GB2358245A

GB2358245A - Photo-acoustic gas sensor

Info

Publication number: GB2358245A
Application number: GB0025434A
Authority: GB
Inventors: Vincent Y Chow
Original assignee: Pittway Corp
Current assignee: Pittway Corp
Priority date: 1999-10-21
Filing date: 2000-10-17
Publication date: 2001-07-18
Also published as: DE10051691A1; GB0025434D0; CA2323675A1

Abstract

A gas sensor comprises two chambers, a first internal transmissive sealed chamber 7-4 housing a reference sample of at least one specific gas to be sensed and a pressure transducer 7-6 7-7 7-8 such as a microphone 7-6, and a second reflective sensing chamber 7-3 which has inflow and outflow 7-14 7-15 of ambient gasses to be investigated. At least two paths of infrared light are directed such that one path A or B passes through the gasses in both the sensing chamber and the internal chamber, whilst a second path (from source 7-2) only passes through the internal chamber, and preferably these two paths are supplied by separate spaced apart light sources 7-1 7-2 which are more preferably operated in alternating fashion. The light sources may be placed behind filters 7-18. The light is absorbed by the gasses in the two chambers according to their composition. Preferably the reference gas is carbon dioxide. The gas sensor is suitable for use in a combined smoke and carbon dioxide sensor unit (see figures 8 9a,b 10a,b).

Description

2358245 GAS DETECTORS

Field of the Invention:

The invention pertains to gas detectors. More particularly, the invention pertains to such detectors wherein ambient gas related signals are generated using single closed cell gas sensors.

Backizround of the Invention:

Smoke detectors have been recognized as being useful in providing early warnings of fire conditions. It has also been recognized that there can be a benefit in combining different types of sensors into a single detector. For example, smoke and carbon monoxide or carbon dioxide sensors have been combined in a single detector. Such combinations can, depending on configuration and on fire type, improve detector performance.

Known types of smoke detectors include photoelectric, ionization-type or projected beam smoke sensors individually or in combination. Various types of gas sensing technologies including solid-state, electrochemical and absorption have been used to implement gas sensors.

Another known gas sensing technology is based on photoacoustic phenomena. Every gas will absorb light energy. Each gas is also unique in the spectrum of light that it will absorb.

A microphone diaphragm or another pressure transducer can detect a pressure wave that results from absorption and convert it into an electrical signal.

Fig. I is a graph illustrating the absorption spectrum of clean air in the range of carbon dioxide and water. Carbon dioxide is strongly absorbed in a 4.1 - 4.5 micron range.

To benefit from the photoacoustic effect, a target gas must be held inside a volume that permits a pressure build up due to gas expansion. This is achieved either by sealing the target gas inside a vessel, a closed cell, or by impeding the gas flow into surrounding air with a restrictive membrane. The vessel technique is called "closed-cell" photoacoustics and the membrane technique is called "open-cell" photoacoustics.

Closed-cell photoacoustic designs have the greatest sensitivity to gas concentrations but they are large, complex and expensive, often costing $5000 or more. Such instruments are usually bench type equipment intended for laboratory usage.

Open-cell designs, on the other hand, can be quite small and low in cost. They unfortunately have poor sensitivity and drift. They are useful in applications where signals can be averaged over long periods of time.

A light pulse directed into a gas filled closed-cell (through an optical window) generates a pressure wave inside that cell. The strength of this pressure wave is proportional to the light energy absorbed. Carbon dioxide will absorb peak light energy at a wavelength of 4.3 microns.

If light energy at 4.3 microns is removed from this light (by carbon dioxide in the atmosphere) before it can enter the cell, the resultant pressure wave that is generated is proportionally reduced. This is the basic principle behind photoacoustic gas detection.

The light beam is designed to pass through a sample chamber before it is allowed to enter the closed cell. If carbon dioxide is present in the sample chamber, it will absorb some of the 4.3 micron energy from this light pulse. As a result, less 4.3 micron light energy will enter the carbon dioxide filled closed cell.

Less carbon dioxide pressure is therefore produced and a lower electrical output signal results. This output signal will increase with less carbon dioxide in the sample chamber and decrease when more carbon dioxide is present.

For carbon dioxide concentrations between 0 to 2,000 ppin, the ppm to voltage relationship is fairly linear. At higher concentrations, the signal attenuation plot follows an exponential and is governed by Beer's Law.

