Acoustics on dual dynamic heads. Two-way speaker systems in an open design. Speaker drawings

The main resonance frequency is the frequency at which the total electrical resistance of the coil increases to its peak maximum;

Quality factor of the electromechanical loudspeaker system. This is a very important characteristic. It shows the degree of inertia of the system - both mechanical and electrical, and determines the rate of attenuation of free oscillations of the monitor;

Nominal frequency range, i.e. frequency region in which the loudspeaker operates to the standard;

Average sound pressure is the pressure developed in a certain frequency range and at a certain point in the sound field when a certain electrical power is supplied;

Characteristic sensitivity;

Frequency response unevenness is the difference between the maximum and minimum pressure in the nominal (or, if necessary, in any other) frequency range. For good loudspeakers it does not exceed 3-4 dB;

Frequency response - graphical representation of the previous parameter;

Directionality - a change in pressure when deviating from the working axis by a certain angle at a constant distance from the center;

Harmonic distortion (usually the 3rd harmonic and higher) - the level of harmonics, expressed as a percentage, that appears when a pure sinusoidal signal is fed to the loudspeaker, in which there are no harmonics;

Intermodulation distortion factor. We need to say more about this parameter. Suppose a signal containing two frequencies, 100 and 1000 Hz, is sent to a loudspeaker. As a result of the interaction of these frequencies, combination frequencies arise (sometimes incorrectly called combination harmonics) with frequencies corresponding to the difference or sum of the upper frequency and a multiple of the lower frequency - in our case, 800, 1200, 600, 1400 Hz, etc. The lower the overall level of these frequencies, the better. An ideal loudspeaker should not generate these frequencies at all, or any others not present in the original signal.

Of the several power parameters, the most important are the following:

Rated power - power at which nonlinear distortions do not exceed a specified limit;

“musical power”, also called “nameplate”, “maximum noise”, “continuous”, etc. - power in a certain frequency range that a loudspeaker can withstand a real or broadband noise signal without damage for some time;

Peak (maximum short-term) power - the power that a loudspeaker can withstand a noise signal for a short pulse (from 0.01 to 1n) without damage;

Horn emitters. The main disadvantage of direct radiating loudspeakers is their extremely low efficiency. The reason for this is the mismatch between the resistances of the mechanical system and the environment. To increase the radiation resistance, it is necessary to increase the size of the emitter, but this will entail an increase in the mechanical resistance of the emitter mass and will not provide a gain in efficiency. Since the diffuser performs two functions: converting mechanical vibrations into acoustic ones and radiating these vibrations into the environment, such a contradiction can only be resolved by separating these functions, which is carried out in horn loudspeakers. The horn also serves to match the resistance of the mechanical system and the environment. A horn is a pipe with a variable cross-section. The inlet of the radiating horn (throat) is smaller than the outlet (mouth). The output hole is the emitter, and the input hole is the load for the mechanical system. Thus, the emitter can be made as large as desired, and the mechanical system can be made small and therefore lightweight.

Types of horns: a - dual; b - sectioned.

Horns are used with different laws of change in cross-section. The most common horns are exponential; Conical ones are used less frequently, since they have a much less uniform amplitude-frequency response. For sharp directivity and a lower limit of the transmitted frequency range, the output aperture of the horn should be increased and a horn of greater length should be selected. To increase the length, the horn is often rolled or folded. We encounter similar phenomena in wind musical instruments: the lower the register of the instrument, the longer its horn.

To concentrate or distance sound waves, acoustic lenses are used, based on the refraction of sound rays when passing from one medium to another with different propagation speeds (for example, the speed of propagation of sound waves in porous materials or in grilles and louvres of plates differs from the speed of propagation in open space) . The disadvantages of the horn include nonlinear distortions caused by the large magnitude and sharp change in the amplitude of sound pressure within one wavelength in the throat of the horn, as well as frequency distortions in conical horns. Horn electrodynamic loudspeakers have two design options: narrow-neck and wide-neck. The area of ​​the horn inlet in narrow-neck loudspeakers is several times smaller than the area of ​​the piston diaphragm; in wide-neck loudspeakers, these areas are either the same or close to each other.

These are the main technical parameters of loudspeakers. It should be noted that passport data should be handled with caution. Some manufacturers sometimes name, for example, the range of reproduced frequencies without indicating the unevenness of the characteristics; in this case, it may turn out that the declared lower threshold of 25-30 Hz is ensured only when the pressure drops by 10 dB or more, which is actually a falsification.

I would like to note that in the 80 years since the invention of the dynamic, the task of conveying the sound of a symphony orchestra, ensemble, voice, etc., one can only be surprised and admire the genius of the design of the loudspeaker itself, audio technology has come a long way: from the phonograph to the DVD - and the loudspeaker is structurally fundamental hasn't changed. Only the technology of its manufacture and materials have changed radically. Considering that such a simple design (consisting of just a few elements: a diaphragm, a coil and a magnetic circuit) faces a huge mass-produced acoustic product, billions of which are used all over the world.

Speaker systems

From the characteristics of loudspeakers, let's move on to the acoustic systems made up of them. Unfortunately, domestic terminology has not yet been established and does not correspond to foreign ones. So, in fact, “speakers” in our terminology, especially in old GOSTs, are called “heads”, and acoustic systems are called “loudspeakers”. In modern professional and commercial environments, the term “acoustic system” is used, and household acoustic systems are commonly called “speakers”, and professional studio acoustic systems “monitors”. Some, confused, simply switched to transliteration from English - “speaker”, in their mouths not the chairman of the Duma at all, but the speaker “in general”. At the same time, a low-frequency “speaker” is a “woofer” or “subwoofer”, a mid-frequency one is a “driver”, and a high-frequency one is a “twitter”, but there is also a Russian definition for it: “tweeter” (by the way, the exact translation of the word tweeter).

