CCM-DCM Average Current Control For Both Continuous and Discontinuous Conduction Modes Boost PFC Converters
CCM-DCM Average Current Control For Both Continuous and Discontinuous Conduction Modes Boost PFC Converters
Abstract—This paper presents a CCM-DCM average current is usually designed for CCM operations, the average inductor
control (CDAC) control method for effectively controlling boost current of which is obtained by sampling in the middle of the
power factor correctors (PFCs). It is capable of operating rising edge of the inductor current at every switching cycle.
in both the continuous conduction mode (CCM), the mixed
conduction mode (MCM) and the discontinuous conduction mode However, if the CCM average current control method is used
(DCM). By estimating the average value of the inductor current for control in DCM, input current distortion may be resulted
in each switching cycle, the CDAC control method efficiency due to inaccurate average current values obtained and poor
improvements and harmonic distortion reductions in boost PFCs inductor current tracking.
operating over wide load range, especially operating in the DCM To achieve good input current shaping over a complete load
mode. Besides, it is mathematically and computationally simple.
Compared with the conventional average current control method, range, a linear CCM controller, a nonlinear DCM controller
simulation results verify the feasibility and performances of the and an operation mode selection circuit were employed in [6],
proposed CDAC control method. [7]. The problem of distorted input current, when operating
Index Terms—Boost PFC, CCM-DCM, average current con- in MCM and DCM, was overcome. However, there were
trol, current estimation. slight distortions in the input current during the transition
between CCM and DCM operations. And when a converter
operates fully in DCM, the distortion in the input current
I. I NTRODUCTION is worse. Moreover, these controllers based on an operation
The power factor correction (PFC) is gaining more and more mode selection circuit require higher development efforts,
attentions because of the urgent need to improve the efficiency complexity, and cost. Thus, it is important to find a universal
of power plants and eliminate input current distortion in order control method for the boost PFC across a complete load
to meet international stringent standards and the quality of range.
power in a utility grid [1], [2]. Compared with other types A typical solution used to avoid separate controllers required
of PFC converters, the PFC based on a boost-type converter for the conduction mode selection is the variable-duty-cycle
is used most widely due to its advantages of good silicon control, which is generally realized by modifying the voltage
usage, smooth current waveform with reduced input filter loop of the conventional control by utilizing an input voltage
requirement, and simple drive and control circuitry [3], [4]. feedforward and additional calculation circuit. The variable-
The boost PFC converter can operate in three conduction duty-cycle control provides effective distortion reduction to the
modes of different inductor currents, with respect to the inductor current distortion, but usually requires a complicated
input voltage frequency, which are the continuous conduction and expensive additional circuitry. A square-root operation and
mode (CCM), the mixed conduction mode (MCM) and the multiplication were added in [8] [9]; Another variable-duty-
discontinuous conduction mode (DCM) [5]. The MCM and cycle control by synthesizing a certain harmonic component
DCM operations of a boost PFC converter are optimal for of the line frequency in the voltage loop requires additional
low to medium power applications, while the CCM operation phase detector and phase-locked loop [10] [11]. Instead of
of a boost PFC converter is optimal for medium and high- modifying the voltage loop, an inductor current sensing circuit
power applications in terms of minimal cost and size. As a is added to implementing variable-duty-cycle control of DCM
conventional control law, the average current control method PFC rectifier. With this addition circuit, the sampled inductor
is most-used for boost PFC converters since it is less sensitive current is corrected or measured properly to represent the
to switching noise as compared to the peak current control average inductor current per switching period. The digital
method, and with fixed switching frequency as compared to the controllers proposed in [12]–[14] deviated the sampling instant
hysteresis current control. The average current control method and calculated a correction factor to correct the average
inductor current of each switching period. Using the correction
L. Wang, Q. H. Wu, W. H. Tang and Z. Y. Yu are with School of Elec- factor, accurate average current values in both CCM and DCM
tric Power Engineering, South China University of Technology, Guangzhou are obtained, so that good input current shaping is achievable
510641, China. W. Ma is with Chongqing University ofScience and Technol-
ogy, Chongqing 401331, China over the complete load range. However, these control methods
Corresponding author is Professor W. H. Tang, E-mail: also require a division and a square root calculation for the D-
[email protected], Tel: 0086 020 87112999. CM duty ratio. Besides, the calculation of the correction factor
The work is funded by the National Natural Science Foundation of China
under Grant 51377188 and Guangdong Innovative Research Team Program of each switching period consumes considerable resources of
(NO. 201001N0104744201). a controller. In [15], by adding an additional average sensing
circuit, the input current distortion was reduced with direct L L"
sensing of the average inductor current. Then, the additional /
correction factor calculation is not required. However, the L /
direct sensing of the average inductor current is naturally L
X
suited for DCM. Under CCM, the average inductor current X & L & X 5
cannot be sensed accurately because there is no enough time
for the reset pulses. And the design of the sensing capacitance
leads to higher development effort.
