DIY 2-way speaker system. Making speaker systems with your own hands. For the frisky and the computer

Making sound speakers with your own hands - this is where many people begin their passion for a complex, but very interesting matter - sound reproduction technology. The initial motivation is often economic considerations: prices for branded electroacoustics are not excessively inflated, but outrageously brazen. If sworn audiophiles, who do not skimp on rare radio tubes for amplifiers and flat silver wire for winding sound transformers, complain on forums that the prices for acoustics and speakers are systematically inflated, then the problem is really serious. Would you like speakers for your home for 1 million rubles? pair? If you please, there are more expensive ones. That's why The materials in this article are designed primarily for very beginners: they need to quickly, simply and inexpensively make sure that the creation of their own hands, all of which cost tens of times less money than a “cool” brand, can “sing” no worse or at least comparable. But probably, some of the above will be a revelation for the masters of amateur electroacoustics- if it is honored with reading by them.

Column or speaker?

A sound column (KZ, sound column) is one of the types of acoustic design of electrodynamic loudspeaker heads (SG, speakers), intended for technical and informational sounding of large public premises. In general, an acoustic system (AS) consists of a primary sound emitter (S) and its acoustic design, which provides the required sound quality. Home speakers for the most part look like speakers, which is why they are called that. Electroacoustic systems (EAS) also include electrical part: wires, terminals, isolation filters, built-in audio frequency power amplifiers (UMPA, in active speakers), computing devices (in speakers with digital channel filtering), etc. Acoustic design Household speakers are usually placed in a cabinet, which is why they look like columns more or less elongated upward.

Acoustics and electronics

The acoustics of an ideal speaker are excited over the entire range of audible frequencies of 20-20,000 Hz by one broadband primary source. Electroacoustics is slowly but surely moving towards the ideal, but the best results are still shown by speakers with frequency division into channels (bands) LF (20-300 Hz, low frequencies, bass), MF (300-5000 Hz, mid) and HF (5000 -20,000 Hz, high, high) or low-midrange and high-frequency. The first, naturally, are called 3-way, and the second - 2-way. It’s best to start getting comfortable with electro-acoustics with 2-way speakers: they allow you to get sound quality up to high Hi-Fi (see below) at home without unnecessary costs and difficulties (see below). Sound signal from UMZCH or, in active speakers, low-power from the primary source (player, sound card computer, tuner, etc.) is distributed among frequency channels separation filters; this is called channel defiltering, just like the crossover filters themselves.

The rest of the article focuses primarily on how to make columns that provide good acoustics. The electronic part of electroacoustics is the subject of a special serious discussion, and more than one. Here you only need to note that, firstly, at first you do not need to take on close to ideal, but complex and expensive digital filtering, but use passive filtering using inductive-capacitive filters. For a 2-way speaker, you only need one plug of low- and high-pass filters (LPF/HPF).

There are special programs for calculating AC staircase separating filters, for example. JBL Speaker Shop. However, at home, individual tuning of each plug for a specific speaker, firstly, does not affect production costs in mass production. Secondly, replacing the GG in the AC is required only in exceptional cases. This means that you can approach filtering the frequency channels of speakers in an unconventional way:

1. The frequency of the LF-MF and HF section is taken to be no lower than 6 kHz, otherwise you will not get a sufficiently uniform amplitude-frequency response (AFC) of the entire speaker in the midrange region, which is very bad, see below. In addition, with a high crossover frequency, the filter is inexpensive and compact;
2. The prototypes for calculating the filter are links and half-links of type K filters, because their phase-frequency characteristics (PFC) are absolutely linear. Without this condition, the frequency response in the crossover frequency region will be significantly uneven and overtones will appear in the sound;
3. To obtain the initial data for calculation, you need to measure the impedance (total electrical resistance) LF-MF and HF GG at the crossover frequency. The 4 or 8 Ohms indicated in the GG passport are their active resistance on DC, and the impedance at the crossover frequency will be greater. The impedance is measured quite simply: the GG is connected to an audio frequency generator (AFG), tuned to the crossover frequency, with an output of no weaker than 10 V into a load of 600 Ohms through a resistor of obviously high resistance, for example. 1 kOhm. You can use low-power GZCH and high-fidelity UMZCH. Impedance is determined by the ratio of audio frequency (AF) voltages across the resistor and GG;
4. The impedance of the low-frequency-mid-frequency link (GG, head) is taken as the characteristic resistance ρн of the low-pass filter (LPF), and the impedance of the HF head is taken as ρв of the high-pass filter (HPF). The fact that they are different is a joke; the output impedance of the UMZCH, which “swings” the speaker, is negligible compared to both;
5. On the UMZCH side, low-pass filter and reflective-type high-pass filter units are installed so as not to overload the amplifier and not take power away from the associated speaker channel. On the contrary, the absorbing links are turned to the GG so that the return from the filter does not produce overtones. Thus, the low-pass filter and high-pass filter of the speaker will have at least a link with a half-link;
6. The attenuation of the low-pass filter and high-pass filter at the crossover frequency is taken equal to 3 dB (1.41 times), because The slope of the K-filters is small and uniform. Not 6 dB, as it might seem, because... filters are calculated based on voltage, and the power supplied to the GG depends on the square of it;
7. Adjusting the filter comes down to “muting” a channel that is too loud. The channel volumes are measured at the crossover frequency using a computer microphone, turning off the HF and LF-MF in turn. The degree of “jamming” is determined as the square root of the channel volume ratio;
8. Excessive volume of the channel is removed with a pair of resistors: a damping one of fractions or units of Ohm is connected in series with the GG, and in parallel with both of them - a leveling one of greater resistance, so that the impedance of the GG with the resistors remains unchanged.

Explanations for the method

A technically knowledgeable reader may have a question: does your filter work for a complex load? Yes, and in this case, it’s okay. The phase response of K-filters is linear, as stated, and the Hi-Fi UMZCH is practically ideal source voltage: its output resistance Rout is units and tens of mOhms. Under such conditions, the “reflection” from the GG reactance will be partially attenuated in the output absorbing section/half-section of the filter, but for the most part will leak back to the UMZCH exit, where it will disappear without a trace. In fact, nothing will pass into the conjugate channel, because... ρ of its filter is many times greater than Rout. There is one danger here: if the impedances of the GG and ρ are different, then power circulation will begin in the filter output – GG circuit, causing the bass to become dull, “flat”, the attacks on the midrange to be drawn out, and the highs to become sharp and whistling. Therefore, the impedance of the GG and ρ must be adjusted precisely, and if the GG is replaced, the channel will have to be adjusted again.

Note: Do not try to filter active speakers with analog active filters on operational amplifiers (op amps). It is impossible to achieve linearity of their phase characteristics in a wide frequency range, which is why, for example, analog active filters have never really taken root in telecommunications technology.

What is hi-fi

Hi-Fi, as you know, is short for High Fidelity - high fidelity (sound reproduction). The concept of Hi-Fi was initially accepted as vague and not subject to standardization, but an informal division into classes gradually developed; The numbers in the list indicate, respectively, the range of reproduced frequencies (operating range), the maximum permissible nonlinear distortion factor (THD) at rated power (see below), the minimum permissible dynamic range relative to the room’s own noise (dynamics, ratio of maximum to minimum volume), the maximum permissible unevenness of the frequency response in the midrange and its rolloff (decline) at the edges of the operating range:

• Absolute or full - 20-20,000 Hz, 0.03% (-70 dB), 90 dB (31,600 times), 1 dB (1.12 times), 2 dB (1.25 times).
• High or Heavy - 31.5-18,000 Hz, 0.1% (-60 dB), 75 dB (5600 times), 2 dB, 3 dB (1.41 times).
• Medium or basic – 40-16,000 Hz, 0.3% (–50 dB), 66 dB (2000 times), 3 dB, 6 dB (2 times).
• Initial – 63-12500 Hz, 1% (–40 dB), 60 dB (1000 times), 6 dB, 12 dB (4 times).

It is curious that high, basic and initial Hi-Fi roughly correspond to the highest, first and second classes of household electroacoustics according to the USSR system. The concept of absolute Hi-Fi arose with the advent of condenser, film-panel (isodynamic and electrostatic), jet and plasma sound emitters. The Anglo-Saxons called high-end Hi-Fi “Heavy” because High High Fidelity in English is like butter.

What kind of hi-fi do you need?

Home acoustics for modern apartment or a house with good sound insulation should satisfy the conditions for basic Hi-Fi. A high one there, of course, won’t sound worse, but it will cost a lot more. In a block Khrushchev or Brezhnevka, no matter how you isolate them, only professional experts distinguish between initial and basic Hi-Fi. The reasons for such roughening of the requirements for home acoustics are as follows.

Firstly, the full range of sound frequencies is heard by literally a few people in all of humanity. People gifted with a particularly fine ear for music, such as Mozart, Tchaikovsky, J. Gershwin, hear high Hi-Fi. Experienced professional musicians in a concert hall confidently perceive basic Hi-Fi, but 98% of ordinary listeners in a sound-measuring chamber almost never distinguish between initial and basic Hi-Fi.

Secondly, in the most audible region of the midrange, a person dynamically distinguishes sounds in the range of 140 dB, counting from an audibility threshold of 0 dB, equal to the intensity of the sound flux of 1 pW per square meter. m, see fig. on the right are curves of equal loudness. A sound louder than 140 dB is already pain, and then damage to the hearing organs and contusion. An expanded symphony orchestra at a powerful fortissimo produces sound dynamics of up to 90 dB, and in the halls of the Bolshoi Opera, Milan, Paris, Vienna Opera Houses and the Metropolitan Opera in New York it can “accelerate” to 110 dB; so is the dynamic range of leading jazz bands with symphonic accompaniment. This is the limit of perception, louder than which the sound turns into still tolerable, but already meaningless noise.

Note: rock bands can play louder than 140 dB, which was what Elton John, Freddie Mercury and the Rolling Stones were fond of in their youth. But the dynamics of rock do not exceed 85 dB, because... Rock musicians cannot play the most delicate pianissimo even if they want to - the equipment does not allow it, and there is no rock “in the spirit”. As for pop music of any kind and movie soundtracks, this is not a topic at all - their dynamic range is already compressed during recording to 66, 60 and even 44 dB, so that you can listen to anything.