An ideal speaker system should have only one full-range loudspeaker that reproduces the full frequency range of 20-20,000 Hz. However, since different and often mutually exclusive requirements are placed on a loudspeaker when operating in different frequency bands, it is almost impossible to make such an ideal loudspeaker, at least at an affordable price. Therefore, the vast majority of modern acoustic systems have two or more heads operating in different frequency bands. A low-frequency loudspeaker is always a diffuser speaker, a mid-frequency loudspeaker is also a speaker, but sometimes there are mid-frequency horn-type ones. High-frequency loudspeakers are produced as diffuser, horn and dome (dome, bullet). The two-way system is usually used for so-called “near-field monitors”, i.e. located directly near the sound engineer's head. One speaker in such a system reproduces low and medium frequencies, the other - high frequencies. To separate frequencies, there is a separating filter inside the housing (in foreign terminology, crossover). In this case, the frequency of separation of the input electrical signal for supply to the low-frequency and high-frequency speakers is selected slightly higher than the lower limit of the range of the high-frequency loudspeaker. The RF loudspeaker power rating is also taken into account. 3-band systems consisting of a low-frequency loudspeaker (woofer), mid-frequency (mid-driver), and high-frequency (tweeter) reproduce the audible frequency range much better. Working in a limited range of “own” frequencies improves the sound of low- and mid-frequency speakers and reduces distortion, because The high-order harmonics generated by these speakers are higher than the cutoff frequency of the filter and are correspondingly suppressed.

Acoustic design

P
The front and rear surfaces of the oscillating piston emit oscillations in antiphase: when the front surface at time t 1 creates compression of the medium, then the opposite surface of the piston, at the same moment t 1, creates a vacuum.

Compression and rarefaction spread in different directions (Fig. 18.6). Under certain conditions, bending around the piston, the waves interfere with oscillations arising from the opposite side (phase) and their sum tends to zero. This phenomenon is called acoustic short h short circuit (AKZ). The occurrence of a short circuit reduces the output of the acoustic power of the emitter (piston) in the region of those frequencies at which the emitted wavelength is large compared to the size of the piston (diffraction conditions). This phenomenon occurs at low frequencies of LF sound waves.

H To avoid AKZ at low frequencies, it is necessary to install a screen so that vibrations from the compression area do not bend around the piston and eliminate the phenomenon of interference. The screen is installed in combination with the emitter. This technique is called acoustic screen design (design). The simplest type of design is a shield (Fig. 18.7). To completely eliminate the short circuit, it is necessary to install a shield whose linear dimensions of the plane were greater than half the length of the LF sound wave λ:

d > λ/2;( 6.1.1)

A standard acoustic screen according to GOST 16122-84 has a size of 1350 x 1650 m.

A closed box (CL, Closed Box) is a second-order design (Fig. 6.1.3 A and Fig. 6.1.4). Compared to other types of loaded design, it is less sensitive to deviations in characteristics. Its main advantages: excellent impulse response. This theoretically allows you to obtain a flat frequency response. Disadvantage = low efficiency, which requires increased amplifier power, and increased level of even harmonics due to the asymmetrical load of the diffuser.

A – closed box, B – bass reflex, C – passive radiator

H
The resonance frequency and total quality factor of the head when installed in a closed box with a volume Vc commensurate with the equivalent Vas increase. Thus, when installing a head in a cell with a volume equal to an equivalent one, its resonant frequency and quality factor increase by 1.41 times, in a box with a volume of 0.5Vas = by 1.73 times, and so on.

The next most common type of acoustic design is a bass reflex. Speakers with an Fs/Qts value of 90 or more are suitable for use in a bass reflex. Of all the possible designs of double-action systems, the bass reflex is the most widely used (FI, Vented Box, Ported Box, Bass Reflex). This is a resonant system. The mass of air contained in the FI at its tuning frequency behaves like a diffuser, being a source of sound vibrations. A passive radiator is a type of FI in which the mass of air in the tunnel is replaced by the mass of the moving system of the passive radiator. A conventional dynamic head is most often used as a passive radiator, sometimes with a remote magnetic system.

Structurally, it is made in the form of a closed box with two holes

The emitter (piston) is placed in one hole, the other hole is free, and has a design in the form of a small pipe of volume V. The bass reflex frequency is ƒ f, (Fig. 18.10).

With slow oscillations (8Hz - 10Hz) spring C in (Fig. 18.10). connecting both masses m does not have time to deform, since it has a large elastic resistance z:

z=1/(ω·С in) ; (18.1)

As a result, both masses m p and m b move with the same phase. In this case, the wave emitted by the hole is shifted by 180 o in phase compared to the wave emitted by the piston. An increase in frequency leads to a decrease in the elastic resistance of the air in the box and the spring C in begins to deform. As a result, a phase shift occurs between the oscillations of both masses m p and m b, increasing with increasing frequency and reaching 180 o at the box resonance frequency. Thus, the air in the hole and the piston oscillate in antiphase, and the waves emitted by them will be in phase and interfering and reinforcing each other. The bass reflex resonance frequency ƒ f is, as a rule, chosen equal to the resonance frequency ƒ 0 of the head (piston), i.e. in the low-frequency operating range (Fig. 18.10). With a further increase in frequency, sound emission from the hole does not occur, since the inertial resistance of the air in the hole ω·m in becomes extremely large. At these frequencies, the bass reflex is similar to a closed box. The internal surfaces of the bass reflex, as well as the box, are covered with sound-absorbing material.

Figure 18.11

In the diagram fig. 18.11 power amplifier, which is a signal source for a loudspeaker, with open circuit voltage and the output resistance is converted into a voltage generator simulating a generator with an output value of acoustic pressure, after the generator the total resistance is the sum of the active resistance of the voice coil and the output resistance of the amplifier. M as is the acoustic mass of the moving system, the attached mass of air from the front and back sides of the diaphragm. C a s - acoustic flexibility of suspensions. R as is the acoustic resistance of the moving system. M av is the acoustic mass of air in the phase-inverted tube.

