EP0914597A1

EP0914597A1 - Method and device for sampling dispersed streams of material

Info

Publication number: EP0914597A1
Application number: EP98925623A
Authority: EP
Inventors: Wolfgang Witt; Stephan Röthele
Original assignee: Sympatec GmbH
Current assignee: Sympatec GmbH
Priority date: 1997-05-20
Filing date: 1998-05-20
Publication date: 1999-05-12
Also published as: US6357305B1; KR20000029455A; DE19721104A1; WO1998053295A1; JP2000515637A

Abstract

The invention relates to a method for sampling dispersed streams of material. According to said method, a partial stream for sampling is separated from the main process stream for subsequent analysis, via a sampling surface. Said sampling surface is smaller than the surface passed over by the main process and is fixed independently thereof. Whilst the partial stream is being separated from the main process stream, the sampling surface scans the surface over which the main process stream has passed, along a flow curve. The invention also relates to a device for carrying out the inventive method.

Description

Method and device for sampling from disperse material flows

The invention relates to a method for sampling from disperse material streams, in which a partial analysis stream is taken from this main process stream for subsequent analysis in a main process stream and a device for carrying out this method.

The size distribution of the particles or droplets, hereinafter referred to as particles, is of great importance for the production of disperse solids and emulsions, since they essentially determine the reactivity, the transport properties and the stability of the material system.

Knowledge of the particle size distribution (TGV) therefore makes it possible to optimize the production process and produce it to a desired quality. For this purpose there is a device for determining the particle size distribution necessary, which can be easily integrated into the process at various points without noticeably disrupting the process flow.

Devices for determining particle size distributions with various measuring principles have long been known. Devices that are based on laser diffraction (LB) increasingly dominate in many applications, since they combine high measurement accuracy with good stability of the results, short measurement times, large measurement range, low measurement range lower limit and simple handling.

These devices take advantage of the fact that a particle illuminated by monochromatic, coherent light, depending on its size, deflects parts of this light to different extents, small particles deflecting the light more than large particles. This deflection of light is known as diffraction.

In a conventional arrangement according to FIG. 1, a laser 1, followed by an expansion optics 2, generates an extensive parallel measuring light beam which illuminates the particles 8 introduced in a measuring cell 7. The diffracted light is imaged by means of a converging lens 14, the Fourier optics, onto a photodetector 16 with a multiplicity of elements which, together with a downstream electronics, allows the intensity distribution to be measured precisely. The particle size distribution can be calculated from this intensity distribution by means of an evaluation unit 20 using known algorithms.

Such a device is equally suitable for determining the particle size distribution of disperse solids, suspensions and emulsions. Known calculation methods based on Fraunhofer diffraction provide particle size distribution regardless of the optical properties of the particles and the surrounding medium. The applicability of these devices is, however, limited to the range of low particle concentration, since work is preferably carried out in transmitted light and the measuring zone must let the diffracted light pass. In addition, the particle concentration should be chosen so low that the diffracted light is not again diffracted on subsequent particles. This process, known as multiple scattering, can be taken into account when calculating the particle size distribution. The known algorithms are limited to specific particle shapes and require precise knowledge of the optical parameters of the material system, which is not the case for most particles.

The high mass flows in the range of a few tons per hour, which are common in production processes, therefore generally make it necessary to first take a sample from the process, and to lower the particle concentration of this sample by adding the medium surrounding the particles, i. H. to be diluted so that the maximum permissible particle concentration in the measuring zone is not exceeded. This method is only permissible for those material systems in which the dilution does not change the particle size distribution.

