What Is a Class-D Amplifier?

Class D amplifiers are amplifiers that drive speakers by controlling the ON / OFF of the switching unit. Class D amplifiers were first proposed in 1958, and have become increasingly popular in recent years. Class D amplifiers have developed tremendously in the past few generations. System designers have greatly improved the durability of the system and improved its audio quality.

Class D amplifier

Class D amplifiers are amplifiers that drive speakers by controlling the ON / OFF of the switching unit. Class D amplifiers were first proposed in 1958, and have become increasingly popular in recent years. Class D amplifiers have developed tremendously in the past few generations. System designers have greatly improved the durability of the system and improved its audio quality.

Chinese name

Class D amplifier

By

Control switch unit

Types of

Amplifier

Time

1958

Class D amplifiers were first proposed in 1958, and have become increasingly popular in recent years.

In conventional transistor amplifiers, the output stage contains transistors that provide instantaneous continuous output current. Many possible types of amplifiers for audio systems include Class A amplifiers, Class AB amplifiers, and Class B amplifiers. Compared to class d amplifier designs, even the most efficient linear output stage consumes a lot of power in their output stage. This difference makes class d amplifiers have significant advantages in many applications because low power consumption generates less heat, saves printed circuit board (pcb) area and cost, and can extend battery life in portable systems.

The linear amplifier output stage is connected directly to the speaker (in some cases via a capacitor). If the output stage uses a bipolar junction transistor (bjt), they usually work in a linear mode with a large collector-emitter voltage. The output stage can also be used

Although the class D amplifier's low power consumption is a powerful driving force for its audio applications, there are some important issues that design engineers need to consider, including:

Output transistor size selection, output stage protection, sound quality, modulation method, anti-electromagnetic interference (EMI), lc filter design, system cost

Output transistor size selection

The output transistor size is selected to reduce power consumption over a wide range of signal conditioning. When conducting large ids, ensure that vds is small, and the on-resistance (ron) of the output transistor is required to be small (typically 0.1 0.2). But this requires a large transistor with a large gate capacitance (cg). The power consumption of the switched capacitor gate drive circuit is cv2f, where c is the capacitance, v is the voltage change during charging, and f is the switching frequency. If the capacitance or frequency is too high, this "switching loss" will be too large, so there is a practical upper limit. Therefore, the choice of transistor size is a compromise between minimizing ids × vds losses and minimizing switching losses during conduction. In the case of high output power, power consumption and efficiency are mainly determined by conduction loss, while in the case of low output power, power consumption is mainly determined by switching loss. Power transistor manufacturers try to minimize the ron × cg of their devices to reduce the total power consumption in switching applications, thereby providing flexibility in switching frequency selection.

Output stage protection

The output stage must be protected from many potentially hazardous conditions:

Overheating: Although the power consumption of the output stage of a class d amplifier is lower than that of a linear amplifier, if the amplifier provides very high power for a long time, it will still reach a level that will harm the output transistor. To prevent the danger of overheating, a temperature monitoring control circuit is required. In a simple protection scheme, when the temperature measured by an on-chip sensor exceeds the thermal shutdown safety threshold, the output stage is turned off and remains until it cools down. In addition to a simple binary indication of whether the temperature has exceeded the shutdown threshold, the sensor can provide other temperature information. By measuring the temperature, the control circuit can gradually reduce the volume level, reduce power consumption, and keep the temperature well within the limits, rather than forcing no sound during thermal shutdown.

Output transistor overcurrent: If the output stage and speaker are properly connected, the output transistor will have a low on-resistance state, but if these nodes are not careful to short-circuit with another node or positive and negative power supply, a huge Current. If unchecked, this current can damage transistors or peripheral circuits. Therefore, a current detection output transistor protection circuit is required. In a simple protection scheme, if the output current exceeds a safety threshold, the output stage is turned off. In a more complicated scheme, the current sensor output is fed back into the amplifier in an attempt to limit the output current to a maximum safe level, while allowing the amplifier to operate continuously without shutting down. In this scheme, if the current limit protection is invalid, the last resort is to force shutdown. An effective current limiter also keeps the amplifier safe from transient large transient currents due to speaker resonance.

Undervoltage: Most switching output stage circuits only work properly when the positive supply voltage is high enough. If the supply voltage is too low and undervoltage conditions occur, problems can occur. This problem is usually handled by an undervoltage lockout circuit, which allows the output stage to work only when the supply voltage is greater than the undervoltage lockout threshold.

