Beruflich Dokumente
Kultur Dokumente
ABSTRACT
For the conventional PFC rectifiers, the high efficiency cannot be achieved over a
wide input range, and the efficiency will be greatly decreased at low input
voltages. In order to overcome the efficiency bottleneck under low line input, a
flexible mode bridgeless boost power factor correction (PFC) rectifier is proposed
in this paper. According to the input voltage, the proposed rectifier can be flexibly
adapted to the suitable operating mode to obtain the maximum efficiency.
Meanwhile, the circuit components can be reused by different operating modes, so
the extra cost is low. In the proposed rectifier, a back-to-back bridgeless boost
PFC topology is adopted at high line conditions and a three-level bridgeless boost
PFC topology is rebuilt to reduce the switching losses at low line conditions.
Compared with the traditional bridgeless boost PFC rectifier, an extra low-voltage
bidirectional switch (usually composed of two switches) is added, so the increased
cost is low. In addition, the low common mode noise can be achieved at both high
and low line conditions due to the direct connection between the input mains and
the output electrolytic capacitor. The detailed principle analysis about the proposed
rectifier is presented in this paper. Finally, an experimental prototype is built to
verify the feasibility and the effectiveness of the proposed topology.
EXISTING SYSTEM
It seems to be much difficult to further reduce the conduction losses for the
bridgeless PFC rectifiers; so many bridgeless PFC rectifiers have been modified to
improve the efficiency via soft switching techniques. According to the ways of
realizing soft switching, these modified PFC rectifiers can be divided into several
groups. The first group adopts the discontinuous current mode (DCM) or critical
current mode (CRM) instead of continuous current mode (CCM). The rectifiers
working in DCM or CRM mode are only suitable for low power applications (<
300 W) due to the restriction of switch current stress. Additionally, a large input
filter must be employed to suppress the high frequency components of
the pulsating input current, which increases the overall volume and cost of the
rectifier. The second group involves adding some auxiliary components (including
the capacitors, inductors and active switches) to the traditional bridgeless PFC
rectifiers, which usually makes the circuit more complex and decreases the
reliability of system. Besides, the voltage or current stress of the power devices is
also significantly increased due to the resonance between the capacitors and
inductors. That is, the higher-rated or, usually, more expensive components are
needed.
PROPOSED SYSTEM
Microcontroller circuit
INTRODUCTION
N order to meet the harmonic regulations and standards, such as the IEC 610000-3-
2 [1], the power supplies with active power factor correction (PFC) feature are
required for various types of electronic equipment. Meanwhile, the increasing
requirements of high efficiency have been forcing the designers to look for any
potential opportunities to reduce the converter losses. The most common solution
is the conventional PFC rectifier shown in Fig. 1 which is widely used for various
kinds of computers, workstations, servers and so on [2]-[3]. However, it suffers
from high conduction losses since the input current always flows through two
bridge diodes and a power switching device. Besides, the efficiency decrease
caused by conduction losses is more obvious when the input line voltage is low,
which usually leads to low power density since a large heat sink is needed. In order
to overcome the shortcomings of the conventional PFC converter, many
researchers have made great efforts to develop the bridgeless boost PFC rectifiers
which can achieve Ihigher efficiency by reducing the number of power
components in the line current path [4]-[9], [11]-[13]. In these boost bridgeless
PFC rectifiers, the most practical and potential topologies are the dual-boost
bridgeless PFC rectifier (DBBL PFC) shown in Fig. 2 and the back-to-back
bridgeless PFC rectifier (BTBBL PFC) shown in Fig. 3 since both of them have
lower common mode (CM) noise interference and better reliability compared with
the others. Nevertheless, it can be noted that the magnetic core utilization is low
for the DBBL PFC rectifier. Although a multiple-winding, multicore inductor can
be used to improve the utilization of the magnetic material [9], the efficiency of the
rectifier is also reduced, which is not expected. For the above two rectifiers, the
high efficiency can be achieved when the input voltage is high (rating value is
220~240 Vrms). If the input voltage decreases, the total conduction and switching
losses although are reduced than the topology shown in Fig. 1, however are still
high with the increase of input current, the efficiency under low input voltages is
still much lower than that under high input voltages. It seems to be much difficult
to further reduce the conduction losses for the bridgeless PFC rectifiers, so many
bridgeless PFC rectifiers have been modified to improve the efficiency via soft
switching techniques. According to the ways of realizing soft switching, these
modified PFC rectifiers can be divided into several groups. The first group adopts
the discontinuous current mode (DCM) or critical current mode (CRM) instead of
continuous current mode (CCM) [9], [14]-[17]. The rectifiers working in DCM or
CRM mode are only suitable for low power applications (< 300 W) due to the
restriction of switch current stress. Additionally, a large input filter must be
employed to suppress the high frequency components of the pulsating input
current, which increases the overall volume and cost of the rectifier. The second
group involves adding some auxiliary components (including the capacitors,
inductors and active switches) to the traditional bridgeless PFC rectifiers [18]-[21],
which usually makes the circuit more complex and decreases the reliability of
system. Besides, the voltage or current stress of the power devices is also
significantly increased due to the resonance between the capacitors and inductors.
