how to choose length of the mos device in our design

Status
Not open for further replies.
S

srikanthp

Newbie level 4
Joined
Jul 2, 2005
Messages
5
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,328
hi i would like to clear a small doubt
suppose i want to design a CS amplifier with active load .
i will then use the eqn Id = 1/2*kn*(w/l)*(Vgs-Vt)²
i would like to bias the ckt at 50 µA and Vgs=1.5 V.
applying the eqn above i would get the value for w/l .
fine

just tell me how to choose the value for the length of channel first (l)
i find the simulation results vary as i change length keeping the aspect ratio constant .
pl help me out guys
lot of thanks in advance
 

the equation u used was not the one used by the simulator.
the equation is more complex than the simple square law. actually, it will contain λ effect, body effect, .... , but from the simplest way (first order approximation) u should include the λ effect in ur equation. i.e. Id=1/2*kn*(w/l)*(Vgs-Vt)²*(1+λVds).
 

Hi.
I agree with Btrend completely. But let me add something to make his word clear. λ is inversely proportional to "L", so with that equation he (Btrend) mentioned, Id will change with L, even if aspect ration remains constant. Besides, "L" specially in input devices will change input referred noise and in cascode devices (e.g. in a conventional cascode opamp) the magnitude of "L" will change ro but at the same time it will change the nondominant pole. So there are some trade-offs in choosing the right "L" for each device. Usually, we start our design with minimum "L" and if it is needed, we will increase it little by little.

Regards,
EZT
 

Status
Not open for further replies.

Similar threads

Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn more…