TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 8, August 201
4, pp. 6055 ~ 6062
DOI: 10.115
9
1
/telkomni
ka.
v
12i8.546
4
6055
Re
cei
v
ed
De
cem
ber 2
3
, 2013; Re
vi
sed
May 20, 20
14
; Accepte
d
Ju
ne 5, 2014
Three-stage Amplifier Adopting Dual-miller with
Nulling-resistor and Dual-feedforward Techniques
Zhou Qiann
e
ng
1*
, Li Qi
1
, Li
Chen
1
, Lin Jinzhao
1
, Li Hongjua
n
2
,Li Yunsong
1
, Pang Yu
1
,
Li Guoquan
1
, Cai Xuemei
1
1
Colle
ge of Ele
c
tronic Eng
i
ne
erin
g, Chon
gq
i
ng Un
iversit
y
o
f
Posts and T
e
l
e
commu
nicati
o
n
s,
Cho
ngq
in
g 40
0
065, Ch
in
a;
2
Colle
ge of Co
mputer Scie
nc
e and T
e
chno
l
o
g
y
, Cho
n
g
q
in
g Univ
ersit
y
of Posts and T
e
le
communic
a
tio
n
s
,
Cho
ngq
in
g 40
0
065, Ch
in
a
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: zhouq
n@cq
u
p
t.edu.cn
A
b
st
r
a
ct
A high-
gai
n w
i
de-b
andw
i
d
th three-sta
ge a
m
plifier, w
h
ich e
m
p
l
oys d
ual-
m
iller co
mpe
n
sa
tion w
i
th
null
i
n
g
-resistor
and d
u
a
l
-fee
dforw
a
rd co
mpens
atio
n
(DM
CNR-DFC), is
desig
ne
d an
d
analy
z
e
d
i
n
this
pap
er. By ado
pting the tec
h
niq
ue
of DMC
NR-DF
C, the desi
gne
d thre
e-stage a
m
plifi
e
r achi
eves w
e
l
l
perfor
m
a
n
ce
i
n
clu
d
in
g
gai
n-
ban
dw
idth
pro
duct (GBW
) a
nd s
l
ew
rate
(SR). T
he
impr
oved
DMC
NR-
DF
C
three-stag
e
a
m
p
lifier
is
des
i
gne
d
and
si
mulate
d i
n
0.35
μ
m BCD
pr
oc
ess. Si
mu
latio
n
res
u
lts s
how
that
DMCNR-
DFC three-sta
ge
a
m
plifier
ac
hiev
es
a dc
ga
in
of
a
bout 121.
1dB and
GBW
of a
bout 6.1MH
z
w
i
th
52º ph
ase
mar
g
in us
ing
a 5-V
pow
er supply
voltag
e.
Ke
y
w
ords
: du
al-
m
il
ler co
mp
ensati
on w
i
th null
i
n
g
-resistor
,
dual-fe
edforw
a
rd co
mp
ens
a
t
ion, three-sta
g
e
amplifi
e
r
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
The o
p
e
r
atio
nal am
plifier is
a fun
d
a
m
ental b
u
ildi
ng bl
ock i
n
the an
alog
i
n
tegrate
d
system
s and
mixed-sign
al
system
s [1
-4]
.
With th
e d
e
v
elopment
of
mod
e
rn
CM
OS technol
o
g
y,
the cha
nnel
lengths
an
d sup
p
ly voltages of
M
O
SFET tran
sisto
r
s
are
scalin
g do
wn,
and
conve
n
tional singl
e-stage
ca
scode am
p
lifiers
are
not
suita
b
le fo
r
obtainin
g
hig
h
d
c
gai
n a
n
d
large
outp
u
t-swi
ng
simulta
neou
sly. In fa
ct, the to
p
o
lo
gy of thre
e-st
age
amplifier
is a
goo
d tra
d
e
off
betwee
n
dc
g
a
in,
ban
dwidth and power con
s
u
m
ptions. Ho
wever,
th
ree
-
stage amplifi
e
rs
suffer from the close-loop stability problem be
cause they hav
e multip
le-poles and zeros.
Therefore,
th
e fre
que
ncy
co
mpe
n
sation te
chni
que
sh
ould
be
analyzed
an
d di
scusse
d
to
improve the
stability of thr
ee-stage am
pli
f
ier system.
