Lecture 14 ANNOUNCEMENTS 10 38 • Midterm #1 results (undergrad. scores only): • N=74; mean=62.8; median=63; std.dev.=8.42 • You may pick up your exam during any TA office hour. • Regrade requests must be made before you leave with your exam. • The list of misunderstood/forgotten points has been updated. OUTLINE • Frequency Response (cont’d) – – – – CE stage (final comments) CB stage Emitter follower Cascode stage Reading: Chapter 11.4-11.6 EE105 Fall 2007 Lecture 14, Slide 1 Prof. Liu, UC Berkeley 25 CE Stage Pole Frequencies, for VA<∞ w p ,in 1 RThev Cin 1 g m RC ro C w p ,out 1 1 C RC ro Cout 1 g R r m C o Note that wp,out > wp,in EE105 Fall 2007 Lecture 14, Slide 2 Prof. Liu, UC Berkeley I/O Impedances of CE Stage Zin 1 || r jwC 1 g m RC ro C EE105 Fall 2007 Z out Lecture 14, Slide 3 1 || RC || ro jwC CCS Prof. Liu, UC Berkeley CB Stage: Pole Frequencies • Note that there is no capacitance between input & output nodes No Miller multiplication effect! CB stage with BJT capacitances shown w p ,Y ro 1 RC CY CY C CCS w p, X 1 wT 1 RS || C X gm C X C EE105 Fall 2007 Lecture 14, Slide 4 Prof. Liu, UC Berkeley Emitter Follower • Recall that the emitter follower provides high input impedance and low output impedance, and is used as a voltage buffer. Follower stage with BJT capacitances shown • CL is the load capacitance EE105 Fall 2007 Circuit for small-signal analysis (Av) Lecture 14, Slide 5 ro Prof. Liu, UC Berkeley AC Analysis of Emitter Follower v X vout v • KCL at node X: vout v vin vout v v v 0 1 1 RS r jwC jwC v v vout • KCL at output node: g mv 1 1 r jwC jwCL C R 1 ( jw ) a S C C C C L C C L gm vout gm C R C vin a( jw ) 2 b( jw ) 1 b R C 1 S L S EE105 Fall 2007 Lecture 14, Slide 6 gm r g m Prof. Liu, UC Berkeley Follower: Zero and Pole Frequencies vout vin C 1 ( jw ) gm a ( jw ) 2 b ( jw ) 1 RS C C C C L C C L a gm b RS C C RS 1 gm r CL gm • The follower has one zero: gm wz 2fT C • The follower has two poles at lower frequencies: j w j w 1 a ( j w ) b ( jw ) 1 1 w w p1 p2 2 EE105 Fall 2007 Lecture 14, Slide 7 Prof. Liu, UC Berkeley Emitter Follower: Input Capacitance • Recall that the voltage gain of an emitter follower is Av Follower stage with BJT capacitances shown ro RL 1 RL gm • CXY can be decomposed into CX and CY at the input and output nodes, respectively: C C X 1 Av C 1 g m RL C 1 CY 1 C g m RL Av Rin r 1RL EE105 Fall 2007 C Cin C 1 g m RL Lecture 14, Slide 8 Prof. Liu, UC Berkeley Emitter Follower: Output Impedance ro Circuit for small-signal analysis (Rout) 1 v i X g m v r jwC vx v ix g m v RS Z out v X RS r C jw r RS r RS iX r C jw 1 1 EE105 Fall 2007 Lecture 14, Slide 9 jw r RS / RS r C jw 1 1 / r C 1 Prof. Liu, UC Berkeley Emitter Follower as Active Inductor Z out v X RS r C jw r RS r RS iX r C jw 1 1 CASE 1: RS < 1/gm jw r RS / RS r C jw 1 1 / r C 1 CASE 2: RS > 1/gm capacitive behavior inductive behavior • A follower is typically used to lower the driving impedance RS > 1/gm so that the “active inductor” characteristic on the right is usually observed. EE105 Fall 2007 Lecture 14, Slide 10 Prof. Liu, UC Berkeley Cascode Stage • Review: – A CE stage has large Rin but suffers from the Miller effect. – A CB stage is free from the Miller effect, but has small Rin. • A cascode stage provides high Rin with minimal Miller effect. ro Av , XY 1 vX 1 g m1 vY gm2 CX 2CXY EE105 Fall 2007 Lecture 14, Slide 11 Prof. Liu, UC Berkeley Cascode Stage: Pole Frequencies Cascode stage with BJT capacitances shown (Miller approximation applied) w p, X ro 1 RS || r 1 C 1 2C 1 w p ,Y 1 1 CCS1 C 2 2C1 g m2 Note that w p ,Y w p ,out EE105 Fall 2007 Lecture 14, Slide 12 gm2 2fT 2 C 2 1 RL CCS 2 C 2 Prof. Liu, UC Berkeley Cascode Stage: I/O Impedances ro 1 Z in r 1 || jw C 1 2C1 EE105 Fall 2007 Z out RL || Lecture 14, Slide 13 1 jw C 2 CCS 2 Prof. Liu, UC Berkeley Summary of Cascode Stage Benefits • A cascode stage has high output impedance, which is advantageous for – achieving high voltage gain – use as a current source • In a cascode stage, the Miller effect is reduced, for improved performance at high frequencies. EE105 Fall 2007 Lecture 14, Slide 14 Prof. Liu, UC Berkeley