Optimizing speech intelligibility for sports stadiums.
In November 2013, Sound & Communications published Part 1 of “Optimizing Speech Intelligibility for Sports Stadiums,” which included an overview of acoustical simulation and modeling programs currently in use by architectural acousticians in developing optimal sound distribution and clarity for stadium audiences. This concluding installment provides specific examples of technology, techniques, systems and acoustic treatments engaged in assuring the highest quality speech intelligibility possible.
Arena Thun (www.arenathun.ch/de/willkommen) in Thun (as in moon) Switzerland is commonly referred to as the “Jewel” of Swiss Super League stadiums. With just over 10,000 seats, the arena’s compact design puts fans closer to the action than most other sports venues. Advantageously situated at the foot of a breathtaking Alpine panorama, its location, coupled with its coherent “boutique” architectural concept, coalesce to make it an exceptionally inviting destination.
In an equally well-conceived decision, the developers linked the stadium to the Panorama Center (www.panoramacenter.ch), an ultra-modern 530,000-square-foot shopping mall. This congenial proximity provides fans with a host of shopping, dining and entertainment options before and after sporting or cultural events. With fine (and fast food) restaurants, a spacious parking garage and five comfortably appointed VIP rooms of varying size, the facility is well suited for conferences, concerts, sporting events, trade shows, banquets, and a variety of business and social events.
Room & Electro Acoustics
Arena Thun’s developers and architects (Pool Architekten, Brügger Architekten, Itten-Brechbühl) wisely engaged the WSDG team at the early stages of facility design. Our mandate was to customize the acoustics for the stadium and all the arena’s public areas. The ground-up project began in 2010 with a series of meetings between the sponsors, architects and contractors. An extensive list of concerns was outlined, with a special focus on the materialization of the underside of the stadium roof (see Figure 3), which plays a major part in controlling the acoustical ambiance to which the fans are exposed.
By extending the roof over the entire audience seating area, the stadium provides an intense, inspirational communal experience, while maintaining a comfortable intimacy, excellent speech intelligibility and limiting noise intrusion into the residential neighborhood. The specified solution consists of a polyester-based, waterproof, self-adhesive material directly applied to the underside of the metal roof panels. The material has particular acoustical absorption characteristics while also controlling humidity and water condensation.
To deliver a clean digital signal distribution and control program for its audio evacuation system, WSDG developed an electro-acoustic approach comprised of Bosch Praesideo network and control units, Shure ULX microphones, Soundcraft Si Compact mixing desk, BSS Soundweb DSP, Crown CT amplification and JBL PD5200 and JBL AM 7215 stadium loudspeakers. This technology package ensured high quality speech intelligibility and targeted distribution of voice messaging, advertisement audio, game coverage and ambient music content throughout the stadium’s public and shopping areas. Permanent signal monitoring and key component and network connection redundancy make the system fully compliant with the strict fire safety and audio evacuation system requirements.
Stadium Electro-Acoustical Systems
How do we arrive at our recommendations for stadium electro-acoustical systems? As discussed in Part 1, a number of programs are employed to generate data based on the architectural plans for a proposed venue. We are particularly focused on requirements for two primary technical parameters: the STI (Speech Transmission Index) and the SPL (Sound Pressure Level), both of which must meet or exceed minimum criteria while also being consistent within certain limits in a given area. Exact specifications differ from country to country because each proprietary organization (FIFA, UEFA, Olympic Committee, etc.) may have slightly different goals driving its decisions.
The primary consideration is that the STI should be even on all audience seats. The mean value of the STI is not sufficient to guarantee this, so the standard deviation (basically describing how much the values differ from the mean value) must be considered. A requirement could look like this: The mean STI shall be equal or higher than 0.5 (or 50%, rated as “fair,”), while the standard deviation shall be equal or less than 0.05 in each audience block. STI values lower than 0.45 are considered “poor”; higher than 0.6 are rated as “good.”(see Figure 4).
