e design: Rethinking Classroom Acoustics

Note:
This article was mined from a State of Florida website, e design, before the state pulled the plug on it. I wish I had downloaded more from this excellent former site. My apologies to anyone whose copyright I have treaded on.
Fred Tepfer, June 2001

Rethinking Classroom Acoustics: Part One

ASHRAE Winter Meeting Seminar
HVAC Noise in Classrooms: Overcoming Barriers to Learning

Posted 06 April 1999

	A seminar on classroom acoustics at the ASHRAE (American Society of Heating, Refrigerating, and Air-Conditioning Engineers) Winter Meeting (Chicago, January 1999) was surprisingly well-attended. Around 200 people turned out to hear an update on this important, but seemingly mundane, issue. The reason for the exceptional turnout was soon made clear. There will likely--and soon--be major, mandatory changes in what is considered an acceptable quality of classroom acoustics under the auspices of the Americans with Disabilities Act (ADA).
	The Access Board, which provides support for the implementation of the ADA, received a complaint regarding poor classroom acoustics and the potential impediments that such environments may cause for the hearing-impaired student. In other words, it was suggested that noisy and reverberant classrooms may act as a barrier to learning; as a barrier to educational access. The Access Board has been researching this issue and appears likely to support this claim. Several of the presentations at the ASHRAE seminar strongly suggested that the claim of denied access is not-at-all far-fetched. Summaries of the four presentations that comprised the seminar follow.
	Seminar Introduction
	During a short introduction to the seminar, it was noted that ASA and ANSI (the Acoustical Society of America and the American National Standards Institute) are forming a task group to develop standards for background noise and reverberation time for classrooms. These standards are likely to provide the basis for federal rulemaking in the year 2001; such rulemaking will probably impact all classroom design efforts. The effect of such rules is very likely to change mainstream thinking about the selection and design of classroom HVAC systems.
	Ted Carnes: Acoustical Criteria Made Simple
	Ted Carnes provided a review of acoustical design criteria as an opening to the seminar. Background sound levels are typically expressed in terms of dB(A), NC, or RC values. dB(A) refers to an A-weighted single-number sound pressure level measurement obtained with a sound level meter. The (A) weighting is intended to approximate the response of the human hearing system, while a (C) weighting scale is more responsive to low frequency sounds. Carnes noted that if the dB(C) measurement exceeds the dB(A) measurement in a space by more than 7 dB (decibels), then the (A) scale measurement may not be a good indicator.
	NC stands for "noise criterion," which is a single-number rating of background noise level obtained from octave-band sound pressure levels plotted on NC templates. A newer NC rating — NCB — has been introduced, with the "B" standing for "balanced." RC stands for "room criterion," which is another single-number evaluation that attempts to address sound quantity and quality. Information on RC may be found in the ASHRAE Applications Handbook. Carnes noted that RC is generally preferred (at least in ASHRAE circles) to NC, and that it was desirable that a sound spectrum lie along an RC contour. RC (Mark II) was mentioned as a new variant of RC, but a variant that was very hard to verify in the field.
	Other indicators of background noise that are sometimes encountered include SIL and PSIL. SIL stands for "speech interference level" and is the average of sound pressure levels at the 500, 1000, 2000, and 4000 Hz frequencies. PSIL is similar, but includes only the 500, 1000, and 2000 Hz frequencies. Speech Transmission Index (STI), Rapid Speech Transmission Index, Articulation Index, and percent loss of consonants (in speech) are also used as indicators of acoustical conditions in classroom-type spaces.
	Carnes summarized a review of articles dealing with school design that showed a general focus on dB(A) as an indicator, with an occasional mention of SIL and NC. References to RC were rare in such articles, although Articulation Index was often mentioned. In response to a question, Carnes noted that NCB may be a good indicator of classroom background noise if speech is the primary concern.
	Peggy Nelson: The Effect of Acoustical Barriers to Learning in Classrooms
	Nelson provided an eye-opening look at classroom acoustics from a student's perspective. Typical classrooms often have background noise levels of from 35 - 45 dB(A) when unoccupied — with the lower number corresponding to HVAC system "off" and the higher number to HVAC system "on." When occupied, the corresponding numbers become 58 - 62 dB(A). The result of such background noise levels is a low signal-to-noise ratio. Signal-to-noise ratio is a critical determinant of good hearing; low ratios are not good.

