비디오 시리즈
Precision Labs 시리즈: 오디오 기본 사항
이 비디오 시리즈를 통해 최신 전자 제품의 오디오의 기본 사항에 대해 알아보세요. 이 비디오는 신규 및 숙련된 오디오 엔지니어를 대상으로 합니다. TI의 오디오 전문가가 제공하는 다양한 오디오 직렬 인터페이스 형식, 오디오 사양, 잡음 유형 등을 알아보세요.
Audio amplifier basics
발표자
리소스
Hello, and welcome to our TI Precision Labs Audio Fundamentals video. Today, we'll be discussing the basics of audio amplifiers. All amplifiers take in low power signals in the form of alternating current and amplify its power by means of increasing its current and voltage. This amplified signal is then driven into the designer's desired electrical load.
In the case of an audio amplifier, this load is often the speaker. Modern amplifiers fall into four main classes, class A, class B, class AB, and class D. We will discuss these later in this video. At the core of all amplifiers, since the first practical amplifier was invented, is some kind of electronic valve that can modulate the flow of electricity.
The three main forms of electronic valves, or triodes, that are used in amplifiers are the vacuum tube, bipolar junction transistor, BJT, and the metal oxide semiconductor field effect transistor, MOSFET. For the scope of this video, we will be discussing modern amplifiers, which are typically implemented using BJTs or MOSFETs.
These three electronic valves all behave in slightly different manners, but the basic function is the same. As we discussed, all triodes can be thought of as electricity valves. In much the same way that a physical valve can control the flow of tens or hundreds of PSI of water with a small control signal, a vacuum tube, BJT, or MOSFET can control the flow of tens or even hundreds of volts of electricity with a small control signal.
When selecting an audio amplifier, there are some key specifications that can help you make the decision that is best for your application. The first specification that you should consider is the maximum continuous output power. This defines how loud the amplifier can be. The output power needed will depend greatly on the application the amplifier will be used.
This specification can be found on the product page for an amplifier on TI.com, or it can be found in the electrical characteristics section of the amplifiers data sheet. It is often specified for different load impedance's, distortion levels, and supply voltage configurations. Efficiency should also be considered. Efficiency describes the ratio of the power that is turned into sound to the total power that is consumed by the amplifier.
The less efficient an amplifier is, the more heat that will be dissipated in that amplifier. This may require a large heat sink, and for some applications such as phones, tablets, and laptops, a large heat sink is not an option. The efficiency specifications of an amplifier can be found in the electrical characteristics section of that device's data sheet, and this specification is often given for different load impedances and output power levels.
Total harmonic distortion plus noise, or THD plus N, is the final key spec we will discuss in this video. Ideally, the output of an audio amplifier would be an exact replica of the input signal with some scaling factor. However, in reality, all amplifiers will add a certain amount of noise and harmonic distortion to the amplified signal.
THD plus N describes the ratio of signal to noise and distortion in the output of that amplifier. This ratio is given in either a percent or a decimal value. 1% is an industry standard level for an acceptable amount of distortion. However, a keen ear can hear distortion lower than 1%. This specification can be found in the electrical characteristics section of the device's data sheet and will often be specified for different output power levels, frequencies, and measurement bandwidth.
Next, we will discuss some of the common amplifier classes. The first amplifier we will discuss is the class A. Class A amplifiers are a single transistor that is externally biased at some non-zero DC current. If we look at the graph on the right, which shows the generic relation between the base-2 emitter voltage, Vbe, and the collector current, Ic, of the BJT.
With proper design, the transistor in a class A amplifier can be biased such that the relation between the collector current and the base-2 emitter voltage is almost perfectly linear across the input voltage of that amplifier. This effect gives class A the lowest distortion of all the amplifiers we will discuss. A side effect of this is that the bias current, Icq, of the amplifier is non-zero, which subsequently means that the amplifier is dissipating substantial power, even when no input signal is applied.
It is for this reason that class A amplifiers are inefficient. Low distortion makes class A amplifiers a favorite amongst audiophiles. However, for most practical applications, the class A does not meet modern efficiency requirements.
