Page 28 - Fall 2007

efit while increasing the likelihood of not being accepted by employees.
The above data are helpful as input to the programmed level control, but are limited to being samples. An adaptive masking system responds to the unique events of each minute of each day. By continually tracking the disturbance potential, the severity of the impact is lessened further. One beneficial feature of adaptive masking is that it responds to all transient sounds, so it will handle unexpected or exterior sounds, such as roadway or aircraft traffic. At the present time only one adaptive system is available.
Improved equalization methods
Most masking specifications permit a range of levels in any one or one-third octave band, typically it is +/- 2 dB. This permits up to 4 dB difference in contiguous bands. Experience has shown that although the spectrum contour is very important in open offices, it is just as important for the contour to be smooth. Newer systems permit a relatively smooth version of the chosen spectrum contour to be creat- ed in a short time.
Most sound masking generators are supplied with no high or low pass filter settings, and flat band pass filter settings. The source is pink noise. Standard methods require the person equalizing to make a set of spectrum measurements in the area of interest, average them, and then compare the results with a desired spectrum. The needed corrections are entered with a mechanical slider or a software slider. In practice this method is iterative and time consuming, considering that large systems may have multiple channels of equalization.
The first improvement is that modern masking systems permit initial spectra to be pre-set. The user can create pre- set spectra by saving and recalling a file created on a previous project. One manufacturer has created pre-set files for the three most common masking speaker locations: above a sus- pended ceiling, in an open ceiling, or under a raised floor. In both methods, a reasonably correct initial masking spectrum can be set. Since every building is acoustically unique, a pre- set spectrum is close but is not likely to contain the correc- tions needed to offset the acoustical impedance variations created by the environment into which the speaker looks. Therefore, a set of masking spectrum measurements is still required. But instead of hand calculating the required cor- rections and tediously entering them into the generator chan- nel, there now are better ways to accomplish this task.
First, it is necessary to choose the desired masking spec- trum. The modeling software mentioned above can be used to generate that spectrum. It may be chosen from the editable database of spectra, it can be developed from modeling the sound attenuation of the facility, or it can be developed from actual sound attenuation measurements. Modeling is the shortest method and has proven to be reasonably accurate, despite the inability to solve the wave equation.
Second, it is necessary to determine the actual masking spectrum by measurement. A set of spatially diverse spectra must be collected. These data may be saved to an Excel file, or to a file that the equalization software is capable of read- ing, or can be downloaded directly to the equalization soft- ware. The software then calculates the corrections needed to
match the two. The corrections can be entered manually into the equalizer, or they can be saved to a file that later is to be read by the generator software.
There are a few systems that have this capability. One manufacturer uses a hand-held computer with a microphone and a wireless connection to the generator, so that equaliza- tion is real time. In each case, the equalization is rapid and the spectrum is smooth and very close to that desired.
Conclusions
It appears that the technological evolution of sound masking equipment and methods has not found its way fully into the marketplace, but the availability of new products are showing signs of changing that. Increased understanding of the relationship between disturbance potential (measurable) and distraction (hard to measure) will permit manufacturers to develop systems that improve privacy dynamically.AT
References for further reading
1 Director of Central Intelligence Directive 6/9, “Physical Security Standards for Sensitive Compartmented Information Facilities,” November 2002.
2 “Health Insurance Portability and Accountability Act”, 45 CFR Parts 160 and 164 (2002).
3 J. P. Schieber, “Analytical Study in the Laboratory of the Influence of Noise on Sleep”, Final Report, Center of Studies of Physiology Applied to Work, Strasbourg Faculty of Medicine, April 1968.
4 “Information on Levels of Environmental Noise Requisite to Protect Public Health and Welfare with an Adequate Margin of Safety”, Document 350/9-74-804, U.S. Government Printing Office (1974).
5 A. H. Suter, “Noise and Its Effects,” Administrative Conference of the United States, November 1991 (see http://www.nonoise.org/ library/suter/suter.htm). Last accessed 9/30/2007.
6 B. Griefahn, “Environmental noise and sleep. Review—Need for further research,” Applied Acoustics 32, 255¬–268 (1991).
7 “Standard Test Method for Objective Measurement of Speech Privacy in Open Offices using Articulation Index,” American Society for Testing and Materials, E1130–02 (2002).
8 H. McGregor and R. Chanaud, Patent Number 4,098,370, “Vibration masking system,”4 July 1978.
9 “Occupational Noise Exposure,” Occupational Safety and Health Administration, 29 CFR Part 1910.95.
Dr. Robert Chanaud has been involved with acoustics since 1958 and sound masking since 1972. He has written a manuscript on sound conditioning offices and has recently completed a manual on sound masking.
Photo: Robert Chanaud and his wife, Jo.
26 Acoustics Today, October 2007