|
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 Homemade UPS for imported transceiver. Encyclopedia of radio electronics and electrical engineering
Encyclopedia of radio electronics and electrical engineering / Civil radio communications Many radio amateurs, for sure, came up with the following thought: "How absurd it turns out! Transceivers are steadily decreasing in size and weight, and power supplies are still heavy and bulky." The author of this article thought about the same. The result of these reflections was the development of a power supply unit, which at the moment managed to visit many radio expeditions and rallies, where, under rather harsh conditions, without turning off for days, it powered imported transceivers of more than ten different models at full output power from both a stationary lighting network and from petrol stations. Few observations Interesting conclusions can be drawn by creatively comprehending the parameters of imported transceivers, given in their "User's Manual" and in the "Maintenance Manual" and which even an experienced radio amateur often passes by. Judge for yourself. Do I need voltage stabilization for a transceiver, the supply voltage of which, according to passport data, can vary within ±15% of the nominal value of 13,8 V, in accordance with GOST, the mains voltage can vary within ±10%? Supporters of hard, up to millivolts, stabilization in power supplies can be recommended to measure the supply voltage fluctuations directly at the transceiver connector, that is, taking into account the voltage drop on the wires, and also try to power the transceiver from a car battery. In the first case, you can see a voltage drop of about 0,5 V, and in the second case, with a battery, even more, and the voltage can fluctuate both negatively and positively. Is it worth it after such arguments to strive so carefully to stabilize the voltage in the power supply? Looking at the schematic diagram of the transceiver, you can even more assert yourself in the opinion that you should not spend extra effort on stabilization. The transceiver itself has its own efficient internal power supply system for individual nodes. In general, it can be divided into three branches: a +5 V voltage regulator for powering all digital microcircuits, a +9 V voltage regulator for powering the preliminary stages of the transceiver path, and, finally, the power supply system for the transmitter output stage. Only the power amplifier of the transceiver receives full voltage from the power connector, and even then it passes through internal filters and fuses. It is protected from exceeding by a powerful zener diode, designed for a voltage slightly higher than the maximum permissible voltage, connected in parallel to the supply circuit after the fuses. The constancy of the output power is maintained by the ALC system. In switching power supplies, ripples with a conversion frequency are easily filtered out using small capacitance and, accordingly, small capacitors connected after the output rectifier. Technical task All the above considerations formed the basis of the idea of the design that now feeds the author's transceiver. The idea is unusual, non-traditional, and it was to create a converter of AC mains voltage to DC voltage close to the nominal (13,8 V), with the necessary load capacity, but without stabilization losses. Obviously, this device had to use the principle of high-frequency conversion of the rectified mains voltage. Additional requirements for the design - the simplicity of the circuit, if possible, the absence of scarce, imported expensive parts, maximum efficiency and the lowest possible level of impulse noise. From previous experience, it was clear that it is unlikely that it will be possible to completely remove impulse noise from a source during its home-made manufacture. Therefore, it was decided to use quartz stabilization of the conversion frequency and make this frequency as high as possible. A high conversion frequency allows better filtering of interference, while reducing the size of the power supply. Quartz stabilization with a "round" value of the conversion frequency, for example, 50 kHz, made it possible to concentrate the affected areas in a narrow band. After mounting the working layout in a steel perforated case, the interference from the source became completely invisible. But do not think that they have disappeared altogether. In fact, their level is so low that it is masked by the noise of the ether. The result is a device with the following parameters: power supply voltage - 220 ± 10% V; voltage without load - 15,2 V; voltage in receive mode - 14,7 V; transmission voltage in SSB mode (100 W, compression 25 dB) - 13,5 V, in CW mode (100 W) - 12,5 V; minimum efficiency - 85%. The power supply has dimensions of 100x60x80 mm and a weight of about 350 g. Operating principle At first glance at the block diagram of the power supply (Fig. 1), nothing new can be found in it, in comparison with the already known block diagrams of similar devices, and this is a completely correct conclusion. This design uses well-known circuit solutions, but the element base is new.
