GB2345134A - Gas sampling and detecting system - Google Patents

Abstract

The system for detecting a target gas in an atmosphere has a Venturi V connected to an atmosphere A to extract a sample SG there from while the sample is simultaneously diluted with the venturi driver gas CG from a source B. A gas detector 12 is in communication with the Venturi output line 11.

Description

Detector systems This invention is concerned with detector systems, and relates in particular to the sampling and detecting of gases as might be required for leak testing. More specifically, the invention concerns the method of a sampling technique described in our co-pending British Patent Application No: 98/28,293.2 (P1486 : Publication No :.,...,...).

Many commercial products are or employ containers loaded with some liquid or gaseous material-typical examples might be an aerosol can of fly killer, or a bottle of scent-and it is an obvious requirement that they should not leak, lest the contained material noticeably seep away before the end of the product's shelf life, possibly contaminating the environment or posing a fire hazard as well as causing an economic loss. It is therefore highly desirable to check such filled containers for leakage as they are being made, and a presently-favoured way of carrying out such a check is to sample the atmosphere around the container, and then to employ some form of detector to show up the presence in the sample of one or other of the constituents of the container's contents that might be leaking out, slotting the sampling means and the detector into the manufacturing process-the production line, say-as a stage therein. Instances of this, and specifically detection (employing a photoionisation detector, or PID) as part of a line filling cans with an aerosol material and/or a propellant therefor, are described in our aforementioned Application.

Products manufactured for the mass market are often constructed very quickly-for example, in a line filling aerosol cans, ten thousand cans may be filled every hour, so that several filled cans may be produced every second, requiring tests on seal integrity to be performed at production speeds. Manufacturers may demand very reliable leak testing on all parts in a production line to low leak detection thresholds.

Detection of gross leaks, as occasioned by a faulty seal, is relatively easy, even at such a high throughput rate. However, 100% leak detection to low leaking levels, as might be occasioned by a pin-prick hole in an aerosol can or a faulty valve, is, with present-day systems, virtually impossible at the necessary reliability demanded by manufacturers on their production lines. The main difficulty stems from the absolute requirement to bring the carrier gas potentially conveying leaked material from a manufactured part to the necessarily remote detectorto do this very rapidly, and within the available fraction of a second before the next manufactured part is delivered to the testing station; while this may often be achieved simply by having a large throughput of gas relative to the volume of the pipework between the leak source and gas detection means, a serious problem arises when a grossly-leaking (or severely contaminated) part enters the testing station, which is that so much leak material is entrained with the carrier gas that the detector (and its associated pipework) itself becomes severely contaminated. As a result, the detector and pipework require considerably longer to be flushed out and so become ready to make a meaningful test upon the next manufactured part. The consequent delay often necessitates a stop in the manufacturing process, with a consequential loss in production.

In the Specification of our aforementioned Application it is suggested that this problem of detector contamination and the consequent period to flush out the system can be overcome, and a leak anticipated, by using delivery means (to convey to the detector a sample of the carrier gas/leaked material combination) which incorporates sampling and dilution means for taking and thinning a sample to form a subsample which is then supplied to the detector. More particularly, this earlier invention is a leak testing station which comprises: a sealable enclosure within which a container to be tested may be positioned ; gas supply means for passing through the enclosure a carrier gas in which can be entrained the relevant content material if leaking from the container; and a detector operable to detect the relevant content material, and delivery means to convey thereto carrier gas emanating from the enclosure; wherein the delivery means taking gas from the enclosure and delivering it to the detector comprises a primary delivery channel taking gas directly from the enclosure and feeding a secondary channel supplying sample gas directly to the detector, there being dilution means for mixing that gas with diluent in the secondary channel, to form a diluted sub-sample, before it is supplied to the detector.

