pole very near to origin in NGCC opamp

[rishabh_31ec]

Hi,
I am designing four stage NGCC opamp in TSMC 180 nm process using Cadence Mixed signal design domain, keeping all mos have uniform length
L = 360nm.
I have designed whole circuitry in proper way so as;
gm1 = 114µs & gmf1 = 113.87µs. ( gm1 = transconductance of PMOS of diffrential amplifier )
gm2 = 184µs & gmf2 = 183.5µs
gm3 = 294µs & gmf3 = 293.7µs
gm4 = 982µs.
but when I simulate my circuitry I have found that a pole appears very near to zero i.e. @ 50.97.
I have calculated by considering UGB is near to 3 MHz and second pole is @ 3UGB, third and fourth pole are@ 5UGB then how the pole appears @ 50.97 my dc gain is ≈102 dB.

Is this due to fact that,
dc gain x 3 db bandwidth ( determined by first pole ) = UGB
If yes then how can I remove my problem without degrading gain, else what is the exact problem and it solution?

I have attached schematic of 3 stage NGCC just for idea.

 

[tompham]

Hi rish

Please read this

https://zanran_storage.s3.amazonaws.com/analog.postech.ac.kr/ContentPages/17514730.pdf

 

[D.A.(Tony)Stewart]

Hi, then how the pole appears @ 50.97 my dc gain is ≈102 dB.Is this due to fact that,dc gain x 3 db bandwidth ( determined by first pole ) = UGBIf yes then how can I remove my problem without degrading gain, else what is the exact problem and it solution? .

YES Try to reduce capacitance by 10x and reduce gain by 10 to increase GBW by 10