METHOD AND APPARATUS FOR RECOGNIZING THE PRESENCE OF LEAKAGES FROM SEALED CONTAINERS

Abstract

A method for recognizing the presence of leakages from sealed containers includes defining a detection zone in which a sealed container will be placed, putting the detection zone in communication with at least one gas sensor through at least one duct, introducing a flushing gas into the detection zone through the duct, placing a container in the detection zone; and sucking a gas flow from the detection zone through the duct and transferring it to a first sensor. The gas flow that reaches the first sensor is either transferred to a second sensor or to the first sensor for a second time. The signals generated by the two sensors, or the two signals generated by the same sensor, are processed for determining the presence of a gas leakage in the container.

Claims

1. A method for recognizing the presence of leakages from sealed containers, said method comprising the steps of: defining a detection zone (13) in which a sealed container (CT) will be placed; putting said detection zone (13) in communication with at least one gas sensor (219a, 219b) through at least one duct (21); placing a container (CT) in said detection zone (13); sucking a gas flow from said detection zone (13) through said duct (21) and transferring said gas flow to a first sensor (219a, 219b), the method being characterized in that it further comprises a step in which said gas flow that has reached the first sensor is either transferred to a second sensor or is transferred to the first sensor for a second time, and in that the signals generated by the two sensors, or the two signals generated by the same sensor, are processed for determining the presence of a gas leakage in said container (CT).

2. The method according to claim 1, wherein said gas flow sucked from the detection zone is transferred from the first sensor to the second sensor in a seamless manner.

3. The method according to claim 1, wherein said gas flow sucked from the detection zone is transferred for a first time to the first sensor in a first direction and for a second time to the first sensor in the opposite direction.

4. The method according to claim 3, wherein the two signals generated by the sensors, or the two signals generated by the same sensor, are compared with each other so as to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

5. The method according to claim 4, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively.

6. The method according to claim 4, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively at a rate higher than a predetermined noise threshold.

7. An apparatus for recognizing the presence of leakages from sealed containers, said apparatus comprising: a detection zone (13) adapted to receive a sealed container (CT); a first gas sensor (219a, 219b) capable of generating a signal indicative of the presence of gas(es); a duct (21) communicating with said detection zone (13) and said first gas sensor; a suction fan (25) provided with an inlet port (25a) from which air is sucked and which communicates with said duct (21), the apparatus being characterized in that it further comprises a second sensor capable of generating a signal indicative of the presence of gas(es) and arranged between the first sensor and the suction fan, or means capable of transferring back to the first sensor an air flow that has been sucked from said detection zone and has left the first sensor, and in that it further comprises processing means for processing the signals generated by the two sensors, or the two signals generated by the same sensor, for determining the presence of a gas leakage in said container (CT).

8. The apparatus according to claim 7, wherein said first and second sensors are arranged in series along said duct, and wherein the first sensor is located upstream of the second sensor in the direction in which air is sucked by the suction fan.

9. The apparatus according to claim 7, wherein said means include a reversible fan or a reversible suction fan that are capable of generating a flow in a direction opposite to the direction in which the flow was transferred from the detection zone to the sensor.

10. The apparatus according to claim 9, wherein an electronic control unit is provided, programmed to compare the signals generated by the two sensors, or the two signals generated by the same sensor, and to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

11. The apparatus according to claim 8, wherein an electronic control unit is provided, programmed to compare the signals generated by the two sensors, or the two signals generated by the same sensor, and to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

12. The apparatus according to claim 7, wherein an electronic control unit is provided, programmed to compare the signals generated by the two sensors, or the two signals generated by the same sensor, and to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

13. The method according to claim 2, wherein the two signals generated by the sensors, or the two signals generated by the same sensor, are compared with each other so as to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

14. The method according to claim 13, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively.

15. The method according to claim 13, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively at a rate higher than a predetermined noise threshold.

16. The method according to claim 1, wherein the two signals generated by the sensors, or the two signals generated by the same sensor, are compared with each other so as to generate a signal indicative of the presence of a gas leakage when at the same instant the level of the second signal exceeds the level of the first signal.