Acoustic load. The diffuser of a dynamic head in a closed design experiences significantly different resistance when moving forward and backward. Load asymmetry is a potential source of nonlinear distortion. Therefore, back in the mid-70s, acoustic systems appeared in the design of which this drawback was eliminated by introducing an additional acoustic load on the front surface of the diffuser. Similar solutions can be used to limit the amplitude of diffuser oscillations in double-acting systems. There are no reliable methods for calculating acoustic load; experimentation is necessary.

Figure 18.12

Acoustic loading can be implemented in various ways. In the simplest case (Fig. 18.12 A), a reflective surface (Reflex Body) is placed in front of the diffuser. However, this solution worsens the sensitivity of the speaker and its frequency response at medium frequencies. In some modern designs, a lenticular-shaped body of rotation is used to improve the frequency response and directivity pattern (Fig. 18.12 B). For the same purpose, you can use a reflective surface located at an angle (Fig. 18.12 B). The wedge load partly acts as a short horn, which contributes to the acoustic amplification of a certain frequency range. As a further development of this idea, acoustic systems with a resonator appeared (Fig. 18.12 D). After this, there was only one step left to take in the design of bandpass loudspeakers.

P
voice loudspeakers. A common feature of all bandpass speaker designs is the presence of one or more resonant chambers and the installation of a dynamic head inside the housing. Since these systems are no longer direct radiation systems, their design and manufacture are very complex. Therefore, mainly fourth-order designs have become widespread (Fig. 18.13 A). Bandpass loudspeakers of the sixth (Fig. 18.13.B, C) and eighth (Fig. 18.13.D, E) orders are less common.

Figure 18.13

Bandpass loudspeakers: A – closed box resonator, B – double-acting bass reflex, C – sequential bass reflex, D – sequential double-acting bass reflex, D – sequential double-acting bass reflex

Bandpass acoustic design is used exclusively for subwoofers. The advantage of a bandpass loudspeaker is its high efficiency, but the impulse and phase characteristics are very mediocre and deteriorate with increasing order. For all designs, except for a closed resonator box, it is desirable to use an infra-low-pass filter (as for a classic bass reflex).

In addition to the considered designs of bandpass loudspeakers with one dynamic head, speakers with two heads are also known. The design is obtained by combining two identical strip systems. One of the chambers becomes common, its volume doubles. (Fig. 18.14 A, B) shows two design options for the fourth order, and Fig. 18.14 B for the sixth order.

ABOUT
One of the advantages of such designs is that they do not require a special monophonic amplification channel: each head can be connected to its own channel of a stereophonic UMZCH.

Figure 18.14

Twin heads. In almost all of the designs considered, dual dynamic heads can be used. To do this, heads of the same type are installed using one of the methods shown in Fig. 18.15. The resulting design can be considered as a new low-frequency dynamic head with completely different properties. The theoretical values ​​of the total quality factor and the frequency of the main mechanical resonance of the resulting system are calculated as the geometric mean of the corresponding values ​​of the original heads. Since when doubling, heads of the same type with fairly similar parameters are usually used, we can assume that these parameters will remain virtually unchanged. However, the bound volume of air enclosed between the diffusers of the heads increases the effective mass of the moving system, lowering the frequency of the main mechanical resonance of large heads to 80% of the original.

Figure 18.15 Installation of double heads: A - face to face, B - back to back, C - in the back of the head, D - with associated volume

Until now, wood remains the main material for the manufacture of speaker cabinets. It is taken into account that wood has its own acoustic properties, and the introduction of its own overtones by the body is undesirable. They are combated both by special damping structures and by using chipboard (chipboard), which we so dislike in furniture, instead of solid “clean” wood. Chipboard does not have any structure (which is the linear fibers of wood), therefore it is less susceptible to resonances. The outside of the chipboard is finished with various coatings, including those imitating wood (veneer), but this finishing is purely decorative.

Along with the traditional use of wood, attempts to use other materials - plastic, metal, stone - continue. There are quite a large number of plastic acoustic systems, usually small in size (near field), which sound quite acceptable and are cheap due to the manufacturability of the housings. However, attempts to create large plastic housings for loudspeaker systems have not yet been successful (from the point of view of acoustics, of course, and not “box construction”). The fact is that a large case must also have a large mass, otherwise such resonances begin to “walk” in it that their suppression is much more expensive than, for example, in a wooden case.

Metal speaker enclosures have been quite effective and have become popular lately. This is due, in particular, to the widespread use in studio practice of computers with traditional cathode-ray picture tube monitors, which are adversely affected by speaker magnets if they are too close. The metal body of the speaker system is in this case a screen. In addition, the metal is easy to manufacture and provides the necessary rigidity for acoustic requirements.

The use of stone also produces interesting results. There is no need to talk about the manufacturability of the enclosures here, but the acoustic results are excellent. However, the problem is solved by a compromise - the use of synthetic material, which makes it possible to combine the ease of production of the body with the massiveness and rigidity of the stone.

However, despite the active search for new materials, the “good old” wood remains the main one.

For a long time, the traditional arrangement of speakers on the front wall of the case in the form of a “snowman” (low-frequency loudspeaker at the bottom, mid-frequency loudspeaker in the middle, and high-frequency loudspeaker at the top) suited users. However, it has been noticed that the distance from the centers of different speakers to the listener is often different, and the sounds from them do not reach the listener strictly in phase. The amount of non-synchrony is extremely small, but the problem exists. The solution was found in various types of so-called coaxial loudspeakers. In the simplest cases, the high-frequency speaker was fixed in front of the center of the low-frequency diffuser cone, but, naturally, without physical contact with it. Another, more complex, but also more elegant way of creating a point emitter was proposed by the famous English company Tannoy. In their now classic system, the tweeter diaphragm is located behind the woofer magnet. In the core of the low-frequency loudspeaker there are channels through which air pressure from the high-frequency membrane passes in the direction of radiation of the low-frequency diffuser, which is also a horn for high frequencies. This way, ideal pinpoint radiation is achieved.

