GB2411953A - High sensitivity gas detector - Google Patents

Abstract

A gas detector includes two laser diodes 1,14 driven by circuits 3,15 operating each laser at a wavelength close to a different optical absorption line of the same target gas; whilst scanning the lasers across their respective absorption lines at non-harmonically related electrical frequencies and wavelength intervals. Radiation from the laser diodes is collected and collimated by optical elements with different, non-harmonically related effective focal lengths and thicknesses. This radiation is transmitted through a monitored space and illuminates an optical detector 7. The electrical signal from the detector is processed to calculate separate estimates of the quantity of target gas present in the monitored space, these subsequently being compared and used in combination with previous results to calculate the reading that will be output by the gas detector. In an alternative arrangement a single laser diode may be driven by two bias voltages and two non harmonically related signals in order operate the diode at two central wavelengths and scan around each of the central wavelengths.

Description

2411 953 1 1

HIGH SENSITIVITY GAS DETECTOR

This invention is an apparatus for the detection or measurement of gases with high sensitivity and reliability.

Laser Diode Spectroscopy (LDS) is a well-known technique used for the sensitive detection of a wide range of simple gas molecules. The LDS technique relies upon the fact that the output wavelength from a laser diode varies with the applied drive current, this variation in wavelength being used to scan the laser's wavelength over a selected optical absorption line of a target gas to be detected or measured. When optical radiation from such a laser diode is transmitted through a gas sample and is focussed onto a suitable optical detector, the signal received by the detector corresponds to the product of the laser's output waveform and the transmission spectrum of the gas sample being illuminated.

The quantity of target gas in the sample path can be determined by measuring the change in received intensity when the laser's wavelength corresponds to the wavelength of the optical absorption line of the target gas. The amount of absorption produced by a given quantity of gas can be determined using Beer's law: I = IO.e where I is the received intensity at the detector, lo is the intensity incident upon the sample being illuminated, cr is the absorption cross section of the target gas at the absorption wavelength and n is the total number of target gas molecules in the measurement path.

Laser diode spectroscopy techniques have been widely used in instrumentation to measure gases in industrial processes and to monitor atmospheric pollutants.

However, equipment using LDS techniques has only enjoyed limited success in the detection of toxic or flammable gases for safety related applications. The main reason that LDS based equipment has not been widely employed for the detection of toxic or flammable gases for safety related applications is that such applications demand an extremely high level of reliability and in particular very low false alarm rates. Whilst existing LDS based equipment might be able to detect or measure fractional absorbances of the order of 1 x10-to 1 X10-5 with reliability acceptable for process control or atmospheric monitoring applications, such small fractional absorbances cannot be detected with an acceptably low false alarm rate for safety applications.

As a consequence of an unacceptably high probability of false alarms, LDS based gas detectors are not currently employed for the detection of toxic and flammable gases for which such detectors would otherwise be suitable. This invention introduces a number of enhancements to the LDS technique with the object of producing an LDS based gas detector suitable for use in a wide range of safety applications. i) 2

The invention is described by way of examples and with reference to the accompanying drawings in which: Figure 1 shows a simple LDS based gas detection or measurement system.

Figure 2 shows the typical variation in laser diode output power with applied drive current for a laser diode used in a simple LDS based gas detection or measurement system.

Figure 3 shows the typical variation in output wavelength with applied drive current for a laser diode used in a simple LDS based gas detection or measurement system.

Figure 4 shows the ideal transmission spectra for a single target gas absorption line to be scanned by a simple LDS based gas detection or measurement system.

Figure 5 shows the drive current applied to a laser diode in a simple LDS based gas detection or measurement system.

Figure 6 shows the Fourier transform of the detector signal for a simple LDS based gas detection or measurement system when there is no target gas in the monitored space.

Figure 7 shows the signal from the optical detector when there is a substantial quantity of target gas present in the monitored space of a simple LDS based gas detection or measurement system.

Figure 8 shows the Fourier transform of the detector signal when there is a substantial quantity of target gas present in the monitored space of a simple LDS based gas detection or measurement system.

Figure 9 shows a laser diode drive current waveform comprising two alternating bias and frequency components.

Figure 10 shows the combined Fourier transform of the detector signal when there is no target gas present in the monitored space of a system with a laser driven as shown in Figure 9.

Figure 11 shows the combined Fourier transform of the detector signal when there is a substantial quantity of target gas present in the monitored space of a system with a laser driven as shown in Figure 9.

Figure 12 shows the absorption spectrum for 25ppm.m of hydrogen sulphide between 1 585nm and 1 595nm.

Figure 13 shows the absorption spectrum for a 100 metre path through the Earth's atmosphere at 30 C, 1 00%RH between 1 585nm and 1 595nm.

Figure 14 shows the absorption spectrum for 25ppm.m of hydrogen sulphide between 1 589nm and 1590.1 nm.

Figure 15 shows the absorption spectrum for a 100 metre path through the Earth's atmosphere at 30 C, 100%RH between 1589nm and 1590.1 nm.

Figure 16 shows the shape of a gas absorption line compared to a half and full cycle of a sinusoid, typical of that produced by coherence / fringe effects.

Figure 17 shows a dual laser, dual frequency, dual wavelength laser diode gas detection or measurement system in accordance with the claimed invention.

Figure 18 shows a dual laser, dual frequency, dual wavelength laser diode gas detection or measurement system with dual transmitter optics in accordance with the claimed invention.

Figure 19 shows a dual laser, dual frequency, dual wavelength laser diode gas detection or measurement system with dual transmitter optics and dual receiver optics in accordance with the claimed invention.

Figure 20 shows a dual laser, dual frequency, dual wavelength laser diode gas detection or measurement system with dual receiver optics in accordance with the claimed invention.

Figure 21 shows a triple laser, triple frequency, triple wavelength laser diode gas measurement system in accordance with the claimed invention.

Figure 22 shows a dual laser, dual frequency, dual wavelength laser diode gas detection or measurement system with separate optical paths for each laser diode in accordance with the claimed invention.

A simple, LDS based gas detection or measurement system is shown in Figure 1 and comprises a laser diode 1, mounted on a temperature stabilised mount 2, driven by a laser diode drive / modulation circuit 3, the output from the laser diode being collected and collimated by optical element 4, the resulting beam being transmitted through a monitored space 5, to a receiver optical element 6, which focuses the received radiation onto a detector 7, the signal from the detector being amplified by amplifier 8 and digitised by ADC 9, then processed by signal processing system 10 to calculate the quantity of target gas in the sample path.

