The following lists many requirements of electroacoustic transducers , but according to the principle of electroacoustics , it is impossible to meet all the requirements. Some demands are mutually exclusive and conflicting with each other, and electro-acoustic engineers need to choose between the ebb and flow of each other’s factors. This means that electro-acoustic is not only an engineering discipline, but also an endless craft art.
High sound pressure ( level )
mean high sensitivity in speaker terms , that is, the output of the transducer is large. This is not necessarily the case with the microphone, and the sensitivity of the output voltage waveform of the microphone does not have to be very high. Because the output signal can be amplified by the post-amplifier, the electronic SNR of the current amplifier is very high, and it will not be distorted by amplifying a few dB . Generally, the sensitivity of 6phi, 9.7phi and other condenser microphones is mostly between (-42dBV per Pa) and (-36dBV per Pa) . A few such as digital MEMS mics may have a sensitivity of (-26dBV per Pa) . The current sensitivity That’s enough.
Returning to the category of radiators, high sound pressure can be discussed from the buzzer. The buzzer usually plays a single tone or two or three frequency points , and these frequency points are usually around its resonance frequency. There is a difference in application between the buzzer and the speaker. The speaker emphasizes a flat frequency response, and will not sacrifice flatness because of the high sound pressure at a certain point; however, the frequency points of the buzzer are very few, and the resonance point The sound pressure should be as large as possible to achieve the effect of warning or reminder.
The resonant frequency of the buzzer is designed between 2khz~4khz , these are the sound ranges that the human ear is more sensitive to, because the sound in this frequency band will resonate in the HRTF ear canal part of the human ear, which is related to the length of the ear canal. The diaphragm of the electromagnetic buzzer is a thin silicon steel sheet and is matched with a weight plate, so the resonance frequency is lower than that of the piezoelectric buzzer; the piezoelectric buzzer performs bending mode vibration in order to increase the amplitude and limit the diaphragm , so the copper sheet will be pasted on the piezoelectric ceramic. The bending mode means that the electric signal uses the indirect piezoelectric effect (indirect piezoelectric effect) to cause the actuator to generate stress in the direction of d31 , and the horizontally stretchable piezoelectric material is attached to the copper substrate, which will cause the overall Vertical bending vibration. The rigidity of the copper sheet is higher, and the mechanical resonance makes the resonance frequency of the overall piezoelectric buzzer much higher.
Because the internal damping of the vibration plate is low, the buzzer has a high Q value, and there will be an obviously raised frequency response section at the resonance point. The advantage of a higher Q value is that when the customer uses the optimized frequency point , the sound pressure can be higher, but it also has a disadvantage, that is, when the customer does not want to use this frequency point , the response sound pressure of other frequency points is obviously lower. Therefore, we must first determine the prompt tone and warning audio segment that the customer wants to sound, and then develop the corresponding buzzer.
Some customers ask for two resonance points. At this time, we can refer to AATC’s wideband buzzer design , such as part number ADI-F10H3C1. He additionally designed another resonant cavity on the main cavity to generate another similar resonant frequency point . On the frequency response curve, the response between two resonant frequency points is connected and pulled up to form a wider frequency band. In other design methods, a sound guide tube can also be placed in front of the buzzer to achieve Helmholtz resonance.
As far as speakers are concerned, sound pressure is also an important consideration, but it may not be as absolute priority as buzzers . At present, for example , the 1511 and 1609 series used in handheld devices , the 1206 and 1508 series used in watches and glasses , everyone is constantly increasing the sound pressure level to above 100dB ; the 3411 series commonly used in notebooks , we also hope to have a sound pressure level of 50cm Under ranging, the sound pressure can greatly exceed the benchmark value such as 75dB . Under high sound pressure, we should also pay attention to the echo cancellation (Acoustic Echo Cancelling, hereinafter referred to as AEC) function of the terminal electroacoustic products . When the radio microphone is placed closely next to the speaker, such as in the case of a smart speaker or earphone, the microphone will receive extremely high sound pressure and corresponding vibration. Echo during calls can be removed by AEC ICs , and vibration interference can be removed by foam, rubber and other shock-absorbing materials.