17. The method according to claim 16, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively.

18. The method according to claim 16, wherein the signal indicative of the presence of a gas leakage is generated when the condition according to which at the same instant the level of the second signal exceeds the level of the first signal has repeated at least twice consecutively at a rate higher than a predetermined noise threshold.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0044] Some preferred embodiments of the invention will be provided by way of non-limiting examples with reference to the accompanying Figures, in which:

[0045] FIG. 1 is a side perspective view of an equipment in which the apparatus according to the invention is incorporated;

[0046] FIG. 2 is a schematic representation of a preferred embodiment of the apparatus according to the invention;

[0047] FIGS. 3A to 3E are graphs of as many tracer gas concentration signals;

[0048] FIG. 4 is a graph comparing two tracer gas concentration signals of different intensities;

[0049] FIGS. 5A to 5C are graphs of as many gas concentration signals that do not indicate a leakage;

[0050] FIG. 6 is a diagram of the comparison circuit of the second embodiment of the invention;

[0051] FIG. 7 is a graph of a pair of gas concentration signals generated by a pair of sensors in the second embodiment of the invention.

[0052] In all Figures, the same reference numerals have been used to denote equal or functionally equivalent components.

DESCRIPTION OF A PREFERRED EMBODIMENT

[0053] Referring to FIG. 1, according to the embodiment of the invention illustrated, a sample container to be tested is placed in detection zone 13, defined in an equipment 10 incorporating the apparatus according to the invention, by means of a positioning assembly 51. According to this embodiment, positioning assembly 51 includes a pair of conveyor belts 53, 55 for the introduction or entrance of the container into detection zone 13 and for the extraction or exit of said container from detection zone 13, respectively. Preferably, said positioning assembly 51 further includes a pair of side guides 55a, 55b for correctly positioning the container in detection zone 13, preferably centrally of zone 13.

[0054] Optionally, the method according to the invention includes a step in which the sample container undergoes a compression or squeezing step, for promoting possible gas spillage. Preferably, said squeezing step is performed by means of a squeezing assembly including rotatable rollers.

[0055] Reference will now be made to FIG. 2 for describing a preferred embodiment of an apparatus 11, made in accordance with a particular embodiment of the invention and arranged to implement a method of recognizing leakages capable of considerably increasing the sensitivity of the recognition itself.

[0056] Referring to FIG. 2, there is schematically shown a detection apparatus 11 made in accordance with a preferred embodiment of the invention and including a detection zone 13. Detection zone 13 is arranged to receive a sealed container CT that is to be checked to ascertain the presence of possible leakages, i.e. of openings capable of putting the content of container CT in communication with the surrounding environment outside the container. In accordance with a preferred embodiment of the invention, detection zone 13 is defined by a supporting structure 15 including a frame 17 and it communicates with the outside environment.

[0057] In FIG. 2, reference numerals 219a and 219b denote two gas sensors serially connected in a same duct 21 for the gases coming from detection zone 13.

[0058] Both gas sensors 219a, 219b are arranged to generate an electrical signal indicative of the presence of a specific gas in a gas mixture passing through said sensors 219a, 219b. In a particular embodiment of the invention, said gas is CO.sub.2 and sensors 219a, 219b are infrared CO.sub.2 sensors each including a measurement cell equipped with an IR emitter and a corresponding photodetector. The gas mixture to be analysed, when passing through the measurement cell in sensor 219a or 219b, causes an alteration in at least one parameter of an electrical signal passing in an electrical circuit associated with the photodetector. The alteration is proportional to the amount of CO.sub.2 being present, i.e. to the CO.sub.2 concentration in the mixture passing through sensor 219a, 219b. In other embodiments, gas sensors of different type could be provided to detect CO.sub.2 with different modalities, or to detect gases of different kinds, for instance He or H.sub.2. Such sensors are known to the skilled in the art and therefore they will not be described in more detail.

[0059] In the embodiment illustrated, apparatus 11 includes a suction fan 25 having an inlet port 25a, through which air is sucked, communicating with said duct 21, and an outlet port 25b for exhausting the air sucked to the outside. Always with reference to the embodiment illustrated, duct 21 includes a first segment 21a connected between detection zone 13 and the first sensor 219a, a second segment 21b connected between sensor 219a and the second sensor 219b, and a third segment 21c connected between the second sensor 219b and suction fan 25.

[0060] In this preferred embodiment of the invention, segment 21a communicates with detection zone 13 through a diffuser 29. According to the invention, a single detection zone 13 could be equipped with a plurality of diffusers 29. For instance, diffusers 29 surrounding container CT passing in zone 13 could be provided, so that substantially the whole of the side surface of container CT passing in zone 13 is submitted to the effect of the air suction by diffusers 29.

