ECEN 607(ESS)

Low Drop-Out Voltage Regulators

1
Analog and Mixed-Signal Center, Texas A&M University
 Power Management
• Why do we need power management?
– Batteries discharge “almost” linearly with time.
– Circuits with reduced power supply that are time dependent
operate poorly. Optimal circuit performance can not be obtained.
– Mobile applications impose saving power as much as possible.
Thus, the sleep-mode and full-power mode must be carefully
controlled.
– What is the objective of a power converter?
• To provide a regulated output voltage

Voltage

Battery (i.e. Li-ion)

Regulated Voltage

Time 2
What are the conventional power converters?
– Low drop-out linear regulator (LDO)
– Switch-inductor regulator (switching regulators)
– Switch-capacitor regulator (charge pump)
Why do we need different Power Converters Types?
– Different applications
– Desired efficiency and output ripple

Can we combine them?

+
Battery SR LDO VREGULATED
-

+
Battery CP LDO VREGULATED
-

+
Battery LDO CP VREGULATED
-

What is the purpose of combining several converters? 3

 Linear Regulator: Principles

+
RC – Vo must be constant and R  R
C LOAD

RLOAD VO
VBAT – VBAT is changing as a function of time

RLOAD
-
VO  VBAT
RLOAD  RC
Thus in order to keep constant Vo, the value of the controlling resistor
RC yields:
V  V V  V 
R  R  
C
 1  R  
LOAD
BAT
  R  
LOAD
BAT
 O
LOAD
LDO

V  O  V  V 
O O

How can we automatically pick the value of RC such that Vo= Vdesired, reg-
voltage? RC
+

VC
Feedback RLOAD VO
VBAT
Control
4
-
 How can we implement RC and the Feedback
Control?
ID

NMOS Transistor
VGS
PMOS Transistor
a b a b
a b

VC = VGS ILOAD VC = -VGS = VSG

VDO,n= VSAT+Vgs Vdo = ILOADRC VDO,p= VSD(SAT)

Can we implement RC as follows? For which applications?

RC1 RC2 RC1 RC2
a b a b

VC MN MP VC

MN and MP are transistors operating in the Ohmic region. Discuss the option5.
How the feedback control could be implemented?
VO VO

Error Amplifier R1 Error Amplifier R1

VC,PMOS VC,NMOS
R2 R2

VREF VREF

• Remarks
– Make sure the closed loop is negative
– For an ideal op amp gain, the differential input is zero, i.e.

R  R 
V V2
0 OR V V  V  1  1
V
R R
O UT REG O REF
 R 
O REF
1 2 2

– VREF is a Bandgap voltage which is also supplied by VBAT=VIN.

6
 The efficiency is defined as:
P V I V V V
 
OUT OUT LOAD
  OUT IN DO

PIN
V I IN LOAD
V V
IN IN

V
 1 DO

V BAT

• Where VDO is the voltage drop across the pass transistor,

i.e. VDS
• The output error voltage (EVO) is defined as:
R
E  C
100%
R R
VO
C LOAD

OR

V V
E  VO
OUT  MAX OUT  LOAD
100%
V OUT  MAX

7
 LDO Analysis
• Let us analyze the basic LDO architecture. First, we will consider ideal
components, then the non-idealities are introduced together with the
accompanied design challenges to tackle

VIN

Error Amplifier
Ref VX gm( Vx – ViN ) rop
VIN = VBAT PMOS Pass Transistor
AEA VIN AEA( VDIV - VREF) Vo

R1
R1
Io
VDIV VDIV RL

Load (RL)
R2 R2

Basic LDO Topology Small Signal Representation

1 1 1  1  1 

VO      VIN   g m   VDIV   g m AEA   g m AEAVREF
 (1)
 rop RL R1   rop   R1 

1 1 V
VDIV     O  0 (2)
 R1 R2  R1 8
Solving the (1) and (2), Vo becomes:

Vin (1  APT )   V REF APT AEA 

Vo 
rop
 [1  APT AEA  ]
RL
Where: APT  g m rop ,   R2 /( R1  R2 ), and ( R1  R2 )  RL

Thus Vo can be expressed as:

Vin APT  V A A 
Vo   REF PT EA
 (1  APT AEA )  (1  APT AEA )

VinT / AEA V T
If T  APT AEA  Vo yields: Vo   REF
 (1  T )  (1  T )