High Efficiency
In the public address system, in order to make the sound spread farther and more efficiently, a compression driver is used . The cavity in front of the diaphragm of the compression horn is very small, which gathers a lot of energy, and the rear of the monomer will guide the sound out to connect to the horn amplifier (horn). Because the sound cavity of this type of monomer is very small, an equal voice coil is usually used. The voice coil is longer than the magnetic gap, and the important part of the conductor is in the important area of the magnetic circuit, and the magnetic efficiency is very high. Due to the small sound cavity , there is no way to use a short voice coil, which will require more magnet paths, which will increase the weight in addition to increasing the volume.
Due to the limited amplitude of this type of monomer, the voice coil is not likely to move out of the magnetic gap, so the high distortion problem caused by equal voice coils can be avoided. Because the acoustic impedance from such a small sound guide tube and opening is very high, in order to match the impedance to the outside atmosphere smoothly, the amplitude curve of the horn needs to be specially designed. For example, the exponential type is usually better than the linear type, because the exponential type can ensure that the value of dA /dx remains constant, and the impedance will not change too much, but gradually changes. If the impedance changes too much, it is equivalent to the sound wave transitioning from one medium to another, which will inevitably generate a little reflected wave. In addition to losing energy, the reflected wave will also cause the original sound field to be scattered . Therefore , in order to increase the efficiency of the compression speaker , it is necessary to use a horn. to use.
Electro-acoustic transducers in underwater acoustics are often used for ultrasonic sonar detection. At this time, it is necessary to efficiently propagate sound waves over long distances and accurately. In order to increase efficiency, ultrasonic transducers generally use ceramic piezoelectric transmitters. This is because the mechanical impedance of piezoelectric ceramics matches the impedance of water, which can improve efficiency. Moreover, the rigidity of the ceramic capsule is high, and the high-frequency segmentation movement is less obvious, and it can support the ultrasonic threshold beyond the audible threshold , which is very suitable for underwater sonar detection. But one of the disadvantages of high efficiency is that if the water is filled with sound wave energy, it may have a negative impact on the surrounding marine ecology, such as for cetaceans and dolphins . Cetaceans rely on sound waves to communicate with each other. Underwater sound sources such as sonar or underwater blasting may disrupt the communication or migration of cetaceans . Therefore, sonar experts will first understand the frequency bands and sound characteristics of cetaceans communication, and Be careful not to emit similar acoustic signals.
But the efficiency is not always high. For example, although the efficiency of a short voice coil (underhung) is high, but at a large amplitude, its BL magnetic field strength and magnetic force lines are unstable, which will cause distortion. On the contrary, although the efficiency of the long voice coil (overhung) is not so high, because some voice coils are not affected, the distribution of the surrounding magnetic force lines is relatively stable and the distortion is low. Long voice coils are often used on woofers , because the stroke of the woofer is longer and it needs to work in the linear region of the magnetic return . It seems that the distortion of the long voice coil is low, and it is very popular in high-fidelity speakers , but its disadvantage is that it is not easy to dissipate heat. This is because it is best to dissipate heat from the voice coil close to materials with high thermal conductivity such as magnets and yokes , rather than air. For example, in a woofer, the heat dissipation of the voice coil outside the magnetic circuit is not easy, and it may be necessary to enhance the thermal conductivity of the metal of the voice coil tube , structure ventilation, or even add ferrofluid , etc. Therefore, high efficiency cannot be achieved blindly, and other conflicting factors need to be considered.
The following is another case that does not need to blindly pursue efficiency, but the definition of efficiency below is not the same as that of radiators. As far as the microphone is concerned, the sound field of the omnidirectional microphone is closer to a spherical surface, and the efficiency is relatively high. This is because his diaphragm has only one side, which can take in the sound field, which is a certain definition of efficiency. However, the sensitivity of the directional microphone to each direction is different, which can be seen from the notch and side lobe of the directivity diagram , which means that the sound collection efficiency is much lower than that of the omnidirectional, and the incident sound waves at some angles will not be converted at all. into electrical signal output.
