Technical hints on loudspeakers

premise

Speakers are the elements of a sound system that convert an electrical signal into an acoustic wave. They are the last elements of the chain and, for that reason, they are very much responsible for the impact on human ears.

For these reasons, the design and implementation of these elements also requires an adequate knowledge of the science of psychoacoustics and reproduction techniques of sound in the environment.

This area of design requires designers to have a deep knowledge of various theoretical and technical fields. Knowledge of electronic networks, the fundamentals of electro-magnetism, acoustic mechanics and psycho-acoustics all play a part. 

All this makes the design of speakers a fascinating field for many people.

Before going on let’s take a look at the following links:

• Acoustics
• Speaker
• Diffuser

speakers for home use

There are essentially two types of speakers for home use: passive and active ones.

Active speakers are composed of a set of transducers or speakers, coupled with an audio amplifier. Simply, an active speaker is a case with speakers, containing an electronic circuit for signal amplification. This kind of speaker does not need amplification, but only the external sound source to be amplified It however, requires a power supply. Small computer speakers are generally of this type.

Passive speakers do not contain any signal amplification circuitry; they receive an already amplified signal. They are not connected to the power supply but only to an external amplifier. This type of speaker is typically that of a normal home entertainment center.

This section will deal with passive speaker construction techniques. The difference between the two types of speakers is minimal and mainly in circuit configuration. The techniques to achieve a correct sound reproduction are the same.

We must bear in mind that a system of sound reproduction, must reproduce in a home environment the stage or recording studio study where the concert was actually recorded. The psycho-acoustic and environment analysis are essential to achieve credible results.

The perception of sound

The study of sound perception has radically changed the speaker construction techniques.

The two main areas in which we should focus our attention are:

• biological and psychological perception of sound,
• Reconstruction of sound levels and distances.

Sound and the ear

The ear’s behavior and its response varies in both frequency (pitch) and intensity (volume). It appears that the listener has different perceptions of sound depending on intensity and frequency. For example, the perception of "lows" increases with volume. At low volumes, the listener hears only the high notes of the featured female vocalists and, at higher volumes, the sound balances out bringing back the lows and eventually they appear even more prominent than high and medium frequencies.

Furthermore, the ear has a resonance frequency, due to its shape and length, of just above 2500 Hz. This means that the ear is very sensitive to sounds in which this frequency is prominent (a child’s cry ) and that they are detectable even at low volumes.

See the graphs produced by Fletcher and Munson in the 1930s.

Stereo and the brain

The brain can be "tricked". In the sense that it, because of its natural condition, it processes simultaneously Information coming from two separate sources: the ears. Providing the ears with slightly different sounds, we can trick the brain into processing the sounds as if they came from a specific position and distance. The ability of the brain to "reconstruct" a virtual environment is an essential element of all the solutions adopted for the realization of excellent sound systems or Hi-fi.

A very important aspect regarding the perception of sound, is the meaning that the brain assigns to various frequencies. At high frequencies (usually over 1500 Hz), the brain interprets the depth of the sound source based on detectable differences in intensity at different frequencies. Low frequencies are more "enveloping" and less well localized even though the brain processes them based on their proximity by processing the arrival times of various the waves. That means that when the listener hears several high notes of different intensities, he/she will tend to believe the louder ones to be more distant than the weaker ones. Just as the sound becomes lower (midrange) the difference in intensity becomes less important to the reconstruction of the source’s distance; at low frequencies, the sound that reaches the ear first becomes more important.

All this has important implications on the choice of home speakers and on the position of the ideal listening point.

Environment

The environment is absolutely crucial for the reproduction of sound. The environment concept includes the size and shape of the listening room, the occupation of space, the listening position.

The size and shape, in particular, are more relevant to the reproduction of the bass notes (lows). If the listening room is a normal apartment room, usually not bigger than 5x7 meters, the limits of proper representation of low frequencies will be of 50 Hz. Below this frequency the sound wave will tend to not be detected correctly: it will tend to spread out, or to resonate in the listening environment creating dangerous (for high fidelity) stationary sound waves. Other stationary waves can be formed due to the other dimensions of the room, for example around 120 Hz relative to the height of the ceiling.