Output transistor turn-on timing: mh and ml output stage transistors (see Figure 6) have very low on-resistance. Therefore, it is important to avoid the situation where mh and ml are turned on at the same time, because it will generate a low-resistance path from vdd to vss through the transistor, thereby generating a large inrush current. In the best case, the transistor is hot and consumes power; in the worst case, the transistor may be destroyed. The break-before-make control of the transistor prevents inrush current conditions by forcing both transistors to turn off before a transistor is turned on. The time interval when both transistors are off is called non-overlapping time or dead time.

Figure 6. Break-before-make switching of output stage transistors

Note: switching output stage = switching output stage

nonoverlap time = nonoverlap time

on = ON

off = disconnect

Sound quality

In a class d amplifier, several issues must be addressed to achieve good overall sound quality.

"Click": Clicking when the amplifier is on or off is very annoying. Unfortunately, they are easy to introduce into a class d amplifier, unless special attention is paid to the modulator state, output stage timing, and lc filter state when the amplifier is muted or non-muted.

Signal to noise ratio (snr)

In order to avoid the hiss noise caused by the noise floor of the amplifier, snr should usually exceed 90 db for low power amplifiers for portable applications, snr for more than 100 db for medium power designs and 110 db for high power designs. This is achievable for various amplifiers, but specific noise sources must be tracked during amplifier design to ensure a satisfactory overall snr.

Distortion mechanism

Distortion mechanisms include non-linearities in the modulation technique or modulator implementation, and the dead time used by the output stage to address the inrush current problem.

In a class d modulator output pulse width, information including the amplitude of the audio signal is usually encoded. The additional dead time used to prevent the inrush current of the output stage will introduce a non-linear timing error. The distortion generated in the speaker is proportional to the timing error relative to the ideal pulse width. The shortest dead time to avoid shock is often the most advantageous to minimize distortion; for a detailed design method to optimize the distortion performance of the switch output stage, see In-depth reading 2.

Other sources of distortion include: mismatches in the rise and fall times of the output pulses, mismatches in the timing characteristics of the output transistor gate drive circuit, and non-linearity of the lc low-pass filter components.

Power Supply Rejection (psr)

In the circuit shown in Figure 2, power-supply noise is almost directly coupled to the output speaker, with minimal suppression. This happens because the output stage transistor connects the power supply to the low-pass filter through a very low resistance. The filter suppresses high frequency noise, but all audio frequencies pass, including audio noise. For a detailed description of the effects of single-ended and differential switch output stage circuit power supply noise, see Further reading material 3.

Without solving the distortion and power issues, it would be difficult to achieve a psr better than 10 db, or a total harmonic distortion (thd) better than 0.1%. Even worse, thd tends to high-order distortions that are detrimental to sound quality.

Fortunately, there are some good solutions to these problems. Using feedback with high loop gain, as used in many linear amplifier designs, helps a lot. The feedback from the lc filter input will greatly increase the psr and attenuate all non-lc filter distortion sources. The lc filter nonlinearity can be attenuated by the speakers included in the feedback loop. In a well-designed closed-loop class d amplifier, psr>; 60 db and thd <; 0.01% high-fidelity sound quality can be achieved.

But feedback complicates the design of the amplifier because the stability of the loop must be met (a very complex consideration for higher-order designs). Continuous-time analog feedback is also necessary to capture important information about pulse timing errors, so the control loop must include analog circuitry to handle the feedback signal. This can increase die cost in the implementation of integrated circuit amplifiers.

To minimize IC costs, some manufacturers prefer not to use or use the least amount of analog circuitry. Some products use a digital open-loop modulator and an analog-to-digital converter to detect power supply changes and adjust the modulator's behavior to compensate, which can be found in further reading 3. This improves psr but does not resolve any distortion issues. Other digital modulators attempt to precompensate for expected output-stage timing errors, or to correct non-ideal modulators. This will deal with at least some, but not all, sources of distortion. For applications that require less stringent sound quality, these open-loop Class-D amplifiers can be used for processing, but for optimal sound quality, some form of feedback seems necessary.

Modulation technology

Class D amplifier modulators can be implemented in a variety of ways, with extensive related research and intellectual property support. This article only introduces the basic concepts.

