The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> problem with PAC analysis
https://designers-guide.org/forum/YaBB.pl?num=1200305082

Message started by aaron_do on Jan 14^th, 2008, 2:04am

Title: problem with PAC analysis
Post by aaron_do on Jan 14^th, 2008, 2:04am

Hi all,

I just got a new PDK, so i decided to test an old LNA circuit using the new PDK. The gain and noise seemed to be fine, but there is a problem with IIP3. I ran both the PSS and PAC analysis. I set the PSS analysis to have one tone with frequency 2.45 GHz and amplitude prf where prf was swept from -80 dBm to -20 dBm (actually I tried sweeping several different ranges). For the PAC analysis, i set one tone with PAC magnitude = prf, and the frequency was set at absolute, 2.452 GHz.

I ran the simulation and found that the linear gain of the 2.452 GHz tone was -22 dB instead of +25 dB when plotted in the PAC form. I checked the PSS form, and plotted the linear gain for the 2.45 GHz signal and found it to be 25 dB as expected. Therefore, the problem is with the PAC analysis.

Anyway i've double and triple-checked everything, and have even gone back to the old PDK (which seems fine), but i can't isolate the problem. Does anybody know of the problem i'm speaking of and how to fix it?

thanks,
Aaron

Title: Re: problem with PAC analysis
Post by didac on Jan 15^th, 2008, 12:07pm

Hi aaron,
Time ago we found also a similar problem after PDK update, in our case it was only in the extracted simulation. The new PDK extracted bad an it causes convergence errors in PSS for the IIP3 analysis, after notification to CADENCE and the foundry we had to go back to the old PDK meanwhile the foundry gives us new extraction rules and then the problem was solved. Could you check with other colleagues if they found the same behaviour you describe?, maybe it's a similar case that we encountered and it's not a Spectre problem and it's caused by foundry models.
Hope it helps,

Title: Re: problem with PAC analysis
Post by aaron_do on Jan 15^th, 2008, 4:28pm

thanks,

yeah it seems others were having the problem too, and we've contacted the foundry.

cheers,
Aaron