Common Drain Stage (Source Follower) Claudio Talarico, Gonzaga University

Similar documents
3. Basic building blocks. Analog Design for CMOS VLSI Systems Franco Maloberti

Lecture 37: Frequency response. Context

Advanced Current Mirrors and Opamps

ECE-343 Test 1: Feb 10, :00-8:00pm, Closed Book. Name : SOLUTION

ECE 546 Lecture 11 MOS Amplifiers

Assignment 3 ELEC 312/Winter 12 R.Raut, Ph.D.

1/13/12 V DS. I d V GS. C ox ( = f (V GS ,V DS ,V SB = I D. + i d + I ΔV + I ΔV BS V BS. 19 January 2012

ECE 415/515 ANALOG INTEGRATED CIRCUIT DESIGN

ECE-342 Test 3: Nov 30, :00-8:00, Closed Book. Name : Solution

Exact Analysis of a Common-Source MOSFET Amplifier

ECEN 326 Electronic Circuits

Lecture 13 MOSFET as an amplifier with an introduction to MOSFET small-signal model and small-signal schematics. Lena Peterson

Lecture 050 Followers (1/11/04) Page ECE Analog Integrated Circuits and Systems II P.E. Allen

First and Second Order Circuits. Claudio Talarico, Gonzaga University Spring 2015

Stability and Frequency Compensation

EE 330. Lecture 35. Parasitic Capacitances in MOS Devices

CMOS Analog Circuits

Common Source Stage Miller Approximation ZVTC Analysis Backgate Effect Claudio Talarico Gonzaga University

ECE-343 Test 2: Mar 21, :00-8:00, Closed Book. Name : SOLUTION

EE105 Fall 2015 Microelectronic Devices and Circuits Frequency Response. Prof. Ming C. Wu 511 Sutardja Dai Hall (SDH)

Lecture 23 Frequency Response of Amplifiers (I) Common Source Amplifier. December 1, 2005

Lecture 15: MOS Transistor models: Body effects, SPICE models. Context. In the last lecture, we discussed the modes of operation of a MOS FET:

University of Toronto. Final Exam

ECE 342 Electronic Circuits. 3. MOS Transistors

Electronic Devices and Circuits Lecture 18 - Single Transistor Amplifier Stages - Outline Announcements. Notes on Single Transistor Amplifiers

Electronic Circuits Summary

ECE 523/421 - Analog Electronics University of New Mexico Solutions Homework 3

Chapter 13 Small-Signal Modeling and Linear Amplification

Lecture 10 MOSFET (III) MOSFET Equivalent Circuit Models

Lecture 04: Single Transistor Ampliers

Design of Analog Integrated Circuits

Lecture 10 MOSFET (III) MOSFET Equivalent Circuit Models

Lecture 24 Multistage Amplifiers (I) MULTISTAGE AMPLIFIER

Biasing the CE Amplifier

EE105 Fall 2014 Microelectronic Devices and Circuits

EECS 105: FALL 06 FINAL

EE 240B Spring Advanced Analog Integrated Circuits Lecture 2: MOS Transistor Models. Elad Alon Dept. of EECS

The Miller Approximation

MOS Transistor Theory

Amplifiers, Source followers & Cascodes

Frequency Response Prof. Ali M. Niknejad Prof. Rikky Muller

Radivoje Đurić, 2015, Analogna Integrisana Kola 1

ECE315 / ECE515 Lecture-2 Date:

Lecture 3: CMOS Transistor Theory

Chapter 10 Feedback. PART C: Stability and Compensation

6.012 Electronic Devices and Circuits Spring 2005

Circuit Topologies & Analysis Techniques in HF ICs

Circuits. L2: MOS Models-2 (1 st Aug. 2013) B. Mazhari Dept. of EE, IIT Kanpur. B. Mazhari, IITK. G-Number

MOSFET: Introduction

Homework Assignment 09

EE105 Fall 2014 Microelectronic Devices and Circuits. NMOS Transistor Capacitances: Saturation Region

Lecture 23 - Frequency Resp onse of Amplifiers (I) Common-Source Amplifier. May 6, 2003

Lecture 5 Review Current Source Active Load Modified Large / Small Signal Models Channel Length Modulation

University of Pennsylvania Department of Electrical Engineering. ESE 570 Midterm Exam March 14, 2013 FORMULAS AND DATA

Circle the one best answer for each question. Five points per question.