In this paper, a CDAC control method is proposed to
implement the variable-duty-cycle control. By estimating the L"
average inductor current with a reset integrator, accurate L X W
average inductor currents are obtained in both CCM and
X
DCM. The organization of this paper is as follows. In Section
II, current waveforms in CCM and DCM boost PFC are Fig. 1. The CCM average current control for a boost PFC converter.
analysed. Section III illustrates the circuit configuration, mode
analysis, and operation of the estimation circuit as well as J 7
the CDAC control method. The performances of the proposed
G 7 G 7 G 7
CDAC control method are analysed in Section IV. Section V
concludes this paper.
W
II. C URRENT WAVEFORMS IN CCM AND DCM PFC
L" L"
A. Conduction mode boundaries L"
A boost PFC converter can operate in any conduction mode,
depending on the input and output specifications and converter L
parameters. The boundaries among its conduction modes are " W
obtained from the following equations [16]:
Fig. 2. Inductor current in DCM.
Vin2 − p
L1 = (1)
4fs Pout i− av are lower than the inductor current samples i− sample .
Thus, sampling in the middle of the rising edge of the inductor
Vin2 − p Vin− p current cannot give the correct average values in DCM. This
L2 = (1 − ) (2)
4fs Pout Vout is the one of the two main causes of input current distortion.
where (1) divides the CCM and MCM regions, which rep- The other cause of input current distortion is that the linear
resents the minimum inductor value for the CCM operation CCM average control is unable to ensure input current shaping
and (2) divides the MCM and DCM regions, which denotes whereby the converter exhibits nonlinear characteristics in
the maximum inductor value for the DCM operation. These DCM. Assuming that the switching frequency of the boost
equations depend on the input voltage peak (Vin− p ), switching PFC converter is much higher than the line frequency, the
frequency (fs ), output power (Pout ) and output voltage (Vout ) average current waveforms of CCM average current controlled
of the converter. PFC converter resemble the sinusoid of the input voltage,
because the switching ripple of the inductor current is rela-
tively small, as shown in Fig. 3(a). However, the converter
B. Current waveforms in any conduction mode
exhibits nonlinear characteristics in DCM, and the CCM
As a conventional control method for a boost PFC converter, average current controller is insufficient to quickly correct
the CCM average current-mode control is commonly used, the errors that caused by inaccurate average current values in
since it can be easily implemented in a digital form. Fig. 1 DCM. Then the envelope of the inductor current is constructed
shows a boost PFC using the CCM average current controller. to be sinusoidal, while the average inductor current is highly
In the CCM average current control, the average inductor distorted, as shown in Fig. 3(b). This results in poor inductor
current is obtained by sampling in the middle of the rising current tracking and input current distortions. The analysis
edge of the inductor current at each switching cycle, and of input current distortion in CCM average current controlled
accurate average inductor current values are obtained in CCM PFC converter [15] is given as follows.
boost PFC. However, unlike in CCM boost PFC, the inductor Assuming that the input voltage is purely sinusoidal and it
current in DCM boost PFC goes to zero before the end of has no distortion, the input voltage is defined as
the switching period. As shown in Fig. 2, i− sample , i− p and
i− av represent the currents sampled in the middle of the rising uin (t) = Vm sinωt. (3)
edge of the inductor current, the peak of the inductor current
and the actual average inductor current, respectively, and it where Vm is the amplitude of the input voltage and ω is the
can be seen that the actual average inductor current values angular frequency of the input voltage.
2017 IEEE Electrical Power and Energy Conference (EPEC)
L" W
L" W D "
D !
D
D #
S S ZW
W Fig. 4. The normalized average inductor current waveforms.
L" W L L"
L" W L /
/
L
X
X & L & X 5
L" W
L"
$
L" % &
W
X W
L
X
Fig. 3. Inductor current and average current waveform shapes. (a) CCM
Fig. 5. Proposed CDAC controlled boost PFC.
operation; (b) DCM operation.
Then, the rectified voltage is If don is constant, according to (8), it can be seen that the
shape of the input current is only dependent on α. When α
ug (t) = Vm |sinωt| (4) is small, input current is almost sinusoidal. As α approaches
1, input current is more and more distorted. The normalized
In a switching cycle, the peak inductor current i− p is average inductor current waveforms as a function of α is
ug Vm |sinωt| shown in Fig. 4.
i− p = don Ts = don Ts (5)
L L
III. P ROPOSED CDAC CONTROL METHOD
where don is the duty cycle and Ts is the switching cycle.
