MOS substhreshold operation

[ngtdat]

Hello,
As i read in the Gray Meyer's book, the drain current of MOSFET in weak inversion region is constant with different Vds. But when i simulated the operation of MOS in Cadence, the results show that the drain current is also change with Vds as in the strong inversion. Can anyone explain for me why? Does Cadence neglect the weak inversion region or it is because of other factors?

 

[Dominik Przyborowski]

In BSIM3 and BSIM4 models the only discontinuity exists between inversion regions so if OP of your transistor is quite deep in weak inversion You shouldn't see nothing "numerical".
Remember that any drain current equations showed in books are only the first approximations based on number of assumptions/simplifications. And if You checked Gray's book They are written about Id equation in weak inv. that Vds is ALMOST constant and it's true for very long devices.
Check equations in EKV model, which You can find in Enz, Krummenaher and Vittoz original article from 1996 or in their book "Charge based MOS model".

 

[dick_freebird]

What a textbook long channel device does, and what a
short channel real device does, diverged a few decades
back. Lambda for starters, add in DIBL and you're nowhere
near Kansas anymore.