CE/CS Amplifier Response at High Frequencies

Errata of CMOS Analog Circuit Design 2 nd Edition By Phillip E. Allen and Douglas R. Holberg

Multistage Amplifier Frequency Response

MOS Transistor I-V Characteristics and Parasitics

LECTURE 130 COMPENSATION OF OP AMPS-II (READING: GHLM , AH )

ID # NAME. EE-255 EXAM 3 April 7, Instructor (circle one) Ogborn Lundstrom

EECS 141: FALL 05 MIDTERM 1

UNIVERSITY OF CALIFORNIA College of Engineering Department of Electrical Engineering and Computer Sciences

The Devices. Digital Integrated Circuits A Design Perspective. Jan M. Rabaey Anantha Chandrakasan Borivoje Nikolic. July 30, 2002

Sample-and-Holds David Johns and Ken Martin University of Toronto

Electronics II. Final Examination

Microelectronics Main CMOS design rules & basic circuits

MOS Transistor Theory

DESIGN MICROELECTRONICS ELCT 703 (W17) LECTURE 3: OP-AMP CMOS CIRCUIT. Dr. Eman Azab Assistant Professor Office: C

ECE 342 Electronic Circuits. Lecture 25 Frequency Response of CG, CB,SF and EF

Integrated Circuit Operational Amplifiers

Homework Assignment 08

Chapter 4 Field-Effect Transistors

ECE 497 JS Lecture - 12 Device Technologies

Lecture 28 Field-Effect Transistors

Lecture 23: Negative Resistance Osc, Differential Osc, and VCOs

MOS Transistors. Prof. Krishna Saraswat. Department of Electrical Engineering Stanford University Stanford, CA

ECE 6412, Spring Final Exam Page 1 FINAL EXAMINATION NAME SCORE /120

Practice 7: CMOS Capacitance

ECE 6412, Spring Final Exam Page 1

ECE 546 Lecture 10 MOS Transistors

Analog Integrated Circuit Design Prof. Nagendra Krishnapura Department of Electrical Engineering Indian Institute of Technology, Madras

ECE 255, Frequency Response

Introduction and Background

The transistor is not in the cutoff region. the transistor is in the saturation region. To see this, recognize that in a long-channel transistor ifv

and V DS V GS V T (the saturation region) I DS = k 2 (V GS V T )2 (1+ V DS )

V in (min) and V in (min) = (V OH -V OL ) dv out (0) dt = A p 1 V in = = 10 6 = 1V/µs

Microelectronic Circuit Design 4th Edition Errata - Updated 4/4/14

The Devices. Jan M. Rabaey

ECE 342 Electronic Circuits. Lecture 6 MOS Transistors

Lecture 17 Date:

ECE315 / ECE515 Lecture 11 Date:

ELEN 610 Data Converters

Practice 3: Semiconductors

GEORGIA INSTITUTE OF TECHNOLOGY School of Electrical and Computer Engineering

EE 330 Lecture 22. Small Signal Modelling Operating Points for Amplifier Applications Amplification with Transistor Circuits

1. (50 points, BJT curves & equivalent) For the 2N3904 =(npn) and the 2N3906 =(pnp)

Figure 1: MOSFET symbols.

Whereas the diode was a 1-junction device, the transistor contains two junctions. This leads to two possibilities:

Introduction to CMOS RF Integrated Circuits Design

Transcription:

Common Drain Stage (Source Follower) Claudio Talarico, Gonzaga University

Common Drain Stage v gs v i - v o V DD v bs - v o R S Vv IN i v i G C gd C+C gd gb B&D v s vv OUT o + V S I B R L C L v gs - C gs S r o v gs -b v o v o V B R L C sb +C L EE 303 Common Drain Stage 2

CD Bias Point V DD Assume V B V OUT, so I D I B R S V IN V OUT V IN V GS V S V OUT I B R L V GS V T + I D 0.5C OX W / L ( ) V T V T 0 +γ PHI +V OUT PHI V B V DS V DD V OUT V DD V IN +V GS For M to be in saturation V DS > V OV V GS +V T! V DD V IN +V GS > V GS +V T! V IN < V DD +V T This is almost always the case in practical realizations EE 303 Common Drain Stage 3

