Classification device

Abstract

Classification device comprising a plurality of blades arranged at suitable intervals in the circumferential direction and arranged radially or eccentrically, and a rotor with classification chambers between the blades, by which fine particles in a fluid are classified such that, as a fluid flows into the classification chambers from the outer circumferential side to the inner circumferential side, particles larger than a separation particle size are moved to the outer circumferential side and particles smaller than the separation particle size are moved to the inner circumferential side, characterized in that the blades have a constant height in the direction of rotation of the rotor and the thickness increases towards the outer circumference in the circumferential direction, wherein the thickness t(d) of the blades at the position of the diameter d of the classification chambers can be determined by means of the following equation 15: t ( d ) = 1 N [ π d − QT × 1 D 1 2 × 2 × 894 d ⋅ n 2 × 18 η 9.8 ( ρ 2 − ρ 1 ) ] where Q: the flow rate, N: the number of classification chambers in the circumferential direction, D1: the particle size, n: the rotational speed of the rotor, η: the viscosity of the fluid, ρ1: the specific gravity of the fluid, ρ2: the specific gravity of the particles and T is the height (constant) of the blades.

Description

[Technical field]

[0001] The present invention relates to a device that classifies fine particles in a gas or a slurry. [Background technology]

[0002] There are dry classification devices that have a rotor 2, in which, as in Fig. Figure 1 shows blades 1 arranged circumferentially at uniform intervals, radially from the center of rotation or eccentrically from the center of rotation, and classifying fine particles in the air by rotating the rotor 2 at high speed, as well as moisture classification devices for classifying fine particles in a slurry.

[0003] Fig.Figure 2 shows the schematic overall structure of a classification system comprising a dry classification device 3, which includes the rotor 2 internally. A material feed device 5 supplies the classification device 3 with material along with air. The material is classified into coarse and fine material by the rotor 2, which rotates at high speed. The coarse material is discharged from the classification device 3 and collected in a container 6, while the fine material passes through an outlet chamber 8 surrounding a drive shaft 7 coupled to the rotor 2 and flows into a bag filter 11. In the bag filter 11, the fine material is separated from the air and collected in a container 12. An example of such a dry classification device is disclosed in the following patent document 2.

[0004] Fig.Figure 3 shows the schematic overall structure of a classification system comprising a wet classification device 14. A material slurry is fed to the classification device 14 from a slurry tank 15 by means of a slurry pump 16. The material slurry is classified into a coarse material slurry and a fine material slurry by a rotor 17 rotating at high speed. The coarse material slurry is discharged from the machine by the classification device 14, while the fine material slurry passes through a hollow drive shaft 18 coupled to a rotor 17 and is collected in a tank 19. An example of such a wet classification device is disclosed in the following patent document 1.

[0005] The classification at the in Fig.The classification of particles in the rotor 2 shown in Figure 1 is achieved by the fact that, while a gaseous substance or a slurry (hereinafter referred to as "fluid") flows into the interior of the rotor 2 and moves towards the inner circumferential side, the particles in the fluid are subjected to a centrifugal force due to the rapid rotation of the rotor 2 and to a reaction due to the fluid flowing in the inner circumferential direction opposite to the direction of action of the centrifugal force. By means of a particle size that balances both forces, the particles are classified into coarse material with a large diameter and fine material with a small diameter. This is explained using examples of particles 10 placed in arbitrary classification chambers 9 between the blades 1 of the rotor 2 shown in Figure 1. Fig. 1. Flow into rotor 2 as shown, which rotates at high speed.

[0006] At the position of diameter d of the classification chambers 9, the particles 10 are subject to a centrifugal force F acting diametrically outwards, and, due to the flow of the fluid towards the inner circumferential side, to an opposing effect due to the flow resistance R, acting in the opposite direction to the centrifugal force F, assuming that the particles 10 are round, their diameter is given as D, the rotational speed of the rotor 2 as n, the specific gravity of the fluid as ρ1, the specific gravity of the particles 10 as ρ2 and the gravitational acceleration as g, the centrifugal force F can be expressed by the following equation 1. F=16πD3⋅g(ρ2−ρ1)⋅0.001118n2⋅(d / 2)=16πDa⋅9.8(ρ2−ρ1)⋅d⋅n22×894

[0007] On the other hand, the flow resistance R is expressed using Stokes' theorem by the following equation 2, given the viscosity as η and the inward linear velocity of the fluid as s: R=3π⋅D⋅η⋅s

[0008] The linear velocity s can be determined by the following equation 3, where the arc area on the circumference of the circle is at the point of the diameter d. Fig. The classification chambers 9 shown in 1 (hereinafter referred to only as the circular arc area) are expressed as A, the number of classification chambers 9 in the circumferential direction as N and the flow rate of the inwardly directed fluid as Q. s=QA⋅N

[0009] Since the circular arc area A shown in equation 3 results from the multiplication of the length of the circular arc on the circumference at the position of the diameter d and the length (height) of the rotor's axis of rotation, and since the classification chambers are present as a plurality, namely N ≥ 1, the circular arc and the chord length of the circular arc are small, so that this is an approximation of the product of the chord length in the form of the cross-sectional area at the chord and the length (height) of the rotor's axis of rotation. Therefore, in the present description, both are considered essentially identical.Similarly, with regard to the thickness of the blades in the circumferential direction at the position of the diameter d (hereinafter referred to as "thickness of the blades") and the chord, as well as the gap, the arc length and the chord length are approximated as the circular arc length of the circumferential direction between the blades at the position of the diameter d (hereinafter referred to as "gap between blades") and the chord, so that in the present description both are treated as essentially identical.

[0010] The separation particle size D1 of the particles 10, at which the centrifugal force F = the flow resistance R at the position of the diameter d of the classification chambers 9, can be determined by the following equation 4 using the above equations 1 to 3. D1=QN⋅1A⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1)

[0011] At the position of the diameter d of the classification chambers 9, the classification is based on the separation grain size D1 of the grain size at which the centrifugal force F and the flow resistance R are balanced, where particles where R > F are particles with a smaller grain size than the separation grain size D1, which move towards the inner circumferential side, while particles where R < F are particles with a larger grain size than the separation grain size D1, so that they move towards the outside of the diameter direction.

