Power Lines: “There WILL Be A Test!”
Gauging your knowledge of power as it relates to sound systems

By Pat Brown

With the conclusion of the three-part series on power in the previous issue (April 2003), it’s time to check your knowledge power, or actually, your knowledge about power. Rather than submit you to the typical “right or wrong” questions with exact numerical answers, I’ve elected to provide a different means of self-evaluation. The test is “open book”, based upon the information shared in the article series and other resources, and I can tell you up front that the answer to every question is “It depends!” But what’s really being asked is “WHAT does it depend on?”

At first glance, it may seem that the question is not even related to audio. Don’t be fooled. Principles are principles. You already know this stuff - you just may not know that you know it! Each question also paves the way for a short review of the concept. That’s it. Relax. Take out a sharpened number 2 lead pencil and... Sorry about that.

1) I want to paint my living room walls and need to buy paint. How much will I need if the ceiling height is eight feet?

Obviously, estimating of the amount of paint requires more information. What we need to know is the area to be covered, which can’t be determined by the ceiling height alone. The total length of the walls is needed to get the area (length x height). The paint store would also need to know how many windows are in the walls (they can subtract this area from the total), and how absorbent the surface is (one or two coats?). Only then can the required amount of paint be determined.

It’s equally ridiculous to calculate an amplifier’s output power by using its peak voltage rating. As with the wall, the area of a waveform must be known to determine how much power is generated. This requires amplitude information (like ceiling height) and knowledge of length (time). We also need to know how much to subtract for higher crest factors (less intense program - like windows in the wall). And lastly, we need to know how much the load will soak up (porosity of the surface). Think of one coat as eight ohms and two coats as four ohms. And two ohms? Don’t even think about it!

2) Which stock will yield the greatest earnings?

We’ve all learned this one the hard way. Stock A has some high amplitude values, but doesn’t last long. Stock B has lower “highs” but is more consistent over time. Like painting walls and electrical waveforms, it’s all about area. An amplifier can have a very high peak rating, but may fizzle when loaded for long spans of time (that all-day outdoor show). Make sure that you look at the long-term continuous output power when shopping for amplifiers. Short-term peak ratings are large numbers, but they don’t tell the whole story.

3) Which song will make the loudspeaker hotter?

This should be obvious by now. Grungy, highly compressed rock and roll has a much lower crest factor (more area) than an “audiophile” recording of a sitar solo. Both types of music may occasionally light the clip light, but the R&R is much more likely to toast the loudspeaker.

4) How much must I increase the power applied to a loudspeaker to make it a little louder?

A bunch - 3 dB represents a modest change in sound level, yet a 3 dB increase requires the amplifier to generate twice the power. So every time you turn it up “a little”, you are doubling the power to the loudspeaker. No wonder so many loudspeakers succumb to the last song of the evening.

5) How much of an amplifier’s rated power will the amplifier likely have to generate in a music playback system?

Not much. Given a typical crest factor of 20 dB for live music, the amplifier’s output power could, on average, be about one watt per 100 watts of rated power. That kilowatt monster that you bought with the home improvement loan will likely need to generate about 10 watts continuous. If you break out the compressor/limiter, you may get this up to 100 watts, but that’s about it.

6) So why buy a big amplifier? After all, they’re expensive!

Loudness and generated power are all about area. Clarity is all about headroom. If program peaks get clipped by a small amplifier, it sounds like trash. Amplifiers must be oversized relative to their average output power by a factor of 10 to 100 to allow for signal peaks. This translates into 10 to 20 dB of headroom. If you have deep pockets and an understanding wife, go for 20 dB. If not, spend some of the money saved on a hard limiter to make the program peaks “fit” though the amplifier. All of this makes sense only if you look at power using the decibel. (Hint: refer specifically to part one of the series in the January 2003 issue.)