I/O DC characteristic () Example: V IN 2.5V; I 400µA; V DD 5V; W/L00µm/µm 5 4.5 DC Transfer Function um Technology Vout V DD -V th 3.43 V 4 3.5 Vth M triode 6 Source: Razavi M operation V out [V] 3 2.5 2.5 5 0.5 4 VdsVdd Vout VovVin Vout Vth 0 0 2 3 4 5 6 7 8 V in [V] 3 2 M sat. A V (@ V IN 2.5V) 0.82 Math artifact due to ideal current source (a current source can take any voltage value across it!!) 0 0 2 3 4 5 6 7 8 V in [V] M triode V out (min) 0 V out (max) V DD -V th EE 303 Common Drain Stage 4

I/O DC characteristic (2) I Example: V IN 2.5V; I 400µA; V DD 5V; V B 0.9V; W/L00µm/µm; 5 4.5 4 3.5 DC Transfer Function um Technology element 0:m 0:m2 model 0:simple_n 0:simple_n region Saturati Saturati id 452.26u 452.26u ibs -3.035f 0. ibd -50.0000f -3.035f vgs.968 900.0000m vds 3.6968.3032 vbs -.3032 0. vth 833.4784m 500.0000m vdsat 363.3703m 400.0000m vod 363.3703m 400.0000m beta 6.8484m 5.656m gam eff 600.0000m 600.0000m gm 2.4885m 2.2606m gds 33.0095u 40.0000u gmb 54.7852u 758.2384u cdtot 72.3963f 93.222f cgtot 256.5245f 256.4846f cstot 246.466f 286.334f cbtot 66.38f 28.5634f cgs 203.334f 203.334f cgd 5.337f 50.3996f * measurement av0 808.9285m V out [V] 3 2.5 2.5 0.5 M 2 off M sat 0 0 2 3 4 5 6 7 8 M 2 triode M sat V in [V] M 2 sat M sat Reduced output (and input) swing - V out (max) V DD V OV - V out (min) V OV2 VoutVds2 Vov2 Vds Vov M 2 sat M triode EE 303 Common Drain Stage 5

CD Voltage Transfer v o ' % sc & Ltot $ + sc " gs + viscgs gm i o RLtot # ( v v ) 0 v i C gd +C gb + v gs - C gs v gs v o v v o i g m + sc g gs m + sc + sc gs Ltot + R Ltot LHP zero C Ltot C L + C sb C Ltot R Ltot R Ltot R L g mb r o v o v i + R Ltot s a v0 z + sc gs ( ) + s C + C gs Ltot + R Ltot often R Ltot /b s p EE 303 Common Drain Stage 6

Low Frequency Gain V DD a v0 g m Ltot L o gm + gmb RLtot R Interesting cases Ideal current source; PMOS with source tied to body; no load resistor; R L, r o, b 0 V o R r V i Y in C L a v0 V i V DD I B V o R L Ideal current source; NMOS; no load resistor; R L, r o, b 0 C L a v0 g m gm + g mb (typically 0.8) Ideal current source, PMOS with source tied to body; load resistor r o, b 0, R L finite gm av0 g + m EE 303 Common Drain Stage R 7 L

Low frequency input and output resistances R in R out v in v s R S / /b v out v in r o R L a v0 / (g m +/r o +/R L ) R in R out r o (/g m ) EE 303 Common Drain Stage 8

High Frequency Gain a v0 g' m +/ r o +/ R L a v ( s) v v o i a v0 s z s p z g C m gs p g C m gs + R + C Ltot Ltot Three scenarios: z < p z > p z p a v (s) a v (s) a v (s) The presence of a LHP zero leads to the possibility that the pole and zero will provide some degree of cancellation, leading to a broadband response a v0 a v0 a v0 f f (infinite bandwidth?!?) f EE 303 Common Drain Stage 9

CD Input Impedance Hint: ic gs (v i -v o ) By inspection: Y in sc gd + sc gs Consistent with insight from Miller Theorem ( a v (s)) v i Y in Gain term a v (s) is real and close to unity C gs v gs up to fairly high frequencies C gd +C gb + v gs - C Ltot R Ltot v o Hence, up to moderate frequencies, we see a capacitor looking into the input A fairly small one, C gd, plus a fraction of C gs EE 303 Common Drain Stage 0