Known light-based gas detectors on the market today are based on "open-cell" photoacoustics ornon-dispersive infrared (NDIR). The latter is much more popular because it represents a well understood technology and favored because of its lower cost.

An NDIR type sensor is illustrated in Fig. 2. NDIR is also an optical absorption technology. A light source 49 with reflector 48 is pulsed by an external drive circuit. This light enters a sampling chamber 50 enclosed, at least in part, by a permeable membrane 5 1.

Instead of sensing pressure, NDIR uses a pyroelectric element 53 that is sensitive to changes in heat. Light passing through the sampling chamber 5 0 will increase or decrease in energy based on the degree of gas absorption. This variation in light energy can then be sensed using a pyroelectric; element 53.

If a specific bandpass filter 52 is placed in front of the pyroelectric element 53, the response can be tailored to match specific ranges or windows of wavelengths. For example, a carbon dioxide filter located in front of the element 53 results in a carbon dioxide gas sensor. NDIR detectors, however, are not as sensitive as photoacoustic, detectors due to their inherently poorer signal to noise ratio.

High performance photoacoustic instrumentation incorporates a dual beam, closed cell structure with ratiometric processing. This technology offers up to 15 times more sensitivity than NDIR based gas sensors especially in the presence of interference gasses. Lab instruments with photoacoustic technology are routinely used to benchmark performance of other gas detecting products.

Fig. 3 illustrates a dual beam photoacoustic detector for sensing carbon dioxide, for example. A common light source 33 illuminates two separate optical paths A and B. This design contains four chambers.

Light from source 33 is collected by a common reflector 32 and directly illuminates two chambers 36 and 37. Chamber 37 is the sample gas chamber having an inlet 44 and an outlet 45. A gas to be sensed flows into and through chamber 37.

Chamber 36 is sealed and filled with nitrogen. It serves as a reference chamber. Nitrogen does not absorb any light in the wavelength region of interest.

Light energy from source 33 enters both chambers 36 and 37 before entering carbon dioxide filled chambers 40 and 41. Optical filters 46, 47, 38, 39 are all carbon dioxide wavelengths selective with respect to incoming radiant energy.

Carbon dioxide in chamber 37 will absorb energy from the light entering chamber 37. The degree of absorption is a function of the carbon dioxide density in chamber 37. The remaining light energy then enters detector chamber 41.

The nitrogen in chamber 36 does not absorb the incoming radiant energy. That incident light then enters chamber 40. It is then converted in chambers 40, 41 to pressure waves via gas expansion.

The capacitive membrane 42 flexes in the direction of chamber 41 or depending on the vector of this differential pressure. This flexing varies the membrane capacitance which can be sensed by an external AC circuit. The peak amplitude sensed is inversely proportional to the amount of carbon dioxide in sample chamber 37. A chopper wheel 34 breaks the light beam up into pulses to permit AC amplification and synchronization.

Photoacoustic pressure signals will normally be the same for optical paths A and B with no carbon dioxide present in the sample cell 37. Any carbon dioxide in the sample cell will cause signal attenuation at 4.3 microns since 4.3 micron energy will be absorbed from the light entering chamber A.

As a result of the absorption, the corresponding sensing chamber 41 will experience a lower pressure while the pressure in reference chamber 40 remains unchanged. This pressure imbalance forces the membrane to flex into chamber 4 1. The resultant amplifier output will now show an increase in PPM of carbon dioxide present in sample chamber 37. This technology is complicated and involves integration of many sensitive components.

Even though the dual bearn closed cell photoacoustic detector of Fig.

3 is very sensitive; it suffers from many drawbacks that limit its applicability.

Among other problems, such detectors configured for a single gas can cost more than $10,000. As a group, they are large, heavy, and require special filaments plus accurate temperature control. They are also sensitive to vibration, needs forced-air cooling and often incorporate a moving chopper wheel assembly. Synchronization is critical for proper operation and this instrument must be "zeroed" before every analysis. These major drawbacks have restricted such units to laboratory or manual applications.

Because ofsuch drawbacks, NDIR-type sensors have gained popularity in gas detection. They tend to be simpler in structure. NDIR, however, has its own challenges. The major commercial obstacles are speed of response, calibration and high cost.

