How Effective is My Smoke Detector?

While most Audiology clinics concentrate their efforts on fitting hearing aids, little attention is paid to life saving alerting devices for the hearing impaired such as smoke/fire detectors.  The purpose of this article is to provide the clinician and consumer with comprehensive information about general commercially available alerting devices and why they are ineffective for individuals with hearing impairment in the Canadian market, based upon current research.  Further, we will arm the reader with information about the most effective devices for their hearing needs. 

Statistics

House fires are rated to be the fourth largest accidental killer and approximately 3000 individuals perish in residential fires each year in the United States.  Although kitchen fires are the most common of the residential fires, the majority of fatalities occur in fires that start elsewhere.  Most of the fatalities that are seen in residential fires occur between the hours of 11 PM to 7 AM.  In 40% of residential fire fatalities, smoke detectors were present and working.

According to the National Fire Protection Association, the leading causes of fire deaths in Canada include the following:

• Residential fires caused by smoking materials that ignite upholstered furniture in living room/family room areas at night time
• Smoking materials that ignite bedding in sleeping areas at night, and
• Cooking appliances that are left unattended at night.
(http://www.praxiom.org/nfpa.htm)

Despite the fact that these incidents start off as smoldering fires, they often develop into flaming fires.  According to the National Fire Protection Association, most people do not die from extended smoldering fires condition; they die from the lethal conditions related to flaming fires.  In cases where people died in fires even though the smoke alarm was activated, some explanations included:

o 13% suicides/homicides
o 21%  clothing fires
o 26% device not in area of origin
o 20% victim is physically challenged
o 16% alcohol a factor
o 1% victim re-enters the fire scene
o 1% unattended infants
o 2% unknown

Research has shown that in some fires where fatalities occurred and smoke alarms were present but did not activate 85% of the incidents occurred due to a dead or missing power source.
http://www.firesafetycouncil.com/english/pubsafet/safact.htm

Statistics from Ontario between the years of 1995 to 1997 indicated that fewer individuals die in home fires when a smoke detector is present and is functioning.  In homes where smoke detectors were present and activated, there are approximately 12 deaths per 1000 home fires.  In homes where there was no alerting device or the device failed to activate, the death toll is significantly higher at 17 deaths per 1000 home fires.  The studies failed to identify variables such as hearing loss.

According to the National Fire protection Agency, within the first 30 seconds, a fire will ignite and grow.  At 1:04 minutes, the fire will spread and smoke will fill the room.  At 1:35, the smoke layer descends and the temperature will exceed 190 degrees (F).  At 2:30 minutes, the temperature will exceed 400 degrees F.  At 2:48, smoke will pour into other rooms.  At 3:03, the temperature can exceed over 500 degrees F.  At 3:20, a safe escape from the fire is considered to be difficult.  At 3:41, flashover occurs and the temperature will reach up to 1400 degrees F.  At 4:33, flames will be visible from the exterior of the house.

Types of Smoke/Fire Alerting Systems

There are essentially two types of smoke detectors.  Photoelectric detectors comprise of 20% of sales and are most effective for slow, smoldering and smoky fires.  Ionization detectors comprise of 80% of sales and are most effective for flaming fires.

Ionization Smoke Detectors 

This alerting device uses a small amount of radioactive material which ionizes the air in the sensing chamber.  The chamber becomes a conductive which permits the incoming current to flow between two charged electrodes.  When smoke particles enter the chamber the conductivity of the air chamber decreases.  When this reduction is reduced to a predetermined level the alarm is activated.

Best to respond to flaming fires 

• Faster to respond to black/smoldering fires that involve “man-made” materials (PVC wiring, wall coverings, carpets) or fires that burn combustible materials rapidly and spread quickly (paper burning in a waste basket or a kitchen grease fire) which account for 70% of house fires.
• Generally slightly slower to respond to white or gray smoky fires that involve natural material (wood, cotton)
• Lowest cost and most commonly sold (statistics show that they comprise of 80-90% of those sold)
• Some models have up to a ten year battery life
• Lowest cost and most commonly sold
• Some models have a temporary silence feature that will allow silencing without removing the battery

Photoelectric Smoke Detectors

A photoelectric alerting system consists of a light emitting diode and a light sensitive sensor in a “sensing chamber”.  The presence of smoke particles that enter the chamber will affect the light beam which in turn sets off the alarm.

