how does the vth affect the gm of nmos differential pairs?

[robicyang]

differential i.d gm

hi:
i want to design a constant gm rail-to-rail op with n-well process. the pmos pairs and the nmos pairs work in weak inversion and there are current switchs to flow the tail current when the pairs is inactive. the body of the pmos pairs connects to its source and the body of the nmos pairs connects to vss. simulation show that the gm of the nmos pairs increace with the common-mode voltage increacing and the gm of the pmos pairs maintian constant. i think this because of the common-mode voltage affecting the vth of the nmos pairs and so affecting the gm of the nmos pairs. but i don't know how the vth affect the gm of the nmos pairs in weak inversion. i know in weak inversion gm=Id/(n*VT), and all the parameters not change, why the gm changes? does anybody know why, please tell me. thanks

 

[alasliu]

vt vth nmos

do you have schematic, i think maybe tail current change.

 

[robicyang]

differential pair weak inversion

do you have schematic, i think maybe tail current change.

yes, i check the tail current, it increaces 2.5%%, but the gm increaces 6%. i think this is not correct.

 

[robicyang]

differential pair nmos body effect

i know why. because of the body effect, vth increace with common-mode voltage increacing, and so the Cjs decreace, and gm in weak inversion increace.