[0012] JP 2002 - 143 707 A concerns the provision of a wet stirred ball mill which improves the separation performance of balls and a drive unit, i.e. the drive force required to crush a slurry of a unit weight.

[0013] In the device for separating foreign substances of JP S 53 - 73 668 A, several drive blades are provided in the lower and upper parts of a rotating element for rotating the element by receiving the liquid and gas injected from corresponding injection tubes, as well as guide blades for directing the foreign substances located on the rotating element downwards. Between the upper guide blades, a section with a large number of holes and a section without holes are provided. A perforated plate is arranged on the rotating element, and a separate liquid outlet and a gas outlet are provided at suitable locations on the main body of the perforated plate.

[0014] JP S 62 - 201 679 A discloses a classification device for powdered raw material with a raw material feed section provided in a lower part of a classification section and a fine powder outlet section provided above the classification section to separate powdered raw material into coarse powder and fine powder.

[0015] In JP 2010 - 253 394 A, a classification mechanism is provided which prevents particles from being forced to the surface of an inner housing wall and retained there by a vortex flow generated on the outer circumference of a classification rotor, in a particle classification mechanism in which a basket-shaped classification rotor is installed in the housing.

[0016] JP H 06 - 190 343 A discloses a material separation device for classifying material discharged from a grinding zone into a fine part and an oversized part, with an inlet for receiving the material to be ground and an outlet for discharging the ground material, wherein the airflow velocity and the rotational speed of rotor blades are adjusted so that particles of the desired size pass through the rotor and are collected in a centrifugal separator.

[0017] JP H 10 - 277 490 A deals with the problem of carrying out a timely inspection of the wear of a fine powder conveying section in a classifier. [Previous state documents][Patent documents] [Patent document 1] JP Patent Publication 2002-143707 A [Patent document 2] JP Patent Publication 2011-072993 A [Summary of the invention][Technical problem]

[0018] In the conventional device disclosed in patent document 1, as shown by equation 4, the separation particle size D1 is converted to a linear velocity s = Q / (AN), G = (d·n) expressed by equation 3. 2 The function of the centrifugal effect G and the diameter d, expressed as ) / (2×984), increases with increasing linear velocity s and decreasing centrifugal effect G and diameter d. In the conventional device, the height and thickness of the rotor blades are constant in the diametrical direction, so the circular arc area decreases towards the inner circumference, and according to equation 3, the linear velocity increases. Furthermore, the centrifugal effect G decreases as the diameter d decreases.

[0019] In the classification device disclosed in patent document 2, under conditions of constant blade thickness and circular arc area in the diametric direction of the classification chambers, the blade height is increased towards the inner circumference by setting the linear velocity. However, the increase in blade height is insufficient to compensate for the decrease in the centrifugal effect G due to the reduction in diameter d. Even when the linear velocity s, expressed as Q / (AN) in equation 4, is set, an increase in the separation particle size D1 due to the diameter d is unavoidable. That is to say, while the classification device disclosed in patent document 2 can somewhat reduce the growth rate of the separation particle size compared to a conventional device, it cannot prevent an increase in the separation particle size altogether.

[0020] As explained above, although the separation particle size D1 increases towards the inner circumference in the respective classification devices, if coarse particles larger than the separation particle size D1 fall in due to a turbulent flow of the fluid on the outside of the high-speed rotating rotor, they reach the inner circumference by mixing in on the inner side if the particle size difference to the separation particle size is small, and are thus easily collected.

[0021] The separation particle size D1 is determined, as previously stated, using equation 4, but it can also be expressed as follows: With the height of the blades 1 in relation to the leaf surface in Fig. Given 1 orthogonal direction as T, the space between blades as E and the thickness of the blades 1 as t, the circular arc area A at the position of the diameter d is given by A=E⋅T=(πd−tN)TN expressed, where the separation particle size D1 is due to equation 5 and equation 4. D1=Q(πd−tN)T⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1) becomes.

[0022] If the flow rate Q, the number N of classification chambers, the height T of the blades 1, the thickness t of the blades 1 and the separation particle size D1 are specified as setpoints and the viscosity η and the specific gravity ρ1 of the fluid as well as the specific gravity ρ2 of the particles 10 are specified as constants, the separation particle size D1 becomes a function of the diameter d, such that, as can be seen from equation 6, the separation particle size D1 becomes smaller as the diameter of the classification chamber 9 increases, and conversely, the separation particle size D1 becomes larger as the diameter d decreases.

[0023] Next, a simulation was performed to calculate the extent to which the particle size D1 changes with respect to the diametric direction of the rotor 20 for the settings in Table 1 below. The result is shown in Table 2 below, along with the centrifugal effect G, the circular arc area A, the gap E between the blades, and the linear velocity s as the fluid's velocity towards the inner circumference of the rotor. [Table 1] SETTINGS DIAMETER POSITION FLOW RATE SHOVEL HEIGHT SHOVEL THICKNESS NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific particle weight Specific fluid weight d Q T t N n η ρ2 ρ1 m m 3 / s m m - rpm kg / m·s kg / m 3 kg / m 3 0.40 0.00001 0.01 0.005 12 2500 0.001 2300 1000 0.35 0.00001 0.01 0.005 12 2500 0.001 2300 1000 0.30 0.00001 0.01 0.005 12 2500 0.001 2300 1000 0.25 0.00001 0.01 0.005 12 2500 0.001 2300 1000 0.20 0.00001 0.01 0.005 12 2500 0.001 2300 1000 [Table 2] CALCULATION VALUES DIAMETER POSITION SEPARATOR SIZE CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES LINEAR VELOCITY d D1 G A E s m µm - m2 m m / s 0.40 0.92 1398 0.00100 0.100 0.00084 0.35 1.05 1223 0.00087 0.087 0.00096 0.30 1.24 1049 0.00074 0.074 0.00113 0.25 1.49 874 0.00060 0.060 0.00138 0.20 1.89 699 0.00047 0.047 0.00176

[0024] Patent document 2 discloses a classification method in which classification is carried out using a rotor with a constant circular arc area A, which is obtained by increasing the height of the blades as the radius decreases, and with a constant linear velocity s.