• Best to respond to white/gray slow smoldering  or slow-burning fires (ie. Cigarettes burning in couches or bedding) which account for 30% of house fires.
• Less prone to nuisance alarms (ie. Cooking)
• In an ideal world, installing both types of smoke alarms would be the most efficient at enhancing fire safety, if individuals are not hearing impaired.

Standards

The National Fire Protection Association (NFPA) does not test, label or approve the performance of fire/smoke alerting devices, but rather, recommends.  The current ANSI standard (ANSI S3.42-1990) (July 1996) indicates that smoke detectors must generate an Audible Emergency Evacuation Signal.  The standard specifies that a Temporal Three (T-3) pattern (a 3 second ½ second on phase followed by an alternating off phase and one second resting period for three minutes or 180 seconds).  According to this standard, the intensity of the signal must be 75 dBA at pillow level and 85 dBA at a distance of 10 feet.  In Canada, a T-1 signal is used in most commercially available units.   There are currently no frequency requirements.  All smoke detectors distributed in Canada are regulated under the Hazardous Products Act, and must meet performance requirements set out in standards developed by the Underwriters' Laboratories of Canada (ULC).   Fire/Smoke alerting systems fall under the CAN/ULC-S531 standard.  As part of these standards, smoke detectors must come with directions for installation, testing and maintenance. It is up to the manufacturers, importers and retailers of smoke detectors to ensure that their units comply with the existing ULC standards.   CAN/ULC-S531 is closely harmonized with the equivalent smoke alarm standards of the US, and is periodically verified through statistical analysis, research, fire modeling, and testing performed by experts throughout Canada and the world.  Yet, there remains no standard for frequency of the tone despite the wealth of research that indicates that a 3100 Hz tone is ineffective for individuals with hearing impairment. 

In addition to regulating smoke detectors under the Hazardous Products Act, Health Canada monitors the safety of smoke detectors in the marketplace. If concerns are raised about the safety of certain models, Health Canada and the Underwriters' Laboratories of Canada (ULC) will work together to correct the situation or to issue a safety advisory if appropriate.

Effective March 1, 2006, every home in Ontario must have working smoke detectors on every storey of the home and outside of sleeping areas.  This relatively new law applies to single family dwellings, semi-detached and town homes.  In addition, every smoke detector must be replaced every ten years.  Fines for non-compliance range from $235 -50,000 and tenants of rental units can be fined for intentionally disabling installed alerting units.

Research

Most normal hearing adults will wake up to a signal <75dBA, however; only 17% of normal hearing children will awaken to the same signal (Bruck& Horasan, 1995; Bruck & Brennon, 2001).  Several studies have investigated the most effective signal to awaken a hearing impaired individual.  Du Bois, Ashley, Klassen, & Roby (2005) investigated the efficacy of five different alerting signals (3100 Hz pure tone-the most commonly used alerting device on the market, a 450 Hz square wave signal, a continuous and intermittent shake system, and a 110 candela strobe light).  Subjects included a group of 34 normal hearing individuals, 45 hard of hearing individuals, and 32 deaf individuals, all adults.  The researchers concluded that the most effective alerting devices for individuals with normal hearing were any alerting system that was tested.  For the hearing impaired subjects, a low frequency tone with the intermittent bed shaker was the most effective combination.  For the deaf individuals, the vibrotactile (bed shaker) were most effective, particularly, the intermittent shaker.  Dubois et al. took into account the stages of sleep, however; it was unclear at what time stages the patients awoke.  If it exceeded 3:20 minutes, a safe escape would have been highly unlikely.

More recently, Bruck & Thomas (2009) investigated the efficacy of a 3100 Hz pure tone and a 420 and 520 Hz square wave device all presented in the T-3 format (current standard), a bed and pillow shaker, and a strobe light that also followed the T-3 format.  In this study, subjects were tested in a stage 3 or 4 slow wave sleep pattern (considered to be the most difficult to awaken from by a signal).  38 subjects with a mild sloping to moderately severe binaural sensorineural hearing loss were used.  With the audible alarm performance, 95% of the individuals woke up at maximal intensity levels.  With the 520 square wave device, 100% of the subject awoke.  Only 84% awoke to a 3100 Hz at maximal levels.  The strobe light was the most ineffective alert, resulting in 57% of the subjects waking when the alarm was at maximal levels, but only 27% awoke at benchmark testing levels.  89% of the subjects awoke with the bed shaker, and 97% with the pillow shaker at maximal intensity levels (compared to 80% and 84% respectively at benchmark levels).  The authors concluded that the most effective alerting device for individuals with a mild to moderately severe bilateral sensorineural hearing loss was the 520 Hz square wave smoke detector, while the least effective was the strobe light. 

