Method and device for identifying defects in capsules in filters for the tobacco processing industry

Abstract

1. A method for identifying defects in capsules in filters for the tobacco processing industry, comprising the steps of: a microwave measuring device measuring a value indicative of the density and / or moisture content of said filter along the length of said filter; The method comprises: a. comparing the measurement to a threshold value and, if below said threshold, indicating that there is a capsule defect in said filter; b. determining a value of the degree of variation of said measurements in an area in front and / or behind the measurement location where at least one measurement is below said threshold; c. comparing said value of said degree of variability to a predetermined capsule threshold; i. if the capsule threshold is exceeded, a broken capsule is present; ii. if the capsule threshold is below, the capsule is missing; A method comprising:

Description

[Technical field]

[0001] The present invention relates to a method and apparatus for identifying defects in capsules in filters for the tobacco processing industry. [Background technology]

[0002] Filters in the tobacco processing industry can be in the form of filter strands, cut filter rods or as filter pieces connected to the cigarette. For taste reasons, capsules with flavoring substances are incorporated in the filter, which capsules are broken in the filter just before the cigarette is smoked, so that the flavoring substances are released in the filter. A commonly used flavoring substance is, for example, menthol. The advantage of using this type of filter, with one or more capsules incorporated in it, is that the flavoring substances are already located in the filter and are not burned together with the tobacco, but are added to the tobacco smoke without being burned. Thus, the production of substances harmful to health due to the combustion of the flavoring substances is prevented. Likewise, the deterioration of the flavoring substances due to the combustion with the tobacco is also excluded.

[0003] Various parameters of the capsules can be measured with suitable sensor systems during production and / or during inspection of the final product to monitor their quality. Here, capsule-related parameters are, for example, the position and spacing of the capsules in the filter as well as the capsule condition, i.e. whether the capsules are intact or broken.

[0004] EP 2249670 discloses various sensor systems for inspecting capsules, capable of detecting the state and presence of capsules in the filter strand.

[0005] EP 2207027 discloses a method for measuring the mass or density and / or moisture content in a number of divided units in a non-metallic substrate. A microwave measuring device, consisting of at least one microwave resonator generating at least one resonance mode, measures in a locally limited measurement range. The microwave measuring device measures at least one characteristic variable of the resonance mode in each case when only the substrate is located in the measurement range without the divided unit and when the divided unit is at least partially located in the measurement range. An evaluation unit subtracts the measured variable for the substrate from the measured value for the divided unit and determines the moisture content and / or the value of the mass or density of the divided unit from the at least one difference using a characteristic map.

[0006] In fact, intact capsules show a pronounced local maximum in the density signal of the filter. In case of capsule breakage, the liquid flavoring first spreads in the filter and then evaporates. The shell of the broken capsule remains at the location of the capsule. Depending on the size and nature of the capsule and its shell, as well as the degree of its breakage, the shell of the capsule that is left behind results in one or more density signals. The amplitude of the density signal can be used to reliably identify a defective capsule in the filter. Thus, if the capsule is defective, the filter can be discarded from the product. However, it cannot be reliably distinguished whether a capsule is missing or whether the capsule is already destroyed. Thus, it cannot be reliably distinguished between a capsule defect in the case of a missing capsule and a capsule defect in the case of a destroyed capsule. There is therefore a need in the art to obtain additional information about capsules and capsule defects in the filter in a reliable manner with little effort. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] European Patent No. 2249670 [Patent Document 2] European Patent No. 2207027 Summary of the Invention [Problem to be solved by the invention]

[0008] The object of the present invention is to provide a method and a device with which it is possible to accurately and reliably evaluate measurement signals for defects in capsules of filters in the tobacco processing industry. [Means for solving the problem]

[0009] This object is achieved according to the invention by a method having the features of claim 1 and by a device having the features of claim 13. Advantageous embodiments are the subject matter of the dependent claims.

[0010] The method according to the invention for identifying capsule defects in filter strands for the tobacco processing industry uses a microwave measuring device to measure the density and / or moisture content of the filter in the longitudinal direction of the filter.

[0011] In a first method step, the maximum value of the measured values ​​is compared with a threshold value in a specific area of ​​the measurement profile. In this case, the measured values ​​indicate density and / or moisture content, meaning that the value of density and / or moisture content can be determined directly from said measured values ​​or with the aid of additional variables. If the threshold value is below, a capsule defect is identified in the filter. In order to analyze the capsule defect precisely, the degree of variation of the density and / or moisture content is recorded in the area in front of and / or behind the measurement position. The area around the target position of the capsule is taken as the measurement position. In a second comparison step, the value of the degree of variation determined for the area is compared with the capsule threshold value for the measured value. If the capsule threshold value is exceeded, a broken capsule is identified. If the value is below the capsule threshold value, a missing capsule is identified in the filter. In each case, the value of the degree of variation is compared with the capsule threshold value. The selection of the measurement position is particularly relevant for the identification of a missing capsule. A missing capsule can only be identified as a defect if the capsule is in a state where it should be there. Therefore, the target or expected position is selected as the measurement position.

