This article refers to the address: http://
introduction
In recent years, the technology of Class D amplifiers has developed rapidly, and the most common ones are those used in low-power products with less than 50W per channel. In these low-power applications, Class D amplifiers are inherently more efficient than traditional Class AB amplifiers because the output stage of a Class D amplifier is typically only turned on or off, with no intermediate bias stage. However, this efficiency advantage has not been widely appreciated by designers for a long time, because Class D amplifiers also have significant drawbacks: high device cost, poor audio performance (compared to Class AB amplifiers), and Output filtering is required.
In recent years, due to the following two main factors, this situation is gradually reversed, making Class D amplifiers attract widespread attention in many applications.
First of all, it is the market needs. Some of the advantages of Class D amplifiers have driven the rapid growth of the two terminal devices market for mobile phones and LCD flat panel displays. For mobile phones, the speaker and PTT (Push-to-Talk) mode requires high efficiency of the Class D amplifier to extend battery life. The development of LCD flat panel displays puts a need for "cool running" on electronic devices because the increase in operating temperature will affect the display color contrast. The high efficiency of Class D amplifiers means lower power consumption when driving electronic devices, resulting in less heat generation and better image display for LCD flat panel displays.
The second factor affecting the application of Class D amplifiers is the development of their own technology. According to market needs, some manufacturers have improved Class D amplification technology to make Class D amplifiers more affordable and have similar audio performance to Class AB amplifiers. In addition, some new Class D amplifier output modulation schemes can also reduce EMI for practical applications.
Some of the new Class D amplification schemes are based on the old-fashioned PWM-type architecture, but use more complex modulation techniques to achieve no-filtering in low-power systems. Efficiency metrics can be tested and verified, but some designers still suspect that there will be widespread EMC/RFI compatibility issues for products based on these new technologies. In fact, a good PCB layout and short speaker connections ensure that EMI radiation is greatly reduced to meet FCC or CE standards.
Application difficulty
In some applications, the physical layout requires long speaker connections. Such speaker connections have antenna effects and RF radiation must be tightly controlled. In fact, the longer the speaker connection, the lower the frequency at which it produces radiation as an antenna. At the same time, some applications require EMI radiation below the CE/FCC standard to comply with automotive electronics specifications or to avoid interference with other low frequency circuits. Faced with such diverse needs, these applications often become difficult to overcome.
The most representative application difficulty is the flat-panel TV. Since the speakers are usually arranged on the outer edge of the device, it is often inevitable to use long speaker connections. If there is still an analog video signal, it is not enough to meet the RF radiation requirements of the FCC or CE (these standards are only for frequencies above 30 MHz); it is often necessary to suppress the fundamental frequency of the switch to avoid interference with the video signal. If the traditional LC filters used in early PWM amplifiers are used, they need to be analyzed to ensure that they can effectively suppress the high frequency switching transients generated by the new amplifiers.
PWM type D amplifier
Traditional Class D amplifiers are typically designed based on pulse width modulation (PWM) principles. Its output can be configured as a single-ended or fully differential bridged load (BTL). Figure 1 shows a typical BTL output waveform for a PWM Class D amplifier. Fast switching times and near-rail-to-rail swings make these amplifiers very efficient. However, these characteristics allow the amplifier to have a wide output spectrum that can cause high frequency RF radiation and interference. Therefore, using such a scheme typically requires the use of an output filter to suppress unwanted RF radiation.
Figure 1. Waveform of a traditional pulse width modulation (PWM) scheme |
As shown in Figure 1, if the inverting and non-inverting output loops of the device have a high degree of matching, the two symmetric output signal waveforms will have a small common mode (CM) signal on the speaker or wire (the trace at the bottom) line). Note: The 50% duty cycle represents the zero input signal (idle state). Therefore, a differential low-pass filter can be designed to attenuate high-frequency components in the signal waveform (produced by fast switching) while retaining useful low-frequency components for output to the speaker.
New generation modulation technology
As the demand for Class D amplifiers continues to grow, some manufacturers have recently introduced a new generation of modulation schemes that can independently control the two half bridges of the H-bridge. This modulation scheme has two main advantages:
When the audio signal is weak or idle, there is almost no differential switching signal on the load. The quiescent current loss is improved over conventional PWM designs.