So what are the best devices available for individuals with hearing impairment?

Most devices that are commercially available emit a 3100Hz pure tone, the least effective tone for individuals with mild to moderately severe sensorineural hearing loss.

Currently, there are two devices that emit a 520Hz signal; the Louden Low, and the Lifetone. The Louden Low is a hard wired device that would be typically is placed in a hallway, a disadvantage due to the alerting signal at a distance. The Lifetone is a regular smoke detector and alarm clock and has a bed shaker alerting system. It is designed to work in conjunction with existing smoke detectors and has the advantage of being proximal to an individual’s ear, as the system is portable.  It recognizes and reacts to a T-3 signal that is emitted from most currently available alarm products.  Since most new build houses have a carbon monoxide detector proximal to smoke detectors, there is an additional advantage, as the Lifetone would also be able to pick up the T-4 signal from either and convert the alert into a 520Hz square wave.  This makes the Lifetone an effective and efficient alarm clock, fire/smoke (either photoelectric or ionization) and carbon monoxide alerting system for individuals with hearing impairment in countries that use the T-3 signal.  Given the fact that research indicates that the 520 Hz square tone and bed shake system are the most effective alerting system for individuals with hearing impairment, the Lifetone is the system of choice, if a T-3 signal is used in alerting devices.  Unfortunately, Canada’s standard T-1 signal renders this device useless for hearing impaired individuals that reside in Canada.  The manufacturer is aware of this and is closely looking at making this device compatible for the T-1 signal and for Canadian households.

On a lighter note, Air Water Safety Service, a fire extinguisher company based in Kobe, Japan, has developed a smoke alarm for hearing impaired individuals that emits a strong odour of wasabi.  Two years in the making, the device emits a chemical compound allyl isothiocyanate and an LED alert also starts to flash when activated.  It  works in rooms roughly 50 square feet, hardly ideal for the typical Canadian household.  According to the Manufacturer, in unspecified tests (unknown as to the variables explored) nearly all individuals (hearing, hearing impaired, and deaf) woke up within two and a half minutes of activation.  The manufacturer also indicated that they have determined the optimal amount of airborne wasabi to waken people, yet not hurt their eyes in the process.  The current price of this device is approximately $560 US a redesign of this product may bring it closer to $225.  The company is also targeting sales at loud karaoke venues where at times, noise can drown out the alert of an alarm.  This device is likely not the best alerting detector for sushi restaurants.

Please do not hesitate to contact us if you require additional information or if you would like a demonstration of any of our products.

References:

http://cfaa.ca/industrynews.html

http://www.firesafetycouncil.com/english/pubsafet/safact.htm

http://www.tbs-sct.gc.ca/pol/doc-eng.aspx?section=text&id=13576

http://news.cnet.com/8301-17938 105-104439 19.html

Bruck, D. (1998) Non-awakening in children in response to a smoke detector alarm,  Fire Safety Journal, 32.  

Bruck, Dorothy and Brennan, Patricia (2001) Recognition of Fire Cues During Sleep. In: Second International Symposium on Human Behaviour in Fire, 2001, London.

Bruck, D., Horasan, M. (1995) Non-arousal and Non-action of Normal Sleepers in Response to a Smoke Detector Alarm, Fire Safety Journal, 25, 125-139. 

Bruck, D. & Thomas, Ian (2009).  Smoke alarms for sleeping adults who are hard-of-hearing:  Comparison of Auditory, Visual, and Tactile Signals.  Ear and Hearing, vol 30:1, 73-80.

Bruck, Dorothy and Thomas, Ian (2007) Waking effectiveness of alarm (auditory, visual and tactile) for adults who are hard of hearing. Project Report. The Fire Protection Research Foundation, Australia.

Bruck, Dorothy and Thomas, Ian and Ball, Michelle (2007) Waking effectiveness of alarms (auditory, visual and tactile) for the alcohol impaired. Project Report. Fire Protection Research Foundation, Vicotoria, Australia.

Thomas, Ian and Bruck, Dorothy (2008) Strobe lights, pillow shakers and bed shakers as smoke alarm signals. In: Ninth International Symposium on Fire Safety Science, 21-26 September 2008, Karlsruhe, Germany.