Technology can be fun when it involves drooling over new gear features, ogling bold new designs, and wallowing in the benefits of an enhanced feature set. But technology stops being fun and becomes more like taking a math test when it involves, well, math.
On the other hand, a little technology savvy can increase your understanding of your gear and can often get you out of a jam. Knowing a few basic principles can help you solve an assortment of annoying problems that commonly plague gigging bands.
To get you started, we'll discuss five basic yet important electrical and audio concepts that musicians regularly encounter. We'll tackle some of the more technical aspects of these subjects using simple, everyday language, keeping the math and jargon to a minimum, in order to make these seemingly daunting concepts clearly understandable to the technically challenged musician. Of necessity, we are going to offer abbreviated and sometimes quick-and-dirty explanations. Be forewarned that in almost every instance there are plenty of exceptions, “yeah buts,” and so on. If you want greater precision and detail than we can offer here, we recommend starting with Scott Wilkinson's book Anatomy of a Home Studio, published by ArtistPro (www.artistpro.com).
Once you've mastered the principles presented in this article, you'll be able to do a number of cool things to impress your friends and bandmates:
- Safely choose or wire an extension speaker cabinet.
- Understand the function of a crossover in a biamped or triamped system.
- Know in what situations a balanced audio line will deliver better results than an unbalanced line.
- Optimally combine the signal of two mics aimed at the same source.
- Grasp how sound-pressure level, decibels, and frequency influence perceived loudness.
Are you relaxed and ready to learn? Good, then let's get technical!
P.A. systems are often modular affairs, with separate speaker cabinets and power amplifiers. This provides more versatility, as you can often mix and match speaker systems to suit the job. But before you start using your Crown power amp with your guitar player's Celestion cabinets, you have to know that you can safely match up those particular components. If you get it wrong, you could permanently damage your amplifier, speakers, or both. (Of course, a speaker system designed for guitar amps might not be the best choice for a P.A., but that's a separate question.)
The most important factors in matching an amp with a speaker or speaker cabinet are their power and impedance ratings. These specs are related, so we'll discuss them together. We are not going to get into much detail because that would require a separate article; we're just going to give you enough to assemble a properly matched system.
Keep in mind that a given power amplifier is designed to deliver a specific amount of AC power, and that a given speaker system is designed to handle a specified amount of AC power. If your amp delivers too much or too little juice to the speaker, the amplifier, the speaker, or both can be damaged. The amplifier and speaker are a team, and they need to be properly matched. We're going to discuss how to make a good match
Don't Impede Me!
You undoubtedly have heard other musicians talk about high- and low-impedance inputs on mixers, power amps, speakers, and guitar amps. Indeed, impedance is an important factor in matching parts of the signal chain, and it pays to understand at least a little bit about it.
For practical purposes, an electrical current always meets some opposition when flowing through a circuit, even if that circuit is merely a straight copper wire. Therefore, a speaker's electronics always present a certain amount of opposition to the AC electrical flow coming from the power amplifier. The opposition to alternating current is called impedance because it impedes, or hinders, the flow of electrons. Impedance is represented by the letter Z in equations, and it's measured in ohms, which is symbolized by the Greek letter omega (ž). (Impedance is closely related to resistance, but the two are not identical. The quick-and-dirty explanation is that resistance is the opposition to direct current, rather than AC. Also, impedance is frequency-dependent, while resistance isn't.)
Every speaker system has an impedance rating that indicates how much opposition its circuitry presents to the signal coming from the power amplifier. Common impedance ratings for speakers are 2, 4, 8, and 16ž.
If the speaker impedance is high, it won't let much current flow, so it doesn't demand much from the voltage source (the power amp). In such a case, the speaker is said to present a small load to the power amplifier. On the other hand, if the impedance is low, a lot of current can flow, and that puts a high demand on the power amp — that is, it presents a large load.
Remember: high impedance, small load; low impedance, large load. Got it? Good, let's talk about power.
Figure 1: Wiring diagrams for connecting speakers in series, parallel, and series-parallel configurations
Electrical power measures how much work a given voltage and current can do when presented with a specific impedance load. This is why we said earlier that the concepts of electrical power and impedance are related. Electrical power is measured in watts, the symbol for which is the letter W.
When you check the power-output rating of an amplifier, therefore, you'll see that power rating expressed as a number of watts of power into a number of ohms of impedance. On the back of the power amp, you'll often see a notation that reads “250W/channel @ 4ž.” This means that each channel outputs 250 watts, assuming the amp channel is connected to a speaker cabinet providing a 4 ž load.
