This article refers to the address: http:// All electronic devices that include audio power amplifiers, such as stereo TVs and multi-channel AV receivers, usually have an important indicator of output power, which is the maximum volume that can be supplied, which is for many consumers. Said to be an important indicator. For the manufacturer, it is not only the output power, but also the thermal stability of the function in the worst case. Test standards in this regard will vary from company to company. Two amplifiers are commonly used to provide output power to televisions, Class AB and Class D amplifiers. The transition to Class D amplifiers is primarily due to the need for flat panel televisions (LCD or plasma), as heat dissipation is a problem because of the limited space in such machines. The current test standard is the standard developed when only Class AB amplifiers were used in the year. This article will discuss whether the standard is still applicable to Class D amplifiers. Maximum output power The maximum output power is the total power that the amplifier can provide at a given time and within the specified frequency and total harmonic distortion (THD). For example, in the power test method specified by the US Federal Trade Commission (FTC), it is required to perform one hour of preheating of the amplifier with a 1 kHz sine wave with a specified output power of 1/8. The amplifier must then be able to provide the specified power for 5 minutes, of course, within the specified THD and frequency range. The load is typically a 4Ω or 8Ω resistor, depending on the nominal speaker impedance. Since most TV sets do not have ports for external speakers, there is no way to test the output of the amplifier, so there is no standard for power measurement. The usual method of calibrating power is to use a 1 kHz signal with 10% THD for at least 10 minutes. Thermal stability This test is used to verify the thermal performance of the entire device. During the test, place the device in a test chamber of the specified maximum ambient temperature (usually 40 ° C). The temperature inside the device will rise, which makes the amplifier to withstand higher ambient temperatures. Use the speaker of the device itself as the load. Tests can be performed with test signals of different amplitudes and different waveforms, which will be discussed in the next paragraph. This test takes several hours. The measurement uses an infrared thermometer or thermocouple, but the measured values ​​are compared to the specifications specified in the safety standard, such as the highest PCB temperature and junction temperature. To pass the thermal stability test, neither the amplifier nor the speaker can be damaged. This functional test checks for potential damage by evaluating temperature characteristics. Test signal This thermal stability test attempts to simulate a worst-case real-world situation that will result in audio tracks in DVD and TV broadcasts. In order to guarantee the same test results in each test, engineers should use standard test signals. Once the final conditions are determined, it should also provide a stable temperature reading. A sine wave can provide a stable reading, but because its amplitude changes over time, it cannot simulate program content such as music or speech. The amplitude of the program signal should be a full range of signals, from muting to overdriving (clipping). The amplitude distortion of the program signal can be well described by the crest factor, which is the ratio of the peak power to the average power of the music or speech signal, expressed in dB. The source signals discussed above are not related to the thermal evaluation of the amplifier output signals we are concerned with. Not only must there be volume and sound control in the signal chain to allow for a large enough gain, but also a fixed power supply that limits the peak output voltage. Therefore, the crest factor will change when someone turns up the volume: because the peak is limited and the average power is still rising, the crest factor will decrease (this is different from the change in the amplifier's input signal). The lowest crest factor depends on the amount of distortion the consumer can accept and the maximum gain setting of the device. In any consumer application, the ideal worst case test is the lowest crest factor. Similarly, speaker manufacturers have also studied the appropriate test signals, and the speakers must be processed without damaging and severely distorting the output signal of the amplifier. Most manufacturers use the IEC268-5 standard, where the test signal is: pink noise signal, that is, after filtering (that is, high-pass after 40Hz, low-pass filtering of 5kHz, filter is a second-order filter) Various frequency distributions to restore the long-term frequency distribution of music sounds (Figure 1). The crest factor of the test signal specified in IEC268-5 is 6dB, which is the worst-case indicator. The average power that can be handled with this signal speaker is referred to as "continuous power," but most manufacturers have announced "program power," which is 3 dB higher than the former, with a discontinuous signal (sequentially cyclically Pass for one minute, break for one minute) test. Therefore, the speaker can process the clipped signal with a 9dB crest factor. The peak power involved in the crest factor refers to the peak power provided by the amplifier. The amplifier's