	Nelson further noted that Articulation Index results can vary greatly with age of the subjects (for example, first graders versus adults). Under good acoustic conditions, the age of the listeners has little impact on Articulation Index results (ratings). Under poor acoustic conditions, the age of the listeners makes a difference to Articulation Index scores — with younger listeners giving lower Articulation Index results than adults in the same space. Further, approximately 15 percent of children were estimated to have at least a slight hearing loss. School records, when properly analyzed, show a substantially greater lack of progression for students with hearing loss than those without.
	Sigfrid Soli: Factors Affecting Children's Speech Communications in Classrooms
	Soli presented a summary of the components that contribute to classroom communications and a powerful review of the acoustic problems faced by many students and teachers. A typical graph of intelligibility versus background noise is shown below. The conclusion that can be drawn from this graph is that low-background-noise environments provide a "level playing field" for many (if not all) students; high-background-noise classrooms start to impose barriers to hearing for a range of students.

	In theory, an adult, native-English speaker can hear acceptably well with a 0 dB signal-to-noise ratio. This is not the situation, however, in the typical classroom. The ability to sort signals from among noises is not fully developed until children reach their teens. Use of English as a second language also imposes difficulties to comprehension of speech, as do other conditions not uncommon in school classrooms. Soli estimated the effects of such conditions on required signal-to-noise ratio as follows: development status (age) can impose up to a 3 dB "penalty" hearing status (loss of hearing) can impose up to a 4 dB penalty lack of language proficiency can impose up to a 3 dB penalty the range of individual differences in hearing can impose a 3 dB penalty excessive reverberance in a room can impose a 2 dB penalty. The bottom line is that some students may require a 15 dB increase in signal-to-noise ratio to be able to hear as well as adults in a given classroom. If this does not occur, their learning will be impaired.
	Measurements in San Diego classrooms have shown the following typical conditions: teacher speech effort of 68 dB(A) quiet voice to 78 dB(A) loud voice background noise levels of 60 - 65 dB(A) equivalent. Such conditions do not provide an adequate signal-to-noise ratio for many students. In some cases this ratio is as low as 3 dB, while 15 dB may be required. Soli recommended a background noise level of around 35 dB(A) as a means of ensuring the availability of a 15 dB signal-to-noise ratio. Jerry Lilly: Designing Quiet HVAC Systems for Classrooms
	Lilly's presentation focused upon the contribution of HVAC systems to classroom noise and he was to-the-point. Low-first-cost HVAC systems will result in problems. One air supply to an entire classroom will not work. In-room air-conditioning units will not work. VAV boxes located directly above a classroom's acoustic tile ceiling may cause problems; typical radiated sound pressure levels may approach 65 dB (at 125 Hz), while ducted sound pressure levels may approach 70 dB.
	Lilly suggested that NC ratings for diffusers are often mis-applied, with estimated noise levels for one diffuser being assumed to apply for multiple diffusers. He recommended that the desired NC -6 be used when selecting 2 diffusers for a space; desired NC -9 for 4 diffusers; desired NC -12 for 8 diffusers in a space. Considering reverberation time, Lilly suggested that the NRC (noise reduction coefficient) of the ceiling times the percentage of coverage equal 0.65, and the NRC for walls times the percentage of coverage equal 0.25.
	Mark Schaffer: The Cost of Noise Control in Classroom HVAC Systems
	Schaffer presented some estimates of the cost of upgrading existing HVAC systems to provide the background noise levels recommended in the previous presentations. The cost per classroom to upgrade a rooftop HVAC system was estimated to be $1900. This included costs to modify the air-handling unit, a VAV box, the ductwork, the diffusers, and to add a sound trap. The rooftop scenario is on the good side of the picture. An upgrade to a closet-based HVAC system was estimated at $4700. It was recommended that ARI 350 (the Air-Conditioning and Refrigeration Institute, Standard 350) sound ratings be equal to or less than one-tenth the desired room NC plus 7 (assuming that dB(A) -7 = NC). Typical costs to reduce the noise impact of HVAC systems in classrooms were estimated by Schaffer to be around $1-5 per square foot. At a national average school construction cost of $77.50 per square foot, this is an additional cost of around 1-6%.
	The most pointed question at the end of the seminar concerned the fiscal impact of the suggested acoustical design criteria on local governments. Would this not be another unfunded mandate? The response was, in general, don't shoot the messenger--and there is a serious problem that should be addressed somehow.
	Walter Grondzik, Editor