Matching the impedance of the load (the speakers) to the output of the amplifier is crucial for achieving maximum efficiency in a system. Maximum efficiency means that all of the power is being used to drive the speaker, and none is being wasted as heat.
If an amp expecting a resistance of 4ž encounters a lower impedance (a larger load), such as 2ž, it will work harder and harder to deliver current to keep up with the current-sucking load. Eventually, it will heat up and burn out. On the other hand, if the amp encounters a higher impedance (a smaller load) — say, 8ž — it simply will deliver half as much power (in theory), which is wasteful but generally not dangerous in and of itself.
However, you can still have problems if the amp's power drops so low that it can't properly drive the speakers. In that event, the amp can start distorting (clipping) the signal, and distortion can rip speakers apart. In fact, assuming we aren't talking about extremes, you're more likely to blow out a speaker by using an amp that is not powerful enough than by using one that is too powerful.
So if you're faced with a mismatch, remember, assuming you have sufficient power, “Four into two won't do; four into eight is great.” The same applies to other impedance ratings: an amp designed to work with 8ž speakers might be fine with a 16ž speaker system but not with a 4ž or 2ž speaker system.
It is important to note that in the real world, cutting the speaker impedance in half does not necessarily cause the amplifier to deliver exactly twice the power. There are many places in a circuit where power is lost, including the speaker wire. Higher current can cause greater losses in transistors and the power supply, as well as in the wire. Heat, that mortal enemy of electronic equipment, also adversely affects an amp's performance. Eventually, if you abuse the amp, its protective circuits should kick in — but some amps don't use a lot of protective circuitry and expect you to behave yourself, so if you abuse them, they'll simply blow up. Having put out power-amp fires onstage in midshow, we can tell you that this is not a fun addition to your light show.
Finally, keep in mind that there are several ways to rate power in an amplifier, including continuous power (the long-term average output power with typical program material), program power (maximum average levels over the medium term, typically up to a minute), and peak power (calculated for short-term peaks, usually about a tenth of a second). For P.A.s, you generally want the program-power rating, preferably measured in watts RMS.
When One Is Not Enough
When you put two speakers in a cabinet or wire two cabinets to the same power-amp channel, how they're wired affects the total impedance presented to the amplifier. For example, a cabinet housing two 8ž speakers can have a total rating of 4, 8, or 16ž, and wiring multiple cabinets or speakers can get a mite complicated. Let's take the fear out of dealing with a mishmash of speaker setups.
If you have two speakers you want to install into a cabinet, or you want to change the existing wiring, you can hook them up in series, parallel, or series-parallel. (The same rules apply when driving more than one speaker cabinet with one power-amp channel.) Fig. 1 shows how you connect the terminals of the speakers (or entire speaker cabinets) to produce series, parallel, or series-parallel configurations. Each scheme yields a different total impedance.
In a series setup, you merely add the impedances together. We know we said we'd keep the math to a minimum; this is easier than it looks. Let's assume we have an 8ž speaker and two 4ž speakers.
For a series setup, the equation is simple:
Z1 + Z2 + Z3 = Z
So in our example (Fig. 1a):
8ž + 4ž + 4ž = 16ž
Thus, if we wire the speakers in series, our power amp will be dealing with a 16ž system.
If we wire the speakers in parallel, and all three speakers have the same impedance, the formula is easy: the impedance of one speaker divided by the number of speakers. So if we had three 8ž speakers wired in parallel, the equation is 8ž ÷ 3 = 2.67ž. If our power amp can handle 2ž loads, that should work fine, but if the amp is looking for a 4ž load, this is going to make our amp work hard. With an amp designed for an 8ž load, this system is going to be bad news.
If the speakers are of varying impedances, things get more complicated. This looks scary but is easier than it appears:
In our three-speaker example (Fig. 1b), that gives us:
That will be ruinous for our amplifier; we can't do it.
Calculating the impedance for a combination of series and parallel wiring is just a matter of applying each equation as needed. In our example, we can do this two different ways.
First, let's wire one 4ž and one 8ž speaker in series, then wire that combination in parallel with the other 4ž speaker (Fig. 1c):
The series: 4ž + 8ž = 12ž
Wiring our 12ž system in parallel with our other 4ž speaker gives us:
Let's see what happens if we wire our two 4ž speakers in series, and then wire that combination in parallel with the 8ž speaker (Fig. 1d).
