At your request, David Coate Consulting (DCC) analyzed the acoustics associated with the 225 Liberty Street project located in New York City. The project entails the build out of two presentation/meeting areas, each of which is circular in shape. Acoustics in circular rooms is often difficult to address because of focusing and reverberation effects. Such effects can include flutter echoes as well as the perception that a sound source is in closer proximity to the listener than it actually is.
Three specific issues were analyzed:
• Reverberation in Forum 1 (1st floor)
• Reverberation in Forum 2 (2nd floor)
• Sound transmission through the glass partition (2nd floor)
Acoustical treatments to address these issues are analyzed and recommended.
Reverberation Time Measurements
The Reverberation Time (RT60) is the time it takes for reverberation to decay 60 decibels. Speech intelligibility decreases with increasing reverberation time. In addition, discrete echoes between hard parallel surfaces can cause focusing, poor imaging and other acoustical issues. Recommended RT60 values for the two forums are in the range of 0.6 to 1.3 seconds in the 500 Hz and 1000 Hz frequency bands. The lower end of this range is desirable from a speech intelligibility standpoint.
The RT60 is measured by playing back pink noise (flat frequency response) bursts through a test loudspeaker and measuring the sound level decay with a Type I octave frequency band analyzer. Table 1 shows the results of the Forum 1 measurements. Gaps in the data were caused by extraneous noises occurring during the tests. A large data set was acquired to average out these gaps plus the expected spatial variation in the data. As can be seen from Table 1, the RT60 values at 500 Hz and 1000 Hz are 1.0 and 0.9 seconds, which already meet the design criteria.
Table 1. Measured Reverberation Time (RT60) Forum 1
Table 2 shows the results of the Forum 2 measurements. The RT60 values at 500 Hz and 1000 Hz are 1.9 and 1.2 seconds which are higher than the design criteria. The 2.6 second value at 250 Hz is particularly problematic. During the tests, significant flutter echoes and focusing due to the circular room were evident.
Table 2. Measured Reverberation Time (RT60) Forum 2
DCC developed a rudimentary acoustics model to calculate the RT60 for existing conditions for both forums. Interior dimensions and architectural finishes were included in the model. Figure 1 shows measured RT60 versus modeled RT60 with reasonable agreement between the two datasets (within an average of 0.1 seconds). Once it is determined that existing conditions are modeled accurately, a model of proposed treatments can be developed.
Figure 1. Forum 1 RT60 Measured, Modeled, and with Treatments
Even though Forum 1 already meets the design criteria, it would benefit acoustically by treating the projection screen wall with absorption treatment. DCC assumed that this wall would be covered with 2” thick Owens Corning 703 semi-rigid fiberglass panels. Several approaches are available to hide this material. Acoustically transparent fabric is often used in such applications and can be stretched to fit curved surfaces. If installed properly, these treatments can be made to look very similar to GWB or plaster surfaces. An example of this type of fabric can be found at http://www.clipsoceilingwall.com/how-it-works/acoustics/ or https://fabricmate.com/.
Figure 2 shows measured RT60 versus modeled RT60 for Forum 2. These two datasets are within an average of -0.2 seconds difference, but the agreement is not as good as the datasets in Forum 1. The likely reason for this is due to the very noticeable focusing effects in this space. Software-based ray tracing techniques would achieve better modeling results but are outside the scope of this present task. Regardless, it is evident that additional robust absorption treatments are needed in this space.
Figure 2. Forum 2 RT60 Measured, Modeled, and with Treatments
The primary culprit of focusing in this space appears to be caused by the vertical 9’ circular GWB area just below the ceiling (treated with spray-on Pyrok acoustical treatment). DCC assumed that this entire area would be treated with 2” thick Owens Corning 703 semi-rigid fiberglass panels and hidden with acoustically transparent fabric.
In addition, our treatment model also assumed that heavy drapes would cover the existing glass perimeter walls. For this analysis, we assumed cotton curtains (0.5kg/m2), draped to 75% area 130mm from wall. This treatment would provide needed absorption as well as eliminate the specular reflections off the glass.
The treated RT60 values at 500 Hz and 1000 Hz are 0.8 and 0.7 seconds which would meet the design criteria.
Glass Wall Sound Transmission
DCC employed a test loudspeaker and pink noise source to measure the Noise Level Reduction of the glass wall separating Forum 2 from the adjacent corridor. Based on the measurement data, it appears that this is not laminated glass (with high sound reduction performance), but rather ½” inch insulated glass. Figure 3 shows the sound level of a raised male voice (at 3’) attenuated by this glass wall compared with measured ambient sound levels in the hall. For this example, speech would not likely be discernable since it is below ambient sound levels.