Xiao Wang
New Member
Offline
Posts: 2
|
Interestingly when I turn off augumented pnoise:
Regular (non augumented Pnoise, turbo solver) (("104.5u" "Error") ("104u" ("Va" 0.2584785 51591.15 "VmidP" 0.8837754 -166661.5 ) ((1.58022e-11 9.134028e-15) 4.516825e-10 -108.0128) ) ("103.5u" "Error") ("103u" "Error") ("102.5u" ("Va" 0.172236 62908.75 "VmidP" 0.9860699 -16984.06 ) ((1.721947e-11 8.37348e-15) 4.715021e-10 -107.6398) ) ("102u" "Error") ("101u" ("Va" 0.07004445 73383.45 "VmidP" 0.9939822 -141.7666 ) ((1.685612e-11 6.915519e-15) 4.665039e-10 -107.7324) ) ("100.5u" ("Va" 0.2955239 44910.93 "VmidP" 0.6764826 -404095.6 ) ((1.141454e-11 8.563347e-15) 3.838876e-10 -109.4254) ) ("100u" "Error") ("99.5u" "Error") ("99u" ("Va" 0.261198 51149.22 "VmidP" 0.8747783 -178366.0 ) ((1.559862e-11 9.128474e-15) 4.487559e-10 -108.0691) ) ("98.5u" "Error") ("98u" ("Va" 0.2123941 58195.96 "VmidP" 0.9657643 -50861.31 ) ((1.735005e-11 9.133293e-15) 4.732865e-10 -107.607) ) )
The results are extremely consistent. Regular (non augumented) pnoise is also based on Demir's algorithm from what I understand. These are also fairly consistent with the good results from the augumented mode. however I do get this warning:
Warning from spectre during PNoise analysis `pnoise'. WARNING (SPCRTRF-15091): The shooting small signal analysis is unable to find an oscillator mode for the oscillator (the cuicuit does not behave like an oscillator), To fix the problem, 1) Tighten the simulation tolerance for the PSS analysis by either using a tighter `errpreset' value (for example, errpreset=conservative), or by using a smaller `reltol' value (for example, reltol=1e-5). Rerun the simulation.
and it suggests turning on augmented pnoise, lol. This problem is a real head scratcher. Is there a chance these problems indicate a possible problem with the augmented pnoise for the shooting solver?
Thanks, Xiao
|