All Class D amplifier modulation technologies encode the relevant information of the audio signal into a series of pulses. Generally, the pulse width is related to the amplitude of the audio signal, and the pulse spectrum includes useful audio signal pulses and useless (but unavoidable) high-frequency components. In all schemes, the total integrated high-frequency power is roughly the same, because the total power of the waveform in the time domain is the same, and according to the parseval theorem, the time domain power must be equal to the frequency domain power. However, the energy distribution varies greatly: in some schemes, there are high-energy tones above the low-noise background, while in other schemes, energy is shaped to eliminate high-energy tones, but the noise background is higher.

The most commonly used modulation technique is pulse width modulation (pwm). In principle, pwm compares the input audio signal with a triangular or ramp wave operating at a fixed carrier frequency. This produces a series of pulses at the carrier frequency. During each carrier period, the duty cycle of the pwm pulse is proportional to the amplitude of the audio signal. In the example in Figure 7, the audio input and triangle wave are both centered at 0 v, so for zero input, the duty cycle of the output pulse is 50%. For large positive inputs, the duty cycle is close to 100%, and for large negative inputs, the duty cycle is close to 0%. If the audio amplitude exceeds the amplitude of the triangle wave, full modulation will occur. At this time, the pulse train stops switching and the duty cycle is 0% or 100% in a specific period.

pwm is attractive because it allows 100 db or better audio band snr at a few hundred kilohertz pwm carrier frequency (low enough to limit the switching losses in the output stage). Many pwm modulators are also stable at almost 100% modulation, allowing high output power in principle to reach the overload point. However, pwm has several problems: first, the pwm process adds inherent distortion in many implementations (see in-depth reading 4); second, the resonance of the pwm carrier frequency generates emi in the AM radio band; and finally, The pwm pulse width is very small near full modulation. This causes problems in most switching output stage gate drive circuits, because their drive capabilities are limited and they cannot be switched properly with the extremely fast speed required to reproduce a short pulse width of a few nanoseconds (ns). Therefore, full modulation is often not achieved in pwm-based amplifiers, and the maximum output power that can be achieved is less than the theoretical maximum, that is, only the supply voltage, transistor on-resistance, and speaker impedance are considered.

An alternative to pwm is pulse density modulation (pdm). The number of pulses in a given time window (pulse width) is proportional to the average value of the input audio signal. Its individual pulse width is not arbitrary like pwm, but a "quantization" multiple of the modulator clock period. 1-bit sigma-delta modulation is a form of pdm.

A large amount of high-frequency energy in sigma-delta modulation is distributed over a wide frequency range, rather than concentrated at multiples of the carrier frequency like pwm, so the potential emi advantage of sigma-delta modulation is better than pwm. At the mirror frequency of the pdm sampling clock frequency, energy still exists; but in the typical clock frequency range of 3 mhz to 6 mhz, the image frequency falls outside the audio frequency band and is strongly attenuated by the lc low-pass filter.

Another advantage of sigma-delta modulation is that the minimum pulse width is one sampling clock period, even for signal conditions near full modulation. This simplifies the gate driver design and allows safe operation at theoretical full power. Nevertheless, 1-bit sigma-delta modulation is not often used in class d amplifiers (see further reading 4), because traditional 1-bit modulators can only stabilize to 50% modulation. At least 64 times oversampling is also required to achieve sufficient audio band snr, so the typical output data rate is at least 1 mhz and power efficiency is limited.

Self-oscillating amplifiers have recently been developed, such as the one described in Further Reading 5. This amplifier always includes a feedback loop, which determines the modulator's switching frequency based on the loop characteristics, instead of an externally provided clock. High frequency energy is often flatter than the pwm distribution. Because the feedback can get excellent sound quality, but the loop is self-oscillating, it is difficult to synchronize with any other switching circuit, and it is difficult to connect to a digital audio source without first converting a digital signal to an analog signal.

The full-bridge circuit (see Figure 3) can use "tri-state" modulation to reduce differential emi. In the traditional differential working mode, the output polarity of half-bridge a must be opposite to that of half-bridge b. There are only two differential operating states: output a high and output b low; output a low and output b high. However, there are two other common-mode states where the two half-bridge outputs have the same polarity (both high or low). One of these two common-mode states can be combined with the differential state to produce three-state modulation. The differential input of the lc filter can be positive, zero, or negative. The zero state can be used to indicate a low power level, replacing the switch between the positive and negative states in a two-state scheme. During the zero state, the differential action of the lc filter is very small. Although the common-mode emi is actually increased, the differential emi is reduced. Differential advantage is only applicable at low power levels, because the positive and negative states must still be used to provide high power to the speaker. The changing common-mode voltage level in the three-state modulation scheme is a design challenge for closed-loop amplifiers.

pwm principles and examples

Note: sample audio in = sample audio input

pwm out = pwm output

triangle wave = triangular wave

pwm concept = pwm principle

pwm example = pwm example

sine = sine wave

audio input = Audio input

pulses = pulses

pwm output = pwm output

emi processing

The high-frequency components of a class d amplifier deserve serious consideration. If not properly understood and processed, these components can generate a large amount of emi and interfere with the operation of other devices.

