JP2010117376A - Method and apparatus for leak testing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method and apparatus for leak testing applicable to containers having various sizes and to various stored materials, when a stored component is liquid. <P>SOLUTION: In a method for manufacturing a closed storing container, which is a method for filling the container with a material having a liquid component, forming a liquid-coated area in the container, closing the container hermetically, and leak testing the container to determine presence/absence of leakage, for leak testing, the surrounding space of the container is evacuated by using a vacuum pump to apply a pressure difference between the inside and the outside of the container, the pressure in the surrounding space is lowered to or below a vapor pressure value of the liquid component while observing the pressure in the surrounding space, and for observing a pressure value of the surrounding space as a leakage determination signal, if the pressure value of the surrounding space reaches the vapor pressure value, the presence of leakage from container is determined. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention generally relates to a technique for performing a leak test of a closed container using at least one liquid component as a storage material.

  Leak test techniques are known in which a closed vessel is introduced into a test cavity and the cavity is depressurized by a suction pump after the cavity is sealed. When the inside of the test cavity reaches a predetermined pressure, that is, when the periphery of the container under test reaches the predetermined pressure, this pressure is maintained substantially constant if there is no leak in the container. If there is a leak in the area of the container in which air is constrained, the air flow from the container will increase the ambient pressure. If there is a leak in the area of the container where the containment is constrained, whether this leak significantly increases the ambient pressure depends on the type of containment, the viscosity of the containment, and the solids in the containment. It depends heavily on whether particles are present and, as is obvious, it depends very much on the size of the leak.

  Other methods for accurately detecting a leak in a product container are known, regardless of whether the leak is in an air restraint area or an area in contact with the contents. One such method is the subject of co-pending European Patent Application No. 0 791 814 and US Patent Application No. 08/862 993, in which impedance measurements are made by electrode configuration near the outer wall of the container. In detail to do, we propose to do resistance measurement. As soon as liquid leaks from the leak location, the leaked liquid contacts each pair of impedance measuring electrodes and causes a significant change in the impedance measured between the electrodes.

European Patent Application No. 0 791 814 US patent application Ser. No. 08 / 862,993 JP-A-4-279850 JP-A-8-334429 Japanese Patent Laid-Open No. 9-61285 No. 58-57940

  Nonetheless, the above method requires significant additional costs in terms of installing an impedance measurement configuration within each test cavity, particularly for multi-cavity type in-line inspection devices. In addition, such a method cannot detect a minute leak that is much smaller than 1 micron, and is limited by the shape of the container and the type of material contained.

  The main object of the present invention is such that it can be applied to very different sized containers and to very different types of containment materials, when at least one of the containment components is a liquid. Another object is to provide a leak test method and apparatus.

  It is another object of the present invention to provide a leak test method and apparatus that is relatively inexpensive with respect to electronics and equipment, and therefore can perform very economical tests.

  Still another object of the present invention is to provide a leak test method and apparatus that have a short measurement cycle and can perform measurement with very high accuracy even in such a short measurement cycle.

  The object is a method for performing a leak test of at least one closed containment container, wherein the container contains at least one liquid component and by applying a pressure difference towards the surrounding space of the container. In the case where a pressure difference is applied across at least a part of the wall of the container to be tested and the pressure value in the surrounding space is observed as a leak determination signal, the pressure in the surrounding space is This is achieved by a method of establishing a pressure difference by lowering it below the vapor pressure value of at least one liquid component.

  In the present invention, if there is a leak in the container, the liquid is sucked out to the external decompression surrounding space, and if the surrounding space has a constant volume, the sucked liquid reaches the vapor pressure at the ambient pressure. It was developed based on the idea that it would evaporate at that time. Such evaporation of the liquid will result in a significant pressure rise compared to the ambient pressure that would be established under the same measurement conditions when there is no leak in the container.

  The pressure observation in the test cavity into which the container is introduced when the vapor pressure of the liquid that can leak is reached is shown to be a very accurate technique for the leak test. It should be noted that according to such a technique, it is possible to accurately detect a leak of a container having a contained product over a very wide range, and it is possible to accurately detect a leak as small as 0.02 μm at the present time. .

  It should also be noted that the volume of the test cavity is not critical, so that the technique according to the present invention makes it possible to test batches of multiple containers simultaneously. In this case, even a leak of one container forming such a batch can be detected with high accuracy.

  When the ambient pressure of the leaking container drops to the internal pressure of the container, some amount of liquid is sucked out of the container, and as soon as the ambient pressure reaches the vapor pressure, the sucked liquid is Start to evaporate. If the circumference of the container is a constant volume, the evaporation of the liquid will cause an increase in pressure, and the pump that is reducing the ambient pressure must in this case also remove the liquid vapor. Significant measurements can be made especially after the surrounding space of the container has become a pressure below the vapor pressure. Nonetheless, a pumping capacity that can evacuate the space around the container under test to a pressure well below the vapor pressure, in particular to a pressure at least 2 orders of magnitude, preferably to a pressure of at least 3 orders of magnitude. It is preferable that

  A pressure change that can recognize a leak becomes detectable as soon as one of the plurality of liquid components begins to evaporate if the container contains a plurality of liquid components, As the vapor pressure value, a method of selecting a larger vapor pressure value among the vapor pressures of several liquid components and reducing the ambient pressure of the container to the vapor pressure value or less is recommended.

  As is well known, since the vapor pressure is a function of temperature, in some cases it may be advantageous to heat, for example, the surroundings of the container to a predetermined temperature in order to stabilize the vapor pressure of the predetermined liquid, Is carried out at room temperature, i.e. the vapor pressure to be reached is considered at room temperature around 20 ° C., the method and apparatus according to the invention is simplified.

