65 nanometers is the next goal of the Zelenograd plant Angstrem-T, which will cost 300-350 million euros. The company has already submitted an application for a preferential loan for the modernization of production technologies to Vnesheconombank (VEB), Vedomosti reported this week with reference to the chairman of the board of directors of the plant, Leonid Reiman. Now Angstrem-T is preparing to launch a production line for microcircuits with a 90nm topology. Payments on the previous VEB loan, for which it was purchased, will begin in mid-2017.
Beijing crashes Wall Street
Key American indices marked the first days of the New Year with a record drop; billionaire George Soros has already warned that the world is facing a repeat of the 2008 crisis.
The first Russian consumer processor Baikal-T1, priced at $60, is being launched into mass production
The Baikal Electronics company promises to launch into industrial production the Russian Baikal-T1 processor costing about $60 at the beginning of 2016. The devices will be in demand if the government creates this demand, market participants say.
MTS and Ericsson will jointly develop and implement 5G in Russia
Mobile TeleSystems PJSC and Ericsson have entered into cooperation agreements in the development and implementation of 5G technology in Russia. In pilot projects, including during the 2018 World Cup, MTS intends to test the developments of the Swedish vendor. At the beginning of next year, the operator will begin a dialogue with the Ministry of Telecom and Mass Communications on the formation of technical requirements for the fifth generation of mobile communications.
Sergey Chemezov: Rostec is already one of the ten largest engineering corporations in the world
The head of Rostec, Sergei Chemezov, in an interview with RBC, answered pressing questions: about the Platon system, the problems and prospects of AVTOVAZ, the interests of the State Corporation in the pharmaceutical business, spoke about international cooperation in the context of sanctions pressure, import substitution, reorganization, development strategy and new opportunities in difficult times.
Rostec is “fencing itself” and encroaching on the laurels of Samsung and General Electric
The Supervisory Board of Rostec approved the “Development Strategy until 2025”. The main objectives are to increase the share of high-tech civilian products and catch up with General Electric and Samsung in key financial indicators.
Probably, everyone who has ever listened to an SDR receiver or transceiver could not remain indifferent to its reception, and especially to the convenience that manifests itself in the fact that stations on the range can not only be heard, but also seen. An overview of the range in the panorama of the SDR transceiver allows you to quickly and visually find stations in the reception band, which significantly speeds up the search for correspondents during contests, and during everyday work on the air. Using the “waterfall”, the history of signals on the band is visually tracked and you can easily switch to an interesting correspondent. In addition, the panorama itself shows us the frequency response of the received stations, their band and width of radiation, which allows us to quickly find a free area on the range to call other radio amateurs.
This is only if we talk about the visual part of SDR, but we also shouldn’t forget about signal processing, both for reception and transmission. Full control of the width and everything that is in the reception band. With the correct selection of the necessary parameters in the settings menu items, the transmit signal also sounds great.
But there is one circumstance: in order to make SDR work, you need additional devices: a computer itself with a high-quality sound card, on which the main signal processing takes place, and a good monitor with a high screen resolution. Naturally, you need appropriate software for it and for the SDR transceiver, which is not cheap. All this already entails certain specific requirements for the computer knowledge of the radio amateur. Which is not always, and not everyone has, unfortunately.
There is one more drawback. If this is not noticeable during reception, then during transmission, due to the specific processing of the sound signal in the computer, there is a significant signal delay of more than 150 ms, which completely eliminates the normal operation of self-monitoring in all types of radiation. The only thing that can save you is an additional control receiver or a friend who also has an SDR transceiver that will record the received signal.
Nowadays, with the advent of the generation of affordable microprocessors from STM, it is possible to develop devices that can partially replace some of the basic functions of large computers. Namely, DSP audio processing and transceiver control, as well as graphical display of information on the transceiver display.