The operation of a simple LDS system such as that shown in Figure 1 is illustrated by Figures 2 to 8. Figure 2 shows the variation in output power from the laser diode as the drive current is increased, this being essentially linear when operating above the laser diode's threshold current. Figure 3 shows how the output wavelength of the laser diode varies with drive current, this effect being used to scan the laser's wavelength across the target gas' absorption line of interest. Figure 4 shows the ideal wavelength dependent transmission in and

-

around the region of a chosen target gas absorption line resulting from introduction of a quantity of target gas into the monitored space. (It should be appreciated that the absorption spectra of gases contain many absorption lines and that Figure 4 shows a small, ideal region around a particular absorption line chosen for making measurements.) Figure 5 shows the current signal applied to the laser diode. This signal contains a DC component which effectively maintains the laser diode above threshold and operating at the correct mean wavelength and a sinusoidal component at frequency f, which scans the laser diode's wavelength back and forth over the region containing the absorption line. In the absence of target gas in the monitored space, the signal from the system's detector will reproduce the waveform shown in Figure 5. Figure 6 shows a Fourier transform of the detector signal in the absence of target gas in the monitored space, there just being a single frequency component at the modulation frequency f. Figure 7 shows the output from the system's detector during a single wavelength scan when a substantial quantity of target gas is present in the monitored space. Note the deviation from a simple sinusoidal waveform produced by absorption of optical radiation when the wavelength of the laser diode scans through the region of the target gas' absorption line. The absorption by the target gas introduces an absorption feature whenever the wavelength of the laser crosses the region of the absorption line, which in Figure 7 occurs approximately half way up the positive excursion of the modulation cycle. Figure 8 shows a Fourier transform of this detector signal, which due to the presence of absorption features now contains harmonics of the modulation frequency f, the size of these harmonics depending upon the amount of target gas in the monitored space and the precise shape of the absorption line in relation to the sinusoidal modulation waveform. Various methods can be used to measure the relative size of these harmonic components and to subsequently detemmine the amount of target gas in the measurement path, the most popular of which is synchronous detection and measurement of the second and / or third harmonics. Alternatively, the signal can be digitised as is shown in Figure 1 and then processed using digital signal processing techniques to measure the magnitude of the various frequency components in the detector signal and determine the amount of gas present in the monitored space.

Laser diode spectroscopy techniques similar to those illustrated in Figures 1 to 8 have been widely used in instrumentation to measure gases in industrial processes and to monitor atmospheric pollutants, where fractional absorbances of the order of 1X104to 1x10-5 are detected or measured with reliability acceptable for such applications. However, equipment using LDS techniques has only enjoyed limited success in the detection of toxic or flammable gases for safety related applications. The only safety related application where any quantity of LDS based equipment has been employed has been for the detection of hydrogen fluoride at aluminium smelters or hydrocarbon alkylation plants. This has been possible because hydrogen fluoride has a very large absorption cross section in the 1310nm region, making it possible to detect ppm concentrations of HE by detecting fractional absorbances of the order of 1 x 1 ok to 1 x 10-3, which is relatively straightforward. !

The main reason that LDS based equipment has not been widely employed for the detection of toxic or flammable gases for safety related applications is that such applications demand an extremely high level of reliability and in particular very low false alarm rates, which cannot be achieved by existing LDS based equipment. With the consequences of false alarms from gas detectors including the shutting down of large industrial or petrochemical plants, personnel donning safety equipment and commencing evacuation procedures; and a loss of confidence in a gas detection system, users of fixed gas detection equipment are looking for false alarm rates for each gas detector of lower than 1 per 100 years.

Whilst existing LDS based equipment might be able to detect or measure fractional absorbances of the order of 1x1o4to 1x10-5 with reliability acceptable for process control or atmospheric monitoring applications, such small fractional absorbances cannot be detected with an acceptably low false alarm rate for safety applications.

Hazardous gases which could usefully be detected by LDS based detectors if the false alarm problem could be overcome include hydrogen sulphide, ammonia, hydrogen chloride, hydrogen cyanide, methane and vinyl-chloride monomer.

An object of this invention is to enable LDS based equipment to detect fractional absorbances as low as 1 x1 o4tO 1 X10-5 with a false alarm rate that is acceptably

low for use in safety related applications.

Another object of this invention is to enable LDS equipment to reliably detect fractional absorbances as low as 1 X1o4to 1 X10-5 whilst operating over open measurement paths of 5 to100 metres length in environments typical of those found at petrochemical installations, requiring the equipment to endure extreme weather and temperatures, objects moving through the monitored space, contamination building up on exposed optical surfaces and high levels of electromagnetic interference.

When attempting to reliably detect fractional absorbances of 1 X104 to 1 X10-5 using an LDS system similar to that shown in Figure 1 with a false alarm rate that is low enough for use in safety related applications, the designer encounters three main problems, these being system noise, absorption(s) by atmospheric gases and coherence / fringe effects.

System noise is introduced by virtually all of the active components used in the LDS system of Figure 1, including the laser diode drive circuit 3, laser diode 1, detector 7, amplifier 8 and ADO 9. These differing noise sources exhibit complex frequency and probability distributions, making it practically impossible to determine their influence upon false alarm rates in a regime where effects with probabilities as low as once in a thousand years are potentially significant. All that can be stated with any confidence is that for an LDS based system to experience a system noise induced signal deviation of less than 1 X10-5 for a period long enough to cause a false alarm just once in one hundred years of operation requires an exceptionally high system signal to noise ratio (>1X106: 1).

In practice, even with careful design and selection of components, subsystems and signal processing routines, achieving such a high system signal to noise ratio is not possible. Furthermore, even if it were possible to achieve such a high system signal to noise ratio in ideal conditions, the signal losses associated with the operation of an LDS system outdoors over a useful path-length preclude achieving such a signal to noise ratio in operational service. Therefore, any LDS based system looking to detect fractional absorbances of 1X10-5 with an acceptably low probability of false alarms for use in safety related applications must address the problem of the signal to noise ratio requirement associated with LDS systems similar to that of Figure 1.

When making optical measurements along an open path through the atmosphere it is essential to consider the effects of absorption by the gases which constitute the atmosphere. In particular, atmospheric gases such as oxygen, carbon dioxide and water vapour exhibit strong optical absorption at wavelengths from the near infrared to the far infrared, which is the wavelength region of main interest for LDS systems. When assessing wavelengths at which to make measurements of a particular target gas it is necessary to ensure that there are no strong atmospheric absorption lines at wavelengths very close to that of the target gas' absorption line; and also that any continuum absorption by the atmospheric gases will not attenuate radiation at the candidate wavelength to such an extent that the system's signal to noise ratio will be unduly compromised. Additionally, when looking to detect fractional absorbances as low as 1x10-5, it is necessary to consider the effects that might be introduced when attempting to make measurements in the far wings of strong atmospheric absorption lines, this because even if the line is relatively distant and the atmospheric transmission is acceptably high, the curvature of the transmission in the far wings of a strong absorption line can look similar to the curvature produced by a small absorption produced by the target gas.