[0061] Hereinafter, the operation principle of this embodiment of the detection method according to the invention will be explained in more detail.

[0062] Referring to FIG. 3A, there is shown the graph, against time, of the variation of the CO.sub.2 concentration, measured by means of an indicative signal generated by anyone of said CO.sub.2 sensors 219a, 219b. The graph in FIG. 3A relates to an operation cycle of apparatus 11 according to the invention, when no sample to be tested is present or when the sample is perfectly hermetic, i.e. wholly free from leakages.

[0063] In accordance with the preferred embodiment of the method according to invention, at time T.sub.0 detection zone 13 of an apparatus 11 made in accordance with the invention is substantially free from tracer gas. The atmosphere in the detection zone can be rich, for instance, in nitrogen or contain a gas mixture having a high nitrogen concentration.

[0064] At time T.sub.1 the suction step is started to suck air from detection zone 13, through the same diffusers 29. Air sucked from detection zone 13 by means of suction fan 25 flows along duct 21 and is intercepted first by sensor 219a and then by sensor 219b, both of which detect for instance 400 ppm, i.e. the typical atmospheric concentration of CO.sub.2.

[0065] At time T.sub.2 the suction step is stopped and nitrogen, for instance, is again introduced into detection zone 13 to flush the detection zone from tracer gas residues. Nitrogen sucked through duct 21 is intercepted by sensors 219a, 219b arranged along duct 21, which detect again 0 ppm CO.sub.2, since nitrogen is again the only gas licking said sensors 219a, 219b. At time T.sub.3 the cycle is stopped.

[0066] Referring to FIG. 3B, reference is now made to a sample container to be tested passing in detection zone 13 of apparatus 11 containing a tracer gas that is assumed to be CO.sub.2.

[0067] FIG. 3B shows the graph, against time, of the variation of the CO.sub.2 concentration measured by means of an indicative signal generated by CO.sub.2 sensors 219a or 219b. The operation cycle is substantially the same as in the preceding case, yet, at time T.sub.1, the sample container to be tested, which has a micro-hole from which CO.sub.2 leaks, is made to pass at constant speed in detection zone 13. In the interval between time T.sub.1 and time T.sub.2 sensor 219a or 219b detects a CO.sub.2 leak, as it can be appreciated from FIG. 3B. The CO.sub.2 concentration at sensor 219a or 219b progressively increases up to a maximum, and then decreases as the passing sample, and consequently the micro-hole, is moving away from detection zone 13. At time T.sub.2, when the container being tested has already passed through detection zone 13 and consequently the micro-leak has moved beyond diffusers 29 through which gases have been sucked, suction is stopped and the flushing step with introduction of pure nitrogen, i.e. a gas substantially containing 0 ppm of CO.sub.2, is started again. At time T.sub.3 the cycle is stopped.

[0068] The operation cycle of apparatus 11 described above with reference to FIGS. 3A and 3B can also be carried out by using compressed air (400 ppm, dashed line in the diagram of FIG. 3B) instead of pure nitrogen (solid line in the graph of FIG. 3B) as flushing gas, or by using other gas mixtures where the CO.sub.2 concentration is lower than that due to the micro-leak detected.

[0069] Referring to FIG. 3C, there is shown the graph, against time, of the variation of the CO.sub.2 concentration measured at sensor 219a or 219b in case in the case of two passing samples exhibiting gas leaks of different amounts, namely a small amount (dashed line) and a great amount (solid line). As it can be appreciated, the shape of the curve of the signal indicative of the variation of the concentration of tracer gas, CO.sub.2 in the example illustrated, is substantially always the same. As it will become even more apparent from the following description, experiments carried out have actually allowed determining that the graphical appearance of the signal indicative of the gas concentration in interval T.sub.1-T.sub.2 has a Gaussian-like behaviour. What is different obviously is the signal intensity, which depends on the size of the opening causing the leakage, on the tracer gas concentration in the gas mixture spilling from the container and on whether and how much the sample is mechanically stressed by the squeezing assembly, if any (the stronger the squeezing, the higher the leakage intensity detected by sensor 219a or 219b).