The minimum pulse, common mode (CM) switching signal helps to reduce turn-on and turn-off transients. The idle state DC level (filtered) of the BTL output pin is close to GND. Therefore, the mismatch or stray capacitance of the filter components (which may cause audio noise when the amplifier is turned on or off) can be minimized.
Obviously, this new technology has some advantages, but the amplifier output will no longer be symmetrical. The signal waveform shown in Figure 2 (taking the MAX9704 stereo Class D amplifier as an example) has a high common-mode component.
Figure 2. Modulation scheme for Maxim's MAX9704 stereo Class D amplifier |
The output filter requirements of such Class D amplifiers are different from those with conventional differential inputs and complementary PWM outputs. Compared to PWM, the output of the MAX9704 modulation scheme often contains a high common-mode signal, which needs to be considered when designing the output filter. As shown in the examples that follow, the effects of traditional differential filter topologies are often less than ideal.
Figure 3a shows a conventional PWM type D output LC filter with ideal values. For simplicity, it can be assumed that the speaker load has an ideal 8 resistance and ignores the DC impedance of the inductor. The problem can be solved with some simple SPICE simulations. Figure 3b shows the frequency response of the filter in Figure 3a to the differential input signal. The response curves of the two output nodes (FILT1, FILT2) relative to GND are given. The device values ​​shown in the figure have an ideal second-order roll-off above the 30kHz frequency and ideal transients. The in-band group delay characteristics remain flat for 4 μs.
Figure 3. (a) Traditional differential mode passive LC filter, (b) Frequency response of the differential input signal, (c) Common mode signal frequency response. |
Figure 3c shows the output of the same filter for a common mode input. Similarly, the response curves of both outputs are relative to GND. The output (Y-axis offset) has large spikes and significant under-damping. Combined with the equivalent circuit of the filter under the common mode signal (Figure 4), it is easy to understand why this result occurs. Since the ideal matching inductor and capacitor are used in the simulation, the differential signal on the resistive load is zero, so there is no attenuation of the LC component. L1 and C1 resonate (L2 and C3 are the same) to produce a peak. In the time domain (not shown), this situation will have large overshoots and oscillations. Note that when entering a common mode signal, C2 will introduce a zero. Therefore, the cutoff frequency of the filter (referred to herein as the resonant frequency may be more accurate) will be higher than the cutoff frequency at the differential input.
Figure 4. The equivalent circuit of the traditional LC filter in Figure 3a under common mode input. |
At this time you may ask, is there a problem? If the output spectrum common mode energy is zero at this frequency, then there is no problem. However, if the peak frequency is exactly equal to the Class D amplifier switching frequency, a large output voltage amplitude will appear on the speaker and wires. At the same time, the spread spectrum modulation (SSM) mode of the MAX9704 will cause the underdamped filter to introduce considerable noise above the audio band. The spread spectrum mode is pin-selectable. At this time, the high-frequency switching energy is “white noiseâ€, and the noise amplitude can be reduced by randomly adjusting the switching time cycle by cycle. This spread spectrum scheme simplifies EMI compatibility design in filterless applications.
Under-damped Common Mode Response Problem One of the solutions to the above common mode problem is to preserve the basic structure of Figure 3a, but to add damping elements that suppress high resonance common mode signals. Figure 5a shows the RC component connected in series between the two output nodes and GND. If the efficiency requirements in the application are not very high, only one resistor can be connected between the output node and GND, but capacitors C4 and C5 will help reduce the extra power loss on R1 and R2.
The values ​​of C4 and C5 should be weighed: on the one hand, increasing the C4 and C5 values ​​contributes to the R1 and R2 attenuation spikes, and on the other hand, reducing C4 and C5 to reduce the loss in high-pitched audio (up to 20 kHz). If the common mode cutoff frequency is much larger than the differential mode frequency, it is easy to choose, for example, simply increasing the ratio of C2 to C1 and C3 is achieved. Increasing the common-mode cutoff frequency reduces the values ​​of C4 and C5 while increasing the values ​​of R1 and R2, which reduces the audio loss on R1 and R2. If the common mode cutoff frequency is too high, the common mode component on the cable will be too much. Therefore, the ratio of the differential and common mode -3dB frequency points must be properly selected. The filter in this case uses a 1:5 ratio.