As a result, the main components of such a transceiver are eliminate the need for an external computer
. But at the same time, as on an external computer, a convenient service for controlling the transceiver is maintained, various modes of recording signals, both for reception and transmission, with subsequent playback of recordings through headphones or on the air during transmission, saving the necessary information on an external SD- card, which is displayed on its own large display with a wide viewing band, as well as DSP processing and signal generation with all major types of radiation. Such transceivers provide high-quality signal reception, high steepness filters with smooth adjustable boundaries, and an automatic Notch filter. They use multi-band graphic equalizers, compressors, reverbs for transmission, and most importantly, they obtain a minimum delay time. With an external synthesizer, transceiver controllers easily work with analog SDRs. These modern transceivers widely use HiQSDR and HiQSDR-mini 2.0 radio paths, which are controlled by a separate SPI bus, or through a DSP board via the main SPI bus with a minimum of connecting wires.
A few years ago, the production of SDR transceivers began, operating on the principle of direct conversion of a radio frequency signal to an audio IF, in which a simplified (compared to the classical circuit) radio channel board and a specialized computer are located in one housing. The main emphasis here is on software. The main cost of the finished product is also determined by the cost of the software. Flex and Sun SDR equipment are built on exactly this principle.
Currently, the principle of signal processing based on DSP methods has moved to the next stage of its evolution. A new method has appeared for direct digitization of a signal from an antenna, followed by direct generation of a signal from a digital signal, which allows you to get rid of almost all types of problems inherent in both classic and SDR technologies with signal processing hardware.
Radio receivers and transceivers with direct signal digitization have the abbreviation DDC (for Digital Down-Converter). The reverse conversion from digital to analog is abbreviated DUC (for Digital Up-Converter). We are talking about digital signal conversion using a software method. It should be noted right away that the abbreviation SDR (Software Define Radio) - software defined radio - is only a general definition of a class of signal processing technologies, which includes DDC architecture as one of the methods.
Already today, with the advent of a generation of affordable microprocessors, it has become possible to develop devices that can partially replace some of the basic functions of large computers. Namely, DSP audio processing and transceiver control, as well as graphical display of information on the transceiver display. In the DDC architecture, the entire spectrum of signals from 0 Hz to frequencies that the ADC chip can process is instantly digitized. The most modern ADC chips today can operate in a band up to 1 GHz, but their cost today is still very high. At the same time, the most popular and relatively cheap ADC chips digitize the spectrum with a band from 0 Hz to 60...100 MHz, which is quite suitable for amateur radio tasks. After digitizing the signal spectrum in the 0 Hz - 30...60 MHz band, a very large digital data stream is obtained at the output of the ADC chip, which is further processed by high-speed FPGA chips. They implement the DDC/DUC algorithm in software, i.e. digital down or up converter.
A digital downconverter samples the spectrum of the required band and transfers it to a computer for processing - i.e. a digital stream of significantly lower bandwidth and speed is created. The computer carries out software processing of the stream using DSP methods and final demodulation of the signal.
In practice, there is very rarely a need to work with the entire spectrum of signals in the 0 Hz - 30...60 MHz band. The maximum bands that we need for processing are 10...50 kHz for demodulation of AM, FM signals and 3...5 kHz for SSB signals.
This most advanced method of signal processing was implemented in amateur radio transceivers TULIP-DSP and the domestic analogue - Tulip-DDC/DUC.
A similar principle of signal formation is used in transceivers from one well-known company, which began producing new models back in 2015. A fragment of the block diagram of such a transceiver is presented below.
If earlier, a few years ago, even in such advanced transceivers as ICOM IC-756Pro3 and IC-7600, the method of sequential spectrum scanning was used and the process of updating the picture was noticeable - i.e. fast scanning, now observation and signal processing occurs in a complex, in parallel, since frequency tuning occurs instantly using the software method. Due to the fact that a large frequency section of 30...60 MHz is digitized at once, without losing tuning to the current radio station, it becomes possible to see what is happening in the neighboring section of the spectrum. Moreover, by calling the second virtual receiver, you can simultaneously hear what is being said on one and the second band. But two receivers is not the limit. It is possible to call three, five, ten... as many receivers as you like. By mixing their sound in a certain way, you are aware of what is happening on the bands. And the “cloud” graphics will allow you to quickly select the desired station.
