Speaker design and technology

The closed box technology was first invented by AR company. In the early days, it had always adopted a baffle design. Only after AR invented the so-called "air cushion" patent (i.e. closed box) did it truly become a speaker. The characteristics of the closed box are good low-frequency power, quick response, clear and powerful low-frequency, but the diving depth is limited and the low-frequency volume is insufficient.

The transmission line box uses the principle of the sound duct to make the backward sound waves emitted by the unit pass through a section of the sound duct and then be superimposed in phase with the forward sound waves at a certain low frequency, so as to obtain a better low-frequency diving depth. However, the length of the sound duct must be half of the wavelength of the sound wave of tens of hertz, which is about several meters long. In this way, although the phases between the backward sound wave and the forward sound wave are in phase, there is a time delay of tens of milliseconds. Therefore, although the low-frequency volume of most transmission line boxes is good, the transient is poor, and the low frequency is relatively turbid and weak. Looking at the current world Hi-FI market, transmission line designs are basically rarely seen. The author's personal opinion is that transmission line boxes may gradually disappear. The starting points of the bass reflex box and the transmission line are the same. Both make full use of the backward sound waves to enhance the low-frequency dive, but the bass reflex type uses the bass resonator method to replace the distributed parameter bass resonator method of the transmission line half-wavelength sound guide tube. Although the sound waves are also inverted, the additional time delay is only about ten milliseconds, and the bass reflex type has a simple structure, easy to process and easy to adjust. Therefore, the bass reflex type and closed box have become the two major designed systems. Other common ones include the empty paper cone type (passive radiation type), whose principle is the same as the bass reflex type, and it is beyond the scope of this article, so it will not be introduced.

Compared with closed-type speakers, the bass reflex speakers have their own advantages. The advantages of closed-type speakers are power, good transient response, and clarity, which are excellent for listening to classical music, especially chamber music. Therefore, British ATC and others prefer closed-type speakers. The advantages of bass reflex speakers are deep low-frequency dive, sufficient volume, high power handling, and high sensitivity. They are more comprehensive and can be used in various occasions. The advantages and disadvantages of closed-type speakers can only be determined by personal preference.

The design of the No. 1 monitor speaker adopts a phase-inverted design to achieve low-frequency diving depth and adaptability (in fact, it can also be a closed box). In order to ensure that there is no power compression or airflow noise under low-frequency high sound pressure, the diameter of the phase-inverted tube should not be too small. In actual design, the calculation formula is only a reference, and the final adjustment should be made by comparing the sound with the ears.

After designing the size of the box, we need to consider the requirements for the box production: the first is the selection of sound-absorbing materials in the box. Different sound-absorbing materials can significantly change the effective volume of the box. Some audiophiles are misled by some domestic audiophile articles and superstitiously use glass wool or asbestos as sound-absorbing materials. Although glass wool has a high sound absorption effect and has the advantage of adiabatic expansion, in fact, unless the unit is not suitable and the designed box volume is too large, such as some low-end high-Q, high equivalent volume speakers in China, glass wool is needed to reduce the actual volume of the box. There is no great advantage in using glass wool, and the use of glass wool in speakers. Using ordinary sponge, acrylic cotton or long wool cotton, it is very convenient to design a speaker with excellent sound quality. Since the design volume of Monitor No. 1 is about 18 liters, the medium-bubble through-hole sponge was selected during the design.

The selection and processing of the cabinet board has a huge impact on the sound quality of the cabinet. The requirements for the board are considerable rigidity and internal damping attenuation characteristics. At present, the better choice is medium-density fiberboard (MDF), but the quality of domestic medium-density boards is poor, and high-density boards are required to achieve the effect of imported MDF. However, the internal damping effect of medium-density boards is still insufficient, and we use the method of attaching specially formulated asphalt boards to the inner wall to make up for it. The damping characteristics of this asphalt board are relatively stable all year round, and the damping degree is excellent. First, apply a layer of woodworking glue on the cut box board, place the asphalt board in the center, and nail the big-head nails 4-8cm apart. In this way, the box board and the asphalt board become an organic whole, which can basically achieve the ideal of "only the speaker makes sound, and there is no sound from the box board."

