The inclusion of non-linear loads in the AC network, for example, lamps with gas-discharge lamps, controlled electric motors, switching power supplies leads to the fact that the current consumed by these devices has a pulsed character with a large percentage of high harmonics. Because of this, EMC problems can arise when operating various equipment. It also leads to a decrease in the active power of the network.

In order to prevent such negative effects on power supply networks in Europe and the USA, the standard IEC IEC 1000-3-2, which determines the norms for harmonic components of current consumption and power factor for power supply systems with a power of more than 50 W and all types of lighting equipment. Since the 80s of the last century to the present day, these standards have been consistently tightened, which has caused the need for special measures and pushed equipment developers to develop various options for schemes that provide an increase in the power factor.

Starting in the 80s of the last century, microcircuits began to be actively developed and used in the aforementioned countries, on the basis of which simple power factor correctors for rectifiers and electronic ballasts can be easily created.

In the Soviet Union, and later in the Russian Federation, no such restrictions were imposed on electricity consumers. For this reason, power factor improvement has received little attention in the technical literature. In recent years, the situation has changed somewhat, largely due to the availability of imported electronic components, the use of which makes it possible to create active corrector circuits that are reliable in operation and inexpensive in cost.

Distortion power and generalized power factor

The negative impact on the supply network is determined by two components: distortion of the supply network current shape and reactive power consumption. The degree of consumer influence on the supply network depends on its power.

The distortion of the current shape is due to the fact that the current at the input of the valve converter is non-sinusoidal (Figure 1). Non-sinusoidal currents create non-sinusoidal voltage drops on the internal resistance of the supply network, causing distortion of the supply voltage shape. Non-sinusoidal mains voltages are decomposed in a Fourier series into odd sinusoidal components of higher harmonics. The first is the main one (the one that should ideally be), the third, the fifth, etc. Higher harmonics have an extremely negative effect on many consumers, forcing them to use special (often very expensive) measures to neutralize them.

Rice. one.

The consumption of reactive power leads to a lag of the current from the voltage by an angle (Figure 2). Reactive power is consumed by rectifiers using single-operation thyristors that delay the moment of switching on relative to the natural switching point, which causes the current to lag behind the voltage. But even more reactive power is consumed by asynchronous electric motors, which have a predominantly inductive nature of the load. This entails colossal losses of useful power, for which, moreover, no one wants to pay - household electricity meters count only active power.

Rice. 2.

To describe the effect of the converter on the supply network, the concept of total power is introduced:

, where:

- effective value of primary stress,

- effective value of the primary current,

, - effective values ​​of voltage and current of the primary harmonic,

Effective values ​​of voltages and currents of higher harmonics.

If the primary voltage is sinusoidal - , then:

,

,

ϕ 1 is the phase angle between the sinusoidal voltage and the first harmonic of the current.

N is the power of distortion caused by the flow of higher harmonic currents in the network. The average power over the period due to these harmonics is zero, since harmonic and primary voltage frequencies do not match.

Higher harmonics of currents cause interference in sensitive equipment and additional eddy current losses in mains transformers.

For valve converters, the concept of power factor χ is introduced, which characterizes the effect of reactive power and distortion power:

,

is the distortion factor of the primary current.

Thus, it is obvious that the power factor depends on the angle of the current lag relative to the voltage and the magnitude of the higher harmonics of the current.

Power Factor Improvement Techniques

There are several ways to reduce the negative influence of the converter on the supply network. Here are some of them:

Using multistage phase control (Figure 3).

Rice. 3.

The use of a rectifier with taps from a transformer leads to an increase in the number of pulsations per period. The more taps from the transformer, the greater the number of ripples per period, the closer the input current waveform is to sinusoidal. A significant disadvantage of this method is the high cost and dimensions of a transformer with a sufficient number of taps (to achieve the effect, there must be more of them than in the figure). Making a winding element of such complexity is a very difficult task that does not lend itself well to automation - hence the price. And if the developed secondary power supply source is small-scale, then this method is unambiguously unacceptable.

Rice. 4.

Increasing the phase of the rectifier. The method leads to an increase in the number of pulsations per period. The disadvantage of this method is a very complex transformer design, an expensive and bulky rectifier. In addition, not all consumers have a three-phase network.

Usage power factor correctors (PFC)... There are electronic and non-electronic PFCs. Electromagnetic reactive power compensators are widely used as non-electronic KKM - synchronous motors that generate reactive power into the network. Obviously, for obvious reasons, such systems are unsuitable for a domestic consumer. Electronic KKM - a system of circuitry solutions designed to increase the power factor - is, perhaps, the most optimal solution for household consumption.