[0012] The invention is based on the observation that a breakage of a capsule increases the fluctuation of the measurement signal, resulting in a greater value in terms of the degree of fluctuation than a pure filter lacking a capsule and having a uniform density without remains of broken capsules and / or of leaked flavoring substances, and therefore showing a lower degree of fluctuation of the measurement signal. Due to the fact that the second comparison is not based on the measurement signal itself versus density and / or moisture content, but on its degree of fluctuation in the area of ​​the capsule defect, information on the capsule condition can be obtained reliably with less technical effort.

[0013] In a preferred development, a line of best fit is determined for the measurements in a control area around the measurement location, and the degree of variation is determined for the line of best fit. The control area may, for example, have a predetermined distance in front of or behind the measurement location. Preferably, the line of best fit is formed by linear regression of the measurements in the control area. The use of the line of best fit is based on the insight that moisture content and density variations that are not due to the capsule or its contents vary more slowly and locally, and can therefore be better represented by linear regression.

[0014] In a preferred development, the microwave measuring device is designed as a microwave resonator, which measures the change in resonance width and / or the change in resonance frequency. The dielectric properties of the microwave resonator change due to the filter, with or without a capsule inserted therein. This leads to a broadening of the resonance curve and a change in its resonance frequency. During the evaluation of the measurement signal of the microwave measuring device, the change in the width of the resonance curve and the change in the resonance frequency can be evaluated as values ​​of moisture content and density. During the evaluation of the signal, the change in the resonance curve is generally recorded for an empty resonator. It is also possible to operate a suitable microwave measuring device with two or more resonance modes and to evaluate the change in the signal between the various resonance modes. In general, it is sufficient for the method according to the invention to measure the values ​​of the density signal and / or the moisture content signal and to determine their change depending on the degree of variation in the spatial range.

[0015] In a preferred development of the method according to the invention, the value of the degree of variation is determined for a predefined length in the longitudinal direction of the filter. This development length must be adapted to the expected target position for the capsule. The predefined length may, for example, be located in front of and / or behind the target position of the capsule where the measured maximum value for the density and / or moisture content is below a threshold value. Furthermore, the predefined length may, for example, coincide with the adjustment region of the line of best fit.

[0016] In a preferred embodiment of the method, the expected position of the capsule in the longitudinal direction is predetermined. This may be due to, for example, the manufacturing process of the capsule-type filter strand itself, for example, the capsule is inserted into the filter at a predetermined distance from a starting position. The predetermined position may be determined, for example, using a predetermined distance between capsules, in that it is measured when the presence of a capsule exceeds a threshold value, from which the expected position for the additional capsule is determined using a given distance value. During the production of the capsule-type filter strand in the manufacturing machine, it is important that the position of the capsule in the seamless capsule-type filter strand is correct with respect to the knife that divides the strand into individual filter rods.

[0017] In an advantageous development of the method, the value of the degree of variability can be determined in each case in the area in front and / or behind the expected capsule position. In this approach, the value of the degree of variability is calculated continuously, regardless of whether the measured values ​​of density and / or moisture content are below a threshold value or not. The standard deviation is then also evaluated for the past length if the value falls below the threshold value.

[0018] In a preferred embodiment, the filter is a filter rod that is transported in its longitudinal direction. The filter rod is produced from filter tow and transported on one side before being cut. The filter rod can be inspected on a continuous strand and / or the filter strand can be inspected after being cut into individual filter rods. However, the method according to the invention can be used to inspect finished cigarettes, not only during production. In this case, the filter to be measured is a filter piece of a cigarette that is transported in its longitudinal direction.

[0019] Measurement and inspection of the finished filter pieces on the cigarettes is preferably carried out at a test station, although in principle it may take place within the manufacturing process of the cigarettes.

[0020] Experiments have shown that in the method according to the present invention, the capsule and filter materials and the moisture content of the filter do not significantly affect the capsule threshold value.Therefore, the capsule threshold value can also be determined regardless of the capsule and filter materials and the moisture content of the filter, and there is no need to adapt the capsule threshold value when the capsule material or filter is changed.

[0021] Preferably, the degree of variation value is determined as the standard deviation.

[0022] This object according to the invention is also achieved by an apparatus having the features of claim 13. The apparatus is used for measuring defects of capsules in filters of the tobacco processing industry. The apparatus comprises a microwave measuring device and a control device, which may be part of the microwave measuring device. The microwave measuring device measures a value indicative of the density and / or moisture content in the filter, and the control device is designed to compare the measured value with a threshold value and to identify a capsule defect if it is below the threshold value. As in the case of the method according to the invention, the control device is also designed to reliably distinguish between broken and missing capsules using the comparison of the value of the degree of variation with the capsule threshold value. Preferably, the apparatus is part of a test station for filter rods. Likewise, the apparatus may be part of an apparatus for producing filters, for example a filter rod making machine, or a machine for assembling various filter pieces into a combined filter (filter rod combiner). Furthermore, the apparatus may be part of an apparatus for producing cigarettes, for example a cigarette making machine.

[0023] Preferably, the control device is designed to determine the standard deviation as a measure of the degree of variation.

[0024] The invention will now be explained in more detail with the aid of the figures shown below. [Brief description of the drawings]

[0025] [Figure 1]1 shows density signals plotted against location within a filter strand for intact and broken capsules. [Diagram 2] The density signal of the missing capsule is also shown. [Diagram 3] The density signal in a filter rod with four capsules is shown. [Figure 4] Standard deviation values ​​for broken and missing capsules are shown. [Diagram 5] The measured density values ​​for intact and broken capsules, as well as the line of best fit for missing capsules, are shown. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] FIG. 1 is a graph showing the density values ​​measured for a filter rod (plotted without dimensions) as a function of position in millimeters. Two measurement curves are plotted, the solid measurement curve 10 showing the density values ​​measured for a filter strand with intact capsules and the measurement curve 12 showing the density measurements when the capsules are broken. It can be clearly seen that in the range from 56 mm to about 61-62 mm and from 76 mm onwards the two measurement curves have the same density values. In the intermediate region, significantly higher measurements of up to 170 occur when the capsules are intact. The measurements for broken capsules rise to values ​​just above 140.

[0027] An intact capsule can be reliably identified within the filter strand based on an upper threshold value 14 of approximately 158. The measurements of curve 10 significantly exceed the upper threshold value 14, so that an intact capsule is reliably identified. Figure 1 also shows a lower threshold value 16 having a value just under 140. In the example shown, the signal of a damaged capsule can be identified based on the lower threshold value 16.

[0028] The method shown in Fig. 1 would in principle also be suitable for identifying situations where capsules are missing. Fig. 2 shows the course 18 of a missing capsule in a filter strand. The thresholds 14 and 16 shown in Fig. 1 would also be sufficient to distinguish between a broken capsule and a missing capsule.

[0029] However, the method outlined above has a series of important drawbacks in practice. In particular, when using small capsules, the difference between the signal of a missing capsule and that of a broken capsule is not very large. As a result, these capsule defects cannot be reliably distinguished. Furthermore, when put into practical use, it may also happen that the capsule breaks into various fragments, resulting in various weakly pronounced local maxima in the signal, rather than a maximum. Here, a missing capsule cannot be distinguished from a broken capsule. A particularly important drawback from the measurement point of view is that the absolute value of the measurement signal is used for the comparison of both threshold values. These absolute values ​​vary very significantly depending on the environmental conditions, in particular the value of the moisture content. If the moisture content of the filter tow changes, the absolute values ​​change significantly and the same threshold values ​​cannot be used as for dry filter tow. This makes it impossible to put into practical use a measurement method with two threshold values, which does not allow a reliable distinction between a missing capsule and a broken capsule.

[0030] Figure 3 shows the density signals measured for four capsules located within a filter strand, one for an intact capsule and one for a broken capsule. In this case, the filter rod has already been cut and contains the four individual filters.

[0031] In the method according to the invention, the surrounding area is identified as an area with a capsule defect if the measurement signal falls below the threshold value. The standard deviation of the measurement signal is calculated in the surrounding area, for example within the range of + / - 10 mm, preferably + / - 5 mm. FIG. 4 shows the measurement standard deviation for the measurement signals. In this case, the standard deviation is measured for each signal, regardless of whether the measurement signal is below the threshold value or not. FIG. 4 shows that the standard deviation determined for the signal associated with a broken capsule is significantly larger than the standard deviation of the signal in the case of a missing capsule. By using the capsule threshold value 20 for the standard deviation, two capsule defects can be reliably distinguished from one another. A particular advantage of this is also that the value of the standard deviation does not change with the ambient conditions and / or with the material composition and properties, for example the moisture content of the filter tow. This means that the capsule threshold value 20 can be used uniformly for a wide range of applications. Even with small capsules, two capsule defects can be reliably distinguished from one another.

[0032] Overall, the use of the standard deviation has a series of advantages. Especially for smaller capsules, a larger relative difference can be observed between the signals of broken capsules and missing capsules. Thus, the discrimination of the shadow images can be performed with a higher accuracy. Also, the discrimination between broken capsules and missing capsules does not depend on absolute values, so it can be used regardless of the density and moisture content of the filter tow.