In the re
ce
nt past m
any
years, m
any
fr
equ
en
cy compen
satio
n
techni
que
s
of three
-
stage a
m
plifiers
have bee
n repo
rted [5
-16]. For the
architectu
re
of three-stag
e amplifier, th
e
nested Miller compensati
on (NMC
), whi
c
h can provide a
wel
l
stability, is a well-known
comp
en
satio
n
techniq
ue [
9
]. Ho
weve
r, the
NM
C
th
ree-stage
am
plifier
suffers from
en
orm
o
u
s
power
con
s
u
m
ption an
d g
a
in-b
and
widt
h limitati
on when the
num
ber of g
a
in
stage
s in
cre
a
ses,
and
which ha
s a ri
ght half
plane
(RHP)
zero. Mor
eov
er, the gai
n-b
and
width p
r
o
ductio
n
of NMC
three
-
sta
ge
a
m
plifier i
s
onl
y one
qua
rter of that
of
a
single-stag
e a
m
plifier [1
0]. Based
on
NM
C,
many co
mpe
n
satio
n
topol
ogie
s
have b
een repo
rt
ed
to improve th
e band
width
and the
stabil
i
ty
of three-stag
e amplifier, such a
s
ne
ste
d
Gm-C com
pen
sation (NGCC) [11], active feedba
ck
freque
ncy
co
mpen
sation
(AFFC) [1
2], dual
a
c
ti
ve-cap
acito
r
act
i
ve-feedb
ack
co
mpe
n
sation
(DA
C
FC) [1
3], impedan
ce adaptin
g comp
en
satio
n
(IAC) [14],
cro
s
s feedf
orward ca
scade
compensation (CFCC) [
7
], current
buffer M
iller compensati
on (CBMC)
[6], AC boosting
comp
en
satio
n
(ACB
C) [1
5
], reverse ne
sted M
ille
r co
mpen
sation (RNM
C)
[16],
and singl
e-Mi
ller
cap
a
cito
r fee
d
forward fre
quen
cy co
m
pen
sati
on
(SMFFC) [8], and so on. I
n
gene
ral, th
ese
comp
en
satio
n
techniq
u
e
s
ba
sed
on
NMC
achieve
well
perfo
rm
ance. However, the
freq
u
ency
comp
en
satio
n
techni
que
must still be discu
s
se
d and an
alyzed to furt
her imp
r
ove
the
perfo
rman
ce
of three-stag
e amplifier.
In this paper, a three-st
age am
plifier,
whi
c
h
adopts
dual-miller com
pensati
on with
nulling
-re
si
stor an
d dual
-feedforwa
rd
compen
sati
o
n
(DM
C
NR-DF
C
), is d
e
si
gn
ed and a
naly
z
ed.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 605
5 –
6062
6056
Section 2 will
discuss the
conve
n
tional
NMC th
ree
-
st
age amplifie
r. The improve
d
DMCNR-DFC
three
-
sta
ge
amplifier
will
be a
nalyze
d an
d de
sig
ned in
Se
ction 3. Se
ctio
n 4
will give
the
simulatio
n
re
sults of three-stage am
plifier.
Finally, co
nclu
sio
n
s a
r
e
given in Section 5.
2. Analy
s
is o
f
Conv
ention
al
NMC Thr
e
e
-s
tage
Am
p
lifier
Figure 1
sho
w
s the to
polo
g
y of NMC th
ree
-
sta
ge
am
plifier [10]. A
v1
, A
v2
and A
v3
form the
basi
c
th
re
e-stage a
m
plifie
r, an
d thei
r
t
r
an
scond
ucta
nce
s
are
exp
r
esse
d by
g
m1
, g
m2
and
g
m3
r
e
spec
tively. C
m1
and
C
m2
are
the
com
pen
sation
ca
pacito
r
, an
d
R
L
an
d
C
L
are
,
r
e
s
p
ec
tive
ly
,
load-re
si
stor
and loa
d
-cap
acitor. r
o(1,2)
a
nd C
o(1,2)
are
the equivale
nt output re
si
stan
ce an
d the
lumped
capa
citan
c
e
re
sp
ectively. Figu
re 2
i
s
th
e
tran
sisto
r-l
eve
l
circuit
of
NMC th
ree
-
st
a
ge
amplifier.
Th
e first-sta
ge,
se
con
d
-stage
and
la
st-sta
ge of
NM
C t
h
ree
-
stage
a
m
plifier
are
made
up of tran
sist
ors M
1
~M
9
, M
10
~M
13
and M
14
~M
15
res
pec
tively.
