The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> RF Simulators >> Pnoise in Time-Interleaved Circuits
https://designers-guide.org/forum/YaBB.pl?num=1188933362

Message started by Friedel on Sep 4^th, 2007, 12:16pm

Title: Pnoise in Time-Interleaved Circuits
Post by Friedel on Sep 4^th, 2007, 12:16pm

Hey Guys,

I use the Pnoise analysis to verify the noise performance of a sample-hold block.
The SH uses a clock frequency of 200MHz. From the strobed Pnoise analysis I got the noise power density V^2/Hz
with I integrate up to 100Mhz to get the noise power.
The obtained value is close to the expected kT/C value. :)
(thanks all for the help in the past on this topic!!)

The total input stage consists of two time-interleaved SH blocks.
each operating at 200MSs. Thus the total input bandwith is 200Mhz!

Over which bandwidth I have to integarte the noise power density V^2/Hz
if I simulate both SH stages operating in a ping-pong (interleaved) fashion? :-[
Still 100Mhz since each stage converts only 100Mhz?

Thanks

Title: Re: Pnoise in Time-Interleaved Circuits
Post by Frank Wiedmann on Sep 4^th, 2007, 11:40pm

You always have to integrate to half the frequency specified in the pss analysis setup.

Title: Re: Pnoise in Time-Interleaved Circuits
Post by Friedel on Sep 5^th, 2007, 11:21am

Frank,

I agree.

Only to be sure, that my approach is correct:
I use a 400Mhz clock (->beat fre=400MHz). The clock is feed into a divider
generating 2x200MHz clocks shifted by 180deg.

Since my beat frequency is 400Mhz, I have
to integrate over 200Mhz each SH output?
But each SH is only sampled at 200MHz? :-/

What would be your approach to simulated such a circuit correctly
(an easy way to evaluate the noise perfromance of an time interleaved circuit?)

Thanks much
Friedel

Title: Re: Pnoise in Time-Interleaved Circuits
Post by Frank Wiedmann on Sep 6^th, 2007, 12:25am

All signals in a pss analysis must be periodic, so you have to set the pss frequency to the lowest frequency that occurs in your circuit, which is 200 MHz, not 400 MHz. I would be very surprised if your pss analysis converged at all when you set the pss frequency to 400 MHz.