verilogAMS feature to see the noise effects on transient locking process.
If you really want to do analysis with Phase noise, you have to rely on Spectre RF, Hspice RF, Eldo RF or any other Hamonic-Balance Simulator to do Steady-State Simulation. As we know, phase noise itself is caused by non-linear of circuits. And hspice itself cannot afford that feature.