Power amplifier parameter index (below)

Transient Intermodulation Distortion (Transient IntermodulaTIon DistorTIon) is called TIM distortion. It was unclear when the author was discovered, but the TIM measurement method was not published publicly until the 1970s. Since transient intermodulation distortion is closely related to negative feedback, it is necessary to start with negative feedback when discussing transient intermodulation distortion. Negative feedback (Negative Feedback) is a simple and practical control technology widely used in various engineering technology fields. Negative feedback is originally a part of the closed loop control (Close Loop Control) system in the control technology, but Because it is widely used, it is often used as a synonym for closed-loop control. Negative feedback is actually a natural law that is ubiquitous in people ’s daily lives. For example, when we drive a car, if we find that the car deviates from the driving route, we will twist the steering wheel in the opposite direction to make the car Drive back to the correct route. Here our eyes are acting as a negative feedback channel, which is responsible for feeding back the output value (the direction in which the car has to drive) to the excavator (brain), and then the controller compares the output value and the set value (correct direction) to each other Minus), and then send a correction signal (twist the steering wheel) to correct it according to the error after comparison. It can be seen that the function of negative feedback is to invert the output value (become a negative number), and then feed it back to the input, and set Subtract the fixed value to get the error signal, and then the controller will make corrections according to the size of the error.

In the electronic amplification circuit, due to the symmetry of the parts, the temperature change, the interference of noise and other various reasons, while reading the number is amplified, various distortions are inevitably added, and the negative feedback is Can effectively reduce these distortions. To give a simple example, if the amplifier adds a distorted square wave signal when amplifying a sine wave signal, this sine plus square wave signal will be inverted by the negative return line, and then fed to the input, It is subtracted from the original sine wave to make the original signal amplitude smaller. It also contains an opposite square wave. This new signal will also be added to a distorted square wave signal again when passing through the amplifier. An opposite square wave, so that the positive and negative square waves will cancel each other, so that the output signal contains only a sine wave, which significantly reduces distortion. However, the disadvantage of negative feedback is also obvious, because negative feedback makes the input signal and the output signal of the feedback subtracted, which reduces the signal level. If the output signal is to be sold, the signal level is reduced. If the output is to be output When the signal is amplified to a sufficient strength, the amplification rate (gain) of the amplifier must be increased. Fortunately, this is not difficult, especially for the transistor machine. If we increase the amount of negative feedback to reduce the output signal to the same level as the input signal level, that is, there is no amplification at all, this amplifier circuit has a special name, called a buffer amplifier (Buffer Amplifier). Although the signal is not amplified, it is because amplifiers generally have high input impedance and low output impedance. Therefore, buffer amplifiers are often used for impedance matching.

Since negative feedback can effectively reduce the distortion, why would it cause transient intermodulation distortion? It turned out that the problem was in time, among which the transistor machine was the most serious. Compared with the vacuum tube, the transistor has the advantages of durability, small size, light weight and high magnification. Its disadvantage is that the operating characteristics are unstable, and it is susceptible to distortion and even runaway due to temperature and other factors. One solution is to use deep negative feedback up to about 50 to 60dB. Anyway, the amplification rate of the transistor is very high, and it does not matter to sacrifice some. Because of the large depth of negative feedback, the distortion is greatly reduced, so the transistor machine can easily obtain superb technical specifications. However, the trouble is also caused. In order to reduce the high-frequency parasitic oscillation caused by the deep negative feedback, the transistor amplifier generally needs to add a small capacitor between the base and collector of the pre-driver transistor to make the high frequency band The phase is slightly lagging, which is called the lag price or the subtraction price. However, no matter how small the capacitor is, it always takes a certain time to charge. When the input signal contains a high-speed transient pulse, the small capacitor is too late to charge. For a moment, the line is in a state without negative feedback. Since the input signal is not subtracted from the negative return signal, the signal is too strong, and these excessive signals will cause an instantaneous overload of the line (Overload). Because the negative feedback of the transistor is large, the signal is too strong, and it often reaches tens or even hundreds of times. As a result, the output signal is clipped. This is transient intermodulation distortion, which is also called "atomic particle" sound because it occurs most in transistor lines.

By the way, this kind of negative feedback time delay problem is also often encountered in industrial control systems, called the pure delay (Dead Time) problem, the cause of which is mostly because the installation location of the sensor (Sensor) is too far. For example, in a constant temperature water heater, the blast detection is installed at a location away from the heat. The result is that when the detector senses that the water temperature is sufficient, the water temperature near the heater has already been overheated. The result of such a control is necessarily a large swing of water temperature between overheating and subcooling, which is called control overshoot or system oscillation. Pure delay is still a big problem that plagues automatic control technology. There have been thousands of papers about solutions from the 1950s to the present, but there is still no simple and effective way.