[0033] Figure 5 shows the measured density of the filter tow. Two maxima of density values ​​are shown with signal values ​​22 and 28. The maxima 22 and 28 indicate the presence of intact capsules. The signal value 24, due to its lower value, indicates a broken capsule.

[0034] Particularly if the density of the filter tow of the filter is homogeneous, missing capsules can be easily identified by calculation of the standard deviation with respect to the average value of the density of the filter tow. However, if the density of the filter tow is heterogeneous, the standard deviation increases and is only suitable to a limited extent for detecting missing capsules. In an advantageous development, a line of best fit is determined for the measurement points in the area around the target position by means of linear regression. The line of best fit 26 shows, for example, the trend of the density of the filter tow. The degree of variation of the measured values ​​with respect to the line of best fit is in this case a suitable criterion for distinguishing between broken capsules and missing capsules.

[0035] FIG. 5 shows a filter rod with a length of 100 mm, which should contain four capsules. Two of the capsules at positions 22 and 28 are intact, one capsule at position 24 is broken, and one capsule at position 26 is missing. The density of the filter tow in the region between about 50 mm and 80 mm is inhomogeneous and increases monotonically in this region. The standard deviation of the density relative to the average density will increase in this region, even though there are no broken capsules. The figure also shows a line of best fit 26 around the target capsule position (+ / - 5 mm). The degree of variation of the measurements relative to the line of best fit is low. The reason for this is that if broken capsules are present, a non-monotonic signal deviation will occur, which will lead to a high degree of variation of the measurements relative to the line of best fit.

Claims

1. A method for identifying capsule defects in filters for the tobacco processing industry, comprising the steps of: a microwave measuring device measuring a value in the longitudinal direction of said filter for its density and / or moisture content; The method comprises: a. comparing the measurement to a threshold and if the measurement is below said threshold, a capsule defect is present in said filter; b. determining the standard deviation of said measurements in a region anterior and / or posterior to the measurement location where at least one measurement is below said threshold; c. comparing said standard deviation to a predetermined capsule threshold; i. if the capsule threshold is exceeded, a broken capsule is present; ii. if the capsule threshold is exceeded, then the capsule is missing; Including, A method, characterized in that a line of best fit is determined by linear regression of the measurements in areas anterior and / or posterior to a measurement location where the at least one measurement is below the threshold, and the standard deviation is determined relative to the line of best fit.

2. 2. A method according to claim 1, characterized in that the search for a maximum is carried out in an area in front and / or behind the measurement position, where the at least one measured value is below the threshold value.

3. 3. The method according to claim 1, characterized in that the microwave measuring device is designed as a microwave resonator and measures at least one change in the width of a resonance curve of the microwave resonator and / or a change in the resonance frequency of the microwave resonator.

4. 4. The method according to claim 1, wherein the microwave measuring device is designed as a microwave resonator for measuring density signals and / or moisture content signals.

5. 5. The method according to claim 1 , wherein the standard deviation is determined for a predetermined length of an area in front of and / or behind a measurement location in which the at least one measurement value in the longitudinal direction is below the threshold value.

6. 6. A method according to claim 1, wherein the capsule is expected to be in a specific longitudinal position.

7. 7. A method according to claim 6, characterized in that the standard deviation is determined in each case at regular intervals around the expected position of the capsule.

8. 8. The method according to claim 1, wherein the filter is designed as a longitudinally conveyed filter rod or as a filter strand.

9. 9. The method according to claim 1, wherein the filter is designed as a filter piece for a longitudinally conveyed cigarette.

10. 10. The method according to claim 1, wherein the capsule threshold value is independent of the type of capsule material and / or capsule filling.

11. 11. The method according to claim 1, wherein the capsule threshold is independent of the surrounding filter tow.

12. A device for measuring defects in capsules in filters in the tobacco processing industry, comprising a microwave measuring device and a control device for processing the measurements, a. the microwave measurement device measures a value indicative of density and / or moisture content within the filter, and the control device is designed to compare the measurement to a threshold value and identify a defective capsule if the measurement falls below the threshold value; b. the control device is designed to determine a standard deviation for measurements in areas in front and / or behind the measurement location where at least one measurement is below the threshold and to compare the standard deviation with a capsule threshold; i. if the capsule threshold is exceeded, a broken capsule is identified; ii. if the capsule threshold is exceeded, the capsule is identified as missing; An apparatus, characterized in that a line of best fit is determined by linear regression of the measurements in areas anterior and / or posterior to a measurement location where the at least one measurement is below the threshold, and the standard deviation is determined relative to the line of best fit.

13. 13. Device according to claim 12, characterized in that it is designed as part of a test station for filter rods.

14. 14. Apparatus according to claim 12 or 13, characterized in that it is designed as part of a manufacturing machine for cigarettes or filters.

Images