First, we calculate the combined value of the series speakers:
4ž + 4ž = 8ž
Now, let's put that 8ž system in parallel with our 8ž speaker, which is simple, since we have two systems with the same value:
8ž ÷ 2 = 4ž
That should work perfectly.
One final tip: if you're in doubt, wiring in series always results in a greater impedance, and while it might be less efficient, it's the safest way to go.
The best way to deliver highs and lows is through two or more separate speakers, each optimized for that particular range. A woofer, or midrange speaker, simply can't deliver high frequencies the way a smaller tweeter or horn dedicated to that purpose can, and a tweeter or horn is similarly ill equipped to handle the low frequencies that a woofer is accustomed to belching out.
The question is, how do you then direct only the appropriate frequencies to each speaker: lows to the woofer and highs to the tweeter? The answer lies in the crossover circuitry, of which there are two types: active and passive.
An active (powered) crossover is more complex and expensive than a passive system, but it's much more efficient and versatile. In an active crossover system, the different frequency ranges are isolated before the amplifier stage, and then each band is sent to a separate amplifier that only needs to reproduce frequencies in that range.
Because active crossovers use powered electronic circuitry, they lose less signal power and can incorporate more precise and sophisticated filters than passive crossovers can. Since each range is amplified separately, you control the power to each element and can put separate protective limiters for the highs, mids, and lows (in a triamped system). To top it off, you can easily control the crossover frequency via a knob, a feature that would require swapping out the electronic components in many passive systems.
In a passive (unpowered) crossover, the signal is split after the amplifier stage, as it's on its way to the speaker. Simple in-line components, an inductor and a capacitor, act as a low- and highpass filter, respectively.
Figure 2: In a typical two-way crossover, a capacitor (C) acts as a highpass filter and an inductor (L) as a lowpass filter.
In a two-way passive speaker system (woofer and tweeter or horn), current on its way to the woofer would flow through an inductor, which acts as a lowpass filter, allowing only the lows to reach the speaker. For the signal going to the tweeter, the current flows through a capacitor acting as a highpass filter, blocking lows and allowing only highs to pass (see Fig. 2).
In a three-way system — with separate drivers for high, mid, and low frequencies — the current on its way to the midrange speaker would flow through both an inductor and a capacitor, which would block lows and highs, allowing only the middle range to pass (that is, acting as a bandpass filter). A typical division in a three-way system would send the 30 to 800 Hz frequencies to the bass speaker; 800 Hz to 6 kHz to the midrange speaker, and 6 kHz to 20 kHz to the high-end driver.
Passive systems have the advantage of being simple and inexpensive, but they can present problematic loads for power amps at some frequencies, and they add some resistance to the circuit, so the system's sound quality can suffer, especially at higher power levels and at lower frequencies. That said, they can be an appropriate choice for smaller P.A. systems.
If you have to use separate speaker systems for three separate frequency bands, or you have a large P.A., use subwoofers, or for other reasons are using beefier power amps, go with an active crossover. If you just need a simple, two-way system and aren't using large amounts of power, a passive crossover will probably do the job.
Keeping Your Balance
Any electrical circuit contains at least two components: a signal and a ground. In an audio connection, this basic 2-conductor arrangement is termed unbalanced, and it's generally used for instrument cables, patch cords, and speaker cables.
Unfortunately, audio cables are subject to picking up noise in the form of hum from nearby power cables, radio broadcasts, and assorted electrical grunge that degrades your signal. If the signal level is hot enough, as with the cable from a power amp to a speaker, or the cable run is less than 20 or 25 feet, as with an instrument cable or patch cord, the noise will generally not be noticeable, and a regular unbalanced line will be fine.
But if you have a low-level signal or a long cable run, as with a microphone cable, the noise induced into an unbalanced line can be relatively loud, and when amplified along with the signal, it would be very noticeable indeed. Therefore, you need to get rid of the noise while retaining the original signal. That's when a balanced line comes to the rescue.
With a balanced audio connection, before the signal leaves the source, it is split into two paths. The polarity of one copy of the signal is inverted with respect to the original — that is, positive becomes negative and negative becomes positive. When the polarity is inverted, the signal is said to be 180 degrees out of phase with respect to the original signal. The connection is called balanced because it carries two opposite-polarity versions of the same signal.
The original and the inverted signal travel to the destination device on separate wires in a 3-conductor cable. (The third conductor is ground.) Along the way, both conductors pick up approximately the same noise. Since the noise didn't originate in the source device, its phase is the same in both conductors.