As then explained, the actual manner in which the main sample is itself sampled (to form the sub-sample), and the sub-sample then diluted, and the material that is used for this dilution, may of course be any that is convenient-for example, there may be extracted from the sample gas in the primary channel a small volume thereof, and this may be fed along the secondary channel to a point where there may be injected thereinto some other, inert (so far as the leak material and the detector are concerned) gas, perhaps from a cylinder thereof. One way of achieving this might be to use a vacuum pump or such like on the detector exhaust to draw the small volume in along the secondary channel, along with some dilution gas. However, though possible, this is not preferred, because it would require the detector to be leak-tight to ensure that all of the gas flow was actually drawn from the main sample primary channel rather than in part from the local environment (perhaps leading to unwanted and uncontrolled spurious signal responses). Furthermore, the vacuum pump would have to be"throttled"else the flow through the detector would be too high (such throttling is normally detrimental to a pump), or a flow manifold would have to be used to divert some background air into the detector gas stream.

As then further explained, the taking and dilution of a sample of the main sample (to form the sub-sample) is most preferably effected using a (conveniently conventional) gas-driven Venturi extraction system (similar, indeed, to the water-powered vacuum pumps seen in most chemical laboratories). More specifically, the secondary channel contains the Venturi extractor, the vacuum effect of which is used to remove from the main sample in the primary channel a small part thereof, and the gas driving the Venturi is then intimately mixed with the extracted part to dilute it down to give the desired sub-sample itself. The gas-driven Venturi can of course be powered by the same carrier gas as is utilised to provide the main sample in the first place.

Thus, as explained in the Specification of our aforementioned Application, the problem of overloading the detector with gas from a grossly leaking part is solved by the extraction and dilution of a sub-sample, this being preferably performed by a Venturi system.

It has now been appreciated that this use of a Venturi extraction system is in itself an extremely efficient method of removing a sample (or a sub-sample) from a gas (or a main sample of gas), novel and inventive in its own right, and it is to this that the present invention is directed. More specifically, the invention concerns the use, in any sampling and detecting system, of a Venturi system, operatively connected to a volume of gas to be sampled, to extract a sample from that gas and simultaneously to dilute it with the Venturi driver gas. In the context of the present invention a Venturi system is an elongate chamber into and along which projects a driver-gas pipe, the chamber having an input port located behind the nozzle of the pipe and an output port located in front of and spaced from the nozzle, so that when driver gas is injected into the chamber there is a resulting drop in pressure behind the nozzle which is communicatable through the inlet port. Thus, if-as in the invention -the inlet port is connected to the volume of gas to be sampled while the outlet port is connected to the detector, then in operation a sample of gas is drawn through the inlet port into the chamber by the drop in pressure, there mixed with driver gas, and then is blown out of the chamber through the outlet port and on to the detector.

As will become clear from what follows, the Venturi-using gas detector system of the present invention offers distinct advantages over other systems in terms of simplicity of fabrication and operation, improved sensitivity, detector stability, and detector lifetime. More specifically, the system of the present invention overcomes the difficulties of other arrangements by enabling a continuous gas sample to be drawn, without the use of a vacuum pump, from a local ambient atmosphere, or from a gas stream at or near ambient pressure being conveyed from a remote source, irrespective of the conveyed gas flow rate and direction. The invention's system allows the sample gas to be diluted by a conditioning gas without the use of complicated apparatus, and also enables a stable and adjustable flow of gas to be conveyed through a detector, this gas containing a stable and adjustable ratio of sample gas to conditioning gas, unaffected by the flow of sample gas to its point of entry into the gas sampling system.

In one aspect, therefore, this invention provides a gas detection system for determining the presence of a target gas in an atmosphere, the system comprising : a Venturi system operatively connected to the atmosphere so as in use to extract a sample therefrom and simultaneously to dilute it with the Venturi driver gas; and a detector operatively connected to the output of the Venturi system for receiving the diluted sample and detecting any target gas present therein.

In a second aspect, this invention provides a method for determining the presence of a target gas in an atmosphere, the method comprising: sampling the atmosphere in which the target gas may be present using a Venturi system operatively connected to the atmosphere, and diluting the sample within the Venturi system with the Venturi driver gas; and feeding the diluted sample from the Venturi system to a detector able to detect any target gas present in the diluted sub-sample, and then detecting any such present target gas.

