Clipping,impedance, heat, sensitivity, blown drivers, and other annoying mysteries…
Thetopic of speaker power handling has been addressed several times here atAudioholics, but I wanted to make things a bit easier to understand. There wasa superb three-part series done in 2008 by Paul Appolino, and it is far moretechnical than I will or can ever get. The three pieces are brilliantlycomposed, and I suggest you read the whole series. This current article will bemore from the perspective and real-world experiences of a professional audioguy who for many decades has tried to save people from themselves.
If you own speakers andyou have ever…
Let's say you've invested in a new set of upgraded speakers and a sleek new stereo receiver unit. You're on your way to hearing your music in superior ways. However, your car may not be doing your custom audio system justice - and your radio's built-in amplifier might not supply enough power to the.
1)done a smoky burnout in a McDonald’s parking lot;
2)leaped into a backyard pool from a nearby garage roof;
3)stood directly in front of the P.A. stacks at a Weezer concert; or
4)yelled to friends, “Hey, y’all, hold my beer and watch this!”
…youshould probably read this article. You’re the guy most likely to blow up yourspeakers, and it’s totally your fault.
“How many watts will my speakershold??”
Ican’t count the number of times I was asked this question by myoverly-enthusiastic customers when I was an audio salesperson. It seemed as ifevery decibel zealot wanted assurance from me, their Audio Maven, that no matter what music they played at whateverlevel, their amplifier would nevershut down and they would never blow atweeter. And I could sense their consternation when I couldn’t give them ananswer; or, at least, an answer to their liking. The manufacturer of thespeaker might rate the speaker at “150 watts” and I had to explain that damagecould still be done with a 100-watt receiver, or a 50-watt receiver, or a 350watt-per-channel separate amplifier. If you set your mind to it, you can andwill damage any speaker. If your speakers are rated at 100 watts and yourreceiver is rated at 100 watts per channel, you’re still not entirely safe.
Mostof you know the basics of how a speaker works, but here’s a simple explanation.Conventional dynamic drivers (woofers, midrange, tweeters; cones and domes)have a magnet structure, voice coil, diaphragm, basket, and suspensioncomponents. The voice coil consists of a tubular “former” wrapped with thincopper (or less often, aluminum) wire. When an electrical current (the programsignal from the amplifier) is applied to the voice coil, a magnetic field iscreated, making a variable electromagnet. The voice coil moves up and back inthe voice coil gap, moving the diaphragm, which is attached by suspensionpieces (spider and surround), propagating sound waves. Bass frequencies requiremore cone area and movement (excursion.) Tweeters require smaller magnets andvoice coils and smaller diaphragms to reproduce higher frequencies. A crossovernetwork is employed to direct the appropriate frequencies to the drivers bestsuited to handle those frequencies, and to filter out frequencies that don’tbelong in that driver.
“Turn those suckers up to 11 and ripoff the knob!!”
Theproblems start when we expect an amplifier or speaker to perform beyond itscapabilities. To play music louder, more movement is required of the suspendedparts of the drivers; the suspension, voice coil, and diaphragm. The harder thespeaker is played, the more heat is generated within the voice coil, and, atsome point, you will surpass the voice coil’s ability to travel and/or dissipatethat heat. Failures of drivers can be from a physical distortion of the voicecoil (going out of round from heat) which will result in driver failure, or insome cases even the voice coil wire acting like a fuse and opening from heat.In any event…no tunes, man. Bummer.
Powerhandling of a loudspeaker system depends upon many things. First, what is theduration of the particular sound? Music is a series of “peaks,” and while youmay be listening using an average of 1 to 10 watts, peaks can demand ten ortwenty times that much. And a speaker that can handle a 300-watt peak (at a veryshort duration) at a certain frequency might not be able to handle more than 50watts for a duration of ten seconds at the same frequency. A note that is at afrequency extreme of the driver’s range may not be handled with as much graceas a note in the center of the driver’s bandwidth. A 12” bass driver may handlea 200 Hz note just fine, but it might struggle with a 25 Hz tone. If you plottedthe power handling of a loudspeaker from 20 Hz to 20 kHz, the power handlingwould vary a great deal by frequency.
Tweeterscan handle far less power, and that’s usually okay because there’s far lessenergy in their useable range. Looking at tweeter specs as I often do (I haveno life, apparently), I’ve found that even the manufacturers’ power ratingsdon’t tell the whole story. A certain $3,250 Seas tweeter will handle 150 wattsshort term, and 55 watts long term, according to Seas. My own Scan-SpeakRevelator tweeters, used in my dedicated theater, used to have an IEC 17.3 maxpower rating of a very robust (and very suspect) 160 watts RMS. I notice thatrecently Scan-Speak has omitted that specification, so, as you can see, I am asmuch in the dark as you are. And a typical tweeter used in a $500-each speakerwill employ a driver that has a power rating of maybe 25 watts, and that shouldbe cause for alarm. Just not a very loud alarm.
Thereare way too many variables to quote reliable power handling numbers, and ageneralization won’t do much good in the real world. As a sales and marketingguy who was also on product development teams, I was often faced with how torate a loudspeaker’s power handling. Some manufacturers of lower-end speakerstend to inflate this spec by blurting out an arbitrary “1,000 watts,” when thespeaker would be hard pressed to handle 50 continuous watts. While working forTriad Speakers, I had proposed that rather than to rate our speakers at howmuch power they would handle before they failed; an exercise in futility;working with the engineers, we decided to forego power handling specs and list recommended amplifier power for each ofour 100+ models. Basically, we were inferring “Use an amplifier rated at 50-200watts per channel into 8 ohms, but don’t do anything stupid.”