For atmospheric gases such as oxygen and nitrogen which have relatively stable atmospheric concentrations, it is possible to compensate for any small reading offsets that their adsorptions might introduce, either by zeroing the instrument or detector when it is installed, or by applying a correction calculated for the length of the monitored space, this calculation based upon the characterized effects of the atmosphere upon the instrument or detector. However, for atmospheric gases such as water vapour, carbon dioxide and carbon monoxide which exhibit significant variation in concentration depending upon weather, geography and any local emissions of these gases, such compensation is not possible.

Therefore, when designing an LDS system to make high sensitivity measurements along an open atmospheric path it is necessary to pay particular attention to the effects of absorption by water vapour, carbon dioxide and carbon monoxide; and any technique which can reduce the potential for such adsorptions to interfere with equipment using LDS would be highly beneficial.

The diode lasers used in LDS systems exhibit a high degree of spatial and temporal coherence, which means that light reflected or scattered from virtually anywhere within the LDS system or monitored space can interact in a coherent manner with the light proceeding directly along the intended measurement path.

The consequence of such coherent interactions is unwanted amplitude modulation of the light proceeding along the measurement path, such modulations being particularly undesirable if they produce features similar to those produced when the laser's wavelength is scanned across the target gas' absorption line. This problem is exacerbated by the fact that the amplitude of any such modulation is dependent upon the field strength of a particular reflection or scattering source, not upon the intensity of such a source. This means that amplitude modulations of 1 x 10'5 can be produced by reflected or scattered light of 1 x 10- intensity relative to that of the beam with which they are interacting. In effect, reflected or scattered light is capable of modulating the intensity of the light proceeding along the intended measurement path by far more than its own intensity.

With relative intensities of 1x1o-'o capable of producing amplitude modulations of 1 X10-5, coherence / fringe effects are a very significant problem in LDS based systems. Indeed, much work upon the enhancement of systems using LDS has revolved around developing techniques to reduce the magnitude or overall impact of coherence / fringe effects upon such systems. This work has included the development of a number of patented techniques to combat coherence / fringe effects, such as those described in US 4,684,258 and US 4,394,816.

Despite the success of techniques developed to reduce the significance of coherence / fringe effects upon LDS systems, in many instances, coherence / fringe effects still set the limit of detection or measurement for such systems.

Also, whilst these techniques work successfully in relatively benign, controlled environments, they work less well in outdoor or uncontrolled environments and are not sufficient to deal with the challenges presented by extreme environments.

Consequently, if an LDS system is to be used to detect fractional absorbances of 1 X 10-5 in petrochemical safety applications its design must address the coherence / fringe problem in a manner that works with the highest reliability even when exposed to extreme environmental conditions.

This invention addresses the problems of system noise, absorption(s) by atmospheric gases and coherence / fringe effects by a combination of approaches which significantly enhance the ability of a system using the LDS technique to detect fractional absorbances of 1 X104to 1 X10-5 with a low false alarm rate.

The first approach which forms part of the claimed invention addresses the problems associated with system noise and the signal to noise ratio requirements for low false alarm rates.

For an LDS system as described in Figures 1 to 8, modulating the laser at a frequency f and measuring at harmonic frequency f,, the probability of system noise N(f') producing a false alarm size deviation A(f') in a given measurement interval T. can be described by an equation of the form below, where k, is a constant for the system relating noise to probability of deviation P((f'(T))): P((f,(T))) = k1.N0,).T-1'2 In the first approach which forms part of the claimed invention, the laser diode is driven by a current as shown in Figure 9, comprising two components, a bias component and a sinusoidal component, the bias component alternating between two current levels A and B. chosen to operate the laser diode at two mean wavelengths A, and A2, close to two separate optical absorption lines of the same target gas, the sinusoidal component synchronously alternating between two, non-harmonically related electrical frequencies f and f' at which the laser's wavelength is scanned across one or the other of the chosen absorption lines for an interval T/2. When there is no gas present in the measurement path, the combined Fourier transform of the detector signal for a total interval T (where T >> 1/f,) will look like Figure 10, with just two frequency components f and f'.

When there is a substantial quantity of target gas in the monitored space, the combined Fourier transform of the detector signal will look like Figure 11, with sets of harmonics of both f and f'. When measured at frequencies f, and f2 for intervals of T/2, the probability of system noise producing a false alarm size deviation ((of)) during each separate T/2 interval is as follows: P(a(f,(T/2))) = k,.N(f,).T-1/2.42 P((f2(T/2))) = k2.N(f2).T-1'2.42 If measurement frequencies f, and f2 are chosen to be the same order harmonics of f and f and the system noise at f' and f2 is the same, the probability that during a combined measurement interval (T/2 + Tl2) the quantities of target gas Q' and Q2 calculated to be present in the measurement path will exceed a false alarm size deviation ((of)) due to system noise is: P((f,(T/2))) & P((f2(T/2))) = k,.N(f,).T- ''2.,l2 x k2. N(f2).T-'/2./2 Which for N(f,) = N(f2) simplifies to: P(0, (T/2))) & P((f2(T/2))) = 2.k,.k2.Nff,)2.T-' Since in most instances P((f, (T/2))) and P((f2(T/2))) are small, the probability of both measurements suffering noise induced deviations sufficient to exceed an alarm threshold during interval T is very small. By way of example, a system with a noise floor sufficient to achieve an average false alarm rate of 1 in 2 days when modulating at a single frequency around a single absorption line could be improved to achieve an average false alarm rate as low as 1 in 100 years by measuring at an additional frequency and wavelength where the average false alarm rate was 1 in 10 hours and using the second measurement to confirm the first.

In order to achieve the best results, the use of quantities Q' and Q2 for calculation of the amount of gas in the monitored space should be in combination with results for previous measurements, using rules dependent upon the quality of agreement between the quantities. These rules and their intended effects are described below: 1. If Q. and Q2 are in close agreement, a large fraction of the average of Q. and Q2 is added to a balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space.

This enables the output from the apparatus to quickly track changes in the quantity of target gas present in the monitored space when confidence in the most recent measurements is high.

2. If Q. and Q2 are in reasonable but not close agreement, a lesser fraction of the average of Q' and Q2 is added to a larger balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space. This enables the output to take account of the most recent measurements whilst reducing the impact that potential errors in these measurements might have upon the output of the apparatus.

3. If only Q. or only Q2 is in close or reasonable agreement with the running average of previous measurements, the quantity which is not in close or reasonable agreement is rejected whilst a small fraction of the close or reasonably agreeing quantity is added to a larger balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space. This enables measurements in which confidence is low to be prevented from affecting the output from the apparatus; whilst allowing measurements in which there is confidence to contribute to keeping the output from the apparatus up to date.

4. If Q' and Q2 are in poor agreement with each other and the running average of previous measurements, both Q' and Q2 are rejected and the calculation of the quantity of gas present in the monitored space is based solely upon the running average of previous results, the balancing fraction being set to unity. This enables false alarms due to measurements in which confidence is low to be effectively eliminated whilst maintaining the output from the apparatus at the most recent level in which confidence is acceptably high.

Modulation and measurement at a number of non-harmonically related frequencies does not only confer benefits in temms of lessening the impact of classic, thermal noise. Electronic systems are often required to operate in environments affected by electromagnetic interference. Unlike classic, thermal noise, electromagnetic interference tends to be at frequencies which are hamnonically related to the frequencies of operation of the equipment which are the source of the interference. Therefore, the use of modulation at a number of non-harmonically related frequencies reduces the likelihood that electromagnetic interference will affect all measurement frequencies simultaneously, enabling false alarms due to electromagnetic interference to be reduced. ) 10

The second approach which forms part of the claimed invention addresses the problems associated with absorption(s) by atmospheric gases when measuring along open paths. The second approach is illustrated by considering the effects of absorption by atmospheric gases when attempting to detect or measure hydrogen sulphide.

Hydrogen sulphide is a highly toxic gas with a Threshold Limit Value (TLV) of 1 Oppm and relatively weak optical absorption in the wavelength regions accessible to the LDS technique. For an LDS based hydrogen sulphide detector to be of use for safety applications it needs to be capable of detecting a 5 metro diameter cloud of gas containing hydrogen sulphide at 50%TLV (5ppm). This corresponds to 25ppm.m of hydrogen sulphide, which will produce a maximum fractional absorbance of the order of 1 X10-5 in the 1560nm to 1620nm region where the best absorption lines for hydrogen sulphide detection using LDS are found. Figure 12 shows the absorption spectrum of hydrogen sulphide between 1585nm and 1595nm for 25ppm.m of hydrogen sulphide and Figure 13 shows the corresponding absorption spectrum for a 100 metre path through the atmosphere at sea-level, 100%RH, 30 C. The 1585nm to 1595nm wavelength region is particularly good for the detection of hydrogen sulphide, suffering from lower levels of atmospheric absorption than found elsewhere between 1560nm and 1620nm. However, despite this there are still some relatively strong absorption features present due to carbon dioxide, water vapour and carbon monoxide. The conventional approach used in LDS systems to reduce the effect of such adsorptions would be to choose the target gas absorption line at which to make measurements very carefully, to minimise the effects of such lines. In this example, choosing a hydrogen sulphide absorption line in the 1589nm to 1590.1 rim region looks a very good option (See Figures 14 and15), but even in this exceptionally clear region, close inspection reveals a number of significant features arising from atmospheric gases, including complete absorption lines and baseline curvature due to distant, strong lines. The only region where atmospheric absorption is less than 1 x10-5 iS 1589.3nm to 1589.45nm but this is a region where the strongest hydrogen sulphide line will only produce a fractional absorbance of 3X10 for 25ppm.m of hydrogen sulphide. The hydrogen sulphide line at 1589.97nm produces a fractional absorbance of almost 9.3X10 for 25ppm.m but the atmospheric absorption due to a strong water vapour line nearby is 5.5X10-5 and is increasing very rapidly with wavelength. Consequently, there is a significant likelihood of atmospheric water vapour influencing any measurements made at 1589.97nm.

If the laser diode drive current is modulated with a waveform similar to that shown in Figure 9, such that it alternately scans the region around the hydrogen sulphide absorption line at 1589.42nm at frequency f for a period T/2 and the region around the hydrogen sulphide line at 1589.97nm at frequency f' for a period T/2, two calculations of the quantity of hydrogen sulphide present in the path can be made at frequencies f, and f2 and compared. If this comparison is performed in such a manner that a hydrogen sulphide alarm reading is only signalled when both measurements agree within a reasonable tolerance about the presence and amount of hydrogen sulphide gas in the monitored path, false alarms due to atmospheric absorption and / or insuffeient system signal to noise ratio can be greatly reduced. A calculated quantity Q' based upon slightly noisy measurements at 1589.42nm needs to be confirmed by a similar calculated quantity Q2 for the less noisy measurement at 1589.97nm; whilst a calculated quantity Q2 for measurements potentially influenced by atmospheric water vapour absorption at 1589.97nm needs to be confirmed by a similar calculated quantity Q. at 1589.42nm where the effects of absorption by atmospheric water vapour are negligible.

Also, since measurements at 1589.42nm are not affected by water vapour, they Can be used to keep track of any effects of water vapour absorption on the measurements made at 1589.97nm, allowing such effects to be compensated for. When used in this manner, the lower signal to noise ratio of measurements at 1589.42nm is not a significant problem, since the effects of water vapour will not normally change quickly and any Compensation can be based upon the average of measurements made over a number of minutes.

In the example described therefore, the measurements made at 1589.42nm provide three benefits: 1. They enable a reduction in the probability of false alarms due to system noise when compared and combined with the results of measurements made at 1589.97nm.

2. They enable the effects of water vapour absorption upon measurements at 1589.97nm to be reliably discriminated from the effects of any changes in the amount target gas concentration in the monitored space. Using this information it is possible to subtract the effects of water vapour absorption from measurements at 1589.97nm without compromising the ability of the apparatus to detect hydrogen sulphide.

3. They eliminate the possibility that even very sudden changes in atmospheric water vapour concentration can result in false alarms or spurious readings. Even if there has not been sufficient time to use the measurements at 1589.42nm to accurately compensate for the effects of water vapour absorption at 1589.97nm, the 1589.42nm measurements will not confirm the presence of hydrogen sulphide in its absence.

The benefits described cannot be achieved by simply modulating at a single frequency with amplitude sufficient to seen the wavelength range encompassing the two absorption lines to be measured. Firstly, there are a number of very strong atmospheric absorption lines between 1589.42nm and 1589.97nm and if the laser is scanned over these lines whilst measurements are being taken, these lines will introduce harmonic frequency components which will seriously interfere with measurements of the weaker hydrogen sulphide absorption lines. Secondly, scanning and measuring in this manner does not allow the magnitude of absorption at 1589.42nm and 1589.97nm to be measured independently, preventing use of results at 1589.42nm for compensation of the effects of water vapour upon measurements at 1589.97nm. Finally, scanning and measuring at a single frequency does not provide the false alarm rate improvement that can be achieved by scanning, measuring and comparing results for two, non- harmonically related frequencies.

The technique of modulating the laser such that it sweeps two distinct wavelength regions around two separate absorption lines of a single target gas at two differing modulation frequencies f and if, followed by comparison of the two measurement results to confirm that they are within an acceptable agreement tolerance can be beneficially employed in a number of different scenarios including those where: 1. The target gas has a relatively weak absorption line in a region of low atmospheric absorption and a stronger absorption line in a region of higher atmospheric absorption.

2. The target gas has two weak absorption lines in a region of low to moderate atmospheric absorption.

3. The target gas has two strong absorption lines in regions affected by absorption by different atmospheric gas species.

For hydrogen sulphide, there are absorption lines at 1582.13nm, 1589.24nm, 1589.42nm, 1589.54nm, 1589.97nm and 1593.05nm which might usefully be combined in the ways described to produce a highly reliable hydrogen sulphide detector.

The third approach which forms part of the claimed invention addresses the problems associated with coherence / fringe effects, these effects often setting the minimum measurement or detection limit in LDS based instrumentation or detection equipment.

The amplitude modulation produced by coherent interference exhibits a sinusoidal variation with wavelength. For coherent light of wavelength A, the phase difference expressed as a number of wave cycles (kin. between light leaving an optical surface and light returning having been reflected at a distance D from this optical surface is given by: 4)n = 2D I /\ The amplitude modulation produced by coherent interference between light leaving an optical surface and light returning from reflection at a distance D from an optical surface will go through a single sinusoidal cycle for a change in wavelength BA, given by equation: (A IF IDA) = 2D / (O'n -1) Figure 16 shows the amplitude modulation produced when a laser diode's wavelength is scanned over a target gas absorption line 11, a half cycle of a sinusoid 12 and a full cycle of a sinusoid 13, all fitted to the same wavelength interval. What can be seen from Figure 16 is that if the period of the sinusoidal modulation produced by coherence / fringe effects is of approximately the same width as the target gas absorption line, there is a probability that such modulation will start to correlate and interfere with the measurement or detection of the target gas absorption line.

Even with extremely careful design and engineering of an LDS system it is not possible to completely eliminate coherence / fringe effects from such systems, it is only possible to reduce their size or impact upon system performance. For LDS systems making measurements through open atmospheric paths, the situation is made considerably worse by the fact that the designer cannot control what happens in the open part of the system. Light can be scattered or reflected in the open path by rain-drops, snow, fog, mist, people or vehicles moving through the path. This results in light being scattered or reflected at distances and intensities over which the designer has little or no control. For an LDS system making measurements along an open path therefore, there is always the possibility of light being reflected or scattered at a distance which will create coherent fringes with a period which will correlate and interfere with the measurement of a target gas absorption line.

Whilst it is not possible to eliminate coherence / fringe effects from an LDS system, especially one operating along an open measurement path, it is possible to reduce the rate of false alarms arising from such effects by using a system configuration as shown in Figure 17. The LDS system shown in Figure 17 contains two laser diodes 1 and 14, operating at mean wavelengths A, and A2, these wavelengths corresponding to two different absorption lines of the same target gas. Lasers 1 and 14 are on a common temperature stabilised mount 2 and are driven by drive circuits 3 and 15 at electrical frequencies f and f' and scanned over wavelength ranges dA, and dA2 respectively. The outputs from lasers 1 and 14 are collimated by a common optical element 4, aligned such that optical radiation from both lasers reaches receiver optical element 6 after passing through monitored space 5. Receiver optical element 6 focuses optical radiation from both lasers onto a receiver detector 7, at which point the optical signals are effectively combined into a single electrical signal with principal frequency components f and f,. The signal from detector 7 is amplified by amplifier 8, digitised by ADC 9 and processed by signal processing system 16.

The electrical signal from detector 7 contains two sets of independent frequency components proportional to the amount of target gas present in the measurement path. A quantity of target gas Q' is calculated from the amplitude of frequency component f, for measurements made around wavelength A, whilst scanning over a range dA,; and a quantity of target gas Q2 is calculated from the amplitude of frequency component f2 for measurements made around wavelength A2 whilst scanning over a range JA2. These effectively independent measurements of the quantity of target gas in the monitored space can then be compared and treated as described earlier.

When addressing the problems associated with coherence / fringe effects, the use of two laser diodes in a configuration as illustrated by Figure 17 has the following benefits: 1. Wavelengths A, and A2 can be chosen independently, enabling a larger difference between these wavelengths to be realised than is possible when using a single laser to scan two absorption lines of the same target gas. This larger wavelength difference will result in significantly different coherence / fringe modulation periods when light from the two lasers is scattered or reflected by a common surface at a distance large compared to the wavelength (1000 x A). Consequently, the probability of coherence / fringe effects due to scattering or reflection simultaneously producing the same net interference effect at both measurement wavelengths is greatly reduced.

2. The lasers can be scanned over significantly different wavelength ranges, making it possible to ensure that light scattered or reflected from a single surface at a distance D cannot produce a sinusoidal amplitude modulation with a period closely correlated to that of the target gas absorption line for both of the lasers. In effect, there are two characteristic distances D and D' corresponding to BA' and BA2 respectively, at which scattering or reflection could be a problem for one laser but not the other. By deliberately spacing these characteristic distances such that they never coincide and are not harmonically related, scattering or reflection from a single surface cannot significantly interfere with both measurements at the same time.

3. The physical separation in x, y and z between the two lasers reduces the probability that light scattered or reflected from a common surface will introduce modulation with the same amplitude and phase onto both laser outputs. If necessary, the lasers can be deliberately mounted in positions at which common-mode interference will be further reduced.

In addition to the above benefits with relation to coherence / fringe effects, the use of two lasers operating at mean wavelengths A' and A2, scanning ranges BA' and BA2 at frequencies f and f', with measurement of frequency components f' and f2 has the following benefits: 1. Both of the lasers can be operated at their mean wavelengths whilst also producing optimum output power, enabling each measurement of target gas absorption to be performed upon signals of optimal amplitude and consequently optimal signal to noise ratio.

2. The choice of measurement wavelengths is not limited to the scanning range of a single laser. Wavelengths can be chosen with considerable freedom, enabling for instance two strong, distant lines to be measured, or two lines with low atmospheric interference to be measured, or two lines which are subject to interference by different atmospheric gases to be measured.

3. Since each laser is scanning their target gas absorption line at one frequency, one hundred percent of the time, there is no system signal to noise reduction such as that associated with scanning a single laser alternately at two frequencies.

4. The use of two lasers, scanning different wavelengths at different electrical frequencies makes it possible to treat each measurement as being completely independent of the other. Statistical calculations of the improvement in false alarm rate achieved can be relied upon when used to calculate false alarm probabilities that are too low to be measured by any economically justifiable programme of tests.

Another approach which forms part of the claimed invention and which reduces the probability of false alarms or spurious readings due to coherence / fringe effects is illustrated in Figure 18. In the LDS system illustrated by Figure 18, laser diodes 1, 14 are mounted on separate temperature stabilised mounts 2, 20 and their radiation is separately collected and collimated by transmitter optical elements 4, 21 prior to transmission through the monitored space 5 to receiver optical element 6 which focuses the radiation from both transmitter lasers onto optical detector 7. The optical elements 4, 6 and 21 are all chosen to have effective focal lengths and thicknesses which are different from each other by non-harmonic factors.

When addressing the problems associated with coherence / fringe effects, the use of a system configuration and geometry as illustrated in Figure 18 has the following benefits: 1. The separate optical paths within the transmitter for laser diodes 1 and 14 can be designed to ensure that there are no interference fringes with common periods affecting the outputs of both lasers. In particular, by using optical elements with different focal lengths and different thicknesses, it is possible to ensure that the unavoidable fringes resulting from reflection or scattering of light from the surfaces of these elements will have substantially different, non-harmonically related periods.

2. The use of separate optical elements with different focal lengths for each laser diode means that in the event of the build up of snow, condensation or contamination upon any exposed surfaces of the transmitter's optical elements, the optical radiation scattered or reflected back to each laser diode will be different in intensity and distribution. This is important because the interaction between a laser diode and back-scattered or back-reflected radiation is unstable and chaotic. The chaotic nature of this interaction introduces the possibility that oscillations or disturbances can appear on the output of the laser diode with periods that could not occur due to simple coherent interference. With two lasers experiencing different intensities and distributions of back-scattered or back-reflected radiation, the probability that both will simultaneously produce outputs which contain components which look like those of the target gas is greatly reduced.

The use of separate optical paths with optical elements of different effective focal lengths and thicknesses in the transmitter as illustrated in Figure 18 can be extended to include separate optical paths with optical elements of different effective focal lengths and thicknesses in the receiver. Applying this approach to the receiver enables systems with configuration such as those shown in Figures 19 and 20 to be realised, conferring similar benefits to those afforded systems configured as illustrated in Figure 18.

The techniques described and illustrated can not only be applied to the reduction of false alarms for LDS systems detecting small fractional absorbances. The techniques can also be applied to the reduction of noise and improvement of measurement accuracy when it is necessary to make measurements of gas concentrations that produce fractional absorbances too small to be measured accurately using conventional LDS techniques.

When measurement accuracy is important, it is not sufficient to just eliminate spurious readings by identifying circumstances where one or more of the measurements made by an LDS system cannot be relied upon, the measurements produced by the equipment must also be continuously maintained within an accuracy tolerance appropriate to the application.

The requirement to maintain measurements continuously within a prescribed accuracy tolerance whilst measuring gases which produce small fractional absorbances can be met by a system with the configuration shown in Figure 21.

This system has three laser diodes 1, 14, and 17 operating at mean wavelengths A,, A2 and As, driven by laser drive circuits 3, 15 and 18, scanning ranges Dan, dA2 and dA3 at electrical frequencies f, f' and f". The outputs from these laser diodes are collimated by transmitter optical element 4, passed through a monitored space 5 and then focussed onto receiver detector 7 by receiver optical element 6. The quantity of target gas present in the monitored space is proportional to the normalised amplitude of frequency components f,, f2 and f3 produced by absorption of optical radiation by target gas when each laser's wavelength scans across its respective target gas absorption line. In order to achieve the best results, the use of quantities A, Q2 and Q3 for calculation of the amount of gas in the monitored space should be in combination with results for previous measurements, using rules dependent upon the quality of agreement between the quantities. These rules and their intended effects are described below: 1. If Q, Q2 and Q3 are in close agreement, a large fraction of the average of A, Q2 and Q3 is added to a balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space. This enables the output from the apparatus to quickly track changes in the quantity of target gas present in the monitored space when confidence in all of the most recent measurements is high.

2. If either Q' and Q2. or Q2 and Q3, or Q. and Q3 are in close agreement with each other and the running average of previous results, the quantity which is not in close agreement is rejected whilst a large fraction of the average of the remaining quantities is added to a balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space. This prevents measurements in which confidence is not sufficiently high from affecting the output from the apparatus; whilst allowing measurements in which there is high confidence to contribute to keeping the output from the apparatus accurate and up to date.

3. If Q,.Q2 and Q3 are in reasonable but not close agreement with each other and the running average of previous results, a lesser fraction of the average of Q'.Q2 and Q3is added to a larger balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space. This enables the output to take account of the most recent measurements whilst limiting the impact that any potential errors in these measurements might have upon the output of the apparatus.

4. If Q,Q2 and Q3 are in reasonable but not close agreement with each other but not in close or reasonable agreement with the running average of previous results, a still lesser fraction of the average of Q'.Q2 and Q3is added to a still larger balancing fraction of the running average of previous results for calculating the quantity of gas present in the monitored space.

This enables the output to contingently take account of the most recent measurements whilst further limiting the impact that any potential errors in these measurements might have upon the output of the apparatus until measurements in which there is greater confidence are available.

5. If only one of the quantities Q'.Q2 or Q3is in close agreement with the running average of previous results, the other quantities are rejected and a fraction of the remaining quantity is added to a larger balancing fraction of the running average of previous results for use in the calculation of the quantity of gas present in the monitored space. This enables the output of the apparatus to be kept moving in the right direction in the event of there being low confidence in two of the most recent measurements. (This situation should not be common if the system is working correctly.) 6. If Q,.Q2 and Q3 are in poor agreement with each other and the running average of previous results, Q'.Q2 and Q3 are rejected and the calculation of the quantity of gas present in the monitored space is based solely upon the running average of previous results, the balancing fraction being set to unity. This enables spurious readings due to measurements in which confidence is low to be effectively eliminated, whilst maintaining the output from the apparatus at the most recent level in which confidence is acceptably high. (The use of a system configuration as shown in Figure 21 should make the probability of such a condition very low and if this condition persists it is likely that there is a problem which needs to be signalled to the user.) The above rules for the use of quantities Q., Q2 and Q3 in combination with the running average of previous results can be further refined by adjusting the fractions of recent and previous results used in proportion to the quality of agreement between them, such adjustment depending upon where in the agreement quality range for a particular rule the results fall.

