This invention relates to life safety devices that include both a carbon monoxide (CO) sensor and a smoke sensor. In particular, the invention relates to improvements that enhance detection of fires and help to eliminate false alarms.
Smoke detectors, carbon monoxide detectors, and units that combine both smoke detection and carbon monoxide detection have found widespread use in residences and in commercial buildings. Smoke detectors provide early warning of fires, while carbon monoxide detectors can warn occupants of the buildup of deadly carbon monoxide that may be produced, for example, by a malfunctioning heating system, a wood burning stove or a fireplace.
Two types of smoke sensors are in common use: ionization smoke sensors and photoelectric smoke sensors. Ionization smoke sensors typically work better in detecting fast flaming fires, while photoelectric smoke sensors alarm more quickly to slow smoldering fires. Increasing the alarm threshold of an ionization smoke sensor can yield better sensitivity to slow smoldering fires, but the increased sensitivity tends to result in more false alarms.
There are some conditions under which a smoke detector can generate an alarm when no fire exists. Common examples of these types of false alarms are alarms triggered by cooking particles or smoke generated during the cooking of food. Another example is a false alarm triggered by shower steam that reaches a smoke detector. Alarms generated under these conditions are a nuisance and can also result in alarms being given less attention than they deserve when a real fire occurs.
A life safety device having a combination of a smoke sensor and a carbon monoxide sensor offers a reduction in false alarms through the use of an adaptively adjustable smoke alarm sensitivity. When the smoke sensor signal indicates presence of smoke at a smoke alarm threshold level, the smoke alarm threshold is adjusted to decrease smoke sensitivity. An alarm will be generated if the CO sensor signal indicates presence of carbon monoxide, or the smoke sensor signal indicates an increase in smoke to the adjusted alarm threshold, or the smoke sensor indicates continued presence of smoke at the initial smoke alarm threshold at the end of a timeout period. If the CO sensor signal indicates presence of carbon monoxide before the smoke sensor signal indicates presence of smoke, the smoke alarm threshold is adjusted to increase smoke sensitivity.
Smoke sensor 12 is an ionization smoke sensor that produces a smoke sensor signal S that is a voltage that varies as a function of smoke particles. As the number of smoke particles present in the ionization chamber of smoke sensor 12 increases, the voltage of smoke sensor signal S decreases.
CO sensor 14 may be a conventional CO sensor. The output of CO sensor 14 is CO sensor signal C. For example, in one embodiment CO sensor signal C is a current that varies nearly linearly as a function of parts per million of carbon monoxide molecules sensed by CO sensor 14. CO sensor signal C increases with increasing concentration of CO molecules.
Controller 16 is a microprocessor-based control that makes determinations of whether to activate alarm generator 18 based upon smoke sensor signal S and CO sensor signal C. In one embodiment, as a true combination alarm, in the case of CO detection, controller 16 maintains a carbon monoxide alarm threshold COT. When the integrated CO sensor signal C reaches alarm threshold COT, controller 16 causes alarm generator 18 to produce a CO alarm.
In the case of smoke/fire detection (in either a dual function or single function embodiment), controller 16 uses both smoke sensor signal S and CO sensor signal C as a part of the smoke alarm determination. Controller 16 uses a CO/smoke alarm threshold CT and an adjustable smoke alarm threshold ST to make a determination of whether to cause alarm generator 18 to produce a smoke alarm.
One problem encountered with smoke detectors is a tendency to generate a false alarm as a result of cooking particles or smoke generated during cooking. Other sources of false alarms can be hot water running in a shower that generates steam, and dust particles. Cooking particles, steam, and dust particles can cause a change in the output of smoke sensor 12 and potentially cause a false alarm.
The use of an adjustable smoke alarm threshold ST, which changes sensitivity to smoke based upon both smoke sensor signal S and CO sensor signal C, can reduce false alarms while increasing the ability of device 10 to detect slow smoldering fires. The adjustable smoke alarm threshold makes use of several observations. First, fast burning fires typically result in a fast buildup of smoke particles. Second, typical causes of false alarms (cooking, steam, and dust particles) normally do not generate much, if any, CO. Third, a smoldering fire will have both smoke and CO present in detectable amounts, with the CO/smoke alarm threshold CT being reached well before typical smoke alarm thresholds.
Smoke alarm state diagram 20 includes five states: Normal Standby state 22, Smart Hush state 24, Smoke Alarm state 26, Normal Hush state 28, and Smoke Sensitive state 30. As long as signal S from smoke sensor 12 and signal C from CO sensor 14 do not indicate a fire or a carbon monoxide danger, controller 16 remains in standby state 22.