To summarize: in a balanced line, the first conductor carries the original signal and noise. The second conductor carries the inverted signal and practically identical noise. The third conductor is ground. Got it? Good!
Now, at the input of the destination device, the polarity of the already-inverted signal on conductor 2 is flipped back to its original state. But the noise in conductor 2 also gets flipped, so it is now inverted with respect to the noise in conductor 1.
The status at this point: path 1 carries the original signal and noise. Path 2 carries the original signal and out-of-phase noise.
Next, the two signals are summed. When you add together two identical signals with the same polarity, they reinforce each other. That's what happens with our original signal, which is good because it gives us a strong signal. In contrast, when you add together two identical signals with opposite polarities — as is now the case with the noise in our example — you are adding positive to negative, so they cancel each other out. Result: a strong signal and no (or significantly reduced) noise! The simple step of inverting the signal while it travels down the line and then reinverting it at the destination is the ingenious idea behind balanced lines.
Although other connectors are sometimes employed, balanced audio lines usually use XLR connectors like those found on low-impedance mics, or 3-conductor, ¼-inch phone connectors. The connection is wired with one signal going to the tip, one to the ring, and the ground to the sleeve of the phone plug, hence the term TRS. The TRS phone connector is the same connector used in stereo wiring (such as in headphones), but it carries the balanced signals instead of the left and right channels (and ground).
Phase and Polarity
Sound is made up of waves that, when allowed to interact in a reflective acoustic space, collide and combine in complex ways. When two waves of the same type meet at exactly the same time, their peaks and troughs align perfectly and the signal will be reinforced, or its amplitude increased. As we learned earlier, if an entire waveform is 180 degrees out of phase with respect to another, identical, waveform, they cancel completely, resulting in silence.
Figure 3: The top waveform (a) is a clarinet sample; the middle waveform (b) is the same sample but delayed 1.6 ms, putting it out of phase with respect to the original. When the original and out-of-phase versions mix, some frequencies are reinforced and some are attenuated. The resulting waveform (c) looks nothing like the original, and in this case, its amplitude is greatly reduced.
But if two waves are offset in time (say, because one is delayed through circuitry or by a reflection), a phase shift, or comb-filter effect, will result, where some frequencies are reinforced (when the values added together are both positive or both negative), and some are attenuated (when the plus and minus values are mixed), as shown in Fig. 3. When the fundamental is 180 degrees out of phase with respect to the other wave, significant cancellation occurs and the resultant sound often produces an extremely weak and hollow-sounding version of the original. Only your ears can tell you what's desirable and what needs correcting.
Therefore, whenever two sound waves from the same source interact, you need to be concerned with their phase.
Any time you have two mics on the same source, such as a close mic and an ambient mic on a guitar amp, you risk capturing the waves at different times in their cycle, because the different distances equate to a time displacement: the mic farther from the source will pick up the wave at a later point in its cycle. You could solve the problem by moving a mic backward or forward (changing the time factor of the phase shift) and listening to the results. But sometimes you don't have that luxury in terms of time and stage-plot mobility in a live situation. So there's another way: reverse the polarity of the signal using the “phase” or “polarity” switch on the appropriate mixer channel.
Polarity is not the same as phase, though they can produce similar results, and the terms are often used interchangeably. While phase has to do with the time relationship between identical waves or signals, polarity is concerned with the positive and negative voltage values of a signal.
If we start with two identical sine waves and we invert the polarity of one of them, the inverted sine wave will be exactly 180 degrees out of phase with respect to the original. This is the only case where the two different principles achieve the same result, but it can be a lifesaver when dealing with a phase-shift problem onstage.
If you have a signal that sounds weaker through two mics than it does through either mic by itself, you may be close to a 180-degree phase shift, because as we know, combining in-phase and 180-degree out-of-phase versions of the same signal results in the two canceling each other out. If that's so, then reversing the polarity of one mic's signal should fix the problem. Best of all, hitting a polarity-reverse switch is quick, easy, and changes nothing in the stage plot!
Here's another example of how you can use polarity to correct a phase problem. If you use two mics on a snare drum, one facing down on the upper head and the other facing up on the lower head, you risk having the signals weaken each other when they're combined at the mixer, because they are out of phase due to their position (facing each other). However, if you flip the polarity of the lower mic's signal with the channel's polarity-reverse switch, you'll produce a stronger signal.
