The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> fully differential Opamp-SC CMFB..help need simlat
https://designers-guide.org/forum/YaBB.pl?num=1158397497

Message started by ssalim on Sep 16^th, 2006, 2:04am

Title: fully differential Opamp-SC CMFB..help need simlat
Post by ssalim on Sep 16^th, 2006, 2:04am

Hi there,

I am designing a Fully differential single stage opamp with normal resistor divided CMFB, using Cadence 0.18um.

While simulating the opamp with resistor divided CMFB i get a gain of 65db and 200Mhz UGBW. And the results from Transcient analysis is fine.
But when the circuit's resistor in CMFB is replaced by Switched capacitor and PSS and PAC analysis were done, PAC yield me same gain and UGBW. I am happy in it...

When annotated my dc opearting points by PSS, it looked exact as in ac and tran analysis. i am happy in it....BUT..

During PSS i applied input frequency of 100Mhz and Clock freq also 100Mhz. I know for PSS (according to Ken) only clock (no input) is applied. After PSS i have to do PAC. But u see...PAC needs an input. How is it possible to run PSS followed by PAC (one requiring no input and other requiring input) at teh same time. I dont know what i understood is wrong...pls correct me if i have gone wrong.

My tran analysis and transient response using PSS looked different.

PSS reponse showed me the output sinusoidal waveform to be distorted and outputs were not differential. they looked almost similar?

Can i see the transient waveform which i got by Tran analysis using PSS?

Since i got same gain and BW (but different Tran response) using normal ac, tran and PAC, PSS, can i ensure that both my circuits are identical?

I am looking forward for ur great help...please respond .

regards
Sajam

Title: Re: fully differential Opamp-SC CMFB..help need si
Post by jbdavid on Sep 22^nd, 2006, 12:17am

Its easiest for me to think by analogy to DC and AC sim..
if I have a source with a VDC and ACM one will be used in the OP analysis, the other in the AC, XF, SP (the small signal) analyses..

So for that source thats DC for PSS - you use a DC value..
for the PAC, it will add in the PAC magnitude and frequency (thats what different between AC and PAC.. now you need both an ACmag + ACfreq..
spectre:

Code:

mag, phase

if you look at the parameters for port source or vsource elements in spectre reference manual

Code:

spectre -help vsource

will give you the details

Code:

pacmag pacphase

in veriloga module you might use a statement like

Code:

vsource #(.dc(1.2), .pacmag(1)) VIN (vin, gnd);