Much of the cost is focused on the bandpass filter(s) and infrared detector(s) assemblies. Since NDIR technology senses the energy loss in a light beam through a narrow bandpass filter, the quality of this filter is extremely important for performance. The same is true for the pyroelectric infrared detector used. These are the two primary factors that keep NDIR costs high.

A "state-of-the-art" NDIR carbon dioxide detector design recently introduced into the market incorporates an integrated reference. To combat inaccuracy and component aging problems over time, a ratiometric technique (vs.

single value readings) is used to determine gas signal attenuation. To perform this math, however, a reference signal is needed that does not respond to the target gas.

This can be accomplished by integrating a tunable filter into the sampling chamber.

Using a 50% duty cycle, the filter is constantly switched electrically between carbon dioxide and non-carbon dioxide wavelengths. The ratio of these two signals is then used to determine carbon dioxide concentration.

In another commercially available design, two pyroelectric detectors are used each with its own filter, one to pass carbon dioxide and the other not. The ratio of these two signals is then used to determine gas concentration. Both of the above units sell for over $250 in volume quantities (single unit pricing is around $450). This cost is much lower than the detector of Fig. 3 but it is still too high for lower priced markets.

Photoacoustic technology, on the other hand, uses the gas itself as the sensor. The type of gas in the detection chamber is used to detect the presence of the same type of gas in the sample chamber. This perfect match of absorption signature at all wavelengths (and therefore rejection of interferences) is the reason why a photoacoustic approach offers much higher sensitivities than an NDIR approach.

One known "open-cell" photoacoustic carbon dioxide gas sensor is on the market. An "open-cell" design illustrated in Fig. 4 includes a single sample chamber 3 0 having three components. A permeable filter membrane 29 allows gas to diffuse into the chamber but prevents dirt ingress.

A microphone 31 is integrated into the body 28 to sense gas pressure signals. A bandpass filter 27 allows selected external light energy to enter the cell.

Implemented as a carbon dioxide detector, pulses of light enter the chamber through the 4.3 micron bandpass filter 27. This light is absorbed by the concentration of carbon dioxide inside the chamber that diffuses through membrane 29 from the outside air.

The energy absorbed by the carbon dioxide in chamber 3 0 at the instant of the light flash is transformed into a pressure wave that is detected by the microphone 3 1. The amplitude of this signal is proportional to the gas concentration inside the chamber.

Diffusion membrane 29 introduces a time delay between carbon dioxide concentrations inside the chamber and carbon dioxide concentrations in the air.

This delay can be up to 15 minutes for large carbon dioxide swings from 300 ppm to 2,000 ppm.

Problems with "open-cell" designs are known. They often require long integration times to overcome poor signal to noise ratios. They also require a bright light source (and therefore more power) to excite low concentrations of carbon dioxide or other selected gas in air. Since there is no reference signal, the detector is prone to temperature sensitivities and component drifts over time.

An improvement over the design of fig. 4 was disclosed recently. This detector is illustrated in Fig. 5. A light source 17 and reflector 16 illuminate sample chamber 21 and reference chamber 25 through bandpass filters 19 and 18, respectively. Microphones 23, 24 are each located in a respective chamber.

Gas from the outside air permeates into sample chamber 21 through a membrane 20. The pressure waves developed are detected by microphone 23 and ratiornetrically compared to output signals from microphone 24 to cancel out component effects. Unfortunately this design exhibits microphone imbalances that can vary up to 70%, poor signal to noise ratios, and long time constants.

Notwithstanding the various known types of gas detectors, there still continues to be a need for lower cost, higher reliability and less complex detectors than now known. Preferably such detectors would require relative low power, exhibit relatively high sensitivity and an improved signal-to-noise ratios than comparably priced gas detectors.

Summary of the Invention:

A single closed chamber photoacoustic gas detector includes an optically transparent, closed chamber filled with a type of a gas to be sensed. The closed chamber is located at least adjacent to a portion of a sensing region into which gas to be sensed flows.

A sensing source of radiant energy, which could be a laser diode which emits light having a wavelength which is known to be absorbed by the gas to be sensed, injects radiant energy into the sensing region. In other embodiments, incandescent or gas discharge sources could be used.

The injected radiant energypasses through the gas in the sensingregion wherein a portion of the energy therein is absorbed by the gas of interest. The radiant energy continues into the closed chamber.

A microphone is located within the closed gas chamber. Incoming radiant energy which has passed through the sensing region and through the closed chamber is converted to an acoustic signal therein and produces an electrical signal indicative thereof.

A reference source is positioned adjacent to the closed chamber. The reference source, which could be implemented as a light emitting diode, a laser diode or any other type of source with an emission frequency having a wavelength in the region of absorption of the gas of interest, injects radiant energy into the closed chamber. The injected radiant energy in turn produces a reference electrical signal at the output of the microphone.

The sensing source and the reference source can be pulsed alternately at a preselected frequency. A ratio of the two signals could be formed for purposes of minimizing component variations and aging effects. Filters responsive to the wavelength of interest can be interposed between each of the sources and the respective adjacent chamber for improved performance.

In yet another aspect, a sensing chamber can be formed of two, connected, substantially identical housing portions which define an internal sensing volume having an ellipsoid profile. The internal walls of the sensing region can be made reflective by a deposited reflective metal surface such as a chrome surface.

A closed gas supply tube is located at a tapered end of the chamber extending thereunto. The gas supply tube contains a quantity of a gas of interest to be sensed. An acoustic transducer or microphone can be located at an end of the closed tube, displaced from the sensing region.

A sensing source is carried in the housing at an end of the sensing region displaced as far as possible from the gas tube. The source can be triggered repetitively at a predetermined frequency whereupon it injects pulses of radiant energy, of a selected wavelength into the sensing region. The injected pulses are directed by the reflective surfaces toward the closed gas chamber at the far end thereof. Gas input and output ports can be provided into the sensing region. The ports can be provided with appropriate filters to exclude dust, airborne particulate matter, insects and gases not of interest.

A reference source can be carried by the housing adjacent the glass tube for purposes of establishing a reference signal. A ratio can be formed of the sensed and reference output signals.

In yet another aspect, instead of a glass chamber, a closed radiant energy transparent plastic chamber can be used as a container of a suitable gas of interest. It will also be understood that optical filters can be located adjacent to the sensing source and to improve detector performance.

A multisensor detector combines a smoke signal, from a smoke sensor, with gas dynamic profile signal, from a gas sensor. In one embodiment, a carbon dioxide sensor, such as a photoacoustic-type, generates an output signal. A degree of threat parameter can be derived from the gas output signal and combined with the signal from the smoke sensor to make a fire determination. In another aspect, the signal from the gas sensor can also be combined with the smoke signal and its rate of change.

A combination detector includes a common sensing region with a housing. A smoke sensor, photo or ion-type shares the sensing region with a single sealed gas sensor of a type described above. Signals from the sensors can be processed locally, remotely or both locally and remotely. The configuration and arrangement of the sealed gas cell can be consistent with the form factor of the detector's housing.

Numerous other advantages and features of the present invention will become readily apparent from the following detailed description of the invention and the embodiments thereof, from the claims and from the accompanying drawings.

Brief Description of the DrawiWs:

Fig. I is a graph illustrating an infrared absorption spectrum; Fig. 2 is a diagram of a prior art non-dispersive infrared gas detector;

Fig. 3 is a diagram of a prior art photoacoustic gas detector;

Fig. 4 is a diagram of a prior art single chamber, open cell, photoacoustic detector; Fig. 5 is a diagram of a prior art dual chamber, open cell, photoacoustic detector; Fig. 6 is a side sectional view of a closed cell photoacoustic gas detector in accordance with the present invention; Fig. 7A is a perspective view of a portion of a housing for a gas detector of the type illustrated in Fig. 6; Fig. 7B is a perspective view of another embodiment of the detector of Fig. 6; Fig. 8 is a block diagram of another detector in accordance with the present invention; Figs. 9A, 9B illustrate different packaging configurations for the detector of Fig. 8; and Figs. 10A IOB illustrate different gas detector forin factors for the detector of Fig. 8.

Detailed Description of the Preferred Embodiments:

While this invention is susceptible of embodiment in many different forms, there are shown in the drawing and will be described herein in detail specific embodiments thereof with the understanding that the present disclosure is to be considered as an exemplification of the principles of the invention and is not intended to limit the invention to the specific embodiments illustrated.

Figs. 6 and 7A and B illustrate a photoacoustic sensor in accordance herewith. A closed glass chamber I I - 1, for example a test tube, made from soft glass that also has good optical and resonant properties when operated at 4.3 microns contains a quantity of a gas to be sensed. The glass is selected to be sufficiently transparent at the wavelengths, ofinterest and capable ofcontaining the selected gas, as discussed below.

Alternately, an appropriate plastic could be used instead of glass without departing from the spirit and scope hereof. In addition, the closed chamber can be formed in a variety of shapes.

An electret condenser microphone 11-2 is epoxy sealed or integrated 11 -4 into the open end of the tube 11 - I under an atmosphere or more of the gas to be detected. Once sealed, the tube 11 - 1 forms the basis of a one chamber gas photoacoustic detector. If the tube contains carbon dioxide, that will be the sensed gas.

The sample chamber 11-5 enclosing the test tube is shaped with an ellipsoid profile to maximize light ray travel 11-7 between a sensing source 11 -8, filter 11 - 10 and the tube 11-2 to increase sensitivity. The flash frequency of source 11 -7 is 2.8 Hz, which is the- preferred operating frequency for the illustrated configuration. Those of skill in the art will understand that other frequencies might be preferable with other configurations.

With reference to Fig. 7, sample chamber 11-5 can be implemented with two substantially identical molded top [ 11 -A (not shown)] and bottom 11 -B housing elements that can be sonically welded or heat staked together to form the closed sensing volume 11-5C.

When the lower sample chamber section I I -5B is mated to the top half I I -5A outside air will flow into the chamber I I -5C via ports 11 - I 2A and 11 - 12B.

A second light source 12-1 and filter assembly 12-2 are positioned adjacent to the base of the sensor tube I I - 1. This light source 12-1 provides a reference function for ratiometric operation.

Sources 11 -8 and 12-1 are time division multiplexed with a 10 second period. Source 11-8 is ON for 10 seconds flashing at 2.8 Hz, source 12-1 is energized, then source 12-1 and so forth. Other periods could be used.

Those of skill will understand that a variety of radiant energy sources could be used without departing from the spirit and scope of the present invention.

These include solid state sources, laser diodes or light emitting diodes, incandescent sources or gas discharge sources. The details of the source(s) are not a limitation of the present invention.

The closed chamber design is also extremely effective in isolating external sounds from the microphone. Fundamental vibrations mechanically transmitted into the assembly can be filtered out by the electronics and supporting software. Using this approach, only one microphone and one sampling chamber are required for ratiometric operation. This structure substantially reduces design complexity and associated costs.

With carbon dioxide sealed inside the detection chamber 11-1, (assuming that to be the gas to be sensed), the gas sample flow rate into the chamber 11-5c can be very fast or very slow. It can be optimized to match an application. It is not necessary to delay the pressure decay in the sampling chamber (as required with "open-cell") to permit microphone detection.

The sealed detection chamber I I -I can also be filled with a blend of two or more gases for multigas response. In such an application, the reference signal serves as a common signal for calculating concentrations. Another sensing source I 1-8a and filter I I - I Oa are required for each additional gas. The low profile ellipsoid chamber 11-5 accommodates numerous light sources (numerous gases) using time division multiplexed sensing.

The. "closed-cell" design produces excellent signal amplitudes. This means less complexity for support electronics to process the signal. "Closed-cell" photoacoustic detection permits the use of wide tolerance filters (11 - 10) because the target gas inside the closed chamber (e.g., carbon dioxide) is a perfect filter for the gas of interest (also carbon dioxide). This result is due to perfect spectrum matching in absorption profiles made possible only by using the target gas to sense itself thereby achieving maximum rejection of other gas species.

Open-cell designs do not have this advantage because interference gases (such as water in the sample chamber) can easily be excited if filter parameters are not narrow and tightly controlled. The tighter the specifications of the interference filter, however, the higher is the cost. A wide band filter costs, for example, $0.20. A tight narrow bandpass filter is $3.00 or more.

In summary, the present photoacoustic sensor eliminates costly components yet results in a high sensitivity gas detector. It retains many of the benefits of a photoacoustic dual beam instrument. It preserves the power of ratiometric signal processing and removes traditional response-speed limitations common to other sensors.

Detector 10 has enough sensitivity to sense weak absorbing gases and can detect several gases in one instrument with no moving parts. Power requirements can be as low as I OOuA average (24VDQ to support applications in fire and smoke detection.

