Audio power amplifiers are ubiquitous, and there are audio amplifiers in places where music is heard. Generations of electronics engineers are working hard in this field to spread wisdom. The audio amplifier is to reproduce the sound signal realistically and efficiently on the speaker or earphone with a certain volume and power. Reality and efficiency have always been the driving force behind technological advances in the power amplifier industry. The audio frequency range is approximately 20 Hz to 20 kHz, requiring the amplifier to have a good frequency response over this frequency range; depending on the output power, the amplifier can be subdivided into different output power specifications, such as from a few hundred mW headphone amplifier to 2W It is used for small power amplifiers for portable devices, and for medium power and high power amplifiers for home audio systems of 10W and 20W. This article refers to the address: http:// Class AB amplifiers are the most commonly used audio power amplifiers today. The difference from the earliest Class A or Class B amplifiers is that Class AB amplifiers use complementary output stages, which can improve the efficiency of Class A amplifiers by adding a certain bias current at the output to prevent crossover distortion. , providing good sound quality. But it is still a linear amplifier, the output stage transistor operates in a linear amplification state, providing instantaneous continuous output current to the load. When the output transistor operates in a linear state, the voltage drop between the source and the drain is large. The instantaneous power dissipation of the output transistor can be expressed as VDS × IDS, and a considerable amount of energy will be dissipated on the output transistor and converted into heat. Even the most efficient class AB amplifiers are typically only about 60 to 70% efficient. Communication Aluminum Alloy Die Casting Communication Aluminum Alloy Die Casting,Custom High Alloy Die Casting,Hardware Alloy Die Casting,Metal Alloy Die Casting Dongguan Metalwork Technology Co., LTD. , https://www.dgdiecastpro.com
Class D amplifiers use a new way of working. The output stage transistors operate in a switching state, and the output switches between a positive supply and a negative supply to generate a series of voltage pulses. When the output transistor is not conducting, the output stage does not consume any current. The on-resistance of the output stage transistor is typically 0.2Ω, so the VDS is low at turn-on and the power dissipation (VDS×IDS) on the transistor is small. Low-power switching operation makes Class D amplifiers a significant advantage in many applications, such as saving metal area for heat dissipation on printed circuit boards, eliminating the need for dedicated heat sinks and extending battery life in portable devices .
When a Class D amplifier is operating, the input audio signal must first be modulated into a series of voltage pulses. There are many ways to modulate an input audio signal into a voltage pulse. The most common technique is Pulse Width Modulation (PWM). In principle, PWM modulation is to generate a series of voltage pulses of the same frequency as the carrier by comparing the input audio signal with a triangular wave of a fixed carrier frequency. The duty cycle of the PWM pulse is proportional to the amplitude of the audio signal during that carrier cycle, and the change in duty cycle contains the change in the input audio signal. Since the carrier frequency is usually more than 10 times that of the audio signal, the PWM modulation is spectrally free of distortion in the audio range.
Usually the output of the Class D amplifier is connected to the speaker via an LC filter. The LC filter is used to filter out the high frequency portion of the voltage pulse signal and reconstruct the audio signal. We know that the speaker itself has a certain frequency response range, and the human ear is basically not sensitive to signals outside the audio frequency range. From this point of view, the LC filter can be completely removed from the class D amplifier. Let's see what happens when the LC filter is removed. In the absence of any audio input signal, the comparator generates a series of 50% duty cycle pulses that are applied directly across the speaker and filtered inside the speaker to re-establish the DC state of zero input. This process produces a large amount of power dissipation in the resistive load of the speaker, reducing the efficiency of the amplifier.
The three-level H-bridge output structure completely eliminates the LC filter. In the three-level H-bridge output structure, the input audio signal and its inverted signal are simultaneously compared with the triangular wave, and two different voltage pulses are generated and applied to the two half bridges of the H-bridge. The differential pulses of the two strings of voltage pulses are the voltage pulses actually applied across the speaker. We also look at the situation when there is no audio input signal. When there is no input signal, the two strings of voltage pulses generated are in phase and are 50% duty cycle. No differential pulse is generated and there is no power loss. When the input signal is getting larger, a series of forward differential pulses are generated, and a forward current flows through the speaker; when the input signal is reversed, a series of reverse differential pulses are generated, and there is a reverse on the speaker. Flowing through the current. The current on the speaker is generated as needed when the input signal changes, and no excess power is lost.
In the actual circuit, the output stage and the driver circuit always contain various undesired factors, resulting in nonlinear distortion of the amplifier output. First, the output stage transistor has a very low on-resistance. If the upper and lower output stage transistors are turned on at the same time, a low-impedance path from VDD to VSS is generated through the transistor, resulting in a large inrush current. When the voltage pulse changes from high to low or from low to high, the output stage transistor is easily turned on at the same time. Therefore, it is important to avoid simultaneous turn-on of the output stage transistors. To prevent this from happening, both transistors must be forced to turn off before a transistor turns on. The time interval in which both transistors are turned off is called the dead time. The dead time changes the duty cycle of the original PWM pulse and causes distortion in the amplifier output. The shortest dead time is typically used in the design to prevent the output transistors from turning on at the same time and to minimize distortion as much as possible. In addition, the rise and fall times of the output pulse do not match, and the output stage transistor drive circuit parameters do not match, which also changes the duty cycle of the original PWM pulse, which in turn produces distortion in the amplifier output.
