The IIS bus typically has four signal lines, as shown in Figure 1, including serial data input (IISDI), serial data output (IISD0), left/right channel selection (IISLRCK), and serial data clock (IISCLK). The main device that generates IISLRCK and IISCLK.
This article refers to the address: http://
2 IIS audio driver to achieve audio driver has 3 modes: MDD / PDD mode, Wavedev2 mode, UAM mode. They are the same in the same place: the interfaces are the same, they are all stream-driven, interacting with the upper layer of waveapi.dll through the stream interface.
The first MDD/PDD mode is the earliest mode and a common layering mode for other drivers. If you use the MDD (wavem-dd.1ib) provided by CE, there are some restrictions: only one device is supported; one setting only supports one stream; the support for loops is not reliable; the support for stream is weak. Of course, since the source code is provided, you can modify the MDD yourself to break through these limitations.
The second Wavedev2 model is due to the new requirements of the 2000 Smartphone project, which require major changes to MDD/PDD. For example, the above limitation 2, according to the development history of CE, at this time waveapi. The dll does not support the software mixer, which means that only one application can be played at the same time. So according to the situation at the time, CE's multimedia development team designed the Wavedev2 model. This is a single (non-layered) drive mode, and the platform-related modules are all in hwctxt. h and hwetxt. Cpp, in addition to the addition of midi support, software mixer support, S / PDlF interface, gain class interface, forcespeaker interface, and so on. Therefore, to develop Smartphone or PPC, this mode is quite suitable.
The third UAM mode, Unified Audio Model, adds support for DirectSound when developing WinCE 4.2, and some audio devices support hardware mixers. UAM is a good choice for this.
This test uses the drive structure of MDD/PDD. The key points of this drive are described below.
2.1 DMA Control and Drivers Generally speaking, DMA (Direct Memory Access) does not require CPU interference or CPU resources, and can automatically transfer audio data from the system bus to the IIS bus; if the audio averages at the sampling frequency of 44 .1 kHz, 16-bit word length, left and right 2-channel calculation, the code stream is 1.411 Mbps, usually 1 to 3 Mbps, so DMA transmission is necessary.
2.2 Clock Configuration As long as the bit clock and the sampling clock can match, the IIS data format is the same as the master and slave. The DMA configuration is good and the audio can work.
The IIS data format is mainly divided into three types: left-aligned, right-aligned, and IIS format. The sound sounds "weird", that is, the data format is wrong. The frequency calculation method is listed in Table 1.
The IIS master clock frequency can be selected by the sampling frequency. The IIS master clock frequency is generated by the IIS prescaler (IIS master clock frequency = MCLK / prescaler value), so you must choose the appropriate prescaler value and CODECLK sampling frequency type (256 or 384fs), in order to get the appropriate ISLLR sampling frequency (IISLRCK frequency = IIS master clock frequency / CODECLK sampling frequency type); serial bit sampling frequency type (16/32/48fs) can be configured by configuring the serial bit of each channel The number and CODECLK sampling frequency types are completed, and the relationship between them is listed in Table 2.
For example, the crystal frequency is 16.934 4 MHz, which is 44.1 kHz by 384 (this is the sampling frequency).
Bit clock frequency = sampling frequency × data bits × 2 = 44.1 kHz × 16 × 2 = 1.411 MHz
For crystal oscillators of other frequencies or clock frequencies from the bus, calculate the division factor in ISC0N to maximize the fit of CODECLK.
2.3 CODEC Control There are currently three bus control CODECs, SPI, I2C and L3. The L3 bus (L3MODE, L3CLOCK, L3DATA) is controlled by a general purpose I/O port. The L3 interface is actually a serial interface, which consists of three signal lines that complete the data and control signal exchange between the processor and C0DEC. The UDAl341TS is based on the L3 interface.
L3DATA: Processor interface data line.
L3MODE: Processor interface mode signal line.
L3CLOCK: Processor interface clock signal line.