T is the open loop gain. Furthermore for T >>1

Vin V
Vo   REF
AEA  
9
Observe that Vin is attenuated by AEA and Vref is not.
 Line Regulation
The line regulation is a steady-state specification. It can be defined as:

Vo APT APT

LR   
Vin  (1  APT AEA ) 1  APT AEA

V 1
L  
o

V A
R
in EA

For a practical case with non-idealities such as offset Op-Amp voltage Vos

and reference voltage error i.e. Vref ; the line regulator becomes:

Vo 1  R  V  Vos 
  1  1  REF 
Vin AEA  R2  Vin 

Observe that designers should also minimize: Vos

and provide Vref to be independent of VBAT and temperature and process variations.
10
Note that Rc given in page 3 for a transistor can be expressed as:
VDS V  V  Vo 
RC   RLoad DS  RLoad  BAT 
I OUT Vo  Vo 

In order to maintain the regulation the transistor must operate properly

NMOS case
Vin  VDS  VC   ID  Io 

Exercise: Repeat the above case for a PMOS case

11
 Load/Line Regulation
Let us assume the error amplifier is a transconductance amplifier of
value GEA and α is the current gain of the pass transistor i.e.

Thus Io

Vo  I o Req  I o ( R1  R2 ) // RL 
Vo  I o RL R1

Furthermore: GEA ∆VDIV RL

RL
I o  G EAVDIV Req  G EA R2 /( R1  R2 )Vo
Vin Vo+∆Vo

Then, the load regulation can be expressed

as: R2
Vo 1  R1 
   1 VREF
I o GEA  R2 
Or line regulation:
Vo 1  R1  1
   1
Vin GEA  R2  RDS  RL 12
 Efficiency Calculation

Example of efficiency: A 3.3V LDO with 3.7 V < Vin < 4.71V, 100mA < Io < 150mA
Io,q (maximum quiescent current) = 100 μA

I o max  Vout,nominal

( I o max  I o ,q )  Vin,max

150mA  3.3
  100  70%
(150mA  100A)  4.71
The output current can be represented as a pulse for simulation purposes

Io,max OR Io,max

Io,min Io,min

t0 t1 t0 t1 t3 t4
13
 LDO ESR Stability
One of the most challenging problems in designing LDO is the stability problems
due to the closed loop and the parasitic components associated with the pass
transistor and the error amplifier. In fact to compensate the loop stability a large
external capacitor is often connected at the output. i.e.

Vo 1  s / wL
Z CL ( s) 
sC L Im
RESR
Re
-wL
CL
w L  1 / RESRC L

Where CL is of the order of μF with a small equivalent series resistor (RESR).

14
 LDO Parameters 1
• Dropout voltage (Vdo); This is the difference between the minimum
voltage the input DC supply can attain and the regulated output
voltage.
• Input rail range; This is the input supply voltage range that can be
regulated. The lower limit is dependent on the dropout voltage and
upper limit on the process capability.
• Output current range; This is the output current handling capability
of the regulated output voltage. The minimum current limit is mainly
dependent on the stability requirements and the maximum limit
dependent on Safe Operating Area (SOA) of pass FET and also
maintaining output voltage in regulation.
• Output capacitor range; This is the specified output capacitance the
regulator is expected to accommodate without going unstable for a
given load current range.
• Output regulated voltage range; This is the output voltage variation
the regulator guarantees. When output voltage is in this range, it is
said to be in regulation.
• Load regulation; This is the variation in output voltage as current moves from min to
max

15
 LDO Parameters 2
• Line regulation; This the variation in output voltage as supply
• voltage is varied from minimum to maximum.
• PSR; Power Supply Rejection ( or ripple rejection) is a measure of the
• ac coupling between the input supply voltage on the output voltage.
• Load/Line transient regulation; This is a measure of the response speed
• of the regulator when subjected to a fast load/Vsupply change.
• Short circuit current limit; This is the current drawn when the
• output voltage is short circuited to ground. The lower limit is
• determined by the maximum regulated load current and the upper
• limit is mainly determined by the SOA and specified requirements
• Power Efficiency; This is the ratio of the output load power
• consumption to input supply power. Linear regulators are not really
• efficient especially at high input supply voltages.
• Overshoot: It is important to minimized high transient voltages at start-up
and during load and line transients.
• Thermal Shut down: This is needed to protect the part from damage

