DE102010021260A1 - Method for quality control of study objects i.e. snacks, involves determining quality characteristics using NMR technique, directly measuring moved sample in production line, and determining inner and outer geometries of study objects - Google Patents

Abstract

The method involves determining quality characteristics using an NMR technique, and directly measuring a moved sample in a production line. Inner and outer geometries of study objects are determined as the quality characteristics. A subpixel-process is used for geometry detection, where the study objects are made of polymer materials such as elastomers, duromers, thermoplastic elastomers, thermoplastics, metallic materials or textile materials. Physical characteristics such as crystallization or orientation are determined by the polymer materials. An independent claim is also included for a system for quality control of study objects manufactured in an extrusion process.

Description

Method and system for determining the geometry, the internal structure, the type of material and the physical properties of products, in particular of rubber, but also of other materials, which are produced in a continuous process, such as extrusion, by means of magnetic resonance tomography directly in the production line ,

field of use

• – die Maßhaltigkeit der Geometrie des Profils, sowohl die Außen- als auch die Innengeometrie betreffend,
• – das Vorhandensein, die Größe und richtige Positionierung verschiedener Materialbereiche oder Inlays,
• – die physikalischen Eigenschaften, wie der Grad der Vernetzung bei Profilen aus Kautschuk an den verschiedenen Stellen im Prozess,
• – die Güte von Mischung und Verteilung der Ausgangsmaterialien.

Rubber profiles are z. B. used in many ways as sealing or insulating materials. They are produced in the continuous extrusion process in sometimes very long production lines. The starting material, the rubber mixture, is conveyed by an extruder and shaped with a tool in the desired geometry. In the next step, the profile goes through the networking unit. The uncrosslinked rubber is plastically deformable. By networking, the profile achieves dimensional stability and the desired elastic properties. Crosslinked rubber is referred to as elastomer (rubber). Thereafter, further refining processes, for. As flocking or painting, done and the profile is made up or rolled up. In order to meet the technical requirements of the product, a material component is often insufficient. Thus, several material components are processed in the coextrusion process to a profile, which consists of several areas of different materials or inserts (so-called "inlays") made of metal, plastic or textiles are used. Essential quality features of such products are

• The dimensional accuracy of the geometry of the profile, both the outer and the inner geometry,
• The presence, size and correct positioning of different material areas or inlays,
• The physical properties, such as the degree of crosslinking of rubber profiles at the various points in the process,
• - the quality of mixing and distribution of the starting materials.

Only accurate, complete and early knowledge of the product quality makes it possible to optimize products and processes and thus reduce the consumption of resources. In this way, start-up times for the processes can be reduced and the influence of batch fluctuations in the starting materials can be minimized. Typical rubber extrusion lines reach lengths of up to 100 m and more and have corresponding information dead times. A complete inline quality control directly in the process is therefore desirable. Other moldable materials, such as in particular numerous foods, are produced in an extrusion process or a similar continuous process, which is why similar questions arise in these fields as in plastic and rubber extrusion.

State of the art

Today, optical methods based on light-section technology or structured illumination are used for geometry detection and dimensional accuracy monitoring directly in the production line. This technique is z. In the inventions DE 103 28 537 A1 and DE 10 2004 052 508 BV used. It is only possible to detect what can also be detected in the optical beam path. All internal structures of the profile and undercuts remain hidden and can not be detected and determined.

On the "XVII IMEKO World Congress - Metrology for a Sustainable Development", 17.-22. September 2006 in Rio de Janeiro, Brazil, by R. Anchini, G. Di Leo, C. Liguori and A. Paolillo discloses a method with which the inner and outer geometry can be checked at the end of the extrusion line. In this case, a stereo camera system after the assembly unit is attached, which receives the tee edges and misses. The system must therefore be integrated as a last link in the production chain. Any erroneous developments in the process are discovered very late. Even with this approach, it is not possible to identify areas of different properties or materials, as they often differ in their brightness in the camera image hardly or only edge information are evaluated.

The among others B. Blümich: NMR Imaging of Materials, Oxford: Oxford University Press, 2000 known nuclear spin resonance technique (English nuclear magnetic resonance, NMR) It can be used to measure components and to check for presence, to determine the size and correct positioning of different material areas or inlays. Furthermore, physical properties of a sample can be determined. The NMR is based on the magnetic properties of the atomic nuclei of different materials and tissues. Cores with magnetic moment (spin), such as. As hydrogen, align themselves when they are exposed to a strong external magnetic field B ₀ . If they are simultaneously excited by high-frequency electromagnetic waves (HF waves) whose frequency, the so-called "Larmor frequency", depends on the field strength of the magnetic field B ₀ , they absorb energy. In a subsequent relaxation phase, this energy is released again. This spins answer to the suggestion provides a wealth of information about the local magnetic field experienced by each nucleus. They form the basis both for the elucidation of the molecular structure and for the spatially resolved imaging (magnetic resonance imaging, magnetic resonance imaging, MRI). The measured signal amplitude reflects the density of the nuclei present in the sample. The initial magnetization of the samples is described by the relaxation time T ₁ . The relaxation time T ₂ describes the decay of a coherent magnetization caused by the excitation with the rf-waves. Both constants depend on the molecular dynamics and order of the material to be investigated. They provide information about the interaction of the nuclei with their environment.

The invention US 7397241 and US 2004/0058559 A1 describe methods by means of which the proportion of one or more material components in a sample can be determined by means of NMR. They can be used to control the quality of mixtures and to determine the proportions of different material components in a profile. However, the methods are intended for static, non-moving samples and there is no way to measure moving samples directly in a production line.

The invention US 5371464 and US 5519319 describe methods for the determination of physical parameters of polymeric samples, such. B. the degree of crosslinking in rubbers. Both methods, however, are intended for static, stationary samples. No moving samples are measured directly in the production line, but samples are taken and measured while they are at rest.

Disadvantages of the prior art

• – die Maßhaltigkeit der Geometrie des Profils, sowohl die Außen- als auch die Innengeometrie betreffend,
• – das Vorhandensein, die Größe und richtige Positionierung verschiedener Materialbereiche oder Inlays,
• – physikalische Eigenschaften, wie den Grad der Vernetzung bei Profilen aus Kautschuk an den verschiedenen Stellen im Prozess,
• – die Güte von Mischung und Verteilung der Ausgangsmaterialien.

None of the methods listed in the prior art makes it possible to determine all the quality features listed below with a measuring system directly on the production line, even on moving extrudates:

• The dimensional accuracy of the geometry of the profile, both the outer and the inner geometry,
• The presence, size and correct positioning of different material areas or inlays,
• Physical properties, such as the degree of cross-linking of rubber profiles at different points in the process,
• - the quality of mixing and distribution of the starting materials.

Purpose and advantages of the invention

The invention describes a method and system which does not have the disadvantages of the prior art solutions and fulfills the task of determining the quality features described above directly in the production line. The developed process can be used in industrial environments at typical operating temperatures (up to more than 200 ° C) and speeds (take-off speeds up to more than 30 m / min). It can be integrated at any point in the production line.

Solution of the task

The object is achieved by a method and system having the features of claims 1 to 23. The solution of the problem will become apparent from the following description of the method. 1 Fig. 12 schematically illustrates an extrusion line and shows how the method and system of the invention are integrated therein. The material is at the end of the extruder ( 1 ) through a tool to a profile ( 2 ) formed. Thereafter, further processing steps ( 4 ), such. As flocking, painting, packaging or in the case of rubber profiles, the crosslinking of the rubber. The quality control measuring system according to the invention can be installed at any point on the extrusion line. It consists of an NMR sensor and a control, operating and evaluation unit. The latter can consist of a computer or an electronic module.

The NMR sensor of the measuring system ( 3 ) has a passage, the z. B. is cylindrical, through which the extrudate profile ( 2 ) is performed as an examination object. The NMR sensor is based on the known NMR technique and consists of the following essential components: With a strong magnet, the magnetic field B _{0 is} generated, which is oriented perpendicular to the extrusion direction. So-called RF coils serve as antennas for excitation of the sample and for measuring the impulse response emitted by the excited sample. A spectrometer with a corresponding amplifier is used to control the coils. In NMR imaging, the signal is localized via a location-dependent magnetic field, which is superimposed on the magnetic field B ₀ . It is generated with a gradient coil system. The gradient coils are driven by a special imaging module, which is integrated into the spectrometer. For control, evaluation and display, the spectrometer is connected to a computer.

As a magnet with a cylindrical passage, a magnet with a cylindrical Halbachgeometrie or a modified Halbachgeometrie be used ( B. Blümich: NMR Imaging of Materials, Oxford: Oxford University Press, 2000 ).

The process steps are shown in a flowchart in FIG 2 clarified. In the first step (A), an image of the internal structure is generated with the NMR sensor. In this case, either the relaxation time T ₁ , the relaxation time T ₂ or the proton density can be determined spatially resolved and displayed as an image. Since the measurement object, the extrudate, moves, the mapping of the averaging corresponds to the information over a certain length of the sample. The considered length depends on the examined material.