  Also, very accurate leak detection is made possible by measuring the ambient pressure of the container at two consecutive points in time. In this case, the term “point” should be understood to include the predetermined time span required to accurately measure the pressure. Although leak detection can be performed by applying the vacuum action of the vacuum pump to the surrounding space of the vessel and then measuring the absolute pressure value obtained after a predetermined time, the ambient at two specific time points If it is a system which measures pressure, the 1st measured value can be used as a reference value, and the difference between the 2nd measured value and this reference value can be formed. In this case, differential pressure measurement is performed instead of absolute pressure measurement. More specifically, the first pressure signal measured at the first point in time is stored as an electrical signal, and then the first value (which is still stored at this point in time) when the second pressure value is measured. And the second value is determined.

  PCT application WO 94/05991 and US Pat. No. 5,239,859, corresponding to the same applicant as the present invention, disclose a differential pressure measuring method and apparatus in which offset is compensated very accurately. Has been. In the preferred mode of operation of the method according to the invention and in the preferred mode of operation in the implementation of the device according to the invention, such a differential pressure measurement technique is used. Thus, PCT application No. WO 94/05991 and US Pat. No. 5,239,859 are hereby incorporated by reference in their entirety. However, important features will be described in detail within this specification.

  Due to the fact that the relative size of the volume of the test cavity with respect to the volume of the container under test is not at all important, the method and apparatus according to the invention have the following important advantages:

  If the wall of the container under test is capable of withstanding the pressure difference between the internal pressure of the container (usually atmospheric pressure) and the surrounding decompression space, such a container is relatively large and small between the container and the test cavity. Regardless of the relationship, it can simply be introduced into the test cavity forming the surrounding space. Even in that case, a highly accurate leak determination can be performed based on the present invention. Thus, a single and identical test cavity can be used for containers of various sizes and volumes. This is a single container that allows a batch of containers to be introduced into a single test cavity in the surrounding space and occupies only a small percentage of the total cavity volume. However, even if only one container among a plurality of containers forming a batch has a leak, the leak determination can be accurately performed.

  Other important advantages of the present invention are as follows.

  It is common for the containment vessel not to be completely filled, in which case a small amount of air is constrained in the closed vessel. In such containers, if there is a leak in the vicinity of the restrained air or gas, such air is sucked out of the leak location by reducing the ambient pressure. As the pressure of the restraining air in the container gradually decreases, the liquid component in the container begins to evaporate. Such vapor is also sucked out of the leak location. Both the air drawn from the leak location and the vapor drawn from the leak location increase the ambient pressure. Therefore, a leak in the air restraint region of the container causes a change in the ambient pressure. That is, the ambient pressure is changed as if the leak location is in the liquid content coating region of the container wall. Therefore, by appropriately setting the leak detection threshold according to the minimum allowable pressure change in the surrounding space, such a leak point is located in the air covering region or the content covering region. It ’s no longer important.

  If one identical leak located in the air confinement area of the container causes only a smaller pressure change to the surrounding space than the same leak when located in the liquid-covered area, the container will leak. The setting of the threshold value for detecting whether or not it has is governed by this pressure change. Or, conversely, if one identical leak located in the liquid-covered region of the container causes only a small pressure change in the surrounding space than the same leak when located in the air contact region. The smaller pressure change dominates the threshold setting for detecting whether the container has a leak.

  If the container under test has a large leak, the contents of the container will not contaminate the inner surface of the test cavity, in general, it will not contaminate the surroundings of the container, and more specifically, the pump Ambient pressure reduction should be stopped as soon as such a large leak is detected so as not to contaminate the equipment. This is preferably by measuring the resistance by direct current by observing whether the ambient pressure is lowered by the pump action or by measuring the impedance around the container near the wall of the container under test. This can be realized by detecting the scattering of the contents. Impedance measurement can be realized by installing an electrode configuration in the vicinity of the periphery of the container under test and over the entire circumference of at least a part of the container. As soon as the contents of the container are sucked out onto the outer wall, the electrode configuration is short-circuited by such contents and a sudden impedance change will occur. By detecting this impedance change, further lowering of the pressure around the container is stopped.

  The latter technique for quickly detecting large leaks applies particularly to containers that need to be conformally enclosed within a test cavity due to the container wall being unable to withstand pressure differentials. The In such a case, an electrode configuration for impedance measurement can be incorporated along the inner wall of the test cavity that contains at least one container in a snug fit. When testing such containers, i.e. when using conformable cavities, even in this case, to form a space around the container between the outer wall of the container and the wall of the test cavity. A continuous volume of. Such a continuous volume is ensured by a plurality of micro-projections on the wall of the test cavity formed by fitting a support grid or mesh, or preferably by roughening the inner surface of the test cavity. This can be done by preventing outward bulging of the container wall. In this case, the communication space between such minute projections forms a surrounding space for the container.

  When it is determined that a container in the test cavity that forms the surrounding space has a leak, the test cavity may be contaminated with some amount of container contents. In that case, the test cavity is cleaned after removing the leaking container. This cleaning may be performed by evacuation and / or by cleaning with a cleaning gas, preferably by cleaning with nitrogen and / or by heating, for example by using a heated cleaning gas. This is done by a combination of

  When the method and apparatus according to the invention are applied to the testing of a plurality of containers in-line, i.e. a plurality of methods and apparatuses according to the invention operate in parallel on a set of containers. If a container has a leak, the test cavity is not introduced into the test cavity for the next test cycle, and the test cavity is emptied. . During that cycle, the test cavity will continue to be tested in other test cavities, but this test cavity, which may have been contaminated, is cleaned and readjusted. Further, in some cases, it is proposed that the internal pressure of the container is increased to an atmospheric pressure or higher by mechanically pressing the container wall toward the inner side so that the liquid can be squeezed out when there is a leak. The

  To achieve the above object, the present invention is an apparatus for performing a leak test of at least one closed container, wherein the container contains at least one liquid component, and the apparatus includes at least one liquid component. An airtightly closable test cavity, at least one vacuum pump operably connected to the test cavity, and at least one pressure sensor operably coupled to the test cavity. The vacuum pump is such that the test cavity can be evacuated below the vapor pressure of the liquid component at about room temperature, and the pressure sensor is a vacuum-compatible pressure sensor, preferably We propose an apparatus that has a Pirani sensor stage.

  Preferred embodiments of the method and device according to the invention are defined in the dependent claims. The method and apparatus according to the invention are preferably used as defined in claims 64, 65. In this case, apart from performing a leak test on a small-sized container, the present invention is for things such as gasoline or gas, including those installed on trains and those installed on vehicles. Can be used to permanently observe the tightness of a giant tank plant. In this case, an alarm signal is generated as soon as a leak is detected.

It is a graph which shows the temperature dependence of the vapor pressure of a liquid qualitatively. Fig. 1 schematically shows a test device according to the present invention operating on the basis of a method according to the present invention. FIG. 4 is a graph for explaining the operation of the method and apparatus according to the present invention and qualitatively showing the time dependence of the ambient pressure of the container to be tested. It is a functional block diagram which shows the preferable form at the time of embodying operation | movement of the test apparatus by this invention. FIG. 2 is a functional block diagram illustrating a preferred form for implementing evaluation electronics in an apparatus according to the present invention that accomplishes the method according to the present invention. FIG. 3 schematically shows a batch operation of the device according to the invention. FIG. 2 schematically shows a test cavity for testing a container with a flexible wall. FIG. 6 is a perspective view showing half of a test cavity for testing three containers per batch. FIG. 2 schematically shows a double-walled tank used directly for carrying out the method according to the invention using the device according to the invention for investigating tank leaks. FIG. 2 schematically shows a preferred seal in a test cavity of a device according to the invention. FIG. 5 is a graph showing pressure changes during a test cycle, where a container or medical application blister pack is shown as having a large or very large leak. The test is carried out using the test cavity according to FIG. 8 without the need for impedance measurements and thus without using the electrodes 32, 34. FIG. 5 is a graph showing pressure changes during a test cycle, where a container or medical application blister pack is shown as having only minor leaks. The test is carried out using the test cavity according to FIG. 8 without the need for impedance measurements and thus without using the electrodes 32, 34. FIG. 3 is a graph showing pressure changes during a test cycle, where a container or medical application blister pack is shown as being perceived as leak-free. The test is carried out using the test cavity according to FIG. 8 without the need for impedance measurements and thus without using the electrodes 32, 34. FIG. 4 is a signal distribution / functional block diagram showing a simplified preferred form of an evaluation unit for operating on the method according to the invention in an apparatus according to the invention. It is a graph which shows the time change of a pressure, Comprising: The statistical deviation of the pressure change when measuring a container without a leak or measuring only a test cavity without introducing a container is shown. Fig. 2 shows a part of the device according to the invention operating on the preferred mode of the method according to the invention, i.e. functioning to form a dynamic reference value for a leak test using a sequentially updated average value FIG. 3 is a simplified functional block / signal distribution diagram showing a portion of the device to perform. Fig. 6 is a graph simulating qualitatively the time variation of a signal in a preferred method according to the invention, i.e. in the operation of a preferred device according to the invention, i.e. when a dynamically updated reference value is formed for leak recognition. . In a further preferred mode of operation of the method and device according to the invention, i.e. in a mode in which a dynamically updated average signal is formed as a basis for the reference value to be compared with the pressure difference signal evaluated during the container test. FIG. 3 is a simplified signal distribution / function block diagram. Pressure on test cavities operated one after the other in the device of the invention using multiple cavities to show a dynamic update of the average signal such that a leak is recognized based on a reference value for comparison It is a graph which shows a measurement result in arbitrary units with respect to a time axis. FIG. 2 is a simplified schematic diagram of a test cavity according to the present invention, showing a state of being rotated during a test. It is a figure which shows the effect of rotation of the test cavity shown in FIG. 18 in relation to the relative positional relationship of a leak location and an accommodation product. It is a figure which shows the effect of rotation of the test cavity shown in FIG. 18 in relation to the relative positional relationship of a leak location and an accommodation product. FIG. 2 is a schematic diagram illustrating a simplified standard leak configuration for calibration for calibrating the apparatus according to the present invention when performing the method according to the present invention;

  Hereinafter, the present invention will be further described with reference to the accompanying drawings illustrating some specific embodiments and the presently preferred embodiments for implementing the present invention.

FIG. 1 qualitatively shows the change in the steam pressure p _V (T) in the graph of pressure and temperature. At a predetermined temperature T _X, and it reaches the corresponding vapor pressure p _VX, liquid starts to evaporate. Above the vapor pressure curve, the material is in the liquid phase and below it is in the gas phase.

  As shown in FIG. 2, the device according to the invention comprises a test cavity 1. The test cavity 1 includes a cover 3 that can be closed in a sealed manner. A vacuum pump 5 is connected to the test cavity 1. The vacuum pump 5 can be a drag pump, a rotary piston valve pump, a diffusion pump, or a turbo vacuum pump such as a turbo molecular pump. The choice of pump depends on the degree of vacuum to be established in the cavity 1. In addition, a vacuum pressure sensor 7 such as a Pirani sensor is provided. The vacuum pressure sensor 7 measures the pressure in the test cavity 1. At least one closed container 9 containing at least some containing product containing at least one liquid component is introduced into the test cavity 1 by opening the cover 3. Thereafter, the test cavity 1 is closed in a sealed manner (airtight manner). By operating the vacuum pump 5, the space around the container 9, that is, the intermediate volume V between the test cavity 1 and the container 9 is reduced.

As shown in FIG. 3, the pressure in the volume V starts from the atmospheric pressure p ₀ and is reduced to a pressure equal to or lower than the value p _V corresponding to the vapor pressure of the liquid component in the container 9. The vacuum pump 5 is selected so that the test cavity 1 can be evacuated to a value that is at least one digit, preferably two digits, more preferably three digits, less than the vapor pressure p _V of the liquid component of the contained product. It is desirable to keep it.

The test is preferably performed at room temperature, ie at a temperature T of about 20 ° C. When the liquid component is water, the vapor pressure p _V of water at room temperature is 20 mbar. In this case, the test cavity 1 can be evacuated to about 10 ⁻² mbar as the vacuum pump 5. It is preferable to prepare such a thing.

If the container introduced into the test cavity 1 has a relatively rigid wall 11 and there is no leakage, the pressure in the volume V is qualitatively determined by the curve (a ). That is, the pressure will drop close to a reachable predetermined pressure value that depends on the type of vacuum pump being used. On the other hand, when the container 9 has a leak at the place 13 as shown in FIG. 2, for example, a small amount of the liquid component 14 in the contained material is pulled from the container 9 through the leak place 13. issued by it and will, immediately at the time the pressure in the volume V becomes p _V, it begins to evaporate in the volume V. Due to this phenomenon, the pressure curve follows (b) as qualitatively shown in FIG. That is, the evaporation of the liquid cancels the action of the vacuum pump 5 and causes a pressure increase in the volume V. In order to finally obtain the vacuum level indicated by the curve (a), the vacuum pump 5 must further remove the vapor. If the leak is located in a region of the container 9 where air is constrained, such as location 13 'in FIG. 2, the vacuum of volume V is initially applied by the vacuum pump 5. Blow out the air from the container. Furthermore, the liquid component in the container 9 starts to evaporate into the container, and this vapor is sucked out through the leak portion 13 '. Again, the pressure curve that would have been followed if only air was being removed by the vacuum pump 5 is not reached and a pressure rise in the volume V occurs.

  A pressure curve in the volume V is observed by the pressure sensor 7. Regardless of the volume V in the test cavity, the clear difference between curves (a) and (b) in FIG. 3 is a few seconds (1 for leaks smaller than 1 micron (0.02 μm)). It has been confirmed by experiments that it is obtained after a time span τ of ~ 3 seconds. In this case, the pressure difference between the leaking container and the non-leaking container is about an order of magnitude. The experiment was conducted for the case where water was contained as a liquid component.

  In order to detect the leakage of the container, for example, after the time span τ, the absolute pressure in the volume V can be measured as an absolute value. However, as described with reference to FIG. ,preferable.

  Referring again to FIG. 2, the pressure sensor 7 is operatively connected to the evaluation unit 15. In the evaluation unit 15, in particular, a leak recognition threshold value is preset (preset), as schematically shown by the presetting unit 17. The output of the evaluation unit 15 is a binary signal indicating whether there is a leak.

As shown in FIG. 4, the output of the vacuum sensor 7 is an input to the storage unit 19 that is controlled by the timing control signal s _{1 as} shown schematically by the switch S. Signal input to a storage unit 19, as a first point, performed at the time t ₁ in FIG. As a second point, at time t ₂ in FIG. 3, the output of the storage unit 19 and the output of the sensor 7 are connected to the input port of the difference forming unit 21. The difference forming unit 21 generates an output signal corresponding to the pressure difference Δp in FIG.

A most preferable configuration example of the evaluation electronics system is shown in FIG. An output signal of the sensor 7 is input to the conversion unit 121. The conversion unit 121 includes an analog-digital converter (analog-to-digital converter) 121a as an input stage, and a digital-analog converter (digital-to-analog converter) 121b at the subsequent stage. The output of the converter 121 is supplied to the differential amplification unit 123. In addition to this, the differential amplification unit 123 directly receives the output signal of the sensor 7. The output of the difference amplification unit 123 is supplied to the further amplification unit 125 corresponding to the case of the difference unit 21 of FIG. The output of the amplification unit 125 is superposed on its own input in the adder 128 via the storage unit 127. The input of the storage unit 127 is supplied from the output of the unit 125. The timer unit 129 performs time control with such a configuration. In order to store the first pressure value from the sensor 7 at time t ₁ in FIG. 3, the timer unit 129 enables a conversion cycle in the unit 121. As a result, the reconverted analog output signal el ₀ appears at the output of the unit 121. At the same time, substantially the same signal from sensor 7 is applied to the second input of unit 123 as signal el. Thus, a zero signal should appear at the output port of unit 125. Nevertheless, in general, a zero offset signal will appear at the output port of unit 125. This signal is stored in the storage unit 127 under the instruction signal from the timing unit 129. In time _{t 2, the} conversion in the unit 121 is not triggered. Therefore, the pressure value at time t ₂ appears directly from the sensor 7 at the input port of the amplification unit 123, and the stored pressure value at time t ₁ appears from the converter 121. Further, the zero offset signal stored in the unit 127 is superimposed as an offset compensation signal. For this reason, the signal obtained at the output of the amplification unit 125 is one in which the zero offset is compensated.

  This allows a very accurate measurement of the pressure difference Δp in FIG.

When the container under test has a large leak, the pressure in the volume V of the test cavity 1 differs from the operation start time of the vacuum pump 5 as shown by a curve (c) in FIG. A response will be shown. This, for example, at an initial time t ₀ by comparing the output signal of the sensor 7 with a predetermined threshold value (not shown), can be easily detected. When the actual pressure value has not reached such a threshold value, the operation of the vacuum pump 5 with respect to the test cavity 1 is stopped. This is to prevent the contents in the container from being ejected in large quantities into the test cavity and contaminating the cavity in the case of a large leak.

  As mentioned above, the method according to the invention functions correctly regardless of the volume V between the test cavity 1 and at least one container to be tested. This makes it possible to simultaneously test a batch 9 'consisting of a plurality of containers 9, as shown in FIG. Thereby, it is possible to accurately detect whether any of the containers 9 has a leak. Further, since the detection accuracy does not change greatly even for different volumes V, a test is performed with one test cavity 1 even if the container has a greatly different shape or a container with a significantly different volume. be able to.

  If the wall of the container to be tested cannot mechanically withstand a pressure load of about 1 bar, it is provided with a cover 3 'to fit the shape of the container 9 as shown schematically in FIG. A test cavity 1 ′ is used. In this case, as schematically shown in FIG. 7, the protrusion 20 prevents the container wall from sticking to the inner wall of the test cavity due to the effect of evacuation. This ensures that a space V to be evacuated by the present invention remains between the container and the test cavity wall. Such protrusions 20 can be realized by fitting meshes or lattices. Alternatively, it can be preferably realized by mechanical roughening of the inner wall of the cavity. In this case, the minute convex shape supports the container wall, and a continuous internal space is formed as the volume V. The

  It is further advantageous to mechanically bias a part of the container wall inward, for example when the cavity cover 3 or 3 'is closed, as indicated by the broken line in FIG. Thereby, the internal pressure of the container 9 can be increased, and when there is a leak, the liquid component of the contained product can be pushed out from the leak location.

  As shown in FIG. 9, the method and apparatus according to the present invention can be used to observe giant tank leaks. FIG. 9 shows a double-wall type tank having an inner wall 23 and an outer wall 25 in particular. A test for the tightness of both of these walls is performed by using the intermediate space of the two walls as volume V in FIG. Such a technique can be applied to, for example, a tank on a vehicle or a tank on a railway vehicle, and can also be applied to a stationary large tank plant such as a gasoline tank. .

  FIG. 8 shows the half 1a of the test cavity 1 when the method according to the invention using the device according to the invention is applied to three containers, which are small plastic containers for medical use arranged at location 29. It is shown. If the test cavities 1 are adapted to the container shape, these containers can have flexible walls. In addition, other techniques have been shown to quickly detect whether any of the containers have a large leak. That is, impedance measuring electrodes 32 and 34 are provided. The impedance measuring electrodes 32 and 34 are integrally installed on the wall of the cavity 1 and are electrically insulated from each other. These electrodes are connected to the impedance measurement unit, preferably to the resistance measurement unit 35. If the liquid content is sucked out of the container wall by evacuating a test cavity, preferably having a roughened inner wall, it is measured between the electrodes 32,34. The impedance changes rapidly and can be detected immediately. The output of the impedance measuring unit 35 stops further evacuation of the test cavity 1 (although the specific configuration is not shown).

  When the test cavity is contaminated by the containment leaked from the leak container, the test cavity can be evacuated and / or by injecting gas, preferably by injecting nitrogen, and / or Cleaning is performed by heating. In FIG. 8, a supply line 36 is shown for the cleaning or cleaning gas supplied under control from the gas tank 37 towards the contaminated test cavity 1. This gas is preferably nitrogen.

  The cavity half 1a in FIG. 8 is hermetically superimposed on the other half. This completes a test cavity 1 similar to that of FIG.

  The present invention is particularly suitable for in-line container testing due to the short measurement cycle, and when such in-line testing is performed, two or more test cavities, especially 1 Several test cavities in a set are provided, for example on a carousel. For these test cavities, the containers (not shown) to be tested are automatically loaded from the conveyor, and these test cavities simultaneously perform the above test procedure. If it is detected that one of the containers tested in such a cavity is leaking, for the cavity that was testing that container, during the next container set measurement cycle, No container is introduced and this cavity is left empty. Meanwhile, the evacuated cavity is cleaned by evacuation and / or gas injection and / or heating as described above.

  A good vacuum seal between the test cavity cover 3 or 3 'and the body of the test cavity 1 or between the two halves 1a forming the test cavity in the case of the test cavity of FIG. It is clear that a stop seal must be established. Preferably, as shown in FIG. 10, at least a pair of parallel seal bodies 28 made of concentric O-ring seal bodies are provided, and an intermediate space 29 between the seal bodies is individually evacuated, Established. If the container to be tested contains a containing product containing two or more specific liquid components, the vapor pressure of the component with the highest vapor pressure, in other words, at the highest atmospheric pressure. The vapor pressure of the component that begins to evaporate is selected for leak detection. In this case, the viscosity must also be taken into account. That is, for the formation of the vapor pressure, a component having a liquid property that can sufficiently pass through even the smallest leak must be selected. By evacuating the test cavity to a pressure well below the vapor pressure of all liquid components, it is not important which component's vapor pressure should be considered.

  In the case of a container in which the time variation of the pressure, measured by the preferred form of the method according to the present invention, using the preferred form of the apparatus according to the present invention, has a large leak (FIG. 11a), in the case of a container with a small leak (FIG. 11b) and in the case of a container without a leak (FIG. 11c).

  These graphs will be described with reference to FIG. 12, which shows a preferred observation unit and control unit by units 15 and 17 in FIG.

In the case of Figure 11a, the timing unit 201 of Figure 12, at time _{t 10,} using a pump facility 105 initiates evacuation of the test cavity 103. This is indicated by the evacuation start signal EVST / t _{10 in} FIG.

After a fixed predetermined time ΔT, eg 0.75 seconds, the output signal A ₅ of the pressure sensor (not shown in FIG. 12) in the test cavity 103 is the first reference signal in the preset source 107. It is compared with the preset value RFVGL. For this purpose, the comparator unit 109 is driven by the timer unit 201 at time t ₁₀ + ΔT.

After the lapse of time [Delta] T, as indicated by an electric signal A ₅ of FIG. 12, actually observed pressure value, as indicated by the curve I in FIG. 11A, when it has not reached the value RFVGL This means that there is a very large leak VGL. This is detected by the comparator 109, and the output signal A ₁₀₉ is generated in the comparator 109. As shown by the characteristics in the block 109 of FIG. 12, when the output signal driven at the time t ₁₁ = t ₁₀ + ΔT of the comparator 109 is, for example, at a high level and indicates the presence of VGL. This is the output at the VGL output. If the pressure around the vessel 103 during the test, i.e. the pressure in the test cavity, reaches the reference value RFVGL and is below this reference value as shown by curve II in FIG. 11a, the VGL output signal is Is not generated.

  As will be described later, when the VGL signal is generated, the evacuation cycle is preferably stopped. This is because contamination of the vacuum pump 105 is or can be caused by a very large leak in the container under test.

As in the case shown by the curve II of Figure 11a, when the VGL is not generated, evacuation continues until a further time t _13. At time t ₁₃ , the timer unit 201 stops the pump equipment 105 and disconnects the pump equipment from the chamber 103 by the valve 106. Further, the timer unit 201 activates the comparator unit 111. For this comparator unit 111, a further reference value RFGL generated by the reference signal generation source 113 is introduced. In t _13, when the pressure value of the surrounding of the test cavity has not reached the RFGL, the comparator unit 111 generates an output signal GL. This signal is a signal indicating that the container under test has a large leak. Again, some actions are taken with regard to further operation of the test system, as described below.

When either the signal VGL or GL is sent from the comparator 109 or 111, the timer unit 201 is in principle because the test has already been completed and the quality of the test container has already been instantly recognized. Is reset. This reset is indicated by signal RS _{201 in} FIG. If not reset, immediately after t ₁₃ , the pressure value A ₅ (t ₁₃ ) around the container is stored in a holding unit or storage unit 117. The output of the holding unit or storage unit 117 is led to one input port of the difference forming unit 119. In this case, the second inputs of a difference forming unit 119, the output A ₅ of the pressure sensor for observing the pressure vessel surrounding the test is connected. As schematically shown by unit 121 in FIG. 12, counting from t _13, after the presettable cycle time T _T, as indicated by the switching unit 123 in FIG. 12, the pressure of the output of unit 119 The difference DP is calculated. This pressure difference DP is supplied to a further comparator unit 125. The comparator unit 125 is activated after a test time T _T. A reference value DPREF is supplied to the comparator unit 125 by a further reference value source 127. As will be described later, the value of DPREF can be changed in a controllable manner according to time. And / or, also the reference value phi _R forms a reference value of dpref, also can be controllably changed with time.

If DP at time t ₁₃ + T _T is greater than reference value DPREF, signal FL is generated in unit 125. This signal is a signal indicating that there is a minute leak in the test container. This situation corresponds to the situation shown in FIG. If DP is smaller than the reference value DPREF, it is determined that the container does not have a leak, and none of the VGL, GL, and FL signals are generated. This situation corresponds to the situation shown in FIG.

  When the VGL signal is generated in FIG. 12, even when the vacuum pump 105 is connected to a single chamber, or the vacuum pump 105 is used in parallel to a plurality of chambers 103. Even when the evacuation pump 105 is applied to an in-line process as described above, the evacuation pump 105 is immediately disconnected from all the test chambers 103. This is because the vacuum pump 105 may be contaminated by leaked container contents due to a very large leak. In this case, as a preparation for such a situation, a spare pump facility can be provided. Under that circumstance, a spare pump is connected to one or more test chambers to continue testing, while the first pumping facility that may have been contaminated is reconditioned. The

  In a multi-chamber type in-line test system, for example, in a rotating conveyor test plant equipped with a plurality of test chambers, a signal FL indicating the presence of a large leak, or in some cases a signal FL indicating the presence of a small leak Is preferably prohibited or “bypassed” further introduction of the container under test into the chamber in which the container with the leak was tested. In contrast, the other chambers are still operational and perform tests on newly supplied containers under test. Such a bypass procedure for a test chamber that has been tested on a container with recognized severe or minor leaks is intended to not affect other test results in that chamber, i.e., This is done in order to prevent the effects caused by the contents of the leak container that may have contaminated the chamber from affecting other test results in the chamber.

  Such bypassed chambers are reconditioned in parallel with performing additional test cycles in other chambers.

  Reconditioning can be performed by heating the chamber, and by cleaning the chamber with liquid and / or gas, particularly with heated gas. Whether the chamber has been properly reconditioned is checked by performing the test as if the container under test was contained within the chamber. In this case, proper readjustment has occurred because the DP in FIG. 12 in the empty chamber is smaller than, for example, DPREF, or a properly set “DP− for empty chamber”. It is indicated by being smaller than "REF" (ECDP-REF).

Such ECDP-REF measures DP _e in a clean and emptied test chamber and uses these measured values DP _e as reference values in testing the chamber for proper reconditioning. Can be prepared by storing.

  Referring to FIGS. 11a and 11b, it can be seen that the setting of the reference value RFGL, especially the setting of the reference pressure difference value DPREF, is very important and has a great influence on the accuracy of the system. In this case, factors such as ambient temperature, ambient air humidity, slight contamination of the pump, etc. have an effect on the temporal change of pressure, and these two important reference values are set to maximum accuracy, especially DPREF. If this is done, there is a possibility of erroneous determination.

FIG. 13 quantitatively shows a pressure curve similar to that of FIGS. 11a to 11c when measurement is performed without a container. In t _13, for statistical deviations, variations in slight pressure value occurs. Thus, before starting container testing in a plurality of test cavity plants, the evacuated airtight test cavity is tested as shown in FIG. 13 to determine the average value (RFGL) _m . The value of RFGL when used in the comparator unit 111 of FIG. 12 or when used in FIGS. 11a to 11c should be (RFGL) _m plus an offset value ΔRFGL. all right. It is pointed out that atmospheric parameters such as ambient air temperature and humidity can be considered constant during a calibration cycle in an empty and adjusted test cavity and when obtaining the measurement results of FIG. Keep it. Nonetheless, when performing on-line testing, these perturbing parameters may change slowly and may change (RFGL) _m .

Even when single test cavities are connected sequentially or when multiple or two or more test cavities are connected in parallel, during multiple tests or in-line tests At all times t _{13 when} it has already been determined that each container does not have a severe leak, the actual output signal of the pressure sensor is input into the averaging unit 130. In the averaging unit 130, the last m actually measured pressure values of a plurality of containers that do not have a severe leak are averaged. The average signal result as output corresponds to (RFGL) _m in FIG. However, it changes with time based on, for example, changes in the atmospheric parameters over time. An offset value ΔRFGL is added to the output average result A5 shown in FIG. This addition result forms a reference value RFGL that changes dynamically. This dynamically changing reference value RFGL is applied to the comparator unit 111 of FIG. This dynamically changing reference value RFGL is shown in FIG. 15 starting from the initial setting as described for the reference measurement in the empty cavity 103, for example.

As can be clearly seen from FIG. 15, the average pressure value A5 (t ₁₃ ) forms the basis for reference to DPREF here. Accordingly, as shown in FIG. 12, reference difference pressure value DPREF, instead of referring to the absolute static value as phi _R, refer to A5.

Further improvements in accuracy were made as described below. This improvement can be adopted individually or in addition to the dynamic upper limit of dynamic RFGL and DPREF based on it. In this case, as shown in FIG. 16, whenever the output signal FL indicates that the container under test does not have a leak at the end of the time interval T _T , the measured value DP of the pressure difference is always obtained. To the averaging unit 135. The output signal of the averaging unit 135 corresponds to the pressure difference average signal DP over the last m test cycles. This average signal is offset by an amount of ΔDP, and the offset result is used as the DPREF signal for application to unit 127 of FIG.

Returning to FIG. 15 above, in this case a constant DPREF signal has been applied, which affects the pressure difference by the technique of averaging DP, as schematically shown by the curve (DPREF) _t . A dynamically changing check value DPREF is provided that varies with changes in such disturbance parameters.

Preparing a dynamically changing signal (DPRF) _t , such as the signal shown in FIG. 15, refers to a dynamically changing A5 value without having to prepare a dynamically changing base value A5. instead of, as indicated by a broken line in FIG. 12, by referring to the static constant value phi _R, is able to be achieved, it is clear.

Evaluation of one or more of the test cavity of the output signal A ₅ are digitally carried out that it is preferable, that is, carried out is preferably that after converting the output signal of each sensor from the analog to digital ,it is obvious.

FIG. 17 shows an actual measurement value DP of a pressure difference continuously measured for a plurality of test cavities in an in-line test plant in an arbitrary unit with respect to the time axis. The calculated average value DP of the pressure difference obtained as shown in FIG. 16 is shown, and (DPREF) _t as shown in FIG. 15 or FIG. 16 is finally obtained. As can be seen from FIG. 17, the average values DP and (DPREF) _t change with time and change with each successive test. Thus, the pressure difference value at A, which is larger than the instantaneous value of (DPREF) _t , is not the influence of the averaged DP, but is a measurement result for a container having a leak as shown in FIG. 11b. It is determined to be a thing.

  Also, if a test of a container in a particular test cavity shows a leak result for a predetermined number of consecutive tests, such as three consecutive times, such a test cavity may also be used for further testing. To be bypassed. This test cavity is then readjusted, as determined that it is contaminated or that the test cavity itself has a leak. Such test cavities may be contaminated by continuously testing leaky containers and may be poorly airtight. These are confirmed at the time of readjustment and, as described above, also confirmed by a test after proper readjustment.

  Also, as described above, the test cavity can be heated to a predetermined temperature for certain types of containers under test, particularly for certain types of contained products. This heating is preferably controlled for each test cavity, for example by negative feedback temperature control. In this case, the temperature dependent vapor pressure of the housed product is set within a predetermined pressure range. Such heating is performed in a preheating cycle before performing the actual test cycle as in FIGS.

  As noted above, container leaks are located on the wall portion that is in contact with the air confined within the container or in contact with the contained product. Judgment is made regardless of whether it is located on the wall. Nonetheless, for certain containments, such as those containing certain components in the liquid, there may be differences in the time course of the pressure difference around the container during the test.

  Accordingly, as conceptually shown in FIG. 18, in some cases, one or several test cavities 103 can be prepared so that the container 9 to be tested can be moved. This is performed by, for example, installing the test cavity 103 so as to be rotatable around the rotation axis A and rotating the test cavity by the rotation shaft 140. In this case, a lead wire for connection to the pressure sensor in the test cavity, a lead wire for connection to the heating mechanism of the test cavity, and the like can be connected through the drive shaft 140. The cavities 1 and 103 are preferably not driven to rotate and are driven to swing as indicated by ± ψ in FIG. In this technique, as conceptually shown in FIG. 19, the leak portion L alternately repeats the state in contact with air and the state in contact with liquid. Therefore, whether the leak location L is at the position shown in FIG. 19a or the position shown in FIG. 19b, the test is always performed by evaporation of the liquid component when there is a leak.

  Whether it is a single chamber test device or a multi-chamber test plant, such as in-line testing, proper functioning of the test device and calibration of the evaluation unit are preferably implemented in the test plant. Preferably, this is done using a standard leak structure attached to. In that case, plant recalibration and / or overall testing can be performed whenever necessary. Such a standard leak structure or a calibration leak structure is shown in FIG.

  In FIG. 20, for example, a needle valve 142 is installed in a line from the test cavity, which is similar to the cavity indicated by reference numeral 103 in FIG. 12, to the vacuum pump 105. The needle valve 142 is adjustable, but preferably is preset by a plant user to a predetermined leak value rather than being variable. A liquid reservoir 144 is connected to the line leading to the vacuum pump 105 via a needle valve 142. The liquid reservoir 144 preferably contains distilled water. The pressurization line and valve 146 allow the reservoir 144 to be adjustably pressurized. The needle valve is set so that distilled water in the reservoir 144 does not enter the connection line between the chamber 103 and the vacuum pump 105, and only water vapor can enter. Nevertheless, by adjusting the pressurization conditions of the water in the reservoir 144 by means of the pressurization line and valve 146, liquids do not enter and contaminate the chambers and / or connecting lines and / or vacuum pumps. It is possible to simulate leaks that vary in various degrees. For plants with multiple test cavities, such a calibration structure with needle valve 142 can be centrally provided as a single unit, and this calibration structure can be applied to all chambers. 103 is connected in parallel. This is preferred in plants where a single centralized pumping facility 105 is provided that functions in parallel for all installed chambers or cavities. Alternatively, such a calibration structure can be provided individually for each installed chamber 103.

  In most cases, it is necessary to attach a resistance measurement as described with reference to FIG. 8 by applying the leak test technique in which the ambient pressure of the container at the time of the test is lowered to the vapor pressure of the liquid containing component or less. It is recognized that there is no. Therefore, the electrode configuration and the measurement unit can be omitted in each test chamber. This reduces the overall plant cost and the plant complexity. The present invention is particularly suitable for testing bottles and blister packs, and can be used as an in-line type in their manufacturing process by checking bottles and blister packs individually, especially for medical applications. Yes. As conceptually shown in FIG. 6, when a plurality of containers 9 are mechanically connected to each other to form a set of containers, for the leak test, such a set of It is clear that it can be considered as one container.

When applying the method and apparatus according to the present invention with respect to the blister pack, the entire test cycle, i.e., the time interval from t ₁₀ in FIG. 11 until the end of T _T, shorter than 2 seconds. This results in very high throughput in an in-line plant with multiple test cavities, eg 24 on a carousel.

DESCRIPTION OF SYMBOLS 1 Test cavity 3 Cover 5 Vacuum pump 7 Vacuum pressure sensor 9 Closure container 11 Wall 13 Leak location 14 Liquid component

Claims (22)

A method for manufacturing a closed containment vessel, comprising:
Filling the container with a material having at least one liquid component, thereby forming an area covered by the liquid in the container;
Sealingly closing the container;
Performing a leak test of the container to determine whether the container is leaking or not leaking;
In that way,
For the leak test,
Applying a pressure differential across the interior and exterior of at least some of the walls of the container by evacuating the surrounding space of the container using a vacuum pump;
While observing the pressure in the surrounding space, reducing the pressure in the surrounding space to a vapor pressure value of the at least one liquid component;
When observing the pressure value of the surrounding space as a leak determination signal, when the pressure value of the surrounding space reaches the vapor pressure value, the container is determined to have a leak;
A method characterized by that. The method of claim 1, wherein
Reducing the pressure in the surrounding space to a value at least two orders of magnitude less than the vapor pressure value. The method according to claim 1 or 2, wherein
When two or more liquid components are contained in the container,
The method is characterized in that the vapor pressure value is a larger vapor pressure value of the vapor pressures of the two or more liquid components. The method of claim 1, wherein
A method characterized in that the test is performed at room temperature. The method of claim 1, wherein
The method of observing the pressure value observed as the leak determination signal after reaching the vapor pressure value. The method of claim 1, wherein
Obtaining a first pressure measurement signal by sampling the observed pressure value at a first point in time;
Thereafter, a second pressure measurement signal is obtained by sampling the observed pressure value at a second point in time,
A method characterized in that a pressure difference formed by these two pressure measurement signals is obtained as a leak determination signal. The method of claim 6 wherein:
Generating the first measurement signal and the second measurement signal as electrical signals;
A method of storing the first signal at least until the second point of time. The method according to claim 6 or 7, wherein
A pressure measuring sensor is provided in the surrounding space, and at the first point in time, the sensor is operatively connected to both input ports of the difference forming unit,
Generating a zero point offset signal from the output port of the difference forming unit;
This zero point offset signal is stored,
The stored zero point offset signal is used to compensate for the zero offset in the difference signal between the two measurement signals. The method of claim 6 wherein:
A pressure measuring sensor is provided in the surrounding space,
Comparing the sensor output signal with one or more predetermined signal values. The method of claim 6 wherein:
A method of storing the first measurement signal by activating an analog to digital converter at the first point in time. The method of claim 10, wherein:
Re-converting the digital output signal from the analog to digital converter into an analog signal. The method of claim 1, wherein
A method of treating a batch of a plurality of containers as if they were one container and testing them simultaneously. The method of claim 1, wherein
Perform impedance measurement at or at least in the vicinity of the wall in the surrounding space, preferably by direct current resistance measurement,
A method of determining whether to further lower the pressure in the surrounding space or to stop the lowering according to a result of the impedance measurement. The method of claim 1, wherein
Providing a test cavity shaped to conform to the outer shape of the at least one container, whereby at least at the part of the wall, between the part of the wall and the wall of the test cavity; A method characterized by maintaining a residual volume as pressure to reduce pressure. The method of claim 1, wherein
Preparing a test cavity for the at least one container to form a test chamber having a volume substantially greater than the volume of the container. The method of claim 1, wherein
Preparing a test cavity for said container;
If the container tested in the test cavity has a leak, at least clean the test cavity,
This cleaning is performed by evacuating the test cavity and / or cleaning with gas, preferably by nitrogen and / or heating. The method of claim 1, wherein
In-line testing of a series of containers in a set of test cavities,
A method characterized by prohibiting a test in a test cavity in which a test has been performed for at least one test cycle when the tested container has a leak. The method of claim 1, wherein
A method of increasing the internal pressure of the at least one container by mechanically pressing at least a portion of the wall inward. The method of claim 1, wherein
For containers where one liquid component is water,
A method of evacuating the surrounding space to 20 mbar or less, preferably about 10 ⁻² mbar or less. The method of claim 1, wherein
Reducing the pressure in the surrounding space to a value that is at least three orders of magnitude less than the vapor pressure value. The method of claim 1, wherein
A method of determining that a leak is present in the container when a leak smaller than 1 micron is present in the container. The method of claim 1, wherein
A method wherein the container is determined to have a leak if a leak of 0.02 microns is present in the container.

Images