The sample is to be taken in such a way that the particle size distribution of the sample corresponds to the particle size distribution of the process in the period under consideration, ie is representative. This in turn requires that all areas of the transport cross-section are recorded with equal weight and that the taking of the sample does not change the particle size distribution at the location of the sample name. If the process particles are integrated in a flowing medium, it is known that the sample can be taken isokinetically, ie that the particles must not experience any change in speed, since otherwise the particle size distribution removed would be noticeably changed. It is also known from laser diffraction that agglomerated particles are determined with their agglomerate diameter. For common tasks, one is preferably interested in the particle size distribution of the primary particles. The agglomerated particles must first be separated from one another before they pass through the measuring zone. To solve this problem known as dispersion, various devices are known which separate dry particles in high-turbulence flows by collisions of the particles with one another, collisions of the particles with the walls or obstacles introduced specifically for this purpose or by centrifugal forces as a result of speed gradients. Devices are known for suspensions in which a liquid is used to separate the particles, partly with the support of special chemical substances and the supply of ultrasound.

To avoid fluctuations in the optical properties of the components, e.g. To eliminate the laser and the coating of the Fourier optics with particles from the calculated particle size distribution, a reference measurement is required from time to time, in which the intensity distribution of the medium surrounding the particles is measured without the presence of particles.

Various devices have now been proposed for the determination of particle size distributions in the process, which meet these requirements only in part, and mostly only with considerable restrictions.

In the simplest known version, the laser diffraction system is flanged directly to the process tube, ie the entire process mass flow has to pass through the measuring zone. The high optical concentrations that occur are taken into account by means of a material-dependent correction of the multiple scatter. A reference measurement is only possible before the start of a production phase (batch). The light source and detector are kept largely clean using gas-flushed tubes or an enveloping flow along windows. Any contamination of the optics which occurs nevertheless falsifies after permanently the result. The particles are not dispersed. Such a device is therefore only suitable for very small pipe diameters with production mass flows in the range of a few kg / h and with short production times (batch times). The analyzed sample is non-representatively defined by the geometry of the measuring zone and the laser beam profile.

In another device, the light source and detector unit are integrated in a rod with an opening transverse to the rod direction, which is immersed in the process mass flow through a flange. The limitations of the aforementioned design are expanded to include non-representative sampling, with this device keeping the windows clean is particularly problematic.

In a further development, the first-mentioned version is operated with a static, non-representative sampling in the bypass to a pipe with a larger cross-section. Sampling takes place from a fixed, definable position of the process tube. The sample is transported by means of a jet pump, which further dilutes the partial analysis flow and is set so that the sampling is carried out as iso-kinetically. The sampling tube is permanently open and exposed to the abrasive process mass flow. No cleaning of the sampling opening is provided. The partial analysis stream continuously passes through the measuring zone. It is accelerated by the jet pump and redirected several times until it is returned to the process. 90 ° bends are used to save overall height. This arrangement is unfavorable in terms of wear. The wear is proportional to the process mass flow and the process time and regardless of the number of required measurements. The reference measurement can only take place at the beginning of a batch.

In a further embodiment, the speed of the process mass flow is initially determined by installing special flow geometries via differential measure the reference pressure sensors and use these values to adjust the pressure conditions in the partial analysis flow to ensure the isokinetics of the sampling. The particle size distribution analysis is carried out by a laser diffraction system attached to the outlet of the tube, in one embodiment the particles are first separated on a filter and when a certain amount of sample is reached, this is measured with a laboratory laser diffraction system. Here too, the sampling location is static, ie not representative of the overall cross-section. The actual particle size distribution analysis takes place outside the pipeline.

Such a sampling stage has also been proposed for the case of flow-through tubes with isokinetic sampling. This in turn results in large partial flows of analysis, which make complex subsequent stages necessary. The actual particle size distribution analysis also takes place outside the pipeline.

In order to improve the sampling situation, DE 35 43 758 C1 proposed that a sampling segment rotate in the cross section of the process tube. In the case of negligible flow, a representative sampling can be guaranteed in the chute. The low division ratio generally requires repeated application of the principle, which, like the need for seals between moving parts in a particle-laden environment, is disadvantageous. The actual particle size distribution analysis takes place outside the pipeline. For this purpose, the partial analysis stream is first collected in a buffer, then transported to the measuring system, whereby the measuring range can be adjusted by screening and blockages of the subsequent stages can be avoided. This was followed by metering into a dry disperser, measuring in a laser diffraction sensor, suctioning off the aerosol and, if appropriate, returning it to the process via cyclone and cellular wheel sluice. This solution proved to be extremely powerful, but at the same time large and complex.

