Printer Friendly

Environmental noise case studies: air-cooled refrigeration chiller installations near residential structures.


The case studies presented herein are air-cooled chiller installations at various schools subject to the same city noise ordinance, which is based on Uniform Building Code (UBC) Ch. 35 which restricts exterior noise levels received at a residential property to 63 dBA. In addition, 7 dB corrections are to be applied for tonality and for nighttime, so that net permissible level is 49 dBA (63-7-7) at residential properties adjacent to the schools. (1) Ordinance conformation was not enough, however. The owners (school districts) and design engineers wished to avoid complaints from the public, even if in some cases the levels would have to be lower than ordinance allowables.


School A is an elementary school receiving a replacement for an aging, single chiller in a three-sided, brick enclosure. The fourth wall of the enclosure is chain-link fence, causing the enclosure to be open to single family residences across the street from the school, the closest of which is 135 feet (41 m) away. There is a history of complaints about chiller noise at those residences.

The location is subject to the municipal noise ordinance, which has a limit of 63 dBA measured at a residential property line under separate ownership. There is a nighttime penalty of 7 dBA. A further penalty of 7 dBA is applied to noises that are impulsive in nature (changing more than 10 dB per second) or having strong pure-tone components. Based on the lack of a strong pure-tone in the measured data (see below), and the possibility for nighttime operation of the chiller, the unmodified nighttime limit of 56 dBA was chosen as the applicable criterion for conformance with the ordinance.

Daytime sound level measurements were taken to characterize noise emissions and propagation for the existing chiller and ambient conditions using a handheld, ANSI Type 1, 1/3-octave sound spectrum analyzer. Octave band equivalent sound levels (Leq, integrated average over measurement time span) were examined for ambient conditions near the existing chiller (with the chiller off) and at the nearest residential property line with the chiller running at 25% capacity. At the time of the measurement (early weekday afternoon), ambient noise levels were 49 dBA, and the chiller running at 25% capacity measured 65 dBA at the nearest residential property line, exceeding the noise ordinance for both daytime and nighttime.

The existing chiller showed strong sound energy in the 250-Hz octave band, measured as approximately 64 dB. The ambient noise spectrum showed a dip in the 250-Hz band. The high sound energy from the chiller in this band compared to the ambient spectrum could result in higher audibility of the existing chiller when compared to other chiller models with similar broadband noise generation.


The proposed replacement chiller is available in three noise reduction packages: Quiet Fan, Ultra Quiet Fan, and Low Noise. The Quiet Fan and Ultra Quiet Fan packages include alternate fan blades and speeds, while the Low Noise package includes a compressor covering in addition to the Ultra Quiet fans. The manufacturer provided sound pressure data at 30 feet (9 m) for all three packages at 25%, 50%, 75%, and 100% loads. For comparison to the measured levels of the existing chiller, the data for 25% load were normalized for distance (-2 dB) and for placement in a reflecting enclosure (estimated at +3 dB).

Figure 2 compares the measured levels of the existing chiller to measured ambient levels and the normalized 25% load manufacturer's data for all three packages. Other than its smoother spectrum, the Quiet Fan package shows no improvement over the existing chiller, with an estimated property line noise level of 65 dBA. The Ultra Quiet Fan package shows a courser spectrum with potential audible tone issues but would be 4 dBA less at the property line, while the Low Noise package shows a smoother spectrum 7 dBA less at the property line. It does not appear that any package will allow the new chiller to be a drop-in replacement that can meet the 56 dBA nighttime noise limit. Selection of the Low Noise package was recommended based on its smoother spectrum and lowest overall broadband noise emission.


The 100% load data for the Low Noise package was normalized to the existing conditions for comparison to the noise ordinance. The manufacturer's data reports 68 dBA at 30 feet (9 m). Normalized to the applicable distance and placement in a reflective enclosure, the predicted property line level is 69 dBA, 6 dBA above the nighttime limit and 13 dBA above the daytime limit. To achieve the additional 13 dBA of attenuation needed to meet the criterion, recommendations were made for modification of the chiller enclosure, the addition of sound absorbing surfaces to the chiller enclosure, and the installation of flexible refrigerant pipe connections to the new chiller.