As re
po
rted i
n
[10], it is a
s
sume
d that g
mi
r
oi
>>
1,
C
mi
>>
C
oi
, C
L
>>
C
mi
and
g
m3
R
L
>>1, here
i=1, 2. Beside
s, with an ad
ditional condit
i
on that g
m3
>>g
m(1,2)
, the open-l
oop
sma
ll-sig
nal tran
sfer
function of NMC thre
e-sta
ge amplifie
r, whi
c
h is
repo
rted in refe
re
nce [10], ca
n be expre
s
sed
as:
2
21
2
12
3
1
2
32
3
2
23
2
2
23
1
2
1
23
2
3
1
()
11
mm
m
mm
m
o
o
L
mm
m
vN
M
C
mm
m
Lm
mm
o
o
L
m
mm
mm
CC
C
gg
g
r
r
R
s
s
gg
g
As
Cg
g
CC
g
gr
r
R
C
s
s
s
gg
g
g
(1)
As an
alyze
d
in refere
nce [1
0], it is a
s
sum
ed that C
m1
=4(
g
m1
/g
m3
)C
L
and
C
m2
=2
(g
m2
/g
m3
)C
L
, and the
gain-ban
dwi
d
th produ
ct (G
BW) of
NM
C three
-
sta
ge a
m
plifier can b
e
given by:
3
1
4
m
L
g
GB
W
C
(2)
Equation
(1
)
and Eq
uation
(2
) indi
cate
that
NM
C three-stage
am
plifier h
a
s on
e-fou
r
th
GBW of sin
g
l
e-sta
ge am
p
lifier and h
a
s
a RHP zero. The RHP
zero deg
rad
e
s the
stabili
ty
signifi
cantly. To furth
e
r i
m
prove
GB
W an
d
st
abi
lity of three-stage
amplifi
e
r, an i
m
pro
v
ed
topology of three
-
stage a
m
plifier, whi
c
h adopt
s
dua
l-miller
comp
ensation with
nulling-re
sist
or
and du
al-fee
d
f
orwa
rd
comp
ensation (DM
C
NR-DF
C
), i
s
analy
z
ed a
nd de
sign
ed i
n
se
ction 3.
-g
m1
+g
m2
-g
m3
`
r
o1
C
o1
`
r
o2
C
o2
C
m1
C
m2
R
L
C
L
v
1
v
2
v
in
v
out
A
v1
A
v2
A
v3
Figure 1. Top
o
logy of NMC Three
-
stage
Amplifier
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Thre
e-stage
Am
plifier Adopting Du
al-m
iller with
Nullin
g-resi
stor a
n
d
…
(Zho
u Qia
nnen
g)
6057
M
1
M
2
M
3
M
4
M
5
M
6
M
7
M
8
M
9
M
10
M
13
M
11
M
12
M
14
M
15
V
in
-
V
in
+
V
ou
t
C
m1
C
m2
`
V
b1
V
b2
V
b3
`
V
b4
V
b5
Figure 2. Tra
n
si
stor-level
Circui
t of NM
C Three-stag
e Amplifier
3. Analy
s
is a
nd Design o
f
Impro
v
ed DMCNR-DF
C
Three
-stage
Amplifier
The topol
ogy
of improved
DMCNR-DF
C
three-
stag
e
amplifier, whi
c
h ad
opts
du
al-mille
r
comp
en
satio
n
with
nulli
ng
-re
si
stor and
dual-fe
edf
o
r
ward
co
mpe
n
sation, is sho
w
n i
n
Fi
gure
3.
The d
c
gain
of DM
CNR-DFC th
ree
-
sta
ge am
plifier is
reali
z
ed
by
ca
scadin
g
g
a
i
n sta
g
e
s
A
v1
, A
v2
and A
v3
. A
v1
is a
hig
h
g
a
in
stage, A
v2
is
high-gain
an
d
wid
e
-o
utput-swi
ng
se
co
nd
-stag
e
, a
nd A
v3
is
the
la
st stage with wid
e
-outp
u
t-swi
n
g.
A
va
and A
vf
form, re
sp
ectively, a fe
edforwa
rd
st
age.
R
m1
-C
m1
and
R
m2
-C
m2
form
, respectively, miller comp
ensation with nu
lling-resist
or. A
va
-A
v2
and
A
vf
-A
v3
form the p
u
sh-p
ull
se
con
d
-stage
and
the p
u
sh-pull
outp
u
t
stage
of thre
e-sta
ge
ampli
f
ier
respe
c
tively,
so DM
CNR-DFC thre
e-stag
e amp
lifier h
a
s
an imp
r
ove
m
ent tran
sien
t respo
n
se.
