CONSTRUCTIVE COMMENTARY
I was pleased to see that Audio has addressed the topic of sound transmission between dwellings with "Muffling the Neighbors" by F. Alton Everest (November 1990) and "Good Walls Make Good Neighbors" by Peter Jurew (December 1990). When not preparing reviews and articles for Audio, I am in private practice as an acoustical consultant. Part of this work involves preparing acoustical studies for proposed residential developments; much of it relates to outdoor noise being transmitted indoors.
Sometimes I am asked to review the drawings for a multi-unit apartment complex or hotel/motel and make suggestions so that the Sound Transmission Class (STC) values will conform to certain criteria.
Occasionally I receive calls from homeowners about excessive indoor noise. Many of these calls relate to outdoor noise such as traffic or commercial/industrial sources. Other calls are typically from more distraught homeowners who, like the people in the Jurew article, have moved into an apartment and discovered that they hear too much sound from their neighbors. It is difficult to help these individuals because the costs of acoustical testing and consulting, plus the construction to effect improvements, are more than they can afford. Now, at last, I can refer the people who call me to the two excellent Audio articles, with hope that they can solve the problems themselves.
A significant number of inquiries re late to what I call "unusual perception." In nearly 20 years of consulting, I have only defined two categories of unusual perception. In the first category, I have investigated two cases where the person was disturbed by tonal noise in the region from 60 to 80 Hz. I amplified the noise in the building and had them listen via headphones and tune in the disturbing noise through a tunable one-third octave filter. The measured SPL of the noise indicated that the clients' threshold of hearing at 60 to 80 Hz was only 20 to 30 dB below the normal value for a young person, and these clients were about 60 years old! (I couldn't hear the noise without amplification.) I am sure that these people would have been highly disturbed by the bass sound of contemporary music coming through their walls! (The sources of sound in these cases were outdoors.) The second category involves perception of sound from non-acoustical stimuli; I have investigated only one of these reports. A woman started hearing tonal-type noises in the region from 500 to 2,000 Hz (identified by comparing a tone from an audio oscillator) just after her community was wired for cable TV. The noise levels in her home were very low, as it was a country town with little traffic at night and no businesses open nearby. The perceived sound did not change when she wore ear protection, and her audiologist said that she did not have tinnitus (ringing in the ears). My investigation ended without finding the source. Another woman indicated hearing high-pitched audio at times when a nearby military installation was testing a high-powered radar, but an investigation was impossible because the military operations were classified. It is difficult to conjecture about the possibilities of such a person hearing her neighbor's audio system via a non-acoustical path.
It is interesting to note that, in today's audio systems, we are pumping high currents via 12-gauge wire into very low-impedance, and perhaps inefficient, speakers.
These examples indicate that perception variations among different people may greatly exceed the variations in the sound transmission paths.
These variations could confound any efforts to retard the passage of sound.
The Everest article is comprehensive and accurate. I have just a few points to add. I think that it may be instructive to note some of the numerical criteria that we commonly work with in the acoustics of dwellings. First, and most important, is the actual noise level indoors: The Department of Housing and Urban Development (HUD) recommends a maximum day/night average sound level (Ldn) of 45 dBA. This is a bit complicated, but suffice it to say that if the level is 45 dBA from 7 a.m. to 10 p.m., and 35 dBA from 10 p.m. to 7 a.m., the Ldn will be 45 dBA. The FNMA specifies an STC of 45, minimum, for partitions between living units, and 50 from public spaces to living units. The FNMA minimum STC values for floors between individual living units and between living units and public spaces are 45 and 50, respectively. The Impact Isolation Class (IIC) values for floors between these spaces are also 45 and 50, respectively. California State regulations for multi-family dwellings require a minimum STC of 50 for party walls and floors. The California Office of Noise Control published a 500-page book of acoustical test reports on many wall and floor structures in 1981. (The book is, regrettably, out of print now). This remains the best reference to estimating wall/floor STC, and determining what to do to improve the STC. These, and many lab test reports on floors that are available from manufacturers of building materials, do include the STC rating plus the Transmission Loss (TL) values at standard third-octave frequencies.