In the area of suspensions, representative solutions have so far been omitted for all solutions. In the usual devices used for this purpose, the sampling takes place via a static tube protruding into the process tube or a pneumatically extendable sampling finger. The sample is conveyed using a dilution liquid. The dispersion is omitted entirely or takes place in a stirred container using ultrasound. The diluted suspension is finally passed through a cuvette into the measuring zone of a laser diffraction system and disposed of externally or returned to the process. For reference measurement, controlled valves switch to pure liquid.

Finally, devices have been realized in which other available sampling systems, such as. B. impact samplers, Y-valves, sampling screws etc. deliver the sample and these are prepared and measured with standard laser diffraction systems. All these devices have in common that they are only suitable for certain special cases, require frequent maintenance and take up a lot of space.

None of the previously known devices can be used universally or made ^¬ light combines a representative sampling of moving media Dis ^¬ pergierung, concentration adjustment, a compact construction and the possibility of the reference measurement process is running.

The object of the invention is therefore to provide a sampling method that is as universal as possible and a compact device for carrying out the method. A representative, continuous sampling should be possible even with high production mass flows. According to the invention, this object is achieved by a method according to the preamble of claim 1, in which the partial analysis stream is withdrawn from the main process stream via an extraction surface which is smaller than the area through which the main process stream flows and is fixed independently of it. The constant sampling area sweeps the area through which the main process stream flows during the sampling of the partial analysis stream along a path curve.

This sampling method according to the invention is distinguished from the method known according to the prior art by a representative and continuous sampling, which can also adapt the analysis flows to high production mass flows.

If one starts with the solution according to DE 35 43 758 C1, as shown in FIG. 2a, for a mass flow Θ (mass flow M / area) in a pipe, the certain values cannot exceed high Mass flows require a large pipe diameter D. In the solution according to DE 35 43 758 C1 with a rotating sampling segment, this inevitably leads to high analysis mass flows m, since the segment angle α is limited at the bottom by the maximum particle size or by a multiple of this particle size. The following applies:

m = —— M = - - Θ i „„ = a D ² = —Θ —- π =

2π 4

Ie the analysis mass flow m increases with the square of the process pipe diameter D. Fig. 2b shows that the known solution with a sampling tube is considerably cheaper here. The following applies: d ² md ¹

Λ ^~ T ^π m = - M = ^ M =

d ² d ² D ²

= ^ ® A _{? Τmß} = —® — π ^

= ^ Θ d ² 4

That the partial analysis flow is now independent of the process tube diameter and proportional to the square of the selectable diameter of the sampling tube d.

Isokinetic sampling with a static sampling tube is not representative. This problem is now solved by the invention in such a way that the sampling tube traverses the cross-section of the process tube at a constant speed so that the entire cross-sectional area or a representative partial area is covered n times during a measurement period. The path curve should be selected so that all areas are traversed only once during a run.

Preferred embodiments of the method according to the invention result from claims 2-9.

Particularly advantageous devices for carrying out the invention result from claims 10-22.

Finally, the devices according to the invention can be used in a particularly advantageous manner for various measuring methods, as claimed in claims 23-26. Devices for determining particle sizes and / or particle size distributions result from claims 27-35.

Further details and advantages of the invention are explained in more detail with reference to exemplary embodiments shown in the drawings. Show it:

1: a laser diffraction system according to the prior art,

2: various sampling geometries for representative sampling according to the prior art via a segment (FIG. 2a) or a sampling tube (FIG. 2b),

3: the spiral path covered by a sampling tube according to the invention,

4: a basic sketch of an embodiment of the sampling tube according to the invention,

5 shows an embodiment of the sampling tube according to the invention with bellows and shield,

6: an embodiment according to the invention with a sampling tube in the park position (with pressure ball seal),

7: a spiral path of the sampling opening in the sampling tube, which begins and ends in the parking position.