[0025] Regarding the separation particle size D1 using this method, the blade height T and the separation particle size D1 were determined in a simulation using equations 5 and 6. The same settings as in Table 1 were used for the rotor speed n, the viscosity η, the flow rate Q, the specific gravity ρ1 of the fluid, the specific gravity ρ2 of the particles, the blade thickness t, and the number N of classification chambers 9. Additionally, the linear velocity s was set to 0.00084 m / s at the outer circumference of the classification chamber, i.e., at the diameter position of 0.40 m in Table 2. The gap E between the blades is calculated using equation 5 as (πd-tN) / N. The result is shown in Table 3. [Table 3] CALCULATION VALUES DIAMETER POSITION FLOW RATE CIRCLE AREA INTERSPACE INTERSPASHES SHOVEL HEIGHT LINEAR VELOCITY SEPARATOR SIZE d Q A E T s D1 m Nm 3 / s m2 m m m / sec µm 0.40 0.00001 0.00100 0.100 0.0100 0.00084 0.92 0.35 0.00001 0.00100 0.087 0.0115 0.00084 0.98 0.30 0.00001 0.00100 0.074 0.0135 0.00084 1.06 0.25 0.00001 0.00100 0.060 0.0167 0.00084 1.16 0.20 0.00001 0.00100 0.047 0.0213 0.00084 1.30

[0026] In Table 3, the blade height T at the gap E between the blades, which is determined by the position of the diameter d, was determined by substituting the constant circular arc area A and the gap E between the blades shown in Table 3 into equation 5. The linear velocity s was determined using equation 3, and the separation grain size D1 was determined by substituting the relevant items from Table 1 into equation 6 in a simulation calculation.

[0027] In the conventional device shown in patent document 1, as can be seen in Table 2, both the separation particle size D1 and the linear velocity s increase towards the inner circumference. In the classification device shown in patent document 2, even with a constant setting of the linear velocity s, the decreasing linear velocity s towards the inner circumference (dn) 2) / (2×894) expressed centrifugal effect G, the separation particle size D1 as can be seen from Table 3, remains unchanged even if the growth rate decreases.

[0028] The present invention aims to provide a classifying device which, in comparison to the aforementioned conventional devices, achieves a sharp particle size distribution with low mixing of coarse material and high classification precision. [Means of solving the problem]

[0029] According to the present invention, in the device for classifying fine particles in a fluid, in which a plurality of blades spaced apart in the circumferential direction and arranged radially or eccentrically, and a rotor with classification chambers between the blades are provided, while a fluid flowing into the classification chambers from the outer circumferential side to the inner circumferential side, particles larger than a separation particle size are moved to the outer circumferential side and particles smaller than the separation particle size are moved to the inner circumferential side and are classified in such a way that the separation particle size becomes uniform in the entire radial area from the outer circumference to the inner circumference of the classification chambers.

[0030] As a method for determining the separation particle size in the radial direction, there are the following three methods, whereby in the present invention it is necessary for a decreasing centrifugal effect G that, to supplement this, the linear velocity s decreases towards the inner circumference, and to realize this is an increase in the circular arc area A directed towards the inner circumference.

[0031] In the first method, the blade thickness is constant in the diametric direction, and the blade height becomes slightly greater towards the inner circumference than in the subject matter disclosed in patent document 2. The second method, like the conventional device disclosed in patent document 1, has a constant blade height and a slightly smaller blade thickness towards the inner circumference. The third method is a combination of the first and second methods, i.e., a method in which the blade height is increased towards the inner circumference and the blade thickness is decreased. [Effects of the invention]

[0032] According to the present invention, by maintaining a constant particle size distribution along the entire radial direction from the outer to the inner circumference of the classification chambers, fine material at or below the particle size is moved towards the inner circumference while being classified over the entire area from the outer to the inner circumference. Even if coarse material larger than the particle size falls in and penetrates, it is easily discharged to the outer circumference because the classification process takes place over the entire area of ​​the classification chamber. This makes it difficult for coarse material to mix in, and consequently, coarse material is hardly mixed into the collected product. This results in a product with a sharp particle size distribution and high classification performance. [Brief explanation of the characters] [ Fig. 1] Cross-section of a rotor forming the classification device. [ Fig.2] Schematic representation showing the overall structure of a system that includes a dry classification device. [ Fig. 3] Schematic representation showing the overall structure of a system that includes a wet classification device. [ Fig. 4] Longitudinal section of the rotor of an embodiment according to the present invention. [ Fig. 5] Sectional view of the rotor in Fig. 4 by AA. [ Fig. 6] Longitudinal section of the rotor of another embodiment. [ Fig. 7] Sectional view of the rotor in Fig. 6 by BB. [ Fig. 8] Cross-section of a modification of the rotor in Fig. 7. [ Fig. 9] Figure showing the rotor used in the classification device of embodiment 1 and its size. [ Fig.10] Figure showing the rotor used in the classification device of Comparative Example 1 and its size. [ Fig.

[11] Figure showing the grain size distribution of the material and the grain size distribution of the particles after classification using the in Fig. 9 and Fig. The classification device shown in Figure 10 represents the classification device. [ Fig. 12] Figure showing the rotor used in the classification device of embodiment 2 and its size. [ Fig. 13] Figure showing the rotor used in the classification device of Comparative Example 2 and its size. [ Fig.

[14] Figure showing the grain size distribution of the material and the grain size distribution of the particles after classification using the in Fig. 12 and Fig. The classification device shown in 13 represents the classification device. [Emphasis of the invention]

[0033] The classification device of the present embodiment comprises a rotor in which blades are arranged radially at uniform intervals around the center of rotation in the circumferential direction, or blades are arranged at uniform intervals around the center of rotation eccentrically to the center of rotation, and the spaces between the blades are provided as classification chambers, wherein one of the preceding methods 1 to 3 is applied in the device classifying fine particles in a fluid by rotating the rotor at high speed, and the classification is carried out in such a way that the particle size is constant over the entire radial direction from the outer circumference to the inner circumference of the classification chambers, wherein the construction of the classification device used in the respective methods is explained below.