Thirdly, natural noises in the quietest living room country house on the outskirts of civilization – 20-26 dB. Sanitary standard noise in the library reading room is 32 dB, and the rustling of leaves in the fresh wind is 40-45 dB. It is clear from this that the 75dB high hi-fi speakers are more than enough for meaningful listening in living conditions; The dynamics of modern mid-level UMZCHs, as a rule, are no worse than 80 dB. In a city apartment, it is almost impossible to distinguish between basic and high Hi-Fi by dynamics.

Note: in a room noisy by more than 26 dB, the frequency range of the selected Hi-Fi can be narrowed to the limit. class, because the masking effect affects the background of indistinct noises, the frequency sensitivity of the ear decreases.

But in order for Hi-Fi to be high-fi, and not “happiness” for “beloved” neighbors and harmful to the owner’s health, it is necessary to ensure the least possible sound distortion, correct reproduction of low frequencies, smooth frequency response in the midrange, and determine what is necessary for sounding a given room electrical power AC. As a rule, there are no problems with HF, because their SOI “go” into the inaudible ultrasonic region; You just need to put a good HF head into the speaker. Here it is enough to note that if you prefer classics and jazz, it is better to take the HF GG with a diffuser with a power of 0.2-0.3 of that of the LF channel, for example. 3GDV-1-8 (2GD-36 in the old way) and the like. If you are “rushed” by hard tops, then the optimal option would be a high-frequency generator with a dome emitter (see below) with a power of 0.3-0.5 of the power of the low-frequency unit; Drumming with brushes is naturally reproduced only by dome tweeters. However, a good dome HF GG is suitable for any music.

Distortions

Sound distortion is possible linear (LI) and nonlinear (NI). Linear distortion is simply a discrepancy between the average volume level and the listening conditions, which is why any UMZCH has a volume control. Expensive 3-way speakers for high Hi-Fi (for example, Soviet AC-30, also known as S-90) often include power attenuators for midrange and high frequency in order to more accurately match the frequency response of the speaker to the acoustics of the room.

As for NI, as they say, they are countless and new ones are constantly being discovered. The presence of NI in the sound path is expressed in the fact that the shape of the output signal (which is sound already in the air) is not completely identical to the shape of the original signal from the primary source. Most of all, the purity, “transparency” and “richness” of the sound are spoiled. NI:

1. Harmonic – overtones (harmonics) that are multiples of the fundamental frequency of the reproduced sound. They manifest themselves as excessively rumbling bass, sharp and harsh midrange and treble;
2. Intermodulation (combination) - sums and differences in the frequencies of the components of the spectrum of the original signal. Strong combinational NIs are heard as wheezing, while weak ones that spoil the sound can only be recognized in the laboratory using multi-signal or statistical methods on test phonograms. To the ear, the sound seems clear, but somehow not so;
3. Transient – ​​“jitter” of the output signal shape during sharp increases/declines of the original signal. They manifest themselves with short wheezing and sobbing, but irregularly, with fluctuations in volume;
4. Resonant (overtones) - ringing, rattling, muttering;
5. Frontal (distortion of sound attack) – delaying or, conversely, forcing sudden changes in overall volume. Almost always occur together with transitional ones;
6. Noise - hum, rustle, hiss;
7. Irregular (sporadic) – clicks, crackles;
8. Interference (AI or IFI, so as not to be confused with intermodulation). Characteristic specifically for AS, IFIs do not occur in UMZCH. Very harmful, because are perfectly audible and cannot be eliminated without major alteration of the speakers. See below for more information about FFIs.

Note:“wheezing” and other figurative descriptions of distortion here and below are given from the point of view of Hi-Fi, i.e. as already heard by experienced listeners. And, for example, speech speakers are designed on SOI at a rated power of 6% (in China - by 10%) and 1

In addition to interference, AS can produce predominantly NI according to claims. 1, 3, 4 and 5; Clicks and crackles are possible here as a result of poor quality manufacturing. They struggle with transitional and frontal NI in speakers by selecting suitable GGs (see below) and acoustic design for them. Ways to avoid overtones are the rational design of the speaker cabinet and the correct choice of material for it, also see below.

You need to linger on harmonic NIs in the speakers, because they are fundamentally different from those in semiconductor UMZCH and are similar to the harmonic NI of tube ULF (low frequency amplifiers, the old name of UMZCH). A transistor is a quantum device, and its transfer characteristics are not fundamentally expressed by analytical functions. The consequence is that it is impossible to accurately calculate all the harmonics of a transistor UMZCH, and their spectrum extends to the 15th and higher components. Also in the spectrum of transistor UMZCHs there is a large proportion of combinational components.

The only way to cope with all this disgrace is to hide the NI deeper under the amplifier’s own noise, which, in turn, should be many times lower than the natural noise of the room. It must be said that modern circuitry copes with this task quite successfully: according to current concepts, a UMZCH with 1% THD and –66 dB of noise is “no”, and with 0.06% THD and –80 dB of noise it is quite mediocre.

With harmonic NI speakers, the situation is different. Their spectrum, firstly, like that of tube ULFs, is pure - only overtones without a noticeable admixture of combination frequencies. Secondly, the harmonics of the speakers can be traced, just like those of lamps, no higher than the 4th. Such a spectrum of NI does not noticeably spoil the sound even at a SOI of 0.5-1%, which is confirmed by expert estimates, and the reason for the “dirty” and “sluggish” sound of homemade speakers most often lies in the poor frequency response in the midrange. For your information, if a trumpet player has not properly cleaned the instrument before a concert and during playing does not splash out saliva from the embouchure in a timely manner, then the THD of, say, a trombone, can increase to 2-3%. And that’s okay, they play and the audience likes it.

The conclusion from here is very important and favorable: the range of reproduced frequencies and the intrinsic harmonics of a NI speaker are not parameters that are critical for the quality of the sound it creates. Experts can classify the sound of speakers with 1% or even 1.5% harmonic NI as basic, or even high Hi-Fi, if the appropriate conditions are met. conditions for the dynamics and smoothness of the frequency response.

Interference

IFI is the result of the convergence of sound waves from nearby sources in phase or in antiphase. The result is surges, even to the point of pain in the ears, or dips of almost zero volume at certain frequencies. At one time, the first-born of the Soviet Hi-Fi 10MAS-1 (not 1M!) was urgently discontinued after musicians discovered that this speaker did not reproduce the A of the second octave at all (as far as I remember). At the factory, the prototype was “driven” in a sound meter using a three-signal method, antediluvian even then, and the position of an expert with an ear for music was not on the staffing table. One of the paradoxes of developed socialism.

The probability of IFI occurrence increases sharply with increasing frequency and, accordingly, decreasing sound wavelength, because To do this, the distance between the centers of the emitters must be a multiple of half the wavelength of the reproduced frequency. At midrange and high frequency, the latter varies from a few decimeters to millimeters, so there is no way to install two or several midrange and high frequency generators in the speakers - then IFI cannot be avoided, because the distances between the centers of the GG will be of the same order. In general, the golden rule of electroacoustics is one emitter per band, and the brilliant rule is one broadband GG for the entire frequency range.

The LF wavelength is meters, which is much greater not only than the distance between the GGs, but also the size of the speakers. Therefore, manufacturers and experienced amateurs often increase the power of the speakers and improve the bass by pairing or quadruplet (putting in a quadruplet) the LF GG. However, a beginner should not do this: internal interference of reflected waves “walking” with the speaker itself may occur. To the ear, it manifests itself as resonant NI: it booms, hums, rattles, it is not clear why. So follow the precious rules so as not to go through the entire speaker over and over again to no avail.

Note: You cannot place an odd number of identical GGs in the AS under any circumstances - the IFIs are then 100% guaranteed

midrange

Novice amateurs pay little attention to the reproduction of mid frequencies - they say, any speaker will “sing” them - but in vain. The midrange is heard best; it also contains the original (“correct”) harmonics of the basis of everything – the bass. The unevenness of the frequency response of speakers in the midrange can give very strong combinational NIs that spoil the sound, because the spectrum of any phonogram “floats” across the frequency range. Especially if the speakers use efficient and inexpensive speakers with a short diffuser stroke, see below. Subjectively, when listening, experts clearly prefer speakers with a frequency response in the midrange, smoothly varying across the frequency range within 10 dB over one that has 3 dips or “bumps” of 6 dB each. Therefore, when designing and making speakers, you need to carefully check at every step: will the frequency response on the midrange “bump” from this?

Note, speaking of bass: rocker joke. So, a young promising group broke through to the prestigious festival. Half an hour later they had to go out, and they were already backstage, worried, waiting, but the bassist was on a spree somewhere. 10 minutes before the exit - he’s not there, 5 minutes - he’s not there either. They wave at the exit, but still no bass player. What to do? Well, we'll play without bass. Failure to do so means instant career ruin forever. They played without bass, it’s clear how. They wander towards the service exit, spitting and swearing. Lo and behold, there’s a bass player, a tough guy, with two chicks. They come to him - oh, you goat, do you even understand how you cheated us?!! Where have you been?! - Yes, I decided to listen in the hall. – And what did you hear there? - Dudes, without bass it sucks!

LF

Bass in music is like the foundation for a house. And in the same way, the “zero cycle” of electroacoustics is the most difficult, complex and responsible. The audibility of a sound depends on the energy flow of the sound wave, which depends on the square of the frequency. Therefore, the bass is heard the worst, see fig. with curves of equal volume. To “pump” energy into the low frequencies, powerful speakers and UMZCH are needed; In reality, more than half of the amplifier's power is spent on bass. But at high powers, the probability of the occurrence of NI increases, the strongest and, of course, audible components of the spectrum of which from the bass will fall precisely on the best audible midrange.