The tapered optical chamber design features an in-situ photoacoustic tube that integrates high signal to noise performance into a low profile package.

The photoacoustic tube is also an excellent omni-directional heatradiation detector. As a thermal sensor, it can easily detect the high frequency flicker signature of fires from a distance.

The detector 10 can be used in combination with fire detectors such as ionization-type, a photoelectric-type smoke detects in the same housing or in displaced housings.

The gas and smoke signals can be processed to establish the presence of an alarm condition such as a fire. One form of processing has been disclosed and claimed in Tice U.S. Patent application filed April 1.9 1999, entitled, System and Method ofAdjusting Smoothing, Ser. No. 09/294,932, assigned to the assignee hereof and incorporated by reference. Other processing can be used without departing form the spirit and scope of the present invention.

Outputs from a smoke sensor could be combined with a rate of change output from the gas sensor. The gas sensor output can also be incorporated into the housing.

Fig. 7B is an illustration of an embodiment of a detector 10' in accordance with the detector of Fig. 6. The various components of the detector 10' carry the designated identification numerals. Radiant energy source 7-1 provides sampling pulses. Radiant energy source 7-2 provides reference pulses. Detector 10' includes carbon monoxide filters 7-18 to exclude frequencies not highly absorbed by carbon monoxide.

Fig. 8 is the block diagram of a multiple sensor detector 80 in accordance herewith. Detector 8 0 includes a photoacoustic gas sensor, comparable to the sensor 10 along with one or more smoke sensor(s) 80-1. Thesensor(s)80-1 could be implemented using a variety of known sensing technologies including photoelectric, ionization-type orprojectedbearn sensing. As illustrated, sensors 10 and 80-1 are carried by a common housing 82.

Detector 80 further includes control circuitry 80-2 coupled to sensors and 80- 1. Dual sensor processing of the type noted above could be performed using circuitry 80-2 local to the sensors. Alternately, circuitry 80-2 via input/output circuitry 80-3 could transmit one or more sensor related values via either a hardwired or wireless medium to other electrical units in a respective system or to a common processing element for further signal processing.

It will also be understood that a plurality of detectors, such as detector 80, could be incorporated into such a system. Such detectors could communicate directly with one another via the medium. Alternately or in addition to the system can incorporate a common control element for carrying out some or all of the signal processing.

In apreferred embodiment, control circuitry 80-2 could be implemented using a programmed processor and executable instructions stored at detector 80.

It will also be understood that the details of the input/output circuitry 80-3 required for bi-directional communication by the medium are not a limitation of the present invention. In yet another embodiment, sensors 10 and 80-1 could be displaced from one another in separate housings.

Figs. 9A and 9B illustrate two different packaging configurations wherein a gas detector of the general type illustrated in Fig. 6 can be incorporated into a smoke detector having a photoelectric, or an ionization-type sensor or both.

With reference to Fig. 9A, housing 9A-10 carries the various sensors. The housing 9A- 10 defines an internal region 9A- 12 wherein can be

carried one or more smoke sensors 9A- 14.

In addition to the smoke sensor or sensors, housing 9A- 10 carries a radiant energy or light source 9A- 16 which proj ects reflected rays 9A- I and direct rays 9A-2 across a reflective sampling chamber 9A- 18 located between source 9A 16 and a closed sensing tube 9A-20. Housing 9A-10 also defines a plurality of slots or openings 9A-22 whereupon airbome gases and smoke can flow into and out of chambers 9A- 14 and 9A- 18. In the process of the in-flow, airborne gases will absorb selected frequencies of the pulses of light from source 9A- 16 for purposes of detecting a respective gas, as described above with respect to detector 10 of Fig.

6. In addition, the airborne particulate matter can be sensed by one or more smoke detectors in chamber 9A- 14.

Detector 9A can also include temperature sensors such as thermistors.

Control circuitry, indicated at 80-2, coupled to the sensors, can be carried by housing 9A- 10.

The detector 9A could incorporate one or more smoke sensors 9A- 14 such as a photoelectric-type sensor, an ionization-type sensor or both in combination with the gas detector. It will be understood that the exact configuration of the smoke sensors in the housing 9A- 10 is not a limitation of the present invention.

Fig. 9B illustrates an alternate configuration of a detector 9B which includes a housing 9B- 1. The housing 9B- 1 can be removably attached to a surface mounted base 9B-2 and could carry therein one or more smoke sensors.