-g
m1
+g
m2
-g
m3
-g
mf
`
r
o1
C
o1
`
r
o2
C
o2
C
m1
C
m2
R
m1
R
m2
R
L
C
L
v
1
v
2
v
in
v
out
-g
ma
A
v1
A
v2
A
v3
A
vf
A
va
Figure 3. Top
o
logy of DMCNR-DFC Th
ree-stage Am
plifier
3.1 Transfer Function and Stabilit
y
Analy
s
is of DMCNR-DFC
Thr
ee-stage Amplifier
As sho
w
n i
n
Figure 3,
g
m1
, g
m2
and
g
m3
are,
re
spe
c
tively, the tran
scon
du
ctan
ce
s of th
e
first-stage, se
con
d
-stage a
nd last-stag
e
of DMCNR-DFC th
ree
-
st
age amplifie
r. r
o(1,2)
and C
o(1,2)
are th
e e
qui
valent outp
u
t re
sista
n
ces and
lump
e
d
pa
ra
sitic
cap
a
cita
nce
of gain
sta
g
e
s
respe
c
tively, and R
L
a
nd
C
L
are, resp
ectively, the loadin
g
re
sist
or an
d loadin
g
cap
a
cito
r. g
ma
and g
mf
are t
he e
quivalent
tran
scond
uc
t
ances of fee
d
f
orwa
rd
sta
g
e
A
va
and A
vf
res
p
ec
tively.
To
analyze the stability of DMCNR-DF
C three-stage a
m
plifier in this
pape
r, the op
en-lo
op tra
n
sfer
function
can
be o
b
taine
d
by analy
z
ing
the eq
uivalen
t
small
-
si
gnal
topolo
g
y a
s
sho
w
n
in
Fig
u
re
3. At
the same time, to simplify the tr
ansfe
r f
unctio
n
without losi
ng accu
ra
cy with the goal
of
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 605
5 –
6062
6058
providin
g a cl
eare
r
in
sight
to the DMCNR-DF
C
thre
e-stage a
m
plifi
e
r, the followi
ng assu
mptio
n
s
are rea
s
on
abl
y made.
(1) Th
e loadi
ng ca
pa
citor C
L
is much
greate
r
than
the compe
n
s
ation
cap
a
ci
tors C
m(1,2)
, i.e.
C
L
>>
C
m(1,2)
.
(2)
Th
e comp
ensation ca
p
a
citors
C
m(1,2)
and
the
loa
d
ing
cap
a
cito
r
C
L
a
r
e
mu
ch
greate
r
th
an t
h
e
lumped o
u
tpu
t
capa
citors o
f
each sta
ge, i.e. C
m(1,2)
an
d C
L
>>
C
o(1,2)
.
(3) T
he comp
ensation cap
a
citors C
m(1,2)
have equival
ent cap
a
cita
n
c
e, i.e. C
m1
=C
m2
.
(4) T
he d
c
ga
ins of all stag
e are mu
ch g
r
eate
r
than 1,
i.e. g
m1
r
o1
, g
m2
r
o2
and g
m3
R
L
>>1.
(5) T
he outp
u
t resi
stan
ce
of all stage
are mu
ch g
r
eate
r
than resi
stors R
m(1,
2
)
, i.e. r
o(1
,
2)
and
R
L
>>
R
m(1,2)
.
(6)
To h
a
ve a
symmetri
c
al
push-p
u
ll
out
put stag
e, the
tran
scondu
ct
ances
g
m3
an
d g
mf
are eq
u
a
l,
i.e. g
m3
=g
mf
.
Based
on
tho
s
e
assu
mptio
n
s, the
op
en
-loop
small
-
si
gnal t
r
an
sfer
function
of
DMCNR-
DFC th
ree
-
st
age amplifie
r
can b
e
given
as:
12
3
3
1
234
11
1
11
1
1
1
dc
LHP
L
HP
LHP
v
dB
nd
nd
nd
nd
ss
s
A
zz
z
As
s
s
sss
pp
p
p
p
(3)
12
3
1
2
dc
m
m
m
o
o
L
A
gg
g
r
r
R
(4)
1
22
1
3
1
1/
LH
P
mm
m
m
z
CR
R
g
(5)
32
1
2
11
3
2
1
1
mm
m
LH
P
mm
m
m
gR
R
z
RC
g
R
(6)
12
2
3
3
21
1
2
1/
mm
m
m
LH
P
mm
a
o
m
o
gg
R
g
z
Rg
C
g
C
(7)
3
23
2
1
1
1
dB
mm
o
o
L
m
p
g
gr
r
R
C
(8)
1
22
1
nd
mm
p
RC
(9)
22
2
22
1
1
mm
nd
mo
m
o
gR
p
R
CR
C
(10)
3
3
m
nd
L
g
p
C
(11)
4
11
2
2
1
//
nd
mo
m
o
p
RC
R
C
(12)
1
3
1
m
dc
dB
m
g
GB
W
A
p
C
(13)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Thre
e-stage
Am
plifier Adopting Du
al-m
iller with
Nullin
g-resi
stor a
n
d
…
(Zho
u Qia
nnen
g)
6059
In this topolo
g
y as sh
own
in Figure 3,
R
m1
>>
1/g
m3
, so it is concl
uded that z
LH
P1
<p
nd1
.