Although the current time delay of the negative return output is not easy to deal with, it is not impossible to solve it. We can simply let it appear, or even if it appears, it will not cause too much damage. There are many methods, for example, only a small amount of large Loop negative feedback, so that there is a negative feedback delay in the name, and the input signal is not too strong. The reduced amount of negative feedback is replaced by a local negative return that only spans 1 amplification stage. The local negative return is short in path and fast in time, and is not easy to induce transient intermodulation distortion. The vacuum tube works stably. It is not necessary to suppress the distortion with a large depth of negative feedback. Moreover, most of the distortion is even harmonic distortion that people love to listen to, so the amplifier does not have the so-called "atomic particle" sound. As for other methods used to prevent transient intermodulation distortion in line design, they will not be introduced one by one because they involve more boring theories.

In addition to preventing transient intermodulation distortion in line design, audiophiles can also take another measure to reduce transient intermodulation distortion, that is, try to use various shielding and filtering measures to reduce various high-frequency interference signals into the amplifier. Although many of these signals are radio frequency interference inaudible to the human ear, because of their high frequency, transient intermodulation distortion is easily induced, overloading the input stage, and preventing the music signal from being amplified normally.

Conversion rate

Transient intermodulation distortion is not only caused by the time delay of the negative return of the amplifier's large loop, but the speed of the amplifier is not fast enough. It is also an important reason. If the amplifier is fast enough, even under the same negative return conditions, transient intermodulation The degree of distortion can also be reduced. The speed of the amplifier is a popular description. The correct statement should refer to the transient response capability of the amplifier (Transient Response). In control theory, transient response and frequency response are two major methods of measuring system performance. Their advantage is that they do not require a detailed understanding of the detailed mathematical model of the entire system. They only need to estimate the system's response curve to a specific input signal based on the system's response curve to a specific input signal to estimate the system's characteristics and thus compensate Or improve. On the contrary, if we know the mathematical model of a system, we can estimate the response mode of the system without testing.

For systems with low accuracy requirements, we can selectively adopt the transient response method or frequency response method to evaluate system performance, and for systems with high requirements, both must be considered. The signals commonly used for transient testing are the unit step function (Step Signal) and the unit pulse function (Impulse). For convenience, the amplifier uses a special form of the former: square wave /. An ideal square wave contains a very high-speed voltage rising edge and falling edge, which is very suitable for testing the transient response of the amplifier.

The response speed of an amplifier is generally measured by the voltage conversion rate (Slew Rate, referred to as "slewing rate" in Taiwan). It is defined as the amplitude of voltage rise in 1 microsecond. If it is measured with a square wave, it is the time required for the voltage to rise from the valley to the peak. The unit is V / us. The larger the value, the higher the transient response. The conversion rate of performance amplifiers can generally be above 25V / us.

The easiest way to improve transient response is to select parts with good high-frequency characteristics. It can also be improved with proper loop negative feedback. This seems to be a contradictory approach, but it is not the case. Transient intermodulation distortion only occurs when the signal speed exceeds the amplifier's transient response capability.

In addition to transient intermodulation distortion, signals that are too fast will also produce another kind of distortion, called ringing, which is essentially the same. When the input signal speed is fast and the amplitude is small, the ringing phenomenon first appears. Only when the speed of this signal is fast to a certain degree will transient intermodulation distortion appear. However, when the signal speed is fast and the amplitude is large, the ringing If it does not happen, it will enter the state of transient intermodulation distortion. The signals that are most likely to cause ringing are all kinds of high-speed but low-frequency high-frequency interference noise, which is one of the reasons why sound equipment must have perfect anti-interference measures.

Interface Intermodulation Distortion

Interface intermodulation distortion is a relatively new and rarely mentioned amplifier specification. The same as the damping coefficient mentioned below, in addition to the amplifier circuit, it also has a lot to do with the speaker. So before introducing these two specifications, let's briefly talk about the characteristics of the speaker in this regard.

Most of the current audio speakers are moving coil horns that use the principle of electric type. Its structure includes a permanent magnet used to generate a magnetic field and a one-person voice coil. In terms of structure, the dynamic coil speaker belongs to a special form of DC motor, because the voice coil only needs to move back and forth instead of rotating, so there is no need to use the carbon brush and commutator (commonly known as "copper head") common on DC motors.

Both AC motors and DC motors are reversible, that is, they can be used as generators under certain conditions. The structure of a DC motor is no different from that of a DC generator. In particular, a permanent-magnet DC motor can generate a certain voltage on its terminals as long as it can rotate its shaft. For a dynamic speaker, as long as we press the diaphragm with our hands, a voltage will be generated on the terminals, the size depends on the speed and amplitude of the pressing.

Due to the effects of loss and non-linearization, it is impossible for the speaker to utilize all the power output by the amplifier and there will be residual power generation. In addition, due to the mechanical inertia of the diaphragm, excess power will be generated in the voice coil. The problem caused by the former is interface intermodulation distortion, while the latter will make the low frequency control of the speaker worse.

Interface intermodulation distortion is related to speaker internal resistance and negative return line. When the electrical energy output from the amplifier cannot be fully converted into mechanical energy, the excess electrical energy will certainly generate an additional back emf in the speaker coil (Back emf), which will be fed back from the speaker wire to the output of the amplifier, and then The magnitude of the internal resistance of the amplifier forms a voltage, which will be fed back to the input by the negative return line, and will be merged with the input signal. The mid-low frequency sound is turbid, and the analytical power and sense of hierarchy are greatly reduced.

