|
Arabic
Bengali
Chinese
English
French
German
Hebrew
Hindi
Italian
Japanese
Korean
Malay
Polish
Portuguese
Spanish
Turkish
Ukrainian
Vietnamese|
ENCYCLOPEDIA OF RADIO ELECTRONICS AND ELECTRICAL ENGINEERING Switching step down stabilizers. Reference data
Encyclopedia of radio electronics and electrical engineering / Reference materials The article brought to the attention of readers describes two pulse step-down stabilizers: on discrete elements and on a specialized microcircuit. The first device was designed to supply automotive equipment with a voltage of 12 V to the 24-volt on-board network of trucks and buses. The second device is the basis for the laboratory power supply. Switching voltage regulators (step-down, step-up and inverting) occupy a special place in the history of the development of power electronics. Not so long ago, every power supply with an output power of more than 50 watts included a step-down switching regulator. Today, the scope of such devices has been reduced due to the reduction in the cost of power supplies with a transformerless input. Nevertheless, the use of switching step-down stabilizers in some cases turns out to be more economical than any other DC-DC converters. The functional diagram of a step-down switching regulator is shown in fig. 1, and the timing diagrams explaining its operation in the continuous current mode of the inductor L are shown in fig. 2. During ton, the electronic switch S is closed and the current flows through the circuit: the positive terminal of the capacitor Cin, the resistive current sensor Rdt, the storage inductor L, the capacitor Cout, the load, the negative terminal of the capacitor Cin. At this stage, the inductor current lL is equal to the current of the electronic switch S and increases almost linearly from lLmin to lLmax.
According to a mismatch signal from the comparison node or an overload signal from a current sensor, or a combination of them, the generator switches the electronic switch S to an open state. Since the current through the inductor L cannot change instantly, then under the action of the self-induction EMF, the diode VD opens and the current lL flows through the circuit: the cathode of the diode VD, the inductor L, the capacitor CBO, the load, the anode of the diode VD. During tlKl, when the electronic switch S is open, the inductor current lL coincides with the diode current VD and decreases linearly from lLmax to lL min. During the Period T, the capacitor Cout receives and gives off the charge increment ΔQcv. corresponding to the shaded area on the time diagram of the current lL [1]. This increment determines the amplitude of the ripple voltage ΔUCout on the capacitor Cout and on the load. When the electronic switch is closed, the diode closes. This process is accompanied by a sharp increase in the current of the switch to the value of Ismax due to the fact that the resistance of the circuit - the current sensor, the closed switch, the recovery diode - is very small. To reduce dynamic losses, diodes with a short reverse recovery time should be used. In addition, the buck regulator diodes must be able to handle large reverse current. With the restoration of the closing properties of the diode, the next conversion period begins. If the switching buck regulator operates at a low load current, it can switch to the intermittent inductor current mode. In this case, the inductor current stops by the moment the switch is closed and its increase starts from zero. The intermittent current mode is undesirable at a load current close to the nominal one, since in this case increased output voltage ripple occurs. The most optimal situation is when the stabilizer operates in the continuous current mode of the inductor at maximum load and in the intermittent current mode, when the load decreases to 10 ... 20% of the nominal. The output voltage is regulated by changing the ratio of the closed state time of the switch to the pulse repetition period. In this case, depending on the circuitry, various options for implementing the control method are possible. In devices with relay control, the transition from the switched on state to the switched off state determines the comparison node. When the output voltage is greater than the set value, the switch is turned off, and vice versa. If you fix the pulse repetition period, then the output voltage can be adjusted by changing the duration of the switched on state of the switch. Sometimes methods are used in which either the time of the closed or the time of the open state of the switch is fixed. In any