To date, pulsed AC-DC converters have a leading position among analogues. The most popular topology for impulse conversions is the flyback topology. Another reason for the popularity is the rather simple and inexpensive way to build a multi-channel power supply, which is provided by simply adding additional secondary windings to the transformer.
Typically, feedback is taken from the output requiring the most accurate output tolerance possible. This output then determines the voltage ratio for all other secondary windings. However, due to the effect of inductance dissipation, it is not always possible to achieve the required accuracy of adjusting the output parameters for various channels, especially in the case of a small load (or no load at all) on the main channel and full loading of the secondary channels.
Post-regulators and pre-loads can be used to stabilize output secondary channels. However, their use increases the final cost and reduces the effectiveness of the product, which makes them less attractive to consumers. This problem is especially acute due to the trend of tightening standards for power supplies in idle or standby mode.
The solution shown in image 1 is called "Active Shunt Regulator" and allows you to achieve parameters in accordance with the introduced standards and at the same time keep an acceptable budget for the end device.
Image 1. Active shunt regulator for multi-channel flyback topology
The scheme works as follows. While the outputs are within regulation, voltage divider R14 and R13 turn on Q5, which turns off Q4 and Q1. When current passes through Q5 in this mode of operation, there is a slight preload for the 5V output.
The nominal voltage difference between 5V output and 3.3V output is 1.7V. When the load on the 3.3V output starts to increase the current consumption without a corresponding increase in the current on the 5V output, the voltage on the 5V output will increase relative to the voltage of 3.3 Q. The moment the difference in nominal voltages exceeds 100mV, Q5 closes, this causes Q4 and Q1 to open, which in turn allows the 5V output current to drive the 3.3V output load and reduce the difference in voltage drift.
The current through Q1 is determined by the resulting voltage difference between the main and secondary channels and allows you to maintain the original voltage ratio regardless of the load, even in the case when the output is 3.3. loaded at 100%, 5V working without load. The consistency of Q5 and Q4 levels out the temperature drift of the parameters, since the change in VB-E of one transistor is compensated by the change in the other. Diodes D8 and D9 are not required, but reduce power dissipation in Q1, eliminating the need for a heatsink.
Because the circuit only responds to relative differences between the two voltages, it is largely inactive at full load and light load. Since the shunt is connected from the 5V output to the 3.3V output, the active power loss in the circuit is reduced by 66% compared to a shunt regulator that is connected to ground. As a result, efficiency remains high at full load and power consumption remains low across the entire load range.
This article will discuss methods for transmitting data over the power wires of devices. Particular attention is paid to the problems that need to be solved by the developer of such communication devices. Examples of the implementation of the receiving and transmitting parts for communication lines via DC power wires, as well as the implementation of a communication channel via AC power wires of 220 Volts with a frequency of 50 Hertz are given. Typical algorithms for the operation of the control microcontroller are described.
A bit of history
The idea of transmitting control signals over power wires is not new. Back in the 30s of the last century, bold experiments were carried out to transmit such signals over the wires of the power network of the city. The results obtained were not very impressive, but do not forget that in those days lamp technology reigned and the element base was not so diverse. Organizational problems were added to all technical problems: there was no single standard - each developer did everything for himself: different frequencies and modulations were used. All this hindered the development of this branch of communication.
The principle of operation of transmitting and receiving devices
The principle of operation of such devices is to transmit high-frequency signals through DC or AC power wires. In AC power lines, most often, signal transmission occurs at the moment the AC passes through zero, i.e., when the power voltage is absent or minimal. The fact is that the level of interference at this moment is minimal. In this case, the signal useful to us is transmitted, as it were, between a series of interference.
Transmission of a high-frequency signal over an alternating current network
A transformer is most often used to transfer a high-frequency signal to a power network. The receiving part usually consists of a coupling transformer and a circuit on which the necessary high-frequency signals are extracted.
Method for transferring high-frequency signals to an alternating current network
In DC power circuits, a similar method of transmitting high-frequency signals is used, but the principle of generating such a signal is different: a powerful key (transistor) briefly shunts the network with its transition. There is a slight decrease in the voltage in the network (Fig. 3).