That is, the higher-rated or, usually, more expensive components are needed.
Another common solution to improve the efficiency is to adopt multilevel
converters which have the advantages of small inductor size, low switching losses,
low device stress, etc. Based on the full-bridge PFC converter shown in Fig. 1, a
traditional three-level boost converter is proposed [22], which suffers from high
conduction losses due to the diode bridge. So an improved three-level PFC rectifier
shown in Fig. 4 is developed to reduce the conduction losses [23]-[25], and the
high efficiency can be achieved. Recently, several new three-level single-phase
bridgeless PFC rectifiers are presented in [26], which have lower conduction losses
than Fig. 4. However, many extra power devices and high-side drivers are needed
for those complicated circuits proposed in [26]. On the other hand, the advantage
of low CM noise in the previous two-level bridgeless PFC rectifiers is lost for all
of the three-level rectifiers mentioned above since the voltage potential of the
output bus in regard to the ground is pulsating. A new concept of flexible converter
is proposed in [27]. There are two or more topologies or operating modes involved
in a flexible converter, where different topologies are formed for different
applications. In order to solve the problem of low efficiency at low input for the
rectifier, a flexible mode bridge less boost PFC (FMBL PFC) converter is proposed
based on the flexible converter concept. The basic design principle can be
concluded as follows. According to the input voltage, the rectifier can be flexibly
adapted to the suitable topology and mode for obtaining the maximum efficiency.
Meanwhile, in order to reduce the extra cost, the circuit components should be
reused as much as possible in different topologies and modes. Based on this idea, a
novel flexible mode bridgeless PFC rectifier (FMBL PFC) is proposed, in which
the high efficiency over a wide input range can be achieved. In the proposed
rectifier, a BTBBL PFC rectifier is adopted at high line voltages and a three-level
bridgeless boost PFC rectifier (TLBL PFC) is formed to achieve high efficiency at
low line voltages. Compared with the traditional bridgeless boost PFC rectifier, an
extra low-voltage bidirectional switch (usually composed of two switches) is
added, therefore the increased cost is low. At both high and low line conditions,
low CM noise can be achieved due to the direct connection between the input
power grid and the output electrolytic capacitor during half line cycle. The detailed
principle analysis about the proposed FMBL PFC rectifier is presented. Finally, an
experimental prototype is built to verify the feasibility and the effectiveness of the
proposed topology. II.THE DERIVATION OF THE NOVEL PROPOSED
TOPOLOGY AND THE OPERATING PRINCIPLE A. The derivation of the
novel proposed topology Fig. 5 shows the asymmetrical BTBBL PFC rectifier,
which is a little different from the symmetrical one shown in Fig. 3. In the
symmetrical BTBBL PFC rectifier, two identical inductors are adopted to make the
circuit look symmetrical because the symmetrical circuit has a better capability to
suppress the CM noise [11]-[12]. However, four fast recovery diodes are needed in
the symmetrical topology, which increases the cost. Therefore, an asymmetrical
structure with single inductor 𝐿 shown in Fig. 5 is used here. From Fig. 5, it can be
seen that the diodes in two half bridges are different: the half bridge connected to
the inductor 𝐿 is made up of two fast recovery diodes and the other one connected
to the input mains 𝑣𝑠 is made up of two slow diodes. When the fast recovery diode
and the slow diode are in series, the switching characteristic is determined by the
fast recovery diode and the slow diode can be treated as short circuit, which will be
