When designing and installing voice evacuation and mass notification systems, intelligibility is required to be achieved. NFPA 72, National Fire Alarm and Signaling Code requires a 0.70 Speech Transmission Index [STI] minimum so that a message is intelligible and therefore understood.
When laying out such a system, the designer must indicate where the acoustically distinguishable space(s) (ADS) are to be within a protected property. Intelligibility is required within each ADS that is identified. Information on both speech intelligibility and ADS can be found within Annex D, Speech Intelligibility of NFPA 72, in both the 2016 and 2019 editions.
Although the NFPA 72 fire alarm and signaling code is updated every three years, the 2013 edition is the one that’s most widely adopted today as building owners move toward compliance of the newer versions.
More NFPA 72 Guidelines
Within Chapter 18 of NFPA 72, Notification Appliances, the designer is advised in 18.4.4.1 of the 2019 edition, that audibility is to be achieved by either producing a sound pressure of at least 15 dB above the average ambient sound level or 5 dB above the maximum sound level having a duration of at least sixty seconds.
For voice messaging, intelligibility is not created by blasting the message over the ambient sound pressure of the space. Paragraph 24.4.4.4 within Chapter 24, Emergency Communications Systems (ECS) requires this only for the alert tone.
This allowance is also found in 18.4.1.6. Within the annex to this paragraph, the following is stated:
“A voice signal must have sufficient audibility to result in intelligible communication.”
NFPA 72 also requires that, when a voice system is activated, all distracting sounds, such as machinery, audio systems and so forth be shut down. This is referred to as ambient noise within NFPA 72.
“18.4.4.5.3 Relays, circuits, or interfaces necessary to stop or reduce ambient noise shall meet the requirements of Chapters 10, 12, 21 and 23.”
Complying with NFPA 72
As I stated above, the designer of a system will identify sources of sound or noise that would prevent the voice message from being heard. If in a manufacturing plant it may be the shutting down of a production line.
If in a sports bar, it would be turning off the sound to the large screen monitors. In a movie theater, it would be turning off the projector and sound from the movie being shown.
There, of course, are many other examples, but the primary objective is to reduce or eliminate the sound pressure level within the protected premises so that the voice message may be first heard and second understood.
But what about background sound, such as sound masking systems? These systems are designed to increase the ambient sound within a space so that speech becomes unintelligible.
While these systems are put in place to prevent one person or persons from hearing the conversations of others, they can also prevent a message from a fire alarm or mass notification system from also being understood.
Typically, you would conclude that such a system may be installed within a SCIF or Secured Compartmentalized Information Facility where there is a need to keep information away from unauthorized individuals or states.
However, these systems are now being marketed and installed in the typical office environment, such as the open office.
ASTM Standard E-1130, Standard Test Method for Objective Measurement of Speech Privacy in Open Plan Spaces Using Articulation Index, provides a means in determining the privacy of conversations in open spaces. The Articulation Index or AI is a measurement in which the privacy of speech can be predicted:
- AI of .05 or less – Confidential Speech Privacy
- AI of .2 or less – Normal Speech Privacy
- AI greater than .2 – No Speech Privacy
Why NFPA 72 Is Important
The human brain is able to understand a message in which 20 percent or more of the message is heard. A sound masking system is designed to lower the AI within a space and thus decrease the ability for a conversation or message to be understood.
These systems, however, are also designed to be only in the background.
As such, while laying out a voice evacuation or mass notification system the designer would without too much difficulty identify sources of sound or noise that would interfere with the achieving the requirements of 18.4.4.1, consideration is most likely not being given to sources of background ambient sound that is producing an AI of .2 or less.
Sound masking can be either direct or indirect. While these systems are designed and installed by firms that specialize in these systems, similar localized effects can occur with background music or water features that may be in a building or lobby.
Sound masking systems maintain their desired results by filling a space with ambient unstructured sound that matches the frequency of the human voice, thereby covering or masking the speech and lowering intelligibility. In effect, sound masking makes a space quieter.
When laying out a voice evacuation or mass notification system, not only should the designer identify the sources of “noise,” but the sources of all ambient sound including the sources of sound masking.
A question that needs to be asked is “is there a sound masking system within the building, or is a sound masking system being considered to be installed?’ If the response is “yes,” then the requirements of 18.4.4.5.3 needs to be met for such a system.
Quiet and privacy is great for an office during normal operations, but intelligibility is required over both in the event of an alarm activation.
Shane Clary has more than 37 years of security and fire alarm industry experience.
He serves on a number of NFPA technical committees, and is vice president of Codes and Standards Compliance for Pancheco, Calif.-based Bay Alarm Co.
