sutapanaki
Advanced Member level 4
Wow, how did you manage to make the ratio 3.9? And why? Why not 4?
Do you know what is the equivalent capacitance looking into the output of the first stage?
- - - Updated - - -
I think your design has a problem. With the gm/Id values you send previously and the currents in the two stages, gm1=130uS and gm2=563uS. So, the UGBW=gm1/Cc = 16.7MHz and the non-dominant pole p2 approximately is at 18MHz from gm7/CL, but it is probably smaller than that because this doesn't include the capacitance seen looking into the 1st stage. The zero is at 72MHz. So your p2 is almost at the UGBW, which will result in phase margin of about 45deg. I'm not surprised at all that your simulated frequency where the response crosses 0 is about 50% lower than what you wanted.
Did you run stability simulations on that amplifier?
You will have to move the non-dominant pole to be at least 3x higher than the UGBW, somewhere around 50MHz and also put a resistor in series with Cc to move the zero to even higher frequencies. Then you will have almost 1st order response, but if you want to get UGBW=19MHz, you will have to increase gm1 by 12 to 15 percent.
Do you know what is the equivalent capacitance looking into the output of the first stage?
- - - Updated - - -
I think your design has a problem. With the gm/Id values you send previously and the currents in the two stages, gm1=130uS and gm2=563uS. So, the UGBW=gm1/Cc = 16.7MHz and the non-dominant pole p2 approximately is at 18MHz from gm7/CL, but it is probably smaller than that because this doesn't include the capacitance seen looking into the 1st stage. The zero is at 72MHz. So your p2 is almost at the UGBW, which will result in phase margin of about 45deg. I'm not surprised at all that your simulated frequency where the response crosses 0 is about 50% lower than what you wanted.
Did you run stability simulations on that amplifier?
You will have to move the non-dominant pole to be at least 3x higher than the UGBW, somewhere around 50MHz and also put a resistor in series with Cc to move the zero to even higher frequencies. Then you will have almost 1st order response, but if you want to get UGBW=19MHz, you will have to increase gm1 by 12 to 15 percent.
