www.fairchildsemi.

com

Application Note 42047

Power Factor Correction (PFC) Basics

What is Power Factor? When the power factor is not equal to 1, the current wave-
form does not follow the voltage waveform. This results not
Power factor (pf) is defined as the ratio of the real power (P) only in power losses, but may also cause harmonics that
to apparent power (S), or the cosine (for pure sine wave for travel down the neutral line and disrupt other devices con-
both current and voltage) that represents the phase angle nected to the line. The closer the power factor is to 1, the
between the current and voltage waveforms (see Figure 1). closer the current harmonics will be to zero since all the
The power factor can vary between 0 and 1, and can be either power is contained in the fundamental frequency.
inductive (lagging, pointing up) or capacitive (leading, point-
ing down). In order to reduce an inductive lag, capacitors are
added until pf equals 1. When the current and voltage wave- Understanding Recent Regulations
forms are in phase, the power factor is 1 (cos (0°) = 1). The In 2001, the European Union put EN61000-3-2, into effect to
whole purpose of making the power factor equal to one is to establish limits on the harmonics of the ac input current up to
make the circuit look purely resistive (apparent power equal the 40th harmonic. Before EN61000-3-2 came into effect,
to real power). there was an amendment to it passed in October 2000 that
stated the only devices required to pass the rigorous Class D
Real power (watts) produces real work; this is the energy (Figure 2) emission limits are personal computers, personal
transfer component (example electricity-to-motor rpm). computer monitors, and television receivers. Other devices
Reactive power is the power required to produce the mag- were only required to pass the relaxed Class A (Figure 3)
netic fields (lost power) to enable the real work to be done, emission limits.
where apparent power is considered the total power that the
power company supplies, as shown in Figure 1. This total
power is the power supplied through the power mains to pro-
duce the required amount of real power.

“Total Power”
Apparent Power
(S) = Volt Amperes = I2Z
Reactive Power
(Q) = vars = (XL – XC) | 2

θ
Real Power Figure 2. Both Current and Voltage Waveforms are in
(P) = Watts = (I2R) Phase with a pF =1 (Class D)

Figure 1. Power Factor Triangle (Lagging)

The previously-stated definition of power factor related to

phase angle is valid when considering ideal sinusoidal wave-
forms for both current and voltage; however, most power
supplies draw a non-sinusoidal current. When the current is
not sinusoidal and the voltage is sinusoidal, the power factor
consists of two factors: 1) the displacement factor related to
phase angle and 2) the distortion factor related to wave
shape. Equation 1 represents the relationship of the displace-
ment and distortion factor as it pertains to power factor.
Figure 3: This is What is Called Quasi-PFC Input,
Irms(1) Achieving a pF Around 0.9 (Class A)
PF = cos θ = Kd ⋅ Kθ (1)
Irms
Causes of Inefficiencies
Irms(1) is the current’s fundamental component and Irms is
the current’s RMS value. Therefore, the purpose of the One problem with switch mode power supplies (SMPS) is
power factor correction circuit is to minimize the input that they do not use any form of power factor correction and
current distortion and make the current in phase with the that the input capacitor CIN (shown in Figure 4) will only
voltage. charge when VIN is close to VPEAK or when VIN is greater