The same applies to spectrum display. In practice, it is rare that the entire 30...60 MHz section is needed at once. If necessary, you can relatively easily select the second, third, fourth and generally as many small streams as necessary from the general digital stream and transfer them to the computer, thereby creating several reception channels simultaneously. This method implements two, three, or as many “virtual receivers” as needed throughout the entire digitization band. For example, we create a separate panorama for the 40-meter range, a separate one for the 20-meter range and for other ranges..., we place them on a separate monitor and now we have the opportunity to observe in real time the conditions of passage in the areas we have chosen.
On the one hand, the presence of mirror stripes is a disadvantage. Since the concept of DD refers to the entire spectrum of digitization, it is possible to significantly unload the ADC input by paying attention to the input circuits of the receiver, which are best made high-quality and tunable. An alternative option is to use low-pass filters with a cutoff frequency of half the clock frequency or bandpass filters in the input circuits. They can further attenuate strong out-of-band signals that are far from the operating band. In this case, the opportunity to review the entire digitization range is lost. Such pre-selection methods are justified if you plan to use the DDC receiver in conjunction with large antennas or in areas with difficult interference conditions.
On the other hand, this drawback provides a technological opportunity by simple means to implement not only reception on the HF band, but also on VHF and even on the VHF bands. It is only necessary to make replaceable bandpass filters with LNAs with bands equal to half the clock frequency.
For example, some DDC receivers have a switchable filter for the MW-LW range, and one of the WiNRADiO DDC receivers and the Perseus DDC receiver have flexibly configurable narrowband filters.
Just 20 years ago we couldn’t even dream of anything like this, when the panoramic attachment to the transceiver was 2 times the size of the transceiver itself and cost 5-10 times more. There is no need to talk about service and quality. SDR technology, which appeared in the early 2000s, made it possible to look at and hear the broadcast in a completely different way. We saw a real live broadcast! Not a static “frozen” picture after a slow scan, but a live broadcast in real time.
If, in order to see a reduced panorama of other bands in the first SDR transceivers with hardware signal conversion, it is necessary to have a separate receiving path for each range, then in the receiving path, made using modern DDC technology, both any part of the range and the entire range are available , and at the same time in parallel with individual sections of it. The implementation of all these possibilities is possible only thanks to the methods of DSP and direct signal digitization.
Regarding amateur radio topics, one of the most popular functions at present and in the near future is spatial signal selection and phase noise suppression methods. Today there is a phase method for signal selection and noise suppression, implemented in hardware. In addition, using mathematical algorithms, any functions for subtracting interfering and adding useful signals generated by a pair, four or more ADCs can be easily implemented.
With the use of these modern developments, it became possible to remotely control the transceiver and remote work on the air. Modern methods of information transmission are capable of transmitting fairly large data streams with virtually no losses. The overall flow of information from/to the transceiver is quite small. Using the IP stack, it becomes possible to use the transceiver as a network segment even without using a computer. By installing the transceiver outside a big city in a fairly quiet area, you can have access to radio broadcasts without leaving your apartment. By organizing guest access to the transceiver, you give your friends the opportunity to work on the air. Another useful function used by special services is the ability to record the entire radio broadcast, or specified pieces of radio broadcast, onto a computer hard drive with delayed processing. This function allows you to quickly carry out statistical processing of signals, search and monitor target signals, and also perform many operations that the average user is not supposed to know about.