analyzed in the next section. Therefore, one end of the input power supply is
always directly connected to the positive terminal of the output capacitor 𝐶𝑜 in the
negative half line cycle or the negative terminal of 𝐶𝑜 in the positive half line
cycle. Then the CM noise is reduced effectively. The high efficiency under high
input voltages for the BTBBL PFC rectifier can be achieved. However, it suffers
from low efficiency due to the rapid increase of switching losses and conduction
losses under low input voltages. Fig. 6 shows the topology of a three-level
bridgeless boost PFC rectifier (TLBL PFC), whose structure is simple. The three-
level bridgeless boost PFC rectifier has the following merits: First, there are fewer
semiconductor components in the current path. When the bidirectional switch 𝑆1𝑆2
is turned off, there is only one fast diode to carry the current in the TLBL PFC,
while two diodes in the BTBBL PFC and DBBL PFC Second, low voltage
MOSFETs can be used in the TLBL PFC rectifier. For the same input current , the
switching losses in the TLBL PFC are less than that in the BTBBL PFC and DBBL
PFC. In addition, if the same filter inductor is used, the operating frequency of the
TLBL PFC rectifier can be reduced for the same current ripple requirement, which
means the switching losses can be decreased further. According to above analysis,
the TLBL PFC should be more attractive for the industry applications. The
efficiency can be increased with low input voltage, such as 110 Vrms input and
400 Vdc output. However, when the input voltage rises, such as reaching the rating
grid voltage 220 Vrms, the expected 400 Vdc cannot be achieved by using TLBL
PFC. It can be seen that the input voltage limit of TLBL PFC comes from the use
of half bridge, which limits its wide application. From the above analysis, the
DBBL PFC and BTBBL PFC can work in a wide range of input voltage but suffer
from low efficiency at low line conditions; the TLBL PFC can achieve high
efficiency, but cannot work under high line conditions. In order to achieve high
efficiency over a wide input voltage range, a potential solution is to extend the
input voltage range of TLBL PFC by disconnecting the midpoint of half bridge
from the negative terminal of input supply and adding some components to convert
the TLBL PFC into DBBL PFC or BTBBL PFC. That is, a flexible mode PFC
rectifier is expected to be developed which can be flexibly adapted to the suitable
topology and operating mode according to the input voltage. Fortunately, the
BTBBL PFC shown in Fig. 5 and the TLBL PFC shown in Fig. 6 look similar to
each other, which makes it possible to combine them into a new topology. On the
basis of BTBBL PFC rectifier, an extra bidirectional switch (composed of two
switches) is added to connect the drain of MOSFET 𝑆2 to the midpoint of the
output capacitors. Therefore, a novel topology is constructed as shown in Fig. 7,
which is the proposed flexible mode bridgeless PFC rectifier. In Fig. 7, 𝑆1𝑆2 and
𝑆3𝑆4 are two bidirectional switches; 𝐷𝐹1 and 𝐷𝐹2 are fast recovery diodes; 𝐷𝑆1
and 𝐷𝑆2 are slow diodes; 𝐿 is the input filter inductor; 𝐶𝑜1 and 𝐶𝑜2 are the output
split-capacitors (𝐶𝑜1 = 𝐶𝑜2 = 𝐶𝑜); 𝑅𝑜 is the load resistance; 𝑣𝑠 is the input line
voltage and 𝑉𝑜 is the output voltage (𝑉𝑜 = 2𝑉𝐶𝑜1 = 2𝑉𝐶𝑜2). It is important to note
that the voltage stress of MOSFETs 𝑆3 and 𝑆4 is only half of the output DC
voltage. Therefore, low-voltage MOSFETs can be adopted here to reduce the extra
conduction losses. For example, the MOSFET with breakdown voltage 250 V can
be used for the output voltage 400 Vdc, and its conduction resistance is only 20
𝑚Ω. That is, the total conduction resistance of the bidirectional switch 𝑆3𝑆4 is
only 40 𝑚Ω and the voltage drop across 𝑆3𝑆4 is usually less than a diode forward
voltage drop.
