We have covered input devices and signal processing, getting an acoustic sound wave into electrical energy and subsequently manipulating it. But how does it come out? For the recording professional, the final medium is magnetic tape or magneto-optical disc; for those in sound reinforcement, we need to concentrate on loudspeakers and their cohorts, amplifiers.
MAKING IT LOUD
Amplifiers are by definition electronic devices that take an existing electrical signal and make it bigger hence the name, "amplifiers." Amplifiers exist in all sorts of electronic circuits, from the internal devices within some microphones to the input channel on a mixing desk, but the really big ones, called power amplifiers, take an analog line-level audio signal and convert it to something that can properly cause a loudspeaker to physically vibrate, and thus, make noise. An analog line-level audio signal is usually measured on the order of millivolts, which is nowhere near enough power
Ideally, the amplifier shall not introduce any nonlinearities into the audio waveform; that is to say, a comparison of the input signal and the output signal of an amplifier should be exact except in terms of amplitude the amplifier makes the hills and valleys of an amplitude-versus-time graph bigger, but the waveform shape should not change. That is the scientific ideal, at least, but in the real world amplifiers can add slight variances to the audio signal. We call any signal not present in the original input signal distortion, which can take many forms.
Harmonic distortion is introduced into an audio system most often when an input signal overloads a given component, resulting in clipping. An electronic component has certain parameters under which it operates to its maximum potential; operate the component at levels below that threshold and one runs the risk of unnecessarily increasing the noise floor; operate the component at levels above that threshold and one is in danger of clipping the waveform. Audio signals, as we know, are sinusoidal in nature; when they are input to a component at a high level, the lovely peaks of the waveform are then flattened because the component cannot handle viewed graphically, the peaks are clipped off, resulting in a high-energy solid line.
Crossover distortion is introduced into an audio system when certain types of amplifiers push-pull designs generate distortion as a waveform travels between the positive output section and the negative output section. At zero, where the handoff between the output sections occurs, distortion can occur. Most modern solid-state amplifier designs do not exhibit this problem.
Power amplifiers come in many shapes and sizes; their selection depends on application.
Slew Rate refers to the time it takes an amplifier to sweep from one state to another. A human analog is that fuzzy time in between waking up and getting that first sip of coffee from a state of relative inactivity to a state of being fully aware. In an ideal universe, the slew rate of the amplifier should replicate exactly the slope of the input signal, but it doesn't, and amplifier designers take into account the amount of time the amplifier needs to really wake up.
What else should I say?
Loudspeakers are transducers they convert electrical energy into sound pressure waves. Loudspeakers are the last device in the sound system chain over which the sound designer has full control. With the exception of room characteristics and acoustic treatment, once the sound wave leaves the loudspeaker cabinet, it's all up to the audience to actually listen, and it is up to the designer to properly utilize loudspeaker equipment to provide an even listening pattern to as much of the venue as is humanly and technologically possible.
Most loudspeaker cabinets and their components share similar design characteristics. Because of the large range of frequencies that are needed to properly reproduce the audible range of sound, in most cases, more than one component is used. A loudspeaker component, called a driver, that reproduces the lower frequencies of sound is called a woofer, while a component that is dedicated to reproducing higher frequencies is called a tweeter. Midrange driver is the name given to a component that is designed to reproduce the middle frequencies. Supertweeters are sometimes used to reproduce very high frequencies; there are some designers who use them in theatrical reinforcement but they are primarily found in old, 1970s-era nightclub sound systems. At the opposite end of the frequency spectrum, subwoofers or sometimes sub-bass cabinets reproduce very low frequencies, down to the point where the listener feels the sound rather than hears it.
Because each component is designed to reproduce a different portion of the frequency spectrum, their construction differs slightly, but all follow a basic principle: a cone of paper or paper-like material is suspended within a frame, and connected to the loudspeaker's input. A permanent magnet surrounds the center of the cone, and as alternating current is applied to the loudspeaker cone, changes in the magnetic field potential causes the cone to move away from or towards the permanent magnet. Study your physics textbook, and you'll figure it out.
Since low frequency sounds have a larger wavelength, low-frequency drivers tend to be larger more mass is needed to convincingly provide enough air movement at the given frequency. Common sizes of woofers range from four inches in diameter to eighteen inches in diameter. Diameters above this size generally become too inefficient for real-world applications due to the amount of mass in the speaker cone; loudspeaker engineers can provide two drivers side-by-side to move a greater volume of air without sacrificing efficiency.
