Why is a constant gm needed Vi+ gm2 gmigI 8m? Cm +SCm8m2+gogh +S一+ Pip2 P1 GBW 8m/C 8, and gbw_gml
Why is a Constant Gm needed ?
Techniques for N-P Complementary Rail-to-Rail Input Stage 1. For input stages with input transistors working in weak-inversion region, using current complementary circuit to keep the sum of IN and ∥ constant[12][6] 2. Using square root circuit (Ip +yIn) to keep constant [3113[16] 3. and 4. Using current switches to change the tail current of input differential pairs [34][56] 4. Using hex-pair structure to control the tail currents of backup pairs [71
Techniques for N-P Complementary Rail-to-Rail Input Stage • 1. For input stages with input transistors working in weak-inversion region, using current complementary circuit to keep the sum of IN and • IP constant [1][2][6]; • 2. Using square root circuit to keep constant [3][13][16]; • 3. and 4. Using current switches to change the tail current of input differential pairs [3][4][5][6]; • 4. Using hex-pair structure to control the tail currents of backup pairs [7];
Techniques for N-P Complementary Rail-to-Rail Input Stage( contd) 5. Using maximum/minimum selection circuit to conduct the output current of the differential pair with larger current, as well as larger gm, to the next stage [8]19 6. Using electronic zener diode to keep constant [10] 7. Using DC level shift circuit to change the input DC leve We will analyze them one by one in the following sections There are still other techniques [12][14[15[1718 interested readers may check these references
Techniques for N-P Complementary Rail-to-Rail Input Stage( cont’d ) • 5. Using maximum/minimum selection circuit to conduct the output current of the differential pair with larger current, as well as larger gm,to the next stage [8][9]; • 6. Using electronic zener diode to keep constant [10]; • 7. Using DC level shift circuit to change the input DC level [11]. • We will analyze them one by one in the following sections. • There are still other techniques [12][14][15][17][18], interested readers may check these references
Rail-to-Rail Input Stage, Structure 2 Basic idea For an input differential pair, using a 1st order approximation, gm=gmy+8mp=√KPR(W/D)lx+√kPWL)pl Square Root 12 Summing Biasing M3 M Circuit and Circuit
Rail-to-Rail Input Stage, Structure 2 • Basic idea • – For an input differential pair, using a 1st order approximation
Rail-to-Rail Input Stage, Structure 3 [3]4116 Using current switches to change the tail current of input differential pairs VDD In VI MI V 2M4 2A Ibn=lb 1:3 .VSS
Rail-to-Rail Input Stage, Structure 3 [3][4][6] • Using current switches to change the tail current of input differential pairs