REV. 0.9.0 8/19/04

AN-42047 APPLICATION NOTE

than the capacitor voltage VCIN. If CIN is designed using the V

input voltage frequency, the current will look much closer to Input Voltage Charging Bulk Input
the input waveform (load dependent); however, any little (Full Rectified) Capacitor Voltage (Vcin)
interruption on the mainline will cause the entire system to
react negatively. In saying that, in designing a SMPS, the Input Current
hold-up time for CIN is designed to be greater than the fre-
quency of VIN, so that if there is a glitch in VIN and a few
cycles are missed, CIN will have enough energy stored to 0 90 180 270 360 Deg
continue to power its load.
Figure 6. Voltage and Current Waveforms in a
Simple Rectifier Circuit
D1
V1 – + In order to follow VIN more closely and not have these high
Vo (to PWM)
amplitude current pulses, CIN must charge over the entire
cycle rather than just a small portion of it. Today’s non-linear
loads make it impossible to know when a large surge of cur-
Cin R1
rent will be required, so keeping the inrush to the capacitor
RTN constant over the entire cycle is beneficial and allows a much
smaller CIN to be used. This method is called power factor
Figure 4. SMPS Input Without PFC
correction.
Figure 5 represents a theoretical result of VCIN(t) (shown in
the circuit in Figure 4) with a very light load, and hence, very Boost Converters the Heart of Power
little discharge of CIN. As the load impedance increases, Factor Correction
there will be more droop from VCIN(t) between subsequent
Boost converter topology is used to accomplish this active
peaks, but only a small percentage with respect to the overall
power-factor correction in many discontinuous/continuous
VIN (e.g. with the input being 120V, maybe a 3-5 volt droop.
modes. The boost converter is used because it is easy to
As previously stated, CIN will only charge when VIN is
implement and works well. The simple circuit in Figure 7 is
greater than its stored voltage, meaning that a non-PFC cir-
a short refresher of how inductors can produce very high
cuit will only charge CIN a small percentage of the overall
voltages. Initially, the inductor is assumed to be uncharged,
cycle time.
so the voltage VO is equal to VIN. When the switch closes, the
130 current (IL) gradually increases through it linearly since:
Vc(t)
100
1
L∫
IL = VL dt .
0 Vin(t)
Voltage (VL) across it increases exponentially until it stabi-
lizes at VIN. Notice the polarity of the voltage across the
-100
inductor, as it is defined by the current direction (inflow side
-130
0 50 100 is positive). When the switch opens causing the current to
Time (s) change from Imax to zero (which is a decrease, or a negative
slope). Looking at it mathematically:
Figure 5. VIN with Charging CIN
di ∆i
After 90 degrees (Figure 6), the half cycle from the bridge VL = L ≈L ,
dt ∆t
drops below the capacitor voltage (CIN); which back biases
the bridge, inhibiting current flow into the capacitor (via or L times the change in current per unit time, the voltage
VIN). Notice how big the input current spike of the inductor approaches negative infinity (the inductor reverses polarity).
is. All the circuitry in the supply chain (the wall wiring, the Because the inductor is not ideal, it contains some amount of
diodes in the bridge, circuit breakers, etc) must be capable of series resistance, which loads this “infinite” voltage to a
carrying this huge peak current. During these short periods finite number. With the switch open, and the inductor dis-
the CIN must be fully charged, therefore large pulses of cur- charging, the voltage across it reverses and becomes additive
rent for a short duration are drawn from VIN. There is a way with the source voltage VIN. If a diode and capacitor were
to average this spike out so it can use the rest of the cycle to connected to the output of this circuit, the capacitor would
accumulate energy, in essence smoothing out the huge peak charge to this high voltage (perhaps after many switch
current, by using power factor correction. cycles). This is how boost converters boost voltage, as shown
in Figure 8.

2 REV. 0.9.0 8/19/04

APPLICATION NOTE AN-42047

rotating machinery and transformers and noise emissions in

IL Vi many products, and bringing about early failure of fuses and
other safety components. They also can cause skin effect,
Vi which creates problems in cables, transformers, and rotating
L V
Vin L
machinery. This is why power companies are concerned
Vo
with the growth of SMPS, electronic voltage regulators,
and converters that will cause THD levels to increase to
unacceptable levels. Having the boost preconverter voltage
Vin
higher than the input voltage forces the load to draw current
Imax 0 Vo
in phase with the ac main line voltage that, in turn, rids
0 IL harmonic emissions.
Vin
VL
0 Modes of Operation
There are two modes of PFC operation; discontinuous and
Figure 7. Flyback Action of an Inductor
continuous mode. Discontinuous mode is when the boost
converter’s MOSFET is turned on when the inductor current
reaches zero, and turned off when the inductor current meets
D1 the desired input reference voltage as shown in Figure 9. In
V1 – Lp D2 this way, the input current waveform follows that of the input
+
Vo voltage, therefore attaining a power factor of close to 1.
Cin
Iin Inductor Peak Current
PWM Q1 Inductor
CONTROL Current Inductor Average
Current
RTN