You can select the radios you are interested in in
Let us first deal with the general principles of analog-to-digital conversion. The basic principle of digitizing any signals is very simple and is shown in Fig. 17.1, A. At some points in time t 1, t 2, t 3 we take the instantaneous value of an analog signal and, as it were, apply some measure to it, a ruler, graduated on a binary scale. A regular ruler would contain large divisions (meters), each divided into ten parts (decimeters), each of which would also be divided into ten parts (centimeters), etc. A binary ruler would contain divisions divided in half, then in half again, etc. .d. – how much resolution is enough. If the entire length of such a ruler is, say, 2.56 m, and the smallest division is 1 cm (that is, we can measure its length with an accuracy of no worse than 1 cm, more precisely, even half of it), then there will be exactly such divisions 256 and can be represented as a binary number of 1 byte or 8 bits.
Rice. 17.1 . Digitization of analog signals:
A- the basic principle;
b– explanation of the theorem Kotelnikov–Nyquist
Nothing will change if we measure not length, but voltage or resistance, only the meaning of the concept “ruler” will be slightly different. This is how we get successive samples of the signal magnitude x 1 , x 2 , x 3. Moreover, note that with the selected resolution and number of digits, we can measure a value no more than a certain value that corresponds to the maximum number, in this case 255. Otherwise, we will have to either increase the number of digits (lengthen the ruler), or change the resolution towards worsening (stretch her). All of the above is the essence of the operation of an analog-to-digital converter - ADC.
In Fig. 17.1, A The graph demonstrates this process for the case if we measure some quantity that changes over time. If measurements are made regularly at a known frequency (called the sampling frequency or quantization frequency), then only the signal values can be recorded. If the task is to then restore the original signal from the recorded values, then, knowing the sampling frequency and the accepted scale (i.e., what value of the physical quantity corresponds to the maximum number in the accepted range of binary numbers), we can always restore the original signal by simply plotting the points on the graph and connecting them with a smooth line.
But what do we lose? Look at fig. 17.1, b, which illustrates the famous Kotelnikov theorem (as usual, abroad it goes by a different name - Nyquist, in fact, they both formulated it independently of each other). This figure shows a sinusoid of the limiting frequency, which we can still reconstruct by having an array of points obtained at the sampling frequency f d. Since in the formula for sinusoidal oscillation A sin(2π ft) there are two independent coefficients ( A– amplitude and f– frequency), then in order to uniquely restore the appearance of the graph, you need at least two points for each period, i.e. The sampling frequency must be at least twice the highest frequency in the spectrum of the original analog signal. This is one of the common formulations of the Kotelnikov–Nyquist theorem.
Try to draw another sinusoid yourself without a phase shift, passing through the points indicated on the graph, and you will see that this is impossible. At the same time, you can draw any number of different sinusoids passing through these points if their frequency is an integer number of times higher than the sampling frequency f d. In total, these sinusoids, or harmonics (i.e., terms of the signal expansion into a Fourier series - see chapter 5), will give a signal of any complex shape, but they cannot be restored, and if such harmonics present in the original signal, they will disappear forever.
Only harmonic components with frequencies below the limit are restored unambiguously. That is, the digitization process is equivalent to the action of a low-pass filter with a rectangular cut of the characteristic at a frequency equal to exactly half the sampling frequency.
Now about the reverse transformation. In essence, no digital-to-analog conversion occurs in the DACs that we will consider here; we simply express the binary number in the form of a proportional voltage value, i.e., from a theoretical point of view, we are engaged in just a scale conversion. The entire analog scale is divided into quanta - gradations corresponding to the resolution of our binary “ruler”. If the maximum value of the signal is, for example, 2.56 V, then with an eight-bit code we will get a quantum of 10 mV, and we do not know and cannot find out what happens to the signal between these values, as well as in the time intervals between samples . If we take a series of consecutive samples of a certain signal, for example, those shown in Fig. 17.1, A, then we will end up with a step-by-step pattern shown in Fig. 17.2.