SEMICONDUCTOR DIODES
4. Semiconductor Diodes
Introduction
So far we have looked at only so-called passive electronic devices: capacitors
and resistors respond to voltages applied across them by accumulating charge or
passing currents, respectively. In the next two labs, we’ll take a look at two active
devices which behave very differently depending upon what voltages are applied to
them. The diode essentially acts as a one-way switch controlled by voltage. For
one polarity of voltage, if the voltage across the diode is greater than a threshold
value (often ~0.6 Volts), it conducts current with essentially no resistance. If the
voltage is below that value, or has the opposite polarity, the diode acts as an open
switch and conducts no current. The transistor differs from the other devices
we’ve considered so far in that it has three leads or connections. The voltage
applied between two of these leads controls whether current can be conducted
between two others. Although the transistor also acts as a switch, it does not
merely shift between fully off and fully on. Its importance lies in the fact that a
relatively low-power voltage supply can control the flow of a more powerful
current over a range of values. It is this switching and decision-making property of
diodes and transistors, which makes all of modern day electronics, including
computers and telecommunications, possible.
The study of semiconductors and the devices made from them falls under the
category of solid state physics. In this experiment, we will work with one useful
device: the diode. Although we will discuss the theory briefly, you will mainly be
expected to be able to understand how diodes function in circuits.
Energy Bands
The electrons of an isolated atom have discrete allowed energies that we call
energy levels. The Pauli Exclusion Principle states that at most two electrons can
occupy any allowed energy level. For example, Figure 1 shows schematically the
energy levels for a Lithium atom. The vertical scale is associated with increased
energy. The dots denote electrons occupying a given level. In order to minimize
energy the electrons fill the levels from the bottom up.
Electron
Energy
When two isolated atoms are brought close together their electric fields interact
and cause a splitting of the energy levels. Each original energy level splits into
two, one going slightly higher and the other going lower (see Figure 2). (A
detailed explanation of the cause of this will have to be postponed until a quantum
mechanics course.) In a crystalline solid, where many atoms exist close together,
this effect is multiplied many times over. Each energy level spreads out over a
small continuous range of energies called an energy band.
far apart both close together many atoms close
together
Figure 2: Energy level splitting due to atomic interaction
If the original energy levels were spaced closely together the associated energy
bands might overlap, resulting in a larger continuous band. Sometimes, however,
the energy levels are spaced far enough apart that the bands don't overlap. This
“gap” between bands is what provides the interesting physics of semiconductor
devices. Remember that energy level diagrams represent allowed or accessible
states that an electron may occupy. A gap between bands indicates a forbidden
energy range for electrons.
Semiconductors
Semiconductors are a special case of insulators in which the forbidden region
between the valence and conduction bands is relatively small (about 0.5 to 1.5 eV).
In this case an extremely small number of electrons are excited across the energy
gap by thermal excitation and occupy states in the conduction band. These excited
electrons can respond to the applied fields but since their number is comparatively
few, the material as a whole is not a good conductor, so we call it a semiconductor.
The conductivity of a semiconductor is very sensitive to temperature since it
depends on thermally excited electrons.
Doping
In general, the amount of current that a semiconductor can carry is not enough
to make a useful device. Most commercial semiconductors are made by
introducing small amounts of impurities to an intrinsic semiconductor (a process
called doping). We will use silicon as an example.
Silicon (Si) is an intrinsic semiconductor, but in its natural state it conducts very
poorly. Silicon is a group IV element on the periodic chart and has four electrons
in its outermost shell. When silicon is doped with arsenic (As), a group V element,
the arsenic atoms replace silicon atoms at a small number of points on the crystal
lattice. Since arsenic has 5 electrons in its outer shell, it adds a loosely-bound
“extra”electron to the crystal. This extra electron (often called a “donor”electron)
is easily excited into the conduction band as a freely roaming current carrier.
III IV V
5B 6C 7N Group III: -1e-: acceptor Ñ p type
13Al 14Si 15P
31Ga 32Ge 33As Group V: +1e-: donor Ñ n type
49In 50Sn 51Sb
Silicon can also be doped with an element from group III of the periodic table,
such as gallium (Ga). In this case the impurity has only three electrons in its
outermost shell so there is a deficiency of one electron at every point where a
gallium atom replaces a silicon atom. This is called an “acceptor site” since the
gallium would very much like to have a fourth electron to complete its bonds. The
gallium often “steals”an electron from a neighboring silicon atom leaving a “hole”
or empty state in the valence band of the silicon. This “hole” is free to roam
around in the valence band and effectively acts as a positive charge carrier.