Figure 8. PFC Boost Pre-Regulator

The input to the converter is the full-rectified AC line volt-

age. No bulk filtering is applied following the bridge recti-
Gating
fier, so the input voltage to the boost converter ranges (at
Signal
twice line frequency) from zero volts to the peak value of the Figure 9. Discontinuous mode of operation
AC input and back to zero. The boost converter must meet
two simultaneous conditions: 1) the output voltage of the Discontinuous mode can be used for SMPS that have power
boost converter must be set higher than the peak value levels of 300W or less. In comparison with continuous mode
(hence the word boost) of the line voltage (a commonly used devices, discontinuous ones use larger cores and have higher
value is 385VDC to allow for a high line of 270VACrms), I2R and skin effect losses due to the larger inductor current
and 2) the current drawn from the line at any given instant swings. With the increased swing a larger input filter is also
must be proportional to the line voltage. required. On the positive side, since discontinuous mode
devices switch the boost MOSFET on when the inductor cur-
Without using power factor correction a typical switched- rent is at zero, there is no reverse recovery current (IRR)
mode power supply would have a power factor of around specification required on the boost diode. This means that
0.6, therefore having considerable odd-order harmonic dis- less expensive diodes can be used.
tortion (sometimes with the third harmonic as large as the
fundamental). Having a power factor of less than 1 along Continuous mode typically suits SMPS power levels greater
with harmonics from peaky loads reduces the real power than 300W. This is where the boost converter’s MOSFET
available to run the device. In order to operate a device with does not switch on when the boost inductor is at zero current,
these inefficiencies, the power company must supply addi- instead the current in the energy transfer inductor never
tional power to make up for the loss. This increase in power reaches zero during the switching cycle (Figure 10).
causes the power companies to use heavier supply lines, oth-
erwise self-heating can cause burnout in the neutral line con- With this in mind, the voltage swing is less than in discontin-
ductor. The harmonic distortion can cause an increase in uous mode—resulting in lower I2R losses—and the lower
operating temperature of the generation facility, which ripple current results in lower inductor core losses. Less
reduces the life of equipment including rotating machines, voltage swing also reduces EMI and allows for a smaller
cables, transformers, capacitors, fuses, switching contacts, input filter to be used. Since the MOSFET is not being
and surge suppressors. Problems are caused by the harmon- turned on when the boost inductor’s current is at zero, a
ics creating additional losses and dielectric stresses in very fast reverse recovery diode is required to keep losses to a
capacitors and cables, increasing currents in windings of minimum.

REV. 0.9.0 8/19/04 3

AN-42047 APPLICATION NOTE

Continuous Mode:
3
Average Current Mode
2.5
Inductor (Line) Current (A)

The heart of the PFC controller is the gain modulator. The

gain modulator has two inputs and one output. As shown in
2
Figure 13, the left input to the gain modulator block is called
the reference current (ISINE). The reference current is the
1.5
input current that is proportional to the input full-wave-recti-
fied voltage. The other input, located at the bottom of the
1 gain modulator, is from the voltage error amplifier. The error
amplifier takes in the output voltage (using a voltage divider)
0.5 after the boost diode and compares it to a reference voltage
of 5 volts. The error amplifier will have a small bandwidth so
0 as not to let any abrupt changes in the output or ripple errati-
Figure 10. Continuous Mode of Operation cally affect the output of the error amplifier.

