venn_ng
Member level 5
I am designing a traditional 1st order delta sigma modulator. I tried to find the input referred noise of the integrator and switching network (circuit is shown). I set vin+ = vin- =vcm & then I set v to be a square wave at a frequency fclk/2 (to emulate the limit cycle behavior of first-order delta sigma for code 0). This is my periodic steady state point. I am also using a differential stb probe as shown to simulate pstb.
First question, is this how you simulate pss for switched cap integrator (note that this is not a switched cap amplifier but an integrator).
I apply PNOISE on this and find the output referred noise (at the output of differential integrator) & scale back to the input by using the PAC gain function by specifying the output and input nodes. I also tried using the input referred noise function in PNOISE. The results don't make sense, as the numbers are way off.
And for pstb, I see that the DC loop gain of PSTB starts off at 0 (i mean -∞ dB) and goes up at 20 dB/decade. Is this due to the integrator action? Is this how do you stability of an integrator using PSTB.
I read articles on simulating PSS/PSTB/PAC for switched cap amplifier but for integrator I am not sure if this is the right way to do.
First question, is this how you simulate pss for switched cap integrator (note that this is not a switched cap amplifier but an integrator).
I apply PNOISE on this and find the output referred noise (at the output of differential integrator) & scale back to the input by using the PAC gain function by specifying the output and input nodes. I also tried using the input referred noise function in PNOISE. The results don't make sense, as the numbers are way off.
And for pstb, I see that the DC loop gain of PSTB starts off at 0 (i mean -∞ dB) and goes up at 20 dB/decade. Is this due to the integrator action? Is this how do you stability of an integrator using PSTB.
I read articles on simulating PSS/PSTB/PAC for switched cap amplifier but for integrator I am not sure if this is the right way to do.