16
 Conventional LDO: Modeling
Close Loop Schematic Open Loop Transfer Function: TF  H1  H 2  H 3  H 4
VIN
g R  g m3 R A
H  H2 
R A CGS   AV  1CGD  s  1
m1 p

Error Amplifier PND1CGs R C s 1

1
p p
VREF PD
C Rds  g mp RPAR RESRC L   s  RPAR  H4 
R2
H3  R2  R1
PND2
Vout
RPAR  RESR C L  s  1
B CGD R1 CL 1
Av  g mp  R par RPAR  Rds R1  R2 RL Rds 
A D RL I ds
z
1EH RPAR : Output impedance
1EF R2 RESR
RL : Load resistance
VTest
RDS : Drain to source impedance of the pass transistor
Loop Breaking Point for Stability Analysis R1 &R2 : Feedback Resistors

Open Loop Block Diagram Vdropout  200mV  VDSSATPass  200mV

Vdropout : Minimum voltage drop across the input and output
terminalsof the LDO with shich the system is able to regulate.
VDSSATPass : Vdsat of the pass transistor

W  W 
 p Cox  VDSSATPass2    
1 2I D
ID 
2 L  L   p Cox  VDSSATPass2

W 
Note: The error amplifier is a two-stage amplifier without miller compensation
g  2  C  i 17
L
mp p ox load
 Error Amplifier AEA
ROUT

Dominant pole

Cgate

R
01

Two stage amplifier without miller compensation

A g R g R
o m1 01 m8 out

g R g
GBW  m1 o1 m8

Notes: C gate

1. CGATE is connected to output node.

2. Miller Compensation is not required since the dominant pole
is at the output.
18
 Matlab/Simulink Macromodel
Pass Transistor Dominant Pole / Compensation Zero
Transconductance

Non-dominat Pole
due to Pass Transistor
Gate Capacitance Error Amplifier
Nondominant Pole Feedback
Factor

Formulas and Component Values

R2  44e  3 
K C  10F Cgd  30 pF Gm1  15.62 A / V g  80e  6   i
R1  R2 o mp
 0.6e  6 
load

R1  420 K Vdsat  200 mV Cgs  50 pF R p  64k Vout

RL 
Ra  7.36 M C p  100 fF iload
R2  240 K iload  50 mA
1 Vout  3.3V
resr  5   0.1 Gm3  1.35mA / V R   19
DS
  i load

Simulation Results From Simulink
z
Loop Gain and Phase
P ND 2
P
ND 1
P
D

Loop Gain ~ 100dB

Fu ~ 2.7MHz

 1 
Phase boost due to Compensation Zero  Z  
 C R
L ESR 

PM ~ 90°

20
Open Loop Gain and Phase Under Load Variation

Notes:
1-Open Loop System
2- Load Variation (iload varies from 10mA to 50mA)
3- The load variation (iload) was simulated in Matlab using a “for loop”

21
Simulation Results From Matlab
Step Response

Vout varies from 0 to 3.3V

Vin varies from 0 to 3.5V

22
 Power Supply Rejection
Vsupply

 Problem:

 Low frequency and high frequency noise affects the operation of the highly
sensitive circuits

 External noise is mainly coupled through the supply lines

 A regulator (LDO) is mandatory with high PSR

23
 Power Supply Rejection Existing Solutions
VIN VIN
VIN
VVDD
IN

 Current Solutions:
 RC filtering: Larger drop-out voltage, and larger power consumption
 Cascading of LDO: Larger area, power consumption, larger drop-
out voltage
 Combined RC filtering and cascading: Larger area and power
consumption, larger drop-out voltage and complexity

24
 Enhancing PSR over a wide frequency range

Proposed Topology

• The NMOS cascode, MNC, shields

the entire regulator from fluctuations
in the power supply.

• MNC gate needs to be biased at a

voltage above the supply using a
charge pump.

• MNC acts as a voltage follower for

noise at its gate, it is critical to
shield the gate of MNC from supply
fluctuations using an RC filter to
shunt supply ripple to ground.

G. A. Rincon-Mora, V. Gupta, “A 5mA 0.6mA CMOS Miller-Compensated LDO Regulator with -27dB Worst-Case
Power-Supply Rejection Using 60pF of On-Chip Capacitance ,” ISSCC, feb. 2007. 25
 Enhancing PSR over a wide frequency range
• With the help of an NMOS cascode, a charge pump, a voltage reference and
an RC filter to shield the entire regulator from power supply fluctuations, a
5mA LDO regulator utilizing 60pF of on-chip capacitance achieves a worst-
case PSR performance of -27dB over 50MHz.