For a complete discussion on the issue we should also approach the topic of direct field and reverb field. This means that the listener is not reached only by the sounds coming directly from the speaker (direct field), but also by those that bounce on furniture and walls before reaching the ear (reverberant field) and can help to reinforce or cancel certain sounds thus altering the quality of the sound.

We should also consider the materials used in the construction of the building, thin walls allow the sound to exit smoothly and decrease the perception of bass frequencies by the listener inside. For this reason there is a tendency by some speaker manufacturers to adapt their products to the type of market or specific type of housing. The most obvious case is that of U.S. manufacturers that design their speakers to be used in predominantly wooden houses with thin walls, unlike in Europe where brick and concrete are predominant.

The listening room should not be too symmetrical in order to avoid resonances or cancellations. We must always keep in mind that sound waves bounce around everywhere before reaching our ears

The listening position is a complicated thing. No speaker can function properly if poorly placed. We should usually follow isosceles triangle rule in which the distance between the speakers is 2 meters and the distance from the listener  2 and a half meters or in proportion to the listening environment. The angle of the speakers determines greatly the perception of space and depth. Furthermore, individual speakers have different directional emission characteristics; so altering their angle and orientation can significantly improve the predominance of certain frequencies over others. All these simple considerations are simply to clarify the scope and design constraints within which the principal designer of speakers must move.

Basic configurations

There are at least two different configurations:

• multichannel systems
• stereo

Between these two there is a difference mainly in the number of speakers used and in the sound source which changes radically.

The first type requires a number of amplifiers equal to the number of speakers installed and this depends on the standard multi-channel you choose, for example, there are multi-channel systems from 5 even up to 8 channels. To understand the reasons for these setups, we should see the links: Multichannel audio and DTS.

In this simple description, we will restrict ourselves to the 2-channel stereo systems because the design basics are common to all systems.

Another area concerns the single-speaker configuration:

• single driver speaker,
• 2-way speaker,
• multi-speaker.

These configurations are based on different design choices. The reasons for choosing these solutions may be commercial, design and electro-acoustic. There is no guarantee that a single driver can cost less than a 2 or 3 way system.

General characteristics of the speakers

Before focusing on these systems we should know speakers a little better.

They are electromechanical transducers (they convert electric energy into mechanical energy) that work in the audio spectrum between 20 and 20,000 Hz. The acoustic wave is a wave of air that spreads . The speaker is then like a piston in every respect; it is capable vibrating and producing compression and decompression of air at various frequencies.

Each wave’s energy depends on the frequency at which it travels. High-frequency waves develop more energy than low frequencies on the same air mass. This means simply that when a speaker produces a loud high sound it does not need to move volumes of air as large as when it produces a low sound.

So, while low frequency waves need a larger  moving mass,  high frequency waves do not and but only require fast moving air. From this we can deduce the reason why 2-way systems exist and what characteristics our speakers should have. There are speakers of different sizes and weight optimized to reproduce high or low notes.

Another important aspect in the evaluation of speakers is the analysis of dispersion. The most popular speaker,  as we have already mentioned, is the piston type.  Calculations and measurements show that this model behaves differently depending on the frequency of sound emission: up to a certain point the "piston" is drowned in the wave that it produces and at a certain point it is as if it is shot forward. In other words, the speaker appears to be dispersive (emits a spherical wave)  up to a certain frequency and from there on it appears to be directional (emits a portion of a spherical wave).

The formula that allows us to understand how often a given speaker changes from good dispersion to directivity states that this threshold depends on the diameter of the emitting membrane. In order to remain around the limits of good dispersion speaker diameter must be less than or equal to half the wavelength that it is emitting. For example imagine a speaker for medium and low frequencies of 17 cm in diameter; we can say that it maintains good dispersion up to 1000Hz, as this frequency has a wavelength of about 34.4 cm.

One-way

Single driver speakers are characterized by a single speaker, called broadband, capable of reproducing all frequencies necessary. To do this, the speaker must have excellent characteristics in the whole spectrum of music. Its moving parts must be light enough to vibrate at high frequencies and must also be able to move large volumes of air to reproduce low frequencies well. In short, it needs to be big and light.