The above is for different radio angles corresponding to different radio efficiency. We can also see from the radio frequency that the radio efficiency of the directional microphone is low. For example, in the case of a unidirectional microphone, there will be a phase difference between the sound waves arriving at the front and rear diaphragms . This effect is similar to that of a differential input microphone, which will cause the diffracted low frequencies to cancel each other out, making the low frequency response of the unidirectional microphone attenuate quickly and improve the efficiency. lower. One of the solutions is to reduce the damping of the microphone around 100~150hz , so that its frequency response has a peak at 150hz , and then lift the response below 100hz . In terms of application, directional microphones still have their advantages, and they perform better than omnidirectional microphones in certain situations, which is why there is no need to blindly pursue efficiency.
Low Distortion
In order to satisfy high-fidelity ( Hi-Fi) or high-resolution ( Hi-Res) listening enjoyment, general speakers should try their best to achieve low distortion. The distortion higher than fo is better than 1% , and the distortion lower than fo is because the bass stroke is larger, and the nonlinear distortion of BL and Cms will be relatively high, which is acceptable. However, when the threshold is lower than 1% , manufacturers may pay higher costs in order to reduce the distortion even further. At this time, the question is, can listeners really distinguish between 0.5% and 1% distortion ? According to personal and surrounding experience, there is usually no way , in fact, the average person can accept about 2~3% distortion, which is fine. Therefore, for non-audiophile applications such as 3C products, if the price factor is very important, the distortion is about 1~3% . The distortion measured by general instruments may be more theoretical, but the so-called noise is easier to distinguish. The noise may come from the bottoming of the voice coil , rubbing the ring or air leakage, etc. The reason may come from the assembly, the positioning fixture is wrong or the parts have tolerances, or the initial position of the voice coil and the initial position of the voice coil are not considered in the design. stroke. In terms of measurement , this can reflect high-frequency distortion (HOHD) and the Rub & Buzz results that Steve , the founder of SoundCheck , is trying to promote . In practice, manufacturers usually have golden ears to help listen, and there are listening rooms on the production line , where speakers with no noise are enshrined as standard golden samples for contingencies. Continuing from the above, the current measurement industry is beginning to pay attention to perceptual noise, such as Soundcheck ‘s ” Perceptual Rub & Buzz ” , which has a unique algorithm to calculate the equivalent noise in the human ear , the masking effect of the audio , and the human ear Equal-loudness curves, etc., try to estimate the perception level of these actual noises in the human ear nervous system ? This can also echo the intersection of electro-acoustic science and listening art, which is never-ending.
As far as microphones are concerned, condenser microphones with better sound quality also have an Equivalent Noise Level of 20~30dBSPL , which is not audible in general locations. But it will become obvious in a recording room with a noise floor of 20~30dBSPL . Moreover, the noise floor of the recording room is more concentrated in the low frequency band ( because the low frequency is difficult to absorb, and some ventilation or motors are low frequency ) . The noise floor of the room cannot completely cover the noise floor of the microphone. Solutions on the application side, such as covering the microphone head with sound-absorbing material, such as a boom mic for film radio with a layer of velvet cover. Solutions for R&D include reducing the residual stress of the diaphragm , or increasing the area of the diaphragm . When the micro-electromechanical microphone ( MEMS mic) collects sound, it is also necessary to ensure that the sound pipe is isolated from the internal space, and there must be no leakage between the pipes , otherwise the sound field will be very disordered. Pipes can be isolated with materials such as rubber sleeves.