The ze
ro z
LH
P1
can can
c
el
the pole p
nd1
. R
m1
and R
m2
are in an order of magnit
ude, and C
o(1
,
2)
and C
m(1,2)
are the lum
ped
cap
a
cita
nce
and
comp
en
sa
tion capa
cit
ance re
sp
ecti
vely. Based
on
the above a
s
sumptio
n
, the relation of ze
ro z
LHP2
and p
o
le p
nd2
can b
e
obtaine
d as:
32
1
22
22
11
3
2
2
2
1
1
1
1
mm
m
mm
LHP
n
d
mm
m
m
m
o
m
o
gR
R
gR
zp
R
Cg
R
R
C
R
C
(14)
Equation
(1
4
)
sho
w
s that
the
zero
z
LHP2
can
ca
nce
l
the p
o
le
p
nd2
. At the s
a
me time,
C
o(1,2)
are th
e lump
ed
ca
pacita
n
ce, so
the
zero
z
LHP3
and the
p
o
le p
nd4
can
be p
u
sh
ed t
o
a
freque
ncy th
at are highe
r than the unity-gain
freq
uen
cy (UG
F
). So,
Equation (3)
can
be
approximated
as:
33
11
dc
v
dB
nd
A
As
s
s
pp
(15)
Equation
(1
5
)
indi
cate
s th
at DM
CNR-DFC th
ree
-
sta
ge am
plifier is a
pproximat
ed to
a
system
with
two
pole
s
. T
h
erefo
r
e, p
h
a
s
e ma
rgin
(PM
)
of
DM
CNR-DFC three
-
st
age
amplifie
r
can
be app
roxima
tely written as:
3
90
t
a
n
nd
GB
W
PM
a
r
c
p
(16)
To e
n
sure
th
e sta
b
ility of
a unity-gain
feedb
ack syst
em, three-sta
ge amplifier
sho
u
ld
have a ph
ase
margi
n
of at least 4
5
º an
d
60º p
r
eferabl
e in mo
st situ
ations. Th
ere
f
ore, pole
p
nd3
sho
u
ld be la
rger than
GBW, and GB
W of DMCNR-
DFC thre
e-stag
e amplifier
ca
n be written a
s
:
13
3
1
mm
nd
mL
gg
GWB
p
CC
(17)
Acco
rdi
ng to
t
he a
bove
ana
lysis, by
ad
op
ting the
DM
CNR-DFC tech
nique, th
e d
e
s
ign
ed
DMCNR-DF
C
three
-
stage amplifier achi
eves wi
de
r
GBW tha
n
t
he
conventio
nal NMC three-
stage am
plifier for a given
load capa
cita
nce
C
L
.
3.2. Transistor-lev
el Circuit of
DM
CNR-DF
C
Three
-stage Ampli
f
ier
`
M
1
M
2
M
3
M
4
M
5
M
6
M
7
M
8
M
9
M
10
M
13
M
11
M
12
M
14
M
15
V
in
-
V
in
+
V
ou
t
C
m1
R
m1
R
m2
C
m2
`
V
b1
V
b2
V
b3
Figure 4. Tra
n
si
stor-level
Circuit
of DM
CN
R-
DF
C Th
ree
-
sta
ge Am
plifier
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 605
5 –
6062
6060
The tran
sisto
r-level
ci
rcuit
of DM
CNR-DFC th
ree
-
sta
ge am
plifier i
s
sho
w
n
in F
i
gure
4
.
Tran
si
st
or
s M
1
~M
9
form
the first gain
stage A
v1
, and tran
si
stors
M
10
~M
13
form
the second
gain
stage A
v2
. A
v3
is made up o
f
transisto
r M
14
. M
15
forms the feedforwa
rd stag
e A
vf
, and
A
v3
and
A
vf
form the
pu
sh-pull
outp
u
t-stage.