Contrary to what Everest indicates, both STC and IIC ratings are generally measured by acoustical laboratories for floor/ceilings and are stated in their reports. The STC rating relates to reduction of airborne noise, typically speech, and the IIC relates to reduction of footfall noise, which is structure borne sound. The IIC is most easily improved by a thick carpet and pad, but this will not improve the floor's STC. However, there are many products available, along with tested designs for floor assemblies, which have high STC and IIC values. Some of these may be retrofitted to finished dwellings.
The STC rating is centered about the performance of the partition in the 500 Hz region and is useful for predicting the attenuation of speech. When I am working with rock music as the source, I disregard STC in favor of the Transmission Loss value at 125 Hz. When the source is mechanical equipment, such as air conditioners, I use the TL value at 250 Hz for calculations. A workable rule of thumb, therefore, is to measure the A-weighted sound level (dBA) in the source room and subtract the appropriate TL value to estimate the (dBA) level in the receiving room. I note that the United States Gypsum Co. specifies a Mechanical Transmission Class (MTC) value for many partitions. This MTC number is centered on 250-Hz performance.
Although the Noise Criteria (NC) curves in the Everest article are all correct, they will be difficult for the reader to work with, as the reader is probably limited to measurements with a handheld sound level meter, which measures A-weighted decibels. To work with the NC criteria, an octave-band analyzer is needed. That is why I gave the estimating procedure above.
Although I hate articles with math in them, there is one simple relation that may lend an understanding of Everest's graphs of sound transmission loss of walls. It is called the Mass Law, and expresses the obvious fact that sound attenuation increases with the mass of the wall:
TL = 20 (log f) + 20 (log W) - 33
where f is the frequency in Hz and W is the weight of the wall in pounds per square foot.
The audiophile may recognize that the slope of TL versus frequency is 6 dB per octave, the same as for an RC equalizer, rising with increasing frequency. If frequency or weight is doubled, TL increases by 6 dB. The formula is accurate for a monolithic wall such as concrete, but a well-designed stud wall should perform better than mass law at low frequencies. At high frequencies, the TL of real walls does not continue to rise, due to resonance phenomena and flanking paths.
Adding a complete extra wall, as reported by Jurew, can be very effective but is not usually recommended. A complete partition is very heavy and should not be placed on an existing floor unless a structural engineer indicates that the building will support the added weight. Local building codes will probably forbid this construction without a permit. The materials listed by Jurew may weigh half a ton; I am worried that a reader may build a wall that will end up in the apartment below.
Also, the footprint of the wall may significantly reduce the floor area of a small room. The wall's STC may be much greater than those of the flanking paths, such as doors, windows, ductwork, floors, etc., and in those cases the wall represents costly overkill. It is usually more cost-effective to add materials to the existing wall, such as furring strips, resilient channels, and drywall layers, with (uncompressed) insulation in the intervening air spaces. This kind of improvement, combined with the neighbor's cooperation in relocating speakers and possibly adding materials to his side, is an optimum solution.
The references in the Everest article are excellent, but some may be too technical 'or the audiophile/home carpenter. I would like to add two books to his list that may be more understandable to the layperson. Sound Control Construction, originally available from the United States Gypsum Co., is now out of print but is helpful if you can find it in a library. Quieting: A Practical Guide to Noise Control, written by R. D. Berendt, E. L. R. Corliss, and M. S. Ojalvo, was published by the National Bureau of Standards in July 1976 and is now available from the National Technical Information Service in Springfield, Va. Call (703) 487-4650 and ask for Handbook 119.
In addition, one should consult the Sweet's Catalog File, available in libraries, for data and design information in the catalogs of building material manufacturers. A favorite book with acoustical data is mentioned in section nine of Sweet's, Fire Resistance Design Manual, and is available for $6.50 plus postage from the Gypsum Association, 810 First Street N.E., Suite 510, Washington, D.C. 20002 or by calling (202) 289-5440. I trust that you will find the bulk of these comments constructive, no pun intended.
(adapted from Audio magazine, Dec. 1991)
= = = =