[0034] Fig. 4 and Fig.Figure 5 shows a rotor of the classification device used for the first method, wherein Fig. 4 a longitudinal section of the rotor 23 is and Fig. 5 a cut through line AA in Fig. 4 shows, where the one in Fig. The rotor 23 shown in Figure 5 has the same cross-sectional area as the one in Figure 5. Fig. Rotor 2 shown in 1 has a different longitudinal section surface.

[0035] The separation particle size D1 for rotor 23 is determined as described above. That is, with regard to the separation particle size D1, at the position of the diameter d of the in Fig. 5 of the rotors shown 23 the centrifugal force F and the flow resistance R are balanced, the height of the in Fig. The equation 6 is derived from the following: 5. The blades 21 shown in the direction orthogonal to the blade surface, expressed as a function of the diameter d by T(d). D1=Q(πd−tN)T(d)⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1) where, at the position of diameter d, as described above, particles with a larger grain size than the separation grain size D1 are discharged in the diametrically outward direction, and particles with a smaller grain size than the separation grain size D1 are moved in the diametrically inward direction. In equation 7, Q, N, t, A, n, η, ρ1, and ρ2 are as in equation 6, where Q is the flow rate, N is the number of classification chambers, n is the rotational speed of the rotor 23, η is the viscosity, ρ1 is the specific gravity of the fluid, t is the thickness of the blades, and ρ2 is the specific gravity of the particles contained in the fluid.

[0036] The circular arc area A(d) of the classification chambers 22, expressed as a function of the diameter d, is given by the space between the blades as E(d) and the height of the blades as T(d) at the position of the diameter d. A(d)=E(d)⋅T(d)=(πd−tN)T(d)N expressed, and can therefore be determined by equation 7 and equation 8 using the following equation 9: A(d)=QN⋅1D12⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1)

[0037] To ensure that the separation particle size D1 remains constant, the flow rate Q, the number of chambers of the classification chambers 22, the rotational speed n of the rotor, the viscosity η of the fluid, the specific gravity ρ1 and ρ2 of the fluid and the particles are given as setpoint values ​​with respect to the circular arc area A(d) at the position of the diameter d and are constant, so that the circular arc area A(d) in equation 9 becomes A(d)=C / d, a function of the diameter d and inversely proportional to the diameter d.

[0038] Here, C is the constant expressed in the following equation. C=QN⋅1D12⋅2×894n2⋅18η9.8(ρ2−ρ1)

[0039] Using equation 8 and equation 9, the height T(d) of the classification blades 21 is determined as a function of the diameter d by the following equation 11: T(d)=Qπd−tN⋅1D12⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1)

[0040] The blade height T(d), which ensures that the separation particle size D1 remains constant throughout the classification chamber from the outer to the inner circumference, is determined using equation 11. Equation 11 expresses the blade height T(d) as a function of the diameter d, decreasing as the diameter d increases towards the outer circumference and increasing as the diameter d decreases. Consequently, rotor 3, as shown in Fig.Figure 4 shows a cross-sectional shape with an extended inner circumference. The space E(d) between the blades is expressed using equation 8 as E(d) = (πd-tN) / N, and increases proportionally to the diameter d towards the outer circumference ( Fig. 5).

[0041] Using the rotor 23 of the present embodiment, in which, as described above, the blade height T increases towards the inner circumferential side, the flow rate Q, the blade thickness t, the number of classification chambers N, the rotor speed n, the viscosity η, the specific gravity ρ1 of the fluid, and the specific gravity ρ2 of the particles were set as in Table 1, and the particle size D1 was set to the value in Table 4 below. Using equations 3, 8, 9, and 11, the linear velocity s at the diameter d position, the arc area A(d), the gap E(d) between the blades, and the blade height T(d) were calculated in a simulation. The calculation result is shown in Table 5 below.To adapt to the separation particle size D1 of the rotor outer circumference at the diameter position 0.40 m, the minimum value of the separation particle sizes D1 shown in Table 2, which were determined by inserting the relevant items shown in Table 1 into equation 6 in a simulation calculation, the separation particle size D1 was set to 0.92 µm in Table 4. [Table 4] SETTINGS DIAMETER POSITION FLOW RATE SHOVEL THICKNESS NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific fish partridge e-weight Specific fluid weight SEPARATOR SIZE d Q t N n η ρ2 ρ1 D1 m Nm 3 / s m - rpm kg / ms kg / m 3 kg / m 3 µm 0.40 0.00001 0.005 12 2500 0.001 2300 1000 0.92 0.35 0.00001 0.005 12 2500 0.001 2300 1000 0.92 0.30 0.00001 0.005 12 2500 0.001 2300 1000 0.92 0.25 0.00001 0.005 12 2500 0.001 2300 1000 0.92 0.20 0.00001 0.005 12 2500 0.001 2300 1000 0.92 [Table 5] CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES SHOVEL HEIGHT LINEAR VELOCITY d G A(d) E(d) T(d) s m m2 m m m / s 0.40 1398 0.00100 0.100 0.0100 0.00084 0.35 1223 0.00114 0.096 0.0118 0.00073 0.30 1049 0.00133 0.073 0.0182 0.00063 0.25 874 0.00159 0.061 0.0261 0.00052 0.20 699 0.00199 0.047 0.0424 0.00042

[0042] Equation 11 of the above embodiment determines the height T(d) of the blades for a constant separation grain size D1 in the diametric direction in the classification chambers.

[0043] Fig. 6 and Fig. Figure 7 shows the rotor of the classification device used for the second method, wherein Fig. 6 a longitudinal section of the rotor 25 is and Fig. 7 a cut through line BB in Fig. 6 shows. As in Fig.As shown in Figure 6, the height T of the blades 26 in the diametric direction of the rotor 25 is constant, and the thickness t(d) of the blades in the peripheral direction changes such that it behaves as shown in Figure 6. Fig. Figure 7 shows that the diameter of the blades 26 increases slightly from the inner to the outer circumference, and the classification chambers 27 are formed with a widening towards the inner circumference. The thickness t(d) of the blades 26 at the inner circumference need not necessarily be zero, although zero is preferred. If the thickness of the blades 26 at the inner circumference is zero, the diameter of the blades 26 at the inner circumference can be reduced and the length of the blades increased in the diametric direction, thereby enabling sufficient classification.

