PSS simulation problem in Cadence

[freewing]

When I run PSS simulation of a vco, PSS is not successful for some cases. But when I run transient simulation, the vco does oscillate and everything is fine. The beat frequency is also correct. So does anybody know what's the problem and how to fix it? Thanks.

 

[steer]

Try to run an initial tran simulation for sufficiently long time. In most cases this solves the convergency problem.

 

[BigBoss]

PSS simulation is very tricky way to predict oscillation frequency. Transient is better with higher accuracy levels ( increase abstol,iabsoltol,vabstol in simulation option ).
But PSS will also converged if you take care some critic nodes such as floating nodes, short circuited components etc. PSS doesn't like unusual circuit topologies and to avoid them, it should be used from simpler to more complex circuit design.

And at all, of course realistic models. Don't forget that there is no circuit but models..

 

[uncle_urfi]

After all is said and done, the PSS simulations turn out to be optimistic by about 7-10 dBC/Hz.

Pls. read this paper as a reference.

Jacobsson, H., et al, “Low-Phase-Noise Low-Power IC VCOs for 5-8-GHz Wireless
Applications”, IEEE, Trans. MTT, Vol. 48, No. 12, December 2000, pp. 2533-2539

It may be a setup problem ----> goto spectre manual
It may be a topology problem ----> goto your professor/boss
It may be a shorter simulation than is necessary ----> Kill a weekend

Try not to include the effects of the transient startup.
Try some initial conditions.

Try using an ideal mixer on chip and setup up a phase noise simulation as you would in the lab to get a handle on what is going on since you mentioned that the transient simulation is working. (kill several weekends)



Use ADS to simulate phase noise, if the circuit is not too too big. Again you will be off by 3-5 dBc/Hz (says the paper above)

Hope it helps.

 

[steer]

After all is said and done, the PSS simulations turn out to be optimistic by about 7-10 dBC/Hz.

Not in my experience. I see PSS/PNOISE closely matches my measurements.