The housing 9B-1 includes an upper region which incorporates a plurality of openings 9B-3 to provide for the ingress and egress of airborne particulate matter, typically smoke, and gases. The members 9B-4 of a plurality of radiant energy sources are also positioned in common sensing region 9B- V.

A photoacoustic tube 9B-5 is centrally located relative to the radiant energy sources 9B-4. The gas sensor of detector 9B operates in accordance with the previously discussed principles of detector 10 of Fig. 6. An electronics packaging 913-6 can be carTied adjacent to the sensing region 9B-l'. An upper section 9B-7 of the housing 9B- I can be coated with an internal reflective surface such as chrome, or its equivalent to provide for reflection of radiant energy pulses from sources 9B-4 into the sensing tube 9B -5.

Detector 9B can also incorporate a plurality of thermistors 9B-8 which provide a temperature sensing function in addition to the smoke and gas sensing function. It will be understood that neither the exact configuration of the smoke sensors nor thennistors 9B-8 are a limitation so the present invention.

Figs. 10A and 10B illustrate alternate form factors for gas detectors such as the detector 10 of Fig. 6. Configuration 1 OA illustrates photoacoustic tube 1 OA- 1 oriented perpendicularly to a gas sampling chamber 1 OA-2. Chamber 1 OA-2_ includes a plurality of openings indicated generally at 1 OA-3 for ingress and egress of airborne gases which in turn can be sensed with detector 1 OA.

A sampling source of radiant energy 1 OA-4 is located at one end of the sampling chamber 1 OA-2. Photoacoustic tube 1 OA- 1, as discussed previously, is a scaled container of the gas to be sensed which includes at one end a microphone and related circuitry 1 OA-5 as discussed previously.

It will be understood that configuration 1 OA could be incorporated into a circular or cylindrical smoke detector housing which might include a sampling chamber 1 OA-6 for one or more photoelectric, ionization-type or obscuration-type smoke detectors. Thermodetectors could also be incorporated into the detector 1 OA.

Fig. 10B illustrates an alternate configuration of a detector 10B.

Components which are common to detector 1 OA carry corresponding identification numerals and were discussed previously.

The above described gas sensors exhibit a variety of performance advantages. This sealed cell design permits the use of relatively short radiant energy pulses to stimulate the sensor and enable the sensing process. Such radiant energy pulses tend to be fully absorbed when injected into the sealed photoacoustic, gas carrying tube. This will produce a highly reproducible pressure wave with little or no response time delays. In contradistinction, open cell designs must balance response time with the pulse rate of the radiant energy source along with the sensor's leak rate to achieve a proper response or signal to noise ratio.

Since the gas sensor or the present application is responsive to short radiant energy pulses, strobe light sources such as Xenon flash tubes, pulsed light emitting diodes or pulsed incandescent bulbs can be used. The advantage of being able to pulse the source of radiant energy is that it results in low average power consumption. For example, a radiant energy pulse having a duration of ten milliseconds and requiring a peak current on the order of 70rnA, if energized once every five seconds, has a resultant average current consumption of the order of 140uA_ For C02 monitoring, one sample every 10 seconds is sufficient (70uA average). If suspicious fire activity is sensed, the sampling rate can increase to once every 5 seconds or faster. More data per unit time allows for better detection sensitivity. Using this energy management technique, average power is greatly reduced.

The jump into higher sampling rates can be based on the signal profile of the associated smoke sensor(s). Profiles include, amplitudes, rate of change of slope and profile trend shifts.

The sealed tube design permits a high degree of miniaturization. For gases such as C02, strong absorption requires only about 1 cm. of travel inside the tube (filled with pure C02) to produce strong signals. This permits both tube and air sampling chambers to take on many forms.

In a smoke detector, a 2 inch path length can be used. The path can be arranged in a shallow "V" or "T" configuration to minimize product size, see Figs.

1 OA, B for example.

A 50mm long, 1Omm diameter photoacoustic tube can be selected for C02 is based on convenience and cost. This is a readily available glass tube used in other industries. Other shapes for photoacoustic, cells, however, can be pursued if applications demand custom designs.