[0044] In one embodiment for carrying out the second method, the circular arc area A(d) at the position of the diameter d is defined as E(d) by means of the spaces between the blades in the circumferential direction. A(d)=E(d)⋅T expressed, wherein the gaps E(d) between the blades at the position of the diameter d are determined using equation 9 and equation 12, which are comparative expressions of the diameter d and the circular arc area A(d), through which the separation grain size D1 becomes constant, and by means of equation 13 E(d)=QT⋅N⋅1D12⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1)

[0045] The circular arc area A(d) and the space between the blades at the position of the diameter d, which apart from the blade height T, which is constant at 0.0381 m as shown in Table 6 below, and the thickness t(d) of the blades 26 in the circumferential direction, which was determined by means of the following equation 15, using the same setpoints as in Table 4 by means of equation 12 and equation 13 in a simulation calculation, are shown together with the thickness t(d) of the blades in the circumferential direction, the centrifugal effect G and the linear velocity s in the following Table 6.

[0046] The blade height T was set to 0.0381 m to achieve the required particle size D1 = 0.92 µm, as well as the flow rate Q in Table 4. With a blade thickness t(d) = 0 at diameter d = 0.20 m and a classification chamber number N = 12, E(d) = 0.052 m is calculated using E(d) = πd / N. Substituting this E(d) value and the corresponding entries in Table 4 into Equation 13, the blade height T at diameter d = 0.20 m is determined to be 0.0381 m. The blade thickness t(d) at the position of diameter d is determined using Equation 14 below. t(d)=πdN−E(d)

[0047] If equation 13 is substituted into equation 14 for E(d), the thickness t(d) of the blades 26 is given by the following equation 15, whereby the blade thickness t(d) is determined by a simulation calculation in which the relevant items of table 4 are substituted into equation 15, where the separation grain size D1 is constant 0.9 µm from the outer circumference to the inner circumference of the classification chambers. t(d)=1N[πd−QT×1D12×2×894d⋅n2×18η9.8(ρ2−ρ1)] [Table 6] SETTINGS CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES SHOVEL HEIGHT LINEAR VELOCITY SHOVEL THICKNESS d G A(d) E(d) T s t(d) m - m2 m m m / s m 0.40 1398 0.00100 0.026 0.0381 0.00084 0.0785 0.35 1223 0.00114 0.030 0.0381 0.00073 0.0617 0.30 1049 0.00133 0.035 0.0381 0.00063 0.0436 0.25 874 0.00160 0.042 0.0381 0.00052 0.0236 0.20 699 0.00199 0.052 0.0381 0.00042 0.0000

[0048] The rotor in Fig. 7 The classification chambers 27 taper towards the outer circumference and are radially shaped, but they can also be arranged as in the rotor 31 in Fig. 8 shovel 33, the surface of which increases towards the outside and eccentrically formed tapered classification chambers 32.

[0049] The rotor in a further embodiment of a rotor for carrying out the third method described above is made from a combination of the components described in Fig. 4 shown rotors 23 and rotors 25 and 31 in Fig. 7 and Fig. 8 formed. That is, the height of the shovels increases, as in Fig. 4 shown, gradually increasing towards the inner circumference, so that the inner circumferential side of the rotor is widened, and the thickness of the blades increases, as shown in Fig. 7 and Fig. 8 shown, towards the outer circumference side, so that the classification chambers are wider towards the inner circumference side.

[0050] In this embodiment, the circular arc area A(d) at the position of the diameter d is defined by means of the spaces E(d) between the blades in the circumferential direction and the height of the blades as T(d). A(d)=E(d)⋅T(d)=(πd−t(d)N)T(d)N expressed, where the thickness t(d) of the blades 26, which is expressed as a function of the diameter d, is given by the following equation 17, where T of equation 15 = T(d), t(d)=1N[πd−QT(d)×1D12×2×894d⋅n2×18η9.8(ρ2−ρ1)] determined, and where by substituting equation 17 into equation 16 the height T(d) is obtained T(d)=QE(d)⋅N⋅1D12⋅2×894d⋅n2⋅18η9.8(ρ2−ρ1) becomes.

[0051] Furthermore, the space E(d) in equation 18 is determined by the following equation 19. E(d)=πN⋅{b⋅d2−b⋅d2−a⋅d1d2−d1×(d2−d)}

[0052] In Equation 19, d1 is the inner circumference diameter and d2 is the outer circumference diameter of the classification chambers, a is the gap coefficient between the inner circumferential blades in the form of (πd1-Nt1) / πd1, and b is the gap coefficient between the outer circumferential blades in the form of (πd2-Nt2) / πd2. Furthermore, t1 is the thickness of the blades 26 at the inner circumferential end and t2 is the thickness at the inner circumferential end. It follows that the difference between the circumferential gap of the inner circumferential diameter d1 and the circumferential gap of the outer circumferential diameter d2 can be expressed by π(bd2-ad1) / N. Any diameter d between the diameter d2 and the diameter d1 can be determined by the following equation 20 by dividing this difference proportionally by (d2-d) / (d2-d1), and the preceding equation 19 is obtained from term 20. π(b d2−ad1)N⋅d2−dd2−d1

[0053] The thickness t(d) of the blades 26 is expressed as t(d) ={πd-N·E(d)} / N, so that t(d) is obtained from the following equation 21 when the equation 19 is substituted into E(d). t(d)=πdN−πN⋅{b⋅d2−b⋅d2−a⋅d1d2−d1×(d2−d)}

[0054] Using the same settings as in Table 4, except for the blade thickness, with a set to 1 and b to 0.8 respectively, and using the space between the blades E(d) at the position of the diameter d determined by Equation 19, values ​​were calculated in a simulation with respect to the blade height T(d) by Equation 18 and the thickness t(d) of the blades by Equation 21, which together with the circular arc area A(d) and the centrifugal effect G are shown in the following Table 7. [Table 7] CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES SHOVEL HEIGHT LINEAR VELOCITY SHOVEL THICKNESS d G A(d) E(d) T(d) s t(d) m - m2 m m m / s m 0.40 1398 0.00100 0.084 0.0119 0.00084 0.0209 0.35 1223 0.00114 0.076 0.0150 0.00073 0.0157 0.30 1049 0.00133 0.068 0.0195 0.00063 0.0105 0.25 874 0.00159 0.060 0.0265 0.00052 0.0052 0.20 699 0.00199 0.052 0.0381 0.00042 0.0000

[0055] As shown in the respective embodiments for carrying out the first to third methods, a rotor is required which has a blade shape that makes the separation grain size constant in the diametrical direction of the classification chambers.

