Classes of amplifiers

The system of letters, or classes of amplifiers , is a classification used to characterize the amplifying, it assigns a letter for each diagram of electronic amplifier.

These diagrams are defined by the relation between the form of the entry signal and that of exit, but also by the duration during which an active component is used during the amplification of a signal. This duration is measured in degrees of a sinusoidal Signal test applied to the entry of the amplifier, 360 degrees representing a complete cycle. In practice the class of amplification is determined by the Polarization electronics components of the stage of exit of the amplifier, or the calculation of the point of rest.

Introduction

At the origin, the classes of amplifiers allowed to know the way in which was polarized the amplifying component (class has, B, AB and C). Thereafter, with the arrival of the amplifiers with cutting, other classes were added to allow to include the technical solutions used there to amplify the signal (class D, E,…).

Required to polarize an amplifying component (tube or transistor) comes from their strong non-linearity as well as theirs unilaterality while running. For example, a bipolar transistor starts has to lead only to the top of a tension threshold (see the characteristic $V_ {Be} = F (I_b)$ on figure opposite). If it is wanted that the transistor amplifies (the curve $I_c = \ I_b$ beta), it is thus necessary to impose on the transistor a tension higher than this tension threshold. The output voltage is fixed by the line of load $I_c=f (V_ {this})$ which is not represented on graphics Ci against because dependant on the assembly used.

In order to stage has these defects, one “polarizes” the amplifying component. Polarization fixes the state at rest (when any signal is not applied to him) transistor. The choice of not of rest influences much the behavior of the amplifying component during its operation. The classes of amplifications allow to know the way in which it was polarized and thus to have information on the characteristics of the total amplifier.

For the linear amplifiers, each class defines the proportion of the entry signal which is used by each active component to arrive at the amplified signal (see figure opposite), which is also given by the angle of conduction has : ; Classify a: the totality of the entry signal (100%) is used ( has = 360°). ; Classify b: half of the signal (50%) is used ( has = 180°). ; Classify AB: More half but not totality of the signal (50-100%) is used (180° < has < 360°). ; Classify C: Less than half (0-50%) of the signal is used (0 < has < 180°).

For the amplifiers “with cutting”, the transistor is used in commutation. In these assemblies, the transistor either is blocked (no current crosses it) or saturated (the tension has these terminals is quasi null). The angle of conduction of a transistor in assemblies with cutting is thus null. For the amplifiers with cutting, the classes are useful has to distinguish technologies used and either the angle from conduction.

In the continuation of the article, in order to simplify the illustrations, a bipolar Transistor is used like amplifying component, but it can be replaced by a Field-effect transistor or a tube. The assembly used to illustrate this article is an assembly known as “transmitting commun run”. Always for reasons of simplification, it will be represented without its circuit of polarization and its condensers of connection C₁ and C₂. Lastly, the frequency of the signal which one applies in entry will be supposed sufficiently large to consider that the condenser of decoupling C₃ “court-circuite” R₄ resistance.

The characteristics of the bipolar transistor used to illustrate the article are idealized characteristics. On these figures, the characteristics of the transistor its represented in blue, the various signals in red and the line of load of the assembly in green In practice, the active components are not also linear and these non-linearities are at the origin of distortions in the output signal. In order to reduce these distortions, there is generally recourse has a form or another of Contre-réaction.

Classify has

In an amplifier of class has, the active components are always in conduction. These amplifiers amplify all the entry signal, thus limiting the distortions on the output signal. They do not have a good Rendement: they dissipate a constant power some is the amplitude of the entry signal. Thus, these amplifiers reach their best performance when that the amplitude of the output signal east in extreme cases of what the amplifier can provide. The best performance of an amplifier of class has depends on the topology of the assembly used: thus, the theoretical best performance of these amplifier is of 50% in the case of a connection by transformer or “ push-pull ”, of 25% in the case of a direct connection and between 6% and 25% for a capacitive connection.

For an assembly transmitting commun run has, a polarization in class has means that the electrode biases $V_ {be_0}$ and $V_ {ce_0}$ were selected so that the amplifier does not saturate when one applies an entry of maximum amplitude to him.

Because of their poor yield, the amplifiers of class has are generally used to make amplifiers of small power. If one needs a class has strong power, the energy losses will become very important. For each Watt delivered with the load, the amplifier will dissipate some, at best, another Watt. The class has important power will need food of very strong power but also of broad squanderers in order to will dissipate lost energy.

The amplifiers of class has are generally used to carry out stages preamplifiers, audio amplifiers, amplifiers high frequencies has broad band as well as oscillators high frequencies. One of the traditional applications of the class has is the Paire differential, which is exceptionally linear, and constitutes the basic stage of many more complex circuits, like much of audio amplifiers and the majority of the operational amplifier . Although the majority of the audio amplifiers use a stage of exit in class B. the amplifiers of class B are generally used to produce basic amplifiers and intermediate frequency. In this case, they are used in configurations known as “ push-pull ”. The assemblies push-pull have two transistors: to amplify the negative part of the signal and a second for his positive part, the two transistors functioning of “classifies B”. The totality of the signal being amplified, the assemblies push-pull have a rate of distortion lower than the amplifiers classifies B “basic” while keeping a good output.