rated output power is measured with a 3dB sine wave, so the long-term power handling capability of the speaker is 6dB, which is less than the rated power of the amplifier. The worst-case long-term test signal for the entire device is IEC268 noise, which has an RMS power 9dB lower than the peak output power and 6dB lower than the maximum sine wave power, which is the maximum output power of the sine wave test instrument. When considering the thermal design of the amplifier, there is no reason to require more power than the speaker specified. The integrated amplifier is usually thermally protected, so the worst case scenario is no sound, which is automatically reset after the amplifier has cooled down again. Setting the thermal limit of the amplifier to a lower level is actually an effective means of protecting the speaker due to permanent damage caused by speaker overload. Classification of amplifiers Two audio amplifiers can be used in the TV, namely Class AB and Class D. We will analyze the specific performance of these two types of amplifiers in the above tests. Class AB amplifiers are a low-cost, heavy-duty solution, but they consume too much power and require a large heat sink. Class D amplifiers have higher efficiency, but the disadvantage is that the price is too high. However, this is compensated by the need for less heat dissipation (small heat sink, or no need for a heat sink) and the small size of the IC. However, the system still needs to pass the thermal test, so the test strategy determines the cost of the amplifier. To simplify the comparison, it is assumed that both amplifiers are FETs, not bipolar output transistors. For a given supply voltage (V CC ), load (R I ), and R DSON (impedance of the output transistor all-on), the maximum output voltage is the same for both types because this is the maximum output power. Also assume a bridged load (BTL) output, ie the output current flows through both transistors and R DSON is doubled (Figure 2). Power consumption varies greatly for different types of amplifiers. Let us start with DC analysis and the output voltage is Ua (the output power P=Ua2/R I ): Class AB: Dab =[( Ua/ R I ) * (V CC -Ua)] + I Q *V CC The resulting power is the product of the output current and the voltage drop across the output resistor. Class D: Dd = (Ua/R I ) 2 * 2*R DSON + I Q * V CC The generated power is mainly composed of resistive losses, (output current) 2 * R Both amplifiers have a constant coefficient: I Q * V CC , where I Q is the quiescent current. Class AB amplifiers use this current to reduce crossover distortion, while in class D amplifiers, this current represents switching losses. The magnitude of this current is the same in both amplifiers. Further analysis can be performed by simulation. Consider a common TV application that uses both a 12V power supply, an 8Ω speaker, and the following parameter data: V CC = 12 V R I = 8 Ω R DSON = 0.3 Ω I Q = 0.02 A The first thing to determine is the efficiency, which is calculated by the following equation: Figure 3 illustrates the efficiency of a sine wave and also gives the distortion of the output signal. This distortion is caused by clipping, which in turn can be said to be caused by a limited power supply. The following equation is used to calculate the maximum output voltage swing: At 10% THD, the output power is 10W, which is the maximum output power specified by the system. As shown in the graph in Figure 3, Class D amplifiers provide much higher efficiency and output power than Class AB amplifiers. In the whole figure, the class D amplifier is only worse than the class AB amplifier at two points: Zero input: Both amplifiers consume only static power, assuming the same. Unlimited overload: The output has become a square wave and is always saturated, as is the AB class. At this point, both amplifiers have the same efficiency, power consumption, output power (15.56 W), and distortion (43.5%). Since efficiency is very important for battery-powered devices, most battery-powered device designers are concerned about the power consumption of the amplifier. Figure 4 shows the power dissipation curves for two amplifiers (note: the input uses a sine wave with variable gain). At 10W rated power, the power consumption of Class AB and Class D amplifiers is 2.53W and 0.994W, respectively. At the lower input stage, the power dissipation of the Class D amplifier is lower, while the power consumption of the Class AB amplifier is increased. What does this have to do with real-world applications? When is the amplifier used for music or voice amplification? In this regard, a good simulation can be performed using a noise signal whose amplitude distribution is similar to that of music and achieves consistent results. In order to compare the results with the actual listening situation and the power handling capabilities of the loudspeaker, we must change the x-axis variable from power to crest factor. The crest factor reflects the relationship between the average output power of the system and the peak power, where the peak power is 15.56W. The crest factor of an ideal noise source is infinite: its amplitude distribution conforms to a "normal distribution" with a distinct difference but no peak voltage limit. This distribution will change when we add the signal to the emulated amplifier whose output signal is limited by the power rail. The average (RMS) voltage will vary as the gain of the system changes. Increasing the RMS voltage will lower the crest factor because the peak reference remains the same. Clipping is rare when the crest factor is high, but it often occurs when the gain is increased. Figure 5 shows the noise of 3dB crest factor when the output signal is severely clipped. For the sake of simulation, we do not pay attention to the "color" of the noise, but the actual test should use IEC268-5 signal, because some amplifiers are less efficient at high frequencies. When we change the gain, we can calculate the power consumption for all possible peak value values ​​(see Figure 5). Between 15dB and 12dB, where the music power is very concentrated, it is severely clipped, which will force most users to lower the volume. 