[0056] The illustrated rotors 23, 24 and 31 of the respective embodiments are used in a vertically oriented classification device, but they can also be used in a horizontally oriented classification device. [First embodiment]

[0057] As a rotor in a Fig. In the dry classification device 3 shown in section 2, a rotor 42 was installed. Fig.The rotor of the size shown in Figure 9 is manufactured, in which forty blades 41, whose thickness at the outer diameter of 200 mm is 5 mm and whose thickness at the inner diameter of 165 mm is 0, satisfying equation 15, are arranged radially in the circumferential direction at a specific distance from the center of the rotor. A material consisting of 73.5 kg / h heavy calcium carbonate with the physical properties shown in Table 9, an average grain size D50 of 2.50 µm, a maximum grain size D100 of 13.20 µm, and the properties shown in Table 8 and Fig. The grain size distribution shown in 11 was combined with 550 Nm 3 / h (0.152 Nm 3Air was supplied to the classification device 3 and classification was carried out according to the setting conditions based on the setting values ​​in Table 10 and the calculated values ​​in Table 11. Then the fines collected in a vessel 12 were subjected to a measurement, the result of which was the particle size and its percentage, as shown in Table 8 below. Fig. Figure 11 shows the measured fines. The average particle size D50 was 1.24 µm and D100 was 5.86 µm. The measurement was performed using a laser diffraction / scattered light particle size distribution measuring device from HORIBA, Ltd. (trade name LA-700). [Table 8] grain size (µm) FREQUENCY IMAGINARY % COMPARISON EXAMPLE 1 FREQUENCY OF FINE GOODS % EXAMPLE OF EXECUTION 1 FREQUENCY OF FINE GOODS % 0.296 0.339 0.00 0.389 0.00 0.11 0.445 0.00 0.17 0.57 0.51 0.11 0.27 1.15 0.584 0.21 0.48 1.95 0.669 0.40 0.87 3.21 0.766 0.79 1.60 5.06 0.877 1.49 2.86 8.41 1.005 2.62 4.75 10.73 1.151 4.14 7.05 12.28 1.318 5.77 9.15 12.46 1.51 7.06 10.34 11.25 1.729 7.69 10.35 9.22 1.981 7.69 9.48 7.06 2.269 7.32 8.27 5.25 2.599 6.90 7.13 3.92 2.976 6.59 6.10 2.76 3.409 6.45 5.29 1.95 3.905 6.42 4.48 1.30 4.472 6.36 3.76 0.80 5.122 6.07 2.95 0.40 5.867 5.39 2.14 0.11 6.72 4.30 1.42 0.00 7.697 2.98 0.81 8.816 1.77 0.29 10.097 0.90 0.00 11.565 0.43 13.246 0.17 15.172 0 17.377 [Table 9] SPECIFIC SURFACE MEDIUM GRAIN SIZE REST AT 45 µm SEVEN BULK WEIGHT DOP ABSORPTION WHITENESS HUMIDITY cm 2 / g Aµm Bµm % g / ml ml / 100 g % % AIR-PERMANENTNESS PROCEDURE AIR-PERMANENTNESS PROCEDURE DIAMETER AT 50% INCUMULATIVE PORN SIZE DISTRIBUTION JISSTANDARD SEVEN JIS K5101 STATIC METHOD ACCORDING TO JISK5101 (DOP USE) SPECTROSCOPICAL COLORIMETER / WHITE DEGREE METER JIS K0068 21,000 1.1 2.5 0 0.3 33 43 94 [Table 10] SETTINGS DIAMETER POSITION FLOW RATE NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific particle weight Specific fluid weight SHOVEL HEIGHT d Q N n η ρ2 ρ1 T m m 3 / s - rpm kg / m·s kg / m 3 kg / m 3 m 0.2000 0.153 40 7000 0.000018 2700 1.2 0.15 0.1942 0.153 40 7000 0.000018 2700 1.2 0.15 0.1883 0.153 40 7000 0.000018 2700 1.2 0.15 0.1825 0.153 40 7000 0.000018 2700 1.2 0.15 0.1767 0.153 40 7000 0.000018 2700 1.2 0.15 0.1708 0.153 40 7000 0.000018 2700 1.2 0.15 0.1650 0.153 40 7000 0.000018 2700 1.2 0.15 Table 11] CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT SHOVEL THICKNESS CIRCLE AREA INTERSPACE INTERSPASHES LINEAR VELOCITY SEPARATOR SIZE d G t(d) A(d) E(d) s D1 m - m m2 m m / sec µm 0.2000 5481 0.0050 0.00161 0.0107 2.38 2.31 0.1942 5321 0.0042 0.00166 0.0111 2.30 2.31 0.1883 5161 0.0033 0.00172 0.0115 2.23 2.31 0.1825 5001 0.0025 0.00177 0.0118 2.16 2.31 0.1767 4842 0.0017 0.00183 0.0122 2.09 2.31 0.1708 4682 0.0008 0.00189 0.0126 2.03 2.31 0.1650 4522 0 0.00194 0.0130 1.97 2.31

[0058] The separation particle size D1 of 2.31 µm in Table 11 is the separation particle size determined at the outer circumference of the classification chamber using the thickness t(d) of the blade 41 set to 5 mm, and was determined by substituting the relevant items from Table 10 into Equation 6 in a simulation calculation. The blade thickness t(d) at the respective diameter positions in Table 11 was determined by substituting the separation particle size D1, set constant in the radial direction of the classification chambers, and the relevant items from Table 10 into Equation 15, whereby the inner circumference diameter at t = 0 was determined using Equation 15. Furthermore, the centrifugal effect G was determined by substituting the rotor speed n from Table 10 into G = (d·n 2) / (2×894), the circular arc area A, with t as t(d) in equation 5, by substituting the previously determined blade thickness t(d) and the relevant items in Table 10 into equation 5, the linear velocity s by substituting the previously determined circular arc area A and the relevant items in Table 10 into equation 3, and the gap E(d) between the blades using the thickness t(d) and equation 14, each determined. As shown in Table 8, the maximum grain size 100 of embodiment 1 at this time was 5.867 µm. [Comparison example 1]