After talking about the design of the cabinet, we have to talk about the design of the crossover, the heart of the speaker. There is no doubt that the crossover is the soul of a pair of speakers, it directly determines the tone of the speakers, but some audiophiles in the past have some mechanical understanding of the crossover. For example, someone once wrote an article "denouncing" the first-order crossover (-6dB/oct), believing that the ideal filter with infinite attenuation is the best choice, that is, the higher the order of the filter, the better, and the greater the attenuation slope, the better. This view only superficially sees the role of the crossover "division", and secondly, mechanically sees the problem that the crossover is composed of high-pass and low-pass filters.

Regarding the frequency divider, I would like to talk about a little philosophy. China once unilaterally promoted "one divided into two" and ignored "two combined into one". In fact, I think that "one divided into two" is the surface, and "two combined into one" is the essence. The timbre of the tweeter is bright, which can be likened to "yang", and the timbre of the bass unit is deep and deep, which can be likened to "yin". Yin and yang have opposing sides, but also mutually inclusive sides. Therefore, the ancient Tai Chi black and white fish pattern is not as simple as a circle split into two halves, but the tail of the yang is the head of the yin, the head and tail are connected, and they are mutually generated and transformed. Moreover, the eyes of the black and white fish are the yang in the yin and the yin in the yang, so-called the yin and yang. Back to the frequency divider, its role is not so much to divide yin and yang, but to blend and melt yin and yang into one furnace. Imagine that if the woofer is the only sound below the crossover point, and the tweeter is the only sound above the crossover point, then the timbre will be split, and the sound and image positioning will jump up and down, and it is impossible to achieve a good sound quality. On the contrary, if the unit quality is of good quality, the woofer is the main part below the crossover point and the tweeter is the auxiliary part, and the opposite is true above the crossover point. At the crossover point, due to the ingenious scheduling of the crossover, the two "yin" and "yang" of the tweeter and woofer units with different timbres are organically integrated into one, and a better timbre can be obtained.

From the above analysis, we can see that, under the premise of guaranteed unit performance, a lower-order crossover is definitely better than a high-order crossover, especially considering the larger phase change and time delay caused by high-order filtering, the advantages of low-order filtering are self-evident. Of course, low-order filtering has very high requirements for the unit. If the crossover point is low and the filter order is also low, the power burden of the tweeter will be greater. This is one of the reasons why the tweeter filter uses a second-order filter and the bass uses a first-order filter in the design of Monitor No. 1.

The decision of the order of high and low pass filtering must also take into account the phase connection of the unit itself at the crossover point. The advantages of low-order filtering are self-evident, as the phase difference between high-pass and low-pass filtering can be used to compensate for the large phase change and large time delay caused by unit filtering. Of course, low-order filtering has very high requirements for the unit. If the crossover point is low and the wave order is also low, the power burden of the high unit will be greater. This is one of the reasons why the high-frequency filter uses the second order and the bass uses the first order in the design of monitoring things.

The decision of the order of high and low frequency filtering must also take into account the phase connection of the unit itself at the crossover point. The phase difference between the units is compensated by the phase difference of the high and low pass filters to obtain a relatively flat phase-frequency response, and a flat phase-frequency response is a necessary condition for the pure and natural mid-frequency sound. I recommend that when N601 and T301 are matched, when the crossover point is within 2.5-3KHz, it is better to use a high pass filter that is one order higher than the low pass filter.

The second is impedance compensation. Since the role of bass inductor at the crossover point cannot be ignored, it is better to add an appropriate impedance compensation circuit. However, the compensation must not be excessive, otherwise it will cause a significant degradation to the tone. It is best if the impedance after compensation is not lower than the minimum impedance value of the impedance curve before compensation. For the adjustment of treble sensitivity, an L-type resistor attenuation network is preferred. A single resistor attenuation will not save much cost and has no benefit to the tone.

Finally, I have to mention that bi-wiring is definitely the best. As for the problem of too much treble after some friends said that some single-wire cabinets were changed to bi-wiring, it is because the original design was single-wire. In order to offset the masking effect of bass crosstalk on treble, the sensitivity of treble was increased a bit. After changing to bi-wiring, the treble attenuation should be increased a bit. For those designed for bi-tone, there is no such problem. Regarding bi-wiring, you can refer to one of my articles in "Audio World".

After the designed speaker is made into a sample box, a series of tests and adjustments need to be carried out, including impedance test, amplitude-frequency test, phase-frequency test, distortion frequency response curve, water drop diagram or transient response, etc., and improvements should be made if problems are found. As a professional designer, human ear calibration adjustment must be carried out only after no problems are found in the test, because there is no art without a scientific foundation, "without the skin, where can the hair be attached". For amateur production, if there is no test condition, this can only be left vacant. You must carefully study the test curve provided by the manufacturer, and thoroughly understand the original design, and spend more time on the human ear calibration to make up for the lack of test conditions.