Among the three control methods, I2C is the most common. I2C is divided into two types: register mode and I/O analog mode. The I/O analog mode has good portability. Only the I/O analog mode I2C driver can be divided into 8 bits, 9 bits, and 16 bits. Whether to have a sub address, whether it can be read and written, whether it is compatible with the SCCB bus.
2.4 The volume control node uses more volume adjustments. Figure 2 is a general model of a volume control node.
The gain at 1 is determined by the volume control function of the player, and the maximum is 0 dB. In other words, the original signal strength can only be restored at most.
The gains at 2 and 3 are controlled by the Coded IC itself. The WM8731 does not generate a gain function. It is easy to introduce signal distortion at two places, generally set to O dB. When the codec increases the volume, the gain is mainly increased at three places.
4, 5 is determined by the power amplifier, the maximum is also O dB, portable power amplifier is usually current type, relying on the amplification current to push the speaker.
When the gain of 1+2+3 is more than O dB, the signal of 1 kHz will produce distortion, but the intensity of most music is less than the strength of 1 kHz test square wave, so the sum of these three terms can be compared with O The dB is slightly larger, but not too large, otherwise it will cause signal distortion.
a. The application is equivalent to manually adjusting the volume in the control panel by calling waveOutSetVolume.
b. When adjusting the volume of the MediaPlay player, it is possible to judge whether the gain at 1 is changed by the message tracking, that is, the DSP digital output gain of the ARM.
c. When adjusting the volume in the control panel, you will find that the CODEC's amplifier register value will also change. The guess is that the control related registers are implemented via the IIS bus because no registers have been found to be overwritten by I2C in the IoControl message.
By analyzing the method of adjusting the volume, there are five nodes as shown in Figure 2 that can be controlled, with the aim of minimizing the maximum distortion of the volume: in the case of maximizing the output gain at one point, the two PCM Volumes are set to 0 dB (the magnification is the easiest here) Introducing distortion), the maximum volume of undistorted sound can be obtained when the power is placed at the maximum; if you want to increase the volume, you can only sacrifice the distortion, and the human ear can receive up to 10% THD (Total Hamonic Distortion). In the case, it is mainly to adjust the gain at three places.
3 Effective way to increase the volume
   1 Under the premise that C0DEC and amplifier can not be changed, it is important to choose the appropriate speaker (different speakers have different effects). The calibers vary in size and the cones are deep and shallow. When selecting a horn, it is generally required that the rated power of the power amplifier is more than twice the rated power of the horn, and the actual maximum power of the horn is 2 to 3 times its rated output power. The sensitivity parameter of the speaker is very important, generally O. It is about 85 dB at 1 W, depending on the sensitivity at rated power. Sensitivity is used to measure the efficiency of converting electrical energy into sound. It is meaningless to say that the rated power does not speak the rated power. The low sensitivity at rated power is not conducive to the "heating does not sound" of the resistance wire.
2 Increase the power amplifier voltage, according to P = U · U / R, a small boost voltage, you can get the square level of power boost. For example, from 4 V to 6 V, the power can be increased by 2.25 times.
3 improve the sound cavity design.
4 In principle, it is not recommended to exchange the fidelity for the volume. If it is necessary, it must be strictly controlled at THD <10%.
3.1 Matching of the amplifier and the speaker and selecting the output power of the amplifier must be greater than the output power of the speaker, otherwise it will not only affect the sound effect, but also accelerate the damage of the amplifier. If the selected speaker impedance is higher than the output impedance of the amplifier, it will affect the output power of the amplifier; when the impedance of the speaker is too low (such as less than 4Ω), the power amplifier used does not match the rated output power. In this case the distortion will increase. If the impedance of the horn meets the requirements, the rated power is slightly smaller than the rated power of the power amplifier, the distortion is relatively small, and the sound quality of the horn is good.
Speaker selection:
1 large diameter, deep paper cone, high conversion efficiency, greater power consumption; small caliber, too shallow paper cone, high frequency response is not good.
2 When the paper cone of the same diameter is lightly pressed by hand, the more laborious speaker has a high resonance frequency and a large dynamic range.
3 hard, dense paper cone speakers, high-frequency performance is generally better; coarse, soft paper cone speakers, the sound quality is generally softer.