Most designers use off-the-shelf loudspeaker systems which are usually a wooden box with a woofer or two and a tweeter or two; only a few designers actually use bare components in different locations. Off-the-shelf loudspeaker systems come in many shapes, sizes, weights, constructions, and even colors. The more important characteristics of a loudspeaker are frequency response, power handling, and coverage angle, although their physical shape and electrical impedance play significant roles as well. The designer can specify the proper type of cabinet given its application whether it is a loudspeaker dedicated to reproduce music, play sound effects, reinforce vocals, hang underneath a balcony, line the front of the stage, or some combination of uses.
Frequency response is the range of frequencies a component or system of components can faithfully reproduce with very little deviation. A loudspeaker cabinet, comprised of a woofer and a tweeter in a box, may tout a frequency response of 47 Hz - 15,000 Hz, ±3 dB. These numbers mean that the speaker will happily reproduce all frequencies between 47 Hz (very low) and 15 kHz (very high). It doesn't necessarily mean that the loudspeaker cannot reproduce frequencies above or below the specs, but the performance of the speaker at higher or lower frequencies may suffer. Manufacturers often provide a frequency response graph, detailing the response of the speaker from 20 Hz - 20 kHz, showing peaks and valleys as the loudspeaker's response varies. These calculations are usually made with a measurement microphone on-axis with the loudspeaker in a free field at a distance of approximately one meter and an input signal such as pink noise (all frequencies with equal energy per octave). These numbers, then, are the best-case scenario of the speaker's performance.
Another important measurement is the speaker's coverage angle. Although a loudspeaker has a front and a back and generally it will sound better when the listener is in front of it, just how much to the left or the right can we go before the loudspeaker starts to lose its efficacy? Since low-frequencies have a long wavelength, they are able to penetrate walls and bend around obstacles. Many engineers consider a low-frequency cabinet omnidirectional and this is mostly true; turn up the bass on your home stereo and stand behind the loudspeaker listening to the low frequencies. High frequencies behave in a different fashion. Because of their shorter wavelength, the sound waves dissipate quicker and are restricted to a given field of focus. Loudspeaker manufacturers thus provide coverage angle data. A loudspeaker may exhibit a 40°V x 90°H coverage pattern. This measurement indicates that, using the center of the high-frequency component as the 0° axis, the loudspeaker can effectively reproduce its touted frequency response 20° above and below the 0° axis, and 45° left and right of the 0° axis. Rotating the speaker 90° will of course mean that the speaker will be effective 45° above and below 0°, and 20° left and right of 0°. Coverage angle helps designers choose loudspeakers. If multiple loudspeaker cabinets, fed the same or essentially the same input signal, both cover a given area, phase cancellation in the form of comb filtering will occur due to differences in arrival times at the listening position. The result? Bad sound. Using time-alignment and level adjustment, it is possible to alleviate these problems and create a good sound image for the listener, but that's for another chapter.
Manufacturers often also provide a polar graph of the loudspeaker's response, which uses both the frequency response and the coverage angle. The polar graph is circular, with 0° being the axis point (center of loudspeaker, usually), and response versus physical location can be easily seen; as we travel 45° degrees away from the axis, we notice the horizontal coverage getting worse and worse; at the rear of the loudspeaker, high frequency coverage is at a minimum until we circle back around, approaching 315° (-45°).
dome tweeters versus compression drivers, horns, bi-radials,
Baffles, vented LF, corner, half-space, quarter-space.
Frequency response, shape, impedance, powered, unpowered, coverage angle.
Coverage angle, intended coverage, system designs
Speaker selection should depend on the type of production and also the type of program material one is planning to use them for. In the past, the standard for music program reinforcement was the Altec "Voice of the Theatre" A-7 cabinet. One can probably still find them in old movie theatres and even in stage theatres. Old '70's rock concerts used A-7s. These are very nice cabinets. They are also behind the times.
