AU2004201100B2

AU2004201100B2 - Fire Detection Method and Fire Detector Therefor

Info

Publication number: AU2004201100B2
Application number: AU2004201100A
Authority: AU
Inventors: Tido Krippendorf; Waldemar Ollik; Heiner Politze; Ralf Sprenger
Original assignee: Novar GmbH
Current assignee: Novar GmbH
Priority date: 2004-03-16
Filing date: 2004-03-16
Publication date: 2009-11-12
Anticipated expiration: 2024-03-16
Also published as: AU2004201100A1

Description

S&F Ref: 670404 AUSTRALIA PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT Name and Novar GmbH Address Dieselstr. 2 of Applicant: 41469 Neuss Germany Actual Heiner Politze, Ralf Sprenger, Tido Krippendorf, Inventor(s): Waldemar Ollik Address for Spruson & Ferguson Service: St Martins Tower Level 35 31 Market Street Sydney NSW 2000 (CCN 3710000177) Invention Title: Fire Detection Method and Fire Detector Therefor The following statement is a full description of this invention, including the best method of performing it known to me/us: 5845c FIRE DETECTION METHOD AND FIRE DETECTOR THEREFOR BACKGROUND OF THE INVENTION Field of the Invention The invention relates to a method for recognizing fires according to the 5 scattered light principle by pulsed emission of a radiation of a first wavelength along a first radiation axis as well as a radiation of a second wavelength which is shorter than the first wavelength along a second radiation axis Into a measuring volume and by measuring the radiation scattered on the particles located in the measuring volume under a forward scattering angle of more than 90' and under a backward 10 scattering angle of less than 900, The invention further relates to a scattered-light fire detector for performing this method. Description of the Related Art A scattered-light detector is known from WO 01/59 737 which is provided especially for installation in ventilation and air-conditioning conduits, which operates 15 according to the aforementioned method and where a first light-emitting diode (LED) emits infrared light and a second LED emits blue light into its measuring chamber. The LEDs are pulsed in an alternating fashion. The radiation produced by the "infrared" LED allows recognizing large particles which are typical for a smouldering fire. The scattered radiation produced by the "blue" LED allows recognizing small 0 particles which are typical for fires with open flames. This is explained by Rayleigh's law, according to which the intensity of the scattered light decreases with the fourth power of the wavelength for particles which are smaller than the wavelength. Although the latter is correct, it does not fulfill the actual conditions in recognizing fires according to the scattered light principle. The known fire detector comprises 5 only a single photodetector which supplies only two pieces of information on the scattered light intensities, namely, depending on the embodiment, either the intensity of the forward scattered radiation in the infrared and In the blue wavelength region or the respective intensities of the backward scattered radiations or also the- intensity of the forward scattered radiation in the infrared wavelength region and the backward 0 scattered radiation in the blue wavelength region. The respective arrangement criteria lead to the consequence, however, that the measuring volumes from which the respective scattered radiation is obtained are not identical. From DE 199 02 319, a fire detection method is known in which the alarm decision is made depending on the ratio of the intensity of the IR forward scattered 5 radiation to the intensity of the IR backward scattered radiation. The respective fire detector works optionally with two infrared LEDs and a photodetector or vice-versa with one infrared LED and two photodetectors. The angle under which the forward scattered radiation is measured is preferably 140*, and the angle under which the backward scattered radiation is measured is preferably 700. The formation of the 10 ratio of the intensities of the forward and backward scattered radiation allows distinguishing bright from dark types of smoke, because bright smoke supplies a high forward scattered signal and a ,comparatively small backward scattered signal, whereas, conversely, dark smoke' supplies a lower forward scattered signal and a comparatively high backward scattered signal. The processing of the absolute 15 intensities or signal level by taking into account the principally lower intensities in the backward scattering region in relationship to the intensities produced in the forward scattering region by the same particles with the same intensity and the simultaneous processing of the ratios or quotients of these signals also allow distinguishing certain deceptive values of smoke. For example, water- vapor in high concentration .0 produces a high forward scattered signal which according to the older state of the art leads to the initiation of an alarm, in this case, to a false alarm. The formation of the quotient from the forward scattered intensity and the backward scattered intensity leads to a value which is characteristic for water vapor, which value is substantially independent of the concentration. By determining this quotient and considering it in 5 the further signal processing it is thus possible to suppress any false alarms that would occur otherwise. The known method and the detector which operates according to this method have a common feature with all other known constructions of scattered-light fire detectors which operate on the basis of infrared light, which feature is the disadvantage of an inadequate sensitivity for small and very small 0 particles. This makes it more difficult to recognize open fires in due time, and especially wood fires whose smoke is characterized by a very small particle size. In the case of a respective hazardous situation it is therefore still necessary to use ionization fire detectors which respond very well to small particles and which work with a preparation of low radioactivity. Due to this radioactive preparation, the 2 3 production of ionization fire detectors is complex and their use is unpopular and even generally prohibited in a number of countries. OBJECT OF THE INVENTION It is the object of the present invention to substantially overcome or ameliorate 5 one or more of the disadvantages of the prior art. SUMMARY OF THE INVENTION The present invention provides a method for detecting fires according to the scattered light principle, comprising: emitting pulsed radiation of a first wavelength along a first radiation axis into a measuring volume; emitting pulsed radiation of a second 10 wavelength which is shorter than the first wavelength along a second radiation axis into the measuring volume; and measuring radiation scattered on particles located in the measuring volume under a forward scattering angle of more than 900 and under a backward scattering angle of less than 900, wherein forward scattered radiations and backward scattered radiations of the first and second wavelengths are measured separately is from each other, and wherein the scattered radiations of the first and second wavelengths are measured on opposite sides of the measuring volume on a same main axis. Preferably, the above-described method further comprises the steps of subtracting from signal levels which correspond to measured intensities of the forward and backward scattered radiations of the first and second wavelengths, corresponding 20 scaled quiescent value levels to produce weighted values; evaluating the weighted values to determine whether an alarm condition exists; and producing at least one alarm signal in response to the determining that an alarm condition exists. Preferably, the step of evaluating the weighted values further includes: forming a first ratio between the weighted values of the forward scattered radiation intensity and the 25 backward scattered radiation intensity of the first wavelength; forming a second ratio between the weighted values of the forward scattered radiation intensity and the backward scattered radiation intensity of the second wavelength; and evaluating the first and second ratios to determine whether an alarm condition exists. Preferably, the step of evaluating the weighted values includes: forming a first 30 ratio between the weighted values of the forward scattered radiation intensities of the first and the second wavelengths; forming a second ratio between the weighted values of the backward scattered radiation intensities of the first and second wavelengths; and evaluating the first and second ratios to determine whether an alarm condition exists.

4 Preferably, the forward scattered radiations of the first and second wavelengths are measured under the same forward scattering angle, and the backward scattered radiations of the first and second wavelengths are measured under the same backward scattering angle. 5 Preferably, the scattered radiations of the first and second wavelengths are emitted into the measuring volume from opposite sides along coinciding radiation axes. Preferably, the first wavelength and the second wavelength are not in an integral ratio with respect to each other. Preferably, the first wavelength lies in the region of the infrared radiation and the to second wavelength lies in the region of blue light or the region of ultraviolet radiation. Preferably, the first wavelength is in the region of 880 nm and the second wavelength is in the region of 475 nm or the region of 370 nm. Preferably, a pulse/pause ratio of the radiation of the first and second wavelengths is greater than 1:10,000. is Preferably, the pulse/pause ratio of the radiation of the first and second wavelengths is approximately 1:20,000. The present invention also provides a scattered-light fire detector comprising: a measuring chamber which communicates with the ambient air and which delimits a measuring volume; a first light emitting diode (LED) that emits infrared 20 radiation into the measuring volume; a second LED that emits blue light into the measuring volume from a different direction than the first LED; first and second photodetectors situated opposite of each other on a common main axis with respect to each other and which measure the radiation scattered by particles situated in the measuring volume, wherein radiation axes of the first and second LEDs enclose an acute 25 angle of less than 900 with the main axis and intersect in a point which is situated on the main axis and is situated in the center of the measuring volume. Preferably, the first and second LEDs are arranged on the same side of the main axis. Preferably, the first and second LEDs are arranged symmetrically to the main 30 axis. Preferably, the first and second LEDs are arranged in a point-symmetrical fashion to the center of the measuring volume, so that radiation axes of the first and second LEDs coincide. Preferably, radiation axes of the first and second LEDs each enclose with the 35 main axis an acute angle of approximately 60*.

5 Preferably, the detector further comprises tube bodies housing each of the first and second LEDs and each of the first and second photodetectors; and diaphragms and radiation traps arranged in the measuring chamber outside of the measuring volume between the first and second LEDs and the first and second photodetectors. 5 Preferably, the first photodetector receives the forward scattered radiation of the first LED and the backward scattered radiation of the second LED and the second photodetector receives the backward scattered radiation of the first LED and the forward scattered radiation of the second LED.

6 BRIEF DESCRIPTION OF THE DRAWINGS A preferred embodiment of the present invention will now be described, by way of an example only, with reference to the accompanying drawings wherein: Fig. I shows a top view intersected at the height of the optical axes of the base 5 plate of the fire detector in a first embodiment, which base plate carriers the measuring chamber; Fig. 2 shows the respective view of a second embodiment, and Fig. 3 shows the respective view of a third embodiment. DESCRIPTION OF THE PREFERRED EMBODIMENTS 10 The method in accordance with the invention assumes the following: depending on the type of the burning material, a wide range of incineration products are obtained which are designed below as aerosols or also as particles for the sake of simplicity. Hot fires produce large quantities of aerosols of small diameter. For example, an aerosol structure or cluster comprising 100 molecules of CO 2 has a diameter of approximately 2.5 nm. Fires with a so-called low energy conversion per unit of time, i.e., so-called smoldering fires, produce aerosols with a diameter of up to 100 pm and partly also macroscopic suspended matter, e.g., ash particles. A scattered-light fire detector which is suitable for recognizing all kinds of fires would 5 therefore have to recognize aerosols with a diameter of 2.5 nm to 100 pm, i.e., it would have to cover a range of five powers of ten. As a result of their high efficiency, infrared-radiating GaAs LEDs have been used exclusively in practice as radiation sources in scattered-light fire detectors, which LEDs radiate at a wavelength X of 880 nm. The intensity of the scattered 10 radiation caused by a particle prirharily depends on the ratio of the diameter of the particle (which is assumed to be a sphere for the sake of simplicity) to the wavelength of the incident radiation. Although the shape and the absorption coefficient of the particle play an additional role, these parameters can obviously not be influenced in the present context. The so-called Rayleigh scattering decreases 15 proportionally to X4 for a particle diameter below 0.1 X. It follows from this that fire detectors working with infrared-radiating LEDs have a steeply dropping sensitivity for particle diameters of less than approximately 90 nm. An additional factor is that the Rayleigh scattering is not omnidirectional but has characteristic maximums at 0' and 1800 and characteristic minimums at 90* and 270*. For particles with diameters of 20 0.1 X to 3 X, which in the case of an. infrared-radiating LED is from approximately 90 nm to approximately 2.5 pm, the Mie effect is relevant, which is even stronger directionally dependent than the Rayleigh scattering and moreover shows destructive and constructive interference effects by interaction of the introduced radiation with the radiation reflected on the particle. Above 3 -- the scattering intensity is 5 substantially independent of the wavelength and depends primarily on the type and the shape of the particle. It follows from this that the low sensitivity of scattered-light fire detectors for hot fires, e.g., open wood fires, is caused by the high wavelength of the infrared radiation in relationship to the diameter of the particles to be detected. This can be i 0 counteracted neither by increasing the amplification of the signal supplied by the photodetectors, nor by increasing the intensity of the introduced radiation, because in both cases the sensitivity of the detector for large and macroscopic particles (e.g., dust, vapors from industrial processes and cigarette smoke) will become too high. 7 By alternately irradiating the measuring volume with infrared radiation and blue light and by separately processing the signals proportional to the received scattered radiation, it is possible, as is principally known from the aforementioned WO 01/59 737, to considerably increase the sensitivity of the detector for particles of 5 small diameter, especially such for which the Rayleigh radiation is relevant. It can be easily shown mathematically that the sensitivity increases by a factor of 10 or more. The increase in the sensitivity of the detector for particles of a small diameter is alone not sufficient for obtaining a secure alarm decision, i.e., for avoiding faIse or deceptive alarms. It is not the case, contrary to the assumption made in WO 01/59 10 737, that the irradiation of the measuring volume with blue light for large and small particles supplies .scattered radiation of approximately the same intensity. Examinations on this part have shown to the contrary that especially small particles supply scattered radiation of very similar intensity in the infrared region and under blue light, both in the forward and', at a lower level, the backward radiation region. 15 As was further observed, it is gnly the addition of the angular dependence of the intensity of the scattered radiation which allows obtaining secure criteria which allow differentiating between deceptive values and consequential products of fires in a manner substantially independent of the kind of the material that is burned. In accordance with the invention, four scattered radiation Intensities are 20 therefore measured in each measuring cycle, namely the forward scattered radiation and the backward scattered radiation in the infrared region and the same values in the blue light region. The corresponding quiescent value level, preferably with a reduction for security purposes (according to a multiplication .of the quiescent value levels with a factor <1, i.e., a scaled quiescent value level), is subtracted from the 5 signal levels which are proportional to the measured intensities, which subtraction is made for increasing the measuring dynamics and in order to simplify the further processing. The thus obtained resUlting values are then compared in an evaluation logic circuit with stored values, especially threshold values. Additional information is obtained by the formation of the quotients of the resulting values and renewed 0 comparison with the stored reference values. The results of these operations can be combined and evaluated on their part, e.g., adjusted to the respective environment in which the detector is used. In this way a number of meaningful intermediate results can be obtained, e.g., for different preliminary alarms and finally also alarm signals. 8 Fig. 1 shows a first preferred embodiment of a detector suitable for performing this method. A spherical measuring volume with a center 1.5 is defined on a base plate 1.7, which measuring volume is schematically indicated with a thin circle. An infrared-radiating LED 1.1a emits radiation along a first radiation axis into said 5 measuring volume. Precisely opposite of the same, there is a blue-radiating LED 1.1b which emits radiation into the measuring volume along a second radiation axis. The first and the second radiation axis coincide. A main axis under an angle of a 1200 to this common radiation axis also extends through the center 1.5 of the measuring volume. A first photodiode 1.2a and 1.2b are arranged opposite of one 10 another on said main axis. As "a result, the main axis on which the respective receiving axes of the two photodiodes are situated encloses with the first radiation axis of the "infrared" LED 1.1a an acute angle p = 60*. The same acute angle Is accordingly enclosed by the main axis with the (second) radiation axis of the "blue" LED 1.1b. As a result, the photodiode 1.2a measures under an angle of 120* the 5 infrared forward scattered radiation as produced by the "infrared" LED 1.1a on particles in the measuring volume and the blue scattered radiation as produced by the "blue" LED 1.1b is measured under a backward scattered radiation of 60*. Conversely, the photodiode 1.2b measures the blue forward scattered radiation which is produced by the "blue" LED 1.1b under an angle c of 120* and the infrared 0 backward scattered radiation which is produced by the "infrared" LED 1.1a under a - backward scattering angle of 60*. In order to avoid any stray reflections, the LEDs and the photodiodes are situated in tube bodies such as 1.6. For the same reason suitably shaped diaphragms such as 1.3a, 1.3b as well as 1.4a and 1.4b are arranged between the LEDs and the photodiodes. Further sensors such as a temperature sensor at 1.8 and a gas sensor at 1.9 are arranged on the base plate 1.7. As is conventional, a circuit board for producing the current pulses for the LEDs 1.1a and 1.1b as well as for processing the electric signals supplied by the photodiodes 1.2a and 1.2b is situated beneath the base plate 1.7. As is also conventional, the base plate 1.7 is housed in a detector housing (not shown) which allows an exchange between the ambient air and the air in the measuring chamber, but at the same time keeps outside light away from the measuring chamber. 9 Fig. 2 shows a second embodiment of the detector with the same components as in Fig. 1, but with a different geometrical arrangement. In order to explain this arrangement in closer detail, the first digit of the respective reference numeral is provided here with "2" instead of "". In contrast to Fig. 1, only the radiation axes of 5 the infrared-radiating LED 2.1a and the blue-radiating LED 2.1b which go through the measunng center 2.5 will coincide. The receiving axis of the photodiode 2.2a encloses an angle al = 120* with the radiation axis of LED 2.1a and with the radiation axis of the blue-radiating LED 2.1b an angle 02 = 60*. The receiving axis of the photodiode 2.2b encloses conversely with the radiation axis of the infrared 10 radiating LED 2.1a an angle al = 60* and with the radiation axis of the blue-radiating LED 2.1b an angle a2 = 120". Accordingly, the first photodiode 2.2a measures the forward scattered radiation of the "infrared" LED 2.1a and the backward scattered radiation of the "blue" LED 2.1b. The second photodlode 2.2b conversely measures the forward scattered radiation which is produced by the "blue" LED 2.1b and the 15 backward scattered radiation which is produced by the "infrared" LED 2.1a. The photodiodes 2.2a and 2.2b can exchange their positions with the LEDs 2.1a and 2.1b, so that the two photodiodes are situated precisely opposite with respect to the measuring center 2.5. This geometrical arrangement of the four components, i.e., that of the two LEDs and the two photodiodes, is less favorable 20 than that of Fig. 1 because only 75% of the four measured scattered radiations originate from the same measuring volume. This is illustrated by the intersecting surfaces between the beams which are shown by omitting the angular dependency both of the intensity of the emitted radiations as well as .the sensitivity of the photodiodes as well as the diffraction effects which occur unavoidably on the edges. .5 In the case of detectors which (as in the embodiment) comprise further sensors such as 2.8 and 2.9, there is an additional factor that the measuring center 2.5 is disposed In a strongly eccentric fashion with respect to the center of the base plate 2.7. This leads to the consequence that the sensitivity of the detector is not omni-directional as in the case of the first embodiment, but that it is dependent upon the direction from 0 which the consequential products from the fire enter the detector and its measuring volume. Fig. 3 shows a third embodiment of the detector with the same components as In Fig. 2, but with a different geometrical arrangement. In order to illustrate this in 10 closer detail, the first digit of the respective reference numeral is provided here with "3" instead of "2". In contrast to Fig. 1, only the receiving axes of the photodiodes 3.2a and 3.2b coincide which pass through the measuring center 3.5. These receiving axes form the main axis. The "infrared" LED 3.1a encloses with the latter 5 an acute angle al = 60* and an obtuse angle 31 = 120*. The "blue" LED 3.1b is situated opposite of the "infrared" LED 3.1a with respect to the main axis, which "blue" LED accordingly encloses with the main axis an acute angle P2 = 600 and an obtuse angle a2 = 120*. As a result, the photodiode 3.2a receives both the infrared forward scattered radiation as well as the blue forward scattered radiation, whereas 10 the photodiode 3.2b receives both the infrared backward scattered radiation as well as the blue backward scattered radiation. Other than is the case in Fig. 2, the two LEDs and the two photodiodes cannot be provided in this arrangement with an exchanged position, because in this case the two photodiodes would simultaneously measure the forward scattered radiation 15 of the one LED and then the backward scattered radiation of the other LED, i.e., supply four measured values of which two would be approximately the same in pairs. As in the case of Fig. 2, only 75% of the four measured scattered radiations each originate from the same measuring volume in the embodiment according to Fig. 3 as well. It is more advantageous than in the case of Fig. 2 in that the measuring 20 volume, even in the case that the detector comprises further sensors such as 3.8 and 3.9, is situated closer to the center of the base plate 3.7, so that the sensitivity of the detector depends less strongly on the direction from which the consequential - products from the fire enter the detector. An additional advantageous aspect in comparison with Fig. 2 is in the geometry according to Fig. 3 that all diaphragms 3.3a, 5 3.3b and 3.4a, 3.4b are arranged close to the measuring volume and are situated in a substantially symmetrical fashion around the same. Under the conditions that are the same otherwise, the positioning of the "blue" LED 3.1b causes a larger diameter of the base plate 3.7 as compared to Fig. 1. Although it applies to all embodiments that the scattered radiations are 0 measured under angles of 120" or 60*, the adherence to these angles is not a necessary precondition for performing the method proposed for implementing the invention. The important aspect is merely that the angles are chosen in such a way that in the forward scattered radiation direction and in the backward scattered 11 radiation direction sufficiently high intensities can be measured on the one hand and sufficiently different intensities can be measured in the forward scattering region and in the backward scattering region of the respective particles for the largest possible number of different consequential fire products. 5 12

Claims

1. A method for detecting fires according to the scattered light principle, comprising: emitting pulsed radiation of a first wavelength along a first radiation axis into a 5 measuring volume; emitting pulsed radiation of a second wavelength which is shorter than the first wavelength along a second radiation axis into the measuring volume; and measuring radiation scattered on particles located in the measuring volume under a forward scattering angle of more than 900 and under a backward scattering angle of less to than 90*, wherein forward scattered radiations and backward scattered radiations of the first and second wavelengths are measured separately from each other, and wherein the scattered radiations of the first and second wavelengths are measured on opposite sides of the measuring volume on a same main axis.

2. A method as claimed in claim 1, further comprising: 15 subtracting from signal levels which correspond to measured intensities of the forward and backward scattered radiations of the first and second wavelengths, corresponding scaled quiescent value levels to produce weighted values; evaluating the weighted values to determine whether an alarm condition exists; and 20 producing at least one alarm signal in response to the determining that an alarm condition exists.

3. A method as claimed in claim 2, wherein the step of evaluating the weighted values further includes: forming a first ratio between the weighted values of the forward scattered 25 radiation intensity and the backward scattered radiation intensity of the first wavelength; forming a second ratio between the weighted values of the forward scattered radiation intensity and the backward scattered radiation intensity of the second wavelength; and evaluating the first and second ratios to determine whether an alarm condition 30 exists.

4. A method as claimed in claim 2, wherein the step of evaluating the weighted values includes: forming a first ratio between the weighted values of the forward scattered radiation intensities of the first and the second wavelengths; 14 forming a second ratio between the weighted values of the backward scattered radiation intensities of the first and second wavelengths; and evaluating the first and second ratios to determine whether an alarm condition exists.

5 5. A method as claimed in claim 1, wherein the forward scattered radiations of the first and second wavelengths are measured under the same forward scattering angle, and the backward scattered radiations of the first and second wavelengths are measured under the same backward scattering angle.

6. A method as claimed in claim 1, wherein the scattered radiations of the io first and second wavelengths are emitted into the measuring volume from opposite sides along coinciding radiation axes.

7. A method as claimed in claim 1, wherein the first wavelength and the second wavelength are not in an integral ratio with respect to each other.

8. A method as claimed in claim 1, wherein the first wavelength lies in the is region of the infrared radiation and the second wavelength lies in the region of blue light or the region of ultraviolet radiation.

9. A method as claimed claim 1, wherein the first wavelength is in the region of 880 nm and the second wavelength is in the region of 475 nm or the region of 370 nm. 20

10. A method as claimed in claim 1, wherein a pulse/pause ratio of the radiation of the first and second wavelengths is greater than 1:10,000.

11. A method as claimed in claim 1, wherein the pulse/pause ratio of the radiation of the first and second wavelengths is approximately 1:20,000.

12. A scattered-light fire detector comprising: 25 a measuring chamber which communicates with the ambient air and which delimits a measuring volume; a first light emitting diode (LED) that emits infrared radiation into the measuring volume; a second LED that emits blue light into the measuring volume from a different 30 direction than the first LED; first and second photodetectors situated opposite of each other on a common main axis with respect to each other and which measure the radiation scattered by particles situated in the measuring volume, wherein radiation axes of the first and second LEDs enclose an acute angle of less than 90* with the main axis and intersect in a point 35 which is situated on the main axis and is situated in the center of the measuring volume. 15

13. A detector as claimed in claim 12, wherein the first and second LEDs are arranged on the same side of the main axis.

14. A detector as claimed in claim 12, wherein the first and second LEDs are arranged symmetrically to the main axis. 5

15. A detector as claimed in claim 12, wherein the first and second LEDs are arranged in a point-symmetrical fashion to the center of the measuring volume, so that radiation axes of the first and second LEDs coincide.

16. A detector as claimed in claim 12, wherein radiation axes of the first and second LEDs each enclose with the main axis an acute angle of approximately 600. 1o

17. A detector as claimed in claim 12, further comprising: tube bodies housing each of the first and second LEDs and each of the first and second photodetectors; and diaphragms and radiation traps arranged in the measuring chamber outside of the measuring volume between the first and second LEDs and the first and second 15 photodetectors.

18. A detector as claimed in claim 12, wherein the first photodetector receives the forward scattered radiation of the first LED and the backward scattered radiation of the second LED and the second photodetector receives the backward scattered radiation of the first LED and the forward scattered radiation of the second 20 LED.

19. A method for detecting fires according to the scattered light principle substantially as hereinbefore described with reference to the accompanying drawings.

20. A scattered-light fire detector substantially as hereinbefore described 25 with reference to the accompanying drawings. Dated 2 October, 2009 Novar GmbH Patent Attorneys for the Applicant/Nominated Person SPRUSON & FERGUSON