Principle of KKM operation

The main task of the KKM is to reduce to zero the lag of the consumed current from the voltage in the network while maintaining the sinusoidal shape of the current. To do this, it is necessary to take current from the network not in short intervals, but throughout the entire period of operation. The power drawn from the source must remain constant even if the mains voltage changes. This means that when the mains voltage decreases, the load current must be increased, and vice versa. For these purposes, converters with an inductive storage and transfer of energy on a return run are suitable.

Correction methods can be roughly divided into low-frequency and high-frequency. If the frequency of the corrector is much higher than the mains frequency, it is a high-frequency corrector, otherwise it is a low-frequency one.

Let's consider the principle of operation of a typical power corrector (Figure 5). On the positive half-wave, at the moment the mains voltage crosses zero, the transistor VT1 opens, the current flows through the L1-VD3-VD8 circuit. After turning off the transistor VT1, the choke begins to give up the energy stored in it, through the diodes VD1 and VD6 into the filtering capacitor and load. With a negative half-wave, the process is similar, only other pairs of diodes work. As a result of using such a corrector, the current consumption has a pseudo-sinusoidal character, and the power factor reaches 0.96 ... 0.98. The disadvantage of this scheme is the large size due to the use of a low-frequency choke.

Rice. 5.

Increasing the frequency of the KKM allows you to reduce the size of the filter (Figure 6). When the power switch VT1 is open, the current in the choke L1 increases linearly - while the VD5 diode is closed, and the capacitor C1 is discharged to the load.

Rice. 6.

Then the transistor turns off, the voltage across the choke L1 turns on the diode VD5 and the choke gives the stored energy to the capacitor, while simultaneously supplying the load (Figure 7). In the simplest case, the circuit operates with a constant duty cycle. There are ways to increase the efficiency of the correction by dynamically changing the duty cycle (i.e. by matching the cycle to the voltage envelope of the mains rectifier).

Rice. 7. Forms of voltages and currents of high-frequency PFC: a) with variable switching frequency, b) with constant switching frequency

Microcircuits for building high-performance correctors from STMicroelectronics

Considering the capabilities of the modern electronics industry, high frequency PFCs are the best choice. Integral performance of the entire power corrector or its control part has become, in fact, the standard. Currently, there is a greater variety of control microcircuits for constructing PFC circuits produced by various manufacturers. Among all this variety, it is worth paying attention to the L6561 / 2/3 microcircuits produced by STMicroelectronics (www.st.com).

L6561, L6562 and L6563- a series of microcircuits specially designed by STMicroelectronics engineers to build highly efficient power factor correctors (Table 1).

Table 1. Power factor corrector microcircuits

Name	Voltage power supply, V	Current inclusions, μA	Consumption current in active mode, mA	Standby current consumption, mA	Output bias current, μA	Power switch current rise time, ns	Decay time of the power switch current, ns
L6561	11…18	50	4	2,6	-1	40	40
L6562	10,3…22	40	3,5	2,5	-1	40	30
L6563	10,3…22	50	3,8	3	-1	40	30

Based on the L6561 / 2/3, an inexpensive but effective corrector can be built (Figure 8). Due to the built-in predictive control system, the developers managed to achieve high accuracy of output voltage regulation (1.5%), controlled by the built-in mismatch amplifier.

Rice. eight.

The possibility of interaction with a DC / DC converter connected to the corrector is provided. This interaction consists in turning off the converter by the microcircuit (if it supports such a possibility) in the event of unfavorable external conditions (overheating, overvoltage). On the other hand, the converter can also initiate the switching on and off of the microcircuit. The built-in driver allows you to drive powerful MOSFETs or IGBTs. According to the manufacturer, based on the LP6561 / 2/3, a power supply with a power of up to 300 W can be realized.

Unlike analogs from other manufacturers, LP6561 / 2/3 are equipped with special circuits that reduce the conductivity of the input current distortion that occurs when the input voltage reaches zero. The main cause of this interference is the "dead zone" that occurs during the operation of a diode bridge, when all four diodes are closed. A pair of diodes operating on a positive half-wave turn out to be closed due to a change in the polarity of the supply voltage, and the other pair has not yet managed to open due to its own barrier capacitance. This effect is enhanced in the presence of a filter capacitor located behind the diode bridge, which, when the polarity of the supply is reversed, retains some residual voltage, which does not allow the diodes to open in time. Thus, it is obvious that the current does not flow at these moments, its shape is distorted. The use of new PFC controllers can significantly reduce the "dead zone" time, thereby reducing distortion.