7) Which subwoofer is better? One that handles 100 watts or 500 watts? You simply can’t tell from the power rating alone. It’s just a “waste disposal” number. How much sound can each produce? This is the efficiency rating. A 15-inch bass horn sitting in a corner and consuming 100 watts continuous could easily be much louder than an 18-inch in a sealed box hanging in free space and consuming 500 watts continuous. It’s not what’s fed in, it’s what comes out. See the stock market question (Question 2, above) for an object lesson on this. 8) Can a 30-pound amplifier really keep up with a 300-pound amplifier? My chiropractor wants to know. Maybe. It depends on what it’s asked to do. Power (like hot water) can be generated “on demand” or it can be pulled from storage. This is why “Anywhere Gas and Electric Company” dams up rivers to create huge reservoirs for turning hydraulic turbines. The water flow can remain constant even through the dry season so that the lights don’t dim when the creek gets dry.

Amplifiers with large, heavy power supplies can typically maintain a more constant current flow under severe conditions - like reproducing low frequency synth tones through subwoofers that can peel paint (see question 1) at 100 feet. Percussive sounds at mid/high frequencies aren’t nearly as “meaty”. You can save your back and your wallet with smaller, lighter amplifiers.

9) Can the loudspeaker’s power rating be trusted? The “Killbox 5000” is rated at 5 kW and the “Lighttones 100” is only rated at 100 watts. Most reputable manufacturers use standardized power testing to rate their loudspeakers. These methods define the type of waveform, time duration, crest factor and a few other metrics. These tests are designed by experts to simulate real-world demands on loudspeakers. They tell you how long the loudspeaker survived under the described conditions. Since they can’t possibly know how the end user will use (or abuse) the product, they can’t guarantee that you won’t blow it up. Power ratings are very useful for comparing the relative differences between products from the same manufacturer. When using them to compare one brand to another, make sure that the ratings are based on the same standard (i.e. AES, EIA-426B, etc.). And even when the standards are used, the “honor system” governs the writing of spec sheets. Since overrated loudspeakers generally don’t kill or maim people, the government doesn’t require validation of power ratings. Only in “Auditopia” are there “data police” that check all the ratings. You can pinch yourself to wake up now. Also remember that large power ratings can be achieved with resistors (which don’t produce much sound!). That “Killbox 5000” might have a couple of water-heater elements as part of its crossover network.

10) What happens if my power amplifier is too large?

It’s better to have an oversized amplifier than an undersized one, provided that one stays within the thermal limits of the loudspeaker. A bigger amplifier is less likely to clip the signal. But there are limits here, too. A piston can only travel so far before it becomes non-linear. If you hook a zillion watts up to a loudspeaker to provide 40 dB of headroom, and then someone drops a mic, you may see impressions of all of the cones in the metal grills. Over-excursion kills fewer loudspeakers than heat, but it must still be considered.

11) Should I use a rubber band or a chain to pull my boat trailer?

I had to get one in regarding reactance. There’s a big difference between the impedance of a real loudspeaker and that of the non-inductive load resistors used to test many amplifiers. Reactive loads reflect power, and the amplifier has to deal with this. Amplifiers with extended bandwidths are often unstable into reactive loads. If presented the choice between a “20 Hz to 40 kHz” bandwidth spec and a “DC to gamma rays” spec, I would pick the former.

TO THE POINT

How did you do? Hopefully this series has provoked some thought. Power ratings are useful for getting a general idea of the performance of a device, but as I have shown, there are many variables and caveats when you assign numerical ratings to amplifiers and loudspeakers. Don’t give these numbers any more or less attention than they deserve. They are but one piece of the puzzle, and probably not the most important piece.

Also, don’t be afraid to do your own power testing. The A/B comparison of two products is still the best way to tell the difference between them, regardless of what the numbers say. ZZ Top at full volume might be a more relevant power test for your system than pink noise. Just be sure going in which party is paying for the toasted voice coils.

Pat Brown, with his wife, Brenda, heads up Syn-Aud-Con, leading audio training sessions around the world. For more info, go to www.synaudcon.com

May 2003 Live Sound International