The addition of a fourth, solid wall to the chiller enclosure was recommended to provide shielding between the chiller and the residences across the street. Additionally, an increase in the wall height of the enclosure was recommended for better containment of chiller noise in all directions. A study of the chiller's ventilation requirements was recommended to determine if the additional wall would reduce air flow to the chiller unacceptably. If the need for additional ventilation is determined, the use of acoustic louvers in the new wall is recommended and/or the addition of acoustical louvers in one of the existing walls.

Installations of spray-on absorptive material or mounted sound absorbing panels were recommended to reduce noise build-up inside the enclosure. The recommendation was for 50% coverage with a material having NRC 0.65 or higher.

The manufacturer indicated that no internal vibration isolation is installed on the chiller beyond isolation mounts for compressors. To further isolate the compressor bodies from the chiller framework, recommendations include the installation of flexible couplers in refrigerant pipes between the compressors and the chiller frame, and the use of resilient mounts where refrigerant piping is attached to the compressor frame. The intent of these recommendations is to reduce vibration in the chiller frame that can be re-radiated as noise by metal structural members or casing cover panels.


Cooling capacity requirements were increased at a middle school, due to a facility expansion. In addition, to reduce noise intrusion into classrooms, the mechanical yard was to be relocated. Two larger air-cooled refrigeration chillers were specified to replace the original chillers. Original and new chillers incorporate rotary helical or "screw" compressors. The new mechanical yard was designed with walls around three sides, with north and northwest open (chain link fence), whereas the former mechanical yard was open only to the northeast. The new chiller installation was near building walls that could reflect sound in the northwest to northeast directions. When chillers were initially operated during commissioning, a noise complaint was received from a neighborhood residence approximately 1000 feet (305 m) to the northwest. The municipal noise ordinance permitted up to 49 dBA at residential property boundaries when tonality and nighttime corrections were applied, as described in Introduction above. A noise study was requested to determine chiller noise conditions due to the chiller installation and to develop and recommend noise mitigation, if necessary to conform to noise ordinance and/or resolve the residential noise complaint. Post-installation performance validation measurements were to be conducted also.

Existing Conditions: On-Site Observations and Measurement Results

Acoustical measurements were conducted on-site to determine chiller noise level and spectrum for normal ambient (without chiller contributions) and for part- and full-load chiller operations using a handheld, ANSI Type 1, 2-channel spectrum analyzer and two Type 1, 1/3-octave, 1-channel analyzers. Equivalent levels (Leq) and statistical percentile levels (Ln) were obtained 5 feet (1.5 m) above ground at locations approximately 60 feet (18 m) from the chillers, at the school property boundary approximately 300 feet (92 m) to the northwest and approximately 1000 feet (305 m) away, in the neighborhood near complaint origin, as shown in Figure 3(b). Other measurements were made within and immediately adjacent to the mechanical yard enclosure to determine and compare spectra of the chillers with some indications of relative contributions from compressors and condenser fans.



Findings and Recommendations

Measurements indicated that chiller noise levels below 49 dBA in the neighborhood conformed to municipal ordinance requirements. Although it is not easily seen in the 1/3 octave spectrum, narrowband measurements showed some tonal peaks at various frequencies, which appeared to represent the new chillers and possibly other air conditioning equipment in the neighborhood. The spectrum near the chillers was very tonal with most significant prominent frequencies in 315-Hz and 125-Hz 1/3 octaves. Projected to the school property boundary, an approximate 49-51 dBA level only marginally conformed to the ordinance. Observations and measurements on and near the chillers indicated broadband noise from condenser fans and tonal noise and vibration from chillers. Concrete retaining walls around three sides of the chiller mechanical yard were contributing to a reverberant build up around the chillers.

