2004 International Conference on

-
Power System Technology POWERCON 2004
Slngapore, 21-24 November 2004

Unity Power Factor Rectifier using

Scalar Control Technique
A.G.Visha1Anand, Neeraj Gupta, V.Rammarayanan

Abshct- The widely used Resistance Emulation technique

. Lad

for eliminating the problem of harmonic pollution of the Power

Distribution system is based on the concept that the reemer
presents a pure resistive load to the ac system. In such a case,
input voltage and current wlll have the ssme waveshape and will
be in phase. Over the recent past, many methodologies based on Fig. I TWMtage PFC Converter -"
the Resistance EmuIation technique have been proposed in the
literature. There has been a consistent attempt to introduce
simplified control, higher efficiency and cost effectiveness h
various emerging techniques. Scalar Control Method based
Resistance Emulation technique, which is being discussed in this
Li-
paper, i s also an endeavor in the same direction. In this paper,
the details of Scalar Control theory, analysis, modeling,
simulation and erperlmental resnlts obtalned by using this
technique have been presented. r
..
w I - 1
wam7 pmu

I. INTRODUCTION Fig. 2 Singre-stage PFC Converter

Conventional power converters with diode capacitor
rectifier front end have distorted input current with high
harmonic content in the line currents. It is well known that
these harmonic currents cause several problems such as
voltage distortion, heating, and noises, reducing the capacity
of the line to supply energy. Owing to this fact and the
adoption of standards such as 1000-3-2 [ 11, there's a need for
power supplies that draw current with low harmonic content
& aIso have power factor close to unity. So far, a variety of Fig. 3 Scalar Control based PFC Converter
passive [24] and active PFC techniques have been proposed.
While the passive PFC techniques may be the best choice at Requirement of two control loops
low power, cost sensitive applications, the active PFC Big storage capacitor
techniques are used in majority of the applications owing to Additional DC-DC converter
their superior performance. While the two-stage approach is a cost effective approach
The active PFC converters can be implemented either in high power applications, its cost-effectiveness is reduced in
using the two-stage approach [5] or the single-stage approach. low power applications due to the additional PFC power stage
Unity power factor and high output voltage regulation are and additional control circuitry. A number of active single
achieved in two-stage approach by the addition of a converter stage PFC techniques [6-91 have been introduced in the recent
for input current shaping at the AC line side of the years as a low cost alternative to this problem. It involves
conventional switching power supplies as shown in Fig. 1. integration of the active PFC stage with the isolated DC-DC
The price that is paid for meeting the objectives of sinusoidal stage as shown in Fig. 2.
input current & output voItage regulation are In this approach only one switch and hence only one
controller is used to shape the input current and to regulate the
A. G. Vishal Anand is with the Department of Electrical Engg., Indian output voltage. Here, the storage capacitor that's used for
Institute of Science, 3angdore-560012 ( [email protected]) storing the difference between the instantaneous input power
Neeraj Gupta is with the Department of Electrical Engg., Indian Institute and constant output power is no longer loosely regulated at a
of Science, Bangalore561H)lZ ( [email protected])
V. Ramanarayanan is with the Department of Electrical Engg., Indian constant value. It happens so because the controller is used to
Institute of Science, BangdoreS6OOl2([email protected]) regulate the output voltage and not Vdc. So the downside

0-7803-8610-8104/$20.00 0 2004 IEEE 862

associated with this approach is that the size and the cost of
electrolytic capacitor is much greater than that used in the
two-stage approach. So the reduction in cost achieved by
using single switch & controller would be offset by
employing a larger sized capacitor.
The single stage PFC converter, used for obtaining UPF at
line side and tight voltage regulation, through scalar control
technique is as shown in Fig. 3.hthe new topology proposed,
the front-end diode bridge rectifier circuit is dispensed with. Load
Also the energy storage capacitor is removed because the
unbalanced energy is now taken care of by output capacitor C
itself. Here the controller employed is DSP based which
senses input current and output dc voltage for achieving the
requisite control. The Scalar Control algorithm used in the
DSP controller results in input sinusoidal current and loosely 2
regulated dc voltage at the output if only input current is used Fig. 4 Singlephase Direct Boost Rectifier
as the sensing signal for DSP controller. While the same
algorithm results in UPF at the line side and tight voltage (3)
"dc
regulation if both input current and output voltage are fed into
the DSP controller. As voltage across the capacitor is
By varying the duty ratio in accordance with the above
Control Law, UPF operation can be achieved. Here, only the
regulakd unlike a conventional single stage PFC converter,
input current & DC bus voltage need to be sensed for
the size and cost of capacitor is reduced. In addition to this,
achieving the control. The Control Law is very simple to
there are other advantages accrued by using this topology.
implement as is evident from (3). The control scheme is also
1. Control algorithm is very simple to implement.
independent of the input voltage.
2. Input voltage need not be measured.
For single phase Scalar Control implementation, the
3. Control is based on carrier (constant switching kquency)
advantages might not be too marked, since four controlled
and is simple to realize.
4. At the cost of one extra active device, number of passive
devices are used. But when the same technique is extended to
three-phase system, the advantages accrued are quite
devices is reduced.
5. Efficiency will be higher on account of the fact that only distinctive, for the control is decoupled in nature and does not
require transformations to the stationary W e of reference as
one device is in the series path of processed power.
is used for a conventional three-phase PFC converter. This
6. The operation is always under CCM.
configuration could also be controlled to obtain bi-directional
power flow, which might not be possible with a three-phase
11. SCALAR CONTROL THEORY
diode bridge on the AC side.
Many UPF Rectifiers [ 10-121 proposed in recent times, However, the focus of the work reported in chis paper is on
have significant improvement over their predecessor circuits. single-phase Direct Boost rectifier. The same idea could also
These UPF Rectifiers have an uncontrolled rectification stage in principle be extended to three-phase operation.
followed by a controlled boosthuck-boost topology. In Scalar
Control technique, at the cost of increasing the number of III. SYSTEM MODELING
controlled devices, the uncontrolled rectification stage can be
totally removed. This method has simpler control algorithm
and easier implementation to its credit than the existing
The Block diagram of the scalar controlled PFC controIler
is shown in Fig. 5. It can be seen that there are two cascaded
techniques.
control loops in the entire scheme. The outer voltage loop
The boost rectifier configuration to which Scalar Control is
applied is as shown in the Fig.4. On applying Volt-sec provides a current reference I,f to the inner current loop.
balance to the line inductance, Under steady state, the current reference is a DC quantity. The
Vd,(1 - 2d) input current is normalized with respect to the reference
v, = current, I,f to generate the Normalized current reference. The
2
where, the terms are as mentioned in Fig. 4. Pulse width modulator shown in Fig. 5 has a standard saw-
For obtaining resistance emulation characteristics,the input tooth ramp generator. The switching instants are produced by
voltage & input current are related by the relation, comparison of the ramp with the Normalized current reference
as shown in Fig. 6. On applying Volt-sec balance on the line
V, =I,R, (2)
inductor,
where Reis the emulated resistance.
From (1) & ( 2 ) ,the Control Law obtained is,

863
 ,V =- va
(1- 2d)
Where
IV. STEADY STATE STABILITY
V, is the applied AC voltage The steady state stability analysis of Scalar Controlled
Boost converter is carried out here. Like current mode
,V is the DC bus voltage
controlled converters, it may also exhibit steady state stability
d is the duty ratio of the switch problem under certain operating conditions due to the
From (4) & (2) the Control Law for the Sca-2 Control base presence of local feedback in its control structure [13]. The
PFC controller is, objective of the analysis here is to quanti@ the steady state
stability condition in terms of circuit parameters and the
(5) switching frequency of the converter.
mere
A
=m

b
t +

2
4 b
T,

Fig.7 Perturbationanatysis to evaluate steady state stability condition

Let I,& I ~ b ethe currents at the start and end of a

switching cycle respectively as shown in Fig. 7. Let AI,be
a small perturbation introduced in I1which produces a change
AI2 in current I2 . From (3,
J4

(7)
%
Fig. 5 Block D i a p of ScaIar Control PFC Converter Differentiating both sides with respect to line current I1
1,. =-vdc (6) Ad
-= - 1
-
Re AI1 2*1d
and is known as the current reference. From the steady state model of the boost converter, we get
dI
L-=V, -V,(l-2d)
4 dt
(91
Differentiating both sides with respect to d ,we get
+-1
2 AI2 ~Vdc (2 * Ts,
(10)
Ad L
Where,
T, i s the switching ftequency
L is the Line Inductance. Multiplying (8) & (lo), we get

_ 1_ A Converter is said to be stable in steady state if all

perturbations die down in subsequent cycles, which implies
that the expression in (1 1) should be less than 1. Therefore the
Fig. 6 Duty ratio command generation