Feedf
orward
stage
A
va
is made
up
of tran
si
stors M
1
~M
9
a
nd
trans
is
tor M
13
. At
the s
a
me time, A
va
and A
v3
form the push-pull stag
e at the output of the
se
con
d
g
a
in
st
age
A
v2
. Those
two
pu
sh
-pull
output
st
age
s
ca
n
effe
ctively improve the
sle
w
-rat
e
of DMCNR-DFC thre
e-sta
ge amplifier.
R
m1
-C
m1
and
R
m2
-C
m2
form the compe
n
satio
n
network
r
e
spec
tively.
4. Simulation Resul
t
s
To v
e
rify
the
cir
c
uit
of DM
CN
R-
DF
C th
r
ee-
st
age
am
plifier
sho
w
n
in Figu
re
4,
NMC an
d
DMCNR-DF
C
three-stage
amplifier a
r
e both de
signed an
d simulate
d in
0.35
μ
m BCD
techn
o
logy wi
th a 5-V power su
pply voltage.
The
simulat
ed op
en
-loo
p freq
uen
cy
re
spo
n
se o
f
NMC thre
e-sta
ge
amp
lifier an
d
DMCNR-DF
C
three-stage
amplifier a
r
e,
resp
ecti
vely, shown in Figure 5 an
d Figure 6 und
er
100-pF load
cap
a
cito
r and
25-k
Ω
load resi
stor. The
NMC th
ree
-
st
age amplifie
r achi
eves the
dc
gain of 1
21.2
d
B, phase m
a
rgin
of 54.4º
and
GB
W of
about 1.8
4
M
H
z, b
u
t DM
CNR-DFC th
re
e-
stage
amplifi
e
r a
c
hieve
s
t
he d
c
gai
n
of 121.2dB
,
pha
se ma
rgi
n
of 52º a
n
d
GBW of a
b
out
6.1MHz.
Sim
u
lation re
sult
s sho
w
that DMCNR-D
F
C
three
-
stage amplifier achi
eves wide
r
G
B
W
than NM
C three-stage a
m
p
lifier.
Figure 5. Ope
n
-loo
p Fre
q
u
ency Respon
se of
NMC T
h
re
e-stage Amplifier
Figure 6. Ope
n
-loo
p Fre
q
u
ency Respon
se of
DMCNR-DF
C
Three
-
stage
Amplifier
Figure 7. Tra
n
sie
n
t respon
se of NM
C th
ree
-
stage am
plifier
Figure 8. Tra
n
sie
n
t re
s
pon
se of DM
CN
R-
DF
C
th
r
e
e-
s
t
ag
e
amp
lifie
r
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Thre
e-stage
Am
plifier Adopting Du
al-m
iller with
Nullin
g-resi
stor a
n
d
…
(Zho
u Qia
nnen
g)
6061
Figure 7
and
Figure 8
sh
ow, respe
c
tively, the simu
lated tra
n
si
e
n
t re
spo
n
se
of NM
C
three
-
sta
ge a
m
plifier and
DMCNR-DF
C
thre
e-stage
amplifier in
a
unity-gai
n n
egative fee
d
b
a
ck
config
uratio
n
to a 125-kHz
1-Vpp
pu
lsing in
put
si
gnal. NM
C t
h
ree
-
stage a
m
plifier a
c
hie
v
es
positive sl
ew
rate (S
R
+
) of
1.43 V/
μ
s an
d negative
sl
ew rate (S
R
-
) of 1.39 V/
μ
s. And DMCNR-
DFC three
-
st
age
amplifie
r achieve
s
S
R
+
of
6.3V/
μ
s a
nd S
R
-
of
6.32V/
μ
s. Si
mulation
results
sho
w
that DM
CNR-DF
C th
ree-stage am
p
lifier achi
eves well SR
than
NMC thre
e-stage amplifier.
Finally, perfo
rman
ce
su
m
m
ary of NM
C three-
stag
e amplifie
r a
nd DMCNR-DFC thre
e-st
age
amplifier a
r
e
given in
Ta
bl
e 1.
From
Ta
ble 1,
the DM
CN
R-
DF
C
th
r
ee-
stage
am
p
lifier a
c
hi
eves a
well pe
rform
a
nce.