When considering the laser diode to be used in systems such as those described for the claimed invention, some of the requirements placed upon this component are difficult to meet with the DEB laser diodes conventionally used in LDS systems. In particular, requirements for relatively large wavelength scanning ranges and for multiple lasers to operate at the right wavelength whilst on a common temperature stabilised mount are not readily met by DFBs. However, these requirements can be met by VCSEL laser diodes, which tune over a significantly larger wavelength range with drive current, this characteristic enabling the required wavelength scan to be realised whilst also facilitating adjustment of the mean output wavelength when the operating temperature is set to a common value. Long wavelength VCSEL laser diodes are virtually ideally suited for use in the claimed invention.

In the examples of the claimed invention described thus far, the laser diode radiation is collected and transmitted through the monitored space, subsequently illuminating a receiver detector. However, the claimed invention can also be beneficially employed in an apparatus where a sample of gas to be measured is drawn into a sample measurement chamber in order to be illuminated and measured using the approaches described. This arrangement might be of particular use in applications such as process control, or where it is not practicable to transmit a measurement beam through the gas without some prior sample conditioning.

In circumstances where it is necessary to measure extremely small fractional absorbances or where optical losses mean that the system signal to noise ratio is still insufficient to achieve the required degree of measurement integrity, there may be a benefit in further extending the approaches described. In particular, instead of scanning each target gas absorption line at a single electrical frequency, each target gas absorption line could be scanned at two, non- harmonically related electrical frequencies, with measurements of the absorption by each line being based upon the magnitude of two, similar order harmonics of the non-harmonically related scanning frequencies. This process could be carried out for each absorption line being scanned and where this process is performed simultaneously, all electrical scanning frequencies employed could be chosen to be non-harmonically related. !

The examples of the claimed invention illustrated in Figures 1, 17 and 18 show the output from the optical detector being amplified, digitised and subsequently processed by a digital signal processing system. This means of processing the signal from the detector is shown because it is particularly suitable for the measurement of multiple frequency components in a common signal. However, it is possible to realise the claimed invention in an apparatus where the magnitudes of the frequency components are determined by amplifying the detector signal and synchronously detecting the various frequency components using multiple synchronous detectors operating in parallel at different frequencies upon the same signal, the outputs from the synchronous detectors being subsequently digitised and used for the calculation of the quantity of target gas present in the measurement path as described earlier. This implementation requires a large amount of analogue electronic circuitry and does not remove the need for digitization of the signal data for further processing in the digital domain; but is a viable implementation of the claimed invention.

The examples of the claimed invention illustrated in Figures 17, 20 and 21 show the outputs from the laser diodes being collected and collimated by a common optical element 4 and transmitted through the monitored space 5 to be detected by a common optical detector 7. This means of collecting, transmitting and measuring the laser diode radiation is a simple and cost effective implementation of the claimed invention. However, it is also possible to implement the claimed invention as illustrated by Figures 19 and 22, with separate optical elements 4 and 21 collecting the laser diode radiation and transmitting it through the monitored space 5 to be focused onto optical detectors 7 and 23 by optical elements 6 and 22 respectively. The electrical signals from detectors 7 and 23 are amplified by amplifiers 8 and 24, digitised by ADCs 9 and 25; and processed by signal processing system 26, the processing of signals for calculation of the amount of gas present in the monitored space being the same as described earlier for the claimed invention.

The means of collection and collimation of the optical radiation from the laser diode(s) need not be limited to the simple optical elements shown in Figures 1, 17, 18 and 19. The optical elements used for this purpose can comprise of a number of separate optical elements combined to perform the required function of laser diode radiation collection, collimation and transmission through the monitored space. Furthermore, these optical elements need not be limited to the free-space optical elements shown. The radiation from the laser diode(s) can be coupled into fibre-optic cable(s) and carried to one or more optical elements which will collimate and transmit the radiation through the monitored space.

Readings or measurements from gas detectors or related instrumentation are output by various means, these mainly depending upon how and by what the readings or measurements are to be used. The means of output for readings or measurements from the claimed invention could include an analogue electrical signal proportional to the concentration or quantity of gas, a digital electronic signal conforming to a defined protocol and containing numerical data conveying the concentration or quantity of gas, a numericalrepresentation of the l concentration or quantity of gas upon a display and the opening or closing of relays at prescribed concentrations or quantities of gas.

Claims (20)

1. An apparatus for the detection or measurement of a specific target gas in a monitored space, the apparatus including a laser diode driven by a current comprising two components, a bias component and a sinusoidal component, the bias component alternating between two levels chosen to operate the laser diode at two mean wavelengths A' and A2, close to two separate optical absorption lines of the same target gas, the sinusoidal component synchronously alternating between two, non-harmonically related electrical frequencies f and ft at which the laser's wavelength is scanned across one or the other of the chosen absorption lines for a prescribed interval, the optical radiation from the laser diode being collected and transmitted through the monitored space and subsequently illuminating an optical detector, the electrical signal from this optical detector being amplified, digitised and processed to determine the magnitudes of frequency components f, f', f, and f2, where frequencies f' and f2 are similar order harmonics of the non-harmonically related electrical frequencies f and f', normalization of the magnitudes of f' and f2 with respect to their fundamentals, calculation of quantities Q' and Q2, separate estimates of the amount of target gas in the monitored space based upon the normalised magnitude of frequency components f' and f2, comparison of quantities Q' and Q2 to determine the quality of their agreement with each other and previous results for measurements made through the monitored space; and applying rules dependent upon this quality, use of Q' and Q2 in combination with previous results to calculate the quantity of target gas present in the monitored space, this calculated quantity of gas being output by the apparatus using conventional means.

2. An apparatus for the detection or measurement of a specific target gas in a monitored space, the apparatus including two or more laser diodes, each laser diode being driven by a bias current which causes it to operate at a mean wavelength close to a different optical absorption line of the same target gas and being scanned across this line by a sinusoidal current component at a frequency which is non-harmonically related to any other scanning frequency used, the optical radiation from all laser diodes being collected and transmitted through the monitored space and subsequently illuminating one or more optical detectors, the electrical signal from the detector or detectors being amplified, digitised and processed to determine the magnitude of components at the fundamental scanning frequencies and similar order harmonics of these fundamental frequencies, normalization of each harmonic with respect to the magnitude of its fundamental, calculation of separate estimates of the quantity of target gas present in the monitored space based upon each normalised harmonic, comparison of these quantity estimates with each other and previous results for measurements made through the monitored space and applying rules dependent upon 1'- this quality, use of these quantities in combination with previous results to calculate the quantity of target gas present in the monitored space, this calculated quantity of gas being output by the apparatus using conventional means.

3. An apparatus as claimed in Claims 1 and 2 where the wavelength scanning ranges for the laser diode(s) are non-harmonically related and have significantly different characteristic distances with respect to the formation of coherent interference fringes.