9dB ​​is the worst crest factor that the speaker manufacturer considers acceptable, and the output at 0dB becomes a full square wave. At 9dB, it will be the best point for thermal evaluation. Class AB amplifiers consume 3.05W and Class D is 0.388W. The ratio between the two is 3.05/0.388 = 7.86, and when performing power tests, the ratio is only 2.53/0.994 = 2.55. This kind of simulation has an important meaning: for class AB amplifiers, the challenge in thermal design is how to pass the noise test. Once the amplifier design is capable of absorbing 3.05W per channel, there will not be too many thermal design issues at the output power of 2.53W per channel. The rated output power can be permanently guaranteed. Since the power consumption obtained in the two tests is similar, a sine wave is used for output power and thermal testing in practical applications. Of course, although tests using sinusoidal signals are easier to set up, the resulting power consumption will be lower than the recommended noise test. In other words, thermal evaluation with a sine wave results in a class AB amplifier with lower power handling than a speaker with the same wattage. For Class D amplifiers, this will be the opposite. The noise test produces 0.338W of power consumption, while the actual power consumption at rated output power is 1W, which is 2.56 times the difference. Therefore, what signal is used for thermal evaluation will result in very large differences. If a sine wave is used in the thermal evaluation of a Class D amplifier, it will cause the system to be too large, which will increase the cost, because: IC suppliers need a larger chip area to reduce R DSON , which is one of the main factors affecting efficiency; Class D amplifiers are larger in package to achieve less thermal resistance between the junction and the PCB or heat sink. Manufacturers need to provide smaller heat sinks or multilayer PCB boards to achieve a smaller R thja , the thermal resistance between the junction and ambient temperature. If the PCB itself is used as a heat sink, careful wiring is required, and a large area of ​​continuous copper surface should be used. Since the copper skin is to transfer heat, multiple good via connections should be used between the layers. Aging test Sometimes a more rigorous test, the Burn-In test, is required in the thermal evaluation. In this test, the maximum voltage that the audio processor can provide is applied to the input of the power amplifier, making the output signal a square-like signal. In the example in this article, the amplifier consumes up to 1.41W per channel and is not much different from Class AB amplifiers. To pass such a test, the Class D amplifier requires 3.6 times higher cooling than the noise test. Summary of this article The conversion of a TV from a CRT to a flat panel requires a smaller amplifier with lower thermal power consumption, so a Class D amplifier is available. Even with traditional sine wave testing, heat can be reduced by a factor of 2.5 in new designs. Engineers must address the new challenge of solving EMI, designing output filters, and using a small amplifier package with a cooling pad. In order to reveal all potential cost-saving methods, including the use of Class D amplifiers, it is now necessary to rethink the test method. The following is the recommended test method: use interrupt burst mode to detect the output power, full power sine wave, the length of time just to get the THD value; use the noise signal or the worst case of the actual application (voice or music) to detect the heat performance. The latter needs to be equipped with a gain setting to limit the clipping of the amplifier so that an acceptable sound is obtained even at full volume. The lighting design of the factory includes indoor lighting, outdoor lighting, station, field lighting, underground lighting, road lighting, guard lighting, obstacle lighting and so on.The lighting design of the factory includes indoor lighting, outdoor lighting, station, field lighting, underground lighting, road lighting, guard lighting, obstacle lighting and so on.The lighting design of the factory includes indoor lighting, outdoor lighting, station, field lighting, underground lighting, road lighting, guard lighting, obstacle lighting and so on. Industrial Lighting Light,High Power Led Flood Lights,Waterproof Led Flood Light,Rgb Led Flood Lights Luxon Optronics Technology Jiansu Co.,Ltd , http://www.luxonim.com
Figure 1: IEC268-5 noise spectral density.
Figure 2: BTL amplifier output stage.
Figure 3: Relationship between efficiency and output power.
Figure 4: Power consumption vs. output power.
Figure 5: Amplifier output voltage with 3dB noise.
Figure 6