Table 1. Perf
orma
nce Summary of Three-stage Am
plifier
Ref. [8]
Ref. [14]
NMC thre
e-
stage amplifier
in this paper
DMCNR
-DF
C
thr
ee-
stage amplifier in
this paper
Process 0.18
μ
m CMOS
0.35
μ
m CMOS
0.35
μ
m BCD
0.35
μ
m BCD
Suppl
y
voltage (
V
)
1.5
1.5
5
5
Loading Capacitive (pF)
250
150
100
100
I
dd
(mA
)
0.023
0.02
0.2
0.2
A
dc
(dB)
~100
110
121.2
121.2
GBW (MHz)
4.4
4.4
1.84
6.1
Phase margin
73º
57 º
54.4 º
52 º
Average SR (V/
μ
s) 1.8
1.8
1.41
6.31
5. Conclusio
n
A DMCNR-DFC thre
e-sta
ge amplifie
r, which ado
p
t
s dual
-miller comp
en
sation wit
h
nulling
-re
si
stor and d
ual
-feedforwa
rd t
e
ch
niqu
e,
ha
s bee
n de
sig
ned an
d anal
yzed in this p
aper.
By adopting t
he du
al-mill
er com
pen
satio
n
with
nulling
-re
si
stor and dual-fe
edfo
r
ward
te
ch
nique
,
the RHP zero
can be remo
ved, and the desi
gne
d DM
CNR-DF
C three-stage
a
m
plifier achiev
es
wide
r GBW
a
nd well S
R
th
an NM
C thre
e-sta
ge am
pl
i
f
ier. Simulation results
show that DM
CNR-
DFC th
ree
-
st
age amplifie
r
achi
eves
well
small-sig
nal
and large
-
sig
nal perfo
rma
n
ce
s.
Ackn
o
w
l
e
dg
ements
This work wa
s sup
p
o
r
ted in part by Nati
onal Scie
nce
Foundatio
n of China (Grant No.
6110
2075, a
nd 613
011
2
4
), Natu
ral
Scien
c
e Fo
u
ndation Proj
ect of CQ
CSTC
(Gran
t
No.
CSTCJJA400
11, and cstc2011jjA1
380
), Scientific
and Technol
o
g
ical
Resea
r
ch Pro
g
ra
m of
Cho
ngqin
g
M
unici
pal Ed
ucation
Commi
ssion
(Gra
nt No. KJ1
2050
3, KJ1
2050
7, a
nd K
J
120
533
),
2013 Progra
m
for Innovation Team Bu
ilding at Inst
itutions of Hig
her Edu
c
atio
n in Chon
gqi
ng
(Grant No. G201
3-4
6
), Special
Proj
ect
of
Inte
rn
et of Thin
gs from Mi
nist
ry of Indu
stry an
d
Information
T
e
ch
nolo
g
y, and Chon
gqin
g
Devel
opme
n
t Plan of Inn
o
vative Youn
g Tale
nts
(Grant
No. cstc2
013
kjrc-qnrc0
126
).
Referen
ces
[1]
Yu SH, W
ang KY, W
e
i SP,
W
ang XF
, Y MJ. A Design of Gain Boost
ed Error Ampli
f
ier Appli
ed to
PWM control.
T
E
LKOMNIKA Indons
esi
an Jo
urna
l of Electri
c
al Eng
i
ne
eri
n
g
. 2013; 1
1
(5): 236
5-23
70.
[2]
Yu F
,
Yang HJ
, Li Gang. A 4th-ord
e
r S
w
itch-
c
apac
itor Lo
w
-
Pass F
ilter for Quartz G
y
rosc
ope Interfac
e
Circuit.
T
E
LKOMNIKA Indons
esia
n Journ
a
l o
f
Electrical Eng
i
ne
erin
g
. 201
3; 11(10): 57
18-
572
4.
[3]
Su SJ, Z
h
a
n
g
HL. T
he Stud
y
an
d Ach
i
evi
ng
of Hi
gh-pr
e
c
ision
D
a
ta-ac
quisiti
on
Bas
e
d o
n
∆Σ
ADC.
T
E
LKOMNIKA Indons
esi
an Jo
urna
l of Electri
c
al Eng
i
ne
eri
n
g
. 2013; 1
1
(8): 445
3-44
60.
[4]
Yu F
,
Yan
g
HJ
, Li G. A
Hig
h
Performanc
e S
i
gma-d
e
lta
AD
C for A
udi
o D
e
coder
Ch
ip.
TE
LKOMNIKA
Indons
esi
an Jo
urna
l of Electri
c
al Eng
i
ne
eri
n
g
. 2013; 1
1
(11)
:6570-
65
76.
[5]
Aamir SA, Harikumar P, Wikner JJ.
F
r
eq
uency
co
mp
en
sation
of h
i
g
h
-spee
d, low
-
vol
t
age
CMO
S
mu
ltistag
e
a
m
plifiers
.
