TYPES OF SPEAKERS
Speakers can be broken down into four major categories: woofer, midrange, tweeter, and full-range. Note that the frequency range overlaps to some extent across speaker types.
The woofer reproduces the very low sounds. Their most effective range is 20 Hz to 1000 Hz range, but they can extend to 3 kHz. Woofers can be any size from about four inches in diameter to over 15 inches in diameter. A diameter of 10 to 12 inches is typical in-home hi-fi speaker systems. Felted paper and polypropylene are common woofer cone materials. Note, in particular, the 12-inch subwoofer. This speaker has dual voice coils and can provide 120 watts of bass range stereo power from the speaker.
The midrange reproduces the broad spectrum of sound from about 1000 Hz to 10 kHz. Their most efficient range is 1 kHz to 4 kHz. The midrange speaker is responsible for playing back singing voices and most all instruments, yet it need not be physically large to do the job well. Midrange speakers come in sizes from about three inches to eight inches in diameter, with the smaller sizes more common.
Most midrange speakers use paper, cloth, or polypropylene cones to reproduce sound. Some midrange speakers use a small plastic dome. The design of the dome provides a wider sound dispersion.
The tweeter reproduces the high or treble sounds in the range of about 4 kHz to 20 kHz (and beyond). Tweeters are usually small — under two inches in diameter. Tweeters use a paper or cloth cone, a plastic or metal dome, or a piezoelectric element and diaphragm to make a sound. Tweeters suffer the most from narrow sound dispersion — the signal travels a narrow corridor through the air. For maximum listening pleasure, the sound should be dispersed to cover a full listening area. Tweeters are often outfitted with horns, baffles, and mechanical “lenses” to help disperse the sound. As with dome midrange speakers, dome tweeters have wider dispersion than conventional cone designs.
A full-range speaker is one that is engineered to reproduce most of the audible sound spectrum adequately. Full-range speakers usually represent a compromise over a system using distinct woofer, midrange, and tweeter speakers, and their frequency response is not as good. No single speaker can accurately reproduce the entire range of human hearing. Full-range speakers are typically used in compact, inexpensive “bookshelf” enclosures.
Coaxial and Triaxial
A coaxial speaker combines a woofer and midrange, or midrange and tweeter, as one unit. A triaxial speaker combines a woofer, midrange, and tweeter as one unit. In each type, the speaker employs multiple cones and voice coils, surrounding a typical heavy-duty magnet. Coaxial and triaxial speakers provide a marked improvement in frequency response over full-range speak-ers, yet take up little room. Many automobile manufacturers claim to have coaxial or triaxial speakers in their sound systems; however, they are not actual coaxial or triaxial speakers because the speakers actually have individual magnets and voice coils for each cone.
Some woofer and midrange speakers have a separate whizzer cone that acts as a tweeter. The whizzer is attached to the cone and voice coil of the speaker; unlike the coaxial and triaxial speakers described above, it lacks its voice coil. This improves the high-frequency response of the speaker, but only marginally. The whizzer is placed around the dust cap in the center of the speaker cone.
Vocabulary time for different kinds of speakers
There is a lot to the process when creating any vocal sounds such as singing or saying, “Hey there!” Vocal cords have to vibrate, the throat changes shape, the mouth creates an assortment of forms, and the tongue flutters around. Before the sound even leaves the mouth, it travels around the head for a brief moment, which gives it fullness. When you have a cold, your sinuses and nose are blocked, and your voice has that bothersome rasp to it. Fortunately, speakers don’t get colds and are a lot more manageable and practical. Speakers work by vibrating a cone made out of several options in materials. The cone affects the air nearby the speaker, and the resulting air pressure causes fluctuations that are heard as sound.
Beyond simply saying “Hey there!” speakers can reproduce the sounds of all sorts of instruments, voice, and just about anything that can be heard by us. Although all speakers produce sounds, they do not do it equally in quality or quantity (volume). There are different measurements, classes, and uses of speakers, made with a diversity of materials. With the plethora of choice of speakers available, it’s essential to know how each one is different, and how to apply the various speaker designs to your setup.