Method for generating high-frequency signals in DC networks
A sensitive detector is installed on the receiving side, on which these voltage drops in the line are distinguished. Further, these signals are fed to the input of an amplifier with the AGC function, after which the received signals are transmitted to the logic block, which can be performed both on low-integration microcircuits, and on a universal microcontroller or a specialized microcircuit, which includes all of the above nodes. Recently, microcontrollers are increasingly used for such tasks due to their low price and great capabilities. Moreover, the use of programmable devices allows you to change the purpose of such devices by loading a new program into them - this is much easier and cheaper than making a new electronic device with a dozen microcircuits ...
Block diagram of a modern PLC modem
Advantages and disadvantages of this type of communication
The advantage of this type of communication is the sharing of an existing wire line of the power network. That is, it is not required to install a communication line, and there is a socket in almost any room.
The disadvantages include both the technical complexity of the device and the low speed when transferring data over distances greater than 100-300 meters.
Also, do not forget that this communication channel can be organized only between those devices that are connected to the same phase of the network and only within the same transformer substation - high-frequency signals cannot pass through the windings of the electrical substation transformer.
In principle, the latter limitation is partially removed by the use of passive or active repeaters of high-frequency signals. They are used both to transmit signals to another phase, and to transmit signals to the line of another transformer.
Technical difficulties in implementing a communication channel
Organization of a reliable communication channel over a power network is not a trivial task. The fact is that the network parameters are not constant, they change depending on the time of day: the number of devices connected to the network, their type and power change. Another of the negative features of the electrical networks of the countries of the former USSR is "hegemony" - powerful transformer substations that feed entire neighborhoods! Accordingly, hundreds of subscribers are connected to one phase of the transformer, in the apartment of each of them there is a large number of various devices. These are both devices with transformer power supplies and devices with switching power supplies. The latter are often made with violations in terms of electromagnetic radiation - interference, which creates a very high level of interference in the power network of the building and the city in particular.
In many countries, compact transformer devices are used to power buildings. One such transformer feeds from 3 to 7 apartments or houses. Consequently, the quality of electricity supplied to subscribers is much higher than in our electrical networks. Also, the resistance between the phase wire and zero is higher. All these factors make it possible to have better conditions for data transmission in an apartment or building than we have in our conditions.
A large number of devices connected to the network leads to low resistance between the phase wire and zero, it can be 1-3 ohms, and sometimes even less. Agree that it is very difficult to “shake” such a low-resistance load. In addition, do not forget that the networks are very large in area, therefore, they have a large capacitance and inductance. All these factors determine the very principle of building such a communication channel: a powerful output of the transmitter and high sensitivity of the receiver. Therefore, high frequency signals are used: the network has more impedance for high frequencies.
No less problem is the poor condition of power networks, both in general and inside buildings. The latter are often performed with violations, even the minimum requirement is also violated: the main line is made with a thicker wire than the outgoing supply lines to the rooms. Electricians know such a parameter as “phase-zero loop resistance”. Its meaning comes down to a simple relationship: the closer to the electrical substation, the thicker the wires should be, i.e., the cross section of the conductors should be larger.
If the cross section of the wires is chosen incorrectly, the laying of the trunk line is done “as it happened”, then the line resistance dampens high-frequency signals. The situation can be corrected either by improving the sensitivity of the receiver, or by increasing the transmitter power. Both the first and the second are problematic. Firstly, there is interference in the communication line, so increasing the sensitivity of the receiver to the level of interference will not increase the reliability of signal reception. Increasing the transmitter power can interfere with other devices, so this is not a panacea either.
Common standards. Standard X10
The most famous of the standards for transmitting commands over the power network is X10. This standard was developed a long time ago, in 1975 by the Scottish company Pico Electronics. Data is transmitted using a burst of pulses with a frequency of 120 kHz and a duration of 1 ms. They are synchronized with the moment when the alternating current passes through the zero value. For one zero crossing, one bit of information is transmitted. The receiver expects such a signal for 200 µs. The presence of a flash pulse in the window means a logical "one", the absence - a logical "zero". The bits are transmitted twice: the first time in direct form, the second time inverted. Usually, modules are implemented as separate devices, but now more and more often they are implemented not on the basis of different components, but using a microcontroller. This reduces the size of the receiver, making it possible to fit "smart stuff" even into an electric lamp socket or a doorbell.