SPL is a bit more complicated. The easy part is assuring even SPL distribution on the audience area: mean SPL in dB(A) ±3dB is a common requirement. This being said, a dB(A) value is an average over all frequencies. Therefore, if we lose all high frequencies in one sector but maintain the low-mid SPL, we might end up with the same dB(A) SPL for the next sector where the high frequencies are OK. And so, we must examine the SPL distribution at, for example, octave frequency bands throughout the venue.
In addition to SPL distribution, there is the question about the maximum SPL that should be possible in the arena. To guarantee fair intelligibility, the reinforced speech SPL should be at least 6dB or better, 10dB higher than the ambient noise. The latter can reach up to 100dB(A) during some phases of the game (e.g., right after a goal) in soccer stadiums. The chances that we would have to evacuate during such a phase are minimal, so we can lower this to a still very high 96dB(A) for evacuation purposes. This leaves us with a requirement of more than 106dB(A) SPL in the audience area for our PA system.
Although this is certainly possible with modern speakers and amplifiers, the downside is that speech intelligibility suffers from high loudness levels due to masking effects. So, although very loud, our PA system does deliver a higher signal/noise ratio (which improves STI), it also causes considerable masking effects that deteriorate STI. A careful balance must be found, designed and ultimately applied during the system calibration process.
If this criteria for STI cannot be met for a certain venue, there are typically two possible corrective approaches: One is to increase the directivity of the PA system and decrease the number of sources. This ensures that the sound energy is only projected to the audience. And, that at every audience position you basically hear only one speaker system (the one that’s precisely pointed at you) rather than two or three that are kinda looking your way, all with slightly different arrival times.
The second approach is to apply absorptive materials (typically at the underside of the roof) to reduce reflections and reverberation within the stadium. This also helps to maximize direct sound at the audience areas and reduce disturbing echoes.
The problem with the first approach is that, due to architectural and structural restrictions, we are not always able to position the speakers at the optimal locations. Also, the sound sources (loudspeakers) for these installations generally need to be quite large (and heavy) in order to produce enough directivity down to the low-mid frequency range. Space and structural load acceptance are frequently insufficient for such large installations.
The second measurement (room acoustics) can be compromised due to additional costs and because the atmosphere in the stadium usually suffers if the “acoustical ambiance” is too dry. That being said, today’s stadium acoustic/speech intelligibility improvement solutions are light years ahead of those of previous generations.
After signing a joint venture between Setha/Prosegur (www.sethaeletronica.com.br), a benchmark company in integrated security and communications headquartered in Madrid, Spain, and systems installation specialists Erhardt (http://erhardtprojetos.com), the sound system installation process for Rio de Janeiro’s largest stadium, the 78,838-seat Maracanã (Figure 5) began in July 2012, when the original design was revised.
A number of sophisticated studies produced extremely accurate projections, which enabled the final layout of PA clusters to be defined. Seventy-eight speakers were distributed on the catwalk in 26 individual clusters formed with three bi-amplified speakers each. Eight additional speakers aimed toward the field cover the entire area for the players, as required by FIFA.
The process of lifting and mounting the 26 clusters was performed by a six-member Erhardt field team and was accomplished over four days at the peak of Rio’s summer, when the average daily temperature in the stadium (with no roof installed at the time), exceeded 105oF. The clusters (Figure 6) were lifted using electrical chain motors and attached to the catwalk by a customized metal frame.
After installing the speaker clusters, the cables were distributed. The distances between the amp racks and the clusters stretched from about 230 feet to 525 feet. For this reason, cable thicknesses varied from 4x6mm2 up to one cable per speaker, and the maximum power loss was in the 15% range (about 1.5dB).
All of the speaker clusters and field speakers were bi-amplified, and each driver and woofer had its own channel. In this way, we achieved total control of the distributed sound, and adjustments could be applied for each box, individually, in groups or in sectors. The Electro-Voice NetMax DSP system processor enabled precise calibration of the sound system (Figure 7 and Figure 8).