Fairchild offers products for all discontinuous and continu- The gain modulator multiplies or is the product of the refer-
ous modes of PFC operation, including critical conduction ence current and the error voltage from the error amplifier
mode (FAN7527B), average current mode (FAN4810), and (defined by the output voltage).
input current shaping mode (FAN4803).
Figure 13 shows the critical blocks within the ML4821
Discontinuous Mode: (a stand alone PFC controller) to produce a power factor of
Critical Conduction Mode greater than 95 percent. These critical blocks include the cur-
rent control loop, voltage control loop, PWM control, and
A Critical Conduction mode device is a voltage mode the gain modulator.
device that works in the area between continuous and dis-
continuous mode. To better explain critical conduction mode The purpose of the current control loop is to force the current
lets look at the difference between discontinuous and contin- waveform to follow the shape of the voltage waveform. In
uous mode in a SMPS design such as a flyback converter. In order for the current to follow the voltage, the internal cur-
discontinuous mode, the primary winding of the transformer rent amplifier has to be designed with enough bandwidth1 to
has a dead time once the switch is turned off (including is a capture enough of the harmonics of the output voltage. This
minimum winding reset time) and before it is energized bandwidth is designed using external capacitors and resis-
again (Figure 11). tors. Once the bandwidth has been designed which in most
cases is a few kHz (to not be affected by any abrupt tran-
sient), it uses information from the gain modulator to adjust
Ipk
the PWM control that controls whether the power MOSFET
is switched on or off.
0
Figure 11. Discontinuous Mode, Flyback Power Supply Ip The gain modulator and the voltage control loop2 work
(Primary Current) together to sample the input current and output voltage,3
respectively. These two measurements are taken and than
In continuous mode, the primary winding has not fully compared against each other to determine if a gain should be
depleted all of its energy. Figure 12 shows that the primary applied to the input of the current control. This decision is
winding does not start energizing at zero, rather residual than compared against a sample of the output current to
current still resides in the winding. determine the duty cycle of the PWM.

Ipk The PWM control uses trailing-edge modulation as shown in

Figure 14.
0
Figure 12. Continuous Mode, Flyback Power Supply IP
(Primary Current)
1The bandwidth is set by Fswitching/6
In critical conduction mode there are no dead-time gaps 2The voltage control loop also needs to be bandwidth limited,
between cycles and the inductor current is always at zero
Again, this is designed using external passive components.
before the switch is turned on. In Figure 9, the ac line current 3The output voltage of a continuous inductor current boost
is shown as a continuous waveform where the peak switch regulator has to be set above the maximum peak of the input
current is twice the average input current. In this mode, the voltage in order to function correctly as a PFC. The output
operation frequency varies with constant on time. should be 1.414 times the maximum input voltage.

4 REV. 0.9.0 8/19/04

APPLICATION NOTE AN-42047

I D ILOAD
DC IN
+
IL
ID
Q
AC +
C DC
IN IPR
IC OUT
DBR RL

IL
– –
RS
IGM
RC RCL Voltage Control Loop
Current Control Loop

IA OUT
ZCF 2

IA–
3 –
–
+ R
+
IA+
4 S Q
OUT
IGM 14
ISINE GAIN
5
MODULATOR
Clock
EA OUT
ZF 6
OSC
INV
ZI 7 –
Ramp
E/A
VREF +

Figure 13. Example of an Average Current Mode PFC Control (ML4821)

Current
Reference

Input

Inductor Current

TON Output

TS

Figure 14. Trailing-Edge Modulation4 Figure 15. Typical Average Current Mode Waveform

4 Trailing edge modulation is when the output switches on when the output of the comparator passes through the trailing edge
of the sawtooth wave created.