4 amplifiers should have sufficient power output, especially for transistor amplifiers. The maximum output power of the speaker should be more than three times its rated power, and the maximum input power of the speaker should be equal to the output power of the amplifier to protect the safety of the speaker.
5 Impedance matching is the most basic requirement: For Class D functions, since PWM is easy to cause high-frequency interference, it is necessary to consider the appropriate inductive reactance to filter. As shown in Figure 3, the impedance and inductive reactance of the coil form a low-pass filter that ideally blocks high-frequency harmonic interference generated by PWM. Here, the frequency at which the gain is a 3 dB is selected as the cutoff point fc=RL/2πL of the high frequency. When the impedance is 8 Ω and the cutoff frequency is 20 kHz, then L = RL / 2 πfc = 8 Ω / (2 π × 20 kHz) = 64 μH. The 8 Ω portable speaker has an inductive reactance of 20 to 100 μH. If the actual inductance is >64μH, the bandpass characteristic will be limited; if the actual inductance is <64μH, the cutoff frequency will be >20 Hz, and noise will be introduced at this time. Therefore, the sensitivity is as close as possible to 64μH when selecting the speaker; for the class AB amplifier, it is not strictly required.
3.2 The sound chamber is designed with a good sound volume. Under the same power, the volume will be louder.
1 sound cavity should be flat, do not have a high and low unevenness.
2 sound hole is 15% to 20% of the sound cavity area (commonly used in mobile phones).
3 sound cavity should be as deep as possible to form a "V" type sound, the effect is better.
4 The front and rear sound chambers should be separated to avoid interference between the front and rear sounds. This principle is consistent with the reason that the speaker emits a much smaller sound than the speaker's sound in the box.
5 front sound chamber: the size of the sound chamber in front of the speaker is mainly determined by the height of the foam above the speaker. Generally, at least O. 2 mm foam. The front chamber mainly affects high-frequency sound, and the effect on SPL (SoundPressure Level) is not too great.
6 After the sound chamber: It should be large enough to achieve the best level of 2 times the equivalent sound volume of the mobile phone speaker; the larger rear sound cavity makes the speaker get better results at low frequencies.
7 The front chamber and the sound hole should be designed reasonably and properly: the front chamber and the sound hole form a Helmholtz resonator, and a resonance peak appears at a certain frequency point. If it is not specially designed, the resonance peak can be adjusted to the high frequency end (>10 kHz), correspondingly requiring the front cavity to be shallow and the sound hole area to be large; if there are special design requirements, for example, to improve the loudness, the resonance peak can be used. Adjusted to 3.4 ~ 6 kHz, but the result will be that the sound is monotonous, and the requirements for the sound source will be harsh.
8 Sealability: The most basic is to separate the front and rear sound chambers of the speaker to ensure good sealing (to ensure the tightness of the sound chamber of the mobile phone as much as possible). The good sealing makes the speaker get better results in the low frequency band (you can get a bigger, more feminine sound).
4 Sound test Because the human ear has different sensory perceptions of audio sound, and subjective differences, I wanted to write an article, specifically to introduce the evaluation and control methods of sound effects, what kind of instruments need to be controlled, and how to experiment. However, due to the limited experimental conditions and my ability, coupled with strong professionalism, I dare not write or fear to write well, but I have to give up. The following is a list of Wolfson Microelectronics plc Jason Fan (for reference only) and expects such articles to appear soon.
1Basic instruments: regulated power supply, millivoltmeter with built-in filter (can measure output noise and output power), distortion meter, sound pressure meter, signal generator.
2 advanced instruments: AP audio analyzer, audio full-frequency scanner (used to test speaker power).
3 The evaluation indicators of the audio system have basic indicators and upgrade indicators.
The basic indicators are: output power, signal to noise ratio, frequency response, distortion, left and right channel separation, left and right channel balance.
Upgrade indicators (measured with an audio analyzer) are: THD+N, dynamic range, FFT.