8: a drive for a sampling tube according to an embodiment of the invention via cardan suspension and push rods, 9: a preferred embodiment of the drive for the sampling tube according to the invention,

10: a sketch for calculating the equation of motion of the opening of the sampling tube,

11: a side view of an entire measuring device with sampling, dispersion, dilution, particle size distribution determination by means of laser diffraction and feedback stage,

Fig. 12: Side view of the device of Fig. 11 in a slightly modified embodiment, rotated by 90 ° and

FIG. 13: a top view of the device according to FIG. 11, from above onto the sampling tube.

FIG. 4 schematically shows a sampling tube 30 which is mounted in the middle of a process tube 32 and is directed at a shallow angle Θ against the product stream (not shown here). The opening 34 of the sampling tube 30 runs along a spiral path, as is shown in FIG. 3 and is represented by the following equations:

x (τ) = r (τ) sin (<s (t) τ) y (τ) = r (τ) cos (ω (t) τ)

D τ τ r (r - _J - - D-

2

τ (t) is to be chosen as a function of time so that the web speed v (t) is constant. The following applies approximately: w. w (t) = r (τ)

With

D: Process tube diameter n: Number of circumferences for the drive to the center t _p : Total time for sampling

To avoid sampling errors, it is necessary to aspirate the partial analysis stream isokinetically. The particle velocity in pipes is generally a function of the radius r measured from the center of the process pipe. In a preferred device according to the invention, the negative pressure in the sampling tube is changed via empirically determined values so that the sampling takes place isokinetically, independently of r.

The sampling tube is exposed to the process mass flow and thus extreme wear. This applies in particular at high particle speeds, e.g. B. in pneumatic conveying. To minimize wear, internals should not have any surfaces perpendicular to the direction of flow. It is also important to limit deflections of the partial analysis flow to small angles in order to minimize wear there too. Finally, all components of the sampling tube should be made of hard metal, ceramic or other wear-resistant materials. For cost reasons, it is necessary to keep the distances as short as possible. Transporting the sample outside the process pipe leads to long pipe lengths and large overall heights if you want to avoid a 90 ° angle.

If one considers the opening 34 of the sampling tube 32, then this describes a spiral path with the path speed v. Since the projection area of the entrance opening changes with cos (θ), this has to be done in order to achieve a representative sampling over the web speed

v (θ) = v ₀ cos (0)

be balanced.

This design does not require the sampling tube to rotate with respect to the process tube. It can therefore be connected without seals using elastic walls. In a preferred embodiment, the connection is made via a metal bellows 36 which, according to FIG. 5, is mechanically protected from the particles of the process mass flow by means of a shield 38.

The occasional reference measurement makes it necessary for the medium surrounding the particles to be guided through the measuring zone without particles. In the known devices, controlled valves and / or controlled dosing devices are used, which allow the particle flow to be interrupted for the duration of the reference measurement. In the proposed sampling system, it is sufficient to withdraw the sampling opening from the product task and / or to close the sampling opening or to connect it to the medium surrounding the particles. The sampling opening must also be cleaned from time to time of caking and blockages caused by coarse particles, fibers, etc. In addition, the sampling opening should not be exposed to the process mass flow in the non-measuring phases in order to minimize wear overall.

In the embodiment shown in FIG. 6, a protected parking position 40 for the sampling tube is provided on the inside of the process tube 32 provided that it provides low resistance to the flowing particles, but at the same time seals the sampling tube 30 securely (e.g. by means of a pressure ball seal 42 according to FIG. 6) or connects it to a connection for the particle-free medium. An integrated scraper ensures cleaning when the pipe opening enters the parking position.

In one embodiment, several parking positions can be distributed on the inside of the process tube so that the sampling opening can be moved from one parking position to another. Using e.g. B. from linear scans over the pipe center, the dwell time of the sampling opening in the process mass flow and thus the wear can be further reduced with limited representation.

In a preferred embodiment according to the invention, the trajectory of the sampling opening is described by a spiral which begins at the parking position on the process tube wall, approaches the center in a spiral and finally ends again on the continued spiral path in the parking position (FIG. 7). Such a trajectory can be created in different ways.