Noise containment jacket or enclosure was recommended for the compressors. (The equipment had been ordered with noise jackets, but only temporary enclosures formed out of duct board, not noise barrier material, had been installed.) Pulsation damper or diffuser for compressor discharge was recommended. Resilient connections between refrigerant piping and chiller chassis/frame was recommended, and to reduce radiated sound, damping was recommended on interior side of sheet metal side panels of chiller. Acoustical absorption treatment (suitable for outdoors) was recommended for the mechanical yard retaining wall enclosure. A barrier wall with height greater than the chillers was recommended to replace the chain link fence at the open end of the yard.

Results and Conclusions

With the noise barrier enclosure and compressor noise jacket recommendations implemented, the chiller noise was significantly reduced, particularly in the northwest direction, so that chiller noise conformed to ordinance limits at the property boundary and was only 43 dBA in the neighborhood. The noise spectrum outside the mechanical yard noise barrier wall showed reduced tonality. The original complaint was not adequately resolved, however, indicating incremental development of receiver sensitivity or receiver hearing a different source and attributing to the chiller installation. The owner and engineer have proceeded with supplemental implementation of the chiller panel damping and minor modifications of refrigerant piping attachment mountings to reduce radiated noise.

This installation is substantially improved, with reductions in noise intrusion into classrooms, reduced noise and tonality at the neighborhood, and it conforms to municipal ordinance limits.



School C is an elementary school with a new building addition and new air-cooled chiller installation. When construction on the new addition was nearly complete, the chiller was turned on and began to draw noise complaints from residential neighbors. The aerial photo shown in Figure 6 illustrates the approximate location and orientation for the new chiller. The design included a new ~7-foot tall (2.1 m) masonry enclosure wall with security fence. The topography in the area is relatively flat. Neighboring residential houses are typically single-story. The new school addition would be three stories tall once completed. Neighbors that were exposed to noise disturbance from the new chiller had already been subjected to noise from two existing chillers, without any complaints. Therefore, during design and equipment selection, adverse conditions were not anticipated from the additional chiller. The new chiller was specified to include a noise control enclosure or lagging around the compressors to contain radiated compressor noise. However, shortly after the new chiller was operational, and before the compressor enclosure was installed, neighbors began to complain about the new chiller's noise.


Existing Conditions: On-Site Observations and Measurement Results

A site visit was arranged to conduct observations and noise measurements with the two existing chillers and one new chiller at different compressor loading and staging conditions. Measurement results were reviewed to determine spectrum and level of the noise from the new chiller, relative to background existing conditions and city ordinance limits. A particular focus was to identify the chiller noise levels near locations of complaints for the purpose of determining mitigation required to mitigate the disturbance complaints. Measurements included equivalent levels (Leq) with the microphone located approximately 5 feet (1.5 m) above ground. During that first site visit, the following findings and observations were made:

1. The new chiller noise received at complainant's property was about 64 dBA, approximately 15 dBA louder than allowed by city ordinance during nighttime hours.

2. The new chiller's noise was tonal. (2) Results showed prominent tonality for the new chiller in the 315-Hz 1/3 octave band, related to the compressors. The 63-Hz, 125-Hz, and 630-Hz 1/3 octave bands also indicated contributions from various harmonics apparently related to condenser fans (63 & 125 Hz) and compressors (315 & 630 Hz).

3. The existing chillers' noise was also tonal, but not as prominent at residential property lines as the new chiller.

4. The shape of the mechanical yard enclosure walls around the new chiller appeared to contribute reflections to reinforce the chiller noise and direct it over the walls towards neighbors as shown in Figure 7(a).


5. The orientation and height of the building wall and roof soffit shown in Figure 7(b) also contribute reflections to reinforce chiller noise, directing it over walls2 towards the neighbors as shown in Figure 7(a).

6. Surface radiation from parts of the chiller produce audible airborne noise that reinforces noise radiated directly from the compressors.

Existing Conditions: Recommendations

Based on evaluations of existing conditions, the following noise mitigation measures were recommended and coordinated for implementation:

1. Complete the installation of planned noise control enclosure or lagging around the compressors to contain greatest radiated tonal noise.

