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: montree_kumngern@yahoo.com, kobchai@telecom.kmitl.ac.th
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.
I NTRODUCTION
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 currentmode 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. C IRCUIT REALZATION
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 R in
(MR1-MR2), the precision current rectifier is current mirrors (MC1-MC4) and current sources and (I
1
-I
5
), and I-V converters are composed of resistors R o1
(MR3-MR4) and R o2
(MR5-MR6). The operation of the half-wave rectifier circuit is as follows.
When the input voltage V in
is applied into the circuit, it is converted to current i z
by negative-type CCII (CCII-) and a grounded resistor R in
. 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]:
R $
1
2 K ( V
DD
# V
TH
)
(1) where K=µ
O
C
OX
(W/L), V
TH
=threshold voltage, V
DD
=-
V
SS
=supply voltage,
O
is the carrier mobility, C
OX
is the gate capacitance per unit area, W and L are the channel width and length, respectively. Therefore the current i z can be written as i z
$
V in
R in
(2)
The MOS transistor MD2 and the current source I
1 generate a constant voltages V
A
, and the MOS transistor
MD4 and the current source I
2
generate constant voltages
V
B
to provide bias voltage for MOS transistors MD1 and
MD3, respectively. The constant voltage V
A
and V
B 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 I
3
and I
4
, respectively, which ensures that two current mirrors are on all the times to the improving frequency response and overall system linearity. When i z
>0, this current is fed through MD1 and then is reflected by CM1 to the output current of CM1 as I o1
(+i z
). In addition, when i z
<0, this current is fed through MD3 and then is reflected by CM2 to the output current of CM2 as I o2
(+i z
). The relations between the input current, i z
, and the output current of
CM1 as I o1
, and the output current of CM2 as I o2
can be expressed as i & 0 ; I o 1
$ % i z
% i ' 0 ; I o 2
$ % i z
I
5
(3)
% I
6
(4)
Since the constant current sources, I
5
and I
6
, compensate the offset currents, I
3
+I
1
and I
4
+I
2
, respectively, from (4) and (4), the currents I o1
and I o2
can be rewritten as
V in
R in
M12
Y
X
CCIIZ
MD1 i z
>0 i z
<0
MD3
M13
MC1
V
A
I
1
MD2
I
3
V
B
V
DD
CM1
MD4
M14
V
DD
MC3
I
2
I
4
V
SS
(a)
CM2
M15 M16 M17
MC2
I
5
I o1
R
01
V
DD
MR3
MR4
V out+
I
6
I o2
R
02
V
SS
V
DD
MR5
MR6
V out-
MC4
V
SS
I
B
M1
V in
M3
Y X
M2 i in
R in
V
DD
MR1
MR2
M4
V
SS
Z i z
V in
R in
Y
CCIIZ i z
X
M5 M6 M7
V
SS
M8 M9 M10 M11
(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, I
5
and I
6
, compensate the offset currents, I
3
+I
1
and I
4
+I
2
, respectively, from (4) and (4), the currents I o1
and I o2
can be rewritten as i & 0 ; I o 1 i ' 0 ; I o 2
$ % i z
(5)
$ % i z
(6)
Suppose R o1
=R o2
=R
V out+
as in
, with the use of the closely matched
MR1 to MR6, we can write the relations between V in
, and
V in
V in
V in
V in
& 0 ; V out %
$ V in
(7)
' 0 ; V out %
& 0 ; V out #
' 0 ; V out #
$ 0 and the relations between V in
, and V out
$ V in
(8)
- as
$ 0 (9)
(10)
This means that the proposed rectifier can be operated as positive and negative half-wave rectifier into a single circuit.
III. S IMULATION R ESULTS
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
I
B
=50µA, I
1
=I
2
=5µA, and I
3
=I
4
=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 R o1
and R o2
.
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 ( .
v out+
, v in v out+
, v in v out-
, v in
(a)
(b)
Fig. 3. Operation of half-wave rectifier at the input signal frequency
250MHz: (a) the input and positive half-wave output V out+
and (b) the input and negative half-wave output V out-
.
v out+
, v out-
100 C
75 C
50 C
(a) v out-
, v in
50 C
75 C
100 C
(b)
Fig. 2. Operation of half-wave rectifier at the input signal frequency
100MHz: (a) the input and positive half-wave output V out+
and (b) the input and negative half-wave output V out-
.
Fig. 4. Output wave forms at different temperature at the input signal frequency 100MHz.
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 I
1
=I
2
=1µA and the input signal frequency 100MHz for 25mV peak
, (a) the input and output V out+
, (b) the input and output V out-
. Note that the precision and frequency range parameters of proposed rectifier can be controlled. Two parameters dependence on the setting of the quiescent currents, I
1 and I
2
, 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. v out+
, v in v out-
, v in
(a) 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.
R EFERENCES
[1] J. Peyton and V. Walsh, Analog electronics with op amps: a source book of practical circuits , Cambridge University Press,
New York, 1993.
[2] R. G. Irvine, Operational amplifier characteristics and applications . Prentice Hall International, New Jersey, 1994.
[3] 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.
[4] S. J. G. Gift, “A high-performance full-wave rectifier circuit,”
International Journal of Electronics , vol. 89, pp. 467-476, 2000.
[5] C. Toumazou, F. J. Lidgey, and S. Chattong, “High frequency current conveyor precision full-wave rectifier,” Electronics
Letters , vol. 30, pp. 745-746, 1994.
[6] 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.
[7] 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.
[8] V. Surakumpontorn and V. Riewruja, “Integrable CMOS sinusoidal frequency doubler and full-wave rectifier,”
International Journal of Electronics , vol. 73, pp. 627-632, 1992.
[9] 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.
[10] 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.
[11] M. Kumngern and K. Dejhan, “High frequency and high precision
CMOS full-wave rectifier,” International Journal of Electronics , vol. 93, pp. 185-199, 2006.
[12] Z. Wang, “2-MOSFET transistor with extremely low distortion for output reaching supply voltage,” Electronics Letters , vol. 26, pp. 951-952, 1990.
[13] K. Hayatleh, S. Porta and F.J. Lidgey, “Temperature independent current conveyor.” Electronics Letters , vol. 30, pp. 2091-2093,
1994.
[14] 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 I
1
=I
2
=1µA and the input signal frequency 100MHz for 25mV peak
, (a) the input and output V out+
, (b) the input and output V out-
.
IV. C ONCLUSIONS
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