“Pumping” NPs is further complicated by the fact that the dimensions of the GG and the entire AS are small compared to the wavelengths of the NPs. Any sound source transfers energy to it the better, the larger its size relative to the sound wavelength. The acoustic efficiency of low-frequency speakers is units and fractions of a percent. Therefore, most of the work and hassle in creating a speaker system comes down to making it reproduce bass frequencies better. But let us remind you once again: do not forget to monitor the purity of the midrange as often as possible! Actually, the creation of a low-frequency speaker path comes down to:

• Determination of the required electrical power of the LF GG.
• Selecting a low-frequency GG suitable for the given listening conditions.
• Selecting the optimal acoustic design (casing design) for the selected low-frequency GG.
• Its correct manufacture in a suitable material.

Power

The sound output in dB (characteristic sensitivity) is indicated in the speaker passport. It is measured in a sound-measuring chamber 1 m from the center of the GG with a measuring microphone located strictly along its axis. The GG is placed on a sound-measuring shield (standard acoustic screen, see figure on the right) and electrical power of 1 W is supplied (0.1 W for GG with a power of less than 3 W) at a frequency of 1000 Hz (200 Hz, 5000 Hz). Theoretically, based on these data, the class of the desired Hi-Fi and the parameters of the room/listening area (local acoustics), it is possible to calculate the required electrical power of the generator. But in fact, taking into account local acoustics is so complex and ambiguous that even experts rarely bother with it.

Note: The GG for measurements is shifted from the center of the screen in order to avoid interference of sound waves from the front and rear emitting surfaces. The screen material is usually a cake of 5 layers of unsanded 3-layer pine plywood with casein glue 3 mm thick and 4 spacers between them made of natural felt 2 mm thick. Everything is glued together with casein or PVA.

It is much easier to proceed from the existing conditions to the technical sound of low-noisy rooms, with adjustments for the dynamics and frequency range of Hi-Fi, especially since the results obtained in this case are in better agreement with known empirical data and expert estimates. Then for initial Hi-Fi you need, with a ceiling height of up to 3.5 m, 0.25 W of the nominal (long-term) electrical power of the GG per 1 sq. m of floor area, for basic Hi-Fi – 0.4 W/sq. m, and for high – 1.15 W/sq. m.

The next step is to take into account actual listening conditions. Hundred-watt speakers capable of operating at microwatt levels are monstrously expensive, on the one hand. On the other hand, if a separate room is not allocated for listening, equipped as a sound-measuring chamber, then their “micro-whispers” at the quietest pianissimo will not be heard in any living room (see above about natural noise levels). Therefore, we increase the obtained values ​​by two or three times in order to “tear off” what we are listening to from the background noise. We get for initial Hi-Fi from 0.5 W/sq. m, basic from 0.8 W/sq. m and for high from 2.25 W/sq. m.

Next, since we need hi-fi, and not just speech intelligibility, we need to move from nominal power to peak (musical) power. The “juice” of a sound depends primarily on the dynamics of its volume. THD GG at loudness peaks should not exceed its value for Hi-Fi in a class below the chosen one; for initial Hi-Fi we take 3% THD at the peak. In trade specifications for Hi-Fi speakers, it is the peak power that is indicated as more significant. According to the Soviet-Russian method, peak power is equal to 3.33 long-term; according to the methods of Western companies, “music” is equal to 5-8 denominations, but - stop for now!

Note: Chinese, Taiwanese, Indian and Korean methods are ignored. For basic (!) Hi-Fi, at their peak they accept a telephone SOI of 6%. But the Philippines, Indonesia and Australia measure their speakers correctly.

The fact is that all Western manufacturers of Hi-Fi GG, without exception, shamelessly overestimate the peak power of their products. It would be better if they promoted their SOI and frequency response flatness, they really have something to be proud of. But the average foreigner will not understand such complexities, but if “180W”, “250W”, “320W” is written on the speaker, that’s really cool. In reality, running the speakers “from there” in a sound meter gives their peaks at 3.2-3.7 nominal values. Which is understandable, because... This ratio is justified physiologically, i.e. the structure of our ears. Conclusion - when targeting Western GGs, go to the company website, look for the rated power there and multiply by 3.33.

Note 9, regarding the peak and nominal designations: in Russia, according to the old system, the numbers in front of the letters in the designation of the speaker indicated its rated power, but now they give the peak. But at the same time the root and suffix of the designation were also changed. Therefore, the same speaker can be designated in completely different ways; see examples below. Look for the truth from reference sources or on Yandex. No matter what designation you enter, the results will contain the new one, and the old one in parentheses next to it.

In the end, we get for a room up to 12 square meters. m peak for initial Hi-Fi at 15 W, base at 30 W and high at 55 W. These are the smallest acceptable values; taking the GG two or three times more powerful will be better, unless you listen to symphonic classics and very serious jazz. For them, it is advisable to limit the power to 1.2-1.5 times the minimum, otherwise wheezing is possible at peak volumes.

You can do it even simpler by focusing on proven prototypes. For initial Hi-Fi in a room up to 20 sq. m is suitable GG 10GD-36K (10GDSh-1 in the old way), for a tall one - 100GDSh-47-16. They don’t need filtering, these are broadband GGs. With basic Hi-Fi it is more difficult; a suitable broadband speaker cannot be found for it; you need to make a 2-way speaker. Here, at first, the optimal solution is to repeat the electrical part of the old Soviet S-30B speaker. These speakers have been “singing” regularly and very well for decades in apartments, cafes and just on the street. They are extremely shabby, but they keep the sound.

The S-30B filtering diagram (without overload indication) is shown in Fig. left. Minor modifications have been made to reduce losses in the coils and allow adjustment to various low-frequency generators; if desired, taps from L1 can be made more often, within 1/3 of the total number of turns w, counting from the right end of L1 according to the diagram, the fit will be more accurate. On the right are instructions and formulas for independently calculating and manufacturing filter coils. Precision parts are not required for this filtering; deviations in coil inductance by +/–10% also do not noticeably affect the sound. It is advisable to place the R2 engine on the rear wall to quickly adjust the frequency response to the room. The circuit is not very sensitive to the impedance of the speakers (unlike filtering using K-filters), so instead of the ones indicated, you can use other GGs that are suitable in power and resistance. One condition: the highest reproducible frequency (HRF) of the LF GG at the level of –20 dB must be no lower than 7 kHz, and the lowest reproducible frequency (LRF) of the HF GG at the same level - no higher than 3 kHz. By moving and moving L1 and L2, you can slightly correct the frequency response in the crossover frequency region (5 kHz), without resorting to such complexities as a Zobel filter, which can also increase transient distortion. Capacitors – film with insulation made of PET or fluoroplastic and sprayed plates (MKP) K78 or K73-16; as a last resort - K73-11. Resistors are metal film (MOX). Wires – audio from oxygen-free copper with a cross-section of 2.5 square meters. mm. Installation - soldering only. In Fig. on the right is shown what the original filtering of the S-30B looks like (with an overload indication circuit), and in Fig. Below on the left is a 2-way filtering scheme popular abroad without magnetic coupling between the coils (which is why their polarity is not indicated). On the right there, just in case, is a 3-way filtering of the Soviet S-90 speaker (35AC-212).

About wires

Special audio cables are not a product of mass psychosis and not a marketing gimmick. The effect, discovered by radio amateurs, has now been confirmed by research and recognized by experts: if there is an admixture of oxygen in the copper of the wire, a thin, literally molecule-sized film of oxide is formed on the crystallites of the metal, from which the sound signal can do anything but improve. This effect is not found in silver, which is why sophisticated audio connoisseurs do not skimp on silver wire: traders shamelessly cheat with copper wires, because... It is possible to distinguish oxygen-free copper from ordinary electrical copper only in a specially equipped laboratory.

Speakers

The quality of the primary sound emitter (S) in the bass determines the sound of the speakers approx. by 2/3; in the midrange and highs – almost completely. In amateur speakers, the IZs are almost always electrodynamic GGs (speakers). Isodynamic systems are quite widely used in high-end headphones (for example, TDS-7 and TDS-15, which are readily used by professionals to control sound recordings), but the creation of powerful isodynamic systems encounters technical difficulties that are still insurmountable. As for other primary IZs (see the list at the beginning), they are still far from being “brought to fruition.” This is especially true for prices, reliability, durability and stability of characteristics during operation.

When getting into electroacoustics, you need to know the following about how speakers are structured and work in acoustic systems. The speaker exciter is a thin coil of wire that vibrates in the annular gap of the magnetic system under the influence of audio frequency current. The coil is rigidly connected to the actual sound emitter into space - a diffuser (at LF, MF, sometimes at HF) or a thin, very light and rigid dome diaphragm (at HF, rarely at MF). The efficiency of sound emission strongly depends on the diameter of the IZ; more precisely, from its ratio to the wavelength of the emitted frequency, but at the same time, with an increase in the diameter of the IZ, the probability of the occurrence of nonlinear distortions (ND) of sound due to the elasticity of the IZ material also increases; more precisely, not its infinite rigidity. They combat NI in IR by making radiating surfaces from sound-absorbing (anti-acoustic) materials.

The diameter of the diffuser is larger than the diameter of the coil, and in diffuser GGs it and the coil are attached to the speaker body with separate flexible suspensions. The diffuser configuration is a hollow cone with thin walls, with its apex facing the coil. The coil suspension simultaneously holds the top of the diffuser, i.e. its suspension is double. The generatrix of the cone can be rectilinear, parabolic, exponential and hyperbolic. The steeper the diffuser cone converges to the top, the higher the output and the lower the dynamics of the speaker, but at the same time its frequency range narrows and the directivity of the radiation increases (the radiation pattern narrows). Narrowing the pattern also narrows the stereo effect zone and moves it away from the frontal plane of the speaker pair. The diameter of the diaphragm is equal to the diameter of the coil and there is no separate suspension for it. This sharply reduces the TNI of the GG, because The diffuser suspension is a very noticeable source of sound, and the material for the diaphragm can be very hard. However, the diaphragm is capable of producing sound well only at fairly high frequencies.