Figure 5. Adding an RC network (a) to each output of a conventional LC filter improves the frequency response of the differential signal (b) and the frequency response (c) of the common-mode signal. |
Figure 5b shows the response of the filter of Figure 5a to the differential input and Figure 5c shows the response of the common mode input. Note: The common mode cutoff frequency is high in Figure 5c (-3dB bandwidth is about 110kHz, differential input is 28kHz) with smooth and reasonably controlled spikes. The cutoff frequency is much higher than the highest audio (also lower than the D-type switching frequency fundamental), so it has a good effect.
Some low switching frequency (200kHz to 300kHz) applications are not suitable for the solution shown in Figure 5c. Other methods and topologies may be required for such products. The MAX9704 stereo D-class amplifier (Figure 6) can be set to 940kHz fixed-frequency mode (FFM) (FS1 = low, FS2 = high), which works best. The MAX9704 operating in FFM mode sets the switching period to a constant value (with three options) via pin selection to meet application requirements.
Figure 6. Typical Application Circuit for the MAX9704 Stereo Class D Power Amplifier |
Figure 7 and Figure 8 show the time domain performance when filtering the MAX9704 using the filter of Figure 5. The load impedance is 8 in both cases. Figure 7 also shows the waveforms of the FILT1 and FILT2 nodes (the trace at the top) and the resulting 1kHz differential output waveform (the trace at the bottom). The noise of the top trace is the residual signal after the output switching signal is filtered (the supply voltage is 15V). Figure 8 is a detailed view of the trace of Figure 7. Note: The ripple is mainly from the 940kHz switching frequency, which appears as a common mode signal on both channels. It should also be noted that there are no higher harmonics on the output, indicating that EMI is effectively suppressed (the initial test frequency for radiated EMI is usually higher than 30MHz).
Figure 7. Signal waveforms generated on FILT1 and FILT2 when driving the Figure 5a circuit with the MAX9704 (shown also at the top trace), and differential output (bottom trace). |
8. The top trace shows the residual ripple voltage in the output of the circuit of Figure 5a. The ripple component is primarily the fundamental frequency of the switching frequency (940 kHz at this time). The second-order roll-off of the filter above this frequency suppresses all higher harmonics very well. The ripple is almost only the common mode component (the trace at the bottom). |
Off instructions
The filter design discussed in this paper assumes a load impedance of 8. The voice coil inductance results in a high frequency range of 20 kHz, and the impedance of most wide range moving coil speakers becomes high. This feature helps achieve high efficiency filterless operation, but when selecting filter components to reduce EMI, consider the rise in impedance.
When attempting to evaluate and characterize Class D amplifiers, audio designers often need to filter even in a lab environment for device selection and evaluation. Even if the final product without a filter can pass the EMC test, the amplifier performance test can still be used to find the problem. Many audio analyzers are designed to measure the THD+N or amplitude response of traditional audio amplifiers and often present errors when testing unfiltered Class D amplifiers. The circuit shown in Figure 5 is suitable for testing (correct loading of 8 resistive loads), but it should be noted that the nonlinearity that may be introduced by the 33μH inductor will limit the THD measurement. Air gap components tend to have the best measurement results, but their size often limits their application in real products!
A tablet computer is an electronic device that integrates mobile commerce, mobile communication and mobile entertainment, with a touch recognition LCD screen, handwriting recognition and wireless network communication functions. At present, the tablet computer has become the most popular electronic product.
1.In appearance, the tablet computer looks like a large-screen mobile phone, or more like a separate LCD screen.
2.In terms of hardware configuration, a tablet computer has all the hardware devices of a traditional computer, and has its own unique operating system, compatible with a variety of applications, and has a complete set of computer functions.
3.Tablet PC is a kind of miniaturized computer. Compared with traditional desktop computers, tablet computers are mobile and flexible. Compared with notebook computers, tablet computers are more compact and more portable.
4.Tablet PC is a digital notebook with digital ink function. In daily use, you can use the tablet computer like an ordinary notebook, take notes anytime and anywhere, and leave your own notes in electronic texts and documents.
Tablet Pc Pad,Mobile Tablet,Scratch Pad Tablet Pc,Tablet Pc,Tablets & Presentation Equipment,Educational Tablet
Jingjiang Gisen Technology Co.,Ltd , https://www.jsgisentec.com