For an assembly has transmitting commun run, a polarization classifies B of it means that the electrode bias $V_ {be_0}$ was selected equal to the tension threshold of conduction of the transistor (see figure opposite). Thus, any negative signal affixed at the base of the transistor will bring it in lower part of its threshold of conduction and will not be amplified. A contrario , any positive signal will bring the transistor in the linear zone of its characteristic $I_b=f (V_ {Be})$ and will thus be amplified.

The assemblies push-pull can suffer from a discontinuity of signal at the place where the two halves of signal exit of each transistor meet. This phenomenon is called the Distorsion of crossing.

One finds amplifiers of class B of the type push-pull in the stage of exit of the continuous amplifier used in: linear feedback loops, the generator of functions, amplifiers in integrated circuits, like in the stage of exit of the majority of the audio amplifier

Classify AB

Class AB is a compromise between the class has and class b: the point of rest of the amplifier is between that of an amplifier of class has and that of an amplifier of class B. Such a method of polarization makes it possible class AB to function as a class has for the signals of low amplitudes then to behave like an amplifier of class B for the signals of strong amplitudes.

The principal disadvantage of this type of assemblies occurs when signals of strong amplitude are amplified: part of the signal is amplified by two transistors (zone of operation in class A) while the remainder of the signal is amplified by only one transistor (zone of operation in class B). Thus, the profit while running of the assembly is not constant during a “cycle” of amplification. This variation of the profit while running generates distortions high-frequencies at the time of the passage between the zone or two components amplifies the signal and that or only one composing amplifies it.

For the amplifiers has tubes, one adds sometimes two subclasses with the classe AB: class AB1, without appearance of current roasts and class AB2 with current roasts. These two subclasses have direction only for the tube amplifiers. Indeed, there exists always current basic in a bipolar transistor, and any variation of tension on the grid of a field-effect transistor will require a current in order to charge or discharge the capacity grid-source inherent in this technology from transistor.

Classify C

The amplifiers of class C amplify less than 50% of the entry signal. The rate of distortion is important, but their theoretical best performance lies between 78% and 100% following the angle of conduction of the amplifier. Certain applications as the Mégaphone S can tolerate an important rate of distortion.

For an assembly has transmitting commun run, a polarization classifies C of them means that the electrode bias $V_ {be_0}$ was selected lower than the tension threshold of conduction of the transistor. Thus, the signal will not be amplified as much as it does not carry the tension base-transmitter of the transistor or above of its limiting tension of conduction (see figure opposite).

The amplifiers of class C are more usually used in the transmitting radio operator, where the rate of distortion can be reduced thanks to the use of a load granted in the amplifier. The entry signal is used to make commutate the active component of passer by with blocked. This tension pulsation creates a current through a circuit tuned. The tuned circuit resounds only for one frequency band, thus eliminating the nondesired frequencies. The desired signal (a sinusoid) is then transmitted to the granted load. If the frequency band to be amplified is not very important, this kind of system functions correctly. The last residual harmonics can be eliminated using a filter.

The amplifiers of class C are used to produce ultrasonic amplifiers, high frequencies selective and microwaves as well as oscillators high frequencies.

Classify D

An amplifier of class D is an amplifier of which all the components of power are used like switches: the amplifying components either are blocked (no current cross them) or saturated (the tension has their terminals is quasi null).

Such stages of exit are used, for example, in the generating of impulses. However, most of the time, this term applies to amplifiers which must reproduce signals whose band-width is weaker than the frequency of commutation. These amplifiers use the modulation of width of impulse, the Modulation of density of impulse or more advanced forms of modulation like the modulation Sigma delta. The entry signal is converted into a series of impulses whose median value is directly proportional to the amplitude of the signal at the moment considered. The frequency of the impulses is generally at least ten times higher than the more high frequency which one wants to reproduce. The spectrum of the exit of such an amplifier contains nondesired frequencies (for example, the quench frequency and its Harmonique S) which must be eliminated by filtering.

The main advantage of the amplifiers of class D is their output. The impulses of exits being of constant amplitude, the switches (in general of MOSFET S, but the tubes or the bipolar transistors are also used) work in mode blocked or saturated rather than with linear mode. That means that apart from their openings or closings, the active components in saturated mode dissipate energy very little. The instantaneous power dissipated in a transistor, which is the product of the current by the tension, weak at the time of the states is saturated and blocked because one or the other of the two units is almost null at the time of these states. The little of losses generated by this type of amplifier makes it possible to use squanderers and food smaller.

The amplifiers of class D are usually used for the control of the electrical motors (one speaks then about Hacheur or Onduleur). They are also used as audio amplifiers. The relative difficulty to obtain an audio good quality made that they were generally used in applications where sound quality is not a determining factor like the miniature audio systems. In the years 2000, one starts to see appearing on the market of the amplifiers of class D of high-quality. Several manufacturers propose modules ready to be used in an integrated audio system. One of the first and prolific sector of application was the Subwoofer S of strong powers for the car. The subwoofers having generally a band-width limited to less than 150 Hz, the frequency of commutation of the amplifier does not need to be as high as for an audio amplifier “normal”. Their output ranges between 80% and 95%.