864
condition €or stability of the scalar control method is given as, other state variables 41 g~ j,
The state variable description of the converter is given by (13)

v.
SMALL SIGNALANALysrs
A linear, low frequency, small signal model of Boost
converter is developed here. It may be noted that, in a dc-dc
converter, when it is operating under steady state, the input where, D is the duty ratio of the converter about the operating
power is equal b the output power in every switching period point & V,is the input voltage to the converter. The linear
T, of the converter. However the same is not true for single-
small signal model of the Boost converter is obtained by
phase resistor emulator rectifiers. The input voltage V, varies perturbing the state variables and the duty ratio input. The
from 0 to V, in a line cycle and under steady state condition, generalized small signal model equation is given as,
the input current I, is proportional to the input voltage. The & -

-=Ax+Bva
A -

+fd (14)
V
2 dt
c
instantaneous input power is, therefore v i -A which is * I

Where, A, B 8 f represent system matrices & x,va ,d are the

g-Re
perturbed variables. Equation (5) representing the steady state
1
not constant within an input cycle T =- whereas the relationship between the system variables & the duty ratio, d
2f when perturbed, will yield the relationship between the small
instantaneous output power - v& is constant. So, unlike dcdc signal 2 & the other system variables as given in (15)
R
converter, the power balance condition between input and
output cannot be satisfied at every switching period of the
resistor emulator rectifiers. However, it may be noted that, the
instantaneous input power is equal to the output power only at Where, i e i s the perturbed value of the emulated resistance,
Re-
the period in which the input voltage is L.Therefore this
JZ Substituting (15) in (14) yields,
switching period is chosen as the equivalent nominal
operating point that represents the entire line cycle for the
derivation of the small signal model. The control structure of
the small signal linear model of the Boost rectifier, switched
by input current shaping modulator, governed by Scalar
Control Law, is as shown in Fig. 8
It is observed ftom the system matrix, ‘A’ of (16) that the
product of off-diagonal terms is zero which implies that the
Small signal Linear system has real poles. Moreover, the term --m Re. system
mod el of the Boost L
matrix ‘A’ introduces additional damping to the system
Converter

$Liz;
response, making the system stable.
From (16), the transfer function of the converter can be
equated as,

je
mod el of the (1 7)
I Modulator I
Fig. 8 Control structure of Low frequency small signal Boost rectifier
The Bode plot representation of (17) is shown in Fig. 9.
Here the control gain transfer function is derived in two For a closed loop control system to be stable, it’s required that
steps [14][15]. First, the low ftequency small signal model of the loop gain crosses over 0 dE? (Unity Gain) with a single
the Boost Converter is obtained in the standard form in terms slope (-ZOdEVdecade). For a given system, transfer function
of duty ratio perturbation 2 as the control input. Subsequently [GIthe design of the closed loop compensator IHI should be
the small signal model of the modulator is derived to replace such that the loop gain IT1 (lGllHl) satisfies the above
2 term in the converter model by the perturbations in and requirement. As per the system transfer function given in (17)

865
the bode plot of the Compensator (PI controller) shown in The Control algorithm is tested for experimental
Fig. 10 satisfies the requisite condition of single slope 0 dB verification on a 1 kW single-phase Boast rectifier unit. The
crossover. AC line input is 138 V (rms]and the DC Bus voltage of the
rectifier is regulated at 400 V. The measured value of boost
inductance is 11 mH. The switching fkquency of the lGBT
t based (IPM Module) converter is chosen to be 10 kHz. The
load resistance is fixed at 250 R.

RC
Fig. 9 Rode plot ofsystem "fer function

\ t
I
b
I 2
- bgw
Fig. 12 (a) Chl: Input Voltage ;Ch2:Input Current
Voltage scale : I 0 0 V/div. Current Scale : 10 mV/A
RC
Fig. 10 Bode plot of Compensat01

VI. SIMWLATION & EXPERIMENTAL RESULTS

The Scalar Controlled UPF rectifier is simulated in Orcad
Pspice. The simulation results obtained, as shown in Fig. I 1
are taken with an input voltage of 230 V @k) and load
resistance of 250 !2. The Line inductance is kept small at 3
mH in order to have IOW voltage drop across it at full load.
The switching frequency is chosen to be 10 ICHZ.The DC bus
voltage is maintained constant at 360 V.

Fig. 12 @) Chl: Input Voltage ;Cb2:lnput cumnt

Voltage scale : 100 Vidiv. Cumnt Scale : 10 mV/A

Fig. 11 (a) lnput Voltage and hput Current in phase

Fig. 12 (c) Cbl: DC Bus Voltage

Voltage scale : 100 V/div.