After measuring the distortion , the next step may be to think about how to reduce the distortion. A further approach is to simulate the distortion before designing the monoblock. This is more complicated than the frequency domain in the theoretical model . Generally, in the equivalent circuit or finite element method, we mostly simulate the frequency domain , because of the help of Fourier transform and related transfer functions. Research in the time domain is more difficult. First, it is necessary to import some nonlinear parameters of materials, even the nonlinear high power functions of BL and C ms . Generally, when deriving acoustic equations, we deliberately ignore terms above the quadratic power , because we assume that the signal is a small signal with limited amplitude, and the sound pressure level is also limited. Distortion prediction at large amplitudes, however, is important when approaching the limits of the system. The second is that these calculations are very complicated and often involve higher-order differential equations, which can no longer be solved by linear algebra, and more time will be spent on finite element simulations. In the JAES journal in October , there was an article ” Favorite problems on acoustic transducers and linear system theory ” , which explored the differences between linear and nonlinear theories.
High Signal-to-noise Ratio
As audio enters the digital era from analog, the noise floor of recorded audio files is greatly reduced, and the noise interference in the transmission process is also greatly reduced. At this time, the remaining obvious noise is from the microphone itself, so we will hope that the microphone The higher the signal ratio ( hereinafter referred to as SNR) , the better. From the previous 60dB , it gradually expanded to 62dB , 65dB , and some manufacturers even advertised 70dB . Generally speaking, the larger the diaphragm , the higher the SNR , but in the era of miniaturization, we cannot blindly increase the size, so another method is to reduce the residual stress of the diaphragm to reduce the noise caused by stress. Another way is to use a dual backplate microphone , the diaphragm is sandwiched between two backplates , which will generate a differential signal during the push – pull movement and maintain the overall spatial gap sound The sum of the impedances is consistent to resist external homogeneous interference and increase SNR . Other methods include reducing the acoustic resistance of the microphone , since these resistances act like resistors in a circuit and create noise . In the finished recording microphone, the damping (mesh) strength can be reduced and the frequency response can be adjusted by electronic equalization ( EQ) , but in the MEMS mic microphone, the acoustic resistance of the monomer is usually determined by the back plate or near the diaphragm Holes are created, which requires more complex actuarial calculations.
As far as MEMS mic is concerned, high SNR sometimes requires the use of digital microphones ( hereinafter referred to as DMIC) . Compared with analog microphones ( hereinafter referred to as AMIC) , DMIC outputs digital signals and is less susceptible to external noise due to codecs interference, so it is often used on laptops . The DMIC on the laptop is often placed in the lens module above the screen. When the signal is transmitted to the C- side of the laptop, it has to pass through the screen switch noise, etc., and the transmission distance is long, so the DMIC is needed to resist interference . If the AMIC wants to resist interference, it is necessary to use a differential output (differential) AMIC . The three lines of the differential output positive, negative and ground can effectively offset the external equal noise source .
Reduction
When the microphone is receiving sound, sometimes it is necessary to include the ambient sound in LIVE or in the field, so that the recording content will not be too dry, or to make the listener have a sense of reality, the omnidirectional microphone is used to collect the surrounding sound, or Use a bidirectional microphone to collect the ambient sound on the opposite side of the speaker with another direction. One of the applications of bidirectional microphones is to use bidirectional microphones in 1 -on- 1 interviews, with the interviewer and the interviewee on both sides.
However, in most applications, in order to remove unnecessary noise, it is generally not desirable to collect too many irrelevant ambient sounds. Therefore, a microphone with high directivity is used to directly aim at a single sound source, such as a speaker or a singer. In the conference room, a shotgun microphone or the like is used to block noise from other angles. The directivity is caused by the pressure gradient design , because the incident sound waves on both sides of the diaphragm are delayed and have different phases, and the incident sound waves at different angles will produce different degrees of cancellation when they reach the diaphragm . The directivity design parameters are the position of the opening to connect to the outside world , the size of the hole, the length of the pipe, and the size of the cavity. It is worth noting that in order to restore the general physical dialogue situation, there will be some slight background sound, and some comfort noise (Comfort Noise Generator, CNG) will be deliberately put in during the call to increase the sense of presence.