Terminology - About Speakers
Before proceeding with the design of speakers, several audio concepts should be understood. Through discovering the meaning of these ideas, you’ll be better able to understand the sections that follow in this guide.
Frequency response is the sonic range of pitches that a speaker system can accurately produce with flatness in response as the target. You’ll hear the term flat thrown around quite a bit in the audio world. It means that one range of frequency isn’t noticeably louder than others. A bassy setup won’t be flat because lower frequencies dominate the mid and higher frequencies. Frequency response is the change of the speaker output over a frequency range. The audio signal amplitude is held constant throughout all frequencies. Frequencies are measured in hertz (Hz) and stretch from about 20 Hz to 20 kHz. A hertz is a signal curve of one cycle per second. The sonic range from 20 Hz to 20 kHz is adequately the range of typical hearing, although many can’t hear high-frequency sounds. An even more typical hearing range for most grown-ups is 30 Hz to about 15 kHz. When a sequence of test sounds with constant amplitude is run through a speaker, every frequency in the 20 Hz to 20 kHz range should have about the same output level as the others. If this occurs, the frequency response is considered to be flat, and is expressed in specs as “20 Hz – 20 kHz, ±0 dB.” Pay attention to that phrasing at it is a very typical spec to see when purchasing speakers and parts. In practicality, no speaker has a perfectly flat frequency response. There is always a deviation around some midrange value.
The decibel is a way to measure signal amplitude. It is a measure of the ratio of the audio system output power compared to the input power. If the sound system resolves all frequencies the same with constant input power, then output power for all frequencies be flat. The higher the dB value, the more power the signal has at the output for a given constant input signal. Of course, the tones in real music are seldom even; that is, the sounds are of varying frequencies at many different input levels.
Nevertheless, the audio system should be able to reproduce those tones at its output accurately. It is a matter of what goes in should come out in the same form. If the speaker tends to favor specific frequencies and drop others, it is said to color the sound. Ideally, the speaker should reproduce all frequencies without coloring. Frequency response can be seen in graphs provided by sellers. Note that the frequency response is a two-fold specification. In one aspect, it’s the frequency range that the system can reproduce at all; and it also shows it’s the “accuracy” to which the frequencies can be reproduced relative to the original recording. This isn’t an entire telltale of the quality of a speaker. This is due to another term called harmonics. This term will be discussed later but consider the fact that the piano and guitar can play similar notes but sound entirely different. This is due to harmonics which color the frequency. Ideally, speakers wouldn’t color any signal it receives, but there is always going to be some coloration. Some speakers will do this more than others.
The lower and upper frequencies at which the response is 3 dB below the midrange value are usually identified by f, and f” respectively.
Dynamic range is the difference between the softest and loudest portions in a music selection and depends mainly on the source of the audio signal. For example, compact disc players, which have a theoretical dynamic range of 96 dB, have a practical dynamic range of 85 dB to 90 dB. This is much better than other recording mediums.
These figures are really applicable for laboratory tests; it’s unlikely you’ll ever find a piece of music with this kind of range — the 1812 Overture included. A 90 dB range means the power of the loudest signal has 10′ (one billion) times as much power as the softest signal! Even though the signal input and the amplifier have excellent dynamic range, other elements in the audio system, particularly the speakers, restrict the dynamic range. Therefore, if you have a sound system with an overall dynamic range of 60 to 70 dB, it is still a pretty good system.
Sound dispersion is the spreading of sound as it leaves the speaker. A narrow dispersion means that the sound does not spread very far. The listener must be directly in line with the axis of the speaker to obtain the best results. A wide dispersion spreads the sound over a larger area, so the effective listening area of the speaker is increased—dispersion changes at different frequencies, even with the same speaker. Generally, a speaker is omnidirectional (disperses sound equally in all directions) up to the frequency where the diameter of the cone is equal to the wavelength of the sound. Then it becomes directional for higher frequencies (shorter wavelengths).