The invention concerns determining the presence of a target gas in an atmosphere, and can be applied to many situations in which the detection of a target gas is necessary, such as in the domestic gas supply industry, in perfume manufacture, in the solvent industry, and in the liquid petroleum industry. One important use of the invention, as discussed further hereinafter, is in leak testing-for detecting containers that are leaking their content material-and in this context the operation of the invention includes the steps of placing the container in an enclosure at a testing station, taking a sample of the ambient atmosphere around the container within the enclosure, and then using a Venturi system for both sub-sampling and heavily diluting that sub-sample (to around 100th of the concentration: this is discussed further hereinafter) to form a diluted sub-sample which is supplied to the detector.

The target gas to be detected can be any gas that needs to-and can-be detected. Those gases commonly found in containers (such as aerosol cans and the like) are typically the large organic molecules commonly found in scents, solvents, and liquid petroleum-based fuels (mostly the alkanes, from butane on). In general, though, there is an increasing need for the detection of organic gases and vapours or organic volatiles where such materials present a health risk, an environmental threat, a fire risk, or an economic loss.

The invention determines the presence of the target gas in an atmosphere. The atmosphere can of course be any wherein the target gas is to be detected-it could, for example, be the ambient air within a laboratory, or even within an entire factory-but where, as noted hereinbefore, the invention is to enable leak testing for containers, the atmosphere is the volume of gas within an enclosure at a testing station. A typical enclosure conveniently has a platform, on which in use there stands the container being tested, and a removable cover, sealable to the platform, to go over, and enclose, the container (this type of use of the invention is described in more detail in our aforementioned Application).

In such a case the container to be tested can of course be of any sort, and can contain any type of material for which a detector exists. Examples of typical container types are aerosol cans, gas cartridges, and temperature controllers (devices, like car radiator thermostats, that contain a working fluid that expands as it gets hot).

The invention requires a Venturi system operatively connected to the atmosphere so as in use to extract a sample therefrom. The simplest way of achieving this is to provide the Venturi's input port with direct access to the atmosphere to be sampled and tested, and this can be done merely by placing the Venturi chamber, with its input port in the wall thereof, in the relevant atmosphere (and having a multiplicity of input ports can improve sample extraction). However, in a preferred system, when it may be necessary to deal with very substantial leakages, it is best to employ primary means for sampling the atmosphere, and then to convey the thus-obtained sample to sub-sampling means where there can be extracted (by the Venturi) a secondary, or sub-, sample which is then fed directly to the Venturi for dilution and onwards conveyance to the detector. Most preferably, therefore, the invention involves primary means for sampling the atmosphere in which the target gas may be present, and means for conveying the sample from the sampling means to a sub-sampling point to which the Venturi system is operatively connected.

The primary means for sampling the atmosphere may at its most basic be merely a port, in the means conveying the sample onwards, opening into the relevant atmosphere, together with means for drawing some of that atmosphere through the port, into the conveying means, and away to the detector. However, when utilising the invention in association with a leak testing station with an enclosure, while there is indeed such a port the actual sampling is most conveniently effected by positively passing into the container-filled enclosure a (detector-inert) carrier gas in which can be entrained the relevant content material if leaking. The carrier gas, and its mode of supply, may be any appropriate; usually, for example, the gas is air, either from a cylinder of compressed air or directly from the atmosphere (but perhaps through filters and cleaners, to ensure it is not already contaminated).

Following a primary sampling, the invention involves conveying the acquired sample to a sub-sampling point. More specifically, there is conveying means comprising a primary delivery channel taking sample gas and feeding, at a sub-sampling point, a secondary channel leading to the Venturi system the output of which then supplies diluted sub-sample gas directly to the detector. The conveying means can be of any suitable form-a conduit, pipe, pipeline, channel, duct, passage or the like made from any appropriate material, such as a metal (e. g. copper or stainless steel), a plastic, or a glass. Any such pipes are conveniently smooth-walled and of a relatively narrow bore (and preferably narrower when carrying the subsample to the Venturi and thence to the detector).