If smoke sensor 12 senses smoke particles so that smoke sensor voltage S is less than a calibrated initial threshold X, controller 16 transitions from Standby state 22 to Smart Hush state 24. Upon entering Smart Hush state 24, controller 16 lowers the current smoke threshold ST by a set amount, meaning that it will require more smoke to cause device 10 to go into alarm. In the example shown in
Controller 16 will stay in the Smart Hush mode as long as smoke sensor 12 continues to sense some smoke, but CO sensor 14 has not sensed carbon monoxide at a level greater than the CO/smoke alarm threshold CT (which may be, for example, in a range of about 12 ppm to about 24 ppm). As shown in
Two conditions can cause controller 16 to return to Standby state 22 from Smart Hush state 24 without any alarm having been generated. First, if during the timeout period the level of smoke has decreased so that smoke voltage S is greater than X+B, controller 16 returns to Standby state 22. Second, if at the end of a timeout period (e.g. about 8 minutes), the smoke level has decreased so that the smoke sensor voltage S is greater than the initial threshold ST=X, controller 16 will return to Standby state 22. In either case, the change in smoke level during the timeout period indicates a temporary situation, caused, for example, by cooking food, rather than by a fire.
While controller 16 is in the Smart Hush state 24, controller 16 continues to look for two events that indicate a fire condition: (a) continued buildup of smoke or (b) presence of carbon monoxide above the CO/smoke alarm threshold level (CT). As shown in
If CO sensor 14 senses more than threshold level CT of carbon monoxide (C>CT) during Smart Hush state 24, controller 16 enters the Smoke Alarm state 26 and causes alarm generator 18 to produce a smoke alarm. If smoke particles are present so that sensor signal S is between X−A and X+B, and carbon monoxide is sensed at or beyond threshold level CT during Smart Hush state 24, this indicates that a fire is present, and not just a cooking problem, dust, or steam from a shower. Carbon monoxide is always present in real fires. Although some carbon monoxide is present when foods are burned or cooked well done, the level of carbon monoxide is usually at amounts that are below threshold level CT. Therefore, when device 10 senses more than level CT of carbon monoxide at the same time that it is sensing smoke particles, there is a basis for generating the smoke alarm.
If smoke sensor signal S is less than X at the end of the timeout, the smoke particles have not dissipated during the Smart Hush period defined by the timeout. Controller 16 transitions to the Smoke Alarm state 26 and causes alarm generator 18 to generate the smoke alarm.
Once controller 16 is in Smoke Alarm state 26, it will remain in that state until (a) smoke reduces to the level where smoke signal S is greater than X+F (which causes a transition to Normal Standby state 22) or (b) a reset button is pushed (causing a transition to Normal Hush state 28).
When Normal Hush state 28 is active, the current smoke threshold is reduced further to ST=X−G. The alarm generated by alarm generator 18 is silenced as a result of a reset button pressed and will remain silenced during the Normal Hush state 28 until smoke voltage S is greater than X+F (indicating smoke has dissipated), or a timeout of the Normal Hush period has occurred, whichever is earlier. In either case, controller 16 will return to Standby state 22.
If smoke continues to build up so that smoke sensor signal S decreases to the point where S is less than X−G, controller 16 exits Normal Hush state 28 and returns to Smoke Alarm state 26. Upon reentry in Smoke Alarm state 26, controller 16 again activates alarm generator 18.
In some cases, carbon monoxide at a level greater than threshold CT could be sensed by CO sensor 14 before smoke has built up to the point where smoke sensor signal S reaches initial threshold level ST=X. In that case, controller 16 will transition from Standby state 22 to Smoke Sensitive state 30. While in Smoke Sensitive state 30, controller 16 increases smoke threshold ST above the initial threshold to ST=X+H. Since smoke voltage S decreases as smoke increases, the increase in smoke threshold ST makes controller 16 more sensitive to the presence of smoke. If smoke is present at a level so that S is less than X+H, controller 16 will transition to Smoke Alarm state 26.
As long as the amount of smoke does not satisfy the more sensitive threshold ST=X+H, controller 16 remains in Smoke Sensitive state 30 as long as carbon monoxide signal C is greater than CT. As soon as the carbon monoxide level decreases below threshold CT, controller 16 returns to Standby state 22.