At the conclusion of the installation, Bosch and Electro-Voice sent a team of two audio division experts to the site for five days to calibrate the system.
Six Racks Of Amps
For Maracanã’s internal areas, six racks of amplifiers were distributed around the stadium’s perimeter, where they feed more than 4000 ceiling speakers connected by more than 360,000 feet of cables. The sound system was designed to reproduce regular media programs, as well as critical evacuation and emergency messages. All of the components, including the infrastructure, are fully certified for this purpose, in order to provide a secure and efficient installation in accordance with FIFA requirements. Erhardt provided a squad of eight highly trained professionals for 150 days to complete this installation.
In many areas of the stadium, the ceiling had been completed prior to mounting the ceiling speakers. This required that the field team employ all their skills to run the cables without damaging the existing ceiling.
The audio system was designed for evacuation purposes and paging, as well as for regular media reproduction, throughout the stadium. WSDG developed a precisely coordinated grid to enable the numerous sectors and zones to be controlled from a Master Control Room.
The Bosch Praesidio network processor was installed in parallel with the NetMax processor to override potential human failure during an evacuation. After a brief interval (predetermined by the Fire Brigade), the system takes control of the evacuation routine established in accordance with FIFA International standards. Affected areas receive a “pre-alarm” signal. If the Fire Brigade does not reset this alarm, the affected area then receives evacuation instructions first. Following the evacuation, messages are sent to higher floors and then lower floors, followed by the ranks (stadium seating), in intervals required for proper evacuation, minimizing panic and controlling the rush of the spectators.
WSDG’s European associates, Michael Chollet and Thomas Wenger, worked closely with Setha/Prosegur, Technical Coordinator, Vinicius L. Salgueiro, to develop four 1054+ SF Giant Video Wall installations for Maracanã. The largest video displays built for any 2014 World Cup Stadium, each of these units is comprised of 50 individual 50-inch LCD screens (provided by China-based manufacturer, LianTronics, www.liantronics.com). Each of these three-story high Giant Screens weighs 11 tons (including custom-built mounting structures, access ramps and interior staircases to permit service). Modifications for the units’ prefabricated structures included ventilation (Figure 9: Note vents on top and bottom rows) and rain protection.
The installation process, including electrical and signal distribution, took three months and required the individual calibration of each 50-inch module. Audio and video are totally synchronized; including video distributed via a Sony Constellation IPTV System, controlled from the dedicated video control room (Figure 10) connected to 450 individual 50-inch screens dispersed throughout the stadium via Cat5 cables over 3750 Cisco network switches. The switches were configured with four encoders for future expansion and to allow multiple signals to be distributed around the stadium, such as for satellite, an Internal Video Channel and two additional sources, such as cable TV, DVD and/or internet video).
The decision to install four huge videowalls was based on two considerations: FIFA’s requirement that they accommodate eight lines of text, and the need for onscreen text to be visible up to a distance equivalent to 200x the size of the individual characters. A “traditional” two-screen configuration would have been insufficient to provide legible text for fans in seats as far as 200 meters (656 feet) from the screens.
The point that the installation and testing process was accomplished in a highly collaborative environment between the general contractors (Odebrech/Andrade Gutierrez) and Prosegur, and the installer (Erhardt) cannot be overemphasized. This mutually supportive interaction inspired everyone to perform at the highest level of commitment, and was essential to obtaining a truly world class installation. Of course, this attitude is to be expected in a stadium built to showcase the value of training and teamwork. And, it is also interesting to note that Maracanã Stadium is a finalist for the Archdaily 2014 Building of the Year Awards.
Dirk Noy, Partner/GM of WSDG’s Europe Office in Basel, Switzerland, has an M.Sc. in Physics, a Diploma in Experimental Solid State Physics from the University of Basel, and is a graduate of Full Sail Center for the Recording Arts. WSDG Partner/GM Brazil Office Renato Cipriano was graduated as a Civil Engineer from the University of FUMEC in Belo Horizonte, Brazil, and Full Sail Center for the Recording Arts.