It was previously mentioned that at high frequencies diffusers, especially large ones, vibrate mainly in the central part adjacent to the coil. This property was used to create wideband loudspeakers, popular in professional equipment two or three decades ago and still found today. In these loudspeakers, an additional micro-diffuser was glued into the central part of the diffuser, which worked as a coaxial high-frequency loudspeaker. Of course, the result was far from the quality of real coaxial systems, but the high-frequency response of these full-range speakers did improve significantly.

Modern production is extremely standardized. Standards have also been established for the size of loudspeakers - from small to large. Modern speakers are usually measured in inches, and this is convenient: it gives not only the size, but also, as it were, the “product number”.

Even for powerful acoustics, speakers larger than 21" are not used, and eighteen-inch ones are not often seen. Next in order are 15", 12", 10" and 8".

Mid-frequency - 8", 6.5" and 5". High-frequency - 4", 2.5" and 1.5". However, the dimensions of the diffuser are important mainly for low-frequency loudspeakers, directly affecting the lower limit of the range and the sound pressure level.

The real sound picture can only be presented by large acoustic systems (control monitors) of the “far field”, sounding evenly over the entire frequency range and not overloading at the recommended listening level (about 90 dB).

Directivity characteristics

As follows from the theory of acoustics, the ideal source of sound is a “point” emitter, that is, an emitter whose dimensions, compared to the length of the sound wave emitted by it, can be neglected. Unfortunately, real acoustic systems are very far from such an ideal emitter and, moreover, have different radiation patterns for different frequencies of the sound signal. The width of a loudspeaker's radiation pattern is determined by the ratio of the wavelength of the sound signal emitted by it and the geometric size (diameter) of the loudspeaker cone. In addition, the radiation pattern in the region of the joint action of radiation from two loudspeakers depends on the mutual phase shift of their signals, determined by the separating filter circuit of the speaker system.

Today in “speaker building” there are two approaches related to the directionality of acoustic systems. Adherents of the first of them argue: the system must be highly directional in order to eliminate harmful sound reflections. According to this logic, highly directional speakers are required to deliver sound information exactly to the listening area without unwanted “impurities” in the form of reflections from walls and various objects. Well-known examples are speakers built on highly directional coaxial drivers (Tannoy, KEF). Coaxial two-way radiators are mid-frequency and high-frequency loudspeakers assembled on a single magnetic system. The dome “tweeter” is assembled on the internal core of the magnetic system and is located inside the cone diffuser of the mid-frequency loudspeaker, which is a kind of horn-sound guide for sound waves emitted by the “tweeter”. Such radiators have a number of unique features that significantly distinguish them from the mass of other loudspeakers. Firstly, thanks to the design used, the emission centers of the HF and midrange loudspeakers are located practically at the same point, which eliminates the occurrence of phase and time distortions of the signals emitted by them. Secondly, since the radiation of medium and high frequencies is physically carried out from one point in space (conditionally), Uni-Q type emitters have a good radiation pattern at these frequencies due to these serious advantages, the sound of speaker systems with coaxial emitters is characterized by excellent localization of sound sources in space. In European speakers, there are D "Appolito schemes, in which the tweeter is located between two identical low-frequency/mid-range heads - this sharpens the directivity at a number of frequencies, reducing the number of sound reflections from the floor and ceiling. In expensive speakers, sometimes there are entire garlands of tweeters designed for jewelry focus high frequencies A diametrically opposite approach is omnidirectional loudspeakers, or acoustics with circular directivity. Such loudspeakers, by virtue of their design, are completely.

Recently, acoustic systems with open acoustic design - shields or shallow open boxes - have gained recognition among some radio amateurs. Even industrial acoustics have been produced using this design, which have been highly praised by experts. The photo shows a famous Jamo R909 system.
Some of the problems with this solution are outlined in the article, my translation of which is given below.

Preface

Closed designs made it possible to sharply reduce the volume of speakers and radically expand the frequency range downwards. The industry has almost completely switched to producing speakers specifically for closed designs. Entire generations have grown up who have heard nothing but ZY. However, many people think that “they threw out the baby with the bathwater” because they believe that the sound of the middle frequencies, the main ones for the perception of frequencies, has deteriorated.

Therefore, among radio amateurs and some acoustics manufacturers, interest in open acoustic designs has again appeared (hereinafter, for simplicity, we will call them OY). The problem is that today they practically do not produce special speakers for OY because they are in low demand, small companies can produce them for amateurs, but due to the small circulation they will be expensive.

I would like to bring to your attention my free translation of the article by Martin J. King “Designing a Passive Two Way Open Baffle Speaker System”. I think the problems raised and their solutions will be interesting.

--
Thank you for your attention!
Igor Kotov, editor-in-chief of Datagor magazine

Source site (En): by Martin J. King

My comment to the article

The lack of measurements using a microphone and specifically at home causes skepticism. It would be interesting to know the impressions of independent listeners-experts who have not “processed” the constructions by the author. But these are just my dreams.

Reader vote

The article was approved by 47 readers.

14.09.2012

We make our own acoustic speaker with two bass and midrange heads

A private material has been published about how to assemble an acoustic system with dual midrange or woofers. Everything is described in simple language for beginners, but our opinion remains the same as it was, it’s better to buy from us.

Making your own acoustic speaker with a dual head

A double head has some advantages over a single head. For example, it has a smoother amplitude-frequency response, less nonlinear distortion, and the required volume of the acoustic design box is also smaller.

The amplitude-frequency response is smoothed out since the heads that make up the dual head are mutually damped. Each single head, within the limits of permissible deviations, has its own uneven frequency response, due to production technology, so that the frequencies of peaks and dips in the frequency response do not coincide. Some of these peaks with dips in the dual head are mutually compensated. See the diagram in Figure No. 1.

