Hi all,


I am having some trouble with SpectreRF PSS anlaysis. I used a VCVS (voltage-controlled voltage source) in a negative feedback loop for DC biasing, but I found that although DC-analysis was able to find the correct solution, PSS anlaysis was not. In fact even DC analysis failed to find the correct solution when I switched the simulator to Turbo mode. Instead of forcing the two input terminal voltages of the VCVS to be equal, the negative input terminal just went to zero, and the output went to 1000V (my reference was 1 V, and the loop gain was 1000). I am sure it is negative and not positive feedback.

Has anybody experienced similar issues, and have a solution?


thanks,
Aaron

the manual of the pss simulator says it uses a special algorithm to calculate the steady-state of the circuit, which is less accurate than waiting for a transient sim to settle out. it could be that this special algorithm does not work well for your circuit for whatever reason - i second the opinion on running transient analysis to see whats going on.