Download High Frequency, High Precision CMOS Half-Wave Rectifier Montree Kumngern and Kobchai Dejhan

High Frequency, High Precision CMOS
Half-Wave Rectifier
Montree Kumngern and Kobchai Dejhan
Faculty of Engineering and Research Center for Communication and Information Technology,
King Mongkut’s Institute of Technology Ladkrabang, Bangkok 10520, Thailand
Tel: 66-2326-4238, 66-2326-4242, Fax: 66-2326-4554,
E-mail: [email protected], [email protected]
Abstract–In this paper, a high frequency and high
precision CMOS half-wave rectifier circuit with positive
and negative outputs is presented. The proposed circuit
consists of a voltage-to-current converter, a precision
current-mode rectifier with operating in class-AB, and two
current-to-voltage converters. A voltage input signal is
changed into a current signal by the voltage-to-current
converter. The class-AB current mode rectifiers rectify
these current signals, resulting in positive and negative halfwave current signals that are converted to positive and
negative half-wave voltage signals by the current-to-voltage
converters. The circuit exhibits low component counts, high
frequency operation, good temperature stability and
suitable of IC fabrication. The simulation results are used to
verify the performance of the proposed circuit. Simulated
rectifier performance with standard 0.5 m CMOS
technology using !1.2V supply voltage shows the proposed
half-wave rectifier circuit provides an operating frequency
more than 250MHz and excellent temperature stability.
I.
INTRODUCTION
Precision rectifier is one of important nonlinear circuits
extensively used in wattmeter, AC voltmeter, RF
demodulator, function fitting, triangular-wave frequency
doubling, error measurements and RMS to DC
conversions. Usually, a rectifier can be realized by using
diodes but diode rectifier is limited by the threshold
voltage of diodes. As a result diode-only rectifiers are
used in only those applications in which the precision in
the range of threshold voltage is insignificant, such as RF
demodulators and DC voltage supply rectifiers. To
overcome this problem is based on the use of operational
amplifiers (op-amps), diodes, and resistors [1]-[4].
However, the classical problem with conventional
precision rectifiers based on op-amps and diodes is that
during the non-conduction/conduction transition of the
diodes the op-amps must recover with a finite smallsignal dv/dt (slew-rate) resulting in significant distortion
during the zero crossing of the input signal. Moreover,
op-amps-based precision rectifier circuits are limited to a
frequency performance well below the gain-bandwidth
product (GBW) of op-amps [5]. This problem was
improved by designing the rectifier circuit by the use of
the current rectifying techniques [5]-[11]. The current
conveyors-based current-mode rectifier circuits were
presented in [5]-[7] enjoy the high precision. However,
the proposed circuits require a floating resistor and
grounded resistors, which is not ideal for IC
implementation and some of these have problem about
temperature stability [5],[6]. In [7] was realized a current-
mode rectifier circuit based on the use of one current
conveyor and bipolar transistors operating in class-B, this
circuit is a realization that enjoys the high precision
rectifier. However, the two grounded resistors are used
and the high frequency performance is limited in this
scheme. In [8]-[9] proposed the current-mode CMOS
rectifier circuits operating in class-AB. This technique
requires the signal current to be greater than four times of
the bias current to avoid the square-law error of MOS
transistors. However, the circuit features wide-band
capability. Recently, a dual output CMOS half-wave
rectifier circuit has been proposed in [10]. This circuit
composed of two sections. The first is rectifier circuit
uses the V-I converter to change the input voltage into
currents, the diodes to rectify the currents, and the I-V
converters to change the rectified currents into the output
voltage. Second is the bias voltage source circuit, it used
the diodes bias voltages to clad the output voltage
excursion during the zero crossings. The advantages of
this circuit are: (i) all of MOS transistors are used, thus
the circuit suitable for IC fabrication, (ii) employs a
!1.2V supply voltage and provides an operating
frequencies up to 100MHz, and (iii) high accuracy and
good temperature stability by using the bias voltage
source circuit. In [11] proposed a rectifier circuit by using
all MOS transistors suitable for IC fabrication and high
operating frequency up to 200MHz.
