PSRR simulation for Bandgap Reference.

[snoop835]

psrr simulation

Hi guys,

I am new in Bandgap Reference design. Currently I need to investigate the performance of Bandgap Voltage Reference using 1.8V supply. I have run DC sweep and temperature sweep to see its variation across temperature and supply voltage. My question is how to simulate for PSRR? What is the testbench setup for this? Is it in AC analysis or DC analysis? Can anyone suggest me good references or articles on this. I am also looking for good reading materials on Bandgap Reference.

Appreciate your help in advance....cheers

-snoop835-

 

[gdhp]

site:edaboard.com bandgap psrr

an 'ac' siganl at power, run the simulation of ac.

i think the psrr of bg is simialr with the opamp. you can do some reading on the the book of allen.

 

[snoop835]

psrr bandgap simulation

Thanks for the hint gdhp.

Here's the simulation waveforms that I have got after I run ac analysis. I think the result is acceptable. Can somebody tell me why is it in negative dB? Please give comments on this waveform.

 

[Humungus]

bandgap reference simulation

It is very important to keep in mind that PSRR (as well as CMRR) is made of two terms: sywtematic and random. When you perform a simple AC simulation, you'll only see the systematic (which in theory can be corrected) PSRR. Actually, PSRR arises from mismatches among the different signal paths from any source rail to the output. If everything were symmetric, PSRR would be negligeable. So, perfom and AC simulation, but along with a Monte Carlo analysis.

 

[drabos]

Re: psrr bandgap simulation

Thanks for the hint gdhp.

Here's the simulation waveforms that I have got after I run ac analysis. I think the result is acceptable. Can somebody tell me why is it in negative dB? Please give comments on this waveform.

plot the dB(1/vref)

 

[yuvan]

Humungus,
As far as it in a loop the random offset doesn't contribute a lot unless in the feedback. And in this case it is mostly the PMOS the current source in f/b (and also if the opamp gain is too low), so random variation in them shouldn't cause to much of an effect unless the deviation is too big for the positive gain to dominate.
And I think the PSRR shown is bit less & could be increased too, since the supply is 1.8v & we have 0.7v headroom (I guess u r trying to regulate the output to 1.1)