The best way to correctly calculate DAC power cons

  • Detail

Correctly calculate DAC power consumption data

as portable multimedia system designers push battery life to the limit, they are spending unprecedented time studying power consumption data provided by different silicon suppliers. Tit for tat comparisons are often difficult because there are too many variables and the key differences between competing devices are often far from obvious

audio input and output subsystems are particularly difficult because they contain both analog and digital circuits and usually require several different supply voltages. As a result, the data provided by manufacturers for these devices are often irrelevant to the actual use cases, and in some cases even completely misleading. However, familiarity with the basic knowledge of relevant circuits, in-depth understanding of Ohm's law and refusing to trust the manufacturer's face value data can help design engineers see through this confusing fog

what does each power consumption figure include

it may seem obvious, but understanding what circuits are included in each power consumption figure is the key to calculating the overall power consumption of the system. However, for example, "if you only rely on a data manual to carry out this work, it is often easier to say than to do.". Now let's think about the audio output of a portable system. Figure 1 shows all the main function blocks. The last few pieces of the chain (such as digital signal enhancement, DAC, analog mixing and amplification) are usually integrated in one device, which is generally called audio DAC

however, when the data book of such devices lists the DAC power consumption or DAC power supply current, it absolutely refers to the DAC itself and does not include amplifiers and other circuits. So what if playback to headphones? Will it include on-chip signal enhancement functions (such as amplitude limiting, 3D signal enhancement or equalization)? Probably not, because silicon suppliers rarely have the courage to make their devices look worse compared to competitors. Some silicon suppliers even specify that DAC power supply current does not include digital audio interface. Obviously, this is not similar to any actual use case, because the interface must be powered on to receive audio data for playback

what makes things more complicated is that the system architecture of these devices is also different. For example, volume control can be realized by software on CPU, digital part of audio chip, or analog gain programmable amplifier in audio chip. A useful and wise check is to determine what functions are required, check which physical device these audio functions are implemented in, and ensure that the power consumption of each function has been calculated

the power consumption of speakers and headphones usually accounts for a large part of the total power consumption. Since this power is not actually consumed in the IC, it is almost never included in the IC data book. Fortunately, it can be easily calculated from the formula P = v2rms/Z, where VRMs is the RMS voltage of the whole speaker and Z is its impedance (for stereo speakers, don't forget to multiply this number by 2!). The difficulty is to choose an actual VRMs. Although the maximum VRMs can be easily calculated from the swing of the amplifier output, in reality, VRMs depends on the volume setting of the end user. Even at the maximum volume, the VRMs on the treble and bass channels of the same piece of music are different, so it is almost impossible to assume a full-scale signal

in order to make a meaningful comparison between different audio devices, a common benchmark is needed. For example, the Japanese JEITA cp-2905b standard stipulates that the battery life of the system with headphone output should be measured when driving 0.2mw (0.1MW per channel) on 16 loads

what is the signal

amplifiers that drive speakers and headphones are another particularly power consuming device. At present, the common practice in the industry is to list their static power consumption, that is, play them absolutely quietly (in the digital domain, it means a string of zeros). However, as long as an actual signal passes through the system, the power consumption on the amplifier (and load) will increase

undoubtedly, the power supply current of the amplifier should be expressed by a non-zero signal, but what kind of signal should be used? Some standards (such as JEITA cp-2905b) often use a 1kHz sine wave because it is easy to generate. However, it has little in common with any sound or music heard by users in the real world. Pink noise (as defined by IEC standards for loudspeakers) may be closer to the amplifier supply current, although fundamentally no signal can map infinitely changing music

another thing to keep in mind when comparing amplifiers is that their power efficiency depends on the signal amplitude. The exact relationship depends on the amplifier (see Figure 2). For example, under static conditions, class D amplifiers may consume more power than equivalent linear amplifiers due to switching losses. Similarly, since linear amplifiers are more efficient at high volume, their efficiency at full scale can be close to that of class D amplifiers

however, the extreme parts of these signal amplitudes are largely irrelevant, because the battle to determine the battery life mainly starts in the middle of the signal amplitude, where amplifiers in the real world spend most of their time. It is here that class D amplifier has won widespread recognition in the industry, because its power conversion efficiency is much higher than that of linear amplifier