Rice. 17.2 . Restoring the digitized signal from Fig. 17.1, a
If you compare the graphs in Fig. 17.1, A and in Fig. 17.2, you will see that the second graph represents the first, to put it mildly, very approximately. In order to increase the degree of reliability of the resulting curve, you should, firstly, take samples more often, and secondly, increase the bit depth. Then the steps will become smaller and smaller, and it is hoped that at some high enough resolution, both in time and in quantization, the curve will eventually become indistinguishable from a continuous analog line.
Notes in the margins
Obviously, in the case of audio signals, additional smoothing, for example, using a low-pass filter, is simply not required here, since it will only worsen the picture, cutting off high frequencies even more. In addition, all sorts of analog amplifiers themselves will smooth out the signal, and the human senses will also work as a filter. So the presence of steps in itself is unimportant if they are small enough, but a sharp drop in the frequency response above a certain frequency has a fatal effect on the sound quality. Many people with a good ear for music claim that they can accurately distinguish CD-quality digital sound (sampled at a frequency of 44.1 kHz, i.e., with a cutoff at a frequency obviously higher than the level of perception of human hearing, and with a number of gradations at least 65 thousand for the entire range) from real analog sound, for example, from a vinyl record or tape. For this reason, high-quality digital audio is recorded at much higher sampling rates than formally required, for example, 192 and even 256 kHz, and then it becomes truly indistinguishable from the original. True, directly digitized sound is only recorded on discs in the Audio CD format, and for almost all other formats compression is used - compression using special algorithms. If it were not for compression, neither the capacity of modern media nor the speed of computer networks would be enough for recording: just one minute of stereo sound with CD-quality parameters takes up about 10 MB on the media, you can check it yourself.
We will not delve into the specifics of sampling analog periodic signals, since this is a very vast area in modern engineering, primarily associated with digitization, storage, replication and playback of sound and video, and this should, at a minimum, be a separate book. . For our purposes, the information presented is sufficient, and now we will move directly to the task of digitizing and inversely converting a single signal value.
We will start from the end, i.e. with digital-to-analog converters - you will see why below. We will assume that at the input we have numbers in binary form - it does not matter whether it is the result of digitizing some real signal or a synthesized code. We need to convert it to an analog voltage level in accordance with the selected scale.
The simplest DAC is a decimal or hexadecimal decoder-distributor, like 561ID1. If we apply a four-bit code to it, then at the output we will get a logical one for each code value on a separate pin. By connecting a line of LEDs to the outputs of such a decoder, we get a strip (scale) indicator, which, with a resolution of 10 or 16 steps over the entire range, will show the level of a certain value. Moreover, very often for practice such a relatively crude indicator, replacing pointer instruments, is quite sufficient. Special microcircuits are produced to control such discrete scale indicators, which allow the value to be shown not as a separate point or strip, but as a luminous column. There are also microcircuits that can control not discrete, but linear vacuum indicators. There is even a K1003PP1 microcircuit (analogue of UAA180), which converts an analog value (voltage) directly into a control signal for a bar indicator. A rather impressive design can be obtained if in the thermometer circuit according to Fig. 13.3 or 13.4, replace the indicating head with such a microcircuit and a scale indicator - like a full-fledged imitation of a traditional thermometer!
Such a primitive DAC has two drawbacks: firstly, increasing its resolution beyond 16–20 gradations is unrealistic, because then there will be too many outputs. But most importantly, it is intended for the narrow task of visualizing a digital value and is helpless outside this area. A converter that performs the function according to Fig. 1 would be much more widely used. 17.2, i.e., outputting an analog voltage proportional to the code at the input.