REV. 0.9.0 8/19/04 5

AN-42047 APPLICATION NOTE

The line that goes through the saw tooth waveform is the out- Using the continuous mode characteristic, the following
put of the differential amplifier within the current loop con- equations show that the inductor current is proportional to
trol. The output of the differential amplifier (located on the the sinusoidal waveform at the turn-on time. Therefore the
top of Figure 13) goes into an R-S flip flop that controls the inductor current minimum value during one switching cycle
power MOSFET. The average current mode waveform is follows the sinusoidal current reference as shown in Figure
shown in Figure 14. Figure 15 shows the waveform of what a 18. However, the inductor current peak value during one
typical average current PFC device looks like. switching cycle is not controlled to follow the sinusoidal ref-
erence. Therefore the average inductor current might not be
Continuous Mode: sinusoidal. To make the average inductor current close to the
Input Current Shaping sinusoidal reference, the inductance has to be high enough to
make the current ripple small.
Fairchild’s FAN4803 features input current shaping, another di L
control method of the continuous current mode PFC. Figure VL = VIN = L : During on-time
t on
16 shows the internal PFC block of the FAN4803. Unlike the
conventional/typical average current mode PFC controller, di L
the FAN4803 does not need input voltage information and a VL = (VIN − VOUT ) = L : During off-time
t off
multiplier. It changes the slope of an internal ramp according
to the error amplifier output voltage, while the current sense t off V IN
information and the ramp signal are used to determine the VIN • t on = (VOUT − VIN ) • t off , = : CCM condition
TS VOUT
turn-on time. As shown in Figure 17a, the switch is turned on
when the current sense voltage meets the internal ramp sig- t off VIN
nal and the switch is turned off by the internal clock signal. VCS = Vramp = Veao = Veao : Switch off to on
TS VOUT instant
To control the output voltage, the slope of the internal ramp
signal is adjusted. By comparing Figure 17a and Figure 17b, Veao
Rs • i L (t O + t off ) = VIN • sin(ω t)
one can see that the average current increases if the slope VOUT
increases and decreases if the slope decreases.
∴ I L (min) = i L (t O + t off ) ∝ sin(ω t )

VOUT = 400V

RP
VC1
VEAO Gate
Output
4 –
COMP
C1 +
30pF
RCOMP
35µA 5V
CCOMP
R1
CZERO

ISENSE
3 –4
VI SENSE

Figure 16. Example of the Input Current Shaping PFC Controller (FAN4803

6 REV. 0.9.0 8/19/04

APPLICATION NOTE AN-42047

Vramp = Veao (toff / TS) Vcs

Average Current
Vcs = Rs • iL

Vramp
to toff to + toff t o +TS

TS

PFC OUT

Clock

Figure 17a. Typical Input Current Shaping PFC Waveform

Vramp = Veao (toff / TS) Vcs

Average Current
Vcs = Rs • iL

Vramp
to t off to + toff t o +TS
TS

PFC OUT

Clock

Figure 17b. Typical Input Current Shaping PFC Waveform

Inductor
Current
Current
Reference

Figure 18. Input Current Shaping PFC Waveform

REV. 0.9.0 8/19/04 7

AN-42047 APPLICATION NOTE

Leading Edge Modulation/Trailing Edge Fairchild Patented Leading Edge Modulation/

Modulation (LEM/TEM) versus Trailing Trailing Edge Modulation (LEM/TEM)
Edge Modulation/Trailing Edge Modulation Technique
(TEM/TEM) In LET/TEM the PFC and PWM switches are tied together,
Leading edge/trailing edge modulation is a patented Fair- but opening and closing 180 degrees out of phase, so when
child technique to synchronize the PFC controller to the the PFC switch is open the PWM switch is closed and vice
PWM controller. Typically TEM/TEM is used in PFC/PWM versa. Initially when the PFC switch is closed, the PFC
controllers which results in an additional step as well as a inductor is energized, once the PWM switch is closed, both
larger PFC bulk capacitor (as shown below). the output and the PFC bulk capacitor are energized. Figures
20a and 20b show that upon repetition of this cycle, the PFC
Trailing Edge Modulation/Trailing Edge bulk capacitor does not have to be that large because it is not
Modulation (TEM/TEM) powering the output all by itself, the PFC inductor is helping
out as well.
Figure 19a shows the PFC inductor being energized.