CD input impedance for PMOS stage (with Body-source tie) V DD I B b generator inactive Low frequency gain very close to unity r o >> C gs v OUT V out C L a v0 + r o vv IN in C in @ C gd Cin C gd C gd C bsub Very small input capacitance Y in sc gd + sc gs ( a v (s)) Y in sc gd for a v (V out V in ) no current flows through C gs (perfect bootstrap) The well-body capacitance C bsub can be large and may significantly affect the 3-db bandwidth EE 303 Common Drain Stage

CD Output Impedance () Let's first look at an analytically simple case Input driven by ideal voltage source C gd +C gb + v gs - C gs v gs v o Z out g m + g mb s ( C + C ) gs sb AC shorted C sb /b Z out Z L Y L sc L +/R L Low output impedance Resistive up to very high frequencies EE 303 Common Drain Stage 2

CD Output Impedance (2) Now include finite source resistance v g R i C gs (v g -v o ) Z x Z out C gd +C gb v o Neglect C gd i x C sb /b Z L i x ( v o v g )( + sc gs ) v o v g $ Z x v o i x + sr ic gs + sc gs " + sc gs v o # " $ $ + sr i C gs $ $ # % ' + sc gs & % ' ' ' ' & ( ) Neglecting C gd v g v o R i sc gs + R i EE 303 Common Drain Stage 3

CD Output Impedance (3) Two interesting cases Z x ( + sr i C gs ) " + sc % gs $ ' # & NOTE: Typically /gm is small, so this can happen!! R i < / R i > / Z x (s) Z x (s) / / f f Inductive behavior! EE 303 Common Drain Stage 4

Equivalent Circuit for R i > / Z x ( + sr i C gs ) " + sc % gs $ ' # & for s 0 Z x / for s Z x R i for s 0 Z x R R 2 for s Z x R 2 R L R 2 Z x C sb /b Z out R R 2 L R g 2 R i R m 2 i C R i g gs m Z x (R + sl)r 2 R + R 2 + sl R + L R 2 R R + R 2 L + R + R 2 This circuit is prone to ringing! L forms an LC tank with any capacitance at the output If the circuit drives a large capacitance the response to step-like signals will result in ringing EE 303 Common Drain Stage 5

Inclusion of Parasitic Input Capacitance* What happens to the output impedance if we don t neglect C i C gd? * Hint: replace R i with R i Z i R i + sc i R i ( ( )) Z x + sr i C gs + C i! + sc $ gs # & + sr i C i " % ( ) R i ( C gs + C i ) < < C gs R i C i R i ( C gs + C i ) < < R i C i C gs Z x (s) Z x (s) / / f f EE4 EE 303 Common Drain Stage 6

Applications of the Common Drain Stage Level Shifter Voltage Buffer Load Device EE 303 Common Drain Stage 7

Application : Level Shifter V i V GS +V o V o V i V GS V i (V t +V OV ) V DD V i V GS V t +V ov const. V o I B Output quiescent point is roughly V t +V ov lower than input quiescent point Adjusting the W/L ratio allows to tune Vov ( the desired shifting level) EE 303 Common Drain Stage 8

Why is lowering the DC level useful? When building cascades of CS and CG amplifiers, as we move along the DC level moves up A CD stage is a way to buffer the signal, and moving down the DC level EE 303 Common Drain Stage 9

Application 2: Buffer V DD V BB I B R D (large) M v x R out vv out Vv in M X M 2 I B2 R L (small) V B2 Low frequency voltage gain of the above circuit is ~ R big Would be ~ (R small R big ) ~ R small without CD buffer stage Disadvantage Reduced swing V out (max) V DD V OV V out (min) V OV 2 EE 303 Common Drain Stage 20

Buffer Design Considerations I B V DD V B B R D (large) M v x V x Vv in M X VB M2 I B2 R out vv out out R L (small) Without the buffer the minimum allowable value of V x for M x to remain in saturation is: V X > V GSX V thx With the buffer for M 2 to remain in sat. it must be: V X > V GS +V GS2 -V th2 V B2 Assuming the overdrive of M x and M 2 are comparable this means that the allowable swing at X is reduced by V GS which is a significant amount (V GS V OV +V th ) EE 303 Common Drain Stage 2