Two types of emi need to be considered: signals radiated into space and signals conducted through speakers and power lines. The class d amplifier modulation scheme determines the baseline spectrum of the conducted emi and radiated emi components. However, some board-level design methods can be used to reduce the emi emitted by a class d amplifier, regardless of its baseline spectrum.

A useful principle is to minimize the loop area carrying high-frequency currents, because the intensity associated with emi is related to the loop area and how close the loop is to other circuits. For example, the entire lc filter (including speaker wiring) should be placed as close as possible and kept close to the amplifier. Current-driven and return traces should be grouped together to minimize loop area (twisted pair wiring for speakers is helpful). Another thing to note is that when the gate capacitance of the output stage transistor switches, a large transient charge is generated. Usually this charge comes from the energy storage capacitor, thus forming a current loop containing two capacitors. By minimizing the loop area, the effects of transient emi in the loop can be reduced, meaning that the storage capacitor should be charged as close to the transistor as possible.

Sometimes it is helpful to insert an rf choke in series with the power supply of the amplifier. Proper placement of them limits high-frequency transient currents to the local loop close to the amplifier, and does not conduct long distances along the power line.

If the gate drive non-overlapping time is very long, the induced current of the speaker or lc filter will forward bias the parasitic diode at the output stage transistor side. When the non-overlap time ends, the diode bias changes from forward to reverse. Before the diode is completely turned off, large reverse recovery current spikes can occur, creating troublesome emi sources. Minimize emi by keeping the non-overlap time very short (it is also recommended to minimize audio distortion). If the reverse recovery scheme is still unacceptable, a schottky diode can be used in parallel with the transistor's parasitic diode to transfer current and prevent the parasitic diode from being turned on at all times. This is helpful because the metal-semiconductor junction of a schottky diode is essentially unaffected by the reverse recovery effect.

An lc filter with a toroidal inductor core minimizes the effects of stray field transmission lines caused by amplifier current. A good compromise between cost and emi performance is to reduce the radiation from low-cost drum cores through shielding, and care can be taken to ensure that such shielding can acceptably reduce inductor linearity and speaker sound quality.

lc filter design

In order to save cost and PCB area, the lc filters of most class d amplifiers use a second-order low-pass design. Figure 3 shows a differential second-order lc filter. The speaker is used to attenuate the natural resonance of the circuit. Although speaker impedance is sometimes approximated by simple resistance, the actual impedance is more complex and may include significant reactive components. To get the best filter design results, design engineers should always strive for accurate speaker models.

The purpose of common filter design and selection is to minimize the filter response drop under the highest required audio frequency conditions in order to obtain the lowest bandwidth. If the drop is required to be less than 1 db for frequencies up to 20 khz, a typical filter is required to have a butterworth response of 40 khz (to achieve the maximum flat passband). For common speaker impedances and standard l and c values, the following table gives the nominal component values and their corresponding approximate butterworth responses:

Inductance L (H) Capacitor C (F) Speaker resistance () Bandwidth -3-dB (kHz)

10 1.2 4 50

15 1 6 41

22 0.68 8 41

If the design does not include speaker feedback, the speaker thd will be sensitive to the linearity of the lc filter components.

Inductor design considerations: Important factors in designing or selecting an inductor include the rated current and shape of the core, and the resistance of the wire.

Rated current: The rated current of the core should be greater than the maximum current of the desired amplifier. The reason is that if the current exceeds the rated current threshold and the current density is too high, many inductor cores will magnetically saturate, causing the inductance to decrease sharply, which is not what we expect.

An inductor is formed by braiding wires around the magnetic core. If there are many turns of the Rao wire, the resistance related to the total Rao wire length is important. Because this resistor is connected in series between the half bridge and the speaker, it will consume some output power. If the resistance is too high, you should use a thicker Rao wire or choose a magnetic core of other metal materials that requires fewer turns of Rao wire to provide the required inductance.

Finally, don't forget that the shape of the inductor used also affects emi, as mentioned above.

System cost

What are the important factors in an audio system using a Class D amplifier that affect its overall cost? How can we minimize costs?