The sub-sampling point is most conveniently a simple T-junction operatively connected (by the secondary channel) to the Venturi system. Only some of the sampled gas is taken at the sub-sampling point and delivered to the Venturi system ; the rest is sent on along the other branch of the T-junction to an exhaust point, where it is disposed of (usually simply by being vented to the external atmosphere). The exhaust point is conveniently located at a distance from the sub-sampling point sufficient to eliminate the effect of gas from the external atmosphere being drawn in and contaminating the carrier gas at the sub-sampling point.

As specifically applied in the preferred forms of the invention, the secondary channel contains the Venturi extractor the vacuum effect of which is used to remove from the main sample in the primary channel a small part thereof, and the conditioning gas driving the Venturi is then intimately mixed with the extracted subsample to dilute it down to give the desired diluted sub-sample feedable directly to the detector.

In using a Venturi system, a sample (or a sub-sample) can be extracted from the atmosphere, and effectively simultaneously diluted within the Venturi system with the Venturi driver gas. The Venturi driver, or"conditioning", gas can be supplied by a pump or cylinder or other source at positive pressure relative to atmosphere, the positive pressure at source driving the conditioning gas into the Venturi chamber and therefrom into the detector. Conveniently, there is control of the source, in the form of regulators and restrictions, to ensure the constant flow of conditioning gas into the system.

The conditioning gas can be any appropriate gas that is inert relative to the detector and target material, and is preferably one that enhances the sensitivity of the detector to the target material (the detector's sensitivity towards, say, butane may be higher when the butane is in nitrogen mixed with air than when it is in air itself). Thus, when detecting the presence in air of an organic such as butane, using nitrogen as the conditioning gas is preferred.

A useful type of detector is the photoionisation detector (PID), as discussed further hereinafter.

However, PIDs are sensitive to many volatile organic compounds, and therefore the conditioning gas should be substantially free from any such organics. Moreover, the performance of a PID may be affected by high humidity levels, in which case it is preferable for the humidity of the conditioning gas to be low.

Feeding the diluted sample from the Venturi system to the detector is effected conveniently by short, simple smooth-walled pipes of relatively narrower bore (smooth walls and a narrow bore means there is less chance of significant amounts of contaminated carrier gas being trapped in the pipes rather than being flushed out, and the narrower the bore the lower the transit time of a sample, or sub-sample, on its way towards the detector, so the faster-acting the system).

In the invention, a detector is operatively connected to the output of the Venturi system for receiving the diluted sample (or sub-sample) and then for detecting any target gas present therein. The detector to be used in this context may be of any convenient sort, but is preferably a photoionisation detector-PID-or an electrochemical sensor (such a sensor would be suitable for detecting the presence of hydrogen in air). An example of a typical multi-purpose PID is that sold by Ion Science Ltd, Fowlmere, Cambridge under the name PHOTEC GP1.

PIDs are normally capable of detecting concentrations of material (in the carrier gas) that are as low as one part per million, but can be completely swamped, and temporarily disabled, by a concentration as low as one part per thousand. On the other hand, a container leaking because of a pin prick in its casing, or a faulty valve, may release sufficient material to provide, in the carrier gas, a concentration of one part per thousand, while a massively faulty container may provide a concentration of as much as one part in ten.

Clearly, if, to feed the detector, the sample (or sub-sample) taken from the sub-sampling point is diluted to around one hundredth, then the detector will still be able to detect a small leak but will not be overwhelmed by a large one.

As long as the sample (or sub-sample) taken from the basic sample is subsequently substantially diluted then, even though the container (say) being tested were suffering a massive failure, the amount of leaking material actually supplied to the detector would still be sufficiently low not to affect detrimentally the ability of the detector itself to recover, and to be reusable, within a very short time.

Accordingly, the substantial dilution within the Venturi should be sufficient to reduce to a very small level the likelihood of a massive leak overwhelming the detector and yet not so high as to reduce the actual material concentration to below the level at which the detector can detect the leak.

Having passed into the detector, the diluted sample (or sub-sample) is tested and then vented to atmosphere.

It is obviously preferable for the detector gas outlet to be as remote from the sample gas as possible.

Embodiments of the invention are now described, though by way of illustration only, with reference to the accompanying Drawings, in which: Figure 1 shows a schematic representation of one embodiment of the invention, pertinent to the sampling and detection of a gas conveyed from a remote source; Figure 2 is a schematic representation of a Venturi system of another embodiment of the invention; Figure 3 is a schematic representation of a Venturi system with associated sample gas piping in a further embodiment of the invention; and Figure 4 shows a diagrammatic view of a test station incorporating the invention and typical gas flows.

Figure 1 shows a source (A) of a sample gas (SG) which is conveyed along piping (1) to a sub-sampling point (2). Line 1 is sufficiently restrictive to prevent too rapid, and hence turbulent and unstable, flow of the sample gas traversing through it, but not so restrictive as to cause unacceptable delay in the onwards conveyance of sample gas SG. The sample gas SG is at or near atmospheric pressure to enable the Venturi system to draw a sub-sample of gas from the sample line piping 1.

Beyond the sub-sampling point 2 the sample gas travels through an extension pipe (3) to an exhaust point (4) which is open to the atmosphere. The additional length of pipe 3, venting to atmosphere at 4, beyond the sub-sampling point 2 is to prevent gas from the ambient atmosphere being drawn in; for a sample gas source A feeding a pipe 1 of 1 cm diameter and flow rate of 50 cm3 min-l, an adequate length of piping between the sub-sampling point 2 and the vent to atmosphere 4 is 5 cm.

At the sub-sampling point 2 some of the gas sample SG is extracted by a Venturi (generally V). The Venturi chamber (5) is fed along piping (6) by a source (B) of a conditioning gas (CG); this is in turn fed to the narrow driver gas pipe (7) within the Venturi chamber 5.

The Venturi is also connected, at its inlet port (8), by piping (9) to the sub-sampling point 2; injection of driver gas CG through the nozzle (10) of the pipe 7 results in a drop in pressure behind the nozzle, causing a sub-sample of sample gas to be drawn from the sub-sampling point 2 into the piping 9 and thence to the Venturi chamber 5. In order for the Venturi to induce the maximum flow of sample gas SG, the nozzle 10 vents the conditioning gas CG into the chamber some appreciable distance upstream from the inlet port 8 (if the flow of conditioning gas is high and the distance between nozzle 10 and port 8 is small there is little withdrawal of the sample gas to form a subsample).

The Venturi's output port is connected, by detection gas line piping (11), to the detector (12); the piping 11 is co-axial with the injection pipe 7.

And, because the induced flow of sample gas SG increases with increased velocity of the driver gas injected into the chamber 5 through the piping 7, this latter piping is of minimum bore so as to maximise the induced flow of sample gas SG for a given flow rate of conditioning gas CG.

The diluted sub-sample gas exiting the Venturi chamber 5 comprises a mixture of sample gas SG and conditioning gas CG. It is conveyed on through the detection line piping 11; part of it is delivered to the detector 12 while the rest is vented to atmosphere by means of a gas outlet (13) so as to prevent the detector 12 becoming overloaded with diluted sub-sample gas. Once the diluted sub-sample has passed through the detector 12, it is conveyed and vented to atmosphere by further piping and exit port (14,15).

Figure 2 shows a detail of the Venturi used in the embodiment of the invention shown in Figure 1.

The induced flow of sample gas SG may be adjusted by repositioning the Venturi injection piping 7. A seal is made between the Venturi chamber 5 and the conditioning gas line piping 6 by means of a sleeve (16) and nut (17), which sleeve jams the piping 6 in place. Loosening the nut 17, and thus unjamming the sleeve, enables the piping 6 to be pushed into or pulled out of the chamber 5, so as to vary the distance between the nozzle 10 and input port 8; this in turn correspondingly varies the sample gas SG flow relative to the conditioning gas CG flow. The faster the flow of conditioning gas the further behind the nozzle 10 is the point of maximum pressure drop, and therefore the further forward the nozzle should be in order to align that pressure drop point with the sample gas inlet port 8.

Figure 3 shows a detail-the Venturi-of another embodiment of the invention; this embodiment is useful where the sample gas source A is a large enclosure with the apparatus of the invention located therein to detect the presence of a chosen gas. The sample gas is drawn into the Venturi through a plurality of perforations (18) in the wall (19) of the chamber 5 disposed behind the injection pipe nozzle 10. In this embodiment there is no separate sample gas line 1.

Figure 4 shows an entire gas detection system employing a Venturi arrangement of the invention to take a sub-sample from a main sample.

The detection system includes a testing station (20), which essentially comprises a removable container-sized sealable enclosure (21), a platform (22) with seals (23), a primary pipeline (24), a sample gas delivery pipeline (25), a conditioning gas delivery pipeline (26), a valve (27), a Venturi system feeding a detector 12, and indicator/warning system electronics (28) operatively connected to the detector 12.

As shown, a container (29) being tested stands on the platform 22 and is sealed inside the enclosure 21 using the seals 23. The sample gas SG and the conditioning gas CG are, in this case, the same and delivered conveniently into the testing station along sample gas delivery pipeline 25 which has a branch pipeline 26 to deliver conditioning gas CG to drive-the Venturi system. Leaking gas from the container 29 is conveyed, mixed with sample gas SG, along the primary pipeline 24 to the sample gas pipeline 1 via a valve 27.

The extraction, dilution and detection of a subsample of gas occurs as already described hereinbefore.

The detector 12 is operatively connected with an electronic control and measurement system 28. This system 28 supplies the voltage required by the detector 12 and measures and interprets the current output by the detector 12 when in operation. The electronics 28 also provides warning and/or alarm signals when a leak is detected from the container 29 being tested.

Claims (10)

1. Claims 1. A gas detection system for determining the presence of a target gas in an atmosphere, the system comprising; a Venturi system operatively connected to that atmosphere so as in use to extract a sample therefrom and simultaneously to dilute it with the Venturi driver gas; and a detector operatively connected to the output of the Venturi system for receiving the diluted sample and detecting any target gas present therein.
2. 2. A detection system as claimed in Claim 1, wherein there is primary means for sampling the atmosphere in which the target gas may be present, and means for conveying the sample from the primary sampling means to a sub-sampling point to which the Venturi system is operatively connected, and at which a secondary, or sub-, sample is taken by the Venturi system.
3. 3. A detection system as claimed in Claim 2, wherein the sub-sampling point is a simple T-junction in the means for conveying the sample from the primary sampling means, and is operatively connected to the Venturi system.
4. 4. A detection system as claimed in Claim 3, wherein the detector is a photoionisation detector.
5. 5. A detection system as claimed in any of the preceding Claims and substantially as described hereinbefore.
6. 6. A method for determining the presence of a target gas in an atmosphere, the method using a gas detection system as claimed in any of the preceding Claims and comprising: sampling the atmosphere in which the target gas may be present using a Venturi system operatively connected to that atmosphere, and diluting the sample within the Venturi system with the Venturi driver gas; and feeding the diluted sample from the Venturi system to a detector able to detect any target gas present in the diluted sub-sample, and then detecting any such present target gas.
7. 7. A detection method as claimed in Claim 6 and which is applied to the leak testing of a container, which method includes the steps of placing the container in an enclosure at a testing station, taking a primary sample of the ambient atmosphere around the container within the enclosure, and then using a Venturi system for both sub-sampling and diluting that sub-sample to form a diluted sub-sample which is supplied to the detector.
8. 8. A detection method as claimed in Claim 7, in which the actual primary sampling is effected by positively passing into the container-filled enclosure a detectorinert carrier gas in which can be entrained the relevant content material if leaking, and then sampling the thus-loaded carrier gas.
9. 9. A detection method as claimed in any of Claims 6 to 8, in which the conditioning gas is one that enhances the sensitivity of the detector to the target material.
10. 10. A method for determining the presence of a target gas in an atmosphere as claimed in any of Claims 6 9 and substantially as described hereinbefore.