The coil and diffuser or diaphragm together with suspensions make up the moving system (MS) of the GG. The PS has a frequency of its own mechanical resonance Fр, at which the mobility of the PS sharply increases, and a quality factor Q. If Q>1, then a speaker without correctly selected and executed acoustic design (see below) at Fр will wheeze at a power less than the rated one, not to mention peak, this is the so-called. locking the GG. Blocking does not apply to distortion, because is a design and manufacturing defect. If 0.7

The efficiency of transferring electrical signal energy to sound waves in the air is determined by the instantaneous acceleration of the diffuser/diaphragm (who is familiar with mathematical analysis - the second derivative of its displacement with respect to time), because air is an easily compressible and very fluid medium. The instantaneous acceleration of the coil pushing/pulling the diffuser/diaphragm must be somewhat greater, otherwise it will not “swing” the IZ. A few, but not by much. Otherwise, the coil will bend and cause the emitter to vibrate, which will lead to the appearance of NI. This is the so-called membrane effect, in which longitudinal elastic waves propagate in the diffuser/diaphragm material. Simply put, the diffuser/diaphragm should “slow down” the coil a little. And here again there is a contradiction - the more the emitter “slows down”, the more powerfully it emits. In practice, the “braking” of the emitter is done in such a way that its NI in the entire range of frequencies and powers falls within the norm for a given Hi-Fi class.

Note, output: Don't try to "squeeze" out of the speakers what they can't do. For example, a speaker on a 10GDSH-1 can be built with an uneven frequency response in the midrange of 2 dB, but in terms of SOI and dynamics it still reaches Hi-Fi no higher than the initial one.

At frequencies up to Fp, the membrane effect never appears; this is the so-called. piston mode of operation of the GG - the diffuser/diaphragm simply moves back and forth. Higher in frequency, the heavy diffuser can no longer keep up with the coil, membrane radiation begins and intensifies. At a certain frequency, the speaker begins to radiate only like a flexible membrane: at the junction with the suspension, its diffuser is already motionless. At 0.7

The membrane effect dramatically improves the efficiency of the GG, because the instantaneous accelerations of vibrating sections of the IZ surface turn out to be very large. This circumstance is widely used by designers of high-frequency and partly mid-range generators, the distortion spectrum of which immediately goes into ultrasound, as well as when designing generators not for Hi-Fi. SOI GG with a membrane effect and the evenness of the frequency response of speakers with them strongly depend on the mode of the membrane. At zero mode, when the entire surface of the IZ trembles as if to its own rhythm, Hi-Fi up to medium inclusive can be achieved at low frequencies, see below.

Note: the frequency at which the GG switches from the “piston to the membrane”, as well as the change in the membrane mode (not growth, it is always an integer) significantly depend on the diameter of the diffuser. The larger it is, the lower in frequency and the stronger the speaker begins to “membrane”.

Woofers

High-quality piston LF GGs (simply “pistons”; in English woofers, barking) are made with a relatively small, thick, heavy and rigid anti-acoustic diffuser on a very soft latex suspension, see position 1 in Fig. Then Fр turns out to be below 40 Hz or even below 30-20 Hz, and Q<0,7. В мембранном режиме поршневые ГГ способны работать до частот 7-8 кГц на нулевой-первой модах.

The periods of LF waves are long, all this time the diffuser in piston mode must move with acceleration, therefore the diffuser stroke is long. Low frequencies without acoustic design are not reproduced, but it is always closed to one degree or another, isolated from free space. Therefore, the diffuser has to work with a large mass of so-called. attached air, the “swing” of which requires significant force (which is why piston GGs are sometimes called compression), as well as for the accelerated movement of a heavy diffuser with a low quality factor. For these reasons, the magnetic system of the piston GG has to be made very powerful.

Despite all the tricks, the recoil of piston engines is small, because It is impossible for a low-frequency diffuser to develop high acceleration at long waves: the elasticity of the air is not enough to absorb the energy given off. It will spread to the sides, and the speaker will go into locking. To increase the efficiency and smoothness of the moving system (to reduce the SOI at high power levels), designers go to great lengths - they use differential magnetic systems, with half-scattering and other exotic ones. SOI is further reduced by filling the magnetic gap with a non-drying rheological fluid. As a result, the best modern “pistons” achieve a dynamic range of 92-95 dB, and the THD at nominal power does not exceed 0.25%, and at peak power – 1%. All this is very good, but the prices - mom, don't worry! $1000 per pair with differential magnets and rheofill for home acoustics selected for impact, resonant frequency and flexibility of the moving system is not the limit.

Note: LF GG with rheological filling of the magnetic gap are suitable only for LF links of 3-way speakers, because completely unable to operate in membrane mode.

Piston GGs have one more serious flaw: without strong acoustic damping, they can be mechanically destroyed. Again, simply: behind the piston speaker there must be some kind of air cushion loosely connected to the free space. Otherwise, the diffuser at the peak will be torn off the suspension and it will fly out along with the coil. Therefore, “pistons” cannot be installed in every acoustic design, see below. In addition, piston GGs do not tolerate forced braking of the PS: the coil burns out immediately. But this is already a rare case; speaker cones are usually not held by hand and matches are not inserted into the magnetic gap.

Note to craftsmen

There is a well-known “folk” way to increase the efficiency of piston engines: an additional ring magnet is firmly attached with the repelling side to the standard magnetic system from the rear, without changing anything in the dynamics. It is repelling, otherwise, when a signal is given, the coil will immediately be torn off from the diffuser. In principle, it is possible to rewind the speaker, but it is very difficult. And never before has a single speaker gotten better from rewinding, or at least remained the same.

But that’s not really what we’re talking about. Enthusiasts of this modification claim that the field of the external magnet concentrates the field of the standard one near the coil, which causes the acceleration of the PS and recoil to increase. This is true, but Hi-Fi GG is a very precisely balanced system. The returns actually increase a little. But at its peak, SOI immediately “jumps” so that sound distortions become clearly audible even to inexperienced listeners. At nominal, the sound may become even cleaner, but without Hi-Fi speakers it’s already high-fi.

Presenters

So in English (managers) they are called SCH GG, because. It is the midrange that accounts for the overwhelming majority of the semantic load of the musical opus. The requirements for the midrange of the GG for Hi-Fi are much softer, so most of them are made of a traditional design with a large diffuser cast from cellulose pulp along with the suspension, pos. 2. Reviews about midrange GG dome and with metal diffusers are contradictory. The tone prevails, they say, the sound is harsh. Classical lovers complain that bowed speakers squeal from “non-paper” speakers. Almost everyone recognizes the sound of the midrange GG with plastic diffusers as dull and at the same time harsh.

The stroke of the MF GG diffuser is made short, because its diameter is comparable to the wavelengths of the midrange and the transfer of energy into the air is not difficult. To increase the attenuation of elastic waves in the diffuser and, accordingly, reduce the NI together with the expansion of the dynamic range, finely chopped silk fibers are added to the mass for casting the Hi-Fi midrange GG diffuser, then the speaker operates in piston mode in almost the entire midrange range. As a result of applying these measures, the dynamics of modern midrange GGs of the average price level turns out to be no worse than 70 dB, and the THD at the nominal value is no higher than 1.5%, which is quite enough for high Hi-Fi in a city apartment.

Note: Silk is added to the cone material of almost all good speakers; it is a universal way to reduce the SOI.

Tweets

In our opinion - tweeters. As you may have guessed, these are tweeters, HF GG. Spelled with one t, this is not the name of a social network for gossip. Making a good “tweeter” from modern materials would be generally simple (the LR spectrum immediately goes into ultrasound), if not for one circumstance - the diameter of the emitter in almost the entire HF range turns out to be of the same order of magnitude or less than the wavelength. Because of this, interference is possible at the emitter itself due to the propagation of elastic waves in it. In order not to give them a “hook” for radiation into the air at random, the diffuser/dome of the HF GG should be as smooth as possible; for this purpose, the domes are made of metallized plastic (it absorbs elastic waves better), and the metal domes are polished.

The criterion for choosing high-frequency GGs is indicated above: dome ones are universal, and for fans of the classics who definitely require “singing” soft tops, diffuser ones are more suitable. It is better to take these elliptical ones and place them in the speakers, orienting their long axis vertically. Then the speaker pattern in the horizontal plane will be wider, and the stereo area will be larger. There is also an HF GG with a built-in horn on sale. Their power can be taken at 0.15-0.2 of the power of the low-frequency section. As for the technical quality indicators, any HF GG is suitable for Hi-Fi of any level, as long as it is suitable in terms of power.

Shiriki

This is a colloquial nickname for broadband GG (GGSH), which does not require filtering of speaker frequency channels. A simple GGSH emitter with general excitation consists of a LF-MF diffuser and a HF cone rigidly connected to it, pos. 3. This is the so-called. coaxial emitter, which is why GGSH are also called coaxial speakers or simply coaxials.

The idea of ​​the GGSH is to give the membrane mode to the HF cone, where it will not do much harm, and let the diffuser at the LF and at the bottom of the midrange work “on a piston”, for which purpose the LF-MF diffuser is corrugated across. This is how broadband GGs are made for initial, sometimes mid-range Hi-Fi, for example. the mentioned 10GD-36K (10GDSH-1).

The first HF cone GGSH went on sale in the early 50s, but never achieved a dominant position in the market. The reason is a tendency to transient distortion and a delay in the attack of sound because the cone dangles and wobbles from the shocks of the diffuser. Listening to Miguel Ramos play a Hammond electric organ through a coaxial cone is unbearably painful.

Coaxial GGSH with separate excitation of LF-MF and HF emitters, pos. 4 do not have this drawback. In them, the HF section is driven by a separate coil from its own magnetic system. The HF coil sleeve passes through the LF-MF coil. The PS and magnetic systems are located coaxially, i.e. along one axis.

GGSH with separate excitation at LF are not inferior to piston GG in all technical parameters and subjective assessments of sound. Modern coaxial speakers can be used to build very compact speakers. The disadvantage is the price. A coaxial for high-end Hi-Fi is usually more expensive than a LF-MF + HF set, although it is cheaper than a LF, MF and HF GG for a 3-way speaker.

Auto

Car speakers are formally also classified as coaxial, but in reality they are 2-3 separate speakers in one housing. HF (sometimes also midrange) GG are suspended in front of the LF GG diffuser on a bracket, see on the right in Fig. at first. Filtering is always built-in, i.e. There are only 2 terminals on the body for connecting wires.