Since the velocity of sound is 13,200 inches per second (at sea level), the omnidirectional frequency limit can be determined by dividing the velocity by the speaker’s diameter. The omnidirectional limit for a 15-inch speaker is 880 Hz. For a 12-inch, 8-inch, and 4-inch speaker, the respective frequency limits are 1.1 kHz, 1.65 kHz, and 3.3 kHz. This limit is not exact (speaker materials and cone geometry effect, as does the design of the speaker enclosure), so a 1-inch speaker will probably extend to about 4 kHz, and a 2-inch speaker to about 8 kHz.
The sensitivity of a speaker is a measure of its sound output level for specified electrical input power. The manufacturer applies a continuous single-tone signal, having a specified electrical power level, to the speaker. The audio output level is then measured at a specific distance (usually one meter). The sensitivity specification is often used to denote the efficiency of the speaker, or how well the speaker converts electric signals into sound. A speaker that puts out a great deal of sound for a given signal input is said to be efficient (but not necessarily better sounding than another speaker).
Speaker cone motion should faithfully follow the applied electrical signal. Any tendency of the cone to vibrate at frequencies that are not in the input or to continue to vibrate after the signal stops will color the sound and adversely affect its quality. This unwanted cone motion must be damped. Speakers are made with built-in damping mechanisms to tame these unwanted vibrations. Components that damp the action of the speaker are:
• Cone material
A stiff cone doesn’t vibrate as readily as a pliable cone. Most speakers are made with a relatively rigid cone, one that will maintain its durability over years of use. The suspension and spider hold the cone to the frame. The spider is usually made very supple to accommodate the motion of the cone readily. The suspension can be made hard or soft. The amount of stiffness of the suspension is usually referred to as compliance. A low compliance suspension means it is stiff; high compliance denotes a pliable suspension. The size and weight of the magnet determine the sound output for a given input signal, but it also determines electrical damping. A small magnet does not provide much force against the voice coil, so the speaker acts like a car with weak shock absorbers. Damping can be tightened by increasing the size and weight of the magnet. Damping is also determined by the size and design of the speaker enclosure. Ideally, the enclosure complements the damping characteristics of the speaker, working with the design of the speaker to produce clear and natural sound.
Any sound that you hear coming out of your hi-fi speakers that wasn’t originally recorded is deemed distortion. Some distortion is created in the electronic circuitry, including the tuner, amplifier, phonograph, and cassette deck. That distortion can be in several different forms, but the result is usually the same: “fuzzy” sound. Distortion also is contributed by the speakers, no matter how good the design. The goal is to minimize the distortion. The four basic types of speaker distortion are harmonic distortion, noise, transient response, and clipping.
Harmonic and intermodulation distortion are characterized by the presence of frequencies in the speaker output, which are not in the electrical input. These spurious signals are caused by imperfect driver and suspension behavior and are especially prominent when the speaker is driven to very high volume levels. Proper design of the speaker enclosure can significantly reduce this type of distortion.
Speaker noise is usually a raspy sound caused by a damaged component, such as a torn cone. Bits of acoustic batting used to muffle the sound inside a speaker cabinet is another common source of the noise. The batting falls on the cone and vibrates as the cone moves.
The transient response is the time delay as the speaker cone begins from rest to respond to a sudden, sharp, electrical pulse. The transient response (how fast the speaker reacts to make a sound) depends on several factors, including the stiffness of the cone and suspension, and the enclosure design. A stiff cone and suspension act to resist movement, so the speaker may respond slower and cause a delay in the reproduced sound.
Clipping is a distortion that occurs when the speaker cone cannot move as far as required by the audio signal. When the cone “bottoms out,” the speaker is not able to accurately reproduce the sound. Damage may result if the speaker is left operating in this condition, because the cone, bobbin, or suspension may eventually be torn. Clipping is most prevalent in woofers, where the cone must move a great distance to produce a high volume level at very low frequencies.