For audio testing, some standard test signals are generally used, such as left and right channel 1 kHz O dB; left and right channels 30 Hz O dB; left and right channels 100 Hz 0 dB; left and right channels 10 kHz 0 dB; left and right channels 16 kHz O dB; left channel l kHz O dB; right channel 1 kHz 0 dB.
The above instruments are accompanied by methods of use and experimental protocols.
5 Summary In the face of audio and video drivers, first of all, don't be afraid and feel unpredictable. After mastering the basic principles and the above points, the general IIS audio CODEC can be driven. In the process of driving WMXXX series (WM9712/WM8978/WM8960/WM8731), UDAl314, PCMl770, UCBl440, CS4344 and other chips, they have been effectively verified.
When applying CS42L52, it is found that the background noise is obvious, but the sound quality of the earphone is very good, indicating that the noise comes from the power amplifier; there is no initialization in the power-on, and further explanation comes from the power amplifier, and does not change with the volume change. Can not frontal noise reduction, the evasive measures taken later: the sound channel is turned off when there is no DMA transmission. This problem is later compared with the old and new boards. It is found that the inductance parameter in the LC loop connected to the output end of the speaker is self-excited. . After the inductor is replaced by an O Ω resistor, the noise is substantially eliminated.
In ARM, the crystal is most common at 12 MHz and 16.934 4 MHz (27 MHz or 28.XXX MHz in the video system), the system peripheral bus is 50 MHz, can be configured to be accurate 44.1 kHz or 48 kHz, It is necessary to pay special attention to the PLL of each chip. If the frequency is too different, noise will also be introduced and the speech rate will be abnormal.
Selection suggestions:
1 When selecting the type, be sure to stick it on your own board and do not rely on the vendor's DEMO board demonstration.
2 There is no background noise, which can basically be heard in the first 5 s blank time of the genuine song.
3 to see if the chip has been mass-produced and has been adopted, word of mouth is an important factor for reference.
The WM9712l has a four-wire resistive touch interface, which is more common in MP4 video player solutions. WM897x is used more in mobile phones or smart phones; in MP3 or low-cost solutions, PCMl770 accounts for a large share; Wolfson's WM897x system IIS audio protocol not only improves system integration, but also improves The audio quality of the system; WM897x uses DSP microprocessor as the core, which can filter the wind and so on to improve the recording function of the audio system. The new product also uses the equalization of 5-band and 3D audio systems to improve the audio output and programmable resistance filter. Eliminate noise. These systems also typically support microphones and cell phone speakers with clock frequencies between 12 and 19 MHz, further reducing the number of components in the product. In order to achieve a power consumption of 900 mW for high-quality audio speakers and piezoelectric speakers, a digital recording playback limiter is used to prevent excessive output of the speakers.
Sunlight shines on the p-n junction of the semiconductor to form new hole-electron pairs. Under the action of the electric field of the p-n junction, the holes flow from the n region to the p region, and the electrons flow from the p region to the n region. After the circuit is turned on, a current is formed. This is how photoelectric effect monocrystalline silicon solar panels work.
Application fields:
1. Household lamp power supply: such as garden lamps, street lamps, portable lamps, camping lamps, mountaineering lamps, fishing lamps, black light lamps, rubber tapping lamps, energy-saving lamps, etc.
2. Photovoltaic power station: 10KW-50MW independent photovoltaic power station, wind-solar (firewood) complementary power station, various large parking plant charging stations, etc.
3. Solar buildings: Combining solar power generation with building materials to make future large buildings self-sufficient in electricity is a major development direction in the future.
4. Other fields include (1) Supporting cars: solar cars/electric cars, battery charging equipment, car air conditioners, ventilation fans, cold drink boxes, etc.; (2) Solar hydrogen production plus fuel cell regenerative power generation system; (3) Sea water Power supply for desalination equipment; (4) Satellites, spacecraft, space solar power plants, etc.
monocrystalline silicon solar cells, single crystal silicon solar cell, monocrystalline solar module, flexible monocrystalline solar panel, monocrystalline pv panels
Ningbo Autrends International Trade Co., Ltd. , https://www.china-energystorage.com