Here, nonlinear distortions are reduced, because the dual head is a symmetrical electro-mechanoacoustic system, unlike a single one. Because of this, the air resistance on both sides is almost the same. It is due to the design features of the head and the properties of the material. Some types of heads do not have any difference in suspension flexibility when the diffuser moves forward or backward. In the double head, the asymmetry of the distribution of magnetic induction, which is in the gap of the magnetic system and negatively affects the level of the 2nd harmonic, also does not appear.

The low-frequency section requires one powerful dual head. It can be placed on a horizontal board, under which there is a horn, which directs the sound to the listener and matches the mechanical resistance with the air environment of the moving head system.

The volume of the box is reduced, because the resulting flexibility of the suspension of such a head is reduced by half compared to a single one. And the mass of the movable system of the double head increases by the same amount. That is why the frequency of the main mechanical resonance does not change.

It may seem that increasing the number of heads that work per speaker hole makes it possible to reduce its dimensions to an even greater extent. But it is practically impossible to bring the heads so close that the geometric dimensions do not affect the phase shifts of the sound waves emitted by the outermost heads. The length of the wave propagation path, if we count from the innermost to the outermost head, is commensurate with the lengths of the emitted waves. And this ultimately leads to the sound signals being subtracted and distorted. By the way, this is why you cannot double mid-frequency and high-frequency heads. In addition, in this case there will be a noticeable decrease in efficiency.

So, we offer our readers an AC loudspeaker-phase inverter, the useful internal volume of which is 50 liters. Here a dual head made of 6GD 2 is used as a low-frequency emitter. And as mid- and high-frequency ones, 15GD-11 and 6GD-13 are used, respectively. You can also choose other speakers. A double head is installed on an inclined board, because this arrangement of a board with a double head allows for more efficient use of the volume of the box, and this already allows reducing the dimensions of the speaker and its weight.

The main technical characteristics of the speaker are as follows:

Rated power is 12 W

Nameplate power - at least 30 W

Nominal frequency range - 30-18000 Hz

The nominal electrical resistance here is 4 ohms.

Since highly efficient 6GD-2 low-frequency heads are used, with a low rated power (namely, 12 W), the sound volume is not inferior to industrial speakers with a power of 30 W. And if we talk about sound quality, most people prefer the speakers, which are described below. The circuit diagram of the speaker is shown in Figure No. 2, and the design is shown in Figure No. 3.

The box of the speaker (3) is made of chipboard, 2 cm thick, covered with paper that imitates valuable wood species. In it, the double head (17) is fixed on the board (10), and the mid-frequency head (12) and high-frequency head (16) are fixed on the front wall (4). As for the back wall (15), it is removable. The mid-frequency head is isolated from the box by a box (13), which is made of 1 cm thick plywood and secured to the wall (4) using corners (11) and screws.

The output hole at the horn of the dual head (17) is closed with a grille (details 1, 2), and the holes that are located opposite the mid-frequency and high-frequency heads are covered with convex metal meshes (6 and 8, respectively) together with ring decorative frames (5, respectively). and 7). The frame (1) was bent from an aluminum alloy strip with a cross-section of 5 x 20 mm. The rods (2) have a diameter of 4 mm. They are made of stainless steel, then inserted with glue into holes that are drilled in the top and bottom of the frame at 2 cm intervals.

The holes for the bass reflex tunnel and the ring frames of the holes for the remaining heads are bent from a strip of the same material with a cross-section of 5 x 10 mm. The frame of the mid-frequency head (5) is attached using four studs with MZ threads, which are inserted with glue into holes with a diameter of 3.2 mm and a depth of 7 mm. Holes are drilled in the end of the ring on the side facing the panel (4). Before cutting a hole for the head (12), it is necessary to select a groove 20 mm wide and 2-3 mm deep in the front wall according to the outer diameter of the frame (i.e., 5) using a round cutter with a cutter or chisel. When starting to assemble the structure, first secure the head (12), then secure the mesh (6) using wire brackets or nails, and only then install the frame (5), which additionally presses the mesh to the panel (4). Also, the frame (7) of the high-frequency head (that is, 16) is fixed in the groove of the front panel using glue.

To give the speakers a normal appearance, the outer ends of the frame (1), frames (5, 7, 9) need to be polished to a shine, and their side surfaces (both internal and external) must be painted black. The metal meshes (that is, 6, 8) also need to be painted black, as well as the internal surfaces of the bass reflex tunnel, the surfaces of the horn of the double head, also the diffuser holder of the lower head, namely 6GD-2, and also the entire area of ​​the circle that is under the mesh ( 6), the part of the head diffuser holder (12) facing the listener, the heads of the screws that secure it.

Coils L1, L2 from the separating filter are wound on frames using PEV-2 1.3 wire. The frames have a diameter of 35 mm and a length of 100 mm. Each has approximately 460 turns (namely 6 layers of about 75-76 turns).

As for capacitors C1, C3 - these are MBGP, MBGO and the like.

When mounting the speakers, you need to pay attention to the polarity with which the 6GD-2 heads are connected, because if there is an error, an acoustic short circuit occurs. In this case, the outer head is BA1.

To improve the damping of the dual head, you can cover the inside surface of the speaker box with sound-absorbing material or upholster it with this material.

You can replace the 6GD-2 head with an 8GD-1, replace the 15GD-11 head with a 4GD-8 or 5GDSh-5-4, and replace the 6GD-13 head with a 3GD-2. With this replacement, the dimensions of the box are maintained.

Several years ago, radio amateur A. Zhurenkov proposed using dual heads to reduce the lower limit of the frequency range reproduced by a loudspeaker. Unfortunately, this method of extending the range towards low frequencies has not received widespread use in amateur radio practice. And this is most likely due to the lack of an available method for calculating speakers with dual heads. The article makes an attempt to fill the gap and give radio amateurs some recommendations for calculating loudspeakers with dual heads.