Figure 19b shows the energy from the inductor being trans-
ferred into the PFC bulk capacitor.

When the PWM switch is closed, as shown in Figure 19c, the

energy stored within the PFC bulk capacitor is used to drive
the load. Every time this cycle is repeated, the PFC bulk
capacitor has to be fully charged since it is fully discharged
when the PWM switch is closed.

8 REV. 0.9.0 8/19/04

APPLICATION NOTE AN-42047

PFC Section PWM Section

Open
Vout

Signal Current
AC Output
Closed Cap

PFC Bulk
Cap
GND

Figure 19a. Energizing the PFC Inductor

PFC Section PWM Section

Open
Vout

Signal Current
AC Output
Open Cap
PFC Bulk
Cap

GND

Figure 19b. Charging the PFC Bulk Capacitor

PFC Section PWM Section

Closed
Vout

Signal Current
AC Output
Open Cap
PFC Bulk
Cap

GND

Figure 19c. Powering the Output

REV. 0.9.0 8/19/04 9

AN-42047 APPLICATION NOTE

PFC Section PWM Section

Open
Vout

Signal Current
AC Output
Cap
Closed
PFC Bulk
Cap

GND

Figure 20a. Energizing the PFC Inductor

PFC Section PWM Section

Current

Vout
Closed
Signal
AC
Output
Open Cap
PFC Bulk
Cap

GND

Figure 20b. Charging the PFC Bulk Capacitor and Powering the Output

Conclusion
Power companies do not get excited over low power factor
driven devices, plus the extra cost of unused or wasted power
can be quite large. This is why PFC on the device side has
become an important part of the final power system design
for so many products. There are many standards in place
(example, EN 61000-3-2) to drive power consumption to a
power factor of 1 and keep total harmonic distortion to a
minimum. Depending on the output power and the designer’s
needs, a SMPS can be designed with either a discontinuous
or continuous mode stand alone PFC controller, or a continu-
ous PFC/PWM mode device can be used. PFC controllers
are forecasted to grow to $175 million in 2006, and stan-
dards are reducing the minimum power limits on systems
that require PFC, more and more PFC controllers will be
used.

10 REV. 0.9.0 8/19/04

AN-42047 APPLICATION NOTE

DISCLAIMER
FAIRCHILD SEMICONDUCTOR RESERVES THE RIGHT TO MAKE CHANGES WITHOUT FURTHER NOTICE TO ANY
PRODUCTS HEREIN TO IMPROVE RELIABILITY, FUNCTION OR DESIGN. FAIRCHILD DOES NOT ASSUME ANY
LIABILITY ARISING OUT OF THE APPLICATION OR USE OF ANY PRODUCT OR CIRCUIT DESCRIBED HEREIN; NEITHER
DOES IT CONVEY ANY LICENSE UNDER ITS PATENT RIGHTS, NOR THE RIGHTS OF OTHERS.

LIFE SUPPORT POLICY

FAIRCHILD’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES
OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT OF FAIRCHILD SEMICONDUCTOR
CORPORATION. As used herein:

1. Life support devices or systems are devices or systems 2. A critical component is any component of a life support
which, (a) are intended for surgical implant into the body, device or system whose failure to perform can be
or (b) support or sustain life, or (c) whose failure to perform reasonably expected to cause the failure of the life support
when properly used in accordance with instructions for use device or system, or to affect its safety or effectiveness.
provided in the labeling, can be reasonably expected to
result in significant injury to the user.

www.fairchildsemi.com

8/19/04 0.0m 001

Stock#AN30042047
 2004 Fairchild Semiconductor Corporation