of the control methods, it is necessary to limit the inductor current at the stage of the closed state of the switch to protect against output overload. For these purposes, a resistive sensor or a pulse current transformer is used. The choice of the main elements of a pulsed step-down stabilizer and the calculation of their modes will be carried out using a specific example. All the ratios that are used in this case were obtained on the basis of the analysis of the functional diagram and timing diagrams, and the technique [1] was taken as the basis. Let it be necessary to calculate a pulse step-down stabilizer with the following parameters: UBX=18...32 V, Ulx=12V, Iout=5A. 1. Based on a comparison of the initial parameters and the maximum permissible values of the current and voltage of a number of powerful transistors and diodes, we first select a bipolar composite transistor KT853G (electronic switch S) and a diode KD2997V (VD) [2, 3]. 2. Calculate the minimum and maximum fill factors: γmin=t and min /Tmin=(UByX+Upr)/(UBX max+Uson - URdT+Upr)=(12+0,8)/(32-2-0,3+0,8)=0,42; γmax \u12d t and max / Tmax \u0,8d (UByx + Upp) / (UBx min - Usbcl -URdt + Upp) \u18d ( 2 + 0,3) / ( 0,8-0,78-0,8 + 2) \u853d 21, where Upp=250 V is the direct voltage drop across the diode VD, obtained from the direct branch of the current-voltage characteristic for a current equal to Iout in the worst case; Usbcl \u0,3d XNUMX V - saturation voltage of the KTXNUMXG transistor, which acts as a switch S, with a current transfer coefficient in saturation mode hXNUMXe \uXNUMXd XNUMX; URdT = XNUMX V - voltage drop across the current sensor at rated load current. 3. Select the maximum and minimum conversion frequency. This item is performed if the pulse period is not constant. We choose a control method with a fixed duration of the open state of the electronic switch. In this case, the following condition is fulfilled: t=( 1 - γmax)/fmin = ( 1 - γmin)/fmax=const. Since the switch is made on the KT853G transistor, which has poor dynamic characteristics, we will choose the maximum conversion frequency as relatively low: fmax=25 kHz. Then the minimum conversion frequency can be defined as fmin=fmax( 1 - γmax)/( 1 - γmin) =25 103]( 1 - 0,78)/(1-0,42)=9,48 kHz. 4. Calculate the power loss on the switch. Static losses are determined by the effective value of the current flowing through the switch. Since the current shape is a trapezoid, then Is = Iout Dynamic losses on the switch Рsdyn 0,5fmax UBX max(lsmax tf+α llx tcn), where Ismax is the switch current amplitude due to the reverse recovery of the VD diode. Taking lSmax=2lByX, we get Psdin=0fmax UBX max Iout( 5tf+ α∙ tcn )=2 0,5 25 103 32(5 2-0,78-10+6-1,25-2-10) \u6d 8,12 W, where tf \u0,78d 10 6-2 s - the duration of the front of the current pulse through the switch, tcn \u10d 6 XNUMX-XNUMX s - the duration of the decline. The total losses on the switch are: Рs=Рscat+Рsdin=6,54+8,12=14,66 W. If static losses prevailed on the switch, the calculation should be carried out for the minimum input voltage when the inductor current is maximum. In the case where it is difficult to predict the predominant type of losses, they are determined both at the minimum and at the maximum input voltage. 5. We calculate the power loss on the diode. Since the shape of the current through the diode is also a trapezoid, its effective value is defined as Static losses on the diode PvDcTaT=lvD Upr=3,84-0,8=3,07 W. The dynamic losses of the diode are mainly due to losses during reverse recovery: PVDdyn=0,5fmax lsmaxvUBx max toB fmax lByx Uin max tov 25-103 -5-32 0,2 10-6=0,8 W, where tOB=0,2-1C-6 s - diode reverse recovery time. The total losses on the diode will be: PVD \u3,07d PMDstat + PVDdyn \u0,8d 3,87 + XNUMX \uXNUMXd XNUMX W. 6. Choose a heat sink. The main characteristic of a heat sink is its thermal resistance, which is defined as the ratio between the temperature difference between the environment and the heat sink surface to the power dissipated by it: Rg=ΔT/Rrass. In our case, it is necessary to fix the switching transistor and the diode on the same heat sink through insulating spacers. In order not to take into account the thermal resistance of the gaskets and not complicate the calculation, we choose a low surface temperature, approximately 70°C. Then at an ambient temperature of 40°CΔT=70-40=30°C. The thermal resistance of the heat sink for our case is Rt=ΔT/(Ps+Pvd)=30/(14,66+3,87)=1,62°C/W. Thermal resistance during natural cooling is given, as a rule, in the reference data for the heat sink. To reduce the