It is known that when calculating any loudspeaker, they usually proceed from the parameters of the head used in it. Twinning the heads leads to a change in only one of these parameters ~ the total equivalent volume. So, when doubling heads with equivalent volumes V e 1 and V e 2, their total equivalent volume is V e = (V e 1 +V e 2)/4. The entire methodology for further calculation of loudspeakers with dual heads does not differ from the calculation of loudspeakers with single heads, both for a closed box and for a bass reflex.

To accurately determine the equivalent volume of the head, it is recommended to use a measuring box. If it was not possible to obtain a suitable measuring box, you can use the approximate formula to determine the equivalent volume of the head (in lit pax):

where C g is the flexibility of the oscillatory system of the head, cm/g, measured according to the method proposed in; D e - diameter of the diffuser without corrugation, cm. The found value of V e can be used when calculating the loudspeaker box, and after its manufacture, more accurate measurements can be taken. A few words about the efficiency of a loudspeaker with dual drivers. Its dependence on the parameters is described by the expression:

where c is the speed of sound, K is a dimensionless quantity, constant for a given type of head and acoustic design. V is the specified volume of the loudspeaker box.

The above formula shows that the price to pay for reducing the lower limit frequency of the range reproduced by the loudspeaker is to reduce it .

This, however, more than pays off in that when the heads are doubled, all types of distortion of the signal they reproduce are reduced. In addition to the reasons indicated in, another important circumstance contributes to this. The fact is that the unevenness of the sound field inside the loudspeaker box leads to severe unevenness in its frequency response. Uneven distribution of sound pressure inside the box can, in addition, cause deformation of the cone (especially light and thin) of the head, which, in turn, contributes to the occurrence of nonlinear and intermodulation distortions.

In the case of using double heads, all these unpleasant phenomena occur only on the inner head, but on the outer head, due to the damping effect of the air enclosed between the heads, they are significantly weakened.

To eliminate the source of these distortions, it is recommended to limit the frequency spectrum of vibrations supplied to the internal head, depending on the size of the loudspeaker, to 100...300 Hz. The harmful effect of the internal resonances of the box on the quality of playback can also be reduced by installing acoustic resistance panels (ARPs) between the heads or on the back side of the inner head. In both cases, it is recommended to place the PAS in the holes of the diffuser holders of the heads. It should also be borne in mind that the PAS reduces the quality factor of the head, and this can be very useful, since in some cases it will allow the use of a low-frequency amplifier without a current PSC.

It is known that the sound quality of a loudspeaker depends on the uniformity not only frequency response, but also FCHH, Smoothing of the phase response is achieved both in the electrical (by selecting appropriate separation filters) and in the acoustic paths (following the recommendations given in).

A certain alignment of the phases emitted by the heads of sound vibrations can be achieved, for example, by placing the voice coils of the heads in the same plane, perpendicular to the acoustic axis of the loudspeaker. However, this measure often turns out to be insufficient, especially when using heads with significantly different masses of moving systems and with diffusers made of materials of different densities. In the first case, this is explained by the fact that the phase shifts introduced by the heads at medium and higher frequencies, other things being equal, are greater. the greater the mass of the moving system, and in the second, the fact that phase shifts depend on the speed of propagation of sound waves along the surface of the diffuser.

These circumstances force the low-frequency head to be pushed forward in relation to the mid-frequency one, and the mid-frequency head in relation to the high-frequency one. The required additional displacement of the heads can be found experimentally by applying a rectangular voltage with a frequency of 0.7fp to the input of the amplifier with which the loudspeaker operates (here fp is the crossover frequency) and observing the transient process of the signal taken from a measuring microphone installed on the acoustic axis of the heads.

Taking into account the above considerations, dual heads operating in the low-frequency section should be installed according to the drawing. If you decide to use dual heads in the mid-frequency section, then they need to be placed with diffusers facing each other, as recommended in.

A practical example of the use of dual heads is the two-way loudspeaker developed by the author, made in the form of a bass reflex. Its low-frequency section uses dual 6GD-2 heads, and its mid-high-frequency section uses a ZGD-42 head (ZGD-32 is also possible). It works in conjunction with a two-way amplifier, the nominal output power of the low- and high-frequency channels of which is 20 and 10 W, respectively. The crossover filter (crossover frequency 500 Hz) is similar to that shown in. The output impedance of the low-frequency channel of the amplifier is negative - 1.5 Ohms. The nominal range of frequencies reproduced by the loudspeaker is 30... 18000 Hz, the frequency response unevenness is no more than 6 dB.

The loudspeaker housing (700x400x360 mm) is made of 20 mm thick chipboard. The front wall is glued together from two sheets of chipboard, its thickness is 40 mm. The same thickness is the cylindrical cover with a diameter of 300 mm made of the same material, fixed to the outside of the front panel. The hole in the cover with a diameter of 230 mm coincides with the hole in the front panel for low-frequency heads.

One of them is fixed on the inside of the front panel, the other on the outside of the cover. The ZGD-42 head is mounted on the outside of the front panel above the low-frequency unit with a large axis vertically. On the inside it is covered with a cap, the volume of which (about 2 liters) is filled with cotton wool. To increase the rigidity of the box, metal spacers are installed between the front and back, as well as between the side walls. The inner walls of the box are covered with felt 20 mm thick.

The phase inversion pipe (installed on the front panel) has an internal diameter of 80 and a length of 160 mm, including the thickness of the front wall.

The loudspeaker can also be made in the form of a closed box. In this case, a smooth frequency response at low frequencies is obtained with zero output impedance of the bandpass amplifier, and the lower limit of the frequency range reproduced by the loudspeaker increases to 40 Hz. If you install crossover filters with a crossover frequency of 400...500 Hz in such a loudspeaker, then it can be used with almost any amplifier operating into a 4 Ohm load.