In some cases it would be very convenient to control the output voltage supplied to the DC / DC converter using a PFC. L6561 / 2/3 allow this control, called "tracking boost control". To do this, simply install a resistor between the TBO pin and GND.

It should be noted that all three microcircuits are pin-compatible with each other. This can greatly simplify the design of the printed circuit board of the device.

So, the following features of the L6561 / 2/3 microcircuits can be distinguished:

configurable overvoltage protection;

ultra-low starting current (less than 50 μA);

low quiescent current (less than 3 mA);

wide range of input voltages;

built-in filter that increases the sensitivity;

the ability to disconnect from the load;

the ability to control the output voltage;

the ability to interact directly with the converter.

Conclusion

Currently, there are strict requirements for compliance with safety measures and economy of modern electronic devices. In particular, when designing modern switching power supplies, it is necessary to take into account the officially adopted standards. IEC 1000-3-2 is the standard for any high power switching power supply because it defines harmonic current and power factor limits for power systems over 50 W and all types of lighting equipment. The presence of a power factor corrector helps to meet the requirements of this standard, i.e. its presence in a powerful power supply is a simple necessity. L6561 / 2/3 is the optimal choice for constructing an effective and at the same time inexpensive power factor corrector.

Obtaining technical information, ordering samples, delivery - e-mail:

About ST Microelectronics

I.P. Sidorov Yu.A.

Attention. High voltage, life-threatening.

Attention when implementing the above diagram of the power factor corrector, you must have experience working with life-threatening voltages and be extremely careful.

the circuit operates with a life-threatening voltage of 400 volts

If mistakes are made during assembly, the voltage in the circuit can reach 1000 or more volts.

At the time of switching on and checking the assembled circuit, you must use protective goggles.

The schematic electrical diagram (corrected) of the power factor corrector is shown in Fig. one.

rice. 1.Power factor corrector - diagram. open in large size
The previous diagram is open in large size

On the diagram, the functional units are marked with colored blocks:

• Brown - noise filter;
• Blue - soft-start module;
• Red - internal power supply;
• Green - power factor corrector;
• Blue - module for monitoring operating parameters;
• Yellow - forced cooling fan activation module.

On the revised version of the diagram it is noted (available in large size):
red rectangle - new elements of the circuit;
green oval - new points of connection of capacitors C3 and C4.

An interference filter protects the supply network from interference generated by switching key transistors. The filter also protects the circuit from mains noise and mains voltage surges.

The soft start module limits the current consumption from the supply network at the time of the initial charging of the output electrolytic capacitors. This module generates an inverted KKM_SUCCESS signal. When a signal appears (since the signal is inverted - the moment at which the voltage drops below 1V), you can turn on the load connected to the output of the power factor corrector. If this signal is ignored, some elements of the circuit may fail.

The internal power supply generates a constant voltage of 15V (tolerances +/- 2V are allowed). This voltage is used to power the internal PFC circuits.

The power factor corrector is the main part of the circuit. The KKM is made on the ir1155s controller, the operating frequency in this circuit is 160 kHz (deviations are +/- 5 kHz). To amplify the control currents of the switching transistors, a single-channel tc4420 driver is used, the driver provides a current of control signals up to 6A.

The operating parameters control module controls the level of the reduced supply voltage; operating temperature of the KKM, the moment of reaching the rated voltage at the output of the KKM

The forced cooling fan enable module turns on the fans when the corresponding signal appears.

Tables of denominations of elements of the KKM scheme.

When assembling the power factor corrector, use only original accessories. In the case of using non-original components (counterfeit, counterfeit, etc.), the KKM will not work or will not work correctly, etc.

Stage 1. all elements must be installed except for:
R3 - varistor;
L3 - KKM choke
C25.2-C25.4 - output electrolytic capacitors, install only one.

The mounting plate is designed taking into account the installation in the case from the radiator profile. In this case, the walls of the case for elements D1, D9, Q5, Q6 act as a heat sink, and heat removal from the choke L3 will be difficult. The temperature of the choke, in this case, serves as an indicator of the heating of the entire device and therefore the R40 thermistor is installed under the choke.

In the case of using a case of a structure in which the role of a heat sink for elements D1, D9, Q5, Q6 will be a radiator - thermistor R40 must be installed on the surface of the radiator. It is necessary to provide electrical insulation for the radiator housing and thermistor.

Then the circuit board must be cleaned of residual flux and other contaminants.

The circuit board after this assembly step will look like this

rice. 2. The upper part of the KKM circuit board.