Adjustments A, B, F, G, and H to smoke threshold ST are voltage adjustments that correspond to a sensitivity adjustment in picoAmps on the sensitivity scale used by Underwriters Laboratories (UL) to test and characterize sensitivity of smoke detectors. In one embodiment, A is a sensitivity adjustment of 7.0 picoAmps; B is a sensitivity adjustment of 3.5 picoAmps; F is a sensitivity adjustment of 7.0 picoAmps; G is a sensitivity adjustment of 14.0 picoAmps; and H is a sensitivity adjustment of 7.0 picoAmps. In other embodiments, some or all of the adjustments may differ from these values.
Ionization smoke sensors typically work better in detection of fast flaming fires, while photoelectric smoke sensors tend to work better with slow smoldering fires. By using carbon monoxide sensor 14 as part of the smoke alarm determination, and adaptively adjusting smoke alarm threshold ST, as illustrated in
Although the present invention has been described with reference to preferred embodiments, workers skilled in the art will recognize that changes may be made in form and detail without departing from the spirit and scope of the invention.
Number | Name | Date | Kind |
---|---|---|---|
4383251 | Perelli et al. | May 1983 | A |
4688021 | Buck et al. | Aug 1987 | A |
4871999 | Ishii et al. | Oct 1989 | A |
4975684 | Guttinger et al. | Dec 1990 | A |
5084696 | Guscott et al. | Jan 1992 | A |
5159315 | Schultz et al. | Oct 1992 | A |
5168262 | Okayama | Dec 1992 | A |
5172096 | Tice | Dec 1992 | A |
5376924 | Kubo et al. | Dec 1994 | A |
5471194 | Guscott | Nov 1995 | A |
5523743 | Rattman et al. | Jun 1996 | A |
5552763 | Kirby | Sep 1996 | A |
5557262 | Tice | Sep 1996 | A |
5592147 | Wong | Jan 1997 | A |
5691703 | Roby et al. | Nov 1997 | A |
5691704 | Wong | Nov 1997 | A |
5767776 | Wong | Jun 1998 | A |
5798700 | Wong | Aug 1998 | A |
5801633 | Soni | Sep 1998 | A |
5831537 | Marman | Nov 1998 | A |
5945924 | Marman et al. | Aug 1999 | A |
5966077 | Wong | Oct 1999 | A |
5969604 | Tice | Oct 1999 | A |
6057549 | Castleman | May 2000 | A |
6064064 | Castleman | May 2000 | A |
6107925 | Wong | Aug 2000 | A |
6153881 | Castleman | Nov 2000 | A |
6166647 | Wong | Dec 2000 | A |
6229439 | Tice | May 2001 | B1 |
6346880 | Schroeder et al. | Feb 2002 | B1 |
6392536 | Tice et al. | May 2002 | B1 |
6426703 | Johnston et al. | Jul 2002 | B1 |
6515283 | Castleman et al. | Feb 2003 | B1 |
6518574 | Castleman | Feb 2003 | B1 |
6597288 | Amano et al. | Jul 2003 | B2 |
6753786 | Apperson et al. | Jun 2004 | B1 |
6788197 | Thuillard et al. | Sep 2004 | B1 |
6788198 | Harshaw | Sep 2004 | B2 |
6856252 | Pfefferseder et al. | Feb 2005 | B2 |
6897774 | Costa et al. | May 2005 | B2 |
6956473 | Hanood | Oct 2005 | B2 |
6967582 | Tice et al. | Nov 2005 | B2 |
6979260 | Liu | Dec 2005 | B2 |
7019657 | Lovell et al. | Mar 2006 | B2 |
7034701 | Rose-Pehrsson et al. | Apr 2006 | B1 |
7142105 | Chen | Nov 2006 | B2 |
7319403 | Woodard et al. | Jan 2008 | B2 |
7327247 | Tice | Feb 2008 | B2 |
7336168 | Kates | Feb 2008 | B2 |
7528711 | Kates | May 2009 | B2 |
7532117 | Barrieau et al. | May 2009 | B2 |
20020118116 | Tice et al. | Aug 2002 | A1 |
20050200492 | Woodard et al. | Sep 2005 | A1 |
20060273896 | Kates | Dec 2006 | A1 |
20070030156 | Schlager et al. | Feb 2007 | A1 |
20090051552 | Chabanis et al. | Feb 2009 | A1 |
Number | Date | Country |
---|---|---|
0418410 | Mar 1991 | EP |
WO 9641318 | Dec 1996 | WO |
WO 9727571 | Jul 1997 | WO |
WO 2006131204 | Dec 2006 | WO |
Number | Date | Country | |
---|---|---|---|
20080211678 A1 | Sep 2008 | US |