HW 7 - Transition Gates and Multiplexers

116

We can calculate VOV as:

Vtn=VTON+γ(VSB+2ϕF2ϕF)=0.75V+0.5V(VI+0.6V0.6V)

Worst case for the NMOS is when VOV=0VGS=Vtn2.5VVI=Vtn as when that happens we are at the lowest of triode (essentially Ron,n):

2.5VI=0.75V+0.5V(VI+0.6V0.6V)1.426V

When this happens then Ron,n so look at the PMOS resistance:

250Ω=Ron,p=1Kp(W/L)P(VSG|Vtp|)=140μA/V2(W/L)P(1.426VVtp)

We need to find Vtp here:

Vtp=0.75+0.5(2.51.426+0.60.6)=1.010V

Thus:

(W/L)P=1250Ω40μA/V2(1.426V1.010V)=240

Thus:

(W/L)P=240/1

Repeat the process:

Vtp=0.75+γ(2.5VI+0.6V0.6V)=VIVI=1.0743V

thus:

Vtn=0.75+0.5(1.0743+0.60.6)=1.010V(W/L)N=1250Ω100μA/V2(2.5V1.0743V1.010V)=96.2

so then:

(W/L)N=96.2/1

117

Pasted image 20240305230554.png

a

For the first transistor, we'll need:

Vtn=0.70V+0.6(2.5Vtn+0.6V0.6V)Vtn=1.087V

So then VS=2.5V1.087V=1.413V since there's no current going through any FETs, so then if it was in triode then VDS=0V which would contradict VGSVtn. As such then that FET is in saturation.

For the second transistor notice that we are in triode. This is because now we have:

Vtn=0.70V+0.6(1.413VVtn+0.6V0.6V)Vtn=0.803V

But this is a contradiction since then VGS=0.803V so then VS=1.7V which is too much for a voltage drop. Hence we have triode so then VDS=0V so then VS=1.413V. The third transistor follows the same logic. Hence, since VS>VM then the inverter outputs VL.

For (b) notice that VSG=2.5V so then we are definitely not in CO. Here I=0 so then since VSB=0V then we have Vtp=VTOP=0.7. So VOV=1.8V. Notice VSG|Vtp| therefore since I=0 we cannot be in saturation. Therefore, we are in triode, so then VSD=0V therefore, all the FETs carry the 2.5V signal through. Hence the inverter outputs VL.

b

For this one, these are just three "switches" connected in series to the inverter, where no voltage drop occurs (since we have VDD as an input here). Essentially you force triode over all the FETS, so then the inverter input just gets VDD, which it turns into VL.