On this circuit board, the thermistor and the lead-out wire are shrink-wrapped. Since the thermistor will be mechanically attached to the radiator, it is placed in additional heat-shrinkable insulation to increase the strength of the electrical insulation.

rice. 3. The lower part of the KKM mounting plate.

A 12V fan with a current of no more than 0.2A must be connected to the KKM board.

ATTENTION!!! The device operates on a lethal voltage of 400 volts.

The KKM board must be connected to a regulated source of alternating voltage 220V 50 Hz with a current limitation of 0.05 A.

After power is applied, the D8 LED should be on, the voltage at the D5 Zener diode should be within 14-17 volts. In the absence of voltage, it is necessary to check the voltage across the capacitor C12, it should be about 310 volts. If voltage is present, this means the inoperability of the standby power source. A common cause of inoperability is incorrect assembly of the T1 pulse transformer.

The voltage at pin 4 of U1 (ir1155s) should be about 3.62 V, the voltage at pin 6 is about 3.75 V.

Using an oscilloscope, it is necessary to check the operation of the PFC module. To do this, the oscilloscope probe must be connected to pin 6 or 7 of the U3 chip (tc4420). The pulses on the pin should match the following image.

rice. 4. Graph of signals at the output of the tc4420 driver microcircuit.

The pulse frequency should be 160kHz (+/- 5kHz). The pulse frequency is set by the capacitor C10. An increase in capacitance leads to a decrease in frequency.

The amplitude of the signals at the SG pins of the power transistors will be slightly lower than at the pin of their driver (Fig. 5).

rice. 5. Graph of signals at the outputs of SG power transistors.

In this case, the graph of the signal across the resistors Rg (R17, R18) will be as follows (Fig. 6).

rice. 6. Graph of the signal across the resistors Rg (R17, R18).

Further, while monitoring the signals at the driver output, it is necessary to smoothly decrease the voltage. With an input voltage of 150-155 volts, the generation of pulses should stop. After stopping the generation of pulses, the input voltage must be gradually increased, with an input voltage of 160-165 volts, the generation of pulses must resume.

Continuing to smoothly increase the voltage, when 270-280 volts (AC) are reached, the relays should work (you can determine by their characteristic sound). The voltage of the KKM_SUCCESS signal should be no more than 1 volt. Then the voltage must be gradually reduced, when the voltage drops to 250-260 volts, the relays must turn off, the signal at the KKM_SUCCESS output must be more than 5 volts.

Using a hot air gun, it is necessary to heat the thermistor, when the temperature reaches 45-50 C °, the fan should turn on, when the temperature reaches 75-85 C °, the generation of pulses should stop. While the thermistor is cooling down, pulse generation should be resumed in sequence and the fan should be turned off.

Disconnect the power supply.

ATTENTION!!! after disconnecting the power supply, a life-threatening voltage will remain in the circuit for some time (several minutes).

Stage 3. It is necessary to install the remaining elements of the circuit: R3, L3, C25.2-C25.4 and a heat sink for elements D1, D9, Q5, Q6. It is necessary to install a thermistor on the heat sink, ensuring a low thermal resistance between them. It is also necessary to ensure a low thermal resistance between D1, D9, Q5, Q6 and the heatsink. In case of difficult heat transfer to the radiator, these elements will fail.

The quality of the radiator installation, from the point of view of heat dissipation, is conveniently controlled with a thermal imager.

The heat sink must be connected to the Earth bus (there are mounting holes on the circuit board next to the Y capacitors).

It is imperative to check the electrical insulation between the Earth and N or L busbars (the N-L busbars are used to supply power). The breakdown voltage of electrical insulation must be at least 1000 Volts. Insulation breakdown voltage over 1000 Volts should not be checked. This procedure can be performed using a special device - an electrical insulation tester.

ATTENTION!!!. In the event of a violation of the tested electrical insulation, when checking, some elements of the circuit may fail.

An example of an assembly of a power factor corrector is shown in the following images.

Stage 4. Connect the KKM to the mains supply, limiting the consumed current to 10A. After switching on, the voltage at the output of the KKM should be about 385-400 V. The sound of switching on the relay should also be heard. Connect a resistive load of 300 Ohm to the KKM output. The voltage at the output of the PFC should remain within the same range. PF must be at least 0.7.

Connect the KKM to the mains without current limiter. By increasing the load to 2000 watts, the PF should also increase to a value of at least 0.95. The PF graph versus load is shown in Fig. 7.

rice. 7. Graph of dependence of PF on load.