As in other pulsed sources, such as, for example, in any modern TV or computer, the mains voltage is supplied through a filter, then rectified by a diode bridge. The ripples are filtered out by an electrolytic capacitor. The value of the rectified voltage on this capacitor will be approximately 310 V. This voltage is switched by a bridge "H"-shaped circuit on four field-effect transistors. Experts call this node "inverter". From the diagonal of the bridge, a rectangular voltage is fed to a step-down transformer, rectified, filtered and fed to the output of the device. The use of new transistors made it possible to significantly increase the steepness of the fronts at the output of the inverter, which, in turn, made it possible to reduce the time for the through current to flow through the bridge arms at the moment of its switching. This circumstance, in turn, made it possible to obtain a large gain in the efficiency of the cascade and to raise the conversion frequency. The efficiency of the key stage has increased so much that it was possible to completely abandon the radiators for transistors. Moreover, with a maximum converter power of approximately 250 W, the power supply case remains slightly warm for a long time of operation. Insulated gate field effect transistors, unlike bipolar ones, do not have the effect of accumulation of minority carriers in the base region - saturation, which does not delay their switching speed. In addition, they are able to adjust their drain current as the case temperature increases. Another amazing property of them is that they have an infinitely large power gain in static mode, i.e., without consuming power in the gate circuit, they are able to switch significant powers in the channel circuit (drain-source section). Therefore, in the dynamic mode, energy is spent mainly to compensate for the charge accumulated on the interelectrode capacitance of the gate-source during the previous half-cycle of the control voltage. The value of this capacitance is approximately 1000 pF and determines the requirements for the driver - it must provide good edge steepness and a constant amplitude of the pulses applied to the gates of the keys when operating on a capacitive load. The modern element base helped here too. Digital microcircuits of the KR1554 (74NS) series do an excellent job with the task. Schematic diagram of a switching power supply is shown in fig. 2.
The mains voltage of 220 V is supplied to the bridge assembly VD1 of the driver power supply unit through the ballast capacitor C1 and the resistor R2, which dampens the starting current pulse. All diodes of this assembly are shunted with small capacitors C2 - C4 to neutralize their dynamic capacitance. Resistor R1 discharges capacitor C1 after the device is turned off. The driver consists of a 50 kHz crystal oscillator and a powerful stage. The voltages on the gates in the required phases are supplied through a transformer power addition circuit on two ferrite rings. The driver is powered by a separate power node using a ballast capacitor in the mains circuit. The rectified pulsating voltage from the bridge is supplied directly to the zener diode VD2. Usually in such circuits in the zener diode circuit, in series with it, a limiting resistor is placed, but in this case, capacitor C1 itself plays its role. The maximum current that can be obtained from the rectifier depends on the capacitance of this capacitor. Without an additional resistor, the circuit also acquires a number of useful properties: efficiency and load capacity increase. If you look at the voltage waveform on the zener diode VD2, when the filter capacitor C7 and the voltage regulator DA1 have not yet been soldered, the voltage shape, compared to the output voltage shape of a simple full-wave rectifier with filters, looks unusual. Instead of the usual "humps" we will see an almost constant, even voltage, cut through by thin negative pulses that occur at the moment the mains voltage sine wave passes through zero. The amplitude of the pulses is equal to the stabilization voltage of the zener diode +10 V. Capacitor C7 is much easier to filter these pulses than a full-wave rectified sinusoidal voltage. After mounting the stabilizer DA1 and capacitor C11, the first tests can be made. Switch the mains voltage on and off several times at short intervals. If nothing exploded, you can leave the network turned on and check the voltage at the +5 V stabilizer output. Then you need to check the load capacity of the driver power node. This node is not at all afraid of a short circuit, so its load capacity can be roughly estimated by simply connecting a tester, connected as a milliammeter, to the output of the stabilizer - in parallel with the terminals of the capacitor C11. In this case, the arrow of the device should show a current of at least 25 mA. Attention! The elements of the circuit are under the potential of the lighting network and experiments (adjustment, preliminary tests) should be carried out through an isolating network transformer with a transformation ratio of 1: 1, with a power of about 100 W. A stabilized voltage of +5 V is supplied to the driver - microcircuits DD1, DD2. The first of them (DD1) is a microcontroller of the AVR family developed by ATMEL. To work, this chip must first be programmed. The dump of machine firmware codes is shown in the table.
