April 4, 2019
by Ben White and Steve Titus, P.Eng.
Every building project brings unique acoustical challenges for the owner and design teams. A successful design will meet these challenges at the lowest possible cost. However, until a building is constructed, questions about how it will sound and the impact of different design choices can be difficult to answer in simple terms.
The questions can take many forms. How will a different geometry and choice of finishes affect the sound of a new recital hall for different types of music? How much sound bleed from one mix room to another will be acceptable in a new studio? How will an open office sound with or without acoustical treatments? What kind of privacy can be expected if a new meeting room is built using glass walls? How loud will it be in a new condominium development when the subway passes underneath?
In many situations, it is critical to answer these questions before breaking ground. Unfortunately, the answers are often a highly technical abstraction and are not easily understood by non-engineers, owners, or architects. In some cases, the answers are not entirely clear even to acoustical professionals. However, getting the answer wrong can seriously affect the long-term viability of any project whether it is a performing arts complex or a new condominium development.
Not surprisingly, the question ‘how will it sound?’ is best answered with an audio simulation, but this can present its own set of challenges because so many factors must be considered to provide an adequate level of realism. For years, acoustical consultants have turned to simulation software for prediction and have succeeded in designing great spaces by relying on specialized technical parameters and theories—few of which are meaningful to anyone without an acoustical engineering background.
The challenges of acoustical simulations
Modern room acoustics software packages enable different designs to be simulated by synthesizing digital files known as impulse responses. These can be used to superimpose the acoustic signature of any space on any sound source. The most common and cost-effective playback method is to use a set of headphones. Unfortunately, this method has several limitations including:
The first issue is not simple to address with currently available techniques—access to an acoustics laboratory with an anechoic chamber and a very patient test subject would be required. Due to advances in virtual reality (VR) technology, the second issue can be addressed with relative ease but would require a specialized hardware and software setup. The third issue would require dedicated subwoofers and associated processing to separate low and high frequencies.
To address questions involving loudness, such as how loud a subway would be in a new condominium, consultants can rent rooms and equipment to play back the noise at a calibrated level, if they can find the right room. This is challenging because most rooms or venues are too noisy to be used for simulation purposes, especially when the noise is significantly quieter than the room itself. The other issue is trying to provide the same experience to all listeners. A simulation room should limit seat-to-seat variation as much as possible, but this can be difficult to achieve.
If the stars align and the right room is available for the type of simulation to be conducted, renting the room and equipment can be a costly undertaking. Budget considerations usually mean this kind of simulation is limited to large and high-profile projects. For standard-sized projects, owners and architects must rely on their interpretation of the acoustician’s recommendations.
Industry-wide, there is now a move toward using purpose-built simulation rooms employing a technology called ambisonics, a method of encoding directional information with a very high level of detail. This technology allows the sound field of any real or simulated space to be reconstructed in three dimensions.
Moving offices to a larger space in 2017 provided Aercoustics with the opportunity to design one of the first dedicated ambisonic simulation environments at an acoustical consulting firm in Canada. When combined with a VR headset, this technology provides an in-depth understanding of the way a space will sound and look before ground is broken.
Designing a simulation room
As with any acoustically sensitive space, three key design parameters had to be considered: room acoustics, background noise levels, and sound isolation.
Since the room is used to simulate how other spaces will sound, it needed to be as acoustically neutral as possible. To attenuate strong reflections from walls and ceilings, which would harm the accuracy of simulations, acoustically absorptive finishes were used. The walls were fully treated with an acoustically transparent stretch fabric system, backed by 100-mm (4-in.) deep panels of semi-rigid glass fibre insulation to provide acoustical absorption. The ceiling consists of a 25-mm (1-in.) thick suspended mineral wool tile in a T-bar lay-in grid. Additionally, the ceiling cavity behind the tiles was filled with acoustic batt insulation in order to maximize absorption at low frequencies.
While the walls are fully absorptive and parallel wall surfaces are not a concern for flutter echo or other acoustical anomalies, a splayed wall layout was used with an eye on future experimentation with reflective and scattering surfaces in a critical listening environment. It is worth noting that even with full absorptive treatment the room is not entirely free from reflected sound (anechoic) as would be the case in a research environment. However, the magnitude of the acoustic reflections has been reduced sufficiently to avoid interfering with direct sound from the playback system, so convincing building acoustics simulations can be conducted.
Controlling the background noise level is one of the obstacles when renting a space for simulation purposes. To ensure it was not an issue in the Aercoustics simulation room, the mechanical system was designed to be very quiet (noise criteria [NC]-10) and can also be turned off if needed for threshold of hearing (N1) applications. Additionally, all of the amplifiers and equipment required to operate the room are located in a rack outside of the testing space so there are no cooling fans to interfere with the sound. With 12 powered speakers and four subwoofers in the space, audible hiss and hum was a significant concern. This was addressed by selecting very quiet powered speakers and paying close attention to the design of grounding and wiring feeding the powered speakers and subwoofer amplifiers. The audio signal chain is fully digital up to the speaker feeds that use balanced lines for low noise performance.
To provide an adequate level of sound isolation, the walls were built as isolated double-stud walls with two layers of gypsum on 92-mm (3 5/8-in.) steel studs on both sides with a 25-mm air space in between. Each stud cavity is filled with acoustic batt insulation. To minimize sound transfer, there are no windows in the space and the room was designed with a vestibule comprising a sliding glass door, and a heavy solid wood door with full sound seals. When communication to the outside is required, it is achieved using webcams and talkback microphones, in much the same manner as in a professional recording studio.