[0059] A classification device having the same structure as the classification device in embodiment 1, with the exception of the rotor, and which has as its rotor a rotor 44 with the same structure and size as the rotor 42 in Fig. 9 has, apart from the fact that as in Fig.Figure 10 shows that the thickness of the blades 43 is a constant 5 mm, and the same material as in embodiment 1 was used, and classification was carried out under the same conditions. Tables 12 and 13 show the items used in the simulation calculation and the calculation results. The separation particle size D1 in Table 13 was obtained by substituting the items in Table 12 into equation 6, the circular arc area A(d) by substituting the items in Table 12 into equation 5, and the centrifugal effect G by substituting the rotor speed in Table 12 into G = (d·n 2 ) / (2×894), the linear velocity s is determined by substituting the determined arc area A(d) and the relevant items in Table 12 into equation 3, and the space between the blades E(d) is determined by substituting the arc area A(d) and the blade height T in Table 12 into E(d) = A(d) / T.

[0060] The results are shown in Table 8 and together with embodiment 1 in Fig. Figure 11. In comparative example 1, the same measuring device as in embodiment 1 was used and the particle size was measured using the same method as in embodiment 1, wherein the particle size of the fine material in the container 12 in Fig. 2, which was collected, had an average grain size D50 of 1.79 µm and a maximum grain size D100 of 8.81 µm. [Table 12] SETTINGS DIAMETER POSITION FLOW RATE SHOVEL HEIGHT NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific particle weight Specific fluid weight SHOVEL THICKNESS d Q T N n η ρ2 ρ1 t m m 3 / s m - rpm kg / m·s kg / m 3 kg / m 3 m 0.2000 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1942 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1883 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1825 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1767 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1708 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 0.1650 0.153 0.15 40 7000 1.8E-05 2700 1.2 0.005 Table 13] CALCULATION VALUES CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES LINEAR VELOCITY SEPARATOR SIZE G A(d) E(d) s D1 - m2 m m / sec µm 5481 0.00161 0.0107 2.38 2.31 5321 0.00154 0.0102 2.49 2.39 5161 0.00147 0.0098 2.61 2.49 5001 0.00140 0.0093 2.73 2.59 4842 0.00133 0.0089 2.88 2.70 4682 0.00126 0.0084 3.03 2.82 4522 0.00119 0.0080 3.21 2.95

[0061] By means of a dry classification device that divides the rotor into Fig.9, which was manufactured such that the grain size in the classification chambers is constant and the thickness of the blades satisfies equation 15, and a classification device having a rotor with a constant blade thickness, the above-described comparative test was carried out, whereby, as in Fig. Figure 11 shows that the particle size distribution of embodiment 1 shifted to the left compared to the particle size distribution of comparison example 1, and with increasing refinement the separation efficiency increased and the sinking of coarse material decreased. [Example 2]

[0062] As rotor 17 of the wet classification device in Fig. 3 was a rotor 47 which was in Fig.The rotor of the size shown in Figure 12 was manufactured, in which thirty blades 46, each with a thickness of 3 mm at the outer diameter of 86 mm and a thickness of 0 at the inner diameter of 70.2 mm, satisfying equation 15, were arranged radially in the circumferential direction at a specific distance from the center of the rotor. The material used was a slurry of spherical molten silicon dioxide (trade name FB-5SDC) from Denka Co., Ltd. The material used has a mean particle size D50 of 5 µm and a particle size D98 of 14 µm, as well as the particle size distribution shown in Table 14 and in Fig. 14, and was mixed into an aqueous hexametaphosphoric acid solution of 0.2 wt.%, with which a test was carried out 3 times based on the adjustment conditions using the adjustment values ​​in Table 15 and the calculated values ​​in Table 16 determined using the adjustment values ​​in a simulation calculation.The results are shown in Table 14 below and in . Fig. Figure 14 shows the results. The fines slurries collected in tank 19 for each trial were analyzed, and the measurement of the respective particle sizes revealed that the average particle size D50 was 2.7, 2.9, and 2.9 µm, and the average particle size D98 was 5.7, 6.6, and 6.8 µm. The measurements were performed using a laser diffraction particle size distribution measuring device from Shimadzu, Corporation (trade name SALD-3100). [Table 14] grain size (µm) FREQUENCY IMAGINARY % COMPARISON EXAMPLE 2 FREQUENCY OF FINE GOODS % EXAMPLE 2 FREQUENCY OF FINE GOODS % - est1 - est2 - est1 - est2 - is3 0.233 0 0 0 0 0 0 0.291 0 0 0 0 0 0 0.362 0 0 0.363 0.002 0.003 0 0.451 0.018 0.004 0.059 0.017 0.029 0.002 0.563 0.134 0.074 0.121 0.067 0.134 0.048 0.701 0.524 0.397 0.456 0.21 0.411 0.267 0.874 1.349 1.257 1.21 0.616 0.976 0.912 1.089 2.51 2.705 2.456 1.619 2.01 22 1.356 3.668 4.397 4.086 3.275 3.555 3.976 1.690 4.622 5.993 5.836 4.682 5.025 5.531 2.106 5.504 7.345 7.227 10.357 7.45 7.142 2.625 6.538 8.513 10.039 24.636 20.768 19.047 3.271 7.693 14.273 16.675 27.179 25.734 26.218 4.076 8.605 17.045 18.543 21.959 22.639 24.201 5.079 9.182 13.076 12.547 2.448 5.498 4.425 6.329 13.959 13.503 11.328 1.892 3.412 3.461 7.887 16.615 6.109 5.307 0.824 1.746 1.841 9.828 10.515 3.567 2.723 0.205 0.535 0.614 12.247 5.058 1.382 0.872 0.012 0.075 0.107 15.262 2.555 0.33 0.152 0 0 0.008 19.018 0.814 0.031 0 0 0 0 23.699 0.137 0 0 0 0 0 [Table 15] SETTINGS DIAMETER POSITION FLOW RATE NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific particle weight Specific fluid weight SHOVEL HEIGHT d Q N n η ρ2 ρ1 T m N m 3 / s - rpm kg / m·s kg / m 3 kg / m 3 m 0.0860 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0834 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0807 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0781 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0755 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0728 2.78E-06 30 4442 0.001 2300 1000 0.01 0.0702 2.78E-06 30 4442 0.001 2300 1000 0.01 [Table 16] CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT SHOVEL THICKNESS CIRCLE AREA INTERSPACE INTERSPASHES LINEAR VELOCITY SEPARATOR SIZE d G t(d) A(d) E(d) s D1 m - m m 2 m m / sec µm 0.0860 949 0.00300 0.000060 0.0060 0.00154 1.52 0.0834 920 0.00250 0.000062 0.0062 0.00149 1.52 0.0807 891 0.00200 0.000065 0.0065 0.00144 1.52 0.0781 862 0.00150 0.000067 0.0067 0.00139 1.52 0.0755 833 0.00100 0.000069 0.0069 0.00134 1.52 0.0728 804 0.00050 0.000071 0.0071 0.00130 1.52 0.0702 775 0.00000 0.000073 0.0073 0.00126 1.52