I must say that the first version of the power supply was assembled without the use of a microcontroller at all: a separate 100 kHz crystal oscillator, a divider by two, and a start-up delay unit on an RC chain. The device was fully functional. But it had nasty transients during launch. There is no such phenomenon with a microprocessor. The DD1 controller performs three relatively simple tasks: a guaranteed two-second software delay after power-up, generating anti-phase rectangular pulses on its pins 6 and 7, and generating strobe pulses on pin 5. Clock intervals in the micro-computer are set by a ZQ1 quartz resonator with a frequency of 10 MHz. To install the microcontroller on the board, it is desirable to provide a connector. The functioning of the programmed DD1 chip should be checked with an oscilloscope. Pins 6 and 7 should have an anti-phase square wave with a frequency of 50 kHz, and pin 5 should have short negative pulses. The amplitude of the signals must be equal to the supply voltage of the microcircuit +5 V, and the fronts must be steep, without blockages and surges. The current consumption of the DD1 chip is about 6 mA. From the outputs of the controller, the pulses are fed to the inputs of the DD2 chip. These are four D-flip-flops with common clock and reset inputs. It is the use of the DD1 chip that the power supply owes its remarkable properties. The KR1554 series (its imported analogue 74NS) has been developed for a long time and, in my opinion, is undeservedly ignored by radio amateurs. Here are just some of its characteristics taken from the reference book: supply voltage - +1 ... 7 V, current consumption in static mode - no more than 80 μA, output current on a separate output - up to 86 mA, maximum clock frequency - 145 MHz. The last two parameters provide the highest switching speed of the VT1 - VT4 switches, minimizing the time for the flow of through currents through the bridge arms on these transistors, and hence the high efficiency and the absence of radio interference. The chain C22, R4, VD7 is used to auto-reset triggers DD2 at the time the mains power is turned on. Capacitors C16, C17 - blocking. They must be installed near the power pins of the DD1, DD2 microcircuits. After installing the microcircuits on the board, the next electrical measurements should be made. The total current consumption of the processor and triggers without connected transformers T3 and T4 should be about 6,5 mA, and the signal shape at the DD2 outputs should be rectangular, without surges and blockages at the fronts and decays of the pulses. The two output transformers of the driver T3 and T4 are identical in design and are wound with PEV-0,1 wire on ferrite rings of the NM1000, .. NM2000 brand with an outer diameter of about 10 mm. The winding is made of "pigtail" of eight copper conductors with lacquer insulation. Of these, four conductors form the primary winding and are connected in series - beginning to end. The four remaining are secondary and connected as shown in the diagram. Thus, each transformer turns out to be step-down with a transformation ratio of 4: 1. Before winding the wire, the fabric is twisted (4 - 6 twists per centimeter). All sharp edges of the rings, both external and internal, must be rounded off. The use of a circuit of two ring transformers with separate magnetic fluxes made it possible to obtain the required driver power. At first glance, it seemed that it would be enough to excite all the outputs of the DD2 chip in phase and parallel them, but this does not help much. The load capacity of the node depends on the internal resistance of the outputs of the DD2 chip. When the outputs are connected in parallel, their equivalent internal resistance decreases exponentially, with the use of a step-down transformer, it decreases exponentially. This circuit technique made it possible to obtain the necessary load capacity of the driver while maintaining the initial steepness of the fronts and decays of the pulses. Let me remind you that the power of the driver is spent mainly on recharging the interelectrode capacitance of the gate-source transistors VT1 - VT4. This method of power addition, if desired, can be applied in the output stage. How to determine the correct number of turns of transformers T3, T4? The criterion is the degree of increase in the current consumption of the driver when connecting the primary windings of the transformers to the outputs of the DD2 microcircuit. The secondary windings are not loaded. The experiment should start with a relatively large number