In Fig. P4.15, M1 is designed to operate as a current sink and M2 as a current source. If M2 had the source and body tied together, there would be no difference in operation of the two FETs. However,...

In Fig. P4.15, M1 is designed to operate as a current sink and M2 as a current source. If M2 had the source and body tied together, there would be no difference in operation of the two FETs. However, M2 is subject to the body effect, so its operation will differ from that of M1. We wish to investigate this difference and see which of the two devices comes closer to ideal current source/sink behavior. Let both FETs have k 5 500 A/V2 , Vt 0 5 21.0 V, and 5 1y(25 V). Moreover, let 5 0.5 V1y2 and u2 pu 5 0.65 V. (a) Calculate I1 and Ro1 at the edge of saturation for M1. Hence, fi nd vL1(max) for which M1 is still saturated. What is the percentage change in I1 for a 1-V for a per-volt decrease in vL1 below vL1(max)? (b) Calculate I2 and Ro2 at the edge of saturation for M2. Hence, fi nd vL2(max) for which M2 is still saturated. What is the percentage change in I2 for a per-volt decrease in vL2 below vL2(max)? (c) How must W2 be changed if we want I2 at M2’s edge of saturation to equal I1 at M1’s edge of saturation? Does this affect the percentage change in I2 for a 1-V decrease in vL2? Comment on your results.