[0070] Referring to FIGS. 3D and 3E, there is shown the graph, against time, of the variation of the CO.sub.2 concentration in case of samples passing at high speed, when perturbations in the concentration of tracer gas, that is of ambient CO.sub.2 in the example illustrated, occur in zone 13 in interval T.sub.1-T.sub.2

[0071] Referring in particular to FIG. 3D, there is shown the graph, against time, of the variation of the CO.sub.2 concentration measured at sensor 219a or 219b in case of two passing samples exhibiting gas leaks of different amounts, namely a small amount (dashed line) and a great amount (solid line), when a very high and constant background value of tracer gas, CO.sub.2 in the specific case, with variable offset, is present in interval T.sub.1-T.sub.2.

[0072] Referring in particular to FIG. 3E, there is shown the graph, against time, of the variation of the CO.sub.2 concentration measured at sensor 219a or 219b in case of two passing samples exhibiting gas leaks of different amounts, namely a small amount (dashed line) and a great amount (solid line), when a very high and highly fluctuating background value of tracer gas, CO.sub.2 in the specific case, with strong turbulences and variable offsets, is present in interval T.sub.1-T.sub.2.

[0073] As it can be appreciated from FIG. 4, a detection method based on a fixed threshold for the tracer gas concentration has a number of limitations. First, being the threshold fixed, such a detection method is very sensitive to background gas offsets. Second, the instant at which the signal emitted by the sensor and indicative of the tracer gas concentration exceeds the fixed threshold and consequently causes signalling the occurrence of a leak, varies depending on the tracer gas concentration, that is depending on the leak amount. Always with reference to FIG. 4, where signals indicative of a leak of small amount (dashed line) and great amount (solid line) are shown and the threshold is identified by horizontal solid line Th, the instant at which the occurrence of a leak is signalled actually has a time shift T.fwdarw.T′ varying as the tracer gas concentration varies.

[0074] Such an approach in which a threshold fixed relative to the signal generated by a gas sensor is set is moreover scarcely performant in case of micro-leaks of very small amounts, and moreover gives rise to the problem of false positives, i.e. false signallings of leak occurrence. More specifically, referring to FIG. 5A, an example is shown in which a small variation in the tracer gas concentration at the gas sensor, due to a micro-opening in the container, would not be sufficient to allow recognising that the container is not correctly sealed and hence is possibly to be discarded. FIG. 5B shows an example in which a fluctuation in the concentration of a gas of the same kind as the tracer gas introduced in the container, due to causes external to the container, has been misinterpreted as a leak since it is sufficient to generate, at the gas sensor, a signal whose value exceeds the fixed threshold set. FIG. 5C shows an example similar to the previous one, in which background gas turbulences, due to causes external to the container, have been misinterpreted as a leak.

[0075] Too low a fixed threshold would therefore make practically impossible distinguish the transitions due to micro-leaks from the ones due to background noise, which are the majority. The presence of the background noise compels therefore to set the threshold to a value significantly different from zero and, anyway, with an absolute value higher than the noise “peaks”. In the specific case this means therefore that a leak would be recognized only if its amount is much greater than the background fluctuations.

[0076] The measurement method according to the alternative embodiment of the invention, capable of considerably increasing the sensitivity of the detection itself, exploits a principle allowing precisely establishing the instant, i.e. the timing, at which a leak has occurred. Establishing a precise and repeatable timing at which a leak occurrence is signalled allows considerably narrowing the interval of analysis of the measurement on the moving sample near the passage of the sample container affected by the leak to be detected. The precise timing selection makes the measurement method less sensitive to ambient turbulences that can originate signals that are very similar to the signal characteristic of a leak and could therefore misinterpreted as leak-indicative signals.

[0077] As stated before with reference to FIG. 4, by assuming a fixed threshold Th exceeding of which triggers signalling the presence of tracer gas, as the amplitude of the signal indicative of the presence of gas originated by a leak varies, also the delays of instants T, T′ at which the leak is signalled due to the threshold being exceeded vary. More particularly, said delays increase as the signal amplitude decreases. Assuming that the signal generated by sensor 219a or 219b is sent to a comparator device arranged to generate a logical signal “0” when the intensity of the input signal of the comparator is below the threshold value set, and a logical signal “1” when the intensity of the input signal of the comparator exceeds the threshold value set, due to the delays pointed out above the time intervals between the transitions from logical state “0” to “1” do not correspond to the correct time intervals at which the variation of the tracer gas concentration at the sensor has occurred. This effect of the timing dependence on the signal amplitude is referred to as “walk” effect in the scientific literature and, as pointed out above, a timing technique based on a fixed threshold is affected by a significant “walk” effect.