size and weight of the device, you can apply forced cooling with a fan. 7. Calculate the throttle parameters. Calculate the inductance of the inductor: L= (UBX max - Usbkl-URdt - UBout)γmin / [2Ivyx fmax(α-1)]=(32-2-0,3-12) 0,42/[2 5 25 103 (1,25-1)]=118,94 μH. As the material of the magnetic circuit, we choose pressed Mo-permalloy MP 140 [4]. The variable component of the magnetic field in the magnetic circuit in our case is such that hysteresis losses are not a limiting factor. Therefore, the maximum induction can be chosen on the linear section of the magnetization curve near the inflection point. Working on a curved section is undesirable, since in this case the magnetic permeability of the material will be less than the initial one. This in turn will cause the inductance to decrease as the inductor current increases. We choose the maximum induction Bm equal to 0,5 T and calculate the volume of the magnetic circuit: 0=2 cm2, where μ=140 - initial magnetic permeability of MP4 material; μ10=7π·118,94-10 H/m - magnetic constant. According to the calculated volume, we select the magnetic circuit. Due to the design features, the MP140 permalloy magnetic circuit is usually performed on two folded rings. In our case, rings KP24x13x7 are suitable. The cross-sectional area of the magnetic circuit Sc=20,352 =0,7 cm2, and the average length of the magnetic line λс=5,48 cm. The volume of the selected magnetic circuit is: VC=SC λс=0,7 5,48=3,86 cm3>Vp . Calculate the number of turns: We determine the diameter of the wire with insulation based on the fact that the winding must fit into one layer, turn to turn along the inner circumference of the magnetic circuit: mm - inner diameter of the magnetic circuit; k3=13 is the filling factor of the magnetic circuit window with the winding. We select the wire PETV-2 with a diameter of 1,32 mm. Before winding the wire, the magnetic core should be insulated with a PET-E film 20 µm thick and 6...7 mm wide in one layer. 8. Calculate the output capacitor capacity: ·8-·2-32(2·0,3)0,42]=8 µF, where ΔUCout=0,01 V is the ripple range at the output capacitor. The above formula does not take into account the influence of the internal, series resistance of the capacitor on the ripple. With this in mind, as well as a 20% tolerance for the capacitance of oxide capacitors, we select two K50-35 capacitors for a nominal voltage of 40 V with a capacity of 1000 microfarads each. The choice of capacitors with an overestimated rated voltage is due to the fact that with an increase in this parameter, the series resistance of the capacitors decreases. The scheme developed in accordance with the results obtained in the course of the calculation is shown in fig. 3.
Let's consider the stabilizer in more detail. During the open state of the electronic switch - transistor VT5 - a sawtooth voltage is formed on the resistor R14 (current sensor). When it reaches a certain value, the transistor VT3 will open, which, in turn, will open the transistor VT2 and discharge the capacitor C3. In this case, the transistors VT1 and VT5 will close, and the switching diode VD3 will also open. The previously open transistors VT3 and VT2 will close, but the transistor VT1 will not open until the voltage across the capacitor C3 reaches the threshold level corresponding to the voltage of its opening. Thus, a time interval will be formed during which the switching transistor VT5 will be closed (approximately 30 μs). At the end of this interval, transistors VT1 and VT5 will open and the process will repeat again. Resistor R10 and capacitor C4 form a filter that suppresses the voltage surge at the base of transistor VT3 due to the reverse recovery of diode VD3. For silicon transistor VT3, the base-emitter voltage at which it switches to active mode is about 0,6 V. In this case, relatively large power is dissipated on the current sensor R14. To reduce the voltage on the current sensor, at which the transistor VT3 opens, a constant bias of about 0,2 V is applied to its base along the VD2R7R8R10 circuit. A voltage proportional to the output voltage is supplied to the base of the transistor VT4 from a divider, the upper arm of which is formed by resistors R15, R12, and the lower arm is resistor R13. The HL1R9 circuit generates a reference voltage equal to the sum of the direct voltage drop across the LED and the emitter junction of the transistor VT4. In our case, the exemplary voltage is 2,2 V. The mismatch signal is equal to the difference between the voltage at the base of the VT4 transistor and the exemplary one. The output voltage