Holes move through the crystal lattice in the same way spaces between cars
move in a traffic jam: the cars (electrons) move forward to fill up the spaces
(holes) in front of them, only to create another space behind them. The holes move
in the direction opposite the electrons, hence the “effective” positive charge.
Doping with group V elements results in an n-type semiconductor since the
charge carriers (electrons) are negative. On the other hand, doping with group III
elements produces a p-type semiconductor because the effective charge carriers
(holes) are positive.
Diodes
Figure 7b: Diffusion of excess charge in a p-n junction after charge diffusion. The
negative charges which have diffused out of the n-type side into the p-type side
leave a net negative charge there, and vice-versa. This charge distribution causes
the “built-in” electric field shown near the junction.
Reverse Biased Junction
Figure 8 shows a p-n junction attached to a voltage source such that the positive
terminal is connected to the n-type semiconductor. The electric field due to the
applied voltage source adds to the built-in field. (Note that the conductivity of the
p- and n-type regions away from the junction is greater than that of the junction
region so the potential varies mainly in the proximity of the junction.) Hence, the
addition of the second field even further opposes the motion of any holes from the
p-type side which are trying to move to the right and any electrons from the n-type
side which are trying to move to the left.
So, very little current flows.
Figure 8: Distribution of mobile charges in a reverse biased diode
Figure 9: Circuit diagram of a forward biased diode, with resistor added to limit
the current.
Experimental Procedure
Experiment 4-1: Diode Tester
Sometimes, despite precautions, too much current passes through a diode and it
"burns out". This "burn out" is not very flashy; in fact, you wouldn’t know that
anything happened except for the fact that your circuit would malfunction. The
diode would look the same as before and you would be wondering what went
wrong. Obviously, a quick test of a diode’s health would be useful for
troubleshooting purposes.
R = 100
Pre-lab question 2: We want Ch. 1 to display the voltage across the diode
and Ch. 2 to display the voltage across the resistor. Explain why we
couldn’t simply connect Ch. 1 to the top wire (as it’s shown), Ch. 2 to the
middle wire, and the scope ground to the bottom wire, i.e. explain why
we can’t interchange the connections for scope ground and Ch. 2.
(a) (b)
1) Explain how this circuit works by tracing the current paths (a) when the
applied voltage is positive and (b) when it is negative.
2) Build the circuit with a 10 F capacitor in parallel with the resistor to
eliminate most of the AC ripple, leaving a nearly constant DC voltage.
Compare the mean voltage for this circuit (again expressed as a fraction
of the p-p input) to what you found for the half-wave rectifier. Do you
see why this circuit is a better DC power supply?
1) Using the same method as in experiment 4-2, plot the I-V curve of the
zener diode . However, before turning on the power CALCULATE THE
MAXIMUM ALLOWED CURRENT for the circuit, given that the
maximum power that can be dissipated in the Zener is 400mW. (To do
this, remember that P = I V.)
2) Record the value at which the zener diode begins to conduct in the
reverse-biased direction.
1) Construct the circuit in Figure 13. Before turning on the DC voltage, you
need to consider whether this circuit will keep you within the 1/4 W limit
for the resistors and the 400mW limit for the Zener. To do this, suppose
that the pot resistance is high so it doesn't draw any current. Then all the
current goes through the Zener. How much will that current be (given
the known Zener voltage for the diode you have been given? How much
power will be dissipated in the resistor and the Zener? To avoid any risk
to the pot, don't set it for less than 100 ohms.
2) Test the circuit by measuring VL (and hence IL) for a few (e.g. 5) values
of RL between 100 ohms and 1000 ohms. You'll have to disconnect the
pot when you adjust it to check its resistance with the DMM. Over what
range of load currents does this circuit maintain a constant voltage
output?
3) Make sure that you understand how the currents are changing to keep VL
constant as RL is varied. Ask your instructor if you're unsure.
Congratulations! You have created a DC to DC converter, a device that
can be used to provide a constant DC voltage to a circuit independent of
variations in the input voltage or the load.
The full-wave rectifier from part 4-4 and the Zener regulator from part 4-6 can be
used to build a regulated power supply Figure 15. This power supply should
deliver a constant voltage (Vout = Vz) over a wide range of output currents.
Switching loss
In the case of SMD (surface mounted) electrolytic capacitors, there are two basic
marking types. The first one clearly states the value in microfarads and the
operating voltage. For example, using this approach, a 4.7 μF capacitor with an
operating voltage of 25 volts would bear the marking “4.7 25V”. In the other
marking system, a letter is followed by three numbers. The letter represents the
voltage rating according to the table below. The first two numbers represent the
value in picofarads, while the third number is the number of zeroes to be added to
the first two. For example, a 4.7 μF capacitor with a voltage rating of 25 volts
would bear the marking E476. This translates to 47000000 pF = 47000 nF = 47 μF.
Characteristics
Capacitance drift
The capacitance of electrolytic capacitors drifts from the nominal value as time
passes, and they have large tolerances, typically 20%. This means that an
aluminum electrolytic capacitor with a nominal capacitance of 47µF is expected to
have a measured value of anywhere between 37.6µF and 56.4µF. Tantalum
electrolytic capacitors can be made with tighter tolerances, but their maximum
operating voltage is lower so they cannot be always used as a direct replacement.
Polarity and safety
Due to the construction of electrolytic capacitors and the characteristics of the
electrolyte used, electrolytic capacitors must be forward biased. This means that
the positive terminal must always be at a higher voltage than the negative terminal.
If the capacitor becomes reverse-biased (if the voltage polarity on the terminals is
reversed), the insulating aluminum oxide, which acts as a dielectric, might get
damaged and start acting as a short circuit between the two capacitor terminals.
This can cause the capacitor to overheat due to the large current running through it.
As the capacitor overheats, the electrolyte heats up and leaks or even vaporizes,
causing the enclosure to burst. This process happens at reverse voltages of about 1
volt and above. To maintain safety and prevent the enclosure from exploding due
to high pressures generated under overheat conditions, a safety valve is installed in
the enclosure. It is typically made by making a score in the upper face of the
capacitor, which pops open in a controlled manner when the capacitor overheats.
Since electrolytes may be toxic or corrosive, additional safety measures may need
to be taken when cleaning after and replacing an overheated electrolytic capacitor.
Electrolytic capacitors have a larger capacitance than most other capacitor types,
typically 1µF to 47mF. There is a special type of electrolytic capacitor, called a
double-layer capacitor or a supercapacitor, whose capacitance can reach thousands
of farads. The capacitance of an aluminum electrolytic capacitor is determined by
several factors, such as the plate area and the thickness of the electrolyte. This
means that a large capacitance capacitor is bulky and large in size.
It is worth mentioning that electrolytic capacitors made using old technology didn’t
have a very long shelf life, typically only a few months. If left unused, the oxide
layer deteriorates and has to be rebuilt in a process called capacitor reforming. This
can be performed by connecting the capacitor to a voltage source through a resistor
and slowly increasing the voltage until the oxide layer has been fully rebuilt.
Modern electrolytic capacitors have a shelf life of 2 years or more. If the capacitor
is left unpolarized for longer periods, they must be reformed prior to use.
They may also be used in input and output smoothing as a low pass filter if the
signal is a DC signal with a weak AC component. However, electrolytic capacitors
do not work well with large amplitude and high frequency signals due to the power
dissipated at the parasitic internal resistance called equivalent series resistance
(ESR). In such applications, low-ESR capacitors must be used to reduce losses and
avoid overheating.
HIGH EFFICIENCY
Efficiency is the (often measurable) ability to avoid wasting materials, energy,
efforts, money, and time in doing something or in producing a desired result. In a
more general sense, it is the ability to do things well, successfully, and without
waste.[1][2][3][4][5] In more mathematical or scientific terms, it is a measure of the
extent to which input is well used for an intended task or function (output). It often
specifically comprises the capability of a specific application of effort to produce a
specific outcome with a minimum amount or quantity of waste, expense, or
unnecessary effort. Efficiency refers to very different inputs and outputs in
different fields and industries.
For example, one may measure how directly two objects are communicating:
downloading music directly from a computer to a mobile device is more efficient
than using a mobile device's microphone to record music sounds that come from a
computer's speakers.
Efficiency is often measured as the ratio of useful output to total input, which can
be expressed with the mathematical formula r=P/C, where P is the amount of
useful output ("product") produced per the amount C ("cost") of resources
consumed. This may correspond to a percentage if products and consumables are
quantified in compatible units, and if consumables are transformed into products
via a conservative process. For example, in the analysis of the energy conversion
efficiency of heat engines in thermodynamics, the product P may be the amount of
useful work output, while the consumable C is the amount of high-temperature
heat input. Due to the conservation of energy, P can never be greater than C, and
so the efficiency r is never greater than 100% (and in fact must be even less at
finite temperatures).
Common-mode noise currents often follow a large loop area which then radiate to
the environment, adding to the system's EM emissions. This type of noise can also
lead to spurious conducted signals within a system which can cause
communications errors and malfunctions due to signal disturbances. Sensitive
measurement devices can malfunction or misinterpret the noise as data, resulting in
erroneous data. AC line transients, such as line surges due to lightning strikes or
power switching from motor controls, circuit breakers, or relays, can cause both
differential and common-mode disturbances on the AC mains that propagate
through the power supply to circuit or system electronic components or is coupled
across conductors, resulting in malfunctions or damage to the electronic circuits
and systems.
Normal-Mode Noise
Differential-mode noise, also known as normal-mode noise, results from AC
voltage disturbances across signal or power lines or current through them. As
might be expected when reviewing a wiring diagram or schematic diagram, the
noise follows the signal and power paths. Common-mode noise is the AC
disturbance from one or more signals or power lines and an external conduction
path, such as an earth ground or chassis or other conductive material not intended
to conduct power or signal energy. The source of noise may be from the AC mains,
the power supply, or even the electronic circuitry being powered. The effects of
parasitic impedance are often not obvious, but understanding these effects is
essential minimizing them. The impedances along an electronic design's common
path may stem from components added to help filter noise.
Common-mode output noise is often overlooked and not specified. Most attention
to noise analysis is given to the input electromagnetic-interference (EMI) filter and
output differential noise filtering. Yet, there is significant common-mode output
noise present due to conductive and radiated coupling to the output. In most
applications, this is not an issue since the output is grounded either internally at the
power supply or at the end application or significant capacitance to ground is
added on the output to earth.
NoiseSources
A typical power supply consists of an AC/DC rectification stage followed by a
high-frequency DC/DC stage and control circuitry to regulate the output voltage.
Noise from the power supply mainly originates from the switching power
semiconductors. Switch-mode power supplies are much more efficient, smaller,
and more economical than the linear power supplies they have largely displaced
during the past few decades. Power-supply designers have made improvements in
reducing noise generated in the power supply from leaking to connected or nearby
equipment. However, limiting power-supply noise is still a challenge and common-
mode noise is often overlooked, partially due to the lack of specifications defining
a requirement for output common mode noise.
The nature of switching power supplies is that
these circuits employ high differential voltage
(dv/dt) and differential current (di/dt) functions
to achieve the high efficiency, reduced size, and These are the noise sources and
cost. With parasitic capacitance as part of a noise paths possible in a typical
power supply, due to the nature of circuit switch-mode power supply.
materials and EM behavior, most power-supply
designs contain a natural high-harmonic noise
source heavily filtered within the supply, although not perfectly contained within
the boundaries of the power supply.
By using a simplified block diagram for a switch-mode power supply, only a few
parasitic coupling paths are shown. Depending on the power conversion topology
used, the dv/dt and di/dt functions in the power supply can vary greatly. Although
filtering reduces the amount of noise present at the output terminals, the amount of
noise that conducts through the output cables depends on the load and its
impedances, both from differential and common-mode perspectives.
For a better understanding of the noise source impedance, and what will be needed
to filter it, it can be helpful to measure a design's open-circuit voltage and short-
circuit current. The open-circuit voltage can be measured from the output return
lead (or from the other polarity) and the system or earth ground point. This can be
done with an oscilloscope, provided that the test lead is kept suitable short when
measuring at supply frequencies of a few Megahertz or higher. This measurement
can also be performed with a spectrum analyzer, which can provide more details
on the frequencies of the noise. This can be helpful when selecting the proper
component values required if additional filtering is needed. It should be
remembered that the input port of a spectrum analyzer is usually 50{0} (sometimes
75{0}) which could result in loading down the noise source. Monitoring also with
an oscilloscope can reveal how much the noise source is being loaded.