In the next step (B), the inner and outer geometry of the examination subject is determined from the determined image data by means of segmentation algorithms known from digital image processing (see, for example, FIG. C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 ). Relevant areas and dimensions can be set and checked by the user. The information about the geometry extracted from the images can be compared (step (D)) with information about the target geometry stored in a database. The time required for image acquisition depends significantly on the desired nominal resolution. It should be so short ("1 s) that the measured values of a moving sample are only averaged over a very short distance. An inspection system that determines the geometry of the profile inline must determine the position of the edges with an accuracy in the range between 0.1 mm and 0.5 mm. A higher resolution can be achieved by a longer measurement time. When capturing edges (for example, between rubber and air or between different materials), it can be assumed that the "true" location of the edge is never exactly between two pixels. The edges are thus inevitably displayed as a gray value curve. This "true" position of the edge can be achieved with so-called "subpixel" methods ( C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 ) can be computationally reconstructed with an order of magnitude higher accuracy than the nominal resolution of the NMR system. Subpixel methods are based on interpolation of the measured data. The application of the subpixel methods for detecting the edges in the image to the NMR image data constitutes an essential component and advantage of the invention, since it first enables work to be done so quickly and simultaneously with practical resolution that the system is integrated into an extrusion line can be.

In a further step (D), in the case of coextrudates, the area of the individual material regions can be determined with the aid of segmentation methods known from image processing, such as, for example, threshold value segmentation (FIG. C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 ) are determined from the image data. With this knowledge, it is possible to calculate the volume flow of the individual extruder. The determination of the volume flows is an essential prerequisite for the use of the invention for the regulation of the participating extrusion plants (step (E)). This makes it possible for the first time to directly determine the performance of the extruder online and to use this information for quality control and regulation.

Furthermore, physical parameters of the sample can be determined in a spatially resolved manner in step (F) from the NMR parameters T ₁ , T ₂ measured in step (A) and the proton density. For example, the local density of a profile can be determined. This value can be used in foamed profiles to determine the density of foamed areas. By evaluating the relaxation times T ₁ and T ₂ , it is possible to _draw conclusions about physical parameters, such as the degree of crosslinking of rubbers or stresses in the material. Different materials and mixtures can be differentiated spatially resolved with the aid of the measured parameters. Such parameters, in particular the degree of crosslinking for rubber extrusion lines, may be described in step (G) for controlling or controlling an extrusion plant or components of the plant, such as, for example, B. the power of the crosslinking unit can be used.

Advantages of the invention

The method outlined above solves the disadvantages of the prior art and fulfills the objects of the invention, which represents the advantages of the invention.

Description of exemplary embodiments

An embodiment of a method according to the invention for determining the inner geometry is based on 3 to be discribed. With the aid of a system according to the invention, a commercially available rubber profile is examined. The profile was produced in a coextrusion process and consists of two different components.

A picture with 64 × 64 pixels is made, which makes a statement about a profile of the profile which is 10 mm thick in the extrusion direction ( 3a ). The field of view of the measurement is 26 mm × 26 mm, so that the nominal resolution of the measurement, ie the size of a pixel, is 0.4 mm × 0.4 mm. In a next step, the image is interpolated using a standard interpolation method to 512 × 512 pixels. In 3b the contour of the geometry of the profile determined by the method according to the invention is shown. These positions are marked as edges whose gray value in the interpolated image exceeds 50% of the maximum gray value. By using the interpolation method, the position of the edge can be determined with a much higher accuracy than the nominal size of a pixel. Even this simple subpixel method therefore brings considerable advantages in terms of measurement accuracy and measurement duration. In the literature on digital image processing, more complex methods for subpixel accurate edge and geometry detection can be found. Their use can further improve performance over the example method used. Furthermore, in the inclusion in 3a to distinguish the different material areas clearly on the brightness of the pixels.

LIST OF REFERENCE NUMBERS

1

Extruder

2

Messobjekt, kontinuierlich hergestelltes Produkt

3

erfindungsgemäßes Messsystem

4

weitere Verarbeitungsschritte

Fig. 2

A

Verfahrensschritt A

B

Verfahrensschritt B

C

Verfahrensschritt C

D

Verfahrensschritt D

E

Verfahrensschritt E

F

Verfahrensschritt F

G

Verfahrensschritt G

Fig. 1

1

extruder

2

Test object, continuously produced product

3

Measuring system according to the invention

4

further processing steps

Fig. 2

A

Process step A

B

Process step B

C

Process step C

D

Process step D

e

Process step E

F

Process step F

G

Process step G

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 10328537 A1 [0004]
• DE 102004052508 B [0004]
• US 7397241 [0007]
• US 2004/0058559 A1 [0007]
• US 5371464 [0008]
• US 5519319 [0008]

Cited non-patent literature

• "XVII IMEKO World Congress - Metrology for a Sustainable Development", 17.-22. September 2006 in Rio de Janeiro, Brazil, by R. Anchini, G. Di Leo, C. Liguori and A. Paolillo [0005]
• B. Blümich: NMR Imaging of Materials, Oxford: Oxford University Press, 2000 known nuclear magnetic resonance (NMR) technique [0006]
• B. Blümich: NMR Imaging of Materials, Oxford: Oxford University Press, 2000. [0013]
• C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 [0015]
• C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 [0015]
• C. Demand, B. Streicher-Abel, P. Waszkiewitz, "Industrial Image Processing. How optical quality control really works ". Berlin: Springer, 2001 [0016]

Claims (23)

Method for quality control of objects to be inspected in a continuous continuous process, such as the extrusion process, characterized in that the quality features are determined using the NMR technique and the measurement takes place inline, ie directly in the production line. A method according to claim 1, characterized in that as quality features, the inner and the outer geometry of the object to be examined is determined. Method according to claims 1 and 2, characterized in that subpixel methods are used for geometry detection. Method according to claims 1 and 2, characterized in that the geometry contour determined by the invention only at certain points metrologically and then the geometry is determined by software algorithms by interpolation algorithms. Method according to one of claims 1 to 4, characterized in that the geometry can be compared punctually or completely with a stored in a database or predetermined by the user target geometry. A method according to claim 1, characterized in that as a quality feature in examination objects, which consist of several different components, such. As coextrudates, or objects under investigation with inlays, the contour and area or the relative proportion of the individual components is determined. Method according to claims 1 and 6, characterized in that the information determined by the method according to claim 6 for the regulation and control of an extrusion plant, in particular for throughput control, are used. A method according to claim 1, characterized in that for objects of investigation which consist partially or completely of cross-linking materials, the degree of crosslinking is determined directly or indirectly by correlation with other measuring methods by means of the NMR technique used. A method according to claim 1, characterized in that in the case of examination objects which consist of a plurality of different components, the invention determines physical properties such as the degree of crosslinking or the mixing quality for the individual components. A method according to claim 1, characterized in that the examination objects, which consist of polymeric materials, physical properties such as crystallization or orientations are determined by the invention. A method according to claim 1, characterized in that in the case of examination objects which consist of several different components, the invention determines quality features such as the number and type of imperfections, flaw structure or bubbles and bladder structure in foamed components. A method according to claim 8 or 9, characterized in that the determined properties for controlling and controlling an extrusion plant, in particular for controlling the performance of a crosslinking unit is used in crosslinking materials. Method according to one of claims 1 to 12, characterized in that it can be used in the temperature range -30 ° C to 300 ° C. Method according to one of claims 1 to 13, characterized in that the examination objects are extrudates of polymeric materials, in particular elastomers, thermosets, thermoplastic elastomers, thermoplastics or blends. Method according to one of claims 1 to 13, characterized in that the objects under investigation are foods, in particular doughs or expanded foods, such as. B. snacks. Method according to one of claims 1 to 13, characterized in that the examination objects are coextrudates of several components of polymeric materials or have inlays of polymeric, textile or metallic materials. System for quality control of objects of investigation produced in a continuous continuous process, such as the extrusion process, characterized in that the quality characteristics are determined by a process having the quality characteristics characterized in claims 1 to 16. System according to claim 17, characterized in that it consists of two subsystems, the control and operating system and an NMR sensor system. System according to claims 17 and 18, characterized in that the NMR sensor system consists of a magnet, RF coils, gradient coils, corresponding amplifier technology and a spectrometer. System according to claims 17 to 19, characterized in that the magnet has a passage through which the examination object is guided, and permanent magnets are used to generate the magnetic field. System according to claims 17 to 19, characterized in that the magnet has a passage through which the object under examination is guided, and for generating the magnetic field permanent magnets are used, which generate a magnetic field perpendicular to the direction of movement of the object to be examined. System according to claims 17 to 20, characterized in that it can be used both in mobile and in stationary facilities. System according to claims 17 to 21, characterized in that the control and operating system and the NMR sensor system can be present both together and spatially separated.

Images