Car speakers have a specific task: first of all, to “shout out” the noise in the car’s interior, so their designers don’t particularly struggle with the membrane effect. But for the same reason, car speakers need a wide dynamic range, at least 70 dB, and their diffusers are necessarily made with silk or other measures are used to suppress higher membrane modes - the speaker should not wheeze even in a car while driving.

As a result, car speakers are, in principle, suitable for Hi-Fi up to medium, inclusive, if you choose a suitable acoustic design for them. In all the speakers described below, you can install auto speakers of a suitable size and power, then there will be no need for a cutout for the HF GG and filtering. One condition: the standard terminals with clamps must be very carefully removed and replaced with lamellas for unsoldering. Modern car speaker speakers allow you to listen to good jazz, rock, even individual works of symphonic music and many chamber music. Of course, they won’t be able to handle Mozart’s violin quartets, but very few people listen to such dynamic and meaningful opuses. A pair of car speakers will cost several times, up to 5 times, less than 2 sets of GG with filter components for a 2-way speaker.

Frisky

Friskers, from frisky, is how American radio amateurs nicknamed small-sized low-power GGs with a very thin and light diffuser, firstly, for their high output - a pair of “frisky” 2-3 W each sound a room of 20 square meters. m. Secondly – ​​for the hard sound: “fast” ones work only in membrane mode.

Manufacturers and sellers do not classify “frisky” people as a special class, because they are not supposed to be hi-fi. The speaker is like a speaker, like any Chinese radio or cheap computer speakers. However, for the “frisky” ones, you can make good speakers for your computer, providing Hi-Fi up to and including average in the vicinity of your desktop.

The fact is that the “fast” ones are capable of reproducing the entire audio range; you just need to reduce their SOI and smooth out the frequency response. The first is achieved by adding silk to the diffuser; here you need to be guided by the manufacturer and its (not trade!) specifications. For example, all GG of the Canadian company Edifier with silk. By the way, Edifier is a French word and is read “ediffier”, and not “idifier” in the English manner.

The frequency response of “fast” ones is equalized in two ways. Small splashes/dips are already removed by silk, and larger bumps and depressions are eliminated by acoustic design with free access to the atmosphere and a damping pre-chamber, see fig; For an example of such an AS, see below.

Acoustics

Why do you need acoustic design at all? At low frequencies, the dimensions of the sound emitter are very small compared to the length of the sound wave. If you simply place the speaker on the table, the waves from the front and rear surfaces of the diffuser will immediately converge in antiphase, cancel each other out, and no bass will be heard at all. This is called an acoustic short circuit. You cannot simply mute the speaker from the rear to the bass: the diffuser will have to strongly compress a small volume of air, which will cause the resonance frequency of the PS to “jump” so high that the speaker simply will not be able to reproduce bass. This implies the main task of any acoustic design: either to extinguish the radiation from the back side of the GG, or to turn it 180 degrees and re-radiate it in phase from the front of the speaker, while at the same time preventing the energy of the diffuser movement from being spent on thermodynamics, i.e. on the compression-expansion of air in the speaker housing. An additional task is, if possible, to form a spherical sound wave at the output of the speaker, because in this case, the stereo effect zone is widest and deepest, and the influence of room acoustics on the sound of the speakers is the least.

Note, important consequence: For each speaker enclosure of a specific volume with a specific acoustic design, there is an optimal range of excitation powers. If the power of the IZ is low, it will not pump up the acoustics; the sound will be dull and distorted, especially at low frequencies. An excessively powerful GG will go into thermodynamics, causing blocking to begin.

The purpose of the speaker cabinet with acoustic design is to ensure the best reproduction of low frequencies. Strength, stability, appearance – of course. Acoustically, home speakers are designed in the form of a shield (speakers built into furniture and building structures), an open box, an open box with an acoustic impedance panel (PAS), a closed box of normal or reduced volume (small-sized speaker systems, MAS), a bass reflex (FI), passive radiator (PI), direct and reverse horns, quarter-wave (QW) and half-wave (HF) labyrinths.

Built-in acoustics are a subject of special discussion. Open boxes from the era of tube radios; it is impossible to get acceptable stereo from them in an apartment. Among others, it is best for a beginner to choose the PV labyrinth for his first AS:

• Unlike others, except FI and PI, the PV labyrinth allows you to improve the bass at frequencies below the natural resonant frequency of the woofer speaker.
• Compared to FI PV, the labyrinth is structurally and simple to set up.
• Compared to PI PV, the labyrinth does not require expensive purchased additional components.
• The elbowed PV labyrinth (see below) creates a sufficient acoustic load for the GG, while at the same time having a free connection with the atmosphere, which makes it possible to use LF GG with both long and short diffuser strokes. Up to replacement in already built speakers. Of course, only a couple. The emitted wave in this case will be practically spherical.
• Unlike all but a closed box and a HF labyrinth, an acoustic speaker with a MF labyrinth is capable of smoothing out the frequency response of the LF GG.
• Speakers with a PV labyrinth are structurally easily stretched into a tall, thin column, which makes them easier to place in small rooms.

Regarding the penultimate point - are you surprised if you are experienced? Consider this one of the promised revelations. And see below.

PV labyrinth

Acoustic design such as a deep slot (Deep Slot, a type of HF labyrinth), pos. 1 in Fig., and a convolutional inverse horn (item 2). We will touch on the horns later, but as for the deep slot, it is actually a PAS, an acoustic shutter that provides free communication with the atmosphere, but does not release sound: the depth of the slot is a quarter of the wavelength of its tuning frequency. This can be easily verified by using a highly directional microphone to measure the sound levels in front of the speaker and in the opening of the slit. Resonance at multiple frequencies is suppressed by lining the slot with a sound absorber. A speaker with a deep slot also dampens any speaker, but increases its resonant frequency, although less than a closed box.

The initial element of the PV labyrinth is an open half-wave tube, pos. 3. It is unsuitable as an acoustic design: while the wave from the rear reaches the front, its phase will flip another 180 degrees, and the same acoustic short circuit will result. In the frequency response of the PV pipe, it gives a high sharp peak, causing blocking of the GG at the tuning frequency Fn. But what is already important is that Fn and the frequency of the GG’s own resonance f (which is higher – Fр) are theoretically in no way related to each other, i.e. You can count on improved bass below f (Fр).

The simplest way to turn a pipe into a labyrinth is to bend it in half, pos. 4. This will not only phase the front with the rear, but also smooth out the resonant peak, because The wave paths in the pipe will now be of different lengths. In this way, in principle, you can smooth out the frequency response to any predetermined degree of evenness, increasing the number of bends (it should be odd), but in reality it is very rare to use more than 3 bends - wave attenuation in the pipe interferes.

In the chamber PV labyrinth (position 5), the knees are divided into the so-called. Helmholtz resonators - tapering towards the rear end of the cavity. This also improves the damping of the GG, smoothes the frequency response, reduces losses in the labyrinth and increases the radiation efficiency, because the rear exit window (port) of the labyrinth always works with “support” from the side of the last chamber. Having separated the chambers into intermediate resonators, pos. 6, it is possible with a diffuser GG to achieve an frequency response that almost satisfies the requirements of absolute Hi-Fi, but setting up each of a pair of such speakers requires about six months (!) of the work of an experienced specialist. Once upon a time, in a certain narrow circle, a labyrinth-chamber speaker with a separation of chambers was nicknamed Cremona, with a hint of the unique violins of Italian masters.

In fact, to obtain the frequency response for high Hi-Fi, just a couple of cameras per knee is enough. Drawings of speakers of this design are shown in Fig; on the left - Russian design, on the right - Spanish. Both are very good floor-standing acoustics. “For complete happiness,” it would not hurt the Russian woman to borrow the Spanish rigidity connections that support the partition (beech sticks with a diameter of 10 mm), and in return, smooth out the bend of the pipe.

In both of these speakers, another useful property of the chamber labyrinth is manifested: its acoustic length is greater than the geometric one, because the sound lingers somewhat in each chamber before passing on. Geometrically, these labyrinths are tuned to somewhere around 85 Hz, but measurements show 63 Hz. In reality, the lower limit of the frequency range turns out to be 37-45 Hz, depending on the type of low-frequency generator. If the filtered speakers from the S-30B are moved into such enclosures, the sound changes amazingly. For the better.

The excitation power range for these speakers is 20-80 W peak. Sound-absorbing lining here and there - padding polyester 5-10 mm. Tuning is not always necessary and is not difficult: if the bass is a bit muffled, cover the port symmetrically on both sides with pieces of foam until optimal sound is obtained. This should be done slowly, listening to the same section of the soundtrack each time for 10-15 minutes. It must have strong midranges with a steep attack (control of the midrange!), for example, a violin.

Jet Flow

The chamber labyrinth is successfully combined with the usual convoluted labyrinth. An example is the desktop acoustic system Jet Flow (jet flow) developed by American radio amateurs, which created a real sensation in the 70s, see fig. on right. The inside width of the case is 150-250 mm for speakers 120-220 mm, incl. “fast” and autodynamics. Body material – pine, spruce, MDF. No sound-absorbing lining or adjustment is required. The excitation power range is 5-30 W peak.

Note: There is now confusion with Jet Flow - inkjet sound emitters are sold under the same brand.

For the frisky and the computer

It is possible to smooth out the frequency response of car speakers and “fast” ones in an ordinary convoluted labyrinth by installing a compression damping (non-resonating!) pre-chamber in front of the entrance to it, designated K in Fig. below.

This mini-acoustic system is designed for PCs to replace the old cheap ones. The speakers used are the same, but the way they start to sound is simply amazing. If the diffuser is made of silk, otherwise there is no point in fencing the garden. An additional advantage is the cylindrical body, on which the midrange interference is close to minimal; it is less only on the spherical body. Working position – tilted forward and upward (AC – sound spotlight). Excitation power – 0.6-3 W nominal. Assembly is carried out as follows. order (glue - PVA):

• For children 9 glue the dust filter (you can use scraps of nylon tights);
• Det. 8 and 9 are covered with padding polyester (indicated in yellow in the figure);
• Assemble the package of partitions using screeds and spacers;
• Glue in padding polyester rings, marked in green;
• The package is wrapped, gluing, with whatman paper until the wall thickness is 8 mm;
• The body is cut to size and the antechamber is pasted over (highlighted in red);
• They glue the children. 3;
• After complete drying, they sand, paint, attach a stand, and mount the speaker. The wires to it run along the bends of the labyrinth.