4. An apparatus as claimed in Claim 2 where the laser diodes are located in positions calculated to minimise formation of coherent interference fringes with common phase, amplitude or frequency.

5. An apparatus as claimed in Claim 2 where the radiation from each laser diode is collected and collimated by separate optical elements with different, non-harmonically related effective focal lengths and thicknesses.

6. An apparatus substantially as claimed in Claims 1 and 2, where each target gas absorption line is scanned at two, non-harmonically related electrical frequencies and measurements of any absorption by such lines are made by determining the magnitude of the two, similar order harmonics of the non-harmonically related scanning frequencies, this process being carried out for each absorption line being scanned and where this process is performed simultaneously, all electrical scanning frequencies being chosen to be non-harmonically related.

7. An apparatus as claimed in Claims 1 and 2 where the two mean wavelengths close to two separate optical absorption lines of the same target gas are chosen such that either, both are in regions of low absorption by atmospheric gases, or one is in a region of low absorption by atmospheric gases whilst the other is in a region of higher absorption by atmospheric gases, or one is in a region affected by absorption by one particular atmospheric gas species whilst the other is in a region affected by absorption by another atmospheric gas species.

8. An apparatus as claimed in Claims 1 and 2 where the rules governing the use of two estimated gas quantities Q' and Q2 in combination with results for previous measurements made through the monitored space to calculate the quantity of gas present in the monitored space are such that if Q. and Q2 are in close agreement, a large fraction of the average of Q. and Q2 is added to a balancing fraction of the running average of previous results; whilst if Q. and Q2 are in reasonable but not close agreement, a lesser fraction of the average of Q. and Q2 is added to a larger balancing fraction of the running average of previous results; whilst if only Q' or only Q2is in close or reasonable agreement with the running average of previous results, the quantity which is not in close or reasonable agreement is rejected whilst a lesser fraction of the close or reasonably agreeing quantity is added to a larger balancing fraction of the running average of previous results; whilst if Q. and Q2 are in poor agreement with each other and the running average of previous results, both Q. and Q2 are rejected and only the running average of previous results is used.

9. An apparatus as claimed in Claim 2 where three mean wavelengths close to three distinct optical absorption lines of the same target gas are chosen such that either, all are in regions of low absorption by atmospheric gases, or two are in regions of low absorption by atmospheric gases whilst the other is in a region of higher absorption by atmospheric gases, or one is in a region of low absorption by atmospheric gases whilst the others are in regions of higher absorption by atmospheric gases, or all are in regions affected by absorption by different atmospheric gas species or combinations thereof.

10. An apparatus as claimed in Claims 2 where the rules governing the use of three estimated gas quantities Q,Q2 and Q3 in combination with results for previous measurements made through the monitored space to calculate the quantity of gas present in the monitored space are such that if Q,.Q2 and Q3 are in close agreement, a large fraction of the average of I, Q2 and Q3is added to a balancing fraction of the running average of previous results; whilst if either Q' and Q2. or Q2 and Q3, or Q. and Q3 are in close agreement with each other and the running average of previous results, the quantity which is not in close agreement is rejected whilst a large fraction of the average of the remaining quantities is added to a balancing fraction of the running average of previous results; whilst if Qj,Q2 and Q3 are in reasonable but not close agreement with each other and the running average of previous results, a lesser fraction of the average of Q,.Q2 and Q3is added to a larger balancing fraction of the running average of previous results; whilst if Q'.Q2 and Q3 are in reasonable but not close agreement with each other but not in close or reasonable agreement with the running average of previous results, a still lesser fraction of the average of Q'.Q2 and Q3is added to a still larger balancing fraction of the running average of previous results; whilst if only one of the quantities Q,.Q2 or Q3is in close agreement with the running average of previous results, the other quantities are rejected and a fraction of the remaining quantity is added to a larger balancing fraction of the running average of previous results; whilst if Qj,Q2 and Q3 are in poor agreement with each other and the running average of previous results, Q'.Q2 and Q3 are rejected and only the running average of previous results is used. al

11. An apparatus as claimed in Claims 1 and 2 where the results of measurements performed upon target gas lines in regions of known low absorption by atmospheric gases are used to discriminate the effects of absorption by atmospheric gases in regions of more significant absorption by atmospheric gases from genuine changes in target gas concentration, thereby enabling any offsets arising from such absorption to be compensated for.

12. An apparatus as claimed in Claim 2 where all the laser diodes are located closely together on a common temperature stabilized mount and have their outputs collimated by a common optical element.

13. An apparatus as claimed in Claims 2 where the selection of the laser diodes and laser diode bias currents are such that all the laser diodes are simultaneously operating at their correct mean wavelengths with near optimal output power whilst at a common temperature.

14. An apparatus as claimed in Claims 1 and 2 where the diode lasers are VCSELs.

15. An apparatus as claimed in Claims 1 and 2 where the means of collecting the laser radiation and transmitting it through the monitored space includes combinations of free-space optical elements and fibreoptics.

16. An apparatus for the measurement of gas concentration based upon the apparatus claimed in Claims 1 and 2 where gas is drawn into a sample measurement chamber in order to be illuminated by laser diode radiation and measured as described in Claims 1 and 2, the measured gas concentration being output by conventional means.

17. An apparatus substantially as claimed in Claims 1 and 2 where instead of amplifying, digitising and digitally processing the detector signal(s) to determine the magnitudes of the various frequency components, the frequency component magnitudes are determined by amplifying the detector signal(s) and synchronously detecting the various frequency components using multiple synchronous detectors operating in parallel upon the signal(s), the magnitudes of the frequency components being subsequently digitised and used as described in Claims 1 and 2 to calculate the amount of target gas present in the monitored space.

18. An apparatus as claimed in Claims 1 and 2 which detects or measures hydrogen sulphide by measurement of any combination of two or more of the hydrogen sulphide absorption lines at 1582.13nm, 1589.24nm, 1589.42nm, 1589.54nm, 1589.97nm and 1593.05nm

19. An apparatus as claimed in Claims 1 and 2 where the means of output for the concentration or quantity of gas calculated present in the monitored path or sample measurement chamber includes an analogue electrical signal proportional to the concentration or quantity of gas, a digital electronic signal conforming to a defined protocol and containing numerical information conveying the concentration or quantity of gas, a numerical representation of the concentration or quantity of gas upon a display which is associated with or forms part of the apparatus, or the opening or closing of relays at prescribed concentrations or quantities of gas, such relays and the necessary control circuitry either being associated with or forming part of the apparatus.

20. An apparatus for the detection or measurement of gases substantially as herein described above and illustrated in the accompanying drawings.