The active components of a class d amplifier are the switching output stage and the modulator. The cost of constructing this circuit is approximately the same as that of an analog linear amplifier. The real compromise to consider is the other components of the system.

The low power consumption of Class D amplifiers saves the cost (and PCB area) of heat sinks, such as heat sinks or fans. Class D integrated circuit amplifiers are available in smaller and lower cost packages than analog linear amplifiers. When driving a digital audio source, an analog linear amplifier requires a digital-to-analog converter (DAC) to convert the audio signal to an analog signal. The conversion is also required for a class d amplifier that handles analog input, but the class d amplifier for digital input effectively integrates the dac function.

On the other hand, the main cost disadvantage of class d amplifiers is the lc filter. The components of the lc filter, especially the inductor, occupy the pcb area and increase the cost. In high-power amplifiers, the overall system cost of a class d amplifier is still competitive, because the large cost savings in the heat sink can offset the cost of the lc filter. But in low-cost, low-power applications, the cost of an inductor is high. In rare cases, such as low cost amplifiers for cellular phones, the cost of the amplifier IC may be lower than the total cost of the lc filter. Even if the cost considerations are ignored, the pcb area occupied by the lc filter is a problem in small applications.

In order to meet these considerations, the lc filter is sometimes completely eliminated to adopt a filterless amplifier design. This saves costs and PCB area, although the benefits of a low-pass filter are lost. Without a filter, the increase in emi and high-frequency power consumption would be unacceptable unless the speaker is inductive and very close to the amplifier, the current loop area is minimal, and the power level remains low. Although this design is often adopted in portable applications, such as cellular phones, it is not suitable for high-power systems, such as home audio.

Another method is to minimize the number of lc filter components required for each audio channel. This can be achieved by using a single-ended half-bridge output stage, which requires half the number of inductors and capacitors as a differential full-bridge circuit. But if the half-bridge output stage requires a bipolar power supply, the cost associated with generating a negative power supply may be too high, unless the negative power supply already has some other purpose, or the amplifier has enough audio channels to share the negative power supply cost. In addition, the half bridge can also be powered from a single power source, but this will reduce the output power and often requires the use of a large DC blocking capacitor.

adi class d amplifier

All the design issues just discussed boil down to a fairly demanding project. To save design engineers time, adi offers a variety of class d amplifiers ic1, which contain programmable gain amplifiers, modulators, and power output stages. To simplify the evaluation, adi offers a demo board for each type of amplifier. The pcb wiring and bill of materials of these demo boards can be used as a practical reference design to help customers quickly design a proven, cost-effective audio system without having to do "repetitive work" to solve the major design problems of Class D amplifiers.

For example, consider using the ad19902, ad19923, ad19944, and ad199655 dual amplifier ic series products, which are suitable for medium-power stereo or mono applications that require two channels to output 5, 10, 25, and 40 w each. Here are some features of these ICs:

The ad1994 Class D audio power amplifier includes two programmable gain amplifiers, two sigma-delta modulators, and two power output stages to drive full-h-bridge connected loads in home theater, automotive, and pc audio applications. It generates switching waveforms that can drive two 25 w stereo speakers or a 50 w mono speaker with 90% efficiency. Its single-ended input is applied to a programmable gain amplifier (pga) with gains that can be set to 0, 6, 12, and 18 db to handle low-level signals.

ad1994 has integrated protection to protect the output stage from overheating, overcurrent and inrush current. Due to its special timing control, soft-start, and dc offset calibration, the clicks associated with muting are minimal. Its main performance indicators include 0.001% thd, a dynamic range of 105 db, a psr greater than 60 db, and a continuous-time feedback and optimized output stage gate driver using a switching output stage. Its 1-bit sigma-delta modulator is especially enhanced for class d applications to reach an average data frequency of 500 kHz, with high loop gain for 90% modulation, and full modulation stability. Independent modulator method allows to drive external high output power FETs.

ad1994 uses a 5 v power supply for pga, modulator and digital logic, and uses a 8 v to 20 v high voltage power supply for the switching output stage. The relevant reference design meets the requirements of the fcc class b emi standard. When driving a 6 load with 5 v and 12 v power supplies, its static power consumption is 487 mw, its power consumption is 710 mw under 2 × 1 w output power, and its power consumption is 0.27 mw in standby mode. The ad1994 is packaged in a 64-pin lfcsp and has an operating temperature range of 40 ° c to + 85 ° c.