If the PF value does not increase to 0.95 with increasing load, this indicates the incorrect operation of the PFC. Probable reasons for such an incorrect one may be: a resistive current sensor, a choke, errors in the manufacture of a circuit board, counterfeit elements D9, Q5, Q6, C18.1, C18.2, an internal power supply of insufficient power.

Oscillograms of the consumed currents and output ripple.

In the course of stress tests, the efficiency was determined (Fig. 8). If we take into account the error of the measuring instruments, it is likely that the actual efficiency will be 1-2% lower. The efficiency was measured when the PFC was connected to the mains using two additional common-mode filters.

rice. 8. Efficiency of the power factor corrector.

The data for both graphs was obtained at supply voltages of 200 and 240 volts.

Stage 5. After all checks, the discharge resistor R23 can be removed. The assembly and inspection of the POS printer at this stage can be considered complete.

Questions and suggestions write to the e-mail address marked with KKM or PFC.

Cart contents

1. WHY IS IT NECESSARY?

Let's say right away that, contrary to superficial statements, the presence of a power factor corrector in itself does not improve the formal characteristics of the device in which it is applied. On the contrary, the introduction of KKM as a rather complex device so far leads to a noticeable rise in price and complication of the product as a whole (of course, as technology develops, the price will decrease). Nevertheless, even now the introduction of PFC in power amplifiers provides a number of very important advantages that more than compensate for this complication.

The first and most important advantage is the fact that when using amplifiers with PFC with the same wiring, without violating any standards, at least three to four times more powerful amplifiers can be used. By the way, there is no violation of physical (and legal) laws here, and why this happens - we will tell further.

The second, no less important, but rarely mentioned advantage is that it is much easier to provide high energy consumption of a power supply unit with a PFC than a traditional one. Energy capacity is a measure of the power supply's ability to deliver power to the load for a certain period of time without "squandering" the network and not greatly reducing the output voltage. From a practical point of view, the lack of energy intensity leads to the fact that the output power of the amplifier at low frequencies (where it is most needed!) Is much less, and the distortion of other signals in the presence of a low frequency is much higher than when measuring at a frequency of 1 kHz, the results which (sometimes just the desired ones) are advertised in the description. Simply put, with a lack of energy capacity, the amplifier begins to "choke" and distort the signal during loud low-frequency sounds, for example, when a kick drum is hit. Unfortunately, for amplifiers with traditional power supplies, this unwanted effect is the rule rather than the exception. Therefore, if it was necessary to ensure good quality, it was necessary to choose an amplifier with a large power reserve.

The third advantage is that the power supply unit with KKM, according to the principle of operation, stabilizes the output voltage. Therefore, the output power of the amplifier ceases to depend rigidly on the mains voltage - even with a "sagging" mains, full power is given.

Another, completely unexpected advantage is that the network background (the same one), when using only amplifiers with PFC, is, as a rule, 10 decibels lower.

2. WHAT IS IT AND HOW DOES IT WORK?

Despite the variety of really existing devices, the principle of PFC operation can be considered on the following simple example (see Fig. 1).

A power factor corrector is nothing more than an almost ordinary switching regulator, powered by a rectified but unsmoothed mains voltage and stabilizing the voltage at the output storage capacitor C2. The basic principle of its operation is quite simple and is as follows. First, the key S1 is closed for a short time, and the current in the inductor L1, in full accordance with the physics textbook, begins to build up. After some time, the switch opens, and the energy stored in the coil passes through the diode to the output storage capacitor. This cycle is continuously repeated, as a result of which portions of energy are supplied to the storage capacitor, the value of which depends on the input voltage, the magnitude of the inductance and the time of the closed state of the switch. In order for the dimensions of the coil and the losses in it to be small, the value of the inductance is chosen small, and, accordingly, the repetition rate of such cycles is made high enough - tens and hundreds of thousands of times per second. It should be noted that at an excessively high frequency, the switching losses of the transistor used as a switch become very

essential. The most important thing here is that with proper control, the input of such a converter from the mains side will look like some resistance (the current at each time is proportional to the voltage), and at the same time, a certain constant voltage will be maintained on the output capacitor, which is practically independent of the load and mains voltage (!). In this case, there will be no phase shift (cos j 1) * or violation of proportionality between the voltage in the network and the current taken from it.