For this purpose, in the device according to FIG. 8, the sampling tube 30 is gimbaled in the middle of the tube. Controlled deflection of the two gimbals 44, 46 shown in FIG. 8 tilts the pipe by two perpendicular angles θ and φ with respect to the pipe axis. Through a controlled chronological sequence of the tilting, the sampling opening can follow any given path curve. The deflection can e.g. B. done by push rods or hydraulic rams. The drive can be from outside the tube z. B. done by stepper motors. When using push rods, a simple lever mechanism can be used to deflect the force at 90 °. In the embodiment shown in FIG. 9, the spiral path is derived from a simple rotary movement. For this purpose, the sampling tube 30 is supported at one end in an axis 47 running perpendicular to the tube direction. The sampling tube 30 is tilted via a push rod 48, which is articulated via a bearing 50 to a lever connected to the sampling tube 30. The articulation axis 50 is parallel to the axis 47. With its other end, the push rod 48 is articulated via a bearing 49 with a threaded nut 52. The threaded nut 52 runs on a thread 54 which is firmly connected to a discharge tube 56. Via a linkage 58 which can be set in rotation, the sampling tube 30 together with the push rod 48 and the threaded nut 52 can be set in rotation with respect to the discharge tube 56 and the thread 54 rigidly connected to it via a ring gear 60. During this rotation, the threaded nut 52 moves on the fixed thread 54 and thus adjusts the angle Θ of the sampling tube 30 with respect to the tube axis via the push rod 48. The sampling opening rotates around the process tube axis and thus describes a spiral path.

The deflection of the sampling opening r from the center of the process tube as a function of the deflection x of the bearing from the position (x ₀ ) at which the sampling tube is in the center of the process tube describes the equation:

L is the length of the sampling tube. The almost linear relationship shown in FIG. 10 results for L = 168.5 mm, H = 10 mm and I = 79.2 mm. The path curve in FIG. 6 was calculated with these parameters for a thread pitch of 1 mm / revolution. It begins in the parking position on the edge and reaches the edge again via the center without having to change the direction of rotation of the motor. For repeated loading it is sufficient to change the direction of rotation of the motor in the park position, ie at standstill. This can e.g. B. be accomplished via a simple sensor that detects the entry of the sampling tube into the parking position, whereupon a controller stops the drive and reverses the direction of rotation of the drive for a new scan. 10 shows a sketch for calculating the aforementioned equation of motion.

The proposed sampling according to the invention thus has clear advantages over known devices and methods: a) Sampling takes place only during the measuring period, otherwise the sampling opening in the parking position is protected from the process mass flow. The wear of the sampling, the dispersion section and the measuring system is reduced by orders of magnitude, b) A reference measurement is possible at any time in the park position while the process is running. No further measures are necessary for this, c) The drive manages without seals to the process mass flow and only moves small masses. Only a low drive power is required, d) Sampling is continuous, representative of time and place. Due to the large diameter ratio D / d, the flow is only slightly affected by sampling. The partial analysis flow is independent of the process tube diameter and can easily be adapted to the requirements of the measuring system via the size of the opening of the sampling tube. The integrated cleaning mechanism makes an upstream safety screening unnecessary.

The reduction in the free cross-section in the process pipe due to the necessary internals can be compensated for by increasing the pipe diameter below the level of the sampling opening. An influence on the sampling can be largely excluded.

Various methods are known for dispersing disperse solids in gas or liquids. One uses for the dispersion in gases Particle-wall, particle-particle impacts and / or centrifugal forces, such as z. B. occur as a result of velocity gradients in a shear flow. Z serves as an energy supplier. B. a jet pump. Ultrasound is often used to disperse suspensions. In the preferred embodiment according to FIG. 11, a jet pump 62 is mounted coaxially inside the process tube 63 at the outlet of the sampler. By supplying propellant gas or suspension liquid, the vacuum required to transport the particles is created in the sampling tube. The pre-pressure of the propellant of the jet pump 62 is selected so that an isokinetic condition on the sampling tube 30 arises via empirically determined parameters. The medium supplied dilutes the mass flow of analysis. The energy supplied is used for dispersion.