In each switching cycle, the inductor has a voltsecond The above analysis shows that, to achieve satisfactory input
balance, i.e., current shaping, the average inductor current must be obtained
accurately over the complete load range. Thus, in the proposed
ug don Ts = (uo − ug )doff1 Ts CDAC control method, an average current estimation circuit
ug Vm |sinωt| (6) is added to obtain accurate average current in both CCM
⇒ doff1 = don = don
uo − ug uo − Vm |sinωt| and DCM operations and compensate for the nonlinearity of
DCM operationS. The proposed CDAC control follows the
where uo is the output voltage and doff1 is the duty cycle general structure of the conventional two-loop control to build
corresponding to the discharging time of the inductor current. an average current-mode control, as shown in Fig. 5. The
According to (5) and (6), the average inductor current in a configuration of the average current estimation circuit is in
switching cycle is obtained as the dashed box and the operational waveform is shown in Fig.
1 Vm d2on |sinωt| 6. Fig. 6(a) and Fig. 6(b) show the waveforms of the average
i− av (t) = i− p (don + doff1 ) = estimation circuit operation in CCM and DCM PFC converter,
2 2Lfs Vm
1− |sinωt| respectively.
uo
(7)
where fs = 1/Ts is the switching frequency. A. CCM operation
Let α = Vumo , the input current is obtained as
Under CCM operations, during the nth switching cycle,
Vm d2on sinωt the boost PFC converter operates in two states, in which the
iin (t) = (8) switch S1 is on for a duration of don (n)Ts and is off for a
2Lfs 1 − α|sinωt|
2017 IEEE Electrical Power and Energy Conference (EPEC)
Current (A)
10 10
0 0
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Output voltage Output voltage
240 220
Voltage (V)
Voltage (V)
220
200 200
180
160 180
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Voltage (V)&Current (A)
0 0
−20 −20
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Time (s) Time (s)
(a) CDAC control (b) CCM average current control
Fig. 7. Waveforms of the CDAC control and conventional CCM average current control in CCM operation. (a) CDAC control method; (b) Conventional
CCM average current control method.
TABLE I
PARAMETERS OF THE B OOST CONVERTER simple. An average current estimation circuit is configured in
Parameter Symbol Value the proposed CDAC control method. The configuration and
output power at full load Pout− fl 800 W
AC input voltage (the peak value) uin− p 100 V the operation of the estimation circuit have been analyzed and
Filter inductance Lf 500 μH explained with mathematical expressions. By employing the
Filter capacitance Cf 0.47 μF average current estimation circuit, an accurate average inductor
Reference voltage uref 200 V current is obtained in both CCM and DCM. Thus, satisfactory
Switching frequency fs 20 kHz
Inductance L 800 μH input current shaping and a higher power factor are achieved
Capacitance C 450 μF over the complete load range with the CDAC control method.
The proposed CDAC control method has been verified and
TABLE II compared with the conventional CCM average current control
S IMULATION MEASUREMENTS OF THE BOOST PFC IN ALL LOAD CASES based on the simulation results. Waveforms and measured
THD data have demonstrated the performance improvement of the
Operation mode
CDAC control CCM control proposed CDAC control method.
CCM (p.u. load) 2.56 % 2.32 %
MCM (0.3p.u. load) 4.63 % 17.21 % R EFERENCES
DCM (0.1p.u. load) 8.0 % 21.40 %
[1] B. Singh, B. N. Singh, A. Chandra, K. Al-Haddad, A. Pandey, and
D. P. Kothari, “A review of three-phase improved power quality ac-dc
converters,” IEEE Trans. Ind. Electron, vol. 51, no. 3, pp. 641–660,
line period. The same comparison in case that the converter 2004.
operates fully in DCM is also shown in Fig. 9. The CDAC [2] U. Kamnarn and V. Chunkag, “Analysis and design of a modular three-
control method still perfects smaller current distortion than the phase AC-to-DC converter using CUK rectifier module with nearly unity
power factor and fast dynamic response,” IEEE Trans. Power Electron,
conventional method. Thus, with the CDAC control method, vol. 24, no. 8, pp. 2000–2012, 2009.
good input current shaping at the complete load range is [3] R. Redl, “Electromagnetic environmnetal impact of power electronics
achieved. equipment,” Proceedings of the IEEE, vol. 89, no. 6, pp. 926–938, 2001.
[4] D. M. VandeSype, K. D. Gusseme, A. P. V. den Bossche, and J. A. A.
The measurements of THD of the input current, the power Melkebeek, “A sampling algorithm for digitally controlled boost PFC
factor in each cases are summarized in Table II. In table II, converters,” IEEE Trans. Power Electron., vol. 19, no. 3, pp. 649–657,
it is observed that, under CCM operations, the THD of the 2004.
[5] J. Sebastian, J. A. Cobos, J. M. Lopera, and J. Uceda, “The determination
input current increases slightly but still meets the EN 61000- of the boundaries between continous and discontinous conduction modes
3-2 class D standard. Under MCM and DCM, the THD of the in PWM DC-to-DC converters used as power factor preregulators,” IEEE
input current is significantly reduced with the CDAC control. Trans. Power Electron., vol. 10, no. 5, pp. 574–582, 1995.