The high voltage across the storage capacitor makes it easier to ensure the power capacity of the power supply, since the energy content in the capacitor is proportional to the square of the voltage, while the dimensions and weight of capacitors of equal capacity are approximately proportional to the voltage. As a result, a capacitor with a capacity of 2200 μF at a voltage of 430 V contains more than 200 J of energy, and the same capacitor at a voltage of 60 V contains only about 4 J, or 50 (!) Times less. The volume of these capacitors differs only six to eight times. Therefore, to achieve the same energy capacity at low voltages, capacitors of enormous capacity are required - more than 100,000 microfarads in this case. At the same time, for the perfect operation of an exemplary high-quality amplifier, the power consumption of its power supply should be at least 0.5 ... 0.8 J per W of total output power; for concert amplifiers (except for subwoofer), 0.2 ... 0.4 J per Tue That is, the 2x1000 W amplifier must have an energy capacity of the power supply unit of at least 400 J, or 200000 uF at 60V, and preferably three times more.

In practice, the energy consumption of traditional power supplies in the vast majority of amplifiers is much lower, and the reason for this is not only the banal savings of manufacturers on transformers and capacitors. No less important is the fact that a rectifier with high-capacity capacitors is a circuit that loads the network only in short periods of time (during the "tops" of sinusoids), but with large currents (see Fig. 2), where, by the way, it can be seen that the shape of the mains voltage is severely distorted by such rectifiers). Moreover, the better the transformer and the higher the capacity, the more pronounced this phenomenon. It is possible to connect such a power supply unit to the network only if there are soft starters, otherwise the fuses will burn out. Further, any, even a small jump in the mains voltage upward causes a sharp increase in the magnitude of these current pulses, which leads to the failure of the rectifiers. That is why the capacitance of the capacitors (and, accordingly, the power consumption of power supplies) in most amplifiers with a traditional power supply is chosen much less than is necessary to ensure a proper power reserve at low frequencies.

Taking a look at fig. 3, two more circumstances can be noticed.

The first is that the peak current consumption is several times higher than the average. But the useful power is determined by the average current, while the voltage drop across the wires is the peak one. And it turns out to be much more than average.

The second circumstance is that the current consumed by short pulses has a high rate of change, and, accordingly, creates more noise.

Another problem arises in three-phase networks. Due to the fact that the phases of voltages in a three-phase network are shifted by a time much longer than the duration of these current pulses, they cease to be compensated in the neutral wire. Moreover, the current in the neutral wire will be approximately equal to the sum of the phase currents, while in a normal situation the current through it does not at all

should flow, and the neutral wire is usually made thinner than the phase ones. Considering that the current through it becomes more than through the phase ones, and also the fact that the installation of fuses in the neutral wire is prohibited, it is easy to guess that it is not far from here to a fire. Therefore, the value of the harmonics of the current consumption is limited by rather stringent international standards. Traditional power supplies with power above 150 ... 200 W are fundamentally unable to meet these standards. This will lead to the fact that at high capacities, traditional power supplies are simply "outlawed".

All these problems can be avoided if from the mains side the power supply looks like a purely active resistance, like an iron or an incandescent light bulb.

This is exactly how a power supply unit with a power factor corrector works. The problems associated with the instability of the network disappear, and it also becomes possible to provide the necessary energy consumption of the power supply.

It becomes quite obvious that the use of a power factor corrector is not only mandatory (from the point of view of the law), but also absolutely necessary for the "honest" operation of professional high-quality amplifiers.

* Small addition: cos j and power factor are often confused, although they are not the same thing. Cos j is a measure of how much of the current flowing in the wires actually goes into the load (and does useful work), while both voltage and current are assumed to be strictly sinusoidal. If there is no phase shift, cos j = 1. If the phase shift reaches 90 degrees, regardless of the sign, cos j becomes zero - the useful power is simply not transferred to the load.

The power factor is the same as cos j only in the case of purely sinusoidal currents and voltages. If the current or voltage is non-sinusoidal, only the power factor remains applicable, which shows how much of the current that has passed through the wires and heats them is usefully gone into the load. The power factor of a conventional rectifier does not exceed 0.25 ... 0.3, while for a good PFC it is at least 0.92 ... 0.95, i.e. 3-4 times more (that's where the three-fourfold difference comes from!).

V. Dyakonov, A. Remnev, V. Smerdov

Recently, on the market of household and office radio electronic equipment (CEA), more and more equipment appears, the power sources of which include new units - power correctors (KM). The article deals with the use of CM, the principle of their operation, diagnostics and repair.

Most modern power supplies for electronic equipment are switching secondary power supplies with a transformerless bridge rectifier and a capacitive filter. Along with the advantages (high efficiency, good weight and dimensions), they have a relatively low power factor (0.5 ... 0.7) and an increased level of harmonics of the current consumed from the mains (> 30%). The shape of the current consumed by such sources is shown in Fig. 1 with solid lines.

The non-sinusoidal form of the current leads to the occurrence of electromagnetic interference, clogging the AC network, and the failure of another electronic equipment.