is stabilized due to the summation of the mismatch signal amplified by the transistor VT4 with the voltage based on the transistor VT3. Assume that the output voltage has increased. Then the voltage at the base of the transistor VT4 will become more exemplary. Transistor VT4 opens slightly and shifts the voltage at the base of transistor VT3 so that it also starts to open. Consequently, the transistor VT3 will open at a lower level of the sawtooth voltage across the resistor R14, which will lead to a reduction in the time interval at which the switching transistor will be open. The output voltage will then decrease. If the output voltage decreases, the regulation process will be similar, but occurs in the reverse order and leads to an increase in the open time of the switch. Since the current of the resistor R14 is directly involved in the formation of the open time of the transistor VT5, here, in addition to the usual output voltage feedback, there is a current feedback. This allows you to stabilize the output voltage without load and provide a quick response to a sudden change in current at the output of the device. In the event of a short circuit in the load or overload, the stabilizer switches to current limiting mode. The output voltage begins to decrease at a current of 5,5 ... 6 A, and the closing current is approximately equal to 8 A. In these modes, the on-time of the switching transistor is reduced to a minimum, which reduces the power dissipated on it. If the stabilizer does not work properly, caused by the failure of one of the elements (for example, a breakdown of the transistor VT5), the voltage increases at the output. In this case, the load may fail. To prevent emergency situations, the converter is equipped with a protection unit, which consists of a trinistor VS1, a zener diode VD1, a resistor R1 and a capacitor C1. When the output voltage exceeds the stabilization voltage of the zener diode VD1, a current begins to flow through it, which turns on the trinistor VS1. Its inclusion leads to a decrease in the output voltage to almost zero and blown fuse FU1. The device is designed to power 12-volt audio equipment, designed mainly for passenger vehicles, from the on-board network of trucks and buses with a voltage of 24 V. Due to the fact that the input voltage in this case has a low ripple level, capacitor C2 has a relatively small capacitance. It is insufficient when the stabilizer is powered directly from the mains transformer with a rectifier. In this case, the rectifier should be equipped with a capacitor with a capacity of at least 2200 microfarads for the corresponding voltage. The transformer must have an overall power of 80 ... 100 W. The stabilizer uses oxide capacitors K50-35 (C2, C5, C6). Capacitor C3 - film K73-9, K73-17, etc. of suitable sizes, C4 - ceramic with low self-inductance, for example, K10-176. All resistors, except R14, are C2-23 of the corresponding power. Resistor R14 is made of a 60 mm long piece of PEC 0,8 constantan wire with a linear resistance of approximately 1 ohm/m. A drawing of a printed circuit board made of one-sided foil-coated fiberglass is shown in fig. 4.
Diode VD3, transistor VD5 and trinistor VS1 are attached to the heat sink through an insulating heat-conducting gasket using plastic bushings. The board is also fixed on the same heat sink. The appearance of the assembled device is shown in Fig. 5.
Today, the development of switching regulators has been greatly simplified. Integrated circuits have become available (including at a price), which include all the necessary nodes. In addition, semiconductor manufacturers began to accompany their products with a large amount of application information containing typical switching circuits that satisfy the consumer in the vast majority of cases. This practically eliminates the stages of preliminary calculations and prototyping from the development. An example of this is the KR1155EU2 chip [5]. It includes a switch, a current sensor, a reference voltage source (5,1 V ± 2%), a trinistor control unit for protection against overvoltage at the load, a soft start unit, a reset unit for external devices, a remote shutdown unit, a protection unit chips from overheating. Consider a laboratory power supply developed on the basis of KR1155EU2. Technical specifications
The scheme of the device is shown in fig. 6. It differs little from the standard switching circuit, and the item designations of the elements are the same. Here, a control method with a fixed pulse repetition period is implemented, i.e., pulse-width control.
