The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Mixed-Signal Design >> pnoise in spectreRF
https://designers-guide.org/forum/YaBB.pl?num=1204758055

Message started by analog2000 on Mar 5^th, 2008, 3:00pm

Title: pnoise in spectreRF
Post by analog2000 on Mar 5^th, 2008, 3:00pm

Hi everyone,
Do you know if for pnoise analysis I have to have the clocks have 0v amplitude assigned. These clocks drive the gates of switches. The switches are part of the switched capacitor network ?
In Ken Kundert's app note on switched cap noise, the clock amplitude is shown to be 0v

thanks
Best Regards
analog2000

Title: Re: pnoise in spectreRF
Post by thechopper on Mar 5^th, 2008, 5:56pm

Hi,

Everytime I run pnoise analysis I did it with the actual voltage levels for the clock voltage sources in order to properly run the transient analysis. I guess it is the only way the pnoise analysis can calculate the periodic quiescent operating point on both phases.

Tosei

Title: Re: pnoise in spectreRF
Post by analog2000 on Mar 6^th, 2008, 2:45am

Hi Tosei,
I am with you, however, when I do a pnoise with a single mos switch and a 2pF capacitor. I expect ~47uV rms noise level. I do not get this correct # if I have an amplitude on the clock.
If I turn the clock amplitude off then I get the correct 47uV rms in nyquist band ?
thanks
Best Regards
analog2000

Title: Re: pnoise in spectreRF
Post by thechopper on Mar 7^th, 2008, 5:20pm

Hi analog2000,

I do not understand exactly how you can get a noise different from zero when your switch is always off during both phases of your pnoise analysis....who is generating such noise? or may be I´m misunderstanding something?

Tosei

Title: Re: pnoise in spectreRF
Post by analog2000 on Mar 9^th, 2008, 4:59pm

Hi Tosei,
For transient sims, of course, I have the clock with a certain amplitude running.
It is only during the pnoise and pss analysis that the clock amplitude is 0
thanks
analog2000

Title: Re: pnoise in spectreRF
Post by ywguo on Mar 9^th, 2008, 11:19pm

Hi analog2000,

Quote:

In Ken Kundert's app note on switched cap noise, the clock amplitude is shown to be 0v

I am not familiar with the syntax of spectre. However, an example from Ken Kundert's Simulating Switched-Capacitor Filters with SpectreRF is shown below. It is clear that the clocks Vphi1 and Vphi2 is not set to zero.

Code:

Vphi1 (phi1 gnd) vsource type=pulse val0=VSS val1=VDD
+ period=CLOCK_PERIOD width=CLOCK_WIDTH
+ rise=CLOCK_TRANS fall=CLOCK_TRANS
+ delay=CLOCK_TRANS-CLOCK_PERIOD
+ fundname="clock"
Vphi2 (phi2 gnd) vsource type=pulse val0=VSS val1=VDD
+ period=CLOCK_PERIOD width=CLOCK_WIDTH
+ rise=CLOCK_TRANS fall=CLOCK_TRANS
+ delay=CLOCK_TRANS-CLOCK_PERIOD/2
+ fundname="clock"

// PSS analysis (sets periodic operating point)
clockAlone pss fund=CLOCK_FREQ saveinit=yes maxacfreq=6MHz \
writefinal="%C:r.ic" tstab=1.5ms swapfile="swap"
// Measure noise
unsmpldNoise (out gnd) pnoise start=100 stop=25kHz maxsideband=200
smpldNoise (out gnd) pnoise start=0_Hz stop=0.5*CLOCK_FREQ maxsideband=200 \
noisetype=timedomain noisetimepoints=[0] numberofpoints=1

Best regards,
Yawei