"Processor-Controlled" systems have taken the lead in critical reinforcement situations. Processor-Controlled systems involve speakers and a specially-designed-for-a-specific-speaker processor unit, which has all sorts of neat equalization and phase circuitry inside of it that will make its corresponding speakers perform the best (they also include overload-protection circuitry to make sure you don't damage the drivers). Apogee, Meyer, Bose, and Electro-Voice all manufacture processor-controlled cabinets. Apogee, Meyer, and E/V to an extent, have taken the lead in theatrical reinforcement cabinets. The two industry-standard vocal processor-cabinets are the Apogee AE-5 and the Meyer UPA-1C, and Electro-Voice's Delta-Max 1122, to a smaller degree. They all look very similar. They are wedge-shaped, which allows clusters of them to be flown or otherwise placed. These are very clean-sounding cabinets with excellent high- and mid-frequency response. Their low-frequency response is a bit smaller since they employ twelve- to fifteen-inch woofers; to compensate they are designed to be used in conjunction with specially-tuned proprietary subwoofer systems. The AE-5s and UPAs are excellent for vocal reinforcement. Apogee, Meyer, and Bose all manufacture smaller processor-controlled systems-- these can be used for under-balcony fills, surround, or for reinforcement/playback in smaller venues.
These same speakers can be used, in conjunction with their corresponding subwoofer units, for music reinforcement. Meyer has a complete line of units for music reinforcement; if I knew what they were, I'd list them; Apogee also has a complete line, and if they'd send me stuff like I keep asking them to do, I'd list them, too. Using processor systems is ideal, but not necessarily economically viable for small-budget amateur shows. Usually, full-range non-processor-controlled cabinets will suit music reinforcement fine. Cabinets made by JBL, EAW, or E/V should work very well, although they may sound a little less "clean" than a processor cabinet. Look for a very wide frequency response (especially in the low-end if you are doing somewhat of a rock musical), and also look for high efficiency ratings.
For surround-sound or underbalcony fills, check out small monitors made by Yamaha, JBL, or Bose. Many small studio monitors will work well as reinforcement speakers-- just be careful not to overload them with excessive program material. Another feature to look for is mounting options or rigging points. Speakers cabinets with no rigging points will not take well to having holes drilled in their cabinets. This is very very dangerous. As a last resort, build strong frames out of metal (learn to arc weld; it's fun) to hold them, and attach rigging points to these. Remember the safety ratio of 5:1-- if a speaker weighs ten pounds, the rigging materials should be rated for at least fifty pounds.
For strong low-frequency response, for dance-clubs or loud music reinforcement/playback, check out subwoofers. For a theatrical effect, place subwoofers under seating platforms or even in the plenum below. For more information, check out THX Systems or Dolby Surround Systems or Sony Digital Surround Systems used in movie theatres. Some dance-clubs install subwoofers underneath the dance floor. Some even have systems that vibrate the dance floor in time to the music. Weird, but cool.
Proper location of speakers is also key in the whole design. A vocal cluster should not be located twenty feet upstage and slightly to the right. Make your decisions quickly on speaker placement and let the lighting and sets people know them early in the game. [You don't want to have to let the sets people move your speakers two linesets upstage an hour before the house opens and find that they have broken the Speakon connector on one of the main cables. It can happen.] The center vocal cluster, if there is one, should be located, as the name implies, in the center. The idea is to cover the audience as equally as possible-- moving the vocal cluster to the left or right without compensating equally on the opposite side will make for some interesting reflections in the house and will most likely not cover the house equally. The house left and house right stacks should be hung, or stacked, equally in the vertical plane. Otherwise, holes in frequency response may occur in certain parts of the house. Check out the dispersion angle characteristics of each speaker and align them according to that... or simply listen to them in different parts of the house and align them that way. If a surround effect is your goal, place the speakers according to that goal. Try to balance the house between left and right-- don't have lopsided design for reinforcement.
There are, however, designers who don't necessarily use the directional characteristics of speakers in their reinforcement designs (Martin Levan comes to mind). Use some creativity to achieve the sound you want. All this text is meaningless without trying and seeing if it works or not...ke the high-impedance -10dBV unbalanced signal and convert into a low-impedance microphone level balanced signal.
Continue to Sound Reinforcement. Return to the Sound Index.
Comments, Questions, and Additions should be addressed via e-mail to Kai Harada. Not responsible for typographical errors.
http://www.harada-sound.com/sound/handbook/speaks.html - © 1999 - 20002 Kai Harada. 19.05.2002