In a preferred embodiment according to the invention shown in FIG. 12, a jet guide tube 70 with a round cross section connects for solid aerosols. The cross section is dimensioned so narrow that there are as many particle-particle and particle-wall collisions as possible. This is followed by a transition tube 72, in which the cross section increases continuously from round to rectangular. Finally, there follows a beam guide tube 74 with a rectangular cross section, which extends to just before the measuring zone 76. The rectangular cross-section prevents a lens effect of the gas jet cooling during expansion in the measuring zone.

In a preferred embodiment of the invention, blockages in the sampling tube can also be eliminated by temporarily closing the outlet of the jet pump. This can be done, for example, by moving the beam guide tube. The beam guide tube is moved until its wall closes the outlet opening almost completely. The blowing agent then escapes through the sampling tube and rinses it free. In a preferred embodiment according to the invention, a cross-sectional widening with a static mixer follows into which the ultrasonic sonotrode projects. The sonotrode is connected to the ultrasonic vibrator attached outside the process tube via a cladding tube. An air layer between the sonotrode and the cladding tube prevents direct sound transmission into the process suspension. The cross-sectional widening reduces the particle speed and increases the exposure time of the ultrasound. The static mixer evens out the diluted suspension.

For optical methods for determining particle size distribution, such as laser diffraction or image processing, the optical concentration Copt must lie within certain limit values. For monodisperse particles with diameter x, the C _opt for the preferred embodiment according to the invention is calculated as follows:

Here, d describes the diameter of the sampling opening, D the diameter of the process tube, / the extinction coefficient, m the process mass flow, b the width of the beam guide tube, p the specific density and v the speed of the particles in the measuring zone. The relationship can be described in a similar way for non-monodisperse particles.

11, the speed of the particles in a second jet pump 66 is increased such that the particles are preferably in the middle of the measuring zone and necessary windows 68 are not contaminated. The medium supplied dilutes the partial analysis flow and changes the particle speed v. The optical concentration can thus be adjusted statically via D, d and b, dynamically via v. In the device, the particle size distribution is measured after the dilution step using known methods, e.g. B. by means of laser diffraction or image processing. The components required for this are passed through one or more cladding tubes from the outside through the process tube to the dilution stage. The cladding tubes are tightly connected to the cladding tube of the dilution stage. The optical components of the measuring system are protected from the particles of the analysis mass flow by optical windows.

In the device described here, the partial analysis flow is returned via the cladding tube 64 immediately after the measuring zones in the direction of the tube axis. The outlet is open. The coaxial structure and the alignment along the process tube axis is particularly favorable compared to the known methods of returning via lateral connections, since the fast particles only hit the process tube wall at very shallow angles and the wear on this wall is thereby considerably reduced. Since the particles remain inside the process tube, no additional measures are required when the process tube is under pressure. There is no mechanism for return transport (pump).

In a further embodiment according to the invention there is a valve at the outlet, e.g. B. a pinch valve that closes the outlet when necessary and z. B. prevents steam, liquid or solid used for cleaning the process tube from entering the measuring zone.

In addition to optical methods such as laser diffraction or image processing, ultrasonic extinction (UE) plays an increasing role in determining the particle size distribution.

Ultrasonic extinction determines the attenuation of a sound wave for different frequencies. Analogous to laser diffraction, the particle collective is measured here in transmission, ie that originating from a sound transmitter Sound wave penetrates the measuring zone and reaches a shell receiver weakened. From the measured attenuation, the size distribution of the particles can be calculated using known algorithms.

These algorithms specify the frequency range in relation to the particle size to be determined. In practical use, you have to deal with very high frequencies in the range of 100 kHz to a few 100 MHz, which can only be coupled into liquids with sufficient efficiency. Ultrasound absorbance is therefore limited to determining the size distribution of particles in liquids.