Application 3: Load Device Looks familiar? CS stage with diode connected load V i V DD M2 M /(2 +b2 ) V o Advantages compared to resistor load "Ratiometric" Gain depends on ratio of similar parameters Reduced process and temperature variations First order cancellation of nonlinearities Disadvantage Reduced swing V out (max) V DD V th2 a v0 g m2 gm + g mb2 V out (min) V IN V th V OV EE 303 Common Drain Stage 22

CD stage Issues Several sources of nonlinearity V t is a function of V o (NMOS, without S to B connection) I D and thus V ov changes with V o Gets worse with small R L ( ) V t V t 0 +γ PHI +V o PHI I D 2 µc OX W ( L V V GS t ) 2 If we worry about distortion we need to keep ΔV in /2V OV small (à we need large V OV, and this affect adversely signal swing). If R L is small, things are even worse because to obtain the desired ΔV out we need more ΔV in (and so even more V OV ) ΔV out ΔV in g' m +/ R L Small R L means less gain Reduced input and output voltage swing Consider e.g. V DD V, V t 0.3V, V OV 0.2V (V GS V t +V ov 0.5V) CD buffer stage consumes 50% of supply headroom! In low V DD applications that require large output swing, using a CD buffer is often not possible CD buffers are more frequently used when the required swing is small E.g. pre-amplifiers or LNAs that turn µv into mv at the output EE 303 Common Drain Stage 23

Summary Elementary Transistor Stages Common source VCCS, makes a good voltage amplifier when terminated with a high impedance Common gate Typically low input impedance, high output impedance Can be used to improve the intrinsic voltage gain of a common source stage "Cascode" stage Common drain Typically high input impedance, low output impedance Great for shifting the DC operating point of signals Useful as a voltage buffer when swing and nonlinearity are not an issue EE 303 Common Drain Stage 24

CD Voltage Transfer Revisited () The driving circuit has a finite resistance R s Z in CD Core example: if R S is the output resistance of a CS stage it can be relatively large Z out R S v in v out v s Cgd C gs /g m C sb R L C L v in r o /g m R tot C tot This model resembles the model of the CS stage - The main difference is in the polarity of the controlled source - This difference has profound impact on the Miller amplification of the capacitance coupling input and output EE 303 Common Drain Stage 25

CD Voltage Transfer Revisited (2) v out v s a v0 + s C gs + b s + b 2 s 2 a v0 R tot b R s C gs ( a v0 )+ R S C gd + R tot C gs + R tot C tot b 2 R S R tot (C gs C gd + C gs C tot + C gd C tot ) The zero in the transfer function is on the LHP and it occurs at approximately ω T Unfortunately the high frequency analysis can become quite involved. A dominant real pole does not always exist, in fact poles can be complex. In the case the poles are complex a designer should be careful that the circuit does not exhibit too much overshoot and ringing. EE 303 Common Drain Stage 26

CD Voltage Transfer Revisited (3) v out v s a v0 + s C gs + s C gs + b s + b 2 s a g 2 v0 m + s ω 0 Q + s2 2 ω 0 Interesting cases: - A dominant pole condition exist - Poles are complex ω 3dB b τ j b ( ) A dominant pole condition exist for: Q 2 << /4 b 2 b 2 <</ 4 EE 303 Common Drain Stage 27

Poles Location Roots of the denominator of the transfer function: + s ω 0 Q + s2 ω 0 2 0 Complex Conjugate poles (overshooting in step response) for Q > 0.5 s,2 ω 0 2Q ( ± j 4Q2 ) - For Q 0.707 (φ45 ), the -3dB frequency is ω 0 (Maximally Flat Magnitude or Butterworth Response) Poles location with Q> 0.5 jω φ σ - For Q > 0.707 the frequency response has peaking Real poles (no overshoot in the step response) for Q 0.5 s,2 ω 0 ± 4Q2 2Q ( ) Source: Carusone EE 303 Common Drain Stage 28

Frequency Response 0 H(s) 5 H(s) + s ω 0 Q + s2 ω 0 2 Magnitude [db] 0 5 0 5 Q0.5 Q/sqrt(2) Q Q2 20 0 2 0 0 0 0 / 0 EE 303 Common Drain Stage 29

Step Response Underdamped Q > 0.5 Source: Carusone Q0.5 Overdamped Q > 0.5 time Ringing for Q > 0.5 The case Q0.5 is called maximally damped response (fastest settling without any overshoot) EE 303 Common Drain Stage 30