Phase Distortion and the Conventional Loudspeaker
A great many problems in music reproductions can be traced directly to a basic flaw which appears as a necessary characteristic of the conventional loudspeaker - Phase Distortion. Of all the speaker ailments, phase distortion is the most critical, the hardest to cure, and probably the least understood by both stereo enthusiasts, and the average consumer alike.
Using easy-to-understand models, and non-technical terms, this pamphlet will enable you to understand phase distortion and the critical role it plays in the reproduction of music through loudspeakers.
To understand phase distortion, one must first understand the meaning of the words Frequency and Cycles. The concept is easily grasped if we form a mental picture of a person sitting at the edge of a quiet pond, making small waves by striking the water with his hand.
The waves, both large and small, depending on the force exerted, travel across the pond at the same speed, so that the distance from crest to crest remains unchanged once they are set in motion. This crest-to-crest (or trough-to-trough) distance is called wavelength and can be changed by striking the water at a faster or slower rate.
Let us assume that another person is sitting on the other side of the pond counting the number of complete waves (one crest to the next) as they pass a given point in a certain length of time. For example, if the wavelength is long, then not many waves will pass the point in, letís say, a minute. If the wavelength is short, then a large number of waves will be counted. In science a complete wave is called a Cycle, and the standard length of time is the second, which would give us Cycles for every one second, or, simply stated, Cycles per second. The number of cycles per second is referred to as Frequency.
Frequencies with a long wavelength we hear as low tones, those with a short wavelength, as high tones. We do not however, hear all tones. The average human may hear frequencies that range from about 30 cycles/second to 16,000 cycles/second. Perhaps you have had experience with a silent dog whistle. Silent that is, for humans, but not for dogs, whose hearing is attuned to much higher frequencies.
When a musical instrument generates a tone, the tone is composed of the fundamental note and whole family of harmonics. As an example, if a bass note is struck on a piano, a fundamental frequency of 73.42 cycles is produced, along with harmonics or overtones that may range up to 7,000 cycles, all of which are radiating from the same impact point on the string, at the same instant of time.
The interweaving or wave pattern of this family of vibrations creates the tone quality or Timbre of the instrument. It is, in fact, the reason why each guitar, even though made by the same company, has its own unique sound. However, only a single point source can reproduce this slight difference.
To visualize why a single point source is important, picture a large stone and small stone dropped into a pond at the same time and in the same place. The subsequent pattern of large and small waves emanating from this common origin can be directly related to instrument timbre. We say that the waves are in-phase.
On the other hand, if the two stones are dropped at the same time but at different points in the pond, the intermixing of these waves (occurring somewhere in the middle) merely resembles our original example; they do not copy it. They are out-of-phase.
It should seem apparent that in order to reproduce accurately the music on a record and have it in-phase, we need a single source, capable of reacting to all frequencies. In other words, we have to drop our stones at the same time, in the same spot.
Now, a vibrating diaphragm of any size can produce low frequency tones. For example, the diaphragm in a microphone is 1 inch or less while the usual headphone diaphragm is 1/2 to 2 inches. Cones (or drivers) of this size, while large enough to create bass in your ear canal, are too small to move enough air to create bass in a room.
In a conventional speaker the bass is produced by using large massive drivers called woofers. However, their very mass makes them sluggish and slow and the problem then becomes one of producing the high frequencies. This is solved by using a small driver (or a series of smaller drivers) called a tweeter. Typically, mid-range drivers are also used for those frequencies which fall between.
This multiple driver system requires an electronic circuit called a crossover to be inserted in the system to direct the appropriate signals to the various drive units. The end result is that although the conventional loudspeaker produces all frequencies, it is using many sources of propagation, which causes the music to be out of phase and time alignment with the subsequent loss of instrument timbre.
Unfortunately, the loss of instrument timbre is only one of the unpleasant side effects of phase distortion. Another is loss of clarity. Music from the conventional loudspeaker most often seems to be heavy, muffled, and coming from the inside of the speaker box. The listener may feel as if he was listening to a performance through a thick blanket. This inevitably leads to a problem in imaging and presence, as they are known in the audio business.
Imaging is the mental picture one should get of the performers and their stage position. Presence refers to how much a speaker sounds out-of-the-box-and-into-the-room. The more presence a speaker has, the easier imaging becomes.
To understand the final and most damaging side effect of phase distortion, we must return to our pond. This time, let us assume that there is one person on either side of the pond, and that both are making identical size waves. As the waves meet in the middle, the water at any given time will be both choppy and calm in the same area.
When the crests of two identical waves meet their effects become additive and the resulting wave crest is twice as high as either of the originals. For example, the crests of two 2-inch waves would join to form a four-inch wave. The same result occurs at the meeting of two troughs. These cases are referred to as constructive interference.
Logically then, if a 2 inch crest meets a 2 inch trough, their effects are cancelled and the water is smooth. This is termed destructive interference. By the same token, we would expect to find intermediate size waves in areas where the interference was neither completely destructive nor completely constructive.
With this in mind, it is easy to understand how in a conventional speaker with many sources of propagation, some frequencies may be totally or partially cancelled causing a mute, or, in the extreme case, complete absence of musical instruments.
In theory then it would seem reasonable to build a speaker that uses a single source for all the frequencies thereby eliminating the need for woofers, tweeters, crossovers and the attendant problems of phase distortion, time alignment, clarity, sluggishness, etc.
At BSL, this was accomplished with our proprietary Dynamic Loading System (DLS) whereby a single driver is used, and the cabinet becomes the generator of bass frequencies in the same way that the body of the bass violin creates a large amount of bass from so small an object as a vibrating string.
A further advantage of this system lies in the quickness of the bass response since the woofer in this case is small (6.25 in.) and easily controlled by the amplifier.
Copyright ?1998-2007 Brentworth Sound Lab
Last modified: January 13, 2007