2. Install and extend a taller solid barrier wall along north and west sides of the existing chiller enclosure. Add or extend noise barrier wall around the north end of the chiller where the open chain-link fence exists now, as shown in Figure 8.


3. Provide acoustical seals, overlapping barrier materials to seal joints at entry/access doors as shown in Figure 8b.

4. Apply acoustically absorptive surface finish to the existing masonry enclosure wall surfaces facing the chiller.

Post Construction: Observations and Measurement Results

After noise mitigation measures were implemented to extend barrier walls and add acoustically absorptive surfaces, neighbors continued to complain, indicating they did notice a reduction in high-frequency "whine" or what was described by neighbors as "machine" noise but did not notice much reduction in a low-frequency "rumble" or "hum" that was still annoying to them. To evaluate the results for comparison with city ordinance requirements and complainants' descriptions, additional noise measurements were conducted on site. Results indicated the following:

1. The new chiller noise received at complainant's property was reduced by about 5 dBA, almost achieving the city's allowance for daytime noise, but still approximately 10 dBA louder than allowed by city ordinance at night.

2. As shown in Figure 9b, a significant noise reduction of 5 to 10 decibels was achieved in the frequency spectrum from 80 Hz to 1000 Hz, reducing the tonality of the equipment noise received at neighbors' properties.


3. Very little noise reduction was achieved for condenser fan tone in the 63-Hz 1/3 octave band. It was assumed the 63-Hz tone represented the prominent low-frequency "rumble" that neighbors continued to be annoyed by.

Post Construction: Further Recommendations

For additional noise mitigation, the following additional measures were recommended, and are currently being coordinated for implementation. To reduce fan noise, apply silencers supported separately above the condenser fans with a tight clearance to the top surface of the chiller (but not touching). This measure would reduce low frequency tonal noise from condenser fans but would not be likely to significantly reduce the overall A-weighted (dBA) chiller noise level with respect to city noise requirements. However, since the primary complainants indicated their concern was with the lower-frequency "rumbly" noise rather than the higher-pitched "whine," it is hoped that this measure would improve the sound spectrum, improving the sound quality with respect to the neighbors' specific concerns, regardless of dBA city noise limits.


Air-cooled chiller noise is often tonal. Noise reduction that is designed for outdoor chiller equipment to achieve a reduction in A-weighted (dBA) noise level will not necessarily reduce the perceived impact on sensitive receivers. Designers should consider equipment tonality, noise radiating from equipment surfaces and pipes, reflecting building and wall surfaces, topography and lines of sight to potential noise receivers, and existing ambient noise conditions when evaluating the potential impacts of new outdoor chiller equipment on project sites and at neighboring properties.


(1.) Code of Ordinances, Part II, Ch. 21, Article III, Division 1, Sec. 21-52. -- Noise nuisance enumeration (a.9) and (b) Table 1, San Antonio, Texas, 2010

(2.) ASHRAE. 2011. ASHRAE Handbook-HVAC Applications, Ch. 48 -- Noise and Vibration Control. Atlanta: American Society of Heating Refrigeration and Air Conditioning Engineers, Inc.

Jack B. Evans, PE


Chad N. Himmel, PE


Joshua D. Leasure, PE


Jack B. Evans is Principal Engineer of JEAcoustics /Engineered Vibration Acoustic & Noise Solutions, an acoustical consulting firm in Austin, TX. Chad N. Himmel and Joshua D. Leasure are Associate and Sr. Consulting Engineers, respectively, at JEAcoustics, Austin, TX.
COPYRIGHT 2012 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Evans, Jack B.; Himmel, Chad N.; Leasure, Joshua D.
Publication:ASHRAE Transactions
Article Type:Report
Geographic Code:1USA
Date:Jul 1, 2012
Previous Article:Field trials of a waterless home heating and humidification technology.
Next Article:Field-test of the new ASHRAE/CIBSE/USGBC Performance Measurement Protocols: Intermediate and advanced Level Indoor Environmental Quality protocols.

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters |