MXPA98005660A - Gas sensors by passive and active infrared analysis and assemblies multiple channel detectors aplicab - Google Patents

Abstract

A passive source infrared gas detector uses a passive infrared source and the space between the detector assembly and the source of the sample chamber, the gas detector includes an infrared detector assembly to produce first, second and third output: the first output indicates the radiation received in the first non-neutral spectral band which is absorbable by a preselected gas to be detected, the second output indicates the received radiation in a first neutral spectral band from the passive infrared source, and the third output indicates the received radiation in a second neutral spectral band from the passive infrared source: signal processing means manipulate the three outputs to determine the concentration of the gas being monitored, additional detectors can be added to the detector assembly for detect radiation in spectral bands characteristic of additional gases

Description

QAS SENSORS BY PASSIVE INFRARED ANALYSIS AND ASSEMBLIES MULTIPLE CHANNEL DETECTORS APPLICABLE . BACKGROUND OF THE INVENTION This request is a partial continuation of the request in progress of the same applicant serial number. 08 / 422,507, filed on April 13, 1995. The application serial number 08 / 422,507 is incorporated by this reference as if it were fully disclosed here.

FIELD OF THE INVENTION The present invention relates generally to the field of gas sensing devices. More particularly, the present invention relates to gas detectors capable of measuring the concentrations of one or more gases, using an infrared absorption band characteristic of the gas to be detected.

DESCRIPTION OF THE PREVIOUS TECHNIQUE Many gases have characteristic absorption bands that remain within the infrared spectrum. The non-dispersive infrared (NDIR) technique has been widely used in the gas analyzer industry to detect these gases. These gas analyzers use the principle that several gases exhibit substantial absorption at characteristic wavelengths in the infrared radiation spectrum. Typically, a narrow band optical element or an infrared transmission filter is used to isolate the wavelength band of interest in gas analyzers by IMDIR. On the other hand, a prism or diffraction grating is used in gas analyzers that are based on scattering techniques. The NDIR technique, which is generally classified as a non-interactive gas analysis technique, offers numerous advantages over previous interactive types of methods for measuring gas, including electrochemical fuel cells, concreted semiconductor (tin dioxide), catalytic ( platinum granules) and thermal conductivity. These advantages include response speed, gas detection specificity, long-term measurement stability, reduced maintenance and increased speci fi city. In addition, in some cases, the interactive gas sensor may be poisoned in a non-functional state. Depending on the application, this would put human life at risk. Interactive gas sensors are generally non-specific because the reagent that is used to determine the concentration of the desired gas can react with other gases that are present. Naturally this will result in false readings. Also, if the equilibrium of the reaction between the nonspecific gas and the reagent is such that the gas and reactant remain reacted even after the partial pressure of the gas falls into the environment being monitored, the sensor will no longer function properly. and he is poisoned. The response time for NDIR gas sensors is typically shorter than for interactive gas sensors, due to the kinetics of the reaction between the sample gas and the reagent, which contracted how quickly a reactive type sensor can detect a change in the concentration of gas in the environment that is being monitored. Despite the fact that interactive gas sensors are not reliable and that the technique for measuring gas by NDIR is one of the best, NDIR gas analyzers have not been widely applied due to their complexity and high implementation cost. . For years a large number of measurement techniques based on the NDIR principle for gas detection have been proposed and successfully demonstrated. In the past »NDIR gas analyzers typically included an infrared source, a motor-driven mechanical fan to modulate the source, a pump to drive or extract gas into a sample chamber, a narrow bandpass interference filter» a sensitive infrared detector plus an "expensive" infrared optical element and windows to focus infrared energy from the source on the detector. The most notable of these types of analyzers are shown and described in US Pat. No. 3,793, 525 of Burch, and co-inventors "in US Patent No.: 3,811,776 to Blau, Jr., and 4,578,762 to Wong. These NDIR gas analyzers have good performance and have contributed greatly to the overall technical advance in the field of gas analysis over the last two decades. However, its overall size, complexity and cost have prevented its use in many applications. The need for better and cheaper gas anagenizers has led to new inventions. For example, US Patent No. 4,500,207 to Maiden and Wong Nos. 4,694,173 and 5,026,992 have proposed NDIR techniques for the detection of gas that do not use any movable part, for example, mechanical fans. The goal of those patents had been to produce NDIR gas sensors that were more robust and compact, opening them up to the possibility of new applications. In an attempt to further reduce the cost and simplify the implementation of the NDIR technique, a low-cost NDIR gas sensor technique was developed. The inexpensive NDIR technique employs a diffusion type gas sample chamber, of the type described in U.S. Patent No. 5,163,332 issued November 17, 1992 to the inventors herein, and which is incorporated herein by this reference. This diffusion type gas sample chamber eliminates the need for expensive optical elements, mechanical fans and a pump to drive or extract the gas into the sample chamber. As a result, numerous applications have been opened for the NDIR technique "previously considered impractical" due to cost and complexity. A similar guiding principle led to the development of the improved NDIR gas sensor described by Wong in U.S. Patent 5,444,249, issued August 22, 1995.

This patent discloses a NDIR gas sensor of the simple and low-cost diffusion type, which can be micromachined from a semiconductor material »such as Si or SaAs» thereby allowing the entire sensor to be placed in a microcapsule. Even though the low cost NDIR gas sensor technique of US Pat. No. 5,163,332 and the improved NDIR gas sensor of US Patent 5,444,249 have opened a wide variety of new applications, these gas sensors they still require too much energy to be used in many potential gas sensor applications. As a result, low-cost "solid state gas sensors" applications remain limited. If a gas analysis technique could be developed that did not require moving parts, it would have the same degree of specificity as the NDIR technique "outside of low cost and have relatively low energy demands" so that the devices employing the technique could be operated by batteries »for a prolonged period» applications where gas sensors are used and the frequency of their use would increase dramatically. Therefore, although there has long been a need for a simple, compact, low cost gas sensor that has low energy requirements, this need remains unsatisfied. Accordingly, it is a goal of the present invention to further advance the infrared gas analysis technique, by providing a "safe" compact infrared gas sensor of low cost and low power consumption that utilizes infrared absorption. Another goal of the present invention is to provide infrared detector assemblies which can be used in infrared gas sensors according to the present invention.

BRIEF DESCRIPTION OF THE INVENTION The present invention is directed to an infrared gas sensor for detecting the concentration of one or more predetermined gases, which uses a novel technique of infrared gas analysis »called passive infrared (AIP) analysis. The AIP technique of the present invention is simpler than the NDIR gas analysis techniques »known so far» because it no longer needs an "active" infrared source, nor does it need a structurally defined sample chamber. As a result »small» solid state »gas sensors can be built for low cost and low power consumption, to satisfy a number of special applications where it was previously impossible to use currently available NDIR gas analyzers. The present invention recognizes that all objects greater than 0ßK (Kelvin) emit radiation. The present invention takes advantage of this fact by using ordinary objects, such as walls, ceilings, floors, etc., as a "passive" source of infrared radiation. These "passive" sources of infrared radiation can be used effectively in some cases to replace the "active" sources of infrared radiation that had been used almost exclusively until now in all NDIR gas analyzers. The "active" infrared source used in conventional NDIR gas sensors is typically a heated and very hot object (500-1000ßC) such as a nicrome wire embedded in alumina ceramic (Nerst incandescent body) or a wire of tungsten resistance »of a small incandescent bulb. These sources are characterized as "active" sources because they are powered by the gas sensor. On the other hand, a "passive" source, as used herein, is any object that is above 0 ° K »but which is not powered by the power source of the gas detector. Typical passive infrared sources will be used by the infrared gas sensor of the present invention to include walls, carpets, tile floors, »roofs and furnace walls» to name only a few »however» clearly »as it will be recognized by those skilled in the art by the teachings of the present description, the passive infrared sources that can be used by the sensor "Gas of the present invention are virtually unlimited.

However, the temperature of the passive infrared source must be higher than the temperature of the gas to be measured. That is, the law of detailed equilibrium must be observed. Although the temperature of the active infrared sources is very high, the source area is typically quite small. A source area on the order of a few square millimeters is not uncommon. On the other hand »although the temperature of the« typical »interior passive infrared source is only around 300ßK or around 25 ° C» if the • source area used is approximately 1000 times greater than that of conventional infrared sources »then» by using the PlancK equation it can be shown that the spectral radiant emission for the passive infrared source is comparable to that of conventional active sources in the spectral region »from 3 to 20 μm. The passive infrared source area necessary for the proper detection of gas will depend on the expected temperature of the passive infrared source. In the AIP technique employed in the present invention »the passive infrared source» must be characterized. To characterize the passive infrared source, a detector assembly is provided that is capable of measuring the spectral mission from the selected passive infrared source "in two different spectral bands. The spectral bands used to characterize the source are preferably "neutral" spectral bands. The neutral spectral bands are spectral bands that are selected so that they are not absorbed "or only moderately absorbed" by any of the gases typically found in the environment to be monitored. Based on the law of PlancK »the proportion of outputs measured in the two neutral spectral bands can be used to uniquely determine the temperature of the passive infrared source» assuming that the two red spectral bands are close enough »so that the variation of the function of identity for the source is negligible. To determine the concentration of the gas to be detected, the detector assembly also measures the amount of incident radiation in a "non-neutral" spectral band that coincides with an absorption band of the gas to be measured. Therefore, this output is indicative of the concentration of the gas within the angle subtended by the detector assembly to the passive infrared source. Using the measured output in at least one of the neutral spectral bands »the measured output the non-neutral spectral band and the calculated temperature» was it can determine the gas concentration within the angle subtended by the detector assembly to the passive infrared source. According to one embodiment of the present invention, a passive-source infrared gas detector is provided which uses a source at ambient temperature "of higher temperature than the surrounding gas" and the space between the detector assembly and the source "as sample camera. The gas detector comprises an infrared detector assembly for producing a first output »a second output and a third output» being the first output indicating the radiation received by the detector assembly in a first non-neutral spectral band which is absorbable by a preset gas to be detected »the second output of the radiation received by the detector assembly being indicated to a first neutral spectral band from the infrared passive source» and the third output of the radiation received by the detector assembly being indicative second neutral spectral band »from the infrared passive source. Signal processing means are included to manipulate the three outputs in order to determine the concentration of gas that is being monitored. By adding additional detectors to the detector assembly that can detect radiation to spectral bands characteristic of additional gases, the infrared gas detector can be used to monitor the concentration of a plurality of gases.

According to one embodiment of the present invention "a passive source infrared gas detector" is provided comprising: (a) an infrared detector assembly »comprising: (i) a port for receiving radiation therethrough from the passive infrared source »< ii) a first sensor »a second sensor and a third sensor arranged to receive the radiation through the port to produce a first output» a second output and a third output indicating the radiation incident on the first sensor »the second sensor and the third sensor »respectively» (iii) a first narrow-bandpass filter, interposed between the port and the first sensor »the first narrow-bandpass filter producing an output therefrom. indicator of the radiation incident on the first bandpass filter, to a first non-neutral spectral band that is absorbable by a preselected gas to be detected; (iv) a second narrow-bandpass filter »interposed between the port and the second sensor, the second narrow-bandpass filter producing an output therefrom» indicating the radiation incident on the second bandpass filter a a first neutral spectral band »and (v) a third narrow bandpass filter. interposed between the port and the third sensor; the third narrow bandpass filter producing an output therefrom, indicating the radiation incident on the third bandpass filter, to a second neutral spectral band; (b) temperature measuring means for producing a signal corresponding to the ambient temperature of the first »second, and third sensors; (c) signal processing means, adapted to receive the outputs from the first sensor, the second sensor, the third sensor, and the temperature measuring means; and for sampling and at least temporarily storing the outputs of the first sensor »of the second sensor, of the third sensor and of the temperature measuring means, at previously established intervals; the means including signal processing means for: (i) correcting the stored outputs of the first sensor, the second sensor and the third sensor, to compensate for the ambient temperature of the first sensor, the second sensor and the third sensor, respectively, during the period Of sampling; (ii) calculate the temperature of the infrared passive source during sampling. based on the proportion of the corrected values of the outputs of the second and third sensors »(iii) calculate a predicted output for at least one of the second or third sensors. based on the calculated temperature of the infrared passive source »during the sampling period; (iv) calculate an attenuation factor, comparing the predicted output of at least one of the second or third sensors. with the corrected output from the corresponding sensor during the sampling period; (v) correct the stored signal of the first sensor by means of the attenuation factor. (vi) determine the concentration of the gas during the sampling period from the corrected output of the first sensor, and (vii) monitor the concentration of gas in a predetermined function, and provide an output signal based on the v gi lancia In this way, the infrared gas sensor according to the present invention uses a passive infrared source in a novel AIP technique that effectively eliminates the need for an "active" hot infrared source, which is used in conventional NDIR gas measuring devices. onales.

Further »in the AIP technique employed in the infrared gas sensor of the present invention, the space between the infrared passive source» for example »a certain portion of a wall» and the detector assembly »becomes the sample chamber. In other words, the present invention not only eliminates the "active" infrared source, but also eliminates the need for the sample chamber used in conventional NDIR gas analyzers.

Due to the fact that an "active" infrared source is not necessary for the implementation of the present invention "the energy consumption of an infrared gas sensor can be significantly reduced" according to the present invention; thus making the passive and simple "infrared" gas sensor of the present invention "operable with batteries for a prolonged period. Additionally, the size of the sensor can be reduced because a defined gas chamber will no longer be necessary. Accordingly, it is an object of this invention to provide an apparatus and a method for measuring the concentration of one or more gases using a "novel infrared analysis technique" called passive infrared (AIP) analysis. It is also an object of the present invention to provide an improved infrared detector assembly that can be used in the gas sensor according to the present invention. Other objects and advantages of the invention will be better understood from the following description taken in conjunction with the accompanying drawings, in which the preferred embodiments of the invention are illustrated by way of example. However, it must be expressly understood that the drawings have the sole purpose of illustration and description and are not intended to define the limits of the invention.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 shows a preferred embodiment of the present invention, illustrating the detector assembly, the infrared passive source (wall) and the intermediate space between the infrared passive source and the detector assembly, which constitutes the sample chamber. Figure 2 is a graph showing the spectral radiant emission of a blackbody at temperatures of 100 to 1000ß. Figure 3 is an exploded view of a detector assembly according to an embodiment of the present invention.

Figure 4 is an oblique view showing a partial cut-out of the detector assembly illustrated in Figure 3. Figure 5 shows an alternative preferred embodiment for the present invention, illustrating the actual use of a portion of a wall as a "passive" source. Infrared "and the use of a convex spherical reflector to increase the original field of view (FOV) of the detector assembly. Figure 6 shows a schematic drawing for signal processing circuits for a preferred embodiment of the present invention. Figure 7 shows the schematic circuit for the signal processor according to another embodiment of the present invention. Figure 8 is a cross-sectional view taken along line 9-9 of Figure 9 »of another embodiment of a detector assembly according to the present invention. Figure 9 is a longitudinal sectional view of a detector assembly according to the embodiment of Figure 8 »taken along line 10-10. Fig. 10 is an oblique view of the substrate and the interference bandpass filters illustrated in Figs. 8 and 9. Fig. 11 is a top view of the used substrate of the detector assembly fashion illustrated in Figs. 8 and 9. Figure 12 is an enlarged bottom view of the substrate used in the detector assembly mode illustrated in Figures 8 and 9 »showing thermal cells manufactured therein. Figure 13 is an illustration of a preferred construction of an interference bandpass filter for use in detector assemblies according to the present invention. Fig. 14 is an illustration of a filter assembler for use in detector assemblies according to the present invention. Figure 15 is a partial sectional view "taken through the substrate illustrated in Figures 8-10" showing the filter assembler attachment of Figure 14 in actual use. Figure 16 is an exploded view of a detector assembly according to another embodiment of the present invention. Fig. 17 is an oblique view showing a partial cut-out of the detector assembly illustrated in Fig. 16. Fig. 18 is a gas sensor "by passive infrared analysis, according to another embodiment of the present invention. Figure 19 illustrates the detector assembly shown in Figures 16 and 17. For use in an NDIR gas sensor according to the present invention.

DETAILED DESCRIPTION OF THE INVENTION A preferred embodiment of an AIP gas sensor according to the present invention will now be described. Referring to Figure 1. Figure 1 illustrates a detector assembly 3 comprising a signal detector 4 equipped with an interference filter, narrow bandpass filter F (not shown) whose center wavelength (CWL) L3. matches the absorption band of the gas to be measured. Further »the detector assembly 3 includes 2 detectors 5 and 6 source characterizers» respectively equipped with narrow bandpass filters F-a and F3 (not shown) whose CWL La and L3 do not match any of the known gases or vapors. that are commonly found in the atmosphere. In other words, "at the wavelengths Lß and Lß" there should be no absorption bands (or at least extremely weak) for the gases or vapors commonly found in the atmosphere being measured. For air, neutral wavelengths can be found at 3.91 μm, 5.00 μm, and 9.00 μm. If carbon monoxide (CO) is the desired gas to be detected »then the CWL and the entire width at the maximum mean values (FWHM) for the interference filter associated with the detector 4» are selected to be at 4.67 μm and 0.1 μm »respectively. On the other hand, if the C02 is the desired gas to be detected, the CWL and FWHM for the interference bandpass filter associated with the detector 4 are set at 4.26 and 0.1 μm, respectively. As will be recognized by those who are experts in the field »this technique has application to many other gases that have an absorption band in the infrared, including water and fully volatile organic chemicals (TVOC). Typically. the CWL LA of the interference filter Fx associated with the detector 4 will be selected so that it is as close as possible to the mi ad of the absorption band that is being used for the gas of interest. This will ensure that the maximum amount of radiation in the spectral band being monitored is absorbed by the gas, thereby increasing the sensitivity and accuracy of the detector. However, in the case of gases that are very strong absorbers, such as C0ß, it may be necessary to move the CWL LA of the neutral filter FA for the detector 4 »to one side of the absorption band, so that too much light is absorbed in the spectral band that is being monitored. This displacement must be considered when using trajectories: very long, or when the concentration of the gas is very high. This technique can be used to prevent the detector from lacking light within the range of gas concentrations to be monitored. The FWHM of the interference filter F. associated with the detector 4, is preferably selected to be approximately 0.1 μm. so that the detector has a high degree of specificity. Those skilled in the art will recognize "however" that other bandwidths can be selected "depending on the amplitude of the absorption band of the gas being monitored and the degree of specificity desired for the detector. The CWLs La and La of the neutral spectral bands selected for the interference filters of Fß and Fß should be selected as close as possible to the spectral position of Ls .. Although it is not necessary »it is also convenient that L ^ lies between Lß and L3. For example, if CO or COa is to be detected »L3 and L3 can be selected to be 3.91 μm and 5.00 μm respectively. Alternatively, La and L3 can be selected to be 3.91 μm and 9.00 μm. In the present embodiment, the FWHM of F2 and F3 is preferably set to approximately 0.1 μm. The amplitude of the spectral band that pass through F3 and F3 must be sufficiently narrow so that it does not overlap with the absorption line of a gas that would be in the atmosphere. When fixing the CWLs of L and L3 equal to 3.91 and 5.00"respectively" and the FWHM of these detectors at 0.1 μm "no important overlap occurs. Consequently, the outputs of detectors 5 and 6 are not affected by the concentration of the gas to be measured or of any other gas or vapor commonly found in the atmosphere. The detectors 4, 5 and 6 are preferably thermal stack detectors. However, as will be recognized by those skilled in the art, other infrared detectors may be used in the present invention, including schottky photodiodes of platinum silicide. The field of view (FOV) of the detector assembly 3 is determined by the aperture collar 7 attached to the detector assembly shown in Figure 1. The detector assembly 3 subtends an area 8 (corresponding to area A) of the wall that it is used as the passive infrared source for the present invention. The length S of the path of the gas infrared sensor of the present »is defined by the distance between the detector plane 10 of the detector assembly 3 and the wall 9.

The ratio between the area A of the infrared passive source 8 and the solid angle subtended therein by the detector assembly 3, or OM, defines only the sample path length S for the infrared gas sensor »currently described» as follows: Sample path length S = CA / 0] 3 - '' z Since the solid angle OM is a function of the FOV subtended by the detector assembly on the wall and can be adjusted at will by design » therefore »the length S of the sample path for the present invention is an extremely useful variable. In other words »the detection of a gas at low concentration, with an extremely weak absorption band can be made by making the length S of the trajectory very long (several meters) in order to reach the adequate modulation for said detection, in fact »As will be recognized by anyone skilled in the art, the path length S must be set depending on the amount of modulation desired. For example, when a very strong absorber »such as C0a» is being monitored, shorter path lengths should be considered. However, if the desired application requires the detection of gas concentrations on the ppb scale, then longer path lengths can be used. Although virtually any path length can be selected, lengths between 12.7 cm and 3.04 m will typically be suitable, with most path lengths being between about 12.7 cm and 1.82 m. The output Vp. of the signal detector 4 is used to determine the concentration of the gas to be measured. The output of the detector 4 depends on numerous factors. First and foremost is a function of the temperature T and the emissivity e of the infrared passive source 8 as dictated by the formula of spectral radiant intensity illustrated in equation 1 which comes further. Additionally, V ^ also depends on the system's optical output or attenuation. expressed as G (see equation 1 below) and the concentration of the gas to be measured, found between detector assembly 3 and passive infrared source 8. The concentration of the gas to be measured determines the value of the modulation factor M as shown in equation 1 below. Detectors 5 and 6 »that are equipped with Fa and F3 filters. they are used to dynamically characterize the infrared passive source 8 and the real-time ampere for the signal channel monitored by the detector 8. The Z proportion of the detector outputs 5 and 6 only determines the temperature of the source 8. Ad cional Once the temperature T of the source 8 is determined, it is also possible to quickly determine the instantaneous values for the source emissivity e »the system's optical output (or attenuation) S. using equation 1 below" and comparing the actual outputs with the stored values of the respective outputs at the temperature C0 and the emissivity e0 »of a reference black body source, measured while the system was being started. The values for T »e and Q are updated continuously in real time for the output of the signal detector 4» which allows the latter to establish the concentration of the gas to be measured. The "simple" infrared gas sensor described herein is also capable of rejecting the influence of scattered radiation, in virtue of the fact that the infrared passive source 8 is generally never good reflective. Therefore, the amount of diffuse radiation that can find way to the FOV of the optical system is minimal. Furthermore, unless diffuse radiation happens to be in the spectral band defined by the filters of the detector assembly, that is, L3 and L3 »will be rejected. Even if they have energy within the spectral path band of the optical system sensor »the emissivity is probably rather uniform and constant. In such a case »the neutral detectors will simply treat said diffuse radiation as an increase in the temperature 8 of the infrared passive source» with the corrected information related to the signal detector to process it appropriately. The manner in which the concentration of the gas to be measured is determined from the Vx outputs. V ^ .and V3 of the detectors 4, 5 and 6, respectively, are now described with respect to FIGS. 2 and 3.

Figure 2 shows a spectral radiant emission from a black body source at temperatures ranging from 100 degrees K to 1,000 degrees K. Several characteristics of the radiation from a black body source can be derived from these curves. First, the total radiant intensity, which is proportional to the area under the curves, increases rapidly with temperature. The area under the curves, which is being defined by means of the Stefan-Boltzmann equation and, thus »is proportional to the Stefan-Bol constant tz ann» multiplied by the absolute temperature »to the fourth power. Secondly, "the wavelength" of the maximum spectral radiant emittance is shifted to the shortest wavelength "as the temperature increases. This is called the Wien displacement law »which is discussed more fully afterwards. Third, the individual black body curves never cross each other; therefore, the higher the temperature, the greater the spectral radiant emittance at all wavelengths. In conventional NDIR measuring systems that use a black body, the infrared source is normally maintained at a constant and relatively high temperature (750-1,000 degrees K) and, thus, its spectral radiant emittance is typically represented. by one of the curves above 700 degrees K in figure 2, depending on its absolute temperature. In contrast, the present invention is based on the infrared radiation of the infrared passive sources. As a result, the black body curves at around 300 degrees K will typically reflect the radiant emittance of the typical sources used with the present invention. Such is the case with wall 9 in Figure 1. The two narrow spectral bands 1 and 2, illustrated in Figure 2, are centered at 3.91 μm and at 5.00 μm which, as discussed further back, are wavelength bands desirable for neutral detectors 5 and 6, when monitoring CO or CO- .. Because the bands illustrated in. Figure 2 corresponds to the neutral spectral bands that let the F- filters pass. and F3, it would be preferable for them to have a FWHM of 0.1 μm. As can be seen from FIG. 2, the ratio (Z) of the spectral radiant emissions of these two wavelength bands determines the blackbody temperature in a unique manner. The only assumption made in this statement is that the emissivity of the "passive" infrared source is approximately the same within the spectral band joined by 3.91 μm and 5.00 μm. For almost all interior walls that are painted "papered or covered with wood panels" this is a good guess. Before determining the concentration of the gas being monitored, the passive source 8 must be characterized. The manner in which the detectors 5 and 6 dynamically characterize the temperature and emissivity of the infrared passive source 8 for the signal channel 4 »is described below. For the purposes of this discussion, the detectors 4, 5 and 6 will be called detectors D ^ »Dz and D3. Assuming that the three detector outputs VA »Vg. and V3 are referenced initially (ie »initialized)» so that they have the values VÍO »V30 and Vao, respectively» in a known "passive" source of infrared having the temperature T0 »e0 and the area A0 = OM x S2 »where OM is the solid angle corresponding to the FOV of detector assembly 3» subtended by the passive source in the detector assembly »and S is the defined sample path length» can be written: -, o = R ( T0. E0 »L,) A0W r1 (aJ / (2? TSz)) GM volts equation CID where: i = 1. 2 or 3 R (T0, e0, L-,) = The source radiant emittance of the source passive infrared (Watt cm-A μ- *) »which is defined as e0 (lambda) multiplied by the spectral radiant emittance of a blackbody, as defined by Planck's Law» as follows: 2? rhcalambda- B) / (exp (ch / k,!, LambdaT) -l); A0 = Infrared passive source area; w., = FWHM or F-,; r-, = Detectivity of detector D., (Vol ts / Watt); a, = Detector area D, »S = Sample path length Q = System optical output (1005Í = unit); and M = Modulation by the gas to be measured. When the detector assembly 3 faces a passive infrared source 8 in real time, the area A (A is the same as the reference condition A0, because OM and S are fixed by the design in the embodiment illustrated in figure 1 ), The temperature T and the emissivity e, the outputs of Dx are given by the previous CID equation, as follows: V., = R (T, e, L1) A 1r1 (a < / (2pS2)) GH volts, where 1 = 1. For the neutral channels Da (i = 2) and D3 (i = 3). If it is assumed that Wa = W3; rx = x (similar detectors); aa = a3 (same detector areas); G ^ = G3 (both detectors share the common optical system) and M = 1.0 (neutral spectral bands for both D3 and D) 'then the outputs of the detectors D3 and D3, ie Va and V3, are the unique functions from their respective spectral position L3 and L3, the temperature T and the emissivity e of the infrared passive source 8. If it is further assumed that the epsilon emissivity of the infrared passive source 8 is equal for the narrow spectral region bound by Lß and L3 (approximately one miera) »then the ratio of the outputs Z = is only a function of the temperature T of the infrared passive source B »and the spectral positions L.ß and L3. - «- > hacho l -F- to the radiation of the plane or to the plane, write down the plane to the plane with the plane of the plane to the ratio of 1 to 1 to not 1 to r-ac. an-baa aapaotralaa < mtr > do * p »oa-f a -1 onaa« apa. · «• cu nd aa-. < aprop adamanta aapaolaaa. da-barm nan aa manar-a -ngu ar * the temperature gives an art-eu to da ßuarpo naa or an oiartßa partaa or to dominion. ou-po n-ß o »tno? -.« no- The present invention takes advantage of this fact and recognizes that in the spectral regions between 3 and 15 microns and at blackbody temperatures between 250 and 350 degrees K »in reality said proportion can uniquely determine the blackbody temperature. Additionally »once the temperature T of the reference temperature T is determined» the current value of V2 or V3 of the respective neutral detector outputs can be used »to calculate by calculation the changes (if any) for the rest parameters »grouped together as a product in the previous CID equation, that is, the change in emissivity e of the infrared passive source 8, from e0. the change in the optical output G of the system and the change in the responsibility of the detentor due to the aging of the detector itself. Thus, by adding two detectors with neutral spectral bands to the detector assembly of the present invention, the proportion of the Z outputs of these two detectors can be used to characterize in real time the temperature of the infrared passive source 8. It is important to note that »since the changes in the other parameters of the previous CID equation» ie »e» G and r are substantially equal for the two neutral detector channels »the value of the Z ratio which is the only necessary parameter to determine in a unique way the temperature of the infrared passive source 8 can always be obtained firsthand. After this vital information is obtained »the individual preset values of the signal and of the outputs of the neutral detector (VAO» V20? V30 »T0» and e0) can be used for further determinations »by means of calculations» of any changes in the parameters in the CID equation. Since the parameters necessary to determine the concentration of the gas to be measured from the output of the signal channel detector in the CID equation are T »e, G» r and M »and since the first four parameters are dynamically characterized by the two channels neutral detectors for the signal detector channel »the present invention» as illustrated in the present mode »is capable of accurately measuring the concentration of gas without the need for an active infrared source or the accompanying gas sample chamber . The only condition for this signaling is that the infrared passive source needs to be at a higher temperature than the gas that is being monitored. When the infrared passive source and the gas being monitored are in equilibrium, no absorption will be observed because the law of detailed equilibrium requires the gas to emit the same amount of photons it absorbs. Table 1 is a table that illustrates the value of the ratio of the spectral radiant emissions for spectral bands of 0.1 microns, which have central wavelengths of 5.00 and 3.91 microns, as a function of the temperature of the "passive" infrared source. It is assumed that the emissivity values for both spectral bands are equal.

TABLE 1 The above table illustrates how the spectral radiant emissions vary at 3.91 μm and 5.0O μm »as a function of the temperature of the infrared passive source, from 5ßC (27B ° K> to 45 °» C (318ßK) For the purposes of Table 1, the emissivity e is assumed to be 1 both at 3.91 μm and at 5.00 μm In the vicinity of the blackbody curves at 3000K the curves themselves are uniform and there is very little difference with respect to the value of the proportion as a function of the black body temperatures.

EXAMPLE 1 An example of how to calculate the concentration of the gas to be measured as from the outputs of vx »Va» and V3 of the detectors 4 »5 and 6» under a given set of circumstances »using the CID equation is provided then. With reference to the CID equation, the reference conditions for this example are defined as follows: T0 = 298 ° K or 25 ° C e «= 1.000 constant in the whole calculation of the example W_ 0.1 miera for i = 1 (signal ). 2 (neutral) and 3 (neutral) r «a Detector Responsiveness D ^. which is the same for i = 1, 2 and 3 a., = detector area D., »which is the same for i = 1» 2 and 3 and remains constant. S = S - constant throughout the calculation of this example, Gß = System output during the initialization l.O M = modulation factor due to the presence of the gas to be measured = 1.00 for detectors D3 and D3. If a constant C is defined as C = a ^ / íSpS38) »the constant is the same for each detector channel and remains unchanged throughout the calculation in the present example» because a., Is the same for each detector and the length of the trajectory proves to be fixed by design. To the above reference conditions »and assuming that the DA signal detector has a CWL of 4.67 μm» which corresponds to the carbon monoxide absorption band »the neutral detector D3 has a CWL of 4.67 μm» corresponding to the carbon monoxide absorption band »the neutral detector D.» has a CWL of 3.91 μm and the neutral detector D3 has a CWL of 5.00 μm »the measured voltage outputs of the detectors DL »Dz and ° 3 'in the initialization, are as follows» using the CID equation and table 1. Detector output Ds, (signal at 4.67 μm) a.0 = 5.4507 x 10 ~ ß A0r-, CG0M volts m 5.4507 x 10 ~ ß YM (where Y = A0r-, C) Detector output Dß (reference to 3.91 μm) V2 < 5 - = 1.775B x 10-ß YGQM volts = 1.7758 x 10-ß Y volts Detector output D..¡, (reference to 5.00 μm) V30 = 7.S655 x 10 ~ ß YG0M vol s = 7.6655 x 10 ~ ß And volts The gas sensor is initialized by measuring the voltage outputs of each detector when no carbon monoxide is present »and then when a known concentration of gaseous carbon monoxide is present within the field of view of the detector assembly. In that way, a calibration curve can be obtained for the gas sensor, as will be recognized by one skilled in the art. After initiation, the sensor is ready to take measurements in real time. For the present example, suppose you are in the following situation. The temperature of the passive infrared source has been increased to 308ßK or 35ßC and the emissivity e of the infrared passive source 8 is 0.8. The optical attenuation factor G is now 0.9; in other words, there is now an attenuation of the signal of the signal from the infrared passive source 8. Suppose also that the concentration of the carbon monoxide gas present within the field of view of the gas sensor causes a 2% modulation in the signal detected by the signal detector D ^ .. As a result »the modulation factor M decreases from 1.00 to 0.98 for the signal detector. There should be no modulation of the signal to the neutral channel detectors D3, D3 due to gaseous carbon monoxide, since the transfer filters for the neutral channel detectors have been appropriately selected to avoid the monoxide absorption bands. carbon and other gases that can be found in the environment that is being monitored. Under the conditions indicated above, the output voltages for the three detectors would be: Detector output D .. (signal at 4.67 microns) x = 7.6250 x 10-ß (0.8) YGM = 6.1000 X 10 ~ ß Y (0.9) (O.98) volts = 5.3802 X 10 ~ ß Y volts., Detector output D-. (reference to 3.91 micras) Vß - * 2.S517 x 10-ß (0.8) YGM «2.1214 X 10-ß Y (0.9) (1.0) volts = 1.90922 x 10 ~ ß Y volts. Detector output D-y (reference to 5.00 microns) V3 - = 10.488 X.10 ~ ß (0.B) YGM = 8.3904X 10 ~ ß Y (0.9) (1.0) volts = 7.55136 x 10- = Y volts. As with the initialization voltage outputs, frame 1 was used to obtain the spectral radiant bands for each of the wavelengths being monitored. The first step in determining the concentration of gaseous carbon monoxide or another gas to be measured is to calculate the Z proportion of the outputs of the two reference detectors: Z = Voltage (5.00) / Voltage (3.91) = 7.55136 / 1.90922 = 3.9552 Using Table 1, it is determined that the temperature of the infrared passive source 8 is 35 ° C. As noted above, it had been assumed that the area of the infrared passive source 8 and the optical arrangement remained unchanged during the example. If only the temperature needs correction, then the new voltage output for the neutral channel detector of two must be the initiation value of 1.7758x 10- ß Y volts, multiplied by the ratio (2.6517 / 1.7758), which is equal to 2.6517 x lO-ß and volts. Since the two voltages are not equal, it is known that the emissivity e or the attenuation G or both are different than the initialization conditions.

For the measured output of detector D2 (3.91 microns), and that the initiation value for this neutral channel must be at ° C, the product eG can be calculated as follows: eG = (1.90922 X 10- = Y vol ts) / (2.6517 X 10 ~ ß Y volts) = 0.72 It should be noted that if the product of emissivity ey the attenuation G < :, during the initialization is less than 1.0, then the initialization value would need to be normalized to what it should be for an em0 of 1.0 and a G0 attenuation of 1.0. In this way, the ratio of the two voltage outputs will result in the product of instantaneous value of the factor eG. The same information can also be derived from the use of output voltage from the other neutral detector that monitors the 5.0O μm channel. Once it is known that the temperature of the infrared passive source is 35 ° C and the product of eG is 0.72, a corrected output voltage in the channel of 4.67 μm, or signal can be calculated from the measured output of detector D, as follows: 5.3802 X 10- = Y volts (5.4507 / 7.6250) X (1 / 0.72) = 5.3417 x 10 ~ ß Y volts Co o can be observed from the calculation »the voltage of the output for the eG factor and for the temperature, as a result the ratio of this corrected voltage output and the starting voltage output gives the modulation factor M »as follows: 5.3417 x 10 ~ ß Y / 5.4507 x 10 ~ ß Y = 0.98. Thus »the previous methodology correctly predicts the modulation factor for the signal detector D» which is monitoring the 4.67 micron channel. To recap »the first step of the procedure is to obtain the temperature of the new infrared passive source 8» by calculating the ratio of the two reference detectors. The second step is to compare the measured value for any of the two neutral channels with "initiation value" and deduce the "eG" factor. These two bits of information are then used to correct the measured output of the Dx signal detector to 4.67 microns. The ratio of this corrected value and the stored initiation value for the DA signal detector will then produce the modulation factor. The modulation factor is used to give the concentrations of the present gas using a calibration curve that can be stored in the signal processing circuit "as is known in the art. It is important to note that »yes. well the FWHM (ie »Wj) of the neutral detectors was described earlier with respect to Figures 1-2 and Table 1» as equals »it is not necessary that the PIA detector of the present invention be designed to have amplitudes neutral band equal. This is because W., for each neutral detector (as well as for the signal detector) will always be a known parameter »set during the manufacture of a particular PIA detector. In addition, depending on the FWHM value of each of the neutral detectors, the Z proportion of the radiant emittance detected in the two neutral wavelength bands will still uniquely define the temperature of the infrared passive source 8. This is because "as illustrated in equation 1 above" the radiant emittance of the infrared passive source is a function of the area under the blackbody curves defined by Planck's law, multiplied by the emissivity e of the source, which will typically be the same for each of the neutral channels.

The PIA gas sensor described with respect to Figures 1-2 and Table 1, can therefore be alternatively designed so that the FWHM of ix and W3 of Fß and F3 (ie the neutral channels) »are set so that their corresponding detectors measure the spectral radiant emittance from the passive infrared source 8 over a band of several microns instead of the 0.1 μm band suggested above. The spectral bands passed through each of the neutral interference filters can also overlap in a system like this. The only limitations in that configuration are that the two neutral detectors can not measure the spectral radiant eminence for identical spectral bands "and that the emissivity e of the passive infrared source must be relatively constant in each of the spectral bands. The two neutral detectors can not measure the spectral radiant emittance for identical spectral bands "because the Z ratio would always be one in such situations. The emissivity e of the passive infrared source must be relatively constant to ensure that this factor is canceled when the Z-ratio is calculated, thereby allowing the temperature T of the infrared passive source to be determined directly from the proportion of the outputs of the two neutral channels. On the wavelength scale from 8 μm to 14 μm »there is very little absorption by water and C03. Therefore »as an example» the FWHM Wg. of the interference filter Fa can be set so that the light passes between 8 and 14 μm »so that the detector 5 measures the total amount of energy emitted by the passive source in this spectral band. If the spectral band passed through the interference filter F3 is narrower than that allowed to pass through F? ». then the ratio of the amount of energy detected by the detectors 5 and S will still be used to uniquely determine the temperature of the infrared passive source 8 in real time, as described above. The band pass filter F3 preferably passes a spectral band of light that is about half as wide as that passed by the bandpass filter Fa; Thus, F3 could be designed so that the spectral radiant emittance of the passive infrared source would fall within the range of 9.5 μm and 12.5 μm. By setting the FWHM W3 of the interference filter F3 »to about half the amplitude of W-g of the interference filter F3, good variability in the proportion Z» is guaranteed as a function of the temperature of the infrared passive source 8. The advantage of designing the neutral channels in this way is that significantly more energy will be detected by the detectors 5 and 6 »respectively. This improves the signal to noise ratio for the detector »thereby enabling a more accurate characterization of the passive infrared source.

As would be apparent to those skilled in the art from the foregoing, the infrared gas detector according to FIG. 1 can be used to monitor the concentration of a plurality of gases, simply by adding additional detectors to the assembly. detector 3 »and appropriately selecting the CWL of the interference filter F-,, to correspond to the characteristic absorption band of the gas to be monitored. The construction of the detector assembly 3 according to an embodiment of the present invention is illustrated in Figures 3 and 4. As illustrated, the detector assembly is produced in a detector housing 31, such as a TO-5 can. The infrared detectors 4, 5 and 6 are mounted on a housing base 30 of the TO-5 can 31. The infrared detectors 4 »5 and 6 are very close to each other, so that the field of view of each detector overlap substantially to another. While a variety of infrared detectors can be used in the present invention, the detectors 4, 5 and 6 are preferably thermopiles »due to the fact that the thermal cells do not require any energy» have a linear output and have a very good signal-to-noise ratio. Although not required, it is also preferable to link the re-Feeding joints of each of the three detectors to the same thermal heat sink. The film assembly 32 is disposed above the housing base 30. so that the only radiation that can enter the space between the filter assembly 32 and the housing base 30. is the radiation that enters through the walls. three openings 34 located in the filter assembly 32. The openings 34 are located in the filter assembly 32. such that each opening is in axial alignment with one of the detectors. The interference bandpass filters Fx, Fa and F3 cover the openings 34. such that they are interposed between the respective detector and the passive source of infrared light. Additionally »covering the three openings 34 located in the filter assembly 32» with interference filters F »F2, and F ensures that the only radiation that can enter the space between the filter assembly 32 and the housing base 30, is that of the desired spectral bands. The divider 40 is used to prevent light of a spectral band from coming into contact with an infrared detector, intended to measure light from a different spectral band. The CWL and FWHM of the bandpass filters F. Fß and F3 are fixed as described in relation to Figures 1-2 and Table 1. The lid 42 for the can 31 TO-5 acts as an opening collar 7 and, thus, defines the FOV for the detector assembly 3 The upper part of the lid 42 comprises a light transmitting window 44. In selecting the material for the window 44, it is preferred to select a material that is as transmittable as possible for the spectral bands that are being monitored by the detector assembly 3. Preferably the window 44 is also transmitter for each of the spectral bands that they are being watched. Window materials that have relatively uniform transmission qualities, in the range of 1 μm to 10 μm »include silicon» CAF ... and BaFjg. The particularly preferred materials are CaFz and BaFa »due to their high trans isi vity in this scale. In order to save costs, the window 44 can be completely eliminated. However, when the window 44 is included, the detector assembly 3 illustrated in FIGS. 3 and 4 can be hermetically sealed and thereby increase the life expectancy for the detector assembly. Additionally »as the po and grease accumulate on the detector assembly 3» the output signal corresponding to the spectral bands will begin to fall. If the attenuation of the signal becomes too great, the infrared gas detector will not work properly. However, by including window 44 in assembly 3, the strength of the original signal can easily be restored by cleaning window 44. This is not possible if window 44 is omitted. If a larger platform is desired, so that additional detectors and additional bandpass filters can be added to increase the capabilities of the infrared gas detector of the present invention "a TO-8 or larger pack can be selected. For example, said platform could be used if the ability to monitor a plurality of gases is desired. A particularly preferred detector assembly 3 is now described, with respect to Figures 8-15. As illustrated in Figures 8, 9 and 10, the detector assembly 3 includes three infrared detectors 4, 5 and 6, which have been formed on the substrate 50 mounted within the detector housing 31. The housing 31 of the preference detector is a TO-5 can, which consists of a housing base 30 and a cover 42. The cover 42 includes an opening collar 7, which defines a port for receiving radiation within the detector assembly. . The FOV of the detector assembly, therefore, it is limited by the opening collar 7. The cover 42 also preferably includes a light transmitting window 44 which fits into or covers the port defined by the opening collar 7. The light transmitting window 44 it is attached to the cover 42 so that when the cover 42 is attached to the base 30 »the infrared detectors 4» 5 and 6 are hermetically sealed inside the detector assembly 3. The infrared detectors 4 »5 and 6 are supported on a substrate 50 which »in the present embodiment» is made of a semiconductor material »such as Si» Ge »GaAs or the like. Due to its very high proximity »the field of view of the detectors 4, 5 and 6 substantially overlaps. In the present embodiment, the infrared detectors 4, 5 and 6 are preferably micromachined thin film or silicon thermopiles. Each of the thermopiles 4, 5 and 6 covers an opening 52 formed in the substrate 50. The openings 52 function as windows through which the radiation that is passed through the filters Fx, Fz and F3 is detected. of band. As is well known in the art, "thin-film or micro-phantom thermopile" sensors 4, 5 and 6 are fabricated on the bottom side of substrate 50, and any of a number of suitable designs can be employed. Figure 12 is an enlarged view of the underside of the substrate 50 »and illustrates a suitable pattern that could be used for thin film or micro-machined thermopile detectors 4, 5 and 6. A top view of the substrate 50 is provided in Figure 11. As is typical in this field, the hot seals 60 of each of the thermopile detectors 4, 5 and 6 are preferably supported on a thin, electrically insulating diaphragm 54. »Covering each of the openings 52 formed in the substrate 50» and the cold joints 62 are located on the thick substrate 50. Although the three openings 52 are preferably covered by an »thin» electrically insulating diaphragm 54 the detectors Thermopile can also be self-supporting. In operation, the infrared radiation from the infrared passive source enters the detector housing 31 through the window 44. The infrared radiation then strikes the interference bandpass filters F »FB and F" 3 » each of which allows radiation to pass within a previously defined spectral band, the radiation that the interference filters Fx, Fa, and F3 pass through, then collides with the diaphragm 54 or the hot seals 60, if the thermopiles are self-supporting , where it is detected by the infrared detectors 4, 5 and 6 of the thermopile respectively To improve the sensitivity of the detectors 4 and 5 to the incident radiation, the upper side of the electrically insulating diaphragm 54 can be coated with a thin film of bismuth oxide or carbon black during packaging, so that areas with an opening can absorb the incident radiation more efficiently. ermopila 4 »5 and 6 are self-supporting» then the side of the hot joints SO »on which the radiation is incident» can be coated with bismuth oxide or with carbon black »directly-. By placing the cold or reference joints 62 on coarse substrate 50, the reference joints of each of the detectors are inherently joined to the same thermal hub. Therefore, the substrate 50 acts as a heat sink to maintain the temperature of the cold joints 62 of each of the detectors at a common temperature. Additionally, the substrate 50 provides mechanical support for the device.

Although the present embodiment has been described as a single substrate 50, with three infrared detectors »4, 5 and 6» of ter opyl »formed therein» one skilled in the art will recognize that three separate substrates could be used »each of which there was in it an infrared detector with thermopile, manufactured therein, in place of the substrate 50 described in the present embodiment. An electrically insulating diaphragm 54 can be made from many suitable materials well known in the art. including a thin plastic film »such as MYLAR or an inorganic dielectric layer such as silicon oxide» silicon nitride or a multilayer structure »consisting of both. Preferably, the electrically insulating diaphragm 54 is an inorganic dielectric layer. thinner "because such layers can be manufactured easily using well-known semiconductor fabrication methods, and as a result, more sensitive thermopile detectors can be fabricated on the substrate. In addition, the possibility of manufacturing the entire device is significantly improved. Also by employing only semiconductor methods to manufacture the detectors 4. 5 and 6. the substrate 50 will have capsule circuit capabilities "characteristic of the devices that are based on full-scale silicon integrated circuit technology; thus, the electronic signal processing elements for the detectors 4 »5 and 6» if desired can be included in the substrate 50. Many other techniques are known for manufacturing thermopile detectors 4 »5 and 6. on the underside of substrate 50, in the techniques of thermal and infrared detectors. A suitable method for producing thermopile detectors 4, 5 and 6"using semiconductor processing techniques" is described in U.S. Patent No. 5,100,479 issued March 31, 1992, which is hereby incorporated herein by reference. reference. The construction of the detector assembly 3 according to one embodiment of the present invention is illustrated in Figures 3 and 4. As illustrated, the detector assembly is produced in a detector housing 31 such as a TO-5 can. The infrared detectors 4 »5 and 6 are mounted on a housing base 30 of the TO-5 can 31. The infrared detectors 4» 5 and S are very close to each other »so that the field of view of each detector substantial overlap to another. While a variety of infrared detectors can be used in the present invention, the detectors 4, 5 and 6 are preferably thermopiles "due to the fact that the thermal cells do not require any power" have a linear output and have a very high good ratio of signal to noise. Although not required, it is also preferable to link the reference joints of each of the three detectors to the same thermal heat sink.

The filter assembly 32 is arranged above the housing base 30, so that the only radiation that can enter the space between the filter assembly 32 and the housing base 30, is the radiation that enters through the three openings 34 located in the filter assembly 32. The openings 34 are located in the filter assembly 32 »so that each opening is in axial alignment with one of the detectors. The filters Fx »Fa and F3, of interference bandpass, cover the openings 34» so that they are interposed between the respective detector and the passive source of infrared light. Additionally »covering the three openings 34 located in the filter assembly 32» with interference filters Fx »Fa, and F3 ensures that the only radiation that can enter the space between the filter assembly 32 and the housing base 30» is that of the desired spectral bands. The divider 40 is used to prevent light from a spectral band from coming into contact with an infrared detector »intended to measure light from a different spectral band. The CWL and FWHM of the bandpass filters F2 and F3 are fixed as described in relation to figures 1-2 and table 1. The cover 42 for the can 31 TO-5 acts as an opening collar 7 and, so »defines the FOV for the detector assembly 3. The upper part of the cover 42 comprises a light transmitting window 44. In selecting the material for the window 44"it is preferred to select a material that is as transmittable as possible for the spectral bands that are being monitored by the detector assembly 3. Preferably the window 44 is also transmitter for each of the spectral bands that they are being watched. Window materials that have relatively uniform transmission qualities, on a scale of 1 μm to 10 μm. They include silicon. CAF-, and BaFa. Particularly preferred materials are CaFa and BaFjj. due to its high transmissivity on this scale. In order to save costs, the window 44 can be completely eliminated. However, when the window 44 is included, the detector assembly 3 illustrated in FIGS. 3 and 4 can be hermetically sealed and thereby increase the life expectancy for the detector assembly. Additional »as the dust and grease accumulate on the detector assembly 3» the output signal corresponding to the spectral bands will begin to fall. If the attenuation of the signal becomes too great, the infrared gas detector will not work properly. However, by including window 44 in assembly 3, the strength of the original signal can easily be restored by cleaning window 44. This is not possible if window 44 is omitted. If a larger platform is desired, so that additional detectors and additional bandpass filters can be added to increase the capabilities of the infrared gas detector of the present invention "a TO-8 or larger pack can be selected. For example, said platform could be used if the ability to monitor a plurality of gases is desired. A particularly preferred detector assembly 3 is now described with respect to Figures 8-15. As illustrated in Figures 8 »9 and 10» the detector assembly 3 includes three infrared detectors 4 »5 and 6» which have been formed on the substrate 50 mounted within the detector housing 31. The housing 31 of the detector of preference is a can TO-5 »consisting of a housing base 30 and a cover 42. The cover 42 includes an opening collar 7. which defines a port for receiving radiation within the detector assembly. The FOV of the detector assembly is therefore »limited by the opening collar 7» The cover 42 also preferably includes a light transmitting window 44 »which fits within or covers the port defined by the opening collar 7. The light transmitting window 44 is attached to the cover 42 so that when the cover 42 is attached to the base 30 »the infrared detectors 4» 5 and 6 are hermetically sealed within the detector assembly 3. The infrared detectors 4 »5 and 6 are supported on a substrate 50 which »in the present embodiment» is made of a semiconductor material, such as Si, Ge, GaAs or the like. Due to its very high proximity, the field of view of the detectors 4 »5 and 6 substantially overlap.

In the present embodiment, the infrared detectors 4, 5 and 6 are preferably thin-film or silicon-micromachined thermopiles. Each of the thermopiles 4, 5, and 6 covers an opening 52 formed in the substrate 50. The openings 52 function as windows through which the radiation that is allowed to pass through the filters Fx »Fjj» and F3 is detected of band. As is well known in the art, the thin-film or thermopile microphoto-thermopile 4, 5 and 6 detectors are manufactured on the underside of substrate 50 and any of a number of suitable designs can be employed. Figure 12 is an enlarged view of the underside of the substrate 50 »and illustrates a suitable pattern that could be employed for thin film or 4 micropylated thermopile detectors 4 and 5. A top view of the substrate 50 is provided in Figure 11. As is typical in this field, the hot seals 60 of each of the thermopile detectors 4, 5 and 6 are preferably supported on a thin, electrically insulating diaphragm 54. »Covering each of the openings 52 formed in the substrate SO. and the cold seals 62 are located on the thick substrate 50. While the three openings 52 are preferably covered by a thin electrically insulating diaphragm 54, the thermopile detectors can also be self-supporting. In operation »the infrared radiation of the infrared passive source enters the detector housing 31 through the window 44. The infrared radiation then strikes the interference bandpass filters F» F3 and F3 »each of which allows radiation to pass within a previously defined spectral band. The radiation that the interference filters Fx, Fa, and F3 pass through, then collides with the diaphragm 54 or the hot seals 60 »if the thermopiles are self-supporting» where it is detected by the infrared detectors 4 »5 and 6 of the thermopile »Respectively. To improve the sensitivity of the detectors 4 »5 and 6 to the incident radiation »the upper side of the electrically insulating diaphragm 54 may be coated with a thin film of bismuth oxide or carbon black during packing» so that the areas with aperture can absorb the incident radiation more efficiently . If the thermopile detectors 4. 5 and 6 are self-supporting »then the side of the hot seals 60» on which the radiation is incident »can be coated with bismuth oxide or with carbon black» directly. By placing the cold or reference joints 62 on coarse substrate 50, the reference joints of each of the detectors are inherently joined to the same thermal hub. Therefore, the substrate 50 acts as a heat sink to maintain the temperature of the cold seals 62 of each of the detectors at a common temperature. Additionally, the substrate 50 provides mechanical support for the device. Although the present embodiment has been described as a single substrate 50. with three infrared detectors, 4, 5 and 6"thermopile" formed therein, one skilled in the art will recognize that three separate substrates could be used, each which had in it an infrared detector with thermopile, manufactured therein, in place of the substrate 50 described in the present embodiment. An electrically insulating diaphragm 54 can be made of many suitable materials well known in the art "including a thin plastic film" such as MYLAR or an inorganic dielectric layer such as silicon oxide »silicon nitride or a multilayer structure» that consists of both. Preferably, the electrically insulating diaphragm 54 is a thin inorganic dielectric layer, because said layers can be easily fabricated using well-known semiconductor fabrication methods and as a result, 50 more sensitive thermopile detectors can be fabricated on the substrate. . In addition »the possibility of manufacturing the entire device is significantly improved. Also by employing only semiconductor methods to manufacture the 4 " 5 " sensors " the substrate 50 will have capsule circuit capabilities, characteristic of devices that are based on full-scale silicon integrated circuit technology,; thus, the electronic signal processing elements for the detectors 4 »5 and G ^ if desired may be included in the substrate 50. Many other techniques are known for manufacturing thermopile detectors 4» 5 and 6 »on the underside of substrate 50 »in the techniques of thermopile and infrared detectors. A suitable method for producing thermo optic detectors 4, 5 and G, using semiconductor processing techniques, is described in US Pat. No. 5,100,479 issued March 31, 1992, which is incorporated herein by reference. this reference. Referring to FIGS. 9 and 12, the output conductors 56 are connected to the female output connectors 64 of each of the thermopile detectors 4, 5 and 6, in the junction regions 58"using welding or other suitable materials. known. Because the reference joints of the detectors 4, 5 and 6 are thermally derived from each other, it is possible that the reference joints for each of the sensors 4,5 and 6. share a common female output terminal. As a result, it would only take four output drivers instead of six, to communicate the output of the detectors. The output conductors 56 typically connect the detectors 4. 5 and 6. to electronic signal processing elements. As mentioned above, however, the electronic signal processing elements can be included directly on the substrate 50 »and in that case output conductors 56 would be connected to the female input and output terminals of the electronic signal processing elements , instead of the female output terminals of the infrared thermopile detectors 4.5 and 6. As illustrated in FIG. 9, preferably a temperature sensing element is constructed on the substrate 50 »near the cold joints 62 of one of the thermopile detectors. The temperature sensor element monitors the temperature of the substrate 50 in the area of the cold joints and »in such a way» the temperature it measures is representative of the temperature of the cold joints 62. The »output of the temperature sensor element 53 is communicated to the electronic elements signal processors »so that the electronic signal processing elements can compensate for the influence of the ambient temperature of the cold joints of the thermopile detectors. The temperature sensor element 53 is preferably a thermistor, but other temperature sensing elements such as diodes, transistors and the like can also be used. Referring now to FIGS. 8 to 11, interference bandpass filters F-, are mounted. F3 »and F3 on the substrate 50» so that each of them covers one of the openings 52 in the substrate 50. The CWL and FWHM of the bandpass filters F »Fß and F3» are fixed as described with respect to figures 1-2 and table 1 above. Because the interference filters cover openings 52 »the light entering the detector assembly 3 through the window 44 must first pass through the filters xt Fx and F3» before reaching the infrared detector 4 »5 or 6» respectively. Thus »by using three separate openings in the substrate 50» the light that passes through one of the filters is isolated from the light that passes through the other filters. This prevents the crossing between each of the detector channels. Therefore »the light that reaches the infrared detectors 4» 5 and 6 »from the" passive source 8 of the infrared "is the light that falls within the spectral band intended to be measured by the particular detector. Fx »F ^ and F3 are secured to the substrate 50 using a thermally conductive material» such as thermally conductive epoxy.An advantage of securing the filters to the substrate 50. with a thermally conductive material, is that it improves thermal shunting between the filters and the substrate, which is at the same temperature as the reference. or the cold joints 62 of the thermopile detectors 4. 5 and 6. As a result »the background noise of the interference filters is reduced to a minimum. Like the interference bandpass filters FJ. »F2 and F3. they are each above 0 ° K release a certain amount of infrared radiation. The total radiant flux incident on a detector due to its filter »which normally has a temperature close to the ambient» is a function that the filter is also thermally derived to the reference or cold 62 joints of the detector. This takes into consideration how a thermopile works. That is, the output voltage generated by a thermopile is a direct measure of the difference in temperature between the (hot) signal junctions and the (cold) reference junctions of the thermocouples that make up the thermopile. A thermopile is nothing more than a large number of thermocouples connected in series to increase the output voltage of the device. Thus, that a filter is also thermally connected to the reference joints of your thermopile detector can affect the output voltage of the detector. In the worst case »when the filter is not thermally derived to the reference joints, at all» the incident radiant flux in the thermopile detector includes an undesirable deflection of the filter »which decreases the modulation of the desired signal from the passive source 8 of infrared that passes through the filter from the outside to the hot seals of the thermopile. The ratio of the usable signal to the non-usable signal in the hot seals is given by the ratio of the spectral radiant emittance of the infrared passive source 8 in the spectral band passed through the interference filter to the spectral radiated emitters emitted by the filter at all wavelengths, at 295 ° K. this could be as small as 2.3 x lO-3 »» for an interference filter that had a CWL of 4.67 μm and a FWHM of 0.2 μm. However, "in real situations" the filter is always somewhat derivative to the reference electrode of the termite detector "and the ratio of usable signal to the unusable signal is around 0.1 to 0.2. The present assembly mode detects the imine specifically »as far as possible» the undesirable radiant flux reaching the thermopile detector from the filter. This is done by providing a very efficient thermal bypass between the (cold) reference junctions 62 of the thermopile detectors 4 »5 and 6. and their corresponding interference filters Fx. F2 and F3. This, in effect. unify the influence of the filter on the (hot) 60 signal seals of the detector »thus making the radiation passed through the filters from the infrared passive source 8» the only source of the radiation that is measured by the thermopipe detectors to. There is no need to say that this is the only radiation that is important »and that it is now isolated in a useful way to process it by means of the thermopile detector. To further improve the thermal bypass between the filters and the substrate 50. other heat dissipating means can be provided. A) Yes. for example. a metal grid 68 »heat sink» can be deposited on one or both sides of the interference filters F ,? as shown in figure 13. The metal used for the grid must have high thermal conductivity. Gold is particularly well suited for this purpose. Alternatively »as shown in FIGS. 15 and 16» a metal grid »heat dissipating 68 can be incorporated in a mounting fitting 70. The thermal conductivity of the metal grid 68 can be improved by coating the grid with gold. Mounting attachment 70 comprises a portion of louvers 68 and a portion 72 of raised lip. The interference filter F., (corresponding to the FA filters, Fa or F3) is seated in the depression formed by the raised lip 72. To improve the heat transfer between the mounting attachment 70 and the filter F-, the filter F. is preferably attached to the mounting fitting 70. using a thermally conductive material »such as a thermally conductive epoxy. It is then attached to the mounting attachment using a thermally conductive material "to the top of the substrate 50" to cover the opening 52. This is illustrated in FIG. 15, which is a partial sectional view through the substrate 50"in FIG. one of the thermopile detectors D-, corresponding to the detectors 4. 5 or 6. Another detector assembly 79 »particularly preferred» is now described with respect to figures 16 and 17. The detector assembly 79 includes three infrared detectors 4 »5 and 6 (not shown) formed in the semiconductor substrate 80» mounted within the detector housing 31. Infrared detectors 4 »5 and 6 are detectors thin film or micromachined thermopile infrared 'formed in the underside of the substrate BO" as described relative to the embodiment of detector assembly illustrated in Figures 8-12.

The primary difference between the substrate 80 of the present embodiment and the substrate 50 of the embodiment illustrated in FIGS. 8-12 is that the substrate 80 includes an enhanced eyebrow B2"surrounding each of the three openings 52 formed in the substrate. The raised eyebrows 82 give an additional thermal mass to maintain the temperature of the reference (cold) joints of the thermopiles 4 »5 and 6» at the same temperature. The additional thermal mass is convenient in the present embodiment, since the detector assembly 79 according to the present embodiment also includes an active infrared light source 84 operatively mounted within the detector housing 31. As with the embodiment illustrated in FIGS. 8-12, preferably a temperature sensing element 53 (not shown) is constructed on the substrate 50 near the cold junctions of one of the thermocouple detectors so as to monitor the temperature of the cold joints and provide that information to the electronic elements processing signal. The source 84 of infrared light also gives greater flexibility to the detector assembly 79. That is to say it allows the use of the three-channel detector assembly 79, in a traditional NDIR gas sensor »having an active source of infrared or light. alternatively, if the source 84 of infrared light is disabled, the detector assembly 79 can be used in a passive infrared gas sensor. according to the present invention. When the detector assembly 79 is used in an NDIR gas sensor, "the added thermal hammer" provided by the raised eyebrows 82"helps to have the temperature of the reference gaskets at a temperature as uniform as possible when the active source 84 of light Infrared is performing connection and disconnection cycles. This is useful to maintain the sensitivity of the detectors to the modulation in s? signal »due to the presence of gas or gases that are being monitored in the sample path of the detectors. The detector housing 31 of the present embodiment is a can TO-5 »consisting of a housing base 30 and a cover 42. The cover 42 includes an opening collar 7 defining a port for receiving radiation in the detector assembly. The FOV of the detector assembly 79 is limited by the opening collar 7. Because they are very close to each other, the FOV of the detectors 4, 5 and 6 substantially overlap. The cover 42 also preferably includes a light transmitting window 44 which fits into or covers the port defined by the opening collar 7. The light transmitting window 44 is attached to the cover 42 so that when the cover 42 is joins the base 30 »the infrared detectors 4» 5 and 6 are hermetically sealed within the detector housing 31. The material used for the window 44 should be selected as described in relation to the detector assembly modes illustrated in FIGS. Figures 3 »4 and 8-15.

The interference bandpass filters F »F3 and F3 are mounted on the upper part of the raised eyebrows 82» so that each covers one of the openings 52 of the substrate 80. The CWL and the FWHM of the pass filters band FL 'F "a and F3 are regulated as described with respect to the previous Figures 1-2 and Table 1. Because the interference filters cover the openings 52» the light entering the detector housing 31 » through the window 44 »you must first pass through the filters Fx. Fg., 0 F3 before reaching the infrared detector 4. 5 or 6. respectively, in such a way» when using 3 separate openings in the substrate 80 »the light that passes through one of the filters is isolated from the light that passes through one of the other filters.This prevents the crossing between each of the detector channels.Therefore »the light that reaches the infrared detectors 4. 5 and 6 »from the infrared passive source 8» or from the source it activates 84 infrared »if you are using the detector assembly in a conventional NDIR gas sensor "is the light that remains within the spectral band to which the particular detector is intended to measure. The interference bandpass filters Fx »F ^ and F3 are attached to the upper part of the raised eyebrows B2 surrounding the openings 52» using a technically conductive material »such as technically conductive epoxy. An advantage of securing the filters to the raised eyebrows 82 with a thermally conductive material is that it improves the thermal bypass between the filters and the substrate 80 which is at the same temperature as the reference or cold joints of the thermocouple detectors 4. »5 and G. As a result» the background noise of the interference filters is reduced to a minimum. In order to further improve the thermal derivation between the filters and the substrate 80, a metal grid 68 »heat dissipater» can be deposited on one or both sides of the interference filters F +, as shown in Figure 13. The metal used for the grid must have good thermal conductivity. Gold is particularly well suited for that purpose. Finally, as shown in FIG. 14, a heat-dissipating metal grid 68 can be incorporated in a filter assembly attachment 70. Mounting attachment 70 comprises a grid portion 68 and a raised lip portion 72. As illustrated in FIG. 15, an interference filter F., (which corresponds to the filters Fx, Fa and F3) is hinged in the depression formed by the raised lip 72. To improve the heat transfer between the attachment 70 and filter F., preferably the filter F is attached to the mounting attachment 70 using a technically conductive material such as technically conductive epoxy. The mounting attachment is then attached using a thermally conductive material "to the top of one of the raised eyebrows 72 to cover the opening 52. The attachments 86 of the substrate assembly are connected to the female output terminals (not shown) of each of the thermocouple detectors 4 »5 and S» in junction regions 88 »using welding or other well-known materials. Since the reference joints of the detectors 4 »5 and 6» share a common female output terminal in the present embodiment, only four substrate mounting attachments 86 are needed to communicate the outputs of the detectors. The substrate mounting attachments 86 are insulated from the base 30 of the detector housing 31 because they are mounted on the electrically insulating substrate 90 which is preferably constituted by a material selected from the group consisting of aluminum oxide and sodium oxide. beryllium. The output signal of the detectors 4 »5 and 6» is communicated by means of the substrate mounting attachments 86 »through the wire junctions 94» to the electronic elements 92 signal processors. The electronic signal processing means 92 may consist of a plurality of microcapsules attached by matrix to the insulating substrate 90 or a single icrocapsule matrix-bonded to the insulating substrate 90. The output conductors 56 are connected to the input and output of the electronic elements. 92 signal processors »by means of wire junctions 96. The electronic elements 92 signal processors include a source driver 9B for driving the active infrared source" 84"at a known frequency. The source driver 98 drives the active infrared source 84 through the wire junctions 97. The manner in which the active infrared source 84 is to be driven by the source driver 98 for "conventional NDIR applications" is well known in the art and it is not necessary to explain it further here. While the detector assembly 79 including electronic signal processing elements 92, arrayed on the insulating substrate 90, has been illustrated, the electronic signal processing elements 92 could be incorporated directly into the semiconductor substrate 80. Alternately, to simplify the assembly detector 79 »the output conductors 56 could be connected directly to the outputs of the detectors 4» 5 and 6 »using welding or other well-known materials. The output conductors 56"in that situation" would connect the outputs of the thermocouple detectors 4 »5 and 6» of the infrared »to the signal processing circuits on the outside of the detector assembly 79. If the detector assembly 79 is used in a passive infrared gas sensor »according to the present invention» the infrared radiation from the infrared passive source 8 enters the detector housing 31 through the window 44. The infrared radiation then strikes the pass filters of the infrared. interference band F »Fß and F3» each of which allows radiation to pass within a predefined spectral band. The radiation that passes through the Fx interference filters, F3 and * F3 »then collides with the inorganic dielectric membrane shown), which covers each of the openings, or hot joints if the thermopiles are self-supporting, where it is detected by means of thermocouple detectors 4, 5 and 6» of the infrared »respectively. The outputs of each of the detectors are then communicated to the electronic signal processing elements "where they are processed according to the description of the infrared gas passive sensor" provided further back with respect to Figures 1-2 and Table 1 As with the detector assembly 3 described with respect to FIGS. 8-15, the sensitivity of the detectors 4 and 5 to the incident radiation can be improved by coating the upper side of the dielectric membrane (not shown) with a film. thinning of bismuth oxide or carbon black during packaging "so that the opening areas can more efficiently absorb the incident radiation. If the thermopile detectors 4» 5 and 6 are self-supporting »then the side of the hot joints on which radiation is incident, can be coated with bismuth oxide or carbon black »directly. As stated above. because the detector assembly 79 also includes an active infrared source »can be used in an NDIR gas sensor. The use of detector assembly 79 in an NDIR gas sensor "according to the present invention" is described below.

Figure 5 shows a real implementation of a preferred embodiment of a PIA gas sensor 33 according to the present invention. The detector assembly 3 is mounted directly on the printed circuit board (PCB) 11 »which is also a mounting for the electronic elements 12 signal processors» the siren 13 for sounding an alarm »and a power supply 14 per battery. The battery power source 14 is preferably a lithium battery, which must provide sufficient power to operate the system for one to two years. Although the gas sensor PIA hereof is illustrated using a detector assembly 3, the detector assembly 79 described with respect to FIGS. 16 and 17 may also be employed in the present embodiment. The spherical reflector 15 »which is rigidly fixed to the detector assembly 3» is used to increase the FOV of the detector assembly 3. The length of the sample path for the gas sensor »in this case» is again defined by the distance between the detector assembly 3 and the passive source 8 of infrared »which is defined as a portion of the wall 9. The PCB 11 carrying all the above-described components» is housed in an enclosure 16 for protection against handling and against external environments, when it is being used to implement the PIA technique of the present invention. Whoever is skilled in the art will recognize that the FOV of the detector assembly 3 can be similarly increased by using an optical refractor system instead of the optical reflector system. Optical reflector elements are preferred, however »because of their cost. Figure 6 shows the schematic drawing for the signal processing circuits according to a preferred embodiment for the present invention. The signal processor circuits illustrated in FIG. 6 can be used in conjunction with any of the detector assembly embodiments described above. According to the present embodiment, the infrared radiation emanating from the passive source (not shown) is collected within the FOV of the detector assembly 3 on the detectors 17, 18 and 19, respectively representing the Dx signal detector. and the neutral detectors Da and D3. The detectors 17, 18 and 19 are thermopile detectors and their reference joints are thermally bonded to the same heat sink 20. One of the main advantages of thermopile detectors is their linear output (linearly scalable with temperatures from 0 to 70). ° C). Thus »the outputs of the detectors 17» 18 and 19 can be corrected for changes in the ambient temperature »when sensing it in the heat sink 20 of the common reference joint» using the microprocessor 21. Although the present embodiment is illustrated by using the sensor assembly 3, the detector assembly 79 described with reference to FIGS. 16 and 17 could also be employed in the present embodiment. In order to minimize the CD traps, the outputs of each of the three detectors are switched. subsequently with the same working factor »by a low noise manifold 22» controlled by the microprocessor 21 »to the differential input of the same low noise preamplifier 23. Then the amplified signals» are converted by means of a 24 DA converter / D »before feeding them to the microprocessor 21» to process the signal. After the gas that has been measured is detected, the gas concentration can be monitored based on the predetermined function programmed in the microprocessor 21. The concentration can be sent as output or displayed using the cable 25 or in some cases also an alarm signal can be generated by the microprocessor 21 »using the cable 26. The microprocessor 21 is of the low power type and contains sufficient RAM. ROM and EEprom to properly process the signals originated by the detector assembly. It could increase additional the versatility of the passive infrared gas detectors of the present invention "adding a distance measuring device to the gas detector. This would allow the user to quickly and easily modify the sample length S »depending on the application. The distance measuring device could be of the contact or non-contact type. For example »it could comprise a laser diode with a sensor» as is well known in the art. The output of the distance measuring device would be communicated to the signal processor »so that the appropriate sample path length S could be inserted into the CID equation» when calculating the gas concentration. As discussed above, the change in path length is not necessary to calculate the ratio of the outputs of the two neutral channels "because this factor would be canceled" since it would be the same for both detectors. Alternatively, the gas detector can include a switch so that the user can enter predetermined path lengths. For example, the switch could include path length settings that would increase in increments of 30 centimeters "so that the user could measure and then enter the appropriate path length for the installation" in which the gas detector is being used. the present invention. The selection of a particular path length is communicated to the microprocessor 21 »so that it knows the appropriate path length that it will use to calculate the concentration of the gas in the sample volume. For a slightly greater flexibility "a female data input terminal" can be used so that the user can enter any desired path length and the microprocessor 21 will be compensated accordingly during its calculations.

Figure 7 is a schematic circuit for a signal processor according to another embodiment of the present invention. The structure of the circuit is determined by the low level of the expected signals »of the order of 5 to 85 icrovolts. There are three identical preamplifier circuits that differ only in the value of a gain-fixing resistor R4. The amplifiers are built in the form of instrumentation amplifiers that have a very high common signal rejection mode "'because for home operation" large signals can be induced magnetically by 60 Hz electrical power wiring. Magnetic shielding of detectors and circuits should reduce this. Detectors and circuit components should also be protected against rapid temperature changes that may allow thermocouple signals in the components. The thermal and mechanical designs are very important to allow the full capabilities of the electronic circuit. Ul forms the input part of the instrumentation amplifier. It was selected by s? very low input separation voltage »about 0.5 μV» and a very low change with respect to that voltage »with temperature. For a high common mode rejection, the two feedback resistors R2 and R3 must equal more than O.i? "And must have temperature coefficients of 10 pp / degree C or better. The gain of this circuit is determined by the ratio of R2 and R3 to Rl »around 500. The noise level for at 10 Hz is approximately 2 μV pp. This is greater than desirable "but can be filtered later. The deviation and the low input shift »with the temperature» are more important to obtain an appropriate processing of the sampled outputs. The input noise level of the output part of the circuit is much smaller »around O.28 μV» but the displacement is much higher, around 50 μV. and with a higher temperature coefficient. U2 really is another instrumentation amplifier. It is used to provide a high and stable gain of around 400. It is used because it is less expensive than another amplifier and four precise gain-fixing resistors. The expected output is 1 to 2 volts or more »depending on the radiation input to the detector. The gains of the preamplifiers for the other two detectors are smaller »since more radiation is expected at the larger wavelengths of these det ectors. The rest of the signal processing could be handled in many different ways »and an implementation is shown as an example. The three signal channels and a temperature sensor, near the detectors »are selected by a multiplexer and their value is converted to a frequency by a frequency-to-voltage converter. The frequency output can be easily processed by a processor micro (μP) »to determine the temperature of the observed scene» the temperature of the detectors »then the absorption due to the CO gas or other gases to be measured from the signals expected at those temperatures. Another embodiment of a PIA gas sensor according to the present invention is described with reference to FIG. 18. The passive infrared gas sensor 110 illustrated in FIG. 18. comprises a passive infrared source 112, a detector assembly. infrared 3"of three channels" centered in the middle of the infrared passive source 112"and having a port 118 for receiving infrared radiation therethrough; and a concave mirror 120 »spaced from port 118 and facing it» in detector assembly 3 and passive source 112 of infrared. The passive infrared source 112 is preferably concave to increase the surface area of the infrared source "within the field of view of the concave mirror 120 that faces the infrared passive source. In the present embodiment, the infrared passive source 112 comprises a black infrared surface 116 which has been applied to the surface of a non-conducting member 114. The black infrared surface 116 may comprise several materials, including black chromium oxide, bismuth oxide and carbon black. The non-conductive member 114 consists of a plurality of plastic panels due to the light weight and ease of fabrication associated with the plastic. As would be obvious to those skilled in the art, member 114 could also be made of a unitary piece of plastic or other electrically insulating material. The mirror 120 may be any concave reflecting surface, so as to increase the field of view of the detector assembly. Preferably the concave mirror 120 has the lowest possible emissivity, so that all the infrared radiation that is being received through the port 118, in the detector assembly 3, is produced by the infrared passive source 112. The concave mirror 120 should be large enough to comprehend the entire field of view of the detector, at the distance at which it is separated from the detector. This can be calculated using the equation d = 0MxS / 2tr-a - ^ =. where d is the diameter of the concave mirror. OM is the solid angle subtended by the detector assembly in the concave mirror 120 and S is the distance between the detector assembly and the detector assembly. Similarly, the infrared passive source 112 must be large enough to fill the field of view of the concave mirror. The space between the infrared passive source and the detector, on the one hand, and the concave mirror on the other »defines the sample chamber of the passive infrared gas sensor 110» according to the present invention. The infrared radiation emitted by the passive infrared source 112 is reflected by the concave mirror 120 towards the detector assembly 3 through the port 118. As a result »the length of the sample path of the gas sensor 110 by infrared passivity is by at least twice the distance between the detector assembly 3 and the concave mirror 120. This allows the gas sensor 110 to have twice the sensitivity as the passive infrared gas sensor in which the passive source is opposite the detector assembly . Alternatively, the gas sensor will have the same sensitivity using half the space. As explained above, the detector assembly 79 described with respect to Figures 16 and 17 also includes an active infrared source.; in such a way »the detector assembly 79 can be used directly in an NDIR gas sensor. A potential NDIR gas sensing device »according to the present invention» is illustrated in Figure 19. The NDIR gas sensor of Figure 19 comprises an elongated hollow tube 100 having a closed end 102 and an open end 104. In the preferred embodiment »the tube 100 is composed of a metal and has a circular cross section. In other modalities »the cross section is square. The inner surfaces of the tube 100"including the inner surface of the closed end 102" are specularly reflective. According to the present invention »the metal tube 110 is gas-tight» and therefore »filtering apertures are provided, of which the filtering aperture 106 is typical, at locations spaced along the tube 100 to allow it to be monitored the gas that enters and leaves the space inside the tube. Each of the filtering apertures 106 is covered with a semi-permeable membrane 108. The exact number, location and arrangement of the filtering apertures are not crucial. although some provisions may be better than others. The three-channel detector assembly 3 is mounted on the open end of the hollow tube 100 in a manner that closes the open end and prevents gas entry or its exit through the open end of the tube 100. Because it is the active infrared source 84 in the present invention uses the detectors 5 and 6 which are used as neutral detectors to characterize the temperature of the infrared passive source 8 when the detector assembly 79 is used in a passive infrared gas sensor , in accordance with the present invention, are not necessary. As a result, the detector assembly 79 can be used to monitor the concentration of up to 3 different gases in the present mode "simply by passing bandpass filters Fx. F? and F3 that allow spectral bands to pass at three different wavelengths "to which the three different gases that are going to be detected absorb the radiation strongly" and to which other gases that would be present do not absorb. If it is necessary to detect less than three gases »it is possible to disable the unnecessary detector channels. This greatly increases the flexibility of the NDIR gas sensor according to the present invention.

The concentration of the gases to be detected within the sample chamber is determined by the extent to which they absorb the radiation emitted from the infrared reactive source 84. By inserting the detector assembly 79 into the open end of the tube 100 they are arranged first the window illustrated in figure 19 »the detectors 4» 5 and 6 »the interference bandpass filters Fx > F2 and F, and the active infrared source 84 »located within the detector assembly 79» so that all are facing the inner surface of the closed end 102. As a result »some of the radiation emitted by the active infrared source 84 eß reflected »is already directly or indirectly» from the internal surface of the closed end 102 »to the detectors 4» 5 and 6 »where it is detected. The amount of radiation detected in the spectral bands monitored by detectors 4 »5 and 6» can be used to determine the concentration of gases being monitored within the sample chamber defined by the space within the 100 ° tube using techniques well known in this field. The purpose of the semi-permeable membrane 108 is to prevent airborne particles »larger than a predetermined size» from entering the space within the 100 ° tube, while at the same time not appreciably interfering with the free diffusion of the gas that is it is monitoring to and from the space within the tube 100. Undesirable particles include tiny moisture droplets or oil droplets and also include particulate fine matter, eg, dust or smoke particles. If these undesirable particles present in the air enter the space within the tube 100, they would be deposited on the mirror reflecting surfaces, thereby reducing the reflectivity and destroying its specular nature. The undesirable particles would also be deposited on the window 44 of the detector assembly 79 »reducing the transmission of the radiation. All these problems are eliminated by the use of the semipermeable membrane which »in the present embodiment» prevents particles present in the air »greater than 0.3 microns from entering the space within the tube 100. Although the present inventions have been clarified in the lustrative modalities »it will be immediately obvious to those skilled in the art that many modifications of structure» arrangement »proportion» e »elements and materials used in the practice of the described inventions, and otherwise» that are particularly adapted to specific environments and functional requirements "can be carried out without departing from the principles described. For example, "the sensor assemblies described with respect to FIGS. 8-17" were described as three-channel detectors because they are being used in the PIA sensor according to the present invention. However, as will be recognized by one skilled in the art, the detector assemblies of the present invention could be easily modified to have any desired number of channels including one depending on the specific application. Thus, it is to be clearly understood that this description is given only by way of example and not as a limitation for the scope of the described inventions which are claimed in what follows.

Claims (58)

1. NOVELTY OF THE INVENTION CLAIMS
2. An infrared detector assembly characterized in that it comprises: (a) a detector housing having a port for receiving infrared radiation therethrough; (b) a substrate mounted within the detector housing; the substrate having three openings; (c) a first thermopile detector, a second and a third thermopile detector »manufactured on the bottom side of the substrate; the hot seals of each thermopile detector located on one of the openings of the substrate "so as to receive the radiation transmitted through the opening; and the cold seals of each thermopile detector located on the substrate; (d) a first interference bandpass filter »mounted on the upper side of the substrate» so that the first filter covers the opening above the first detector and the first filter is interposed between the port and the first detector, being designed the first interference bandpass filter for passing the incident radiation to a first spectral band "(e) a second interference bandpass filter, mounted on the upper side of the substrate, so that the second filter covers the opening above the second detector, and the second filter is interposed between the port and the second detector; the second interference bandpass filter being designed to let the radiation pass to a second spectral band; and (f) a third interference bandpass filter, mounted on the top side of the substrate, so that the third filter covers the opening above the third detector, and the third filter is interposed between the port and the third detector. »The third interference bandpass filter being designed to let radiation pass to a third spectral band. 2. An infrared detector assembly according to claim 1 »further characterized in that it comprises output conductors extending through the detector housing and electrically connected to the first and second and third thermopile detectors.
3. 3. An infrared detector assembly according to claim 2 »further characterized in that the substrate consists of a semiconductor material and the first» second and third thermopile detectors are selected from the group consisting of thin film thermopile detectors and detectors of micromachined thermopile.
4. 4. An infrared detector assembly according to claim 1 further characterized in that the substrate consists of a semiconductor material and the first, second and third thermopile detectors are micromachined thermopile detectors.
5. 5. An infrared detector assembly according to claim 4, further characterized in that it further comprises: (a) a signal processor manufactured on the substrate; the signal processor being electrically connected to the second and third thermopile detectors »first»; and (b) conductors that extend through the detector housing and electrically connected to the signal processor.
6. 6. An infrared detector assembly according to claim 1 »further characterized in that each of the first» second and third infrared thermopile detectors are formed in an electrically insulating diaphragm covering the opening on which it is located the detector.
7. 7. An infrared detector assembly according to claim 6 »further characterized in that the electrically insulating diaphragm consists of a thin plastic film.
8. 8. An infrared detector assembly according to claim 7 »further characterized in that the plastic film is MILAR < - > .
9. 9. An infrared detector assembly according to claim 6 »further characterized in that the electrically insulating diaphragm consists of a dielectric and inorganic membrane selected from the group consisting of silicon oxide» silicon nitride and a multilayer oxide structure of silicon and silicon nitride.
10. 10. - An infrared detector assembly according to claim 1 »further characterized in that the first» second and third interference bandpass filters are bonded to the substrate using a thermally conductive material.
11. 11. An infrared detector assembly according to claim 1. further characterized in that it further comprises heat dissipating means for improving thermal shunting between the substrate and the first and second and third interference bandpass filters.
12. 12. An infrared detector assembly according to claim 1 »further characterized in that it additionally comprises a" light transmitting window "mounted inside the port.
13. 13. An infrared detector assembly according to claim 12 »further characterized in that the substrate» the filters and detectors are hermetically sealed within the detector housing.
14. 14. An infrared detector assembly "characterized in that it comprises: (a) a detector housing having a port for receiving through the same infrared radiation; (b) a semiconductor substrate mounted within the detector housing, the substrate having three openings; (c) a dielectric membrane covering each of the three openings »and which is formed on the bottom of the substrate; (d) a first »second and third thermopile detectors» the hot seals of each thermopile detector being formed on one of the openings in the dielectric membrane covering the opening »and the cold seals of each thermopile detector being formed on the substrate; (e) a first interference bandpass filter »mounted on the upper side of the substrate» so that the first filter covers the opening above the first detector and the first filter is interposed between the port and the first detector; the first interference bandpass filter being designed to pass the incident radiation to a first spectral band; (f) a second interference bandpass filter »mounted on the upper side of the sub-layer» so that the second filter covers the opening above the second detector »and the second filter is interposed between the port and the second detector; the second interference bandpass filter being designed to let the radiation pass to a second spectral band; (g) a third interference band filter filter »mounted on the upper side of the substrate» so that the third filter covers the opening above the third detector »and the third filter is interposed between the port and the third detector; the third interference bandpass filter being designed to let the radiation pass to a third spectral band; and (h) conductors extending through the detector housing and electrically connected to the first and second and third thermopile detectors.
15. 15. An infrared detector assembly according to claim 14 »further characterized in that the dielectric and inorganic membrane is selected from the group consisting of silicon oxide, silicon nitride and a multilayer structure of silicon oxide and nitride. yes 1 icio.
16. 16. An infrared detector assembly according to claim 14 »further characterized in that it additionally comprises a signal processor manufactured on the substrate» and wherein the first »second and third thermopile detectors are electrically connected to the signal processor» and the conductors are electrically connected to the signal processor.
17. 17. An infrared detector assembly according to claim 14 »further characterized in that the first» second and third interference bandpass filters are attached to the substrate using a thermally conductive material.
18. 18.-An infrared detector assembly "characterized in that it comprises: (a) a detector housing that has a port for receiving through the same infrared radiation; (b) a semiconductor substrate mounted within the detector housing, the substrate having three openings and one raised eyebrow on the top of the substrate »surrounding each of the openings; (c) a dielectric membrane covering each of the three openings »and formed on the lower part of the substrate» (d) a first »second and third thin film thermopile» sensors; the hot seals of each thermopile detector being formed on one of the openings in the dielectric membrane covering the opening; and the cold seals of each thermopile detector being formed on the substrate »(e) a first interference bandpass filter» mounted on the upper part of the raised eyebrow encircling the opening on which the first detector is formed » so that the first filter is interposed between the port and the first detector; the first interference bandpass filter being designed to let the radiation pass to a first spectral band; (f) a second interference bandpass filter »mounted on the upper part of the raised eyebrow encircling the opening in which the second detector is formed. so that the second filter is interposed between the port and the second detector; the second interference bandpass filter being designed to let the radiation pass to a second spectral band; and (g) a third interference bandpass filter "mounted on the upper part of the raised eyebrow encircling the opening on which the third detector is formed" so that the third filter is interposed between the port and the third. detector; the third interference bandpass filter being designed to let the radiation pass to a third spectral band.
19. 19. An infrared detector assembly according to the indication 18 »further characterized in that it further comprises an active source of infrared light» operatively mounted within the detector assembly.
20. 20. An infrared detector assembly according to claim 19 »further characterized in that the inorganic dielectric membrane is selected from the group consisting of silicon oxide» silicon nitride and a multilayer structure of silicon oxide and silicon nitride. icio.
21. 21. An infrared detector assembly according to claim 19 »further characterized in that it additionally comprises output conductors extending through the detector housing» and electrically connected to the thermopile detectors »first, second and third.
22. 22. An infrared detector assembly according to claim 19. further characterized in that it further comprises: (a) a signal processor manufactured on the substrate; the signal processor being electrically connected to the thermopile detectors; and (b) conductors extending through the detector housing and electrically connected to the signal processor.
23. 23. An infrared detector assembly according to claim 19. further characterized in that the first »second and third interference bandpass filters are attached to the substrate using a thermally conductive material.
24. 24. An infrared detector assembly according to claim 19 »further characterized in that the active source of infrared light comprises a tungsten filament.
25. 25. An infrared gas sensor »of passive source» characterized in that it comprises: (a) a detector housing having a port for receiving through the same infrared radiation »(b) a substrate mounted inside the detector housing» having the substrate three openings; (c) a first »second and third thermopile detectors, manufactured on the underside of the sub-layer; arranging the hot seals of each thermopile detector on one of the openings of the substrate »in order to receive the transmitted radiation» through the opening; and the cold joints of each thermopile detector on the substrate being located; (d) a first interphase bandpass filter »mounted on the upper side of the substrate» so that the first filter covers the opening above the first detector, and the first filter is interposed between the port and the first detector; the first interference bandpass filter being designed to pass the incident radiation to a first non-neutral spectral band. which is absorbable by a preselected gas to be monitored; (e) a second interference bandpass filter »mounted on the upper side of the substrate, so that the second filter covers the opening above the second detector, and the second filter is interposed between the port and the second detector; the second interference bandpass filter being designed to let the radiation pass to a first neutral spectral band; (f) a third interference bandpass filter, mounted on the upper side of the substrate, so that the third filter covers the opening above the third detector and the third filter is interposed between the port and the third detector; the third interference bandpass filter being designed to let the radiation pass to a second neutral spectral band; and (g) signal processing circuits, connected to the electrical output β produced by the first, second and third detectors "to produce a signal in response to aliases, representative of the concentration of the gas being measured.
26. 26. A "passive source infrared gaß sender" according to claim 25 »further characterized in that the gas being monitored is at least one selected by the group consisting of CO CO.,. O »and TVOC.
27. 27.- An infrared gas sensor »of passive source according to claim 25» further characterized in that the first »second and third bandpass filters have an amplitude of 0.1 μm to FWHM.
28. 28. - A passive source infrared gas sensor according to claim 27 «further characterized in that the second and third bandpass filters have a central wavelength selected from the group consisting of 3.91 μm. 5.00 μm and 9.00 μm.
29. 29. An infrared gas sensor of passive source, according to claim 28. further characterized in that the first bandpass filter has a central wavelength selected from the group consisting of about 4.26 μm and about 4.67 μm .
30. 30. An infrared gas sensor, characterized in that it comprises: (a) a detector housing having a port for receiving infrared radiation therethrough; (b) a sub-layer mounted within the detector housing. the substrate having three openings; (First, a second and a third thermopile detector, manufactured on the underside of the substrate, with the hot seals of each thermopile detector located on one of the openings in the substrate, so as to receive the radiation transmitted to the substrate. through the opening »and the cold seals of each thermopile detector are arranged on the substrate; (d) a first interference bandpass filter, mounted on the upper side of the substrate» so that the first filter covers the opening on top of the first detector and the first filter is interposed between the port and the first detector, the first interference bandpass filter being designed to let the incident radiation pass to a first spectral band, (e) a second pass filter of interference band »mounted on the upper side of the substrate, so that the second filter covers the opening above the second detector» and the second filter is interposed between the port and the second detector; with the second interference band screen filter being designed, the radiation passes to a second spectral band; (f) a third interference bandpass filter. mounted on the upper side of the substrate »so that the third filter covers the opening above the third detector» and the third filter is interposed between the port and the third detector; the third interference bandpass filter being designed to let the radiation pass to a third spectral band; (g) an active source of infrared light »operatively mounted within the detector assembly; (h) a source exciter "electrically connected to the active infrared source" to excite the active source of infrared light at a predetermined frequency; and (i) signal processing circuits connected to the electrical outputs produced by the first »second and third sensors» to produce a signal in response to them »representative of the concentration of at least one gas being monitored.
31. 31.- An infrared gas sensor »of passive source» according to claim 30 »further characterized in that the gas being monitored is at least one selected from the group consisting of CO» C0β »H30 and TVOC.
32. 32. An infrared gas sensor of passive source »according to claim 30» further characterized in that the first »second and third bandpass filters have an amplitude of approximately 0.1 μm» to FWHM.
33. 33.- A passive source infrared gas sensor »according to claim 32» further characterized in that the second and third bandpass filters have a central wavelength selected from the group consisting of 3.91 μm »5.00 μm and 9.00 μm.
34. 34.- A passive source infrared gas sensor »according to claim 33» further characterized in that the first bandpass filter has a central wavelength selected from the group consisting of about 4.26 μm and about 4.67 μm. μm.
35. 35.- A passive source infrared gas sensor, characterized in that it comprises: (a) a passive infrared source, comprising a black infrared surface; (b) A three-channel infrared detector assembly centered on the infrared passive source; the detector assembly having a port for receiving infrared radiation through it; and (c) a concave mirror facing the detector assembly port and the black infrared surface; the concave mirror being positioned in such a way that the radiation emitted from the infrared passive source is reflected from the mirror towards the port.
36. 36.- A passive source infrared gas detector »according to claim 35» further characterized in that the black infrared surface comprises a material selected from the group consisting of black chromium oxide »bismuth oxide and carbon black.
37. 37.- a passive source infrared gas detector »according to claim 35» further characterized in that the infrared passive source comprises a black infrared surface »concave.
38. 38.- a passive source infrared gas detector »according to claim 35» further characterized in that the three-channel detector assembly comprises three thermopile detectors for the infrared and three interference bandpass filters; each filter being arranged in the optical path between the passive infrared source and one of the thermopile detectors.
39. 39.- An infrared gaß detector »of passive source» characterized in that it comprises: (a) an infrared detector assembly »comprising: (i) a port to receive through the same radiation from the infrared passive source» (ii) a first sensor »a second sensor and a third sensor» arranged to receive the radiation through the port »to produce a first salt» a second outlet and a third »outlet indicating the radiation incident on the first sensor »The second sensor and the third sensor» respectively; (iii) a first narrow bandpass filter, interposed between the port and the first sensor »the first narrow bandpass filter producing an output indicating the radiation incident on the first bandpass filter» to a first band non-neutral spectral, which is absorbable by a selected gas that is being detected; (iv) a second narrow bandpass filter, interposed between the port and the second sensor »the second narrow bandpass filter producing an output indicating the radiation incident on the second bandpass filter, to a first band neutral spectral; and (v) a third narrow bandpass filter, interposed between the port and the third sensor; the third narrow bandpass filter producing a signal indicating output of the incident radiation on the third bandpass filter »to a second neutral spectral band; (b) temperature measuring means for producing an output q? e corresponds to the ambient temperature of the first »second and third sensors; (c) signal processing means, adapted to receive the outputs of the first »second and third sensors» and temperature measuring means and to sample and at least temporarily store the outputs of the first sensor »of the second sensor» of the third sensor and of the temperature measuring means »at pre-set intervals; including the average signal processing means for: (i) correcting the stored-out sails of the first sensor »of the second sensor and the third sensor» to compensate for the ambient temperature of the first sensor »of the second sensor and the third sensor, respectively» at the moment of taking the mueßtra; (ii) calculate the temperature of the infrared passive source at the time of taking the sample »based on the proportion of the corrected values of the outputs of the second and third sensors; (iii) calculate a predicted output for at least? of the second or third sensors "based on the calculated temperature of the infrared passive source" for the sampling period "(iv) calculate an attenuation factor by comparing the predicted output of at least one of the second or third sensors »with the corrected output of the corresponding sensor for the sampling period; (v) correct the stored output of the first sensor »with the attenuation factor; (vi) determine the gas concentration during the sampling period »from the corrected output from the first sensor; and (vii) monitor gas concentration based on a predetermined function and provide an exit signal based on surveillance.
40. 40.- A passive infrared gas detector »according to claim 1» further characterized in that the first sensor, the second sensor and the third sensor each comprise a thermopile detector.
41. 41.- A passive infrared gas detector »according to claim 1» further characterized in that each of the first sensor »of the second sensor and the third sensor comprises a thermopile and each shares a common reference joint.
42. 42.- A passive infrared gas detector according to claim 1 »further characterized in that in the field of view for the first, second and third sensors, it is substantially the same.
43. 43.- A passive infrared gas detector according to claim 1 »further characterized in that the gas being monitored is at least one selected from the group consisting of CO C02» H ^ O »and TVOC.
44. 44.- A passive infrared gas detector according to claim 1 »further characterized in that the first, second and third narrow band pass filters have an amplitude of approximately 0.1 μm to FWHM.
45. 45.- A passive infrared gas detector according to claim 6. further characterized in that the second and third narrow band pass filters have a central wavelength selected from the group consisting of about 3.91 μm »about 5.0 μm and around 9.00 μm.
46. 46.- A passive infrared gas detector according to claim 7 »further characterized in that the first narrow band pass filter has an approximate central wavelength of 4.67 μm.
47. 47.- A passive infrared gas detector according to claim 1 »further characterized in that it additionally comprises a battery power source.
48. 48.- A passive infrared gas detector according to claim 1 »further characterized in that the port comprises a window in a can T05.
49. 49. A passive infrared gas detector according to claim 4. further characterized in that it additionally comprises an optical system that expands the field of view of the detector assembly.
50. 50.- A passive infrared gas detector according to claim 1 »further characterized in that the output of the signal processing means is communicated to an alarm.
51. 51.- A passive source infrared gas detector »characterized in that it comprises: (a) an infrared detector assembly» to produce a first signal »a second signal and a third signal; being the first indicator output of the radiation received by the detector assembly to a first non-neutral spectral band, which is absorbable by a preselected gas to be detected; the second indicator output of the radiation received by the detector assembly being a first neutral spectral band from the passive infrared source, and the third signal being the radiation received by the detector assembly being a second neutral spectral band from the passive source infrared »(b) temperature measuring means to produce an output indicating the ambient temperature of the detector assembly; (Or signal processor means adapted to receive the first, second and third output and output of the temperature measuring means and to sample and store, at least temporarily, the outputs, first, second and third and the output of the outputs. temperature measuring means at pre-set intervals, including the signal processing means »means for: (i) correcting the first» second and third stored »outputs to compensate for the ambient temperature of the detector assembly; (ii) calculate the temperature of the infrared passive source »based on the proportion of the corrected values of the stored second and third outputs; (iii) calculate a second or third predicted departure for the sampling period, based on the calculated temperature of the infrared passive source; (iv) calculate an attenuation value for the sampling period »by comparing the predicted second or third output with the second or third stored» real »output respectively; (v) correct the first output stored with the attenuation factor calculated for the sampling period; (vi) determine the gas concentration during the sampling period »using the first corrected output; and (vii) monitor gas concentration based on a predetermined function "and provide an exit signal based on surveillance.
52. 52.- A passive infrared gas detector, according to claim 51 »further characterized in that the gas to be detected is at least one selected from the group consisting of CO» CO.,., Ha0 and TVOC.
53. 53.- A passive infrared gas detector according to claim 51 »further characterized in that the first non-neutral spectral band» the first neutral spectral band and the second neutral spectral band have an approximate amplitude of 0.1 μm to FWHM.
54. 54.- A passive infrared gas detector according to claim 53 »further characterized in that the first neutral spectral band and the second neutral spectral band have a central wavelength selected from the group consisting of approximately 3.91 μm. 5.Om and 9.0 μm.
55. 55.- A passive infrared gas detector according to claim 54, further characterized in that the first non-neutral spectral band has an approximate central wavelength of 4.67 μm.
56. 56.- a passive infrared gas detector according to claim 54, further characterized in that the first non-neutral spectral band has an approximate central wavelength of 4.26 μm.
57. 57.- A passive infrared gas detector according to claim 51, further characterized in that it additionally comprises a battery power source.
58. 58.- A passive infrared gas detector according to claim 39, further characterized in that the infrared detector assembly is housed in a can T05.