The fidelity of the loudspeaker's music programs in both versions is very high.

LITERATURE

1. Zhurenkov A. Twin dynamic heads. Radio. 1979. No. 5. p. 48.
2. Vinogradova E. L. Design of loudspeakers with smoothed frequency characteristics. - M.. Energy, 1978.
3. Ephrussi M. Calculation of loudspeakers. - Radio 1977 No. 3. p. 36-37.
4. Valentin and Victor Leksin. Single-lane or multi-lane? Radio, 1981. No. 4, pp. 35-38. (RADIO 2, 1983)

An audiophile never rests. He wants to achieve perfect music sound in his home. There is a large selection of acoustic systems (AS) on sale. True, prices are painfully “biting”, but this was not always the case. About 40 years ago I had no choice: there was a Symphony stereo radio on sale (even without a stereo decoder, there was only a connector on the rear wall to connect it), at a price equal to the three-month salary of a novice engineer (330 “full-fledged” Soviet rubles). The speaker radios, in general, worked well, but the lowest frequencies sounded “not convincing” (even with a transistor amplifier) ​​because it’s a closed box, the resonant frequency of the speaker is 70 Hz. And I wanted to listen to the organ with its f n =16 Hz, and a Turkish drum tuned to 20 Hz, finally the same concert grand piano (from 27.5 Hz)! Take a look at Fig. 1, and it will become clear to you what and how (in what range) it sounds.

If something cannot be bought, then the audiophile builds “it” himself. Good speakers for low frequencies were not sold in stores at that time. They were “mined” and brought “from over the hill.” Once, while visiting a friend, I saw amazing speakers from the UK. They were called the “Jordan Watts module.” Small (152x152x50 mm), heavy (3.2 kg), with a diaphragm made of metal (anodized aluminum) with a diameter of 4 inches (10 cm). They had a diffuser stroke of 6.5 mm, a resonant frequency of 41 Hz, a power of 12 W and a frequency band from 30-17000 Hz (at a level of ±3 dB). They were accompanied by a specifications sheet, which showed a drawing of the recommended box (Fig. 2) and a table with seven speaker options. There were designs with two and one heads (modules) of different sizes (all bass reflexes). Depending on the size, different lower reproduced frequencies were obtained. One speaker had f n =20 Hz! True, the sound pressure level was not indicated.

Using an article from the magazine “Radio”, I calculated the phase inverters according to the sizes from the table and got a good (for those times) match with the company’s data. The owner of the speakers built the Juliet speaker to the dimensions shown in the table and was pleased with the results. In this speaker, the resonant frequency of the head almost coincided with the tuning frequency of the bass reflex. At that time, bass reflex calculations were made using empirical formulas and graphs. No theory was developed, but it was believed that the tuning of the bass reflex box should match the resonant frequency of the outdoor speaker. And here (for the Jupiter speaker) - 41 Hz at the head and 20 Hz at the speaker itself. Mystery? Fantastic?!

And, inspired, I decided to build the speaker myself. Which? And the best! With difficulty, I acquired four 25GD-26 heads (directly from Berdsk) for the low-frequency section (two heads per speaker) Based on materials from the Radio magazine and the book by M. Ephrussi, he designed and manufactured solid (90 liters - external volume) boxes made of thick chipboard 20 mm. The front panel drawing is shown in Fig. 3. I took into account all the recommendations for damping the walls: I glued a layer of fiberboard (6 mm) with vibration-absorbing mastic and covered it with vinyl leather. The bass reflexes were set to 25 Hz (the speakers had f r =36-42 Hz).

I started listening. The sound at low frequencies was not satisfactory! I filled the entire volume of the box with cotton wool (almost 3 kg!). The sound got better, but the bass still sounded bad (there was no “lightness” or “power”). By the way, the 35AC-1 loudspeakers that went on sale did not sound any better. According to estimates, the financial costs for my speakers corresponded to the cost of the 35AC-1.

Several years passed, and a book by E.L. Vinogradova appeared, which opened my eyes. The theory presented there clarified everything and put it “in its place.” A well-functioning loudspeaker with a bass reflex can only be built if certain relationships between the parameters of the speaker and the box are observed. Speaker parameters are always ( I emphasize, always!) needs to be measured. There is production variation. Materials age (the flexibility of the suspension changes, the magnet weakens), and then the quality factor increases. Descriptions of speakers that do not provide the parameters of the used heads measured by the authors seem frivolous to me, so I do not advise blindly copying such “masterpieces”.

Basic parameters when measuring speakers:

Some more relationships:

V as / V = ( f h 2 - f b 2 ) ∙ ( f b 2 - f l 2 ) / ( f h 2 ∙ f l 2 ) , Where

f l And f h - frequencies (lower and upper) of the “humps” on the Z-characteristic of the bass reflex;

f s ’ = ( f l ∙ f h ) / f b , Where

f s ’ - resonant frequency of the head, taking into account the added mass of air that occurs when the head operates in the bass reflex;

Where Q b - quality factor of the acoustic design, taking into account losses in the slots of the box and head, in the filling of the box and in the bass reflex pipe;

Rb - head resistance at the bass reflex tuning frequency;

R s - head resistance at resonant frequency.

Having the basic parameters of dynamic heads, you can use nomograms to find the bass reflex tuning frequency f r and its lower operating frequency at a level of -3dB - f 3 .

In accordance with the theory of operation of a speaker in a bass reflex, it is possible to obtain different full power frequency responses. They are called by the names of mathematicians who studied the corresponding curves and mathematical expressions - polynomials that describe these curves. These are the polynomials of Butterworth, Chebyshev, Cauer and others. At certain relationships between the parameters, different frequency responses of the loudspeaker are obtained.