In contrast to laser diffraction, optical transparency does not play a role and multiple scattering and dispersion play a very minor role. It can therefore be measured at very high particle concentrations without dilution and without dispersion. This is an advantage for material systems where dilution would change the size distribution of the particles (e.g. in crystallization).

The damping of the sound wave by the liquid limits especially at high frequencies, i. H. fine particles, the maximum possible distance between sound transmitter and receiver to a few millimeters. As a result, the short distance limits the possible volume flow that can pass through the measuring zone.

A reference measurement carried out from time to time on the particle-free liquid also serves to improve the stability of the results.

The available calculation methods also require knowledge of the akusti ^¬ specific parameters of the fuel system, in particular the extinction function. The ^¬ se can be determined for a group of similar material systems such as comparison with other measurement methods. Ultrasonic extinction is therefore particularly suitable for processes in which dilution would change the size distribution and product change is rare.

In a further preferred device according to the invention, the described sampling is combined with an ultrasound extinction system. In order to maintain the advantage of ultrasonic extinction, to be able to measure particle size distribution without dilution without dispersion, the measuring system is arranged immediately after the sampling stage. The samples are transported via a downstream, regulated pump. It can be z. B. act as a jet pump when the introduction of liquid into the process is harmless. For simple applications, the pressure difference can also be used, which results from the process pipe being constricted before the return outlet.

Claims

Method and device for sampling from disperse material flows

1. A method for sampling from disperse material streams, in which a partial analysis stream is taken from a main process stream for subsequent analysis,

characterized,

that the analysis sub-stream is withdrawn from the main process stream via an extraction area which is smaller than the area through which the main process stream flows and which is defined independently of it, and that the extraction area sweeps over the area through which the main process stream flows during the extraction of the analysis sub-stream along a trajectory curve.

2. The method according to claim 1, characterized in that the removal surface is the circular flap of the mouth of a sampling tube.

3. The method according to claim 1 or 2, characterized in that the main process stream flows through a pipeline.

4. The method according to any one of claims 1-3, characterized in that the constant sampling area is moved on a trajectory over the area through which the main process flow flows that the swept locations contribute equally to the partial analysis stream during a sampling cycle.

5. The method according to any one of claims 1-4, characterized in that the constant removal surface is guided along a spiral path along the surface through which the main process flow flows.

6. The method according to any one of claims 1-5, characterized in that the entry surface of the constant removal surface is closed or partitioned off when the sampling is to be interrupted.

7. The method according to claim 5 or 6, characterized in that the partitioning of the constant removal area is carried out by moving the constant removal area into a parking position which is designed so that it effectively prevents the entry of parts from the main process flow into the sampling tube.

8. The method according to any one of claims 1-7, characterized in that the suction speed of the partial analysis stream is adjustable.

9. The method according to claim 8, characterized in that the partial analysis stream is sucked off at a speed which is equal to the speed in the vicinity of the sampling opening (isokinetic suction).

10. A device for performing a method according to any one of claims 1-9, characterized in that a movable sampling tube of the flow direction of the disperse material flow is arranged in the opposite direction within the main process stream.

11. The device according to claim 10, characterized in that the sampling tube is connected to a suction device in such a way that the suction speed can be adjusted at the extraction surface defined by the opening of the sampling tube.

12. The apparatus of claim 10 or 11, characterized in that the sampling tube is arranged in the middle of a tube through which the main process stream flows.

13. The device according to one of claims 10-12, characterized in that a plurality of parking positions are formed on the tube such that they prevent access to parts of the main process flow.

14. The apparatus according to claim 13, characterized in that a cleaning device is arranged in a parking position, which cleans the sampling opening during the retraction and extension of the sampling tube in the parking position.

15. The apparatus of claim 13 or claim 14, characterized in that the sampling tube in the park position can be connected via a valve with a particle-free flow medium.

16. The device according to any one of claims 10-15, characterized in that the sampling tube is rotatably and pivotably mounted with respect to the process tube axis and can be forcibly pivoted coupled via a push rod during the rotary movement.