G. A. Rincon-Mora, V. Gupta, “A 5mA 0.6mA CMOS Miller-Compensated LDO Regulator with -27dB Worst-Case
Power-Supply Rejection Using 60pF of On-Chip Capacitance ,” ISSCC, feb. 2007.
26
 Stability and PSR Simulations
• Stability test (Open Loop)
Frequency Response

Open loop gain shows a low pass frequency response

AC signal is injected here Loop Gain ~ 65dB Fu ~ 10MHz PM ~ 60°

27
 Stability and PSR Simulations (Continue)
• PSR simulation
AC signal is injected here PSR versus Frequency

PSR curve shows a Quasi-Band Pass frequency

response
it is less than -40 dB up to 10KHz, -30 dB at 100 KHz

28
 Different Compensation Techniques for Stability
Purposes
• Internal zero generation using a differentiator
– An auxiliary fast loop (differentiator) provides both a fast transient detector path
as well as internal ac compensation.
– The simplest coupling network might be a unity gain current buffer.
– Cf senses the changes in the output voltage in the form of a current that is then
injected into pass transistor gate capacitance.

Robert J. Milliken, Jose Silva-Martínez, and Edgar Sánchez-Sinencio “Full on-chip CMOS low-dropout voltage regulator,”
29
IEEE Trans. on Circuits and Systems – I, pp 1879-1890, vol. 54, Issue 9, Sept. 2007.
 Different Compensation Techniques
• Capacitive feedback for frequency compensation
– It introduces a left hand plane zero in the feedback loop to replace the zero generated
by ESR of the output capacitor.
– the capacitor is split into two frequency-dependent voltage-controlled current sources
(VCCS) and grounded capacitors.
– Instead of adding a pole–zero pair with zero at lower frequency than the pole, in this
technique only a zero is added.
– It needs a frequency dependent voltage control current source (VCCS).

30
Chaitanya K. Chava, and Jose Silva-Martinez, “A frequency compensation scheme for LDO voltage regulators,”
IEEE Trans. on Circuits and Systems – I, vol. 51, No.6, pp. 1041-1050, June 2004.
 Different Compensation Techniques
• DFC frequency compensation
– It is a pole-splitting compensation technique especially designed for compensating
amplifier with large-capacitive load.
– DFC block composed of a negative gain stage with a compensation capacitor Cm2,
and it is connected at output of the first stage. Another compensation capacitor Cm1
is required to achieve pole-splitting effect.
– The feedback-resistive network creates a medium frequency zero for improving the
LDO stability.

31
K. N. Leung, and P.K.T. Mok, “A capacitor-free CMOS low-dropout regulator with damping-factor-control frequency compensation,”
IEEE J. Solid-State Circuits, vol.38, no.10, pp.1691-1702, Oct. 2003.
 Different Compensation Techniques
• Pole-zero tracking frequency compensation
– To have pole-zero cancellation, the position of the output pole po and compensation
zero zc should match each other.
– The resistor is implemented using a transistor Mc in the linear region, where its
value is controlled by the gate terminal.

K. C. Kwok, P. K. T. Mok. “Pole-zero tracking frequency compensation for low dropout regulator,”
2002 IEEE International Symposium on Circuits and Systems, Vol. IV, pp. 735-738,May 2002.
32
 Fast Transient Response
• A current-efficient adaptively biased regulation scheme is implemented using a low-
voltage high-speed super current mirror. It does not require a compensation
capacitor.
• The adaptively error amplifier drives a small transconductance (MA9) to modulate the
output current IOUT through a transient-enhanced super current mirror.

Yat Lei Lam, Wing-Hung Ki, “A 0.9V 0.35µm Adaptively Biased CMOS LDO Regulator with Fast
Transient Response,” 2008 IEEE International Solid-State Circuit Conference, February 2008.
33
 Noise in Linear Regulators
• LDO noise is sometimes confused with PSRR
 PSRR is the amount of ripple on the output coming from the ripple of the input.
 On the other hand, noise is purely a physical phenomenon that occurs with the transistors and resistors
(ideally, capacitors are noise free) on a very fundamental level.