201
3 I
EEE Internati
o
nal S
y
m
posi
u
m on C
i
rcuits
and S
y
stems.
Beiji
ng. 2
0
1
3
:
381-
384.
[6]
Yan Z, Mak PI, La
w
MK, Martin
s PR.
Ultr
a
-are
a-efficie
n
t three-sta
ge
a
m
plifier
usi
ng
current b
u
ffer
miller com
p
e
n
s
a
tion a
nd p
a
ral
l
el com
pens
ati
on.
Electron
ics
Letters
. 2012;
48(1
1
): 624-
62
6.
[7]
Cho
ng SS, Ch
an PK. Cross feedfor
w
a
rd c
a
scode
com
p
e
n
s
ation for l
o
w
-
po
w
e
r three-st
age am
plifi
e
r
w
i
t
h
lar
ge cap
a
c
itive lo
ad.
IEEE Journal of Solid-St
a
te Circ
u
its
. 2012; 4
7
(
9
): 2227-
22
34.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 8, August 2014: 605
5 –
6062
6062
[8]
Zhang
L,
Cha
ng Z, W
a
n
g
Y,
Yu Z. C
u
rre
nt-reus
e
sin
g
l
e
miller
fee
d
for
w
ard c
o
mpe
n
sat
i
on
for m
u
lti-
stage amp
lifier
s
.
Electronics L
e
tters
. 2012; 4
9
(2):94-
96.
[9]
Eschauz
ier
R
G
H, Kerklaa
n
LP, Hui
j
sin
g
J
H
. A 10
0-MHz
100-
dB
oper
a
t
iona
l am
plifi
e
r
w
i
th m
u
ltip
ath
nested Mi
ller c
o
mpe
n
satio
n
structure.
IEEE Journal of Solid-
State Circuits
. 199
2; 27(1
2
):1
709-
171
7.
[10]
Leu
ng
KN, Mo
k PKT
. Anal
y
s
i
s
of mu
lt
istage
amplifi
e
r-freq
u
enc
y
c
o
mpe
n
s
a
tion.
IEEE
T
r
ansacti
ons
on
Circuits a
nd Sy
stems-I: F
unda
me
ntal T
h
e
o
ry and Ap
plic
atio
ns
. 2001; 4
8
(9)
:
1041-
10
56.
[11]
You F
,
Embabi S, Sanc
h
e
z-Sin
enci
o
E. Mu
ltistage amplifi
e
r
topo
l
ogi
es
w
i
th n
e
sted Gm-C
compensation.
IEEE Journal
of Solid-State
Circuits
. 19
97; 32(1
2
): 200
0-2
011.
[12]
Lee H, Mok P
K
T
.
Advances
in active- fe
e
dba
ck fre
que
n
c
y
com
p
e
n
sati
on
w
i
th
po
w
e
r
optimizati
on.
IEEE Transactions on C
i
rcuits
and Syste
m
s-I: Regul
ar Pap
e
r
s
. 2004; 51(
9): 1690-
16
96.
[13]
Guo S, Lee H. Dual activ
e
-c
apac
itive- fee
d
back comp
ens
ation for lo
w
p
o
w
e
r l
a
rge- ca
pacitiv
e-lo
ad
three-stag
e am
plifiers.
IEEE Journal of Solid-
State Circuits
. 201
1; 46(2):.45
2
-46
4
.
[14]
Xia
o
P, S
ans
en W
,
Ho
u LG
, W
ang JH, W
u
W
C
. Imped
a
n
ce
ada
ptin
g c
o
mpe
n
satio
n
f
o
r lo
w
-
po
w
e
r
multistag
e
amp
lifiers.
IEEE Journal of So
lid-State Circuits
. 201
1; 46(2): 44
5-45
1.
[15]
Xi
ao
P, Sans
e
n
W
.
AC b
oos
ting c
o
mpe
n
sa
tion sc
heme
for lo
w
-
p
o
w
e
r
multistag
e
am
plifiers.
IE
EE
Journ
a
l of Soli
d-State Circu
its
. 2004; 39(
11)
: 2074-2
0
7
7
.
[16]
Garimell
a A, Rashi
d
MW
,
F
u
rth PM. Reverse nes
ted mil
l
er
compens
ation
using curre
nt buffers in a
three-stag
e L
D
O.
IEEE Transactions
on
Cir
cuits and Syst
em
s-II: Expres
s Briefs
. 201
0;
67(4): 2
50-
254.
Evaluation Warning : The document was created with Spire.PDF for Python.