Note: The letter D is used to indicate this type of amplifier because it is the letter which comes after C , that is not an abbreviation of “digital”. The amplifiers of class D and E are sometimes qualified, wrongly, of numerical. This confusion comes from the form of wave of the exit which resembles a numerical train of impulses. In fact, these amplifiers function on the principle of the modulation of width of impulse. A signal of digital display would be in modulation of coded impulse.

The classes E and F

The amplifiers of class E and F are high-output amplifiers. They are generally used to amplify the frequencies radio. The principle of the amplifiers of class E was published for the first time in 1975 by Nathan O. Sokal and Alan D. Sokal. The amplifiers of class F take again the same principle as the amplifiers of class E but with a load granted to a frequency and some of its Harmonique S, while the load of the amplifiers of class E is granted only for the fundamental frequency. The basic concept of these amplifiers is that the model of the active components, a transistor for example, is a linear combination of two parts: (1) a theoretical switch “perfect”, and (2) a complex network of parasitic elements which are attached to him (capacities, inductances and resistances). After this decomposition, it is possible to eliminate the losses from each component:

the “perfect” switch must become busy at the time of one zero of tension and must be blocked at the time of one zero of current. Thus, the switch lets pass from the current or has a tension on its terminals nonnull, but never both at the same time. The dissipated power being equal to the product current voltage, it becomes null. That can be carried out by modifying the phase (condenser) or the component continues (resistance) entry signal of the transistor.
the imaginary part of the impedance of the parasitic elements can be modified, one by one, by associating them with another passive component which has a combined impedance, leaving thus only the real part of the complex impedance .

In theory, the only remaining losses are due to the real part of the parasitic impedances of the system, which cannot be eliminated. This class of amplifiers is only used for the radioelectric frequency band, where the analysis of the amplifier is done in the frequential field and not in the tension field/running. This class is still divided into subclasses according to the harmonics which are taken into account during commutation to the zero of tension (Switching Zero-Voltage in English) and commutation to the zero of current ( Zero-Current Switching in English). One speaks then about class E/F2, F^-1 class,….

The figure below watch the diagram of an amplifier of class E/F using the principles exposed higher to increase its output.

The switch periodically is opened and closed at the frequency of operation. In general, the cyclic Rapport is of 50%. The winds of stop has a strong impedance in order to provide a function of power source. The other passive components are selected in order to satisfy the following conditions: (1) the terminal voltage of the switch is null at the time of its closing; (2) the temporal derivative of the tension is null when the switch becomes busy. Moreover, Ls and Cs form a resonant circuit at the frequency of operation.

In practice, the switch is replaced by a transistor which functions in mode saturated (passer by) or blocked. The theoretical yield of an amplifier of class E using of the perfect components is of 100%. However, the side circuits suffer from many imperfections preventing them from reaching such an output. These imperfections include the finished switching times, internal resistance and the tension of nonnull saturation of the transistor, as well as the losses in the passive components used high frequency. In practice, the output is of 60% for the frequencies going from 1 to 2 GHz.

This class of amplifiers is especially studied to amplify the rectangular wave trains, like those used for the data transmission numerical. The rectangular impulse or wave trains have specific needs because of their spectral contents. Indeed, they require a faithful reproduction of the high frequencies present in their rising and downward faces, without adding artefact (goings beyond, oscillations, etc) during the process of amplification. Moreover, the low frequencies must also be taken into account because them variations which the load (a line of transmission in general) can generate.

Classify G

The class G was invented in 1976 by Hitachi in order to increase the output of the amplifiers. These amplifiers have several different buses of tensions and they use only the bus of tension having the weakest tension making it possible to provide the wished power in exit. While thus acting, these amplifiers limit the power “lost” in the transistors of exit increasing from there even the output.

The amplifiers of class G are composed of an amplifier of class B as of one or more amplifiers of class C. the amplifier of class B functions all the time and uses, when it is the only one has to function, the bus of tension having low value. The amplifiers of class C use one of the buses of more raised tension and are activated only according to the requirement in charging voltage.

The principal disadvantage of the amplifiers of class G is the complexity brought by the presence of several buses of tension. Thus, the number of buses of tension used is generally of two and seldom exceeds three. They consist of an amplifier of class has having a weak maximum output current and of an amplifier of class B. the amplifier of class has provides the current for the signals of weak power then leaves the step to the amplifier of class B when the current required by the load becomes too important.

Special classes

There exists a multitude of another classes that described in the preceding paragraph. These new classes are generally based on the classes has, B, C and D. the goal of these new classes can be either to improve the characteristics of the “traditional” classes by modifying the topology of the basic assemblies or so that a manufacturer can be different from the others. This last type of classes generally do not reflect any idea of revolutionary amplification, but are rather used at ends of marketing . That is often expressed by the fact that the name of the class is a registered trademark or under copyright.

Classify T

The class T is a registered trademark of TriPath which manufactures audio amplifiers. This new class T is in fact an amplifier of class D standard functioning at a frequency of 650 Khz, with a system of modulation owner .

Random links: