Your questions answered: Speech Privacy for Engineers, Architects, and other Construction Professionals
The September 12, 2018 “Speech Privacy for Engineers, Architects, and other Construction Professionals” webcast presenter addressed questions not covered during the live event.
Modern architectural trends often can create a "speech privacy crisis" in many commercial construction projects. Fortunately, specifiers have some tools at their disposal to address many of the problems that can arise.
This webcast discussed the As, Bs, Cs, and Ds of architectural acoustics and identified how these four acoustic principals affect speech privacy levels in the workplace. We defined the term "speech privacy" and focused on the impact both employees and employers experience. The webcast explored the impact that common construction techniques and interior furnishings pose on the levels of speech privacy one can obtain, keeping in mind modern interior design may conflict with client’s speech privacy goals.
Learn how sound masking contributes to increased speech privacy levels as well as the different general approaches sound masking systems use in terms of deployment, performance tradeoffs, and design parameters. This emphasis on sound masking systems will differentiate between "white noise" systems of the past and modern technologies that often use network infrastructure and contemporary components.
This webcast focused on how specifiers might best approach specifying sound masking systems in their projects to ensure client satisfaction. Information is shared to better explain how systems may be designed, specified, made code compliant, and deployed, making the process seamless for both construction professionals and clients.
Presenter Todd Berger, CET, CTS-D, field sales engineer, Cambridge Sound Management, responded to questions not answered during the live Speech Privacy for Engineers, Architects, and other Construction Professionals webcast on September 12, 2018.
Question: What methods are used to make general setting adjustments of a sound masking system? What is considered average levels for general office spaces?
Todd Berger: The range typically is an overall level between 42 dBA and 48dBA depending on site conditions. It is important that the ambient noise of a building be considered when adjusting levels. When sound masking levels are in the higher portion of this range, more speech privacy will be realized at the expense of occupant comfort. As a general statement, the higher the ambient sound levels in a space, the more sound masking can be tolerated by occupants.
Q: Is there a specific audio curve to be achieved in the masked area?
Berger: Yes, the curve used for sound masking needs to cover a range of 200 Hz to 5 kHz minimum with a center frequency of 1 kHz. The roll-off of the spectrum is approximately -6dB per octave above 200 Hz.
Q: How do masking systems affect the Noise Critera (NC) or sizing decisions for the mechanical engineer?
Berger: Sound masking usually is configured post-HVAC system balancing and therefore is configured to coincide noise emanating from mechanical systems. As the majority of mechanical system noise is at the lower range of the speech privacy band, sound masking usually is adjusted based on the mechanical system noise present, not the other way around. However, in many cases, HVAC can exceed the threshold of sound masking. In such cases, occupants may complain of excess noise coming from HVAC while believing it is coming from the sound masking system.
Q: Do sound levels from two different sources add?
Berger: Yes, if they are coherent (in time/phase with one another). Sound masking system often use noncoherent noise generators (sources not in time) for this very reason to ensure adjacent loudspeakers do not "sum" in the overlap between loudspeakers.
Q: Can sound masking can be heard by humans?
Berger: Yes, sound masking is audible and must be so to "cover up" the signal/sound of the talker. The sound of the sound-masking system reduces the signal-to-noise ratio of the talker’s signal/sound at some given distance, which determines the radius of distraction.
Q: It seems that 48 dBA is a very low level for the conversation; can you address this?
Berger: Remember that the signal of a natural talker is only about 70 dBA. As the talker and listener become further separated by distance, the talker’s signal decreases at a rate of -6 dB per doubling of distance. A sound-masking system set to 48 dBA essentially provides a signal-to-noise ratio in this case at a distance of around 25 ft.
Q: What gauge speaker wire do you use for network and constant-voltage systems?
Berger: Best practices for audio-video (A/V) systems dictate here. Cambridge Sound Management standard practice is 16 to 18-gauge depending on length of the cable run. Network systems have a low voice-coil impedance where constant-voltage system impedances are much higher due to the transformer. Direct field systems (Qt Pro) use Category UTP cabling (24 AWG) multiconductor cabling for distribution.
Q: One of the last items noted was that final tuning must occur when 1.) space is finished, 2.) unoccupied and 3.) after furniture is installed. Can you go a little further into why it must occur after furniture is installed?
Berger: The furnishings will affect the acoustic properties of the room (primarily absorption and diffraction). Therefore, an empty space typically will have a different natural acoustic signature than a furnished one. As sound masking is comprised of signals as experienced by listeners, these furnishings need to be in place at the time of system tuning.
Q: What can consulting-specifying engineers be expected to provide to the typical sound management installer?
Berger: Initial deliverables include a layout of devices based on a reflected ceiling plan (RCP) or floor plan. This should denote devices IDs and be at minimum clouded to include zone demarcation. Cambridge Sound Management will typically provide a drawing on engineer-provided backgrounds denoting these items for the specifier. We also have available project specification sections readily available that are product-agnostic.
Q: What does a typical direct ceiling system cost per 100 sq ft?
Berger: Sound-masking system prices typically range from $1.00 to $2.00 per sq ft installed costs depending on the product chosen and feature set required (paging and music requirements). This can vary slightly based on site conditions and occupancy.
Q: Have there been any psychological effects noted with this system?
Berger: Sound-masking systems have been deployed for nearly 50 years with no adverse effects to building occupants documented. Remember that this is the same frequency range as human speech at a low level. When done properly, occupants will likely not know a system has been deployed.
Q: Can you use internal masking for blocking external sounds, such as a nearby freeway?
Berger: No. Sound masking is a speech privacy product. The scenario you describe is one where the sound transmission class (STC) of the building envelope must be increased to a level where the freeway noise cannot be a nuisance.
Q: Under uniform situations, can people notice masking under quiet scenarios?
Berger: Sound masking must certainly be audible to work. However, when sound masking is very consistent, most occupants will not realize that a masking system is deployed unless they are told (not recommended). Most occupants equate sound masking to air movement from other building systems and simply believe it to be noise from properly-deployed HVAC.
Q: Do systems "autotune?" If so, are additional sensors required?
Berger: There are simply too many variables (acoustically) for an autotune system to work well in sound masking. The listener location is subject to change, there are multiple sound-masking generators in use, there are reflections that must be accounted for, and there potentially are architectural variations that affect the sound experienced by the listener. It takes a (talented) human being to make these tradeoffs in real-world scenarios when tuning a sound-masking system.