• Primary noise sources:

1. Bandgap-Reference (Primary source of noise)
 Possible solution: add a low-pass filter (LPF) to the output of the bandgap.
– Drawback: it can slow down the output startup.
2. The resistor divider ( thermal noise = 4KTR )
 Possible solution: Use resistors as small as possible.
– Drawback: Smaller resistors burn more current through the feedback divider.
 Possible solution: add a capacitor across the top resistor in the resistor feedback divider. At high
frequencies it reduces the close loop gain and thus the noise.
– Drawback: it could slow down start-up time significantly, since the capacitor would have to be
charged by the current in the resistor divider.
3. Input stage of the error amplifier (thermal and flicker noise)
 Possible Solution: Large input drivers.
– Drawback: larger area.

• Second order noise sources:

– Load current and output capacitance (Phase Margin)
• Low phase margin will cause peaking in the close-loop gain and since the close-loop gain amplifies this
noise, the total output noise increases.

34
Reference: J. C. Teel, “Understanding noise in linear regulators” Analog Application Journal, 2Q 2005, www.ti.com/aaj
 LDO Design Example
Conventional LDO
Parameters Specifications

VIN 3.5V

Error Amplifier VOUT 3.3V

ILoad 0mA-50mA
Pass Transistor
VREF
VOUT IQ < 100µA

Technology 0.5µm CMOS

R1
CL
RL

R2
RESR

35
 Design Flow Diagram

Vin Vout ILoad

Vdropout RL

W/L Design Error Amplifier

CGate RDS Cp Rp RA

Calculate non dominant poles: PND1, PND2

Adjust poles and zeros locations using CL and RESR

Check stability

Check transient, PSR, load regulation, line regulation, …

36
 LDO Design Example
Since Vdropout  200mV  VDSSATPass  200mV

W  W 
 p Cox  VDSSATPass2    
1 2I D
ID 
2 L  L   p Cox  VDSSATPass2

Assuming µPCox= 65µA/V2 the pass transistor size can be calculated by:

W
 38,462
L
In order to minimize the gate capacitance, we use minimum length L = 0.6µm
W  38462  0.6m  23mm

The gate capacitance of the pass transistor is given by the following equation:

C  C  g  R  1C
gate GS mp par GD

2 1
where C  W  L  C C W  L C RPAR  Rds R1  R2 RL Rds 
GS
3
ox GD D ox
I ds

37
 Design Example (Continue)
The values of CGS and CGD can be also obtained if we run a dc simulation and verify the operating
point of the pass transistor. Using the last method, we found:

1
R  DS
 15 C  26 pF
GS
C  7.6 pF
GD
g ds

 R 
R1 and R2 are calculated using : V
OUT
 1  2
  V
REF
 R 1 

Assuming a reference voltage of 1.2V and choosing R1 = 240K, We found R2 = 420K .

V 3.3V
R 
L
OUT
  66 R  15 420 K  240 K 66  12
PAR
I LOAD
50mA

W   23e  3 
G  2  C  i  130e  6     50e  3  50mA / V
2

L  0.6e  6 
mp p ox load

C gate
 C  g  R  1C  36.5 pF
GS mp par GD

38
 Design Example (Continue)
Error Amplifier Design and Considerations
– High DC Gain to guarantee high loop gain over the range of loads (AV > 60dB)

– Low output impedance for higher frequency pole created with CGS of pass transistor

– Internal poles must be kept at high frequencies, preferably > fU of the system (~1MHz)

– Low DC current consumption

– Low Noise

Error Amplifier schematic

The error amplifier is implemented using a two stage without miller compensation topology
in order to achieve a gain larger than 60dB and GBW =7MHz using 0.5µm CMOS technology
39
 Design Example (Continue)
Error Amplifier Simulation Results
Magnitude Plot Phase Plot

A  65dB
o
GBW  7.6MHz PM  51

Note: This results were obtained using Cgate = 36.5pF as the load
40
 Design Example (Continue)
z
Pole / Zero Locations
P
ND 2
P
ND 1
P
D

Dominant Pole Location Second non-dominant pole location

1 1
P   1.2 KHz P   7.3MHz
2 R  R C
ND 2
D
PAR ESR L
RC p p

First non-dominant pole location Zero location

1
P 
1
 5.3KHz Z  3.2 KHz
ND 1
RC A Gate
R C
ESR L

Notes:
1- RA was obtained from simulations. Basically, it is the output resistance of the error amplifier.
2-PND2 is greater than 7.3MHz since the phase margin of the amplifier is around 51°. This is
good news since we want this pole to be located above the gain bandwidth product of the
overall system.
3- RESR equal 5 was chosen for stability reasons (see next slide). 41
 Design Example (Continue)
Stability versus RESR

UGB versus RESR Phase margin versus RESR

Note: RESR was swept from 0 to 10

42
 Design Example (Continue)
System Simulation Results
Magnitude Plot (IL=50mA) Phase Plot (IL=50mA)

Phase Margin = 55°

DC Gain = 74dB

UGB = 6.3MHz

Magnitude Plot (IL=100µA) Phase Plot (IL=100µA)

Phase Margin = 90°

DC Gain = 75dB

UGB = 172KHz

43
 Design Example (Continue)
System Simulation Results
VIN Step Plot IL Step Plot

V  3.30352V
V  3.30206V
2
4

V  3.30238V
3

V  3.30238V
1
V  V 1.17mV
Load Regulation = 2
 1
 0.0234V / A
I I
L2
50mA
L1

V V 313V
Line Regulation =
2
 1
 0.0031V / V
I I
L2
100mV
L1

Notes:
1- VIN step from 3.4 to 3.5V
44
2- ILOAD step from 0 to 50mA
Design Example (Continue)
System Simulation Results

PSR versus Frequency

PSR @ 100KHz = -35dB

45
Summary of the Results
Parameters Results

VIN 3.5V

VOUT 3.3V

ILoad_max 50mA
ILoad_min 0mA
IQ 30µA

PSRR@100KHz -35dB

Line Regulation 0.0031V/V

Load Regulation 0.0234V/A

TR 1µs

Technology 0.5µm CMOS

46
 Current Efficient LDO

For low IL, for R1+R2 >> Ro_pass Note that RL is significantly larger Loop Gain:

1 I L Ro _ pass //( R1  R2 )  Ro _ pass Av  AEA g mp Ro _ pass

w p1  
Ro _ passCo Co 2 KI L
 W   p Cox Av  AEA
K   I L
 L p 2
For no-load current
2 K
Av  AEA
w p3 
g mQ1

2 KI bias  IL
C par C par AEA  g ma Roa 47
 Current Efficient LDO
GB  Avw p1  AEA g mp Ro _ pass / Ro _ passCo
GB  AEA g mp / Co  2 AEA KI L / Co  2 g ma Roa KI L / Co

Note that GB will change as I L changes, this effects modify the pole-zero locations and the phase
margin. i.e.,
GBMAX  2 AEA KI L ,max / Co ; w p1 
GBMIN  2 AEA KI L,min / Co ; w p1 

Due to current mirror when load current exist, then we have

2 KI bias  I L / N w p 3,load w p1 GB

w p 3,load  ; make  ,
C par I L I L I L
2
1  C 
 or   par  KN 2 I bias  C par
2 2
NI L  1
C par KN 2 I bias  NI L Co  Co 

1
The zero is located at: w z |Cb Co 
Co RESR
1
Another pole is located at the output of the AEA, w p 2 
RoaC pa
G. A. Rincon-Mora, P. E. Allen, “A low-voltage, low quiescent current, low drop-out regulator,”
48
IEEE J. Solid-State Circuits, vol.33, no.1, pp.36-44, Jan. 1998.
To determine the stability one can consider the open loop gain defined as:
Aopen

 s 
A 1   Avo

 w 
vo
v
A  
fb z

 s  s  
open
v s
ref
1  1  1  
 w  w  w 
  
p1 p2 p3 

where wp1 wz wp2 wp3

1
Avo  g ma Roa g mp Ro _ pass w w w w 
p1 z p2 p3
R
1 2

R 1

Stability imposes the following conditions: A ( jw )  1

open o

Where wo is defined as: ARG  A ( jw )  0

open o

Or A ( jw )  1
open o
where w  GB
o

Phase Margin:   180  ARGA  jw 

M

open

w  w  w  w 
  180  tan   1
  tan 1
   tan 1
   tan 1
 
w  w  w 
w  49
M
z  p1  p2   p3 
 For w  w  GB ; 
o M is the arbitrary phase margin

w   GB   GB 
 180    tan  1 o
  tan  A   tan
1 1
 
 w   tan
1

w 

w
M vo
z   p2  p3

In order to determine an approximated relation between wp3 and GB several assumptions are
required.

Avo >> 1, Then tan-1(Avo)  90°

wo Avow p1 R
  ESR Avo
wz wz Ro _ pass

wo RoaC pa
 Avo
wz Ro _ passCo

Thus one can write

 R   RoaC pa   GB 
 90  MX  tan 1  ESR Avo   tan 1  Avo    tan 1  
R  R C  w 
 o _ pass   o _ pass o   p3 
50
 Cadence Simulation

• This current efficient LDO is implemented using 0.5 μm CMOS process

• The LDO provides a 3.3 V regulated output voltage from a 3.5 V supply, for
a load current IL ranging from 250 μA to 25mA

• The current in the buffer ranges from 20μA (which is only the bias current) in
case of ILmin to 180µA in case of ILmax.

• The output voltage for load

current alternating between ILmin
and ILmax is shown.

51
 Cadence Simulation

iload  25mA
Vout  3.3V

Loop Gain ~ 52dB

Fu ~ 0.78MHz
PM ~ 82°

1
RESR  100m C  10F wz   1M
o
RESRCo
1
Ro _ pass  8.5 C  10F w p1   11.8K
o
Ro _ passCo
1
Roa  900k C pa  1 pF w p2   1.1M
RoaC pa
1
R par  400 C par  3 pF w p3   833.33M 52
R parC par
 Cadence Simulation
• ωz cancels ωp2 and the phase margin can be approximated to:

 GB 
M  90  tan 
1   90  tan 1  2  0.78   89o

 w p3   833.33 

• Which is close to the simulated value of 82o

• The step response for this

case of ILmax is shown where
Vin varies from 0 to 3.5V and
accordingly Vo varies from 0
to 3.3V

• For the case of ILmin of 250µA, the dominant pole becomes even smaller and
very far away from ωp3 and so the phase margin is almost 90o
53
Simulation Results From Simulink
Loop Gain and Phase

Fu ~ 5.4MHz
Loop Gain ~ 51.8dB

PM ~ 90°

54
Open Loop Gain and Phase For Different
wp3 Locations

PM  90°
wp3  100GBW

PM  45°
wp3  GBW

Notes:
1-Open Loop System
2- The location of wp3 was varied
3- The variation (wp3) was simulated in Matlab using a “for loop” 55
Current Limiters Architecture

Courtesy of Tuli Dake from TI

56
ILIM Op Amp

57
 References
[1] G. A. Rincon-Mora, V. Gupta, “A 5mA 0.6mA CMOS miller-compensated LDO regulator
with -27dB worst-case power-supply rejection using 60pF of on-chip capacitance ,”
ISSCC, Feb. 2007.
[2] L.-G. Shen et al., “Design of low-voltage low-dropout regulator with wide-band high-PSR
characteristic,” International Conference on Solid-State and Integrated Circuit
Technology, ICSICT, Oct. 2006.
[3] R. J. Milliken, J. Silva-Martínez, and E. Sánchez-Sinencio, “Full on-chip CMOS low-
dropout voltage regulator,” IEEE Trans. on Circuits and Systems – I, pp 1879-1890, vol.
54, Issue 9, Sept. 2007.
[4] C. K. Chava, and J. Silva-Martinez, “A frequency compensation scheme for LDO voltage
regulators,” IEEE Trans. on Circuits and Systems – I, vol. 51, No.6, pp. 1041-1050, June
2004.
[5] K. N. Leung, and P.K.T. Mok, “A capacitor-free CMOS low-dropout regulator with
damping-factor-control frequency compensation,” IEEE J. Solid-State Circuits, vol.38,
no.10, pp.1691-1702, Oct. 2003.
[6] K. C. Kwok, P. K. T. Mok. “Pole-zero tracking frequency compensation for low dropout
regulator,” 2002 IEEE International Symposium on Circuits and Systems, Vol. IV, pp.
735-738, May 2002.
[7] J. C. Teel, “Understanding noise in linear regulators” Analog Application Journal, 2Q
2005, www.ti.com/aaj
[8] K. N. Leung, P. K. T. Mok, and W. H. Ki, "A novel frequency compensation technique for
low-voltage low-dropout regulator," IEEE International Symposium on Circuits and
Systems, vol. 5, May 1999.
[9] G. A. Rincon-Mora and P. A. Allen, "A low-voltage, low quiescent current, low drop-out
regulator," IEEE J. Solid-State Circuits, vol.33, no.1, pp.36-44, Jan. 1998.

58