In this paper, a high frequency CMOS half-wave
rectifier circuit is presented. The proposed circuit consists
of a voltage-to-current converter, a current-mode
precision rectifier with operating in class-AB to support
the high frequency operation, and two current-to-voltage
converters. In particular, the proposed half-wave rectifier
circuit has the following features:
" The proposed rectifier uses 31 MOS transistors
whereas the proposed rectifier in [10] uses 44 MOS
transistors; hence when the rectifier circuits are
fabricated, the proposed rectifier will be used small
area of chip than the previous rectifier.
" The proposed rectifier using supply voltage about
!1.2V is equal to the rectifier circuit in [10]. However,
the proposed rectifier provides an operating frequency
more than 250MHz, whereas in [10] identified by
authors is 100MHz (simulated results with 0.5 m
CMOS model obtained though MIETEC).
" The proposed half-wave rectifier provides excellent
temperature stability without compensation circuit.
ECTI-CON 2007
The 2007 ECTI International Conference
___________________________________________________________
1
II. CIRCUIT REALZATION
VDD
The proposed CMOS half-wave rectifier circuit
realization scheme consisting of three main components:
the V-I converter, the current-mode rectifiers, and the I-V
converters, as shown in Fig. 1. The V-I converter is
composed of negative-type CCII (CCII-) (M1-M17) and
input resistor Rin (MR1-MR2), the precision current
rectifier is current mirrors (MC1-MC4) and current
sources and (I1-I5), and I-V converters are composed of
resistors Ro1 (MR3-MR4) and Ro2 (MR5-MR6). The
operation of the half-wave rectifier circuit is as follows.
When the input voltage Vin is applied into the circuit, it is
converted to current iz by negative-type CCII (CCII-) and
a grounded resistor Rin. The grounded resistors can be
implemented by two matched MOS transistors are diodeconnected [12]. In Fig. 1(b), assume MR1 and MR2
operating in saturation region and have the same
characteristics. The resistance value of MOS resistor of
Fig. 1(b) can be expressed as [12]:
MC1
I1
MD1
Vin
Y
1
2 K (V DD # VTH )
MD2
VSS
iz<0
X
R02
VB
MD3
VDD
MD4
I6
Rin
I2
I4
Vout-
Io2
CM2
MC3
MR5
MR6
MC4
VSS
VSS
(a)
VDD
M12
M13
M14
M15
M16
M17
VDD
Rin
M1
M2
iin
IB
V in
Y
iz
MR1
X
Vin
Y
iz
CCII-
Z
Z
X
MR2
M4
Rin
VSS
(1)
M5
where K=µOCOX(W/L), VTH=threshold voltage, VDD=VSS=supply voltage, O is the carrier mobility, COX is the
gate capacitance per unit area, W and L are the channel
width and length, respectively. Therefore the current iz
can be written as
V
i z $ in
(2)
Rin
The MOS transistor MD2 and the current source I1
generate a constant voltages VA, and the MOS transistor
MD4 and the current source I2 generate constant voltages
VB to provide bias voltage for MOS transistors MD1 and
MD3, respectively. The constant voltage VA and VB
should be fairly close to the threshold voltage of MOS
transistors MD2 and MD4, respectively, to obtain
precision results, and cause them to operating in class-AB.
CM1 and CM2 is applied by the current sources I3 and I4,
respectively, which ensures that two current mirrors are
on all the times to the improving frequency response and
overall system linearity. When iz>0, this current is fed
through MD1 and then is reflected by CM1 to the output
current of CM1 as Io1 (+iz). In addition, when iz<0, this
current is fed through MD3 and then is reflected by CM2
to the output current of CM2 as Io2 (+iz). The relations
between the input current, iz, and the output current of
CM1 as Io1, and the output current of CM2 as Io2 can be
expressed as
B
i & 0 ; I o1 $ %i z % I 5
i ' 0 ; I o 2 $ %i z % I 6
Vout+
MR4
I5
Z
M3
R$
MR3
Io1
I3
VA
iz>0
CCII-
VDD
R01
MC2
CM1
(3)
(4)
Since the constant current sources, I5 and I6, compensate
the offset currents, I3+I1 and I4+I2, respectively, from (4)
and (4), the currents Io1 and Io2 can be rewritten as
M6
M7
M8
M9
M10
M11
VSS
(b)
Fig. 1. (a) Proposed half-wave rectifier circuit, (b) a voltage to current
converter circuit using CCII- and MOS resistor.