of turns - 30...40 and gradually reduce their number by controlling the driver current. At first, the current increases very slightly, but from a certain point, each removed turn will lead to a sharp increase in current. The number of turns must be left such that the driver's no-load current is on the verge of increasing. In this case, there will be a maximum load capacity and efficiency of transformers. For convenience, experiments can be performed with a single wire. This technique can also be applied to clarify the number of turns of any transformer - both mains and high-frequency. For the described power supply, the total current consumption of microcircuits DD1, DD2 with transformers T3 and T4 at idle, without load, should be about 8 mA. The load capacity of the driver is checked using resistors with a resistance of about 100 ohms, temporarily connected to the secondary windings of transformers T3, T4. An oscilloscope controls the amplitude and shape of the pulses. As for the previous measurements, there should be no squareness distortion, and the pulse amplitude should be about 5 V. After connecting the secondary windings of the transformers to the gate circuits of transistors VT1 -VT4, the driver consumption current will increase to approximately 12 mA. The output stage is assembled according to the bridge circuit. The advantages of this circuit, compared to the more common half-bridge, are obvious: it quadruples the output power, the best efficiency of both the transistors themselves and the output power transformer T2. The field-effect transistors with an insulated gate KP707A used in the power stage have a "right" characteristic of the dependence of the drain current on the gate voltage. This means that the current through the channel, the drain-source section, will only flow if the voltage between the source and the gate is positive. And even then, with a gate voltage of less than 3 V, the transistor is still closed. Therefore, it is advisable to "raise" the amplitude of the buildup pulses above the zero level. Otherwise, the negative half-cycles of these pulses would be wasted - the transistors are still closed! This task is handled by RC chains R6 - R9, C31 - C34 and diodes VD10 - VD13 in gate circuits VT1 - VT4. This technique made it possible to reduce the amplitude of the buildup voltage by half. By the way, the "dead zone" of the gate voltage automatically provides a protective interval between the moments when one bridge arm is turned off and the other is turned on, which reduces the amount of through current through a pair of transistors at the moment they are switched. The output transistors are powered by a mains voltage rectifier assembled according to a bridge circuit on diodes VD3 - VD6. Capacitors C18 - C21 prevent the occurrence of modulating interference penetrating from the network. Capacitor C23 smooths out the ripple of the rectified voltage. If desired, its capacity can be slightly increased. Resistor R5 discharges this capacitor when the power supply is turned off and is mainly intended to ensure the safety of those who like to fall under the residual charge on high-voltage electrolytic capacitors. Resistor R3 (negative temperature coefficient thermistor) provides damping of the charging current pulse of capacitor C23 at the moment the mains power is turned on. At the moment the unit is connected to the network, R3 has an ambient temperature and its resistance is equal to the nominal - 10 ohms. As the power in the load increases, the power dissipated in this element also increases and it begins to heat up. As a result, its resistance drops. It's like he's shorting himself. The use of a thermistor additionally gives the effect of some stabilization of the output voltage of the power supply. It can be replaced by a conventional resistor with a power of about 10 W with a nominal value of 5 ohms. At the input of the power supply there is a two-stage filter L1 and T1, C6, C8 - C10. The pre-filter L1 is made on a ferrite ring with a diameter of about 20 mm with a permeability of 1000...2000 and contains three windings located along the radius at an angle of 120 degrees to each other and having three turns. The winding is carried out with a PVC-insulated mains wire until the entire perimeter of the magnetic circuit is uniformly filled in one layer. The filter transformer T1 uses a ferrite ring similar to L1. Both windings contain 30 turns each, are made with an insulated network wire and are located on diametrically opposite sides of the magnetic circuit. The nominal value of the voltage supplied from the output of the mains rectifier to the output stage is +310 V, and the current flowing through both arms of the bridge without the connected output transformer T2 with the control voltage supplied from the driver must not exceed 12 mA, i.e. 6 mA per arm. Resistors R10, R11 dampen pulses of through currents through a pair of transistors VT1, VT2 and VT3, VT4. They can also be used for oscillographic observation of the amplitude and shape of these pulses. For the first, after the completion of the installation of the output stage, switching on the power supply, we can recommend a reduced supply voltage of 10 ... 15 V supplied from a separate source. The mode of operation of transistors VT1 - VT4 is such that they do not need radiators at all - they are located vertically on the board, in one row, and are slightly blown by a twelve-volt fan 40x40 mm in size, taken from the computer. The fan power is taken from the output of the power supply and fed to the motor through a stabilizer on the DA2 chip. In this case, the device receives sufficient cooling, and the fan is not audible. The output transformer T3 is wound on a pot-shaped ferrite magnetic core of the brand M2000NM1 with a diameter of 30 mm. It is necessary to ensure that the magnetic circuit is without a gap in the core. The primary winding contains 60 turns of PELSHO wire, the winding is made in bulk, the turns are evenly distributed over the frame. The use of a sectioned frame is categorically unacceptable - the primary and secondary windings are wound in two layers, one above the other. Otherwise, the broadband of the transformer is disturbed, oscillatory processes occur and the overall efficiency of the unit is sharply reduced. The secondary winding is shielded from the primary with a strip of copper foil insulated. The screen forms one and a half open turns. For the secondary winding, a bundle of an even number of conductors with a diameter of about 0,1 mm is used, twisted together. Such a home-made litz wire is filled into a heat shrink tube with a diameter of 4 ... 6 mm. This tube is made three turns over the primary winding. Then the conductors are divided by number into two equal groups. The beginnings of the first group are connected to the ends of the second group. Thus, a winding of six turns is formed with a conclusion from the midpoint. After the manufacture of the transformer T1 and its installation - a traditional test: measuring the current of the output transistors in idle mode. It should be about 25 mA at a full supply voltage of +310 V. The secondary winding is loaded on a full-wave half-bridge rectifier on diodes VD8, VD9. The diodes are located on a common radiator - an aluminum plate measuring 30x40 mm. The radiator, transformer T1 and output transistors are blown by a fan. The rectified voltage is supplied to the output connector XS2 through the filter T5, C25 - C3O. The T5 transformer is similar in design to T1, but is made with a thicker wire. The power supply used capacitors K73-17 with a capacity of 0,68 microfarads for a voltage of 400 V (C1) and an imported Rubicon company with a capacity of 100 microfarads for a voltage of 400 V (C23). To increase reliability, we recommend installing resistors R1 and R5 with a resistance of 100 kOhm with a power of at least 1 W, and replace the diodes KD2998 (VD8, VD9) with 2D252A or 2D252B or imported 30CPQ060. Structurally, the power supply was "born", and to this day exists in the form of a well-made, but still a layout. Its appearance is shown in Fig. 3.
The parts are mounted on a board made of double-sided foil fiberglass by surface mounting, without holes, on cut patches. The connections are made with wires in PTFE insulation. Metaposition on the other side of the board is preserved. Author: S.Makarkin (RX3AKT), Moscow
The world's tallest astronomical observatory opened
04.05.2024 Controlling objects using air currents
04.05.2024 Purebred dogs get sick no more often than purebred dogs
03.05.2024
▪ Hydrogen storage in residential areas ▪ MAX5945 Ethernet Power Controller ▪ Solar tornadoes will help green energy
▪ section of the site Amateur radio calculations. Article selection ▪ an officer's article can only be replaced by death. Popular expression ▪ article Clove tree. Legends, cultivation, methods of application ▪ article A glass of milk. Focus Secret
Home page | Library | Articles | Website map | Site Reviews
www.diagram.com.ua |
See other articles Section 
Latest news of science and technology, new electronics:
All languages of this page