About horns

Horn speakers have high output (remember why they have a horn in the first place). The old 10GDSH-1 screams through its horn so loudly that your ears wither, and the neighbors “can’t be happier,” which is why many people get carried away with horns. In home speakers, curled horns are used as they are less bulky. The reverse horn is excited by the back radiation of the GG and is similar to the PV labyrinth in that it rotates the phase of the wave by 180 degrees. But otherwise:

1. Structurally and technologically it is much more complicated, see fig. below.
2. It does not improve, but on the contrary, it spoils the frequency response of the speakers, because The frequency response of any horn is uneven and the horn is not a resonating system, i.e. It is impossible in principle to correct its frequency response.
3. The radiation from the horn port is significantly directional, and its waveform is more flat than spherical, so one cannot expect a good stereo effect.
4. It does not create a significant acoustic load on the GG and at the same time requires significant power for excitation (let’s also remember whether they whisper into a speaking speaker). The dynamic range of horn speakers can be extended, at best, to basic Hi-Fi, and in piston speakers with a very soft suspension (that is, good and expensive ones), the diffuser breaks out very often when the GG is installed in the horn.
5. Gives more overtones than any other type of acoustic design.

Frame

The housing for the speakers is best assembled using beech dowels and PVA glue; its film retains its damping properties for many years. To assemble, one of the side panels is placed on the floor, the bottom, lid, front and back walls, partitions are placed, see fig. on the right, and cover with the other side. If the external surfaces are subject to final finishing, you can use steel fasteners, but always with gluing and sealing (plasticine, silicone) of non-adhesive seams.

The choice of housing material is much more important for sound quality. The ideal option is a musical spruce without knots (they are a source of overtones), but finding large boards of it for speakers is unrealistic, since spruce trees are very knotty trees. As for the plastic speaker enclosures, they only sound good if they are manufactured in one piece, while amateur home-made ones made from transparent polycarbonate, etc. are a means of self-expression, not acoustics. They will tell you that this sounds good - ask to turn it on, listen and believe your ears.

In general, natural wood materials for speakers are difficult: completely straight-grained pine without defects is expensive, and other available building and furniture species produce overtones. It is best to use MDF. The above-mentioned Edifier has long since completely switched to it. The suitability of any other tree for AS can be determined by following. way:

1. The test is carried out in a quiet room, in which you yourself need to first stay in silence for half an hour;
2. A piece of board approx. long. 0.5 m is placed on prisms made from sections of steel angles, laid at a distance of 40-45 cm from each other;
3. The knuckle of a bent finger is used to knock approx. 10 cm from any of the prisms;
4. Repeat tapping exactly in the center of the board.

If in both cases the slightest ringing is not heard, the material is suitable. The softer, duller and shorter the sound, the better. Based on the results of such a test, you can make good speakers even from chipboard or laminate, see the video below.

Having on hand a number of old and very ordinary Soviet speakers, I wanted to assemble speakers on them (MAS - small-sized speakers) and see what could be gotten from this idea.
I think that the design is suitable for repetition, the cost of components is low. If I used rare imported speakers, the possibility of exact repetition would be minimal. The frequency range of the finished speaker is 40…20000 Hz.

Selecting the speaker type

There are 6GD-2 speakers and some others, but they are in short supply and you definitely can’t make small-sized speakers with them; 6GD-2 require boxes the size of a small refrigerator.

To get a more or less low frequency on the shield, it must be very large, this is also not a MAC. For MAS, only FI remains because, in my opinion, there are no Soviet speakers capable of providing bass in a small ZY.

All the woofers I know have a rubber surround (they had foam rubber, but at the moment they require mandatory replacement of the surrounds, and the volume of the box will be too big for MAC) and they sound bad in the midrange, and these frequencies are very important for me because I think that they determine the overall sound quality.

All Soviet MACs are made using a two-way design, have a high crossover frequency, approximately 5 kHz, and therefore sound inexpressive due to the fact that the “rubber” low-frequency speaker does not sound well above 1...2 kHz.

Most industrial speakers, even those assembled using a three-way design, also sound muddy in the midrange because they usually have a “rubber” 15GD-11 installed. This circumstance is the reason for the huge number of articles on the conversion of 25AC, 35AC and their clones. Therefore, a three-way version with a paper midrange speaker was chosen for the project.
In addition, I believe that the best design for a midrange speaker is a shield or a shallow OJ. The shield is preferable in terms of sound, but is inconvenient in design.

Therefore, the choice of column type is as follows: three-way, woofer in a small box with FI, midrange speaker in a small shallow OJ, tweeter on a tiny shield.

The midrange speakers also need to be improved. Soak the corrugation (the entire diffuser cannot be soaked) with a liquid solution of guerlain. The process is described in detail by Shorov and his followers.
But this was not enough. When listening to the assembled and configured speakers, the sound was somewhat harsh. The reason turned out to be the high acoustic quality factor of the speakers. I had to use PAS, which gave a noticeable positive result.

Attempts to modify the 2GD-36 using the method of improving the 3GD-31 did not give a positive result.

AC crossover filter

I was unable to smoothly mix the midrange and high-frequency speakers at a frequency of about 5 kHz; I was forced to succeed only at a frequency of 8...9 kHz, which has its drawbacks. LF-MF section at 500 Hz.

Unlike parallel filters, where the filter for each speaker can be adjusted separately and independently, there is an unwanted interdependency that makes adjustment difficult.
The result is a filter like this.

According to the author of this solution, the circuit has the smoothest joints since everything that is cut off by the low-pass and high-pass filters goes to the midrange speaker. The idea is interesting and everything would be so if there were resistors instead of speakers. But, as I have written many times, speakers, like electromechanical devices, have their own temper and the result may not live up to expectations.

Speaker design

The front panel of the speaker is removable.

The midrange speaker is in the OJ, it is advisable to make it minimally deep. The PAS is cut out of a piece of laminate, the large hole is for the magnetic speaker system, the small ones with a diameter of 10 mm are for the PAS. The number of holes is not critical; about 20 can be made.

The tweeter is mounted on a small rectangular piece of laminate.

Since three independent housings were made, it became possible to move the speakers inward. The midrange is shifted by about 4 cm, the treble is also shifted by 4 cm relative to the midrange. And if the shift of the midrange speaker relative to the bass has little effect on the shape of the measuring pulse, then the shift of the tweeter is quite noticeable, the shift value was chosen optimal for listening to speakers from a distance of 2...2.5 meters.

I would like to note that if all the speakers are traditionally placed in one housing, this shift will be very difficult. Imagine an 8 cm deepening...

One can argue about the usefulness of such a solution, but it was tried on in some speakers for a very long time and I think that it was abandoned due to its low technology.
However, this solution is still used today, but not in wooden, but in cast cases - plastic or metal. To do this, the recess is made flat; some believe that this is a horn, but this is not so - this is a way to move the speakers inward.

I think music lovers will hear the difference; I did not conduct special tests on music, but configured them using computer programs and a microphone.

The height of the speaker is 73 cm, if you listen while sitting in a chair, a stand for the speaker is not needed. The direction of the midrange and tweeter axes was chosen deliberately.

If anyone wants to repeat my “15MAS-1”, you can buy a pair of 15AC-109 in good condition (clones are not suitable) for LF boxes.
You will need to remove the tweeter and plug the hole; you can use a standard FI pipe.

Results

I think it’s better than any Soviet small-sized speaker, which is not surprising because they are two-way.
As for industrial three-lanes, the situation here is ambiguous. It is clear that in terms of power (volume level), my speakers cannot compete with the 25AC and 35AC series of clones.
They cannot compete in terms of the amount of bass, despite the fact that my speakers operate from approximately 34 Hz ​​at a level of -3 dB, a speaker of such a small diameter is noticeably inferior in power to the woofers of the 25GD, 30GD series.

But 2GD-36 sounds softer than 10GD-35. And the main advantage is in the midrange. Rubber 15GD-11 sound clearly worse, although louder. I think the overall balance is in favor of these MACs, but not for lovers of killer bass.

For home use, the volume of these speakers is quite sufficient, despite the low sensitivity of the 25GDN3-4. But for a disco - no. But quality is not needed for a disco.

The purpose of my experiments was to try out some techniques and tuning methods, see what can be done from cheap and affordable speakers, and if the result is positive, offer Datagor readers a new article.

I think that the design is suitable for repetition, the cost of components is low. If I used rare imported speakers, the possibility of exact repetition would be minimal.
The frequency range of the speakers is at least 40...20000 Hz, in fact, according to the standard method, even wider.

The speakers work well.
I don’t give measurement graphs; anyone who wishes can measure it themselves, for example, using the methods I described on Datagor.

Thank you for your attention!

Reader vote

The article was approved by 73 readers.

Greetings to Datagor readers! I want to tell you about creating an acoustic system using 3D printing technology. Using a 3D printer, I was able to build an unusual acoustic system in the shape of a ball, as well as solve a number of additional problems that arise when making acoustics.
I would like to note that I am not at all advocating the use of plastic as the main material for building speakers.

Ever since my student days I had a dream - to make speakers in the shape of balls. But the methods available to me at that time for creating a custom-shaped case did not inspire me in any way. And now, many years later, I got a 3D printer.

Now here is my translation of Troels Gravesen’s article about “the world’s worst dome tweeter Philips AD 0160.” I think he didn’t come across Soviet, or even many modern tweeters.
Most likely, few people have this particular tweeter (tweeter, not to be confused with Twitter), but Troels’ research will be useful for home-made people to evaluate the quality and proper use of tweeters.

Best regards, Sergei

I'll show you with a real example what you can do with old speakers, namely their drivers, to get improved sound.

Unfortunately, not every one of us can afford to have a high-quality speaker system in the house. Now even the cheapest option will cost at least 10 thousand rubles. However, why not buy low-quality speakers that make a squeaking sound? If you are so eager to have your own in your home, you can make it yourself.

