In Search Of The Audio Abode---The Hi-fi House Letters part 3

Harmonics & octaves
Editor: Contrary to Mr. Holt's assertion (in "In Search of the Audio Abode" in the April 1990 issue) that harmonics occur at one-octave intervals, harmonics occur at integer multiples of the fundamental tone. Therefore, the second and fourth harmonics will occur at one and two octaves above the fundamental, but the third harmonic will occur at an octave and a fifth above the fundamental. I'm sure this was just an oversight.---D. Andrew Austin, Philadelphia, PA

Mr. Austin is, of course, correct. Regarding Mr. Koval's letter, a paper presented by David Clark at the 8th AES International Conference in May ("Stereo in Automobiles") reinforced the idea that at low frequencies, when the room dimensions become smaller than half the wavelengths of the sound, the room starts to act as though it were directly pressurized by the cone motion, allowing deep bass to be developed. However, it must be pointed out that subjectively this is often obscured by the booming sound from the often strong and always sparsely spaced axial room modes present immediately above this region in a typically sized domestic room. J. Gordon Holt himself responds to these and other letters concerning his article on good listening rooms as follows:---JA

Several readers have had the gall to question some of the things I said in my article about listening rooms ("In Search of the Audio Abode," Vol.13 No.4). Even though JA tells me I don't have to respond, I'm going to anyway.

The loudest and most gleeful squeals came from people who insisted I had the wrong decibel numbers for my attenuation-with-distance chart. Even JA agreed with them. They were right. I knew, of course, that the strength of a soundwave diminishes by one quarter every time the distance from the source is doubled, but blithely assumed that, since sound pressure is used to measure subjective loudness, it is sound pressure which diminishes by one quarter. It isn't; it is sound intensity, or power. Sound pressure is proportional to the square root of intensity, just as voltage is proportional to the square root of power. Thus, if the intensity diminishes by a factor of four, the pressure will diminish by a factor of two. So the chart was wrong. The corrected one, which is even worse news than the original for people who insist on listening with windows open, was shown in the July issue, on p.34. I suggest, now, that you pull out your copy of the original article, cross out the chart at the top of p.87, and write next to it "See Vol.13 No.7, p.34." All other attenuation figures in the article, including that for attenuation due to glass, are correct.

Some correspondents told me my estimated price (of "under $1000) for retrofitting air conditioning to a forced-air heating system was way out of line, citing "a minimum of $2500" as a more reasonable figure. So much for inflation.

A number of readers whined about the rigidity of my dimensioning recommendations, and asked how much difference a few feet of discrepancy here and there would make. The answer: as much difference as the discrepancy. No one, least of all me, would tell you you can't listen in a cubic room if that's all you have available. It'll just sound like doggy-doo.

The same goes for the questions about flexible walls. The more you have, the more you'll lose deep bass. The system can still sound very good, but not as good as it would if you, not your neighbor, were bathing in those glorious 25Hz fundamentals.

Yes, I've read the AES paper Mr. Koval mentioned, and while I am not prepared to say it is nonsense, I can only say that my own experience has been that small rooms make for very poor deep-bass response. The major problem, as JA said, is the severity of higher-bass-range standing waves, which are only partly controllable by such as ASC Tube Traps. These tend to mask lower-frequency signal which is of lesser amplitude, even though it may, in fact, be rising above virtual zilch. Anyone who has ever compared the low-end range of, say, Altec A-7 loudspeakers in the home with that in a large auditorium (like a movie theater) will know what I'm talking about. In a theater, their low end (such as it is) is full and well balanced with the upper range; in a typical living room, it is absurdly thin.

I can easily understand why automobiles might be a different story altogether, because their acoustical space is so irregular and so broken up by the presense of seats and people in that space as to minimize standing-wave activity. In fact, I've heard bass range and detail from average car radios that I did not think possible from loudspeakers that size.

I agree with Mr. Koval that equalization can do more good than harm to a system's sound, but few equalizers have sufficient resolution (that is, narrow-enough controllable bands) to correct an acoustical peak without worsening an adjacent dip. The problem, usually, is that severe resonances tend to be about 1/6-octave wide or less, while equalizers can rarely control less than 1/3 of an octave, and the spectrum analyzers used to measure what's going on---also typically 1/3-octave---cannot show the damage that is done to the overall response when a single 1/3-active band is set for average flat response relative to the rest of the audio range.

Some people asked what kind of room I finally ended up with. I didn't. I gave up, and am now looking for a house where I can build an adjoining listening room that suits my needs. That doesn't mean you won't have better luck than I had.---J. Gordon Holt

Share | |
Site Map / Direct Links