what about digital circuits

amplifiers are not the only circuits whose static power consumption is lower than the working power consumption, as are other analog circuits (such as mixers and gain programmable amplifiers) and digital CMOS circuits. For CMOS circuits, power consumption is largely a function of the conversion frequency of 1 and 0 status bits, so a signal composed of only 0 status bits (i.e. static) only requires very low power supply current. In order to get meaningful data, all devices should process a real non-zero signal

another factor to consider is the sampling rate of digital audio signals. Most digital and mixed signal circuits convert once per sample, so their average power consumption is directly proportional to the number of samples per second. When comparing audio DAC or ADC, you should pay attention to whether the power supply current indicated in the manual uses the same sampling rate as the benchmark

if you look up the signal chain, the coding quality of the source audio file (such as the bit rate of an MP3 file) can affect the power consumption of the decoder. The bit rate and buffer size determine the frequency of data recovery from the storage medium. This is particularly important in hard disk based systems because each disk read will cause a large spike in battery current

many audio ICs (such as DAC or ADC) can be configured as master or slave devices. In the master mode, the audio IC drives the digital audio interface, so it requires more current than the slave mode. It is not surprising that the power consumption indicated in the manual is usually measured in the slave mode. So does this mean that slave mode is always more popular? Of course not. After all, if the audio IC does not have a drive interface, the corresponding device at the other end must do this work, so the power consumption is only moved from one end of the system to the other end, not eliminated

when the manual indicates the power consumption in the main mode, you must pay attention to the load capacitance, because it determines how much additional current is needed. If the figures in the manual are assumed to be measured under the worst large load capacitance, the actual situation may be much better than the specifications in the manual. On the contrary, IC suppliers may manipulate these power consumption figures artificially by using unrealistic low load capacitors

some audio devices have special clock mode, which can eliminate the need for very power consuming low jitter PLL, but this mode can only be used in main mode. For example, many ousheng

audio DAC and codec have a USB mode, in which the audio clock can be directly generated from a 12Mhz USB clock. In this case, the power saved by the integrated clock usually far exceeds the power consumed in the audio IC

power supply

except for the simplest audio IC, all audio ICS use more than one power rail. A typical circuit includes at least one analog power supply, a digital i/o power supply for audio and control interfaces, and an independent digital core power supply. The total power consumption of IC is the sum of the power consumed on each power rail

one of the most obvious ways to save power is to use the lowest possible voltage for each power supply. For digital i/o voltage, the lower limit may be determined by other system components that the audio IC must be connected to. On the other hand, the digital core voltage can use the lower limit voltage specified under the recommended operating conditions usually in the data book

some data manuals contain the relationship curve between power supply current and voltage under a given working mode. If there is no such diagram in the data book, you can also make some reasonable logical conjectures. For digital logic such as CMOS IC, the current is proportional to the applied voltage. This means that reducing the voltage can get double benefits, that is, reducing the power supply voltage by half can actually reduce the power consumption of this power supply voltage rail by three quarters

things will become more complicated for analog circuits, because analog circuits often contain constant current sources. After halving the analog power supply voltage, the power consumed by the analog part of the IC (excluding any load) is usually between one quarter and one half of the original power consumed

better understand the power consumption data in the manual

in order to make a real and meaningful comparison of the power consumed by different audio ICs, "Before, the gap of plastic raw materials in China was very different every year. The test conditions between audio ICs must be practical and consistent, including the power provided to the load and the nature of the signal (such as pink noise) , sampling rate and power supply voltage

in addition, the functions must reflect the expected actual application situation, all required functions on the IC must be enabled, and any unnecessary functions must be turned off as much as possible. The digital interfaces of the audio IC to be compared should all work in the master mode or in the slave mode, and the load capacitance should also be the same in each case. The master clock of each IC should also be the same. If the clock source of a PLL also needs to come from the audio clock, its power consumption should also be included

of course, in real life, different suppliers tend to adopt different test conditions for their audio IC. However, if we know which factors have the greatest impact on power consumption, system design engineers can quickly find out some key indicators and infer some important data from the supplier's test strips, even the garbage pieces, according to their own actual applications. This enables them to make an in-depth observation of IC power consumption, which is much more meaningful than the title specifications found on the page that can often be in the front of the data book, but the silicone tube does not degrade. When the probing part is healed, the silicone tube needs to be removed twice. The friction in the removal process causes trauma to the initial healing part and the inner wall of the normal lacrimal duct, which not only brings pain to the patient, but also may cause lacrimal duct re adhesion or stenosis, affecting the curative effect

Copyright © 2011 JIN SHI