The above-described power supplies, being single-phase consumers, with a large number of electronic equipment and its irrational connection to a three-phase supply network, can cause phase imbalance. In this case, part of the electronic equipment will work at an increased voltage, and the other at a reduced voltage, which is always undesirable. To eliminate phase imbalance, a neutral wire is usually introduced into a three-phase network, which equalizes the voltage in all phases. However, with a pulsed nature of the consumed current and a large number of its harmonic components, overloading of the neutral wire is possible. This is due to the fact that its cross section is usually 2 ... 2.5 times less than that of phase wires. For safety reasons, do not protect this wire with fuses or circuit breakers. Obviously, under unfavorable conditions, the neutral wire may burn out and, as a consequence, the occurrence of phase imbalance.

In this regard, the requirements for electromagnetic compatibility of secondary pulse sources with the mains are becoming more stringent and the level of higher harmonics of the current consumed from the mains for all single-phase consumers is sharply limited. Currently, the new European standards require an improvement in the form of the consumed current only at consumer powers over 200 W, and in the near future these requirements will be introduced for consumers with a power of up to 50 ... 70 W.

Currently, passive and active correction of the consumed current shape is used.

Passive correction circuits, consisting of inductors and capacitors, provide a power factor that shows the difference in the form of the consumed current from a sinusoid (not worse 0.9 ... 0.95). With constructive simplicity and reliability, passive correction circuits have relatively large dimensions and are sensitive to changes in the frequency of the supply voltage and the magnitude of the load current.

More promising is the use of active CMs, which form a sinusoidal current consumption at the input of a switching power supply, which coincides in phase and frequency with the supply voltage. Such CMs have small dimensions due to work with conversion frequencies of several tens of kilohertz and provide a power factor of 0.95 ... 0.99.

It is possible to form a sinusoidal current at the input of a bridge rectifier of a switching power supply using one of the DC-to-DC converter circuits using the principle of tracking high-frequency pulse width modulation (PWM). In this case, step-up converters are most often used, which have the following advantages:
... the power transistor has a source connection with a common wire, which facilitates the construction of its control circuit;
... the maximum voltage across the transistor is equal to the output voltage;
... the presence of an inductance connected in series with the load provides filtering of high-frequency components.

Consider the principle of operation of an active CM, implemented on a boost converter with a tracking PWM (Fig. 2).

First, consider the operation of the CM circuit without multiplication nodes (PA) and a load voltage sensor (DNV), the role of which is described below. The reference voltage of the sinusoidal form, obtained from the rectified voltage sensor (DVN), is fed to one of the inputs of the control circuit (CS) by a power switch implemented on the MOS transistor VT. The second input of the control system receives a signal proportional to the key current. While the voltage from the DVN is greater than the voltage generated by the current sensor (DT), the transistor is open and energy accumulates in the inductance (Fig. 3 a). The VD diode is closed in this interval (Ti).

When the signals arriving at the control system are equal, the key is closed and the energy accumulated in the inductance is transferred to the load. After the current in the inductance drops to zero during the time tP, the transistor turns on again. The switching frequency of the transistor is many times higher than the frequency of the supply network, which makes it possible to significantly reduce the size of the inductance. In this case, for a half-period of the mains voltage, the envelope of the amplitude values ​​of the inductance current (Fig. 3 b) changes according to a sinusoidal law. The average current value changes in the same way. As a result, the consumed current is sinusoidal and in phase with the supply voltage.

However, the magnitude of the voltage across the load is highly dependent on changes in the input voltage and load current. To stabilize the load voltage, a feedback loop for this voltage is additionally introduced into the control system. The possibility of obtaining a sinusoidal form of the consumed current with simultaneous stabilization of the load voltage is realized using analog multiplication (PA node) of the signals coming from the DVN and from the DNV.
The additional signal obtained in this way in this case becomes the reference voltage for the control system.

The considered principle of CM control is used at load powers up to 300 W. At high powers, it is necessary to form a smoother curve of changes in the consumed current. This can be done when the current in the inductor does not drop to zero (Fig. 3 c and 3d). If in a CM of relatively low power the transistor comes into operation when the inductance current reaches zero, then in powerful CM - at a given value of this current.

Let us consider the work of the CM using the example of a practical circuit shown in Fig. 4. The control circuit is implemented on a specialized microcircuit L6560, the block diagram of which is shown in Fig. 5,

And the purpose of the conclusions is in table. one.