Moreover, all suitable parts and elements can be purchased almost anywhere, and their cost will certainly not be 10 thousand rubles. How to do it yourself? You will learn about this from our article today.

Preparing tools

So, during the work we will need the following materials and tools:

• screwdriver;
• sheet of chipboard or MDF (for making the speaker system housing);
• marker;
• jigsaw;
• PC power supply 400 W;
• radio;
• Bulgarian;
• a pair of acoustic speakers;
• furniture screws and self-tapping screws;
• sealant (it is best to use silicone based);
• voltmeter and glue.

Before assembling, first check the operation of the radio and find out whether it can be powered or not. It is also necessary to test the speakers for sound quality. After this, you can safely begin manufacturing the housing and other elements of the speaker system.

Case manufacturing

As walls for the column, you can use a regular sheet of MDF or chipboard. In this case, it is unacceptable to use plywood, since according to its characteristics it is very flexible and produces a strong resonance. When making a housing for the speaker system, also take into account the fact that the more air remains inside the box, the softer the bass will be. Therefore, leave as much free space as possible, but everything should be in moderation (otherwise such speakers will simply be impossible to transport).

Place the shelf so that the maximum amount of air reaches the speakers. Next, mark the places for cutting with a marker. Now you can safely cut a sheet of chipboard with a jigsaw. We also note that the edges of the cut parts of the tree should be carefully aligned. To do this, use a small construction grinder. Please note that it can work with several discs - for metal and wood. We need the latter option, since when processing such materials, the cutting element of the first type simply wears out and even smokes. Experts recommend using a petal circle.

Now the matter remains small. On the body, mark places for screwing in furniture screws and use a screwdriver to screw them in completely. In the case of screws, first make markings for them and drill through holes. That's it, the housing for the speaker system has been successfully manufactured.

Wall fasteners

Emphasis should be placed on the strength of the wall fasteners. Don't skimp on screws and self-tapping screws. The design of the speaker housing should be as strong and durable as possible. If the number of screws is insufficient, the system walls will rattle heavily under heavy load, thereby only worsening the sound quality.

Thorough assembly

How do you make a three-lane with your own hands? After you have made the case (the so-called “monoblock”), you can begin to thoroughly assemble the structure. Here it is best to use a cordless screwdriver with a 4mm hexagon to tighten the screws. Remember that inside the speaker its own sound load is distributed from minimum to maximum - bottom wall, top, front and side.

How to do it yourself next? At the next stage, the joints should be treated with silicone sealant. This is necessary in order to prevent excess from penetrating out of the case through the cracks. Thus, the level of music playback will become even better. How then do you make an acoustic system with your own hands? After lubricating all the cracks with sealant, you will need to install the speakers and radio. The latter is best purchased assembled. The speakers, together with the radio, are installed through holes made in the bottom wall of the monoblock.

When everything is ready, the final design should look like this: on the back of the monoblock there is a power supply, two speakers on the sides (with each of them located in a separate column) and a car radio in the middle. Making speaker systems with your own hands occurs in a certain order of actions - first the power supply is mounted, and then the radio tape recorder. This will make it much more convenient for you to screw fasteners. But at this stage, the PC has not yet been completely assembled. Next, you will need to reinforce the corners. We will tell you about this in the next section.

How to make a speaker system with your own hands? Strengthening corners

The whole point of the work is to paste over certain parts of the monoblock and then install square or triangular glazing beads on them. It is not necessary to use Moment as an adhesive. Regular PVA will do the job quite well. Before applying glue to the surface of the material, make sure that it is dry and that its surface is free of cracks and bends.

What needs to be done in order for the power supply to work?

To do this, you will need to put a jumper on the wide, large connector (in other words, short it). Here it is enough to use a regular paper clip. Use it to connect two wires (green to black) and check the functionality of the device with a voltmeter.

To ensure that these elements have greater conductivity, after installing the contact, thoroughly solder their connection points. Now insert the block body inside the monoblock and attach it with self-tapping screws. Also treat any resulting cracks with sealant.

About sound-permeable material

At the next stage, the acoustic system, made by yourself, is filled with a special sound-permeable material (here you can use ordinary padding polyester). They need to fill the entire volume of the columns.

However, you cannot apply it to the diaphragm. This sound-permeable material significantly reduces the load on the walls of the system and reduces the speed of sound waves. So, when playing a melody, the speaker design will practically not vibrate. However, you should not follow the principle “the more, the better.” If you overfill the speaker with synthetic padding, it may lose bass and, accordingly, the sound quality will deteriorate significantly.

Fan

If your TV or computer speaker system is designed for high playback power, consider additional cooling elements.

Indeed, under high load, the elements of the speakers become very hot, which can cause their premature failure. And you need to install the fan in such a way that it blows from the inside to the outside, that is, the hot air is brought out into the street (or room). If heat is removed from the radio, overheating of system parts will be prevented, and your speakers will last for a very long time. At this stage, the question of how to make a three-way speaker system with your own hands can be considered closed.

Do-it-yourself bookshelf acoustics

DIY or Do It Yourself

One day I decided to build myself high-quality acoustics for sounding a small room, as well as for use as near-field monitors when working with sound on a computer (hobby). The main requirement is adequate sound in relation to the source. It’s not so much that “the lows are wobbling” or that “the cymbals are ringing,” but rather an adequate natural sound. So, we collect high-quality “shelf holders”.

Number of lanes

In theory, the ideal system is single-band. But, like everything ideal, such a system does not exist in nature. Yes, there are very high-quality broadband speakers from the same "Visaton", but for some reason all well-known manufacturers make two-way bookshelf systems. And when it comes to the floor option, then 3 stripes are not uncommon. There wasn’t much of a question here – the classic two-band version: low-frequency and high-frequency.

Speaker selection

The main requirement for speakers is an optimal price/quality ratio. Those. it shouldn’t be “cheap” for 500 rubles, but also not a mind-blowing “high-end” for $1000. Besides, I was in no hurry. The idea of ​​assembling “shelf speakers” with my own hands came quite a long time ago, and I threw the bait in advance to my good friend, “sick” with sound, with whom we have been constantly and fruitfully communicating on this topic for a long time.

The first to appear was the HF - Vifa XT19SD-00/04 ring-rad. These are high-quality 4-ohm tweeters, quite popular among audiophiles. They were planned for one set, but for some reason they didn’t work out and ended up in my set.

LF arrived second. They turned out to be very decent midbass from the Soundstream Exact 5.3 kit. Here you can read a little about them. It so happened that the tweeters burned out during installation, and the lone woofers themselves turned out to be unnecessary. 4-ohm 5.5" midbass mounted in a cast aluminum basket were immediately purchased.

Now that you have the speakers, you can start creating acoustics.

Active / passive?

Each option has its pros and cons. Firstly, you need to take into account the compactness of the speakers themselves and the associated difficulties in layout in limited space. There is no point in installing it outside. Secondly, individual modules as independent components can be combined in the future, and it is also easier to repair if something happens. And thirdly, active speakers are quite expensive. Because if you make a decent amplifier (and sometimes there is one in each case), then it will turn out to be more expensive than the acoustics themselves. Besides, I already had an amplifier. But in any case, I am in favor of the scheme - passive acoustics + amplifier, it is more universal.

Calculation of housing dimensions

We have decided on the speakers, now we need to understand which housing is optimal for them. Dimensions are calculated based on the sound characteristics of the woofer. There are no recommendations on the manufacturer's website, because... The speaker was intended primarily for car audio. There is no point in keeping special equipment for these purposes unless it is your job. Therefore, a smart guy with a special stand comes to the rescue. As a result of laboratory tests, we obtain a calculated case size of 310 x 210 x 270 mm. During the measurement process, the bass reflex parameters were also calculated.

By the way, many manufacturers publish recommended housing sizes for speakers on their websites. When such information is available, it is logical to use it, but in this case I did not have such data, so I had to do laboratory research.

Housing material

In my opinion, the most optimal material for the case is MDF. It is acoustically neutral and also performs slightly better than chipboard. Plywood is also good, but quality plywood is not easy to find and is more expensive and difficult to process. 22mm MDF sheet was chosen as the source material for the body. In principle, standard 18-20mm is quite enough, but I decided to make a little extra. There is no such thing as too much rigidity.

Housing construction and design

One of the most important stages. Before going for MDF, I advise you to decide on the design so that you can immediately ask the seller to cut the sheet in parts, and at a normal point of sale there are always good machines with accurate and even cuts. It is difficult to obtain such a cut at home.

So, design. The speakers should look at least as good as “industrial” ones, so that there is no feeling of a club of crazy hands. We make not only high-quality, but also beautiful acoustics. In general, there are practically no beautiful, interesting and at the same time structurally simple acoustic systems. Beautiful acoustics are made by the Italian Sonus Faber, stunning in beauty - Magico Mini. But they are all made using precision machines, which, by definition, are not available at home. As an option, you can order the cases from a good “cabinet maker” with hands and CNC. Depending on where and what you order, such work will cost from 10,000 rubles. up to 30,000 rub. along with materials. If the specialist is good, then the speakers will look no worse, or even better, than store-bought ones. In this case, I decided that I would do everything completely myself. Therefore, we look at things realistically and make a design without any bevels, curly cuts, etc. Those. it will be a parallelepiped. The calculated dimensions give a fairly pleasant proportion, and proportion in the design is already half the battle.

What to design in? Although I am related to design by line of work, my knowledge of 3D packages is, to put it mildly, superficial. In this case, the program should be more engineering than rendering. Specialized "Kads" for this purpose are heavy and unnecessary. A solution was found quickly enough - frivolous SketchUp is more than suitable for this purpose. It is so simple and intuitive that I was completely mastered in about an hour. The main thing he can do is quickly create any shapes, set dimensions, use simple textures. I believe that such a program is ideal for “home” purposes. You can easily use it, for example, to design a kitchen or even a small house.

