Several people have asked me recently what power ratings listed on speakers really mean and where they come from, so I thought I'd post it here. By the way, I currently work for a sound engineering company (that I am not supposed to name) most well known for certifying system performance and I have been in the audio/video industry for more than 30 years.
Commercial loudspeakers these days list two power ratings. These are usually given in watts, since that is the most common way amplifiers are rated. Power handling is actually a measure of the thermal characteristics of the speaker system, and really has nothing to do with how loud it will play. The two ratings are called "Peak" and "RMS."
The "Peak" rating is determined by the physical size of the electrical wiring in the voice coil. This is the number of watts at the speaker’s voice coil DC resistance rating that would produce the maximum amount of current the windings are capable of passing. Peak numbers are dependant on the physical size of the wiring and the amount of time the current is present. The AES Peak test is done with a 0.05 second (5 milliseconds) duration pulse of DC power. This corresponds to a square wave signal pulse at 20Hz. The peak rating is the maximum amount of energy the wiring can handle before it fails (i.e.: “smokes.”)
“RMS” in very general laymen’s terms corresponds to “average” in electrical theory. The RMS rating of a loudspeaker is determined mainly by the heat dissipation capability of the loudspeaker’s voice coil system and by how much and how freely the system can move. Better designed loudspeakers tend to have higher RMS ratings than more poorly designed ones that have the same peak ratings because they have better internal air flow. Long-throw loudspeakers tend to have higher RMS ratings because they can move more. It is the voice coil’s movement back and forth that creates cooling in the system and allows the heat generated by the current passing through the windings to be carried away from the windings. Exceeding the RMS rating can cause the windings to heat up too much and separate, mechanically bind, or simply fall apart (i.e.: “smokes.”)
The AES/IEC RMS power handling test is somewhat deceptive as it is performed at single frequencies using tone generators, and not with full musical energy (pink noise.) Some loudspeakers are more thermally efficient at certain frequencies than at others. A manufacturer trying to over-rate a speaker in order to sell it to an unsuspecting customer might list the RMS value at the single best performing frequency (and you will never know that) but when the speaker is powered at that rating with full musical energy, it will blow. Or they might just make the numbers up out of thin air. Good manufacturers (those with reputations to protect) will list RMS power as an average over several test frequencies, giving a better real-life picture of the speaker’s thermal capacity.
In general, loudspeakers with higher power handling ratings tend to be less efficient, since they must use larger gage voice coil windings to handle the current. Larger gage means less movement at a given current flow. Efficiency is usually listed as “sensitivity” in a rating such as db/watt. Higher sensitivity numbers mean the loudspeaker will always “play louder” at a given power input.
I hope this helps. Next I’ll try to tackle my favorite audio myth, the notion that under-powering a loudspeaker is more dangerous than over-powering it.