17. The apparatus according to claim 16, characterized in that the forcibly coupled pivoting movement of the push rod can be generated via a threaded nut screwed onto a thread firmly connected to the discharge tube during the rotational movement of the sampling tube.

18. Device according to one of claims 10-15, characterized in that the sampling tube is mounted in the process tube over two gimbals and can be deflected by time-controlled pivoting of the gimbals.

19. Device according to one of claims 12-18, characterized in that the sampling tube on its path position, which begins at the parking position and ends at the same parking position, can be moved continuously without reversing the direction of rotation of the motor and that only in repeated operation from the parking direction the direction of rotation of the motor must be reversed.

20. Device according to one of claims 10-19, characterized in that the gear for pivoting the sampling tube over movable walls, for example bellows, can be locked without a seal with respect to the process tube.

21. Device according to one of claims 1-20, characterized in that the opening of the sampling tube (removal surface) is smaller than the inner diameter of the sampling tube to avoid blockages.

22. Device according to one of claims 1-21, characterized in that the opening of the sampling tube (sampling surface) can be adapted to the measurement requirements by means of interchangeable caps which can be plugged onto the sampling tube.

23. Use of a device according to one of claims 10-22 in a method for determining particle sizes and / or particle size distributions in pipelines based on laser diffraction, image processing and / or ultrasound extinction, the sampling device, the sample transport, the measurement zone and the sample return within the tube are arranged parallel to the tube axis and a partial flow is removed by sampling, which passes through the device shielded to the remaining process mass flow and is immediately reunited with the process mass flow after the sample return.

24. Use according to claim 23, characterized in that in addition a dispersion within the tube is arranged parallel to the tube axis.

25. Use according to claim 23 or 24, characterized in that in a method for determining particle sizes and / or particle size distributions in pipelines based on laser diffraction and / or image processing, the sampling device, the sample transport, the measuring zone and the sample return in the order mentioned are arranged one after the other within the tube.

26. Use according to claim 23 or 24, characterized in that in a method for determining the particle sizes and / or particle size distributions in pipelines based on the ultrasonic extinction, the sampling device, the measuring zone, the sample transport and the sample return in the order mentioned in succession within the Pipe are arranged parallel to the pipe axis.

27. Device for determining particle sizes and / or particle size distributions containing a device for sampling from disperse material flows according to one of claims 10-22, characterized in that both the sampling device, the sample transport, the dispersion, the measuring zone and the sample return in this order within a pipe are arranged.

28. The apparatus according to claim 27, characterized in that the measuring method is based on laser diffraction, ultrasonic extinction or image analysis.

29. The device according to claim 27 or 28, characterized in that the sample is transported by a jet pump.

30. Device according to one of claims 27-29, characterized in that the dispersion in liquid systems takes place by means of an ultrasonic finger, which is insulated from the process tube via an air layer.

31. The device according to any one of claims 27-30, characterized in that for solid aerosols, a jet guide tube with a round cross section connects to the sampling tube, the cross section being so narrow that there are as many particle particles or particle wall impacts as possible that a transition pipe connects to it, in which the cross section converges tinuieriich enlarged from round to rectangular and that a beam guide tube with a rectangular cross-section adjoins that reaches just before the measuring zone.

32. Device according to one of claims 27-31, characterized in that, preferably in the case of laser diffraction, the dispersion path can be briefly closed by pivoting in front of the outlet opening of the jet pump, and so a reverse flow direction within the sampling tube can be generated to remove blockages.

33. Device according to one of claims 27-32, characterized in that the return of the partial analysis stream via a cladding tube takes place immediately afterwards to the measuring zones in the direction of the tube axis.

34. Device according to one of claims 27-33, characterized in that a valve, in particular a pinch valve, is arranged at the outlet of the device.

35. Device according to one of claims 27-34, characterized in that a second jet pump is connected downstream of the measuring zone, via which the speed of the particles to be measured in the measuring zone is increased such that the particles are preferably in the middle of the measuring zone and thus contribute to suppressing the contamination of the measuring window.