Here is the body design:

Based on the drawing, a diagram of cutting the sheet emerges:

In general, the options are not bad in appearance, but purely structurally they cause difficulties. As a result, it was decided to trim the side walls with ash veneer, and cover the remaining 4 walls around the circumference with leather, or rather with high-quality automotive leatherette. The arquebus is beautiful in itself, but the woofer has a structural overlay on the front side of the housing that will not look very nice. Therefore, it was decided to make an additional decorative overlay (ring) for it, which will press it to the body, and at the same time add beauty to the speaker itself. The construction and design have been decided.

Tools

Before moving on to the next stage, I’ll outline what basic tools are needed for the job:

Circular.

Jigsaw.

Sanding machine.

Straight arms.

Without this kit, it is better to order the cases from a good craftsman.

Sawing

So, we cut the budget MDF sheet. I have already written that it is better to saw on special machines - it is inexpensive, but the results are accurate. But because I decided to make the body myself inside and out, then for the purity of the experiment I sawed it myself with a manual circular saw, and small pieces with a jigsaw with a guide. As expected, the perfect cut did not work out. After the cut, pairs of walls (left-right, front-back, etc.) are installed in pairs, adjusted with a grinder and/or electric planer and checked for perpendicularity with a square. And later during assembly they are finally adjusted after gluing. A loss of 2-3 mm is insignificant. But I still recommend sawing right away “at the base”, you will save a lot of time.

Housing assembly

The walls are glued together with PVA and tightened with screws. First we glue the body without the front wall.

Now there is a hole for the terminal block, as well as a chamfer in order to “sink” it. Initially, according to the design, the terminal block was supposed to be placed at the bottom. But during the process, it became clear that mounting the crossover in the center through the hole for the woofer would not be very convenient, so I moved the hole for the terminal block higher, and the place for the crossover lower.

You can close the box.

Now one of the very important stages is cutting out the holes for the speakers on the front panel. I have already said that the ideal speaker system is a single-way one. Why? Because sound propagates from one source to the listener without a time mismatch due to the (tiny) difference in distance that occurs when using a multi-band system. Therefore, it is best to place the speakers as close to each other as possible. This makes the sound image “dense.” We calculate the holes so that the distance between the edges of the speakers will be approximately 1 cm. The holes are cut with a jigsaw with a circular guide.

After the chamfers are removed, we attach the terminal block and speakers, and then drill holes for future self-tapping screws with a thin drill. Without them, firstly, the MDF itself may “spread open” when screwing in the screws, and secondly, during the final installation the speakers will be more difficult to position evenly. I thought for a very long time about how to position the speakers relative to each other, and came up with the following scheme:

Screw holes on external surfaces must be repaired before final finishing. I used epoxy. In order not to wait for one surface to harden, I sealed each surface with tape and moved on to the next one. When the epoxy had dried, I sanded it with a sander.

The veneer needs to be protected. I coated it with clear yacht varnish.

Now you need to cover the body with leatherette. There are many options for how to do this. I decided to do it as follows. A strip is cut 20 mm larger than the width of the case and slightly longer than the circumference of the case. On each side it is folded by 10 mm, the hem is glued with “special glue 88”. Then, using the same glue, the strip is glued around the circumference to the body. First the bottom (partially), then the back wall, then the top, then the front and the bottom again. At the last stage before gluing, the strip is cut in place and glued end-to-end. I glued all sides at once, i.e. didn't wait for each side to dry. After each side I took a short break (the glue sets quite quickly) and started on the next one.

If you really want to, then the phasic can be somehow refined.

Then holes are cut on the terminal block, “woofer” and “tweeter”. The skin on the terminal block and RF will be recessed downwards, so the diameter of the cutout can be left 5-10 mm smaller. The skin on the woofer will be pressed against the decorative ring, so you need to trim it so that it is not visible.

Final editing

First of all, we mount the crossover. The cross is home-made, based on a good element base. Air-core coils, tweeter film capacitors and MOX resistors are used. I didn’t solder it myself, but ordered it from smart guys.

Now we solder the required pair of wires to the terminal block and fix it on the body. The terminal block and speakers are screwed on with decorative black self-tapping screws with an asterisk head. The cover on the “squeaks” is screwed with similar screws, so it would be logical to use the same ones for the rest. The back wall is ready.

The midbass needs to be slipped under the skin, and pressed down on top with a decorative ring. Solder the remaining couple of wires and mount the speaker.

All? All. We screw the acoustic cable to the terminal block and begin testing.

Tests

The system was tested in the following configurations:

1. Receiver Sherwood VR-758R + acoustics.

2. Computer + Unicorn (USB-DAC) + Homemade stereo amplifier + acoustics.

3. Computer + E-mu 0204 (USB-DAC) + Sherwood VR-758R + acoustics.

A little about the configurations themselves. I personally think that at the moment the ideal option for a home music center is: computer + USB DAC + amplifier + acoustics. Digital sound without distortion is captured via USB and sent to a high-quality DAC, from which it is transmitted to a high-quality amplifier and then to the acoustics. In such a chain the amount of distortion is minimal. In addition, you can use completely different soundtracks: 44000/16, 48000/24, 96000/24, etc. Everything is limited by the capabilities of the driver and DAC. Receivers in this regard are a less flexible and obsolete option. The size of modern hard drives allows you to store almost your entire media library on them. And trends towards subscription to Internet content may eliminate this option, although this is not in the near future and is not suitable for everyone.

I will say right away that in all three configurations the acoustics sounded great. To be honest, I didn't even expect it. Here are some subjective aspects.

1. Adequate and natural sound. What is recorded is what is played back. There are no distortions in any direction. Just as I wanted.

2. Greater sensitivity to the source material. All recording flaws, if any, are clearly audible. High-quality mixed tracks are listened to perfectly.

3. Well-readable bass for such sizes. Of course, you can’t fully appreciate organ music on bookshelf speakers (it’s generally difficult to appreciate it on acoustics), but most of the material can be “digested” without problems. It's hard to expect more from such babies.

4. Very good attention to detail. You can hear every instrument. Even with a rich sound image and decent volume, the sound does not turn into a mess (the amplifier plays an important role here).

5. I would like to make it louder;) That is. The acoustics do not scream, but play smoothly. Although this is also not a small merit of the amplifier itself, because As the load increases, a good amplifier maintains linearity.

6. Long listening does not give you a headache. Personally, this often happens to me, but here it plays all day and nothing happens.

7. Concerns about incorrect panorama and strong dependence of the sound on the position of the listener were not confirmed. As far as I know, car acoustics have a specific sound phasing due to the location of the speakers in the cabin. Namely, I read about this set that its midbass is more universal in this regard. Which was actually confirmed. You can sit in the center in front of the speakers, or stand next to them sideways - the sound is excellent. There is a dependence, but it is very small.

As for the configurations themselves, the highest quality sound was achieved with the second configuration.

Firstly, a very high quality Unicorn DAC was used.

Secondly, the “home-made amplifier” is the know-how of one smart Tolyatti “sound specialist”. Here it is in a nice little aluminum case:

In a nutshell, we were able to find a circuit solution in which the amplifier retains its characteristics when the volume changes, i.e. does not distort the sound at any (constructively permissible) volume. Many amplifiers (even very expensive ones) suffer from this. It was amazing to hear how such an amplifier brought many speakers to life, i.e. made them sound the way they should sound. By the way, some industrial amplifiers (in particular, the Xindak, which is quite good in itself), were also rebuilt according to this scheme, and they got a “second wind”.

Have you compared acoustics with something else, you ask? Yes, for example, with ProAC Studio 110 – these are quite high-quality bookshelf acoustics, here’s a little about them. We compared them and realized that they definitely sound no worse. The “Proaks” may have a slightly lesser dependence of the sound on the position of the listener due to the specific placement of the inverter and the “tweeter”; somehow they cleverly calculated all this. As for the rest, it’s absolutely no worse, even I personally liked my homemade products better, but we’ll chalk that up to subjectivity;) I also put on headphones (quite good Koss) and compared them by panorama, highs and lows. Absolutely identical sound. Even at the bottom. In general, complete delight.

Costing by materials

Mid/bass speakers (pair): 3,000 rub.

HF speakers (pair): 3,000 rub.

Crossover (pair): 3,000 rub.

Sintepon: 160 rub.

Terminal (terminal block): 700 rub.

Screws: 80 rub.

MDF sheet, 22mm: RUR 2,750.

Scotch tape: 30 rub.

PVA: 120 rub.

Special glue 88: 120 rub.

Vibration isolation: 200 rub.

Figured ring-onlay: 500 rub.

Cable: 500r.

Total: 14,160 rub.

Some materials were or were received free of charge and accordingly are not taken into account here.

In custody

In any more or less complex device or complete functional system, absolutely everything is important. When it comes to a music system, the final result is influenced by a large number of factors:

Soundtrack quality.

A device for playing a phonogram.

Digital-to-analog converter.

Signal amplifier.

Wires.

Speakers installed in the acoustic system housing.

Correctly designed for speakers and high-quality assembled housings.

Diagram and accessories for the crossover.

This is a basic but not complete list.

It is wrong to think that the main thing is the amplifier, or the main thing is the wires, or the main thing is the speakers. A home music system is like an orchestra. And if in this orchestra some people play poorly and others play brilliantly, then overall the result will be average. Or, as they said in a very precise example: if you mix a barrel of shit with a barrel of jam, you get two barrels of shit.

There is another extreme. A good system costs incredible amounts of money. This means that each component should cost half a million. And phonograms must be exclusively on Super Audio CDs or branded records. Like a closed society of elite audiophiles. This is all bullshit.

I came to the conclusion that it is quite possible to assemble your own relatively budget system, which can be described in one word “Sounds”. And if, due to its specific features, it is better to use real existing solutions as a DAC or amplifier, of which there are a lot now. Then a correctly made (either independently or to order) acoustic system will sound better than a “branded” one purchased for the same money. Nowadays almost all components can be ordered online. Moreover, many manufacturers publish enclosure diagrams for their respective speakers. There is a lot of software for calculating housing parameters. There are many specialized forums online, and offline there are people with hands. Of course, it is impossible to be an expert in everything. As in any field, the main thing is to know the general principles.

The article does not claim to be the ultimate truth, but I hope that my thoughts and my experience will be useful to someone else.
Site administration address:

DON'T FIND WHAT YOU WERE LOOKING FOR? GOOGLE: