METHOD AND SYSTEM FOR LEAK DETECTION

Description

[0001] The present invention relates to the field of leak detection methods for checking the watertightness of an object, and more particularly to leak detection methods using pressure variation. The present invention also relates to leak detection devices configured to implement such methods.

[0002] Various systems and methods are known for detecting leaks, for example tracer gases, the soap bubble method, etc. In the present case, the invention relates more specifically to the detection of leaks by pressure variation.

[0003] Thus, when a leak is detected by pressure variation, the object being tested, whose level of sealing is to be checked, undergoes a controlled pressure variation. This means, in particular, that a known pressure variation is applied to a given volume inside the object (known as the direct method) or to a closed volume surrounding the object (known as the indirect method).

[0004] Then, after a determined time, the pressure variation is measured in a characteristic volume during a predefined time interval, in order to determine if the object has a leak, and also sometimes to be able to quantify the level of the leak of the tested object.

[0005] Indeed, a leak results in a pressure variation over a given period of time; more specifically, the pressure variation per unit of time can, for example, be related to a leak level by the following mathematical relationship: F = k Δ P Δ t V where F is the leakage, generally expressed in standard cubic centimeters per minute (scm³ / min), ΔP is the pressure change in Pascals (Pa) measured in a relevant volume, Δt is the time interval (in seconds) of the measured pressure change ΔP, V is the relevant volume to be considered (e.g., the internal volume of the object), generally expressed in cubic centimeters (cm³), and k is a multiplicative constant (in Pa⁻¹). Note that the leakage can also be expressed in other ways, for example, as a mass flow rate. The formula that relates the leakage to a physical quantity can take different forms depending on the measurement method used and the physical quantity being studied.

[0006] Thus, regardless of the object or sub-element of the object, it is possible to check if it has leaks and to determine its level of sealing.

[0007] The object to be tested could be an electronic device, packaging, a container, etc.

[0008] Tolerances for the level of sealing (or leakage) can therefore vary considerably in other ways, for example, as a mass flow rate. The formula that relates leakage to a physical quantity can take different forms depending on the measurement method used and the physical quantity being studied.

[0009] Thus, regardless of the object or sub-component, it is possible to check for leaks and determine its level of sealing. The object to be tested could be an electronic device, packaging, a container, etc. Tolerances for the level of sealing (or leakage) can therefore vary considerably depending on the object being tested, its volume, shape, and / or function.

[0010] However, during a leak test of an object, environmental parameters and / or various temporary parasitic phenomena can make it difficult to measure the pressure variation over time and / or the repeatability of such measurements.

[0011] This problem is even more pronounced when leak detection takes place in an industrial environment, such as a factory, where temperature and pressure can vary locally and / or over time depending on the operations performed on the objects being tested or near the leak detection device. Furthermore, the detection device may be used to test the leak tightness of objects on a production or manufacturing line. In this case, it is essential that the leak test be as fast, accurate, and repeatable as possible so as not to disrupt the production or manufacturing line.

[0012] Other solutions such as those described in patent documents published under numbers US 2017 / 254719, EP 2 801 716, US 6 182 501, WO 2010 / 134622, DE 10 2015 226360 or in the publication by MOLLER under ISBN number: 978-1-933742-89-2, are also unsatisfactory in addressing at least one of the problems mentioned above.

[0013] The present invention aims to remedy at least one of the problems mentioned above by proposing a new method for detecting leaks on an object by varying the pressure using a leak detection device according to claim 1, said method comprising the following steps: Connect the object to be tested to the leak detection device; vary the pressure in at least one part of the object being tested or in an enclosure surrounding the object being tested; measure, in a test step, a physical quantity related to the leak level F in said at least one part of the object being tested or in an enclosure surrounding the object being tested, this measurement corresponding to a measurement signal M(t); apply a modeling function A(t) containing at least one relevant component, said exponential component. More specifically, the approximation function can be in the form of a series of the type ∑ n = 0 + ∞ 1 n ! − t τ n , where n is a natural number. The expansion of this type of series is done to order 2, 3 or beyond depending on the acceptable degree of error (on the approximation made).

[0014] Depending on a possible characteristic, the values ​​of the components of the modeling function A(t) are calculated.

[0015] The calculation of the values ​​of the components of the modeling function can in particular be carried out during the test stage from measurement signals M(t).

[0016] According to another possible characteristic, the modeling function A(t) is a continuous and differentiable function, at least piecewise. Advantageously, the function is differentiable n times, where n is a positive integer between 1 and 3.

[0017] According to another possible characteristic, the modeling function A(t) includes at least one component of the type Ce -t / < τ< , where C is a real constant.

[0018] According to another possible feature not covered by the claims, when the measurement signal M(t) has the dimensions of pressure, the modeling function A(t) is of the form: A t = P i + Ce − t τ − 1 − F t where Pi is the initial pressure value, C is a real constant, F is the leakage level of the object being tested.

[0019] According to another possible characteristic, when the measurement signal M(t) is homogeneous to a pressure, the leakage level F of the tested object is equal to C e − t 1 τ − 1 + P i − M t 1 t 1 , Or t 1 is any time during the testing stage.

[0020] According to another possible feature not covered by the claims, when the measurement signal M(t) is homogeneous to a pressure, the components of the modeling function A(t) are determined by the time variation of the measurement signal M(t) at successive times. t 0 + t 1 2 and t 1 and by calculating a ratio R(t 1 ) of said variations, where t 0 is an initial time during the test step and t 1 is a time subsequent to t 0 , and R t 1 = M ′ t 1 − M ′ t 0 M ′ t 0 + t 1 2 − M ′ t 0 .

[0021] Advantageously, the gap between times t 0 and t 1 is greater than 500 ms (for milliseconds), and preferably greater than 1000 ms.

[0022] According to another possible characteristic, the leakage level F of the tested object is equal to M ′ t − M " t 2 M ‴ t .

[0023] According to the invention as claimed, when the measurement signal M(t) is dimensionally equivalent to a flow rate or pressure per unit time, the modeling function A(t) is of the form: A t = C e − t τ + F where C is a real constant and F is the leakage level of the object being tested.

[0024] According to the invention as claimed, when the measurement signal M(t) is homogeneous with a flow rate or pressure per unit time, the leakage level F of the tested object is equal to M t 1 C e t 1 τ

[0025] According to another possible characteristic, when the measurement signal is homogeneous with a flow rate or pressure per unit time, the components of the modeling function A(t) are determined via the measurement signal M(t) at successive times. t 0 + t 1 2 and t 1 and by calculating a ratio R'(t 1 ), where t 0 is an initial time during the test step and t 1 is a time after t 0 , and R ′ t 1 = M t 1 − M t 0 M t 0 + t 1 2 − M t 0 .

[0026] Advantageously, the time difference between times t0 and t1 is greater than 500 ms (ms for milliseconds), and preferably greater than 1000 ms. The time difference is generally a function of the volume of the object being tested, the test pressure, and / or the level of leakage being sought. More specifically, the larger the volume, the higher the test pressure, and / or the smaller the level of leakage being sought, the greater the time difference.

[0027] According to another possible characteristic, the leakage level F of the tested object is equal to M t − M ′ t 2 M " t .

[0028] According to another possible characteristic, the determination of the leakage level F is a function of the ratio of variations with respect to time, such as derivatives, of the measurement signal M(t).

[0029] According to another possible characteristic, the determination of a leakage level F is a function of the measurement signal M(t), of the first, second and / or third derivatives, respectively denoted M'(t), M"(t) and M'"(t), of the measurement signal M(t).

[0030] According to another possible characteristic, the value of the leakage level F is the average of several values ​​of the measurement signal M(t) and / or its variations with respect to time (such as derivatives or others).

[0031] Thus, in order to make the determination of the leakage level F more robust, the modeling function A(t) is applied to several time intervals, consecutive or not, of the measurement signal M(t).

[0032] According to another possible feature, the values ​​of the components τ and C of the modeling function A(t) are calculated beforehand or are predetermined, then stored for use when determining the leakage level.

[0033] In cases where the tested object is always the same, it can be advantageous to predetermine average values ​​for certain components of the modeling function A(t) to improve the speed of leak testing. It should also be noted that it is advantageous to predetermine the values ​​of the τ and C components using a database of past measurements and machine learning, optimization, and / or regression methods on such databases.

[0034] According to another possible characteristic, said modeling function A(t) is determined for a predetermined period of time Δt, advantageously for a period of time Δt between 500 and 4000 milliseconds (ms).

[0035] The acquisition of the measurement signal M(t) (and therefore the detection of leaks) can be carried out over durations ranging from 100 milliseconds to 650 seconds, it is thus possible to determine several successive or sliding fitting functions A(t), in order to average and / or to be able to check several times the components of the modeling function A(t), as well as the leak level F.

[0036] The invention also relates to a leak detection device according to claim 11 configured to implement the leak detection method according to the invention as defined above.

[0037] According to another possible characteristic, the leak detection device is a pressure variation leak detection device.

[0038] The invention will be better understood, and other objects, details, features and advantages thereof will become more apparent from the following description of particular embodiments of the invention, given solely by way of illustration and not limitation, with reference to the accompanying drawings, in which: there [ Fig.1 ], referenced [ Fig.1 [ ], is a very schematic representation of a leak detection device according to the invention; the [ Fig.2 ], referenced [ Fig.2 ], is a graph illustrating an example of pressure variation during a leak detection procedure performed by the device of the [ Fig.1 ] ; there [ Fig.3 ], referenced [ Fig.3 ], is a flowchart illustrating the leak detection method according to the invention.

[0039] There [ Fig.1 ] is thus a very schematic representation of a leak detection device 1 to check the tightness of an object 10. The object tested can be any object whose tightness needs to be checked, packaging, heat exchanger, mobile phone, etc.

[0040] The said device 1 comprises the following: a system 5 for pressurizing or evacuating a characteristic volume relative to the object 10 being tested, i.e., this can be an internal volume of the object (direct method) or a closed volume surrounding said object (indirect method); a first pressure sensor 7 configured to measure the pressure variations of said characteristic volume, said sensor 7 allowing the object 10 to be checked for leaks; a second pressure sensor 17, which is an optional sensor, configured to measure the pressure applied in said characteristic volume by said device 5; air connections 11, such as ducts or pipes, configured to connect the pressurizing device 5 to the object 10 and to a reference 13; an electronic entity 15, such as an electronic circuit, connected to the various pressure sensors 7 and 17 and configured to retrieve the pressure values ​​measured by said sensors 7 and 17.

[0041] Advantageously, the pressurization or vacuum system 5 includes a pressure source 51 which can be, for example, a pump, a compressor, a compressed air supply or a pressurized gas supply.

[0042] It should be noted that in an embodiment not shown, said device 1 includes a third pressure sensor configured to measure variations in environmental pressure, generally atmospheric pressure.

[0043] The first pressure sensor 7 is preferably a differential pressure sensor. The second 17 and / or third sensors are advantageously absolute pressure sensors. In an alternative embodiment not shown, the first pressure sensor 7 is an absolute pressure sensor, and the leak detection device does not include a reference 13.

[0044] Leak detection performed by device 1 of the [ Fig.1 ] can be divided into four main phases, most particularly illustrated in the [ Fig.2 ] : a filling phase I of the characteristic volume of the object tested with compressed air (or any gas, preferably inert, such as nitrogen), the pressure increasing up to a desired pressure value P1; a stabilization phase II, after pressurizing the characteristic volume of the object tested, it is necessary for it to return to thermal and mechanical equilibrium generally during a predefined period, also called the stabilization time, in order to limit the disturbances in the leak measurement by transient phenomena; a test phase III, during which the pressure variation is measured over a time t test, called the test time, in the characteristic volume to determine a leak level of the object tested; an emptying phase IV, during which the pressurized characteristic volume of the object tested is returned to atmospheric pressure.

[0045] It should be noted that leak detection can also be performed under vacuum (or negative pressure). This means that instead of increasing the pressure during the first phase (Phase I), the pressure within the characteristic volume of the object being tested is lowered to a predetermined value. Phases II and III remain unchanged. The fourth phase (or Phase IV) involves increasing the pressure within the object being tested to a value corresponding to atmospheric pressure. Therefore, the filling and emptying phases (Phase I and IV) can be considered to be reversed between pressure and vacuum leak detection procedures.

[0046] The invention, which is a method for detecting leaks by pressure variation using device 1, therefore advantageously fits into one or more phases of a leak detection process illustrated in [ Fig.2 ], particularly at the stabilization phase II and / or test phase III, one of the goals of the invention being to reduce the stabilization time and to be able to detect leaks as quickly as possible (by reducing the test time).

[0047] The said method according to the invention, more particularly illustrated in the [ Fig.3 ], thus includes the following steps: Connect the object 10 to be tested (S1) to the leak detection device (1); vary the pressure S2 in at least a part of the object being tested or in an enclosure surrounding the object 10; measure a physical quantity S3 related to the leak level F in said at least a part of the object 10 being tested or in an enclosure surrounding the object being tested during a test step, this corresponding to a measurement signal M(t); apply a modeling function A(t) (S4) containing at least one exponential component of the form e -t / τ< ,or at least a function approximating said at least one exponential component of the form e -t / τ< , to the measurement signal M(t), where t is the time and τ a characteristic time constant; calculate S5 the values ​​of the components of the modeling function A(t); determine S6 the leakage level F in the tested object as a function of the components of the modeling function A(t) and at least one variation with respect to time of the measurement signal M(t).

[0048] It should be noted that the measurement signal M(t) can be a quantity homogeneous to a pressure, a flow rate or a pressure per unit time (in particular the variation of pressure per unit time).

[0049] According to alternative embodiments not shown, the measurement signal M(t) is obtained via a suitable leak detection device, i.e. configured to measure these quantities and including for example a pressure sensor, a flow meter, etc.

[0050] Furthermore, it should be noted that "variation with respect to time" means a derivative, the string, or any differentiating filter.

[0051] Furthermore, in one embodiment, the exponential component of the form e -t / τ< can also be an approximation function that approximates, over a relevant time interval, said exponential component.

[0052] More specifically, the approximation function can take the form of a series of the type ∑ n = 0 + ∞ 1 n ! − t τ n ,where n is a natural number. The expansion of this type of series is done to order 2, 3 or beyond depending on the acceptable degree of error (on the approximation made).

[0053] The various steps, phases, calculations and / or determinations described above are carried out using the electronic entity 15, entity 15 managing in particular the opening and closing of the various valves 57 of said device 1 accordingly.

[0054] The modeling function A(t) as defined includes at least one component of the type Ce -t / τ< , where C is a real constant. The constant C can be a negative or positive real number depending on the modeling and / or the transient phenomena to be taken into account during the testing step.

[0055] In a first embodiment not covered by the claims, when the measurement signal M(t) has the dimensions of pressure, the modeling function A(t) advantageously takes the form: A ( t) = P i + C ( e -t / τ< - 1) - F where P i is the value of the initial pressure, F the leakage level of the object being tested.

[0056] Whereas in a second embodiment, when the measurement signal M(t) is homogeneous to a flow rate or pressure per unit time, the modeling function A(t) advantageously takes the form: A t = C e − t τ + F where C is a real constant, F the leakage level of the object being tested.

[0057] In the different embodiments, the modeling function A(t) is for example applied continuously or piecewise to the measurement signal M(t) measured by the leak detection device 1, the electronic entity 15 being configured to determine the value of the different components of the modeling function A(t) and to carry out all the operations necessary for this.

[0058] More specifically, when the measurement signal is homogeneous to a pressure (or first embodiment), the modeling function A(t) and the calculation of its components thus makes it possible to determine the presence of a leak F, the leak level F being for example a function of the ratio of variations with respect to time (such as derivatives) of the measurement signal M(t).

[0059] For example, the components of the modeling function A(t) are determined by variations with respect to time (or differentiation) of the measurement signal M(t) at successive times. t 0 + t 1 2 and t 1 and by calculating a ratio R(t 1 ) of said variations, where t 0 R t 1 = M ′ t 1 − M ′ t 0 M ′ t 0 + t 1 2 − M ′ t 0 .

[0060] This allows us to determine the leakage level F of the tested object 10, which, after simplification, is equal to C e − t 1 τ − 1 + P i − M t 1 t 1 (Or t 1 is any time during the testing step).

[0061] According to a variant of the first embodiment, the determination of the presence of a leak, and of the associated leak level F, is a function of variations with respect to time or of first, second, and third derivatives, respectively denoted M'(t), M"(t), and M"'(t), of the measurement signal M(t). The leak level F of the tested object (10) is then equal to M ′ t − M " t 2 M ‴ t .

[0062] In the second embodiment, that is, when the measurement signal is homogeneous with a flow rate or pressure per unit time, the modeling function A(t) and the calculation of its components also make it possible to determine the presence of a leak F, the leak level F being, for example, a function of the ratio R' (t) of the measurement signal M(t) at different times.

[0063] For example, the components of the modeling function A(t) are determined via the measurement signal M(t) at successive times. t 0 + t 1 2 and t 1 and by calculating a ratio R'(t 1 ), where t 0 is an initial time during the test step and t 1 is a time after t 0 , and R ′ t 1 = M t 1 − M t 0 M t 0 + t 1 2 − M t 0 .

[0064] This allows us to determine the leakage level F of the tested object 10, which, after simplification, is equal to M t 1 C e t 1 τ (Or t 1 is any time during the testing step).

[0065] According to a variant of the second embodiment, the determination of the presence of a leak, and of the associated leak level F, is a function of the variations with respect to time, or first and second derivatives, respectively denoted M'(t) and M"(t), of the measurement signal M(t). The leak level F of the tested object (10) is then equal to M t − M ′ t 2 M " t .

[0066] Regardless of the embodiment, the leakage level F is advantageously determined from several time intervals of the measurement signal M(t) and its derivatives. That is to say, several successive or sliding fitting functions A(t) are determined in order to average and / or determine the components of the modeling function A(t) multiple times, as well as the resulting leakage level F.

[0067] The said modeling function A(t) is determined for a predetermined time period Δt, such as a time period between 500 and 4000 milliseconds (ms).

[0068] It should also be noted that the parameter values τ and A are advantageously calculated beforehand and stored, for example in a memory of the electronic unit 15, for use in determining the leakage value. This predetermination of the values ​​of the components τ and C can be achieved using a database of past measurements and / or learning, optimization and / or regression methods on such databases.

Claims

1. Method (100) for detecting leaks on an object (10) by pressure variation via a leak detection device (1), said method (100) including the following steps: - connecting (S1) the object (10) to be tested to the leak detection device (1); - varying the pressure (S2) in at least a part of the tested object (10) or in a chamber surrounding the object (10) to be tested; - measuring, in a test step, a physical quantity (S3) related to the level of leakage F in said at least a part of the object (10) tested or in a chamber surrounding the object (10) to be tested, this measurement corresponding to a measurement signal M(t); - applying a modelling function A(t) (S4) containing at least one exponential component having the form e-t / τ, or at least one function approximating said at least one component, to the measurement signal M(t), where t is the time and τ a characteristic time constant; - determining the leakage level F (S6) in the tested object as a function of the components of the modelling function A(t) and of at least one variation of the measurement signal M(t) over time, characterised in that, when the measurement signal M(t) is homogeneous at a flow rate or at a pressure per unit of time, the modelling function A(t) has the form: A t = C e − t τ + F where C is a real constant and F is the leakage level of the tested object, and the leakage level F of the tested object (10) is equal to M t 1 C e t 1 τ , where t1 is a time after t0.

2. Method (100) according to the preceding claim, characterised in that there is calculation of the values (S5) of the components of the modelling function A(t).

3. Method (100) according to any one of the preceding claims, characterised in that the modelling function A(t) comprises at least one component of the type C e-t / τ, where C is a real constant.

4. Method (100) according to claim 3, characterised in that the values of the components τ and C of the modelling function A(t) are calculated or predetermined in advance, then stored for use in determining the leakage level F.

5. Method (100) according to claim 4, characterised in that the level of leakage F of the tested object (10) is equal to M ′ t − M " t 2 M ‴ t .

6. Method (100) according to the preceding claim, characterised in that, when the measurement signal is homogeneous at a flow rate or at a pressure per unit of time, there is a determination of the components of the modelling function A(t) via the measurement signal M(t) at successive times t 0 + t 1 2 and t1 and by calculating a ratio R'(t1), where t0 is an initial time during the test step and t1 is a time after t0, and R ′ t 1 = M t 1 − M t 0 M t 0 + t 1 2 − M t 0 .

7. Method (100) according to claim 5, characterised in that the level of leakage F of the tested object (10) is equal to M t − M ′ t 2 M " t .

8. Method (100) according to any one of the preceding claims, characterised in that the determination of the leakage level F is a function of the ratio of variations over time, such as derivatives, of the measurement signal M(t).

9. Method (100) according to any one of the preceding claims, characterised in that the value of the leakage level F is the average of several values of the measurement signal M(t) and / or of its variations over time.

10. Method (100) according to any one of the preceding claims, characterised in that said modelling function A(t) is determined for a predetermined period of time Δt, advantageously for a period of time Δt between 500 and 4000 milliseconds.

11. Leak detection device configured to implement the leak detection method according to any one of claims 1 to 10.