Since the constant current sources, I5 and I6, compensate
the offset currents, I3+I1 and I4+I2, respectively, from (4)
and (4), the currents Io1 and Io2 can be rewritten as
i & 0 ; I o1 $ %i z
i ' 0 ; I o 2 $ %i z
(5)
(6)
Suppose Ro1=Ro2=Rin, with the use of the closely matched
MR1 to MR6, we can write the relations between Vin, and
Vout+ as
Vin & 0 ; Vout % $ Vin
Vin ' 0 ; Vout % $ 0
(7)
(8)
and the relations between Vin, and Vout- as
Vin & 0 ; Vout # $ 0
Vin ' 0 ; Vout # $ Vin
(9)
(10)
This means that the proposed rectifier can be operated as
positive and negative half-wave rectifier into a single
circuit.
III. SIMULATION RESULTS
In order to test the ideal designed, the scheme of dual
output half-wave rectifier of Fig. 1 has been simulated
using PSPICE simulation program. For the circuit
simulation, the 0.5µm CMOS model obtained though
MIETEC as listed in [10] is used. The aspect ratios (W/L)
of MOS transistors are: 10µm/0.6µm for M1-M2, M5-
ECTI-CON 2007
The 2007 ECTI International Conference
___________________________________________________________
2
M11, MD1-MD2, MC3-MC4; 40µm/0.6µm for M3-M4,
M12-M17; 30µm/0.6µm for MD3-MD4, MC1-MC2;
4µm/0.6µm for MR1-MR2; 2µm/0.6µm for MR3-MR6.
The supply voltage used is !1.2V. The current sources
IB=50µA, I1=I2=5µA, and I3=I4=100µA. The first test
applied an input signal of amplitude 200mV peak and
frequencies100MHz and 250MHz. The results of these
tests are shown in figures 2 and 3, respectively. As is
evident, satisfied half-wave rectified signals are produced
at all two frequencies. This is as a direct result of the
operation of the class-AB and the fast action of the
voltage-to-current converter using the CCII. The
amplitude errors between the input and output signals in
figure 3 results from the decrease of the gain of the
proposed rectifier for the operation at high frequency. It
can correct by increase the value of Ro1 and Ro2.
Several solutions to improve the poor temperature
characteristic have been proposed. The good temperature
stability can be achieved. However, it is causes of few
problems such as use of many transistors [10], the
precision decreasing at the high frequency [13], and the
solving technique only suitable for bipolar technology
[14]. To demonstrate the temperature performance of
proposed circuit of Fig. 1, we simulate the proposed
circuit in Fig. 1 at the frequency of 100MHz by changing
temperatures form 50(C to 100(.
vout+, vin
B
(a)
vout-, vin
vout+, vin
(b)
Fig. 3. Operation of half-wave rectifier at the input signal frequency
250MHz: (a) the input and positive half-wave output Vout+ and (b) the
input and negative half-wave output Vout-.
vout+, vout100 C
75 C
50 C
(a)
vout-, vin
50 C
75 C
100 C
Fig. 4. Output wave forms at different temperature at the input signal
frequency 100MHz.
(b)
Fig. 2. Operation of half-wave rectifier at the input signal frequency
100MHz: (a) the input and positive half-wave output Vout+ and (b) the
input and negative half-wave output Vout-.
Fig. 4 shows the output waveforms of the proposed
rectifier at temperature of 50(C, 75(C, and 100(C. In Fig.
4, it can see that the proposed circuit very temperature
stable without the compensation technique. In [8], [9]
recommended that the class-AB technique requires the
signal current to be greater than four times of the bias
current to avoid the square-law error of MOS transistors.