A “dumb” method of obtaining such a voltage would consist of the following modification of the method with a decoder-distributor of type 561ID1. To do this, you need to build a divider from a chain of identical resistors, connect it to a reference voltage source and switch the taps of this divider with keys controlled from the decoder-distributor. For a two- or three-digit code, you can use those described in chapter 15 multiplexers type 561KP1 and 561KP2. But for a larger number of bits, such a direct conversion DAC turns into a completely monstrous design. An eight-bit code would require 256 resistors (exactly identical!), the same number of keys and a decoder with the same number of outputs, but an eight-bit code is a rather rough “ruler”, its resolution does not exceed a quarter of a percent. Therefore, in practice, this method is used to build an ADC, and not a DAC (because, despite its complexity, it has one unique property, see below), and here we will not even draw such a circuit.
Let's consider one of the most common methods, which allows for code-voltage conversion without using such monstrous structures. In Fig. 17.3, A shows an implementation option for a DAC based on an op-amp with switched resistors in the feedback circuit. As switching keys, you can use, for example, small-sized electronic relays of the 293 series, i.e., the same type that we used in the design of the thermostat in Fig. 12.9, or specialized keys from the 590 series. However, to implement a switching contact it would be necessary to install two such keys for each digit, therefore the 561 series provides a special chip 561KTZ (CD4066), which contains four identical keys that work exactly as shown in the given diagram.
Rice. 17.3. Schemes used in construction DAC :
a– two-bit DAC with negative output;
b– R–2R chain of arbitrary length;
V– DAC with positive output
These keys are bidirectional, but their outputs work differently. The output, which is designated OUT/IN (in the domestic version, usually simply “Output”), in one state is switched with another input/output, in another it is simply disabled, as usual. And the output, designated IN/OUT (in the domestic version simply “Input”), in one state is connected to the first input, but when the key is broken, it does not “hang in the air” like the first, but is grounded. Thus, if you apply a logical one signal to the key control input of 561KTZ, then the IN/OUT pin of the appropriately connected key is switched to the OUT/IN input, and if the control signal is logical zero, then the IN/OUT pin is shorted to ground. , as we need.
Notes in the margins
Note that there is also a 176KT1 microcircuit (CD4016A, there is no analogue to it in the 561 series, but there is an imported version CD4016B with power supply up to 20 V), with which the 561KTZ is often confused - it has the most common double-sided keys, without grounding. And, despite the fact that these microcircuits are described exhaustively in the classic reference book, erroneous information is often provided in online amateur reference books about 561 KTZ. Of course, it is unlikely that you will have to build such DACs yourself, but just in case, you should take into account that the resistance of the 561KTZ switch, as well as more modern modifications (1561 KTZ or CD4066B), is quite high, on the order of hundreds of ohms, which can affect the accuracy. Although for practical purposes in a number of circuits (but not in the one under consideration!) what is more important is not the absolute value of the resistance, but the difference in this parameter between the keys, which, according to reference books, does not exceed 5 Ohms.
Let us finally consider how this scheme works. To better understand the principles, I drew only a two-digit version. Two digits are four gradations, i.e. the output voltage of the op-amp must take 4 values with equal intervals, in this case these voltages are equal to 0, as well as 1/4, 1/2 and 3/4 of the reference voltage U op. How does this happen?
Let us first consider the circuit in its initial state, when the code at the key control inputs has the value “00”. Since both are lower in the resistor circuit 2R in the initial state are connected to “ground”, i.e. connected in parallel, then their total resistance is equal to R .
Then the top resistor in the circuit R and these two resistors form a divider, the voltage across which is equal to exactly half of U op. Resistor parallel to divider 2R does not participate in voltage division. The keys are open, the resistor chain is disconnected from the op-amp input; and its output will have a voltage equal to 0.