Let's take a look at what happens on the power when playing music. The power supply is directly connected to the speaker through an output stage transistor of a 0.2Ω resistor. The actual power supply always has a certain internal resistance. When playing music, the power supply will generate twice the ripple of the signal frequency. The ripple amplitude will vary with the internal resistance of the power supply. The ripple is directly coupled to the speaker through an output stage transistor of a 0.2 ohm resistor, producing even audible noise in the output signal. From a circuit perspective, this audible noise comes from poor power supply rejection ratio performance, that is, through power coupling, the power supply ripple produces high-order even-order distortion at the amplifier output.
In order to reduce the distortion, a circuit similar to a class AB amplifier can be used to improve circuit performance through closed-loop negative feedback. In principle, the closed-loop negative feedback is to add noise shaping filter in front of the comparator to shape the error noise between the input and output of the amplifier, attenuate the distortion in the audio range, and improve the linearity of the amplifier. This error comes from both the output transistor and the driver circuit, as well as the direct coupling of the power supply ripple, thus improving both the linearity of the amplifier and the power supply rejection ratio of the amplifier. Analogous to linear amplifiers, the noise shaping filter actually provides a large open-loop gain for the loop in the audio range. The optimized design of the closed-loop Class D amplifier achieves good performance with PSRR > 60dB and THD < 0.1%.
PT5306/26 is two high-performance Class D amplifiers recently launched by China Resources Converse Technology. The PT5306 is a single-channel 2.5WD class amplifier and the PT5326 is a dual-channel 2.1WD class amplifier. Both products use a closed-loop negative feedback three-level H-bridge output PWM modulation structure, and the output can be directly connected to the speaker without using an LC filter. Good performance is achieved by careful design of the internal noise shaping filter. From the typical test data, the performance parameters are no less than the same type of products abroad. For example, 3.6V power supply voltage, 8 ohm load, 0.5W output power, 1 kHz frequency THD+N is 0.1%, and 500 Hz frequency THN+N is 0.06%. The internal output of four 0.25Ω on-resistance output transistors forms the H-bridge output stage. With a 5V supply voltage, 8 ohm load, and 1W output power, the efficiency can reach over 88%. These two products can be widely used in a variety of portable multimedia devices, such as mobile phones, MP3, MP4, digital photo frames and low-power USB portable speakers.
The PT5306/26 includes a complete “beep†suppression scheme to quietly turn off or wake up the amplifier. In terms of protection circuits, the PT5306/26 also integrates overheat and overcurrent protection. Although the output power of a Class D amplifier is much lower than that of a linear amplifier, if the amplifier provides very high power for a long time, it will still cause the device to overheat. To prevent the risk of overheating, when the temperature exceeds the thermal shutdown safety threshold, the output stage is turned off and held until the device cools down. In addition, if there is a short circuit between the two outputs, a large current will be generated. If no protective measures are taken, a large current will damage the output stage transistors. Therefore, a maximum current limit needs to be added to the output transistor, and if the output current exceeds the safety threshold, the output stage is automatically turned off.
When the amplifier is applied, it often encounters an input amplitude that is too large. This condition causes the amplifier output to reach the limit state. The closed-loop modulator in the limiting state, even if the input signal is below the overload input amplitude, the modulator output lags behind the input signal and is in an output overload state for a long time. This hysteresis introduces additional nonlinearity at the output of the amplifier when a large signal is input. The PT5306/26 quickly adapts to changes in the input signal after the modulator overload condition disappears through a well-designed modulator overload prevention circuit.
In order to avoid the hum of the amplifier's own noise, a small power amplifier for portable applications typically requires a signal-to-noise ratio of 90 dB. The PT5306/26 achieves a satisfactory signal-to-noise ratio by carefully optimizing each noise source in the circuit.
The PWM modulation structure is simple and easy to implement, but it can cause EMI problems. In principle, the PWM output voltage pulse contains a large amount of energy at the octave of the carrier frequency, which produces a large amount of EMI. For example, the carrier frequency is 500 kHz. From the spectrum of the PWM output pulse, the energy in the carrier harmonic band within 30 MHz is large. The use of three-level modulation can reduce the amplitude of these frequency components. To further reduce the amplitude of these frequency components, it can be achieved by frequency hopping. The PT5306/26 includes a frequency-jitter modulation technique that equally divides the energy of the harmonic frequencies within 30 MHz into the entire frequency band outside the audio range by continuously varying the carrier frequency over a range of frequencies. In general, EMI needs to consider radiation to the space and conduction through the speaker and power lines. The Class D amplifier modulation scheme determines the baseline spectrum of conducted and radiated EMI. If you need to further reduce the EMI of your amplifier, you can use some board-level design methods.
When using a class D amplifier, attention should be paid to the selection of the bypass bypass capacitor CS. When the amplifier is working, the power supply is seriously disturbed. If the bypass storage capacitor at the VDD terminal is unreasonable, the amplifier will interfere with the power supply, which may cause other parts of the system to work abnormally. The best solution is to use a surface mount tantalum electrolytic capacitor for CS. If the cost and PCB layout size limit does not allow the use of tantalum electrolytic capacitors, you can use a ceramic capacitor of 0805 or 0603 package, and the capacitance value is above 4.7uF, which can effectively reduce the power supply interference when the amplifier works.