[6] A. A. Bento and E. R. C. da Silva, “Hybrid one-cycle controller for
boost PFC rectifier,” in in Proc. IEEE 42nd Ind. Appl. Conf.,, 2007, pp.
V. C ONCLUSION 2333–2339.
[7] L. Roggia, F. Beltrame, J. E. Baggio, and J. R. Pinheiro, “Digital control
A CDAC control method for boost PFC converter has been system applied to a PFC boost converter operating in mixed conduction
proposed in this paper. CCM-DCM control is achieved with mode,” in in Proc. Brazilian Power Electron. Conf. COBEP,, 2009, pp.
minimal changes to the conventional CCM current control 698–704.
[8] K. Yao, X. Ruan, X. Mao, and Z. Ye, “Variable-duty-cycle control to
method and no additional algorithms are required for conduc- achieve high input power factor for DCM PFC boost converter,” IEEE
tion mode selections. It is mathematically and computationally Trans. Ind. Electron., vol. 58, no. 5, pp. 1856–1865, 2011.
2017 IEEE Electrical Power and Energy Conference (EPEC)
Current (A)
4 4
2 2
0 0
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Output voltage Output voltage
210 210
Voltage (V)
Voltage (V)
200 200
190 190
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Input voltage & input current Input voltage & input current
5 5
0 0
−5 −5
0.9 0.92 0.94 0.96 0.98 1 0.9 0.92 0.94 0.96 0.98 1
Time (s) Time (s)
(b) CCM average current control
(a) CDAC control
Fig. 8. Waveforms of the CDAC control and conventional CCM average current control in MCM operation. (a) CDAC control method; (b) Conventional
CCM average current control method.
Current (A)
4
4
2 2
0 0
1.2 1.22 1.24 1.26 1.28 1.3 1.2 1.22 1.24 1.26 1.28 1.3
Output voltage Output voltage
210 210
Voltage (V)
Voltage (V)
200 200
190 190
1.2 1.22 1.24 1.26 1.28 1.3 1.2 1.22 1.24 1.26 1.28 1.3
Voltage (V)&Current (A)
Voltage (V)&Current (A)
Input voltage & input current Input voltage & input current
4 4
2 2
0 0
−2 −2
−4 −4
1.2 1.22 1.24 1.26 1.28 1.3 1.2 1.22 1.24 1.26 1.28 1.3
Time (s) Time (s)
(a) CDAC control (b) CCM average current control
Fig. 9. Waveforms of the CDAC control and conventional CCM average current control in DCM operation. (a) CDAC control method; (b) Conventional
CCM average current control method.
[9] Z. Z. Ye and M. M. Jovanovic, “Implementation and performance rectifiers,” IEEE Trans. Power Electron., vol. 25, no. 10, pp. 2683–2692,
evaluation of DSP-based control for constant-frequency discontinuous- 2010.
conduction-mode boost PFC front end,” IEEE Trans. Ind. Electron., [15] J. W. Shin and B. H. Cho, “Digitally implemented average current-mode
vol. 52, no. 1, pp. 98–107, 2005. control in discontinuous conduction mode PFC rectifier,” IEEE Trans.
[10] D. Weng and S. Yuvarajian, “Constant switching frequency ac-dc Power Electron., vol. 27, no. 7, pp. 3363–3373, 2012.
converter using second harmonic injected PWM,” IEEE Trans. Power [16] L. Roggia, F. Beltrame, J. E. Baggio, and J. R. Pinheiro, “Digital current
Electron., vol. 11, no. 1, pp. 115–121, 1996. controllers applied to the boost power factor correction converter with
[11] D. S. Schramm and M. O. Buss, “Mathematical analysis of a new load variation,” IET Power Electron., vol. 5, no. 5, pp. 532–541, 2011.
harmonic cancellation technique of the input line current in DICM boost
converters,” in in Proc. IEEE Power Electron. Spec. Conf.,, 1998, pp.
1337–1343.
[12] K. D. Gussem, D. M. V. de Sype, A. P. V. den Bossche, and J. A.
Melkebeek, “Digitally controlled boost power-factor-correction convert-
ers operating in both continuous and discontinuous conduction mode,”
IEEE Trans. Ind. Electron., vol. 52, no. 1, pp. 88–97, 2005.
[13] S. F. Lim and A. M. Khambadkone, “A simple digital DCM control
scheme for boost PFC operating in both CCM and DCM,” IEEE Trans.
Ind. Electron., vol. 47, no. 4, pp. 1802–1812, 2011.
[14] F. Chen and D. Maksimovic, “Digital control for improved efficiency
and reduced harmonic distortion over wide load range in boost PFC