[0078] Moreover, the signals generated by the gas sensor are generally affected by a significant background noise, which causes an equally significant “jitter” effect, i.e. a fluctuation, in the timing.

[0079] The substantial similarity in the shapes of the curves of the signal indicative of the tracer gas concentration in the gas mixture arriving at the sensor, notwithstanding the variation in the signal amplitude, has advantageously enabled adoption of a substantially walk-free timing technique, consisting in making the transition of the timing logical signal occur when the signal exceeds a threshold that ideally, for each signal, adapts itself to a defined fraction of the maximum of the curve, for instance when the signals attain half their final amplitude.

[0080] Providing such a “floating” threshold is comparable to a so called “Constant Fraction Timing” or “Constant Fraction Discrimination” (CFD).

[0081] Reference will now be made again to FIG. 2 for describing a preferred embodiment of an apparatus 11 made in accordance with a particular embodiment of the invention, arranged to implement the detection method capable of considerably increasing the sensitivity of the detection itself.

[0082] As disclosed hereinbefore, apparatus 11 includes a pair of sensors 219a and 219b connected to each other by duct segment 21b the internal volume of which is known: i.e. the length and the cross-sectional size of said duct segment 21b are known and constant. Such a duct segment 21b separating sensors 219a and 219b substantially forms a corresponding delay line in gas propagation along duct 21.

[0083] Referring also to FIG. 6, corresponding signals M.sub.1 and M.sub.2 coming from the two sensors 219a and 219b are sent to a comparator 210 and output signal M.sub.3 of comparator 210 will indicate the occurrence of a leak from a passing container when the signal of the second sensor 219b exceeds the floating threshold determined by the variable signal of the first sensor 219a at the same time instant.

[0084] This technique advantageously allows having a discrimination time instant independent of the amplitude and less sensitive to jitter and walk.

[0085] CFD discrimination moreover makes the system more performant in case of low intensity leak signals and increases measurement sensitivity. Furthermore, the detection method is less affected by background variations, or turbulence effects, of external CO.sub.2. This detection technique moreover allows preventing false positives, i.e. preventing external fluctuations from being misinterpreted as leak measurements.

[0086] In the example shown in FIG. 7, two switches, i.e. two transitions, 0.fwdarw.1 of the comparator occur in the proper measurement interval. Yet such switches occur at time instants different from the instant at which reading is made. If the switches occur at too close instants, they are considered by the system as being due to background noise and not to events determined by a leak of CO.sub.2.

[0087] In an alternative embodiment of the apparatus made in accordance with this particular embodiment of the invention, the signal of the second sensor is replaced by a second signal of a first sensor in which the gas flow is made to pass a second time in opposite direction. In other words, according to such an alternative embodiment, the gas flow coming from detection zone 13 passes through the first sensor 219a by flowing along the duct in a first direction towards suction fan 25, and then in the opposite direction towards sensor 219a. Clearly in this embodiment a single and unique sensor could even be provided.

[0088] In this embodiment, the arrangement is similar to that shown in FIG. 2, yet without the sensor denoted 219b. Delay lines 21a and 21b coincide, and this single line can have a length at will. In this manner, the same line 21a, 21b becomes a gas accumulation chamber. When the flow is reversed, the Gaussian loss peak having passed through the first and single sensor 219a from the left to the right passes again through that same sensor in the opposite direction, from the right to the left. Sensor 219a has acquired two signals with a known delay and, at this time, it is possible to process the signals by reproducing constant fraction discrimination. Since two Gaussian peaks have been acquired, one due to the passage in one direction and one due to the passage in the other direction, it is possible to digitally process them by reproducing the constant fraction discrimination.

INDUSTRIAL APPLICABILITY

[0089] The invention finds industrial application in several fields, for detecting leaks and micro-leaks from containers of substantially any kind, either compressible or rigid. The invention can also be applied for detecting leakages of liquids, for instance water or beverages, from pressurised rigid containers.

[0090] The invention as described and illustrated can undergo several variants and modifications falling within the same inventive principle.