Fig. 12(a) shows UF'F operation at the line side. Under

constant load conditions, Fig. 12(b) shows an increase in the
U
L
- ,111 r..,
5% 1.- 75-

Hc
I._ 2s-

Fig. 1I (bl DC Bus Voltage regulated at 360 V

866
input current with the decrease in line voltage. Tight [ti] M.Madigan, R.Erikson, E. Ismail, “Integrated high-
regulation of the DC bus voltage under the same condition is quality rectifier-regulators,” IEEE Power Elech-onics
evident fiom Fig, 12(c). The THD of the input current Specialists ConJ(PESC) Record, jun. 1992, pp. 1043 -
waveform calculated from the FFT shown in Fig. 12(d) is 1051.
found to be 12%, which indicates a very high input power [7] S.Teramoto, M.Sekine, R Saito, “High power factor
factor. AClDC converter,” U.S. Patent #5,301,095, Apr.5, 1994.
[8] F.S. Tsai, et al, “Low cost AC-to-DC converter having
input current with reduced harmonics,” U.S.Patent #
5,652,700, July,29,1997.
[9] P. Vinciarelli, “Optimal resetting of the transformer’s
core in single-ended forward converters,” U.S. Patent #
4,441,146, April 1984.
[lo] J.P. Gegner, C.Q. Lee, “Linear Peak Mode Control: A
simple Active Power Factor Correction Control
Technique for CCM,” IEEE Power Elec!ronics
Speciulists ConJ(FEX) Record, june 1996, pp. 196-202.
[ll] D. Maksimovic, Y.Jang, R.Erickson “Nonlinear-Carrier
Control for High Power Factor Boost Rectifiers,” APEC,
1995, pp.635-641.
[12] W. Tang,Y.Jiang, G. C. Hua, F. C. Lee, “Power Factor
Correction with Flyback Converter employing Charge
Frequency (Hz)
Control,” IEEE APEC 1993, pp.293-298.
[13] S. Hsu, A. Brown, L Rensink, R. D. Middlebrook,
Fig. 12 (d) FFT of Input Current “Modelling and Analysis of switching DC-to-DC
Converters in Constant Frequency Current Programmed
Mode,” PESC, June 1979.
W.CONCLUSION [14] F.Dong Tan, R. D.Middlebrook, “A UnifiedModel of
In this paper, a novel active power factor correction Current Programmed Converters,” IEEE Trulrsactions on
technique named Scalar control is proposed and analyzed. Power Electronics, Vol. 10, No. 4, pp.397-408,July
With the proposed method, the power factor of the single- 1995..
phase fiont-end converter is made unity without sensing the [ 151 R. D.Middlebrook, “Topics in Multiple Loop
input voltage. The control is extremely simplified & the same Regulators and Current Mode Programming,” IEEE
control technique could be extended to a three-phase system. Trunsactiom on Power Elecfronics, Vol. PE-2, No. 2,
The conditions for steady-state stability & the small-signal pp.109-124, april1987.
model are derived. The proposed method has the disadvantage [16] R.W.Erickson, D. Maksimovic, “Fundamentals of
of instability under certain conditions because of the nature of Power Electronics : Second Edition, Chapter 16: Power
the control. The results obtained from the simulation of the and harmonics in nonsinusoidal systems,” Publisher:
converter are also verified experimentally, Kluwer Academic Publishers,ISBN 0-7923-7270-0.

WI. REFERENCES
IEC 1000/3/2 International Standard,“Limitsfor
harmonic current emissions (equipment input current
<16A per phase)” ,1995.
R. Red, “ An Economical Single Phase Power-Factor
Corrected Rectifer: Topology, Operation, Extensions,
and design for compliance“, ZEEE AppIied Power
Electronics Conference (APEC),1998.
A.R Prasad, P.D.Ziogas, S. Manias, “A Novel Passive
Waveshaping Method for Single Phase Diode
Rectifiers ”, ZEEE Trunsuctions on Industrial
EIech-onics,December 1990.
E. Maset et al, “Improved Passive SoIutions to meet IEC
1000-3-2 Regulation in IOW cost Power
Supp1ies”Jnternational Telecommunications and Energy
Conference (INTELEC) 1996.
L.H Dixon, Jr., “High power factor pre-regulators for
off-line power supplies,”UnitrodeSwitching Regulator
Power Supply Design Seminar Manual, Paper 12, SEM-
700, 1990.

867