Flat Frequency Response
The frequency response (Frequency Response) of the omnidirectional microphone is relatively flat, because only one side of his diaphragm is open to the outside world. Both sides of the diaphragm of a directional microphone are open to the outside world, which will cause the incident sound waves on both sides to interfere with each other, causing the frequency response to be bumpy in a specific frequency band. The size of the microphone needs to be as small as possible, because the smaller the size, the easier it is for high-frequency sound waves to diffract, and it is not easy to form scattering and reflection. In this way, the sensitivity of the microphone in the high-frequency band where the axis is positively picked up can be avoided from being too high.
As far as speakers are concerned, the rigidity of the diaphragm is related to the mid-to-high frequency. The harder diaphragm will have a higher sound pressure output in the mid-to-high frequency, and it can also be used in a more open field. Ambisonic Systems , acquired by Origin Acoustics , recently introduced the Ribbon Six (LS6) , a ribbon-style horn loudspeaker. The aluminum strip itself vibrates as a conductor. Since it can be made very thin and light , it has a better transient response , and its low quality has a higher sound pressure. In addition, the aluminum ribbon-type diaphragm is fully and evenly stressed, so it also has the characteristics of a flat speaker.
of the diaphragm is related to the low frequency . At present, the electroacoustic technology has encountered the bottleneck of low frequency extension. The material can be softer, but it will introduce more unstable side-to-side swing or split motion; the cavity size is mostly limited by the ID of the end customer , and various sound-absorbing materials have been added to the cavity . In recent years, there have been researches on transmission line speakers, such as bending the sound pipe as an acoustic low-pass filter to guide the bass energy behind the monomer, and at the same time adjust the phase to make the phase between him and the front sound wave Consistent, in order to strengthen the low frequency. The disadvantage of the transmission line design is that it is easy to cause a lot of reflections in the box, which may cause various resonances and noises, so the transmission line must also be covered with various sound-absorbing materials. There is also a more novel way to strengthen the low frequency, called ” Tapped Horn ” , he is similar to a bandpass filter (bandpass loudspeaker) , designing the sound cavity in front of the monomer , but he is more bold, making a horn-like Large bends to diffuse the sound to the back of the monomer. JASA has an article in May this year discussing this, ” Enhancing low frequency sound radiation of electrodynamic loudspeakers with acoustic metamaterial ” , he called the various spaces contained in the horn Helmholtz resonators , that is, resonant cavities one after another. Use these resonators for purely acoustic tuning. Different from other studies, the materials placed in the cavity of this study are acoustic metamaterials .
If the working frequency band of one speaker is limited, coaxial speakers or multiple speakers can be used. In coaxial audio, the entire diaphragm will move when reproducing the bass , but when the treble is limited by the quality control and inertia of the diaphragm, only the middle part can move correctly, so the tweeter is generally placed in the middle .
There are multi-channel speakers in general audio, and there are frequency dividers to divide the frequency. At this time, in addition to the frequency response of the monomer needs to be controlled, the electronic circuit design of the frequency divider is also the focus. For example, the crossover point , the filter slope of the crossover (6dB, 18dB, 24dB per octave) , the position and direction of the monomer placement, etc. In the beginning, you may feel that in the frequency domain , the higher the slope, the better, but its disadvantage is that it will cause too much phase shift. If the phase is not designed properly or violates the minimum phase (minimum phase) and other principles, it is easy to cause the gain of the system to be inconsistent. Be careful of getting out of control; in addition, it looks smooth in the frequency domain , but these filters with steep slopes in the time domain may cause more ripples and larger fluctuations in the impulse response . may not be applicable in your system.
In addition to using electronics to achieve frequency division, it can also be achieved by mechanisms. For example, Japan ‘s Final B1 earphones are advertised to use acoustic mechanisms to divide the sound , and the sound field is very natural and refreshing, and it is impossible to feel that it does not use electronic frequency division. Looking at similar earphone acoustic tuning patents such as Sony , it can be seen that the acoustic low-pass and high-pass filters are usually formed by the cavity and the pipe for tuning. The advantage of this method is that it will not introduce noise from electronic frequency division, but the challenge is the It is very dependent on the design process, and may even require finite element method software to help simulate the sound field.
For example, all the above-mentioned filtering or tuning can be tuned in an analog or digital way. The analog tuning method uses resistors, inductors, and capacitors to make circuits. The advantage is that it is simple and less likely to lose control of the circuit like a digital algorithm. The disadvantage is that it cannot perform real-time adaptive tuning, and the parameter selection of passive components is limited . Digital tuning does not have this problem. Digital tuning is implemented in a programming language, and there is a very intuitive way to manipulate the poles and zeros of the transfer function to design where the frequency response curve should have peaks and where. Edges have valleys .
In the headphone unit, especially in the monitor-level headphones (In-Ear Monitor, IEM) , multi-units such as ring iron , multi-moving iron (ie balanced armature, Balanced Armature, BA , hereinafter referred to as moving iron ) headphones Not uncommon . The moving coil is responsible for the medium and low frequencies, while the moving iron is responsible for the medium and high frequencies. However, in recent years, more and more moving irons for the medium and low frequencies have been developed . Due to innate factors, the dynamic coil will have a boost in the low frequency response , which is helpful when doing ANC , and can also attract European and American consumers with heavy bass, but the disadvantage is that sometimes too much bass will cause a sense of turbidity ( boomy ) . In any case, the moving coil is responsible for the bass, while the moving iron is mainly for the treble. Compared with the moving iron, the bass characteristics of the moving coil are mainly due to the high compliance of its diaphragm. The moving iron is a metal diaphragm . Like electromagnetic and piezoelectric buzzers, the resonance frequency of the metal diaphragm is higher. High, and the sound range is usually higher, so it is more suitable for high frequencies.
moving irons can be inserted is because the moving iron is very small in size and the location of the sound outlet is well designed, so it is easy to stack many moving irons and output sound forward at the same time. Knowles has a background in hearing aids , so the moving iron was originally designed to be placed in a small hearing aid, and it was finally applied to monitor headphones. Recently Knowles also released his own Target Curve , which slightly improves the response in the mid-to-high frequency of Harman Target Curve , especially above 10khz . In this way , the subtle sounds of some instruments such as the high-order harmonics of the violin and the cymbals can be heard more clearly; but some consumers do not like to hear the over-emphasized vocal or airy sound , so the high-frequency response should be strengthened after all. To what extent is still uncertain. What’s more, most people’s hearing degrades with age, and they may not be able to hear sounds after 12khz and 15khz , which is also a consideration.
At this time, you can upgrade Hi-Res and other requirements to the concept of 40khz . Up to now, whether the human ear is sensitive to 20~40khz audio is still debated. I remember that Sabine CEO mentioned in a speech that what people hear is actually the intermodulation distortion (intermodulation distortion) generated by the interaction between the 20~40khz frequency band and the frequency band below 20khz , and they mistakenly regard this as part of high-definition music. But in any case, whether people can perceive it consciously, maybe people know these frequency bands subconsciously, and have a certain impact on psychology. At present, psychoacoustics and music therapy are gradually becoming popular, such as alpha, beta, gamma waves, etc. have been proven in clinical experiments to activate some nerves and brain regions, and there may be more detailed studies in the future.
Directivity
The directivity needs to be higher or lower, there is no certainty, it depends on the application scenario. Directivity is generally required in personal wearable devices or in personalized spatial fields. For example, cars now require each passenger to be able to listen to their own audio independently, which is the concept of sound bubbles . Such a concept requires the seat back, armrest, door, roof and front area around the passenger to form a bubble area, and use the speaker array to confine the sound.
Narrowing down to personal wearable devices, such as smart glasses, also have the function of sound bubbles. For example, when the user is talking, it is best not to let the people around him hear the broadcast content. Therefore, in order to strengthen the audio privacy (audio privacy) , sometimes a few more holes are opened in the glasses mechanism, so that the sounds of a specific frequency band ( usually the voice band ) interfere with each other and offset the external impact of this part of the sound broadcast. spread.
In general large-space PA occasions, lectures, performance venues, and living rooms at home, omnidirectionality is generally required. That is to say, no matter where the audience sits, they hear the same sound. In addition to using arrays to achieve omnidirectionality, sometimes planar radiators (panel radiators) or commonly known as planar speakers are also used. Planar speakers assume that the entire area is evenly stressed, so it is equivalent to emitting a planar sound wave wall that spreads evenly to the audience area. In contrast, the general cone diaphragm , although it is not easy to deform due to bending, the force point of the cone diaphragm is usually limited to the contact area of the inner voice coil , and the sound emitted is closer to the point sound source, and the spherical surface emitted by the point sound source The sound waves will change significantly according to different angles and distances.
Ease of Use
Generally, condenser microphones for recording need 48V phantom power (phantom power) to generate bias voltage, which sometimes causes trouble, so sometimes pre-polarized (pre-polarized) or so-called electret (electret) is used. ) microphone, the internal capacitor has been pre-polarized, and no external power supply is required for use. The same principle also applies to piezoelectric MEMS speakers. Generally, such speakers require a DC bias voltage, but such bias voltage is not necessarily used for polarization. Instead, it is used to drive the diaphragm to a working point in advance, thereby reducing the Distortion, such techniques can be found in USound .
In MEMS microphones, the grounding of GND can also be seen by the manufacturer’s care. Some DMCs have more than one pad for GND , maybe two or even four. For example, the pads of Knowles are connected to each other when they are connected to each other. That is to say, even if the customer is only grounded to one pad and the other pads are not grounded, it will not cause signal noise. However, if some brands do not pay attention to this internal detail, if the customer accidentally misses grounding a pad , the signal quality will be greatly damaged. This is a case that I personally experienced. The representative said that it is very important for manufacturers to make customers ” foolproof ” design.
The appearance of some products is also difficult to do institutional integration. For example, the diaphragm of some earphones is required for sound quality, so the diaphragm protrudes from the frame. It is not flush with the frame, nor is it lower than the frame. This will cause the mechanism to easily damage the floating diaphragm during the design of the opponent or during assembly , causing troubles. In large stereos, if the diaphragm floats out, it is not so serious, because the diaphragm of the audio speaker usually radiates directly into the external atmosphere and will not interfere with the counterpart mechanism.
High Reliability
For example, when the microphone is receiving sound, it needs to be able to resist the noise caused by the surrounding interference and vibration. Even if the microphone is accidentally dropped or hit, it should not affect the sound reception. In this environment, it is suitable to use a dynamic microphone instead of a capacitive one . microphone. The sensitivity of dynamic microphones is generally low, and the dynamic range margin (headroom) is usually high, which can tolerate some sudden loud sound waves. It is also possible to use piezoelectric microphones, such as piezoelectric MEMS microphones , which are often used as waterproof microphones for underwater acoustics. There is dynamic margin on the microphone , and it is desirable to have dynamic margin control when the speaker is operating, but this is usually controlled electronically, such as Dynamic Range Control (Dynamic Gain Control, DRC) and Automatic Gain Control (Auto Gain Control, AGC) , etc., It is to limit the amplitude of the loudspeaker not to exceed the linear point or the overload point.
When used at high temperature for a long time, or when a surge occurs, the speaker needs to be able to cope with the ensuing heat. At this time, our magnetic circuit and basin bracket must be properly dissipated. We can use materials with better heat dissipation, or control the size of through holes . The voice coil used can also increase the heat resistance level, such as changing from ordinary to large black wire, etc.
Low Power Consumption
In terms of microphones, piezoelectric MEMS microphones have lower power consumption because their capacitive ceramic piezoelectric sheets have a very large impedance, so they are often used in portable devices maintained by batteries. Before there is no voice wake-up, the piezoelectric MEMS mic ( digital ) is in a dormant state, and it will enter the operating state instantly when it is triggered, and the power consumption in the dormant state is very low. As far as speakers are concerned, the general rated power and DC impedance are fixed, so it is difficult to reduce energy consumption. In addition , in the field of electro-acoustic ICs , the power consumption of ICs continues to decline with the advancement of wafer technology .