Two-way

2-way speakers are popular because they represent the best compromise. They are characterized by a unit designed for low frequencies (called a woofer) and one for high frequencies (tweeter) that has a light emission surface and is usually much smaller than the woofer.

Each of the two speakers then deals only with a specific set of frequencies and therefore expensive materials and construction techniques are not necessary, unlike in broadband speakers. In order to work correctly the two speakers must be connected to an electronic filter (crossover) that is responsible for distributing the right frequencies to the specific speaker.

A 2-way speaker is therefore characterized by a mobile resonance chamber, inside which there are an electronic filter externally mounted and two dedicated speakers.

Multi-channel

A multi-channel system is very similar to a two-way system with 3 or more speakers. The reason for this is essentially to accentuate the division of the audio bandwidth for 2 reasons: to distribute less frequencies to individual speaker units, thus making then more specific and extending the reproduction limits of the sound spectrum band, where most speakers show a decrease in sensitivity.

In this type of speaker the circuitry of the electronic filter can be very complex and expensive.

Analysis of the specifications and constraints

We have seen so far a general overview of the constraints and of the fundamentals that characterize the creation of a system of loudspeakers for home use.

The designer must necessarily define the constraints and characteristics of the system to be realized.

It is certainly wrong to accept that the choice of a system configuration can "sound" better or worse. The objective is and remains high fidelity and different design choices are not aimed at finding a better or worse quality, but rather to obtain a high fidelity environment and listening position under all conditions and constraints. This set of constraints include economic conditions.

To complete the overview, the client needs to produce precise information on his/her use for the speakers. This is not about what kind of music he/she listens to but about knowing what kind of listening he will use them for (it would be unthinkable that a sound could be reproduced correctly if it came from a symphony orchestra but not from a country band). In order to understand the concept of the kind of listening we should know that of "noise event". But simply, we need to know how the user intends to experience the acoustic event: alone or with friends, sitting or standing in the center of the room like at a party and whether the speakers will be devoted to analytic listening or entertainment, if they are to be used for home theatre or listening to music only.

Since it is often not possible to obtain this information, we can classify at least the most common types of "standard listening":

• analytic listening,
• shared analytic listening
• heterogeneous listening
• distracted listening as background
• involved group listening.

These are just some of the potential different uses and, of course, do not include any other use other than home use (such as professional or public address) which are not covered in this section.

Analytic listening (not shared) means that users consider music as the only subject of their listening. In the case of heterogeneous listening, distracted or group listeners put as subject of the session not the music (not necessarily as a single entity), but, for example, their activities (including work) or entertainment, creating atmosphere for a party or a celebration ...

In the first two cases,  there is the possibility of choosing the listening position, with some attention to the shared configuration. Other types of listening instead might be highly variable, particularly in terms of listening position.

Let’s take a look at a few examples.

Analytic listening

To listen analytically we can assume that the user will seat himself in the best position, so we should respect the rule of the isosceles triangle: 2 by 2.5 meters.

In this case the size of the environment can become the main parameter to optimize. Particularly in small rooms a single driver system can be a solution. The systems are based on single driver broadband speakers characterized by high directivity at high frequencies. This phenomenon forces us to utilize these systems with an unchanging listening position.

The fact that the speaker is directional can diminish the effects of the reverberant field and listening may result to be natural and not "colored", less dependent on the environment.

A system based on broadband speakers would seem to be perfect for listening in small spaces where the listening position does not change.

In the case of shared listening at loud volumes, a two-way system might be appropriate. This configuration allows  greater power distribution than a single driver and ease of placement and versatility.

This is just one example of how to choose your configuration.

Other types of listening

The diffusion of sound in conditions where the music is not the protagonist is offers more solutions, from the least expensive to the most complex.

Generations of electro-acoustic engineers have defined templates to make the correct reproduction of sound levels or the so-called "sound stage" as independent as possible from the listening position.

These models, due to their their complexity, are essentially based on multi-way systems. Different orientations of the axes of emission of each individual speaker drivers can create a situation where the movement of the listener around the ideal listening point does not affect the proper depth perception and sound levels.

This configuration is particularly suitable if the sound event must necessarily be shared with other listeners.

Copyright © 1996 [Audio Service snc] All rights reserved.
Last updated: 11-05-10. http://www.audioservice-bologna.com