There are limits to everything.
Thinkof powering a speaker the way you might think about riding a bicycle. GregLeMond, three-time Tour de France winner between 1986 and 1990, has stated thata professional cyclist can only sustain a flat-out effort for 45 seconds,possibly generating up to 1,500 watts. Typically, this would occur in a sprint.A stage of the Tour de France might take more than six hours, with an average output of between 200-300 watts.As with a loudspeaker and amplifier, output can go higher as the duration ofthe effort gets shorter. With a cyclist, the anaerobic threshold is wheredemand for oxygen exceeds what the athlete can take in and process, and lacticacid accumulates in the muscles faster than it can be dissipated. In the caseof a speaker, the voice coil excursion exceeds its limits (x-max) and/or heatin the voice coil cannot dissipate fast enough. (X-max generally refers to themaximum distance a voice coil can move linearly in one direction from rest,expressed in millimeters.)
Compressionalso occurs as a driver is over-driven. As the voice coil becomes hotter, itsresistance increases and the driver’s sensitivity is reduced, exacerbating theproblem. The harder you drive a speaker that’s already compressing, the lessresponsive it becomes. Compression at lower levels simply dulls the peaks inthe music and may be inaudible, but severe compression makes the music harsh,not dynamic, and high in distortion.
“If it’s too loud, you’re too old!”
Youmay have heard that too little power will blow a tweeter faster than too muchpower, and that’s not exactly true. Usually when a loudspeaker is damaged, it’sfrom too much power for too long a duration, or playing it excessively loud foreven a short period. What is true about the “too little power” statement isthat when a receiver or amplifier is driven into hard clipping, the resultingdistortion has more nasty high frequency content making it easier to damage atweeter. The term “clipping” comes from the visualization of what the signallooks like on an oscilloscope. The top of the waveform looks flattened,compressed, or “clipped off.” I’ve heard the term used with speakers when theyare overdriven to the point where they start to compress transients, either inthe drivers or by crossover saturation, but I prefer to use clipping only as anamplifier term. And while I recommend more amplifier power with low-impedancecapability, over-driving a speaker is still the fastest way to damage it. Factoringin yet another variable, some amplifiers clip gracefully, either by inherentdesign or additional circuitry. This is called “soft clipping.” Other amplifieroutput stages don’t behave as well at clipping, and they produce a cornucopiaof odd-order harmonics that can cause tweeter voice coils to overheat.
Mostof today’s receivers have their output power rated at 8 ohms, and they havedifficulty at lower impedances, due to intentional current-limiting. Theproblem is few speakers are a benign 8 ohm load. Most 8-ohm speakers dip to 4-6ohms at some frequencies and at system resonance, they can measure as high as20 ohms or more. Not only don’t you know how much power your speaker can handle,you don’t know how much power your receiver can deliver at that frequency! A good solid state amplifierwill deliver varying power depending on the frequency and magnitude of theimpedance the speaker presents it. But, the amplifier output should remain aconstant voltage source until it current limits at low impedance and highoutput levels. (Tube amplifiersdon’t react the same way, but that’s a different discussion.)
“Crank up the drums, crank up thebass, crank up my Les Paul in your face!”
Okay,it gets worse, and I’ve mentioned this in other articles. Do you useequalization in your system? Keep in mind that when you increase a frequency by3 dB (mildly perceptible increase), you doublethe power you use at that frequency, and you’re more likely to drive youramplifier or speaker into distortion. And equalization tends to be applied atthe frequency extremes; right where it can do the most damage. For example, ifyou boost 12 dB at 10 kHz on your EQ, you are now using sixteen times the powerat 10 kHz. If peaks were 25 watts before adding equalization, with a 12 dBboost, peaks are now demanding 400 watts. That dog don’t hunt. And simple bassand treble controls on receivers can allow you to boost as much as 18 dB, whichis also a recipe for disaster. Tone controls should not be used the way you useketchup at 2:00 AM on a Sunday morning.
So,how do you avoid damaging your speakers? It comes down to common sense. Don’tboost EQ much at all. Use an equalizer to cutpeaks. Don’t listen at abusively high levels. Sit closer to your speakers, ifyou can. Every time you “half” the distance from your speakers, you gainapproximately 6 dB of headroom. If you have a receiver, use a separateamplifier of higher power and better low-impedance capability. If output is themost important thing to you, consider very efficient speakers. If you play yourspeakers outside, you won’t have much output or bass, and turning up the basswill only hasten the demise of your speakers. You can’t play bass-heavy musicas loudly. You can’t play music that’s mastered hotter as loudly. If you’relistening at a high level and you turn it up higher and the sound doesn’t getlouder; you’ve exceeded your system’s limits. Either the amp or speakers, orboth, are compressing and distorting. And know that if you blow a center channeltweeter, it was because the center channel got the brunt of the power during amovie. If you notice any kind of compression or distortion at all, turn it downa notch or two. One safeguard is to calibrate your system to THX standards, andthen don’t exceed the reference level…by much.
Conclusion
Somespeakers have higher sensitivity and more robust drivers, but in the end, anyspeaker can be damaged if you apply yourself to the task. I could tell you Ihave never blown a speaker, but I would be lying to you. If you’ve chosen theappropriate speakers for how you listen, and the proper amplification fordriving them (and you’re not a maniac), you should be okay. Have fun, but becareful, kids, and always remember that when you blow your speakers...it’stotally your fault.