Add Audio Features
There is a so-called target curve for headphones with a neutral frequency response in the market , but it is not necessarily the one that most people like. This type of neutral curve is often used from the measurement results of speakers with a flat frequency response in a free sound field or a diffuse sound field, without emphasizing any frequency band. But people usually have specific preferences. For example, American customers prefer heavy bass rhythms, and European classical music may maintain the subtle sounds of treble and other instruments. Even each brand has its own frequency response curve target, and the foundry needs to help them reproduce it, which violates the principle of the so-called neutral curve. Knowles may have recently published a modified Harman Target Curve because they want to promote their moving iron , and instead emphasize the frequency band above 10khz . At present, consumers such as earphones and car users can adjust customized EQ curves on the App or the car’s central control panel.
Introduce specific even-numbered harmonic distortion: In order to increase the richness and level of audio, some designers will deliberately introduce distortion into the transducer . This distortion is usually an even-numbered power, which can elevate the sound to another level. Conversely, odd-numbered distortions are less pleasing to the ear, and usually overclipped waveforms such as square waves will imply many odd-numbered harmonics , which sound very harsh to people .
In the performance field, if singers or instrument players want to increase the low-frequency components to make the sound more vigorous and full, they can move closer to the microphone of the directional microphone . This is the application of the proximity effect .
Miniaturization
In terms of speakers, one of the most obvious examples is the development of headphone monomers from 10phi all the way to 6phi or even below 5phi . As far as headphones are concerned, the advantage is to make the overall headphones smaller, or to have more space for batteries or mechanical tuning, but the disadvantage is that such output and low frequency may be limited. In the category of micro speakers, the length and width are usually fixed, and the direction of efforts is usually to reduce the thinness to a lower level, which may be within 2~3mm . This is because 3C products such as mobile phones, watches, glasses and notebooks are becoming thinner and thinner . In terms of microphones, for example, MEMS mic has shrunk from 4*3mm to 4*2 mm, and now it has shrunk to 2.5*2 mm or even 2*2 , etc., all of which are the direction of miniaturization.
Water Proof
For speakers, common waterproof methods are to make airtight speakers , use waterproof glue, and coat the diaphragm with waterproof materials. In the case of microphones, damping is usually connected , or waterproofing is done on the mechanism of the end product, for example, the sound hole is facing downward , the sound pipe is misplaced , the sound pipe is tortuous, the drainage leak is made, or the sound hole is reduced. Waterproof generally includes IPx5, IPx6, IPx7, IPx8, etc., and there are settings for resistance to different atmospheric pressures and water depths. If the speaker or microphone body is waterproof, Haocheng will provide the corresponding damping and deliver it together. If the sound hole of the end product mechanism is waterproof, we will recommend a suitable supplier’s part number or purchase it on our behalf, such as Saati , Gore or Nitto, etc.
Assembly Difficulty
As the industry invests in automated processes, surface mount SMT technology is becoming more and more popular, so customers will request some SMD components. In terms of microphones, in the past, condenser microphones ( ECM) evolved from DIP parts to SMD parts of MEMS mic . As far as the buzzer is concerned, whether it is electromagnetic or piezoelectric, it has also evolved from the past DIP parts to SMD parts. This is not the case with speakers, because the speakers are usually installed on the speaker, far away from the main board of the electro-acoustic product , and it is unlikely to be integrated on the same board.
For products that can be converted into SMD , it should be noted that in addition to changing the structure of the pins to pads , it is also necessary to consider the heat resistance of the glue inside, the plastic case, and the diaphragm . After all, the peak temperature of reflow soldering generally reaches about 250~260 degrees Celsius, and the material cannot be deformed or damaged during the reflow process. Some reflow soldering requirements now require double-sided boards, plus one rework (rework) , so SMD components ideally need to be heat-resistant for three times . Furthermore, if SMD components can be loaded automatically , their own manufacturing process is usually also automated. Therefore, how to change a single production line from manual to automatic requires considering the number of shipments of this model, the importance of customer strategy, the scale and time period of the project, and whether the benefits can be spilled over to other models to evaluate SMD .