Detectors in accordance with the present invention exhibit very fast response to detected gases, for example, carbon dioxide, yet at the same time exhibit small size, low cost and require low power levels. The fast response time of the present sensor overcomes deficiencies of some of the known gas sensors which have a slower response rate. The higher response rate of the present sensor is particularly beneficial when combined with profile processing. More particularly where the gas of interest is carbon dioxide, carbon dioxide predictive profiles can be used to improve the reliability of detection of actual fires as opposed to conditions which might represent false alarms.

A profile, a relatively short snap shot, over time can be used to estimate the predictive value of carbon dioxide concentration patterns. A profile which results from a multi-factor analysis correlates gas amplitude levels, long and short term slopes, and frequency content. As a result, it functions like a predictive barometer for a fire threat. It provides a strong confirmation value to the output signal of a photoelectric smoke sensor.

For example, using such predictive processing, profiles with strong carbon dioxide patterns can be characterized into fires or false alarms with greater confidence levels than is the case with known processing. This is a result of the result of the rapid response characteristics of gas sensors in accordance with the present invention. Such sensors have the capability of accurately attracting large changes in gas concentration as rapidly as the concentration changes.

The threat factor within a carbon dioxide profile can be used to adjust, or throttle the alarm impact of additional data. Hence, a carbon dioxide sensor in accordance with the present invention which exhibits a fast response characteristics makes it possible to closely track rapid changes of the gas to improve discerning between nuisance conditions and real fires.

From the foregoing, it will be observed that numerous variations and modifications may be effected without departing from the spirit and scope of the invention. It is to be understood that no limitation with respect to the specific apparatus illustrated herein is intended or should be inferred. It is, of course, intended to cover by the appended claims all such modifications as fall within the scope of the claims.

Claims

What is Claimed:

1 A detector comprising:

a radiant energy transmissive closed container which defines an internal gas containing region; a sensing region adjacent to the closed container wherein the sensing region is open to an inflow and outflow of ambient gas; a first beam ofradiant energy which extends, through at least part of the sensing region and through at least part of the internal gas containing region; a second beam of radiant energy which extends through only a part of the internal region; and an output transducer which generates an output signal in response to the first and second beams.

2. A detector as in claim I which includes a source of the first beam and a source of the second beam.

3. A detector as in claim 2 wherein the sources are displaced from one another.

4. A detector as in claim 2 wherein the sources are selected from a class which includes a solid state emissive element, an incandescent emissive element and a gas discharge emissive element.

5. A detector as in claim 2 which includes circuitry, coupled to the sources and the transducer, for forming an output indicative of both the first and second beams.

6. A detector as in claim 1 wherein the output transducer comprises a microphone.

7. A detector as in claim 5 wherein the circuitry includes drive elements whereby the sources are energizable at least intermittently.

8. A detector as in claim 7 wherein the sources are energizable at different times.

9. A detector as in claim 1 wherein the internal region includes a sample of at least one gas to be sensed.

10. A detector as in claim 9 wherein the gas sample comprises carbon dioxide.

11. A detector as in claim I wherein the internal region includes a first and a second gas each of which is to be sensed.

12. A detector as in claim I I which includes a third beam of radiant energy which extends through at least part of the sensing region and through at least part of the internal gas containing region.

13. A detector as in claim 12 which includes circuitry for providing a first output in response to the first and second beams and a second output in response to the third and second beams.

14. A detector as in claim 1 which includes a smoke sensor.

15. A detector as in claim 1 wherein the closed container is formed, at least in part, of one of glass and plastic.

16. A detector as in claim 15 wherein the closed container is elongated and symmetrical about an axial centerline.

17. A detector as in claim 16 which includes a housing which defines the sensing region and wherein the closed container is carried therein.

18. A detector as in claim 14 wherein the smoke sensor comprises one of a photoelectric-type and an ionization-type smoke sensor.

19. A method of detecting a gas comprising:

confining a reference gas sample in a region and excluding other gases therefrom; providing a gas sample adjacent to but outside of the region; irradiating both the sample gas and the reference gas with a common beam; detecting the beam after it has passed through both as samples; irradiating only the reference gas sample with a reference beam; detecting the reference beam as it exits the sample region; and processing the detected output to determine the presence of a selected gas.

20. A method as in claim 19 which includes energizing the common beam and subsequently irradiating the reference gas sample.

21. A method as in claim 19 which includes using the common beam to sense the presence of airborne smoke.