If Qt =0,383, V as /V =1,41, f b / f s =1 and Q b >10, we have Butterworth frequency response (maximally smooth). Such speakers are most often built. In them f b =f s , i.e. the bass reflex tuning frequency is equal to the resonant frequency of the head. When Q b =10 nomograms correspond to Fig. 4, for Q b =5 - in Fig. 5.

What if Qt does the existing head differ from these “Butterworth” values? Then, as we see from the nomograms, the relationship will change V as /V , f b / f s And f 3 / f s . Of course, the frequency response will also change: with growth Qt it turns from the most smooth (Butterworth) into the wavy Chebyshev characteristic. And for her f 3 / f s <1, т.е. можно получить АЧХ с нижней воспроизводимой частотой, меньшей, чем резонансная частота динамика. Вот и разгадка английских акустических систем (41 Гц и 20 Гц).

In my speakers the speakers had Qt =0,54, V as / V =90/68=1.32. The heads stood side by side, and V as for two heads was twice as much as for one. And for Qt =0.54 (Fig. 4 and Fig. 5) needed V as /V =0.3, i.e. The volume of the box should be 3 times the equivalent volume of the heads. It turns out: V=V as / 0.3=90/0.3=300 l.

The situation was further aggravated by the active resistance of the crossover filter choke for the low-frequency head in the speakers, which increased Qt by 10%.

But the audiophile never calms down! The thought came unexpectedly. Eureka! I remembered about the dual heads. After all, their equivalent volume is halved compared to one head, the quality factor is preserved, and fs =√( fs 1 ∙ fs 2 ) .

No sooner said than done! I connect two heads, diffuser to diffuser, through a 3 mm spacer and insert this “sandwich” in place of one woofer from the inside of the front panel. It’s okay that the magnet of one head sticks out, and the hole for the second speaker was hastily sealed with a chipboard patch (bottom hole in Fig. 3). The main idea! Measured Qt =0.5 - even less than single speakers. One thing is clear, because the radiation comes from the back of the diffusers, and the effective area has decreased due to the windows in the diffuser holder, and the radiation resistance has increased. The voice coil inductance has been linearized and harmonics have been reduced.

By the way, distortion due to the Doppler effect is not noticeable in the low-frequency section if its upper frequency is not higher than 500-800 Hz, and the lower one is 25-30 Hz. And here the radiation along the axis of the heads is shielded by magnets, and, as Drazen wrote: “These distortions are not audible from the side of the loudspeaker neutral.” Those interested can calculate these distortions using the formula:

K D = (18 ∙ 10 3 ∙ f V √ P a ) / ( f n 2 ∙ d 2 ) [%], P a = P uh ∙ efficiency , Where:

P a - acoustic power, W;

P uh - electrical power, W;

d - diffuser diameter, cm;

f n, f in - corresponding limiting frequencies of the range, Hz.

I have, at P e =10 W, Efficiency =0,1%, f n =28 Hz, f in =800 Hz it turned out:

K D = (18 ∙ 10 3 ∙ 800 √0,01) / (28 2 ∙ 20 2) = 4,5%.

I'm filming Z - characteristics of the speakers, I calculate V as /V =0.29. Now everything matches! I measure fs outdoor: head 1: 34.5 Hz; head 2: 42.8 Hz; dual block (compound): 38 Hz.

According to calculations:

f sk = √(fs 1 ∙ fs 2 ) = √(34.5 ∙ 42.8) = 38.4 [Hz];

V ask / Vk = (41,4 2 – 24 2) ∙ (24 2 – 20 2) / (41,4 2 ∙ 20 2) = 0,29;

f sk = (f 1 ∙ f h) / f b = (41.4 ∙ 20) / 24 = 34.5 [Hz].

I remind you fsk< fsk due to the added mass of air in the box with the bass reflex. I find Qak =5.148 and Q ek =0.883, and also Q bk ≈3.5. Qbk is not enough, but after removing the excess cotton and leaving about 700 g, I am getting closer to Q bk ≈5. Now the nomogram in Fig. 5 is suitable, f bk/f sk ≈0,75; f 3 k / fsk ≈0.7, where I find it from f bk =25.9 Hz; f 3 k =24.2 Hz.

I check the frequency response by ear using a test CD. I hear a frequency of 25 Hz with a slight block, 31.5 Hz is excellent. The sound of musical programs with a Turkish drum is simply pleasing (“Funeral Mass” by Verdi, part - “Lacrimosa”, performed by the Strasbourg Philharmonic Orchestra). And when the drums of the choir and orchestra come on forte fortissimo, everything jumps, including me in surprise. This is what the great Verdi sought! I have only heard such a low-frequency sound from Tappou speakers with a volume of 200 liters and a head diameter of about 380 mm.

Is “Chebyshev” muttering? I didn't notice this. But I noticed losses due to diffraction when I started measuring the frequency response of loudspeakers installed at a distance of 1.5 m from the walls. Let's calculate the frequency at which this rolloff occurs (-3 dB) using the formula:

f = 115 / W = 115 / 0.375 ≈ 300 [Hz], where W - AC width, m.

This value exactly coincided with the measured one. If the speakers are placed in the corners of a room measuring 6x3x2.7 m along a narrow wall, then there is no drop in LF due to diffraction.

Of course, we must take into account that the frequency response of a loudspeaker in an ordinary living room will have many peaks and dips due to sound reflections from walls, ceilings, floors and other surfaces. This is shown in Fig. 6 (curve 1 – frequency response in the room, curve 2 in the sound-measuring chamber).

Let me summarize:

1. If you want to get the lowest frequencies, and the resonant frequency of the speakers is about 1.5 times higher than them, then a “Chebyshev” speaker will help you.
2. To avoid building huge boxes, you can use double heads.
3. Properly configured speakers “according to Chebyshev” do not “mumble”!
4. An audiophile never rests (axiom!).

I still have one question: the peaks on the Z-characteristic are slightly different in height, and it has not yet been possible to align them. Why?