Sound Under Pressure
Many of the concepts dealing with sound have scientific meaning and are expressed in scary-sounding terms like decibels and sound-pressure level. But you don't need to learn math to appreciate how the principles work. So while there are some scientific concepts here, we promise not to introduce any equations, and there will not be a pop quiz at the end of the session.
Let's tackle some of the most common terms and see how they can help when listening to your P.A. or guitar amp. By the time we're done, you'll know exactly what to do when the sound tech says, “Turn that guitar down by 20 dB!” (Okay, you already know the answer: turn the blasted thing down!)
Loudness, which is also known by the nontechy term volume, is a description of how humans experience sound-pressure levels (SPL), or acoustic amplitude. Mathematically describing audio phenomena such as sound-pressure levels (and our ears' response to them) involves a huge range of numbers; for instance, the ratio of the loudest sound we can endure to the softest we can perceive is approximately a trillion to one. If you were to plot changes like this on a graph using a linear scale, it would require a gigantic graph that would be very difficult to understand. Logarithms (which are essentially the opposite of exponents) reduce these large numbers to a manageable scale and make it easier to understand what is going on.
Bels are a logarithmic unit of measurement originally developed at Bell Labs to compare two power values. It turned out that Bels were too large for practically measuring audio circuits, so the scientists developed the decibel, a unit of measurement equal to one-tenth of a Bel.
There are several distinct types of decibels, and each uses a different formula. The decibels used to measure electrical values are not quite the same as those used to measure acoustic phenomena, although they are related. For our purposes, we are specifically dealing with the type of decibel used to measure acoustic sound-pressure levels, the symbol for which is dB SPL. Thanks to logarithms and decibels, rather than say that the level of a pianissimo flute line is two one-thousandths of a dyne per centimeter squared, we can say it's 20 dB SPL.
The table “Relative Sound Pressure Levels” shows the SPL of some common sounds. Notice that as SPL increases additively by 10 dB (for example, 10, 20, 30), the amplitude increases by a factor of 10 (10, 100, 1,000). So a signal that increases by 20 dB SPL is actually 100 times more powerful than the original.
Keep in mind that loudness is something else again, responding nonlinearly, depending on the pitch (frequency). Our ears cannot detect when extremely high or low sounds get louder or softer, as they are not sensitized as much to extreme frequencies. For example, if you increase a 5 Hz signal by 10 dB, you would perceive no change in volume because our eardrums are not sensitive to frequencies that low. However, if you raise the frequency to the range of a baby's cry or a tenor's high note in an aria — sounds whose fundamental frequencies fall in the middle of our hearing range — you'll hear the difference because our ears are very good at detecting subtle changes in amplitude in that frequency range. Obviously, that affects how we listen to music.
How can understanding this relationship of loudness help you in action? Well, if you turn up the highs in a system, it won't be perceived as louder, only brighter, to our midrange-oriented ears. And if you turn up the bass in a two-way P.A. speaker too much, you'll distort the signal before you can hear an appreciable boost in level. Even though you're turning up the bass, you'll hear the distortion first in critical information such as the vocals and acoustic guitar — midrange instruments to which our ears are highly sensitive. So though the bass might be boosted in terms of acoustical power, that will not necessarily translate to perceived loudness. For increased perceived loudness, crank those mids!
Now you know that when the tech says to turn the guitar down 20 dB, that means you need to bring it way down. Come on, pal, are you trying to damage my ears?
Source Decibels SPL Times the TOH (in SPL)
RELATIVE SOUND PRESSURE LEVELS
Note: since decibels are logarithms, zero is an admissible value.
In this case, zero represents the threshold of human hearing.
Threshold of hearing (TOH) 0 1× Rustling leaves 10 10× Whisper 20 100× Computer keyboard 30 1,000× Typewriter keyboard 40 10,000× Quiet voice 50 100,000× Normal conversation 60 1,000,000× Street traffic 70 10,000,000× Vacuum cleaner 80 100,000,000× Shouting 90 1,000,000,000× Walkman at maximum 100 10,000,000,000× Subway 110 100,000,000,000× Jet take-off 120 1,000,000,000,000× Gunshot at close range 140 1014× Instant eardrum damage 160 1016×
Jon Chappellis the author of several books on audio and recording, including his latest, Build Your Own PC Recording Studio (McGraw-Hill, 2003). Visitwww.jonchappell.comfor more information. Steve Ois the editor in chief of Onstage and Electronic Musician magazines.