Let the code now take the value “01”. In this case, a resistor with a nominal value 2R the least significant digit (lowest in the circuit) is switched to the amplifier input. For the resistor chain itself R ‑2R it doesn’t matter whether this resistor is connected to ground or to the input, because the op-amp input potential is equal to the same ground potential. Thus, to the input of the op-amp through a resistance with a nominal value 2R a current will flow, the magnitude of which will be equal to the voltage at its input ( U op/2, as we found out), divided by the value of this resistor ( 2R). The total current value will be U op /4R, and this current will create on the feedback resistor of the op-amp, the resistance of which is equal to R, voltage drop equal to U op/4. You can think differently - consider an inverting amplifier with a gain of 0.5, which is determined by the resistance ratio R /2R, and input voltage U op/2. In total, the output of the entire circuit will be voltage U op/4 (but with the opposite sign, since the amplifier is inverting).
Let the code now take the value "10". Then everything is even simpler - voltage is connected to the input of the op-amp U op through the upper resistor 2R. The gain is the same (0.5), so the output will be voltage U op/2. The most difficult case is when the code takes the value “11” and both resistors are connected. In this case, the op-amp should be considered as an analog adder (see. chapter 12, rice. 12.5, A). The output voltage will be determined by the sum of the currents through the resistors 2R, multiplied by the feedback resistance value R, i.e. it will be equal to ( U op / 2 R + U op /4R)R, or just 3 U op /4.
I looked at this example in such detail to clearly demonstrate the properties of the chain R‑2R. The method of its construction with any number of links is shown in Fig. 17.3, b. Extreme resistors 2R connected in parallel and in total they give resistance R, therefore the next link turns out to consist of the same denominations in 2R and in total it will also give R etc. No matter how long the chain is made, it will divide the input voltage in a binary ratio: at the very right end of the chain according to the diagram there will be voltage U op, on the next branch U op/2, next U op/4, etc.
Therefore, using just two types of resistors, differing by exactly two times, it is possible to build a DAC of basically any capacity. So, an eight-bit DAC will contain 16 resistors and 8 switches (if switched, as in 561KTZ), not counting the feedback resistor, which for clarity we also had equal to R, but can be of any convenient denomination. In integrated DACs, this resistor is often not installed at all in advance, but the corresponding pins are placed outside, so that you can easily obtain any output voltage scale. For example, if in our circuit we make this resistor equal to 1.33 R, then at the output we get voltages equal to U op , 2U op /3, U op/3 and 0.
True, the inconvenience in such a simple circuit is that the output voltages will have the opposite sign, but this problem is easily solved. In Fig. 17.3, V shows the simplest version of a DAC with a “normal” positive output. I leave it to the reader to analyze the operation of this circuit on their own - it is, in fact, even simpler than the inverting version. The disadvantage of this option compared to the inverting one is that the gain is not adjustable, and the scale will be determined only by the value U op. But this drawback can be easily corrected by slightly complicating the circuit. Such DACs are also called multiplying .
Notes in the margins
I will not consider commercial DAC integrated circuits (for example, 572PA1) based on this principle, because in general they work the same, and DACs by themselves, without being used as part of an ADC, are rarely needed. Nevertheless, let's say a few words about the problems associated with metrology. It is clear that it is not easy to obtain exact resistor values when manufacturing a chip of such a DAC, so in practice the absolute values R can have quite a wide spread. Their denominations are carefully coordinated with each other using laser adjustment. The switches' own resistance can also have a great influence on the operation of the circuit, especially in the higher bits, where the currents are greater than in the lower ones. In the integrated version, they even make these keys different - in the higher digits they put more powerful ones with less resistance. And if you try to make a homemade DAC based on the previously mentioned 516KTZ, then the value of R should be tens of kilo-ohms, no less, otherwise the switches will begin to introduce too much error.
Another point is related to obtaining a stable reference voltage, since this directly affects the conversion accuracy, and for absolutely all ADCs and DACs, as we will see later. Currently, advances in electronics have made it possible to almost forget about this problem - all major manufacturers produce reference voltage sources that make it possible to achieve stability of about 16 bits (i.e., 65,536 signal gradations). In addition, you can always contrive to construct a diagram so that the measurements become relative.
The speed of the DAC of the type considered is mainly determined by the speed of the switches and the type of logic used, and in the case of CMOS switches it is not too high - about the same as that of conventional CMOS elements.