To reduce the intermodulation distortion of the interface, the key point is to reduce the negative feedback and the internal resistance of the amplifier (that is, increase the damping coefficient). Many Hi-End transistor amplifiers are designed using this principle. In addition to this, the two-wire connection is also an alternative way to improve, because the separate high and low lines make the low-frequency back-EMF will not affect the high-frequency signal, thereby improving sound quality.

Damping Factor

The ratio of the damping coefficient of the speaker impedance to the amplifier output impedance. As the name implies, the resistance coefficient is the degree to which the physical quantity that changes in a certain process is suppressed. For the speaker, what is to be suppressed is the inertial vibration of the speaker diaphragm in the absence of electrical signal input, which is simply a braking action. The diaphragm of the speaker cannot be braked with mechanical damping, all that can be used is electromagnetic damping. This method requires the system to be in the generator state as much as possible.

The previous discussion mentioned that the speaker can easily enter the state of the generator. When the input electric reading disappears, the speaker diaphragm is still not vibrating due to inertia. This vibration will produce an induced voltage in the voice coil. If the output resistance of the amplifier is low at this time, it is equivalent to connecting a small resistor in parallel to the speaker terminal. The induced voltage on the voice coil will drive a relatively high voltage. The large amount of current flowing through the internal resistance post office of the amplifier means that the speaker becomes the power supply at this moment, and the power output stage circuit of the amplifier becomes the load. According to the law of electromagnetic induction, this current is generated by the vibration of the voice coil in the magnetic field of the permanent magnet, so this voice coil current must generate a force opposite to the vibration direction to counteract the vibration. The smaller the internal resistance of the amplifier, the greater the current and the stronger the effect of counteracting inertial vibration. Because the energy of this current will be consumed as heat on the resistor, this braking method is called "dynamic braking" (Dynamic Bracking) in the motor control technology. The loudspeaker has the largest amplitude when replaying low frequencies, and the most serious inertial vibration caused. If it is not suppressed, the low frequency control will be poor, lacking strength, flexibility and layering, but excessive suppression will make the sound dry.

Due to the existence of the coil resistance of the output train, the damping coefficient is extremely limited. On the contrary, the transistor machine can easily increase the damping coefficient to a few dozens or even hundreds by using multiple tubes and connections. However, it can be different from the requirement of a resistance factor, which results in a variety of different sound combinations between different speakers and amplifiers.

For an amplifier that uses more than half of the negative feedback, the damping coefficient is not the only tool that will brake the speaker, because the speaker's inertial vibration current will generate a certain value of voltage when it flows through the amplifier's output internal resistance. , The negative return line feeds it back to the input immediately, so that the amplifier line thinks that a distortion voltage should not appear, and immediately generates an inverted signal to resist. This is one of the most powerful electric motor braking methods, called "reverse connection braking" (Plugging). But it is also the least used method, because a motor reversed suddenly will produce a huge mechanical impact and damage the machine, but the speaker is originally designed as a device that continuously moves back and forth, so this method is theoretically no problem, However, in practice, there are often problems, and the trouble comes from negative return.

The speaker is not a microphone, and the voltage generated by the vibration of the diaphragm will not be as accurate as the microphone. Therefore, the cancellation voltage generated by the amplifier cannot be completely equal to the vibration and the direction is opposite. The result is that the suppression process is unstable, and the low frequency is not reduced smoothly and quickly. This process is actually very similar to the process of interface intermodulation distortion. The low-frequency control power of some atomic particle amplifiers is not as good as that of amplifiers, which is why.

There are some other specifications for measuring amplifier performance. This article only mentions the concerns of more enthusiasts, plus the frequently disputed specifications. The author is by no means an expert, just because I often need to take into account the technical principles of motors, electronics, and even machinery at the same time. When I have a headache, I find that there are many technologies or problems in the fever field. What has long been understood and recognized by people in the technical field is not esoteric and mysterious. It is just that different industries have different interpretation methods and are confusing. This article should try some specific metaphors to explain and distinguish some often confusing Specifications. Hope that some non-engineering enthusiasts can have a clearer concept.

With the development of amplifier technology, it is believed that it is difficult to make special technological breakthroughs in circuit design and material application. High-quality equipment can only be relied on with a careful and conscientious attitude. It has been neglected in the past. A large number of trivial technical specifications have been improved bit by bit. Out of proportion. The so-called flat and pretty are just relatively speaking. The technology is built with money. There are many so-called high-tech military technologies that use only those scholars from universities and research institutions in various countries. In order to improve their academic status, The theoretical research results published on the channel have no secret at all, but the difficulty is only in the pre-research, design, test, production and quality assurance process technology. Like Hi-Dnd equipment, the cost invested is often astronomical It may only be a process that can be used by transforming old machines alone.

Follow WeChat

Interesting and informative information and technical dry goods

Download Audiophile APP

Create your own personal electronic circle

Follow the audiophile class

Lock the latest course activities and technical live broadcast

related suggestion

Power amplifier parameter index (on)

It has been more than 120 years since Edison invented the phonograph in 1877. From that year ...

Published on 2006-04-17 23:58 • 523 times read