Ceilings in the office are 6 m (20 ft) to 9 m (30 ft) in height; due to the extensive height and interferences with mechanical services, the walls were not constructed from the floor to ceiling. Consequently, Aercoustics designed an isolated ceiling system that was constructed from the outer studs and is composed of two layers of gypsum board, insulation, 92-mm studs and two layers of plywood.
Ideally, the floor would have been isolated to create a room-within-a-room construction using either a discontinuous slab-on-grade or a floating floor. The existing floor is continuous slab-on-grade construction and cutting the slab to isolate was not permitted by the landlord. Building a floating slab would have required ramping and loss of significant floor area, which was unacceptable from a layout perspective. With this limitation on design, testing revealed impact noise through the common slab from hard-soled shoes was the only significant concern. This is readily controlled through administrative means.
An improvement on old methods
Using an array of 12 loudspeakers in a quasi-spherical arrangement, the room can recreate acoustical environments in three dimensions. The perceived quality of the simulation no longer depends on knowledge of a listener’s cranial dimensions, and so spatial realism of the simulation is significantly improved when compared to headphone simulations.
For a recent performing arts project, this allowed validation of the design to a degree that was unachievable in the past. When combined with a VR simulation, reflections from overhead surfaces and the rear wall were remarkable to perceive when these aspects of the design had formerly been nothing more than lines on a page. On a recent feasibility study for the acoustical renovation of the Montréal Olympic Stadium, where an acceptable reverberation time design target was hard to determine, simulation provided the necessary information to develop an effective design, while allowing the client to understand exactly what they would get for their money with different levels of treatment.
Another key aspect leading to the improved realism is the quality of the low frequency reproduction in the 20 to 100 Hz range. Convincing low frequency reproduction of sources such as subway rumble or bass instruments is lacking in headphone simulations since bass is heard but not felt as it would be in real life. The issue is an acoustic phenomenon known as room modes or standing waves.
At modal frequencies where the wavelength in air corresponds to dimensions in a room, energy will build up in the space, and it will ring. All rooms have modes, and these modes can radically change what is heard at specific frequencies depending on the listening position. Moving a few feet can change not only the level by as much as 10 to 20 dB (quiet versus loud), but also the character of the sound (‘tight’ to ‘boomy’). This level of variation can be misleading and would be unacceptable in a precision simulation environment.
A special system was therefore developed for reproduction of low frequency sources. Multiple subwoofers have been placed throughout the room so that at seated height anywhere in the room, the level varies only slightly (+/- 3 dB) from the target frequency response. This ensures an accurate simulation. The use of multiple subwoofers has many advantages over methods used in the past for low frequency control. In some recording studio control room designs, as much as 50 per cent of the floor area might need to be consumed by acoustical treatment to provide a similar level of control at low frequencies.
The entire playback system is controlled using a dedicated digital signal processing (DSP) unit, performing all processing needed to provide a uniform low frequency response and to ensure all speakers are configured with the correct delay and equalization settings. The configuration is fixed and inaccessible to users of the system to avoid the potential issue of an accidental change of settings. The processor is fed over Ethernet using the Dante protocol that allows up to 1024 channels of uncompressed digital audio to be transmitted to the listening room. If desired, a simulation and two-way communication with the client could be conducted using a laptop from any room in the office.
As a result of this level of control, any project affected by low frequency noise can use the technology to get a clear picture of how the intrusion will sound. For example, this approach was used to simulate subway noise intrusion for Toronto’s Koerner Hall with and without isolation. Subways are not the only noise source that can be simulated. This technology has been used in various projects such as simulating helicopter noise through different windows, walls, and ceiling assemblies for a hospital. Weight drop noise from condominium gyms, basketball courts above classrooms, set construction shops beside recital halls, and rock music recording and rehearsal facilities are other applications where this technology has been applied to gain a better feel for the numbers and understand the true implications of various design recommendations.
In conclusion, simulation rooms can be a valuable tool for building professionals. The insight provided will result in savings in time, cost, and frustration. With a true picture of how a building will sound, risk and expectations can be managed more effectively.
Ben White, B.E.Sc., has worked as an acoustical consultant at Aercoustics for four years. He was responsible for the design of the Bridge, a new virtual reality (VR) acoustical simulation environment at Aercoustics, among other in-house technical initiatives. White has worked as acoustical project manager on the Centre for Technology and Innovation at Humber College, the new Rogers Radio Studios, Lynx Music studios, and many other projects in the institutional, performing arts, and media/broadcast spaces. He can be reached at firstname.lastname@example.org.
Steve Titus, B.A.Sc., P.Eng., brings more than a decade of experience to being CEO of Aercoustics Engineering, a privately held firm specializing in acoustics, vibration, and noise control. He has been responsible for the acoustical design and delivery of several high-profile projects such as the Sick Kids Research Tower, Corus Quay, Thunder Bay Courthouse, and the St. Lawrence Market North redevelopment. Titus is co-chair for Canstruction Toronto, and sits on the finance and audit committee of the Consulting Engineers of Ontario (CEO). He can be reached at email@example.com.
Source URL: https://www.constructioncanada.net/building-an-acoustic-simulation-room/
Copyright ©2021 Construction Canada unless otherwise noted.