As mentioned earlier, a high-frequency signal cannot propagate beyond the transformer substation and phase. Therefore, so-called active repeaters are used to obtain communication on another phase. But it must be taken into account that the receiver listens to the signal only at certain points in time. Therefore, either “smart” receivers are used, with modified parameters
This communication standard has both pluses and minuses. Firstly, he developed it a very long time ago, there were no microcontrollers, and all the circuitry was analog, using numerous components. Therefore, the communication protocol is very low-speed: no more than one bit is transmitted in one period of the network. The fact is that the bit is transmitted twice: in the first half-cycle it is transmitted in direct form, and in the second half-cycle - inversely. Secondly, some commands are transmitted in groups. This further increases the communication time.
Also, a significant drawback of this protocol is the lack of confirmation of the receipt of the command by the device. That is, having sent a command, we cannot be sure of its guaranteed delivery to the recipient. It also does not contribute to the spread of this standard.
Own experience. Reinventing the wheel
Having tested in real conditions numerous ready-made devices that allow transmitting commands over the power network, I came to a disappointing conclusion: at home, with a limited budget, without specialized devices and (what to hide?) knowledge, it will not work to invent something ingenious . But nothing and nothing prevents you from making a nice craft for yourself, under your specific conditions. This includes the area of application of such a product, the distances over which commands must be transmitted, as well as the functionality of such a device.
Let's complete some formalities in the form of a kind of technical task for our project:
- the device must transmit data over the wires of the power network;
- data must be transmitted in the "pauses" of the current, i.e., when the voltage in the network is minimal;
- the reliability of the communication channel is ensured both by hardware (by the optimal signal level at the receiving point) and by software (data are transmitted with a checksum to detect damage to the received data, commands are transmitted several times, the fact that the receiving device received the command is confirmed by sending the corresponding signal back to the main device);
- Let's simplify to the required level both the protocols for data exchange between devices in the network and the type of modulation. We will assume that one bit of data is transmitted for 1 millisecond. The unit will be transmitted in the form of a burst of pulses of this duration, and zero - its absence;
- on the network, all devices listen to signals, but only the device to which such a command is addressed executes the received command. That is, each of the devices has its own individual address - number.
The circuitry itself of the executive part of such devices can be different. We are interested in the scheme of the receiving and transmitting parts.
The figure shows a diagram of a real device that transmits commands over a power network. The executive part of the device controls the brightness of the lamp, i.e., is a dimmer.
Let's consider the scheme in more detail. Transformer T1 and diode bridge D1-D4 provide power to the device. Node R8 \ R11, diode D6 and transistor Q1 provide formatting of the signal indicating the minimum voltage in the mains (frequency 100 Hz). Buttons S1-S3 are used for local control of the dimmer operation: they change the brightness of the lamp, allow you to save this parameter by default, as well as the rise and fall time of the lamp. The LED LED displays the dimmer operating modes and the fact of receiving signals. The remaining LEDs display the brightness of the lamp and the dimming time.
Resistors R11 and R12 form a voltage divider and are used to set the "sensitivity" of the receiving part of the device. By changing the resistance ratios of these resistors, it is possible to influence the response of the device to both interference and the useful signal.
The T2 communication transformer is used for galvanic isolation of the receiving and transmitting parts of the device, and also transmits high-frequency signals to the power network of the building.
The transmitting part is made on the transistor Q2 and one of the windings of the transformer T2. Pay attention to the zener diode D5 - it is he who protects the transistor junction from breakdown during short-term high-voltage interference in the network.
The receiving part is somewhat more complicated: one of the windings of the transformer T2, together with the parallel oscillatory circuit L1 \ C2, form a complex circuit of the receiving path. Diodes D8 and D9 protect the microcontroller input from the voltage limit. Thanks to these diodes, the voltage cannot exceed the value of the supply voltage (in our case, 5 Volts) and cannot become negative below minus 0.3-0.5 Volts.
The process of receiving signals is carried out as follows. Polling buttons and working with indications do not have any special features. Therefore, I will not describe their work.