The voltage of the DVN, formed by the resistive divider R1 R2, is fed to the pin. 3 microcircuits L6560. Capacitor C1 at the output of the rectifier acts as a high-frequency filter, and not as a smoothing capacitor, as in traditional circuits. Therefore, its value does not exceed hundreds of nanofarads - microfarad units at load powers of 100 ... 200 W. Additional filtering of RF interference on the pin. 3 is carried out by a capacitor C2.
Resistor R5 acts as a key current sensor, the voltage of which is fed to the pin through the high-frequency filter R4 C4. 4 microcircuits. The power switch is controlled by a signal received from the pin. 7. Taking into account the peculiarities of the KM keys operation (a large dynamic range of the amplitude values ​​of the current), MIS transistors are most often used as them. At high conversion frequencies typical for CM, these transistors have low dynamic losses and are easily controlled directly by microcircuits. To reduce the likelihood of excitation of the circuit, a low-resistance resistor is introduced into the gate circuit of the MIS transistor.

The output voltage feedback signal is removed from the resistive divider R6 R7 and fed to the pin. 1. To reduce the influence of impulse noise arising in the output circuit, between the pin. 1 and 2 of the microcircuit includes an integrating capacitor C3, the capacity of which is hundreds of nanofarads.

When the CM is connected to the network at the first moment, the microcircuit is powered through the resistor R3. As soon as the CM enters the operating mode, a voltage is removed from the additional winding of the inductor L, which, on the one hand, is used as the supply voltage of the microcircuit, and on the other, is a signal for determining the zero inductance current.

At the output of the CM, a filter capacitor C5 is necessarily present, since energy is transmitted to the load in pulses. The capacity of this capacitor, as a rule, is determined at the rate of 1.5 ... 2 μF per 1 W of power in the load.

Recently, leading companies have produced a large number of integrated circuits for control systems of power correctors. Such a number of microcircuits is associated with additional functions that they are capable of performing, although the principle of constructing CM on these microcircuits is practically the same. Additional features include:
... overvoltage protection during transient processes;
... protection against repeated launches;
... protection against damage when starting on a closed load;
... improvement of the harmonic composition at zero crossing of the mains voltage;
... undervoltage blocking;
... protection against accidental surges of input voltage.

The power corrector, as a rule, is not an independent device, but is a part of switching power supplies. To obtain the required levels and polarities of the output voltages, such power supplies contain converters. In this regard, the developers of microcircuits often combine two cascades of control circuits in one case: for the CM itself, and also for the voltage converter.

Table 2 shows the main parameters of control microcircuits of various companies, intended for secondary switching power supplies with power correction.

The main criterion for the operation of the CM is the output voltage level. With an alternating voltage of the supply network of 220 V, the output voltage of the KM is constant and should be 340.360 V. If the voltage is less than 300 V, then this indicates a malfunction. An oscilloscope is needed to further check the CM. With its help, oscillograms are checked in the characteristic nodes of the CM at a nominal load, which can be an equivalent resistor.

Voltage at the gate of the transistor. With a working microcircuit, its output voltage is rectangular pulses of high frequency, much higher than the mains frequency. With a working MIS transistor, the difference in voltage at the output of the microcircuit and the gate of the transistor is practically zero. If the gate of the transistor is broken, a difference in these voltages of several volts appears.

The voltage at the source of the transistor, which is the voltage taken from the current sensor. During normal operation of the CM, the voltage waveform should be similar to the key current waveform shown in Fig. 3. The difference will indicate a possible malfunction of the MIS transistor. Diagnostics of their malfunctions is detailed in.

Voltage on DVN. The shape of this voltage is a rectified sinusoid. With a normally operating rectifier, the resistive divider may malfunction.

To test the microcircuit itself, an additional constant voltage source is required with voltage regulation from 3 to 15 V. This voltage is supplied to the inputs of the microcircuit power circuit when the KM is disconnected from the network. When the voltage of the regulated source changes, the output voltage of the microcircuit is monitored. As long as the supply voltage is less than 12..13 V, the output voltage is zero. With a higher voltage at the output of the microcircuit, an output signal appears with a level that monitors the supply voltage. When the supply voltage drops below 7 V, this output signal drops abruptly to zero. In the absence of such a pattern, it is highly likely that the microcircuit is faulty.

Literature
1. Bachurin V.V., Dyakonov V.P., Remnev A.M., Smerdov V.Yu. Circuitry of devices based on powerful field-effect transistors. Directory. M .: Radio and communication, 1994.
2. V. Dyakonov, A. Remnev, V. Smerdov. Features of the repair of radio electronic equipment units on MIS transistors. Repair & Service, 1999, No. 11, p. 57-60.
[email protected]