ECTI-CON 2007
The 2007 ECTI International Conference
___________________________________________________________
3
This limitation can be corrected by setting the quiescent
currents. The changing of the quiescent currents is
directly proportional to the zero crossing. To demonstrate
this point, we simulate the proposed rectifier in Fig. 1
again. Fig. 5(a) shown the operation of half-wave
rectifier at the bias currents I1=I2=1µA and the input
signal frequency 100MHz for 25mVpeak, (a) the input and
output Vout+, (b) the input and output Vout-. Note that the
precision and frequency range parameters of proposed
rectifier can be controlled. Two parameters dependence
on the setting of the quiescent currents, I1and I2, The low
quiescent current may limit the high frequency capability
of the proposed rectifier. In this case may be is an
advantage for application. When we want to application
in high precision system, the precision rectifier can be
given. Again, if we want to application in high frequency
system, the frequency range can be controlled.
mirrors, and current sources and its combine a positive
half-wave rectifier, a negative half-wave rectifier into
single circuit. Although, the proposed rectifier is based
on previously reported circuit conveyor rectifier but with
the different circuit structure, the proposed half-wave
rectifier yields features superior to the previously
proposed current conveyor rectifier in view of the voltage
employed, the power consumption, the number of devices,
the operating frequency, and the temperature stability.
The configuration described is very suitable for
integrated circuit implementation in both bipolar and
CMOS technology. The performance of the proposed
circuit is confirmed from PSPICE simulation results.
REFERENCES
[1]
[2]
vout+, vin
[3]
[4]
[5]
[6]
[7]
[8]
[9]
(a)
[10]
vout-, vin
[11]
[12]
[13]
[14]
J. Peyton and V. Walsh, Analog electronics with op amps: a
source book of practical circuits, Cambridge University Press,
New York, 1993.
R. G. Irvine, Operational amplifier characteristics and
applications. Prentice Hall International, New Jersey, 1994.
Z. Wang, “Full-wave precision rectification that is performed in
current domain and very suitable for CMOS implementation,”
IEEE Transaction on Circuit and Systems-I, vol. 39, pp. 456-462,
1992.
S. J. G. Gift, “A high-performance full-wave rectifier circuit,”
International Journal of Electronics, vol. 89, pp. 467-476, 2000.
C. Toumazou, F. J. Lidgey, and S. Chattong, “High frequency
current conveyor precision full-wave rectifier,” Electronics
Letters, vol. 30, pp. 745-746, 1994.
A. Khan, M. Abou El-Ela, and M. A. Al-Turaigi, “Current-mode
precision rectification,” International Journal of Electronics, vol.
79, pp. 853-859, 1995.
A. Monpapassorn, K. Dejhan, and F. Cheevasuvit, “A full-wave
rectifier using a current conveyor and current mirrors,”
International Journal of Electronics, vol. 88, pp. 751-758, 2001.
V. Surakumpontorn and V. Riewruja, “Integrable CMOS
sinusoidal frequency doubler and full-wave rectifier,”
International Journal of Electronics, vol. 73, pp. 627-632, 1992.
V. Riewruja and R. Guntapong, “A low-voltage wide-band CMOS
precision full-wave rectifier,” International Journal of
Electronics, vol. 89, pp. 467-476, 2002.
A Monpapassorn, K. Dejhan, and F. Cheevasuvit, CMOS dual
output current mode half-wave rectifier,” International Journal of
Electronics, vol. 88, pp. 1073-1084, 2001.
M. Kumngern and K. Dejhan, “High frequency and high precision
CMOS full-wave rectifier,” International Journal of Electronics,
vol. 93, pp. 185-199, 2006.
Z. Wang, “2-MOSFET transistor with extremely low distortion
for output reaching supply voltage,” Electronics Letters, vol. 26,
pp. 951-952, 1990.
K. Hayatleh, S. Porta and F.J. Lidgey, “Temperature independent
current conveyor.” Electronics Letters, vol. 30, pp. 2091-2093,
1994.
D. Stiurca, “Truly temperature independent current conveyor
precision rectifier.” Electronics Letters, vol. 31, pp. 1302-1303,
1995.
(b)
Fig. 5. Operation of half-wave rectifier at the bias current I1=I2=1µA
and the input signal frequency 100MHz for 25mVpeak, (a) the input and
output Vout+, (b) the input and output Vout-.
IV. CONCLUSIONS
A high frequency and high precision CMOS half-wave
rectifier circuit was presented. The proposed circuit have
distinguish in used of a negative-type CCII, current
ECTI-CON 2007
The 2007 ECTI International Conference
___________________________________________________________
4