Capacitor C1 - input filter. It has a larger capacitance than indicated in the typical switching circuit, which is due to the relatively large current consumption. Resistors R1 and R2 control the current protection level. Their maximum total resistance corresponds to the maximum protection operation current, and the minimum resistance corresponds to the minimum current. With the help of capacitor C4, the stabilizer is smoothly started. In addition, its capacitance determines the restart period when the current protection threshold is exceeded. Resistor R5 and capacitors C5, C6 are elements of frequency compensation of the internal error amplifier. Capacitor C3 and resistor R3 determine the carrier frequency of the pulse-width converter. Capacitor C2 sets the time between a sharp decrease in the output voltage (caused by external causes, for example, a short-term output overload) and the transition of the RESO signal (pin 14 DA1) to a state corresponding to normal operation, when the transistor connected between the RESO and GND pins inside the microcircuit closes . Resistor R6 provides an open collector load for this transistor. If you plan to use the RESO signal with its binding to a voltage different from the output voltage of the stabilizer, then the resistor R6 is not installed, and the open collector load is connected inside the RESO signal receiver. Resistor R4 provides zero potential at the INHI input (pin 6 of DA1), which corresponds to the normal operation of the microcircuit. The stabilizer can be turned off by an external high TTL level signal. The use of the KD636AS diode (its total allowable current significantly exceeds that required in this stabilizer) allows you to increase the efficiency by 3 ... 5% with a slight increase in the cost of the device. This leads to a decrease in the temperature of the heat sink and, consequently, to a decrease in its dimensions and weight. Resistors R7 and R8 are used to regulate the output voltage. When the slider of the resistor R7 is in the lower position according to the diagram, the output voltage is minimal and equal to the reference voltage of the DA1 chip, respectively, when in the upper position the output voltage is maximum. The trinistor VS1 opens with a CBO signal (pin 15 DA1) if the voltage at the CBI input (pin 1 DA1) exceeds the internal reference chip DA1 by approximately 20%. This is how the load is protected from overvoltage at the output. All oxide capacitors are K50-35, except for C1 - K50-53. Capacitor C6 - ceramic K10-176, the rest are film (K73-9, K73-17, etc.). All fixed resistors are C2-23. Variable resistors R2 and R7 - SPZ-4aM with a power of 0,25 watts. They are mounted on the board using brackets. The inductor L1 is wound on two folded annular magnetic cores K20x 12x6,5 made of MP140 permalloy. The winding contains 42 turns of wire PETV-2 1,12, wound in two layers: the first - 27-28 turns, the second - the rest. The stabilizer is assembled on a board made of one-sided foil fiberglass. The board drawing is shown in fig. 7.
The microcircuit, diode and trinistor are fixed on the same heat sink. In this case, the microcircuit in most cases can not be isolated from the heat sink surface, since its flange is connected to pin 8 (GND). Diode and trinistor must be isolated. The heat sink is chosen on the basis of a dissipated power of approximately 15 ... 20 W and an overheat of 30 ° C. You can reduce the size and weight of the heat sink by using a fan (if possible). Pay special attention to the mains transformer and rectifier. The transformer is designed for an output power of at least 150 W and an open circuit output voltage of approximately 33 V. At maximum load, a decrease in the output voltage by no more than 1,5 V relative to the open circuit voltage is permissible. The rectifier is selected for a current of 3,5 ... 2 A with a total voltage drop across its diodes of not more than XNUMX V. The rectifier (in the case of a monolithic version) or individual diodes can be mounted on the same heat sink as the stabilizer. A good alternative to a mains transformer and a rectifier can be a pulse converter. Analyzing the two considered devices, one can see their differences. Obviously, the first stabilizer is cheaper than the second. Moreover, the ways to further reduce the cost of the first one are very obvious (replacement of the KD2997V diode by KD213V with a slight deterioration in efficiency and an expensive permaploy one with a cheap ferrite magnetic core). In the second device, KD213V (as, indeed, KD2997V) is no longer suitable due to inertia, and replacing the magnetic circuit will not lead to a noticeable decrease in cost. Details for the first stabilizer can be found in the desktop of any radio amateur, which cannot be said about the second. However, the first device requires more time at the design stage. In addition, it has a larger number of elements with less functionality. Literature
Author: Yu.Semenov, Rostov-on-Don
Traffic noise delays the growth of chicks
06.05.2024 Wireless speaker Samsung Music Frame HW-LS60D
06.05.2024 A New Way to Control and Manipulate Optical Signals
05.05.2024
▪ On January 4, the Earth will get as close as possible to the Sun. ▪ Motorola Skip NFC tag unlocks smartphone without entering password ▪ The most massive neutron star discovered ▪ Face mask with microphone and speakers
▪ section of the site Fundamentals of safe life (OBZhD). Article selection ▪ article Where is the coldest place in the universe? Detailed answer ▪ Article Commercial Director. Job description ▪ article Repair of rubber hand pumps. Simple recipes and tips ▪ article Climb through the sheet. Focus Secret
Home page | Library | Articles | Website map | Site Reviews
www.diagram.com.ua |
where α=lLmax /llx=1,25 is the ratio of the maximum inductor current to the output current. Coefficient a is chosen within 1,2 ... 1,6. Switch static loss PSstat=lsUSBKn=3,27-2=6,54W.
We take the number of turns equal to 23.
See other articles Section 
Latest news of science and technology, new electronics:
All languages of this page