Does Spectre (Currnetly : Version 13.1.0.127.ISR3) has any option or setup to get better SFDR ?

Charley,

I don't understand the question. Spurious-Free Dynamic Range is a metric for the distortion (or ability to detect a signal above the noise/distortion in the system). It's a function of the design, not of the simulator - so I don't see what you mean about "an option to get better SFDR". Sure you should ensure that your measurement is not limited by the way you're measuring it (so do things like ensure that you are using strobeperiod in the tran analysis to coincide with the samples used for the SFDR calculation, because you don't want the DFT (which should be using a power of 2 samples) to have to interpolate between simulated points), but mostly this is governed by your design and improving it means improving your design...

Regards,

Andrew.

Hi Andrew , 

Sorry ,  I am debuging PostSim result  and spend a long time , since designer's pre-sim SFDR is better , but after extract r PostSim ,

the SFDR is worst ,  The  trend (SFDR ↑  r count  ↓     ,   SFDR ↓  r count  ↑ )  is base on parasitic r network (r  count).

Does spectre has limition on r count in parasitic file ? It means in  "conservative moderate liberal" is dependent  on

r count  in parasitic file? 

Thank you,

Charley 

No, spectre doesn't have a limitation on "r count". Your SFDR may have some dependency on how accurately the parasitics are extracted (I don't know) but I doubt that's anything to do with the simulator...

Regards,

Andrew.