We have talked a lot before, only the technical part of speaker design, for high-end Hi-Fi, this may only account for 40%, while the art part such as human ear tuning accounts for 60% of a good design. Although the technical parts discussed above are complicated, they can be traced after all. As long as you work hard, you can master them thoroughly. The part of "human ear tuning" requires not only frequent listening, rich experience in live music, and a deep understanding of high-end and well-known equipment, but more importantly, it requires a deep cultural and artistic foundation, and understanding of philosophy may be helpful. As the saying goes, "the effort is outside of poetry", only learning to write poetry can only make you a "poet craftsman", only by learning various subjects such as humanities and geography can you combine the strengths of various schools into one furnace and create a brilliant "poetry talent". By analogy, we may say that "the effort is outside of Hi-Fi", which only refers to tuning, and it is a top requirement (for general Hi-Fi, perhaps only appropriate technical accumulation and Hi-Fi experience are needed).

Regarding sound calibration, in order to obtain the neutrality pursued by Hi-Fi, it is necessary to use a wide range of various types of music for sound calibration, because a type of music often only covers a certain frequency range or a few frequency ranges. Using various types of music for sound calibration can adjust the entire frequency range to the best possible level. Among the music signals, I personally feel that the human voice may be the most complex and difficult to reproduce. There is an old saying that "silk is not as good as bamboo, and bamboo is not as good as meat", and this "meat" refers to the human voice, and at the same time, the human voice is the most familiar and most picky. It is very important to reproduce the human voice well. And the human voice is very sensitive to the flatness, balance, and flatness of the mid-frequency band, so the human voice is often one of the most commonly used means for designers to calibrate the sound.

When listening to vocals, pay attention to the lip shape, positioning, and flavor. The clear focus of the lip shape and the accurate and stable positioning are closely related to the flatness of the frequency response and the balance of the left and right channels. The flavor of the human voice is closely related to the flatness of the phase curve. Phase design is a technology that has just emerged internationally in the past decade. The pure and beautiful vocals of the Monitor No. 1 are due to the contribution of the unit quality and the phase design.

Others, such as using cello, cannon sounds, and drum sounds to calibrate low frequencies, violin solo to calibrate mid-high and high frequency tuning, and percussion cymbals to calibrate ultra-high frequency overtone resolution. Large-scale symphonies can calibrate the width and depth of the sound field, as well as the sound and image positioning in large dynamics and complex scenes. The piano has a wide range and rich overtones, and the texture of its sound particles and the timbre of overtones have extremely high requirements for the flat response of each frequency band.

When conditions permit, you can use some internationally renowned and well-reviewed speakers as reference for sound calibration, such as the high and low frequencies of ATC20, or its balance, Harbeth's 3/5A for vocals, ATC and Alon's large-scene reproduction, etc.

Here I would like to remind audiophiles that sound tuning is an extremely complicated task. First, it must be calibrated repeatedly. Second, testing must be done at the same time as the sound tuning. The speakers must not deviate from the theoretical guidance. Sound tuning and testing must be carried out alternately to achieve a high degree of unity between art and theory.

The characteristics of the unit used in Monitor No. 1 have been introduced in many articles, and the product manual also gives the parameters and characteristics, so I will not discuss them here.

Having written this far, I want to reiterate an old tune. I believe that in the Hi-Fi audio industry, we Chinese can absolutely be at the forefront of the world. We can fully grasp some key technologies and develop some advanced technologies of our own. It can be said that it will not take too long for us to catch up with the world's advanced level in technology as long as we work hard. The accumulation of our five thousand years of long culture has given us some advantages in our understanding of Hi-Fi that others do not have, especially the traditional Chinese cultural and philosophical thoughts and the modern Hi-Fi industry have the right time, place and people.

We are determined to produce world-class Hi-Fi products. This is not because we are arrogant, but a realistic goal we set after fully considering the factors mentioned above. Of course, to achieve this goal, our efforts alone are far from enough. It must rely on the care and support of the majority of audiophiles, and also on the emergence of a group of serious Hi-Fi manufacturers rather than a few in the country to form a good industry atmosphere.

We hope that through this event, we can strengthen our ties with audiophiles, gain more support from them, and enable us to take another big step forward on the arduous road to realizing our ideals.