Most integrated DACs are built using the described principle of summing weighted currents or voltages. Another class of digital-to-analog converters are integrating DACs that serve to convert time-varying quantities. These DACs ideally allow you to immediately get a truly analog, continuous signal with no signs of aliasing.
The range of analog-to-digital converters is significantly larger than that of DACs. However, all the variety of their types can be reduced to three varieties: these are parallel ADCs, successive approximation ADCs and integrating ADCs. Let's look at them in order.
Read the article to the end, as it expresses a different opinion from the traditional one regarding recording studio switching.
Electromagnetic interference.
All wires can pick up electromagnetic interference. This may cause noise or humming. This interference may come from any other music or household equipment.
To reduce the influence of such interference, a balanced connection was invented.
Balanced connection.
It is called balanced because the audio signal travels through two different wires in addition to ground. These two wires carry the same signal, except that the signal on one of the wires is inverted. The main purpose of signal inversion is to eliminate noise at the receiving end. These actions help the balanced system to better withstand outside noise.
This is what a balanced wire looks like.
Not a balanced connection.
Unlike balanced, unbalanced signal transmission requires only one wire (central in the cable) and one ground (GND). Since only one wire transmits the signal, this connection is called unbalanced. This method is susceptible to interference that will be transmitted along with the original signal. As the cable length increases, the noise level will also increase. This is why most studio engineers or live performance technicians use balanced connections for very long cables.
This is what a non-balanced wire looks like.
What do you need to remember and know about balanced and unbalanced connections?
Most older analog or digital devices and instruments made before the 90s had unbalanced audio outputs. Connecting such a device or instrument with a balanced wire to the balanced input on a mixer is unacceptable! When you connect an unbalanced audio output with a balanced wire, antiphase occurs; if you have stereo sound (two outputs are used), or if you have mono sound (one output is used), a poor-sounding, quiet sound will appear.
For example, Roland TR 808 or Roland MC 505 have unbalanced audio outputs.
And vice versa, if you connect a device or synthesizer with a balanced audio output with a non-balanced wire. The sound will not lose its depth and antiphase will not arise.
Traditional opinion.
A balanced connection is considered to be more professional as it eliminates noise. It is recommended to use balanced switching in recording studios. For unbalanced instruments or instruments, various unbalanced to balanced converters are available. However...
Unconventional opinion, experienced sound engineers.
Many studio sound engineers do not recognize the balanced connection, believing that it creates problems in the overall mix precisely because the audio signal is inverted. Inversion is when one wire carries a hot, positive signal (plus), and the other carries exactly the same signal, but a cold, negative signal (minus). By overlapping each other, the two signals subtract the noise, and we do not hear the noise. But... not only noise can be subtracted. Along with noise, important sound color, warmth, and richness may disappear. And there is always a danger that with stereo sound, antiphases will begin to arise. Then, in the overall mix, failure and impoverishment of entire musical parts is possible.
The balanced connection was not actually invented for studio recording. It is ideal for live concerts where long cables are involved.
Professional recording studios are usually well shielded around the perimeter from external noise. To combat noise, voltage rectifiers are installed. Devices in racks are placed in such a way that electromagnetic interference is not created. Audio cables are located at a relative distance from various power supplies. With proper placement of studio equipment, using balanced wires becomes unnecessary. The only exceptions are studio microphones with a long wire and a balanced output (XLR).
A conclusion that you don’t have to agree with.
Connect all devices in your studio only with unbalanced wires. Firstly, you certainly won’t be mistaken about which device you have has a balanced output and which one is not balanced. Secondly, you'll get a fat, old-school sound to the mix.
Fight noise by dealing with electromagnetic interference in other ways. Eliminate interference yourself “manually” (screen, move, ground) and do not trust the balanced connection to do this.
Note: I wrote about some methods of dealing with noise