The receive subroutine is waiting for a current zero crossing signal. Upon the occurrence of this event, the analog comparator polling procedure is started, which lasts about 250 microseconds. If no signals were received, then the subroutine starts its work from the very beginning.
When any signal is received (the comparator has given a logical unit at its output), the procedure for analyzing the received signal is launched: for a certain time, the comparator is polled for the presence of a long signal. If the received signal has the required duration, then the received signal is recognized as reliable. After that, the procedure for receiving the required number of bits of data transmitted by the remote device is started.
Having received all the data, they are analyzed for the fact of coincidence with the checksum received in the same package. If the data is received reliably, then the command is recognized as reliable and executed. Otherwise, the received data is ignored and the program is re-executed.
The process of transmitting signals to the network is also completely performed by the microcontroller. If it is necessary to transfer data, the subroutine waits for the starting condition: receiving a current zero crossing signal. Having received this signal, a pause of 80-100 microseconds is maintained, after which a burst of pulses of the required frequency and duration is transmitted to the power network. High-frequency signals pass almost without loss through a small capacitance of high-voltage capacitor C1 into the network. Packets of the required frequency are formed using a hardware PWM generator available in this microcontroller. As experiments have shown, the most optimal signal transmission frequency lies in the range of 90-120 kHz. These frequencies are allowed for use without the need for registration with the relevant regulatory authorities both in Russia and Europe. (CENELEC standard)
And now the answer to the most frequently received question: what is the communication range between such devices? The answer is simple: the communication range is affected by many factors: the quality of the power lines, the presence of "twists" and mounting boxes, the type of load and its power ...
From practice: in a small city, on a power line that feeds 30-50 private houses, in the morning and afternoon (when less electrical appliances are used), the communication range is much higher than in a large city with a hundred apartments on one phase.
I will also answer the second common question: how to increase the communication range? To do this, you can increase the power of the signal transmitted to the power network, as well as improve the receiving part of the device.
The power amplifier can be made on a common TDA2030 or TDA2003 chip (although the parameters declared by the manufacturer are different, but they work well).
The receiving part is more difficult to refine:
- add an input amplifier and AGC;
- add narrow band filters at the input of the device. The simplest solution is this: a serial loop tuned to the desired frequency.
Relay-regulators are shunting and non-shunting.
1) The simplest and cheapest RRs are shunt RRs. Their principle of operation is as follows - when the set voltage amplitude is exceeded, the phases of the generator are shunted (shorted) to each other for a short circuit. In other words, we drive a car and constantly keep full throttle, and we regulate the speed not by dropping the gas, but by pressing the brake. Absurd isn't it? And that's exactly how the shunt relay-regulator works. Naturally, such a block is not very reliable. The increased heat dissipation of the unit itself, wires and connectors often leads to melting, short circuit and failure of the entire circuit, from the generator-relay-regulator to the battery and fuse box. For such a block, the greater the load of consumers, the better. Since in this case, the shunt circuits are less likely to come into operation. And vice versa, without connected consumers, all the power of the generator will be dissipated in the shunt circuit, which will lead to an early failure of the unit. At the moment when the power of the generator is much higher than the total power of the consumers, the bypass circuits are constantly in operation. At the output of such a relay-regulator, the voltage has the form of a sawtooth shape (as in the figure), and not constant as it should be. Such a voltage charges the battery much more slowly and even the light dims with increasing engine speed. I think many have observed such a picture - this is the shunt relay-regulator. Of the advantages of such a block is its low cost and ease of manufacture. Overwhelmingly, the electronic circuits of these particular blocks wander around the forums. Often factories put such blocks on snowmobiles with combined electrical equipment, when both direct and alternating voltage are present on the device for different consumers.
2) Well, the second type of PP is not shunting. I will not describe the circuitry of these blocks, I will simply say that the principle of regulation is based on turning off the output voltage from the RR when the set amplitude is exceeded. When the voltage returns (decreases) to normal, it turns on, and so thousands and even tens of thousands of times per second. In this way, a high voltage stability is achieved, the shape of which tends to a flat horizontal line (see figure). From the generator is taken exactly as much as required by consumers. Heat dissipation is much less, and hence the reliability of such a unit is higher. The advantages of such a block are obvious.