Next, let's discuss the class B amplifier. Class B intends to solve the efficiency problems of class A. Class B amplifiers utilize a complimentary pair of transistors where each transistor conducts one half of the input signal. In this configuration, the input signal itself is responsible for turning the transistors on. Another way to say this is that the transistors are biased at 0 Icq.
This effect greatly improves the efficiency of class B relative to class A. However, since the base-2 emitter connection of a BJT behaves as a diode, the input signal needs to rise to 1 diode voltage above the output before any significant amount of current is drawn by the NPN transistor.
Similarly, the input needs to dip 1 diode voltage below the output before the PNP transistor conducts. This introduces an effect called crossover distortion. Because of this, class B amplifiers are rarely the practical choice.
The class AB amplifier attempts to mitigate the effects of this crossover distortion. As we discussed previously, the base-2 emitter on a BJT behaves as a diode. So if we use diodes to bias the base of each transistor at 1 diode voltage above and 1 diode voltage below the input signal, then we are able to ignore the region of the Vbe versus Ic curve that would introduce crossover distortion.
By doing this, we can ensure that each transistor conducts almost entirely in the linear region of the Vbe versus Ic curve. A side effect of this is that the transistors will be conducting a small current at no input. This effect, paired with the resistive losses in the biasing diodes and resistors, will decrease the class AB amplifiers efficiency relative to class B. However, this sacrificed efficiency comes with the benefit of far less distortion.
And next, we'll explore the class D amplifier. The primary difference between this amplifier and the other classes is that the input signal is pulse-width modulated before it is driven to the transistors, and once amplified, it can be low-pass filtered and leave the original signal. The class D amplifier is able to achieve up to 90% efficiency because the PWM signal will modulate each MOSFET either fully on or fully off, treating it like a switch.
Let's consider an ideal switch as an example. When the switch is open, no current flows through it, and therefore, no power is dissipated in the switch. Similarly, when the switch is closed, there is no voltage across it, and therefore, no power is dissipated in the switch. However, in practice, leakage currents, non-zero RDS(on) values, and non-infinite slew rates will cause some power to be dissipated in the MOSFETs of a class D amplifier.
Class D amplifiers are able to practically achieve 85% efficiency at sub 1% THD plus N. This makes them a great choice for small, portable, battery powered speakers such as those in phones, tablets, and laptops.
However, the high frequency PWM introduces a couple of problems, electromagnetic interference, which can be snubbed by including an LC low-pass filter on the output. And due to the high dV/dt currents of the PWM, the class D amplifier is not suitable to drive capacitive loads such as piezoelectro devices, as the repetitive current spikes will damage the FETs.
And finally, one of the recent innovations in the world of audio is the TI Smart Amp. TI Smart Amps are essentially class D amplifiers with the addition of a DSP, DAC, and IV sensing. This allows for direct digital input, real time signal processing, and speaker protection algorithms, all to be integrated into one package.
This innovation allows designers to safely drive higher volumes, higher bass, and higher quality audio through modern compact speakers such as those that can be found in cell phones, laptops, and tablets. Smart Amps are available in 4 watt to 175 watt versions and can be tuned and configured with their TI provided software such as PurePath Console.
We discussed a lot of amplifiers in this video. Let's have a short review. Class A amplifiers offer great audio quality, but are inefficient. This makes them most suited for audiophiles. Class B was a logical next step in amplifier design as a demonstration that efficiency could be improved, but class B is impractical for audio use due to the large amount of distortion.
Class AB are a good compromise between audio quality and efficiency. Class AB was, and still is, a top choice for high power audio systems. Class D offers the highest efficiency and low distortion in the audio band. This makes it perfectly suited for compact and battery powered applications. And Smart Amps took the efficiency of class D and built on top of that.
Modern TI Smart Amps can be tuned for the best possible audio performance out of modern compact speakers. Smart Amps combine a class D amplifier, a DSP, and a DAC into a single package, all of which is tunable through the TI provided software for each family of products. I hope we have helped you determine which amplifier class is best for your application. For more technical information and product catalog, please visit TI.com. Thank you.