Curbing Common-Mode Noise
Once the noise frequency spectrum, source impedance, and noise frequencies are
known, it is possible to make informed decisions how ways to mitigate the noise:
Ferrite material, clamp on cores, inductors, and toroids are quite effective at
reducing offending currents if the proper material is selected. A number of
manufacturers provide ferrite materials, some with excellent characterization
data showing impedance verse frequency.
For high-frequency noise from 10 to 20MHz and higher, selecting a material
with a high resistivity characteristics is often more effective than using
materials with reactive characteristics. The resistive element helps dampen
and dissipate the energy, while reactive parts will present impedance, but
may contribute to resonances depending on the capacitance of the network.
However, at lower frequencies, having a higher reactive element may be
better since the energy involved could be significant and just providing a
high impedance will reduce the current flow and the common-mode noise.
Capacitors should be selected with knowledge of their tolerance variations
over temperature and applied voltage. Such variations can be significant and
result in degraded power-supply performance in terms of noise. Capacitor
values should be selected to coincide with the minimum impedance at the
frequency of interest.
Checks should be made for resonances when adding capacitances and
inductances to noise filters. A resonant frequency can be calculated and,
depending on the noise level and frequency, it may be useful to measure the
voltage if applying the filter components results in worse performance.
Capacitive coupling between noisy signals should be minimized, by
shielding noisy cables and using separate power and signal lines.
The differential voltage, dv/dt, should be minimized, which can be done by
grounding the output return to chassis/earth ground near the power supply if
possible, and shunting noise to the power supply chassis near the power
output.
Noise should be decoupled from source to load by adding a common-mode
inductor or ferrite cores on the power cables. Ferrite material with a high
impedance should be selected for high-frequency noise.
Common-mode noise is present in many power supplies and can cause interference
and other performance problems. Power supplies should be selected that provide
acceptable common-mode noise performance and any noise source should be
shielded from sensitive circuitry to minimize issues
CONCLUSION
In this paper, a novel FMBL PFC rectifier is proposed, in which the high efficiency
over a wide input range can be achieved. In the proposed rectifier, a BTBBL PFC
rectifier is adopted at high line voltages and a TLBL PFC rectifier is formed to
achieve high efficiency at low line voltages. Compared with the traditional
bridgeless boost PFC rectifier, an extra low-voltage bidirectional switch (usually
composed of two switches) is added, therefore the increased cost is low. At both
high and low line conditions, low CM noise can be achieved due to the direct
connection between the input power grid and the output electrolytic capacitor
during half line cycle. The detailed principle analysis about the proposed FMBL
PFC rectifier is presented. Finally, an experimental prototype is built to verify the
feasibility and the effectiveness of the proposed topology.
REFERENCES
[2] S. Abdel-Rahman, F. Stückler, and K. Siu, “PFC boost converter design guide”
[Online]. Available:http://www.infineon.com/dgdl/Infineon-
ApplicationNote_PFCCCMBoo stConverterDesignGuide-AN-v02_00-
EN.pdf?fileId=5546d4624a56eed8014a62c 75a923b05
[5] J. W. Lim and B. H. Kwon, “A power-factor controller for single phase PWM
rectifiers,” IEEE Trans. Ind. Electron., vol. 46, no. 5, pp. 1035– 1037, Oct. 1999.
[9] Y. Jang and M. M. Jovanovic, “A bridgeless PFC boost rectifier with optimized
magnetic utilization,” IEEE Trans. Power Electron., vol. 24, no. 1, pp. 85-93, Jan.
2009.
[10] J. Zhang, B. Su, and Z. Lu, “Single inductor three-level bridgeless boost
power factor correction rectifier with nature voltage clamp,” IET Power Electron.,
vol. 5, no. 3, pp. 358-365, Mar. 2012.
[11] Y. -S. Kim, W.-Y. Sung, and B.-K. Lee, “Comparative performance analysis
of high density and efficiency PFC topologies,” IEEE Trans.