[0063] The separation particle size D1 in Table 16 is the separation particle size determined at the outer circumference of the classification chambers using the blade thickness t(d) set to 3 mm. This was determined by substituting the relevant items from Table 15 into Equation 6 in a simulation calculation. By setting the separation particle size in the radial direction to a constant 1.52 µm and substituting this separation particle size and the respective items from Table 15 into Equation 15, the blade thickness t(d) at the respective diameter position was determined. Furthermore, the inner circumference diameter d for t = 0 was determined using Equation 15. The centrifugal effect G was calculated by substituting the rotor speed n from Table 15 into G = (d·n 2) / (2×894), the circular arc area by substituting the previously determined blade thickness t(d) and the relevant items in Table 15 into Equation 5, the linear velocity s by substituting the previously determined circular arc area A and the relevant items in Table 15 into Equation 3, and the gap E(d) between the blades using the thickness t(d) and Equation 14. As shown in Table 14, D98 in embodiment 2 was 5.7, 6.6, and 6.8 µm at this time. [Table 17] SETTINGS DIAMETER POSITION FLOW RATE SHOVEL HEIGHT NUMBER OF CLASSIFICATION CHAMBERS ROTOR SPEED VISCOSITY Specific particle weight Specific fluid weight SHOVEL THICKNESS d Q T N n η ρ2 ρ1 t m Xm 3 / s m - rpm kg / m·s kg / m 3 kg / m 3 m 0.0860 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0834 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0807 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0781 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0755 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0728 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 0.0702 2.78E-06 0.01 30 4442 0.001 2300 1000 0.003 [Table 18] CALCULATION VALUES DIAMETER POSITION CENTRIFUGAL EFFECT CIRCLE AREA INTERSPACE INTERSPASHES LINEAR VELOCITY SEPARATOR SIZE d G A(d) E(d) s D1 m - m2 m m / sec µm 0.0860 949 0.000060 0.0060 0.00154 1.52 0.0834 920 0.000057 0.0057 0.00162 1.58 0.0807 891 0.000055 0.0055 0.00170 1.64 0.0781 862 0.000052 0.0052 0.00179 1.71 0.0755 833 0.000049 0.0049 0.00189 1.79 0.0728 804 0.000046 0.0046 0.00200 1.88 0.0702 775 0.000043 0.0043 0.00213 1.97 [Comparative example 2]

[0064] A classification device having the same structure as the classification device in embodiment 2, with the exception of the rotor, and which has as its rotor a rotor 45 with the same structure and size as the rotor 47 in Fig. 12 has, apart from the fact that as in Fig. Figure 13 shows that the thickness of the blades 48 is a constant 3 mm, and the same material as in embodiment 2 was used and a classification was carried out under the same conditions.

[0065] Tables 17 and 18 show the items used in the simulation calculation and the calculation results. The separation particle size D1 in Table 18 was obtained by substituting the items in Table 17 into equation 6, the circular arc area A(d) by substituting the items in Table 17 into equation 5, and the centrifugal effect G by substituting the rotor speed in Table 17 into G = (d·n 2 ) / (2×894), the linear velocity s is determined by substituting the determined arc area A(d) and the relevant items in Table 17 into equation 3, and the space between the blades E(d) is determined by substituting the arc area A(d) and the blade height T in Table 17 into E(d) = A(d) / T.

[0066] The result is shown in Table 14 and together with embodiment 2 in Fig. 14. In comparative example 2, the same measuring device as in embodiment 2 was used, and a particle size measurement was carried out twice using the same method as in embodiment 2, whereby the particle sizes of the fine material, which is in the Fig. The sample collected from the vessel shown in 3 (vessel 19) yielded a mean grain size D50 of 3.3 and 3.5 µm and a grain size D98 of 9.1 and 9.7 µm.

[0067] The particle size distribution of embodiment 2, as shown in the wet classification device, also showed the following: Fig. As can be seen in Figure 14, compared to the grain size distribution of the comparative example 2, there is a shift to the left, whereby the reduction in grain size and the separation sharpness were extensively improved and the sinking of coarse material could be reduced. [Industrial applicability]

[0068] The classification device of the present invention can be used in all industries where various powders of a micron or submicron level are subjected to dry or wet classification, e.g. in the metal industry, the chemical industry, the pharmaceutical industry, the cosmetics industry, the pigments industry, the food industry, the ceramics industry, etc. [Explanation of reference symbols] 1, 21, 26, 33, 41, 43, 46, 48 shovels 2, 17, 23, 25, 31, 42, 44, 47 rotor 9, 22, 27, 32 Classification Chambers

Claims

[1] Classification device comprising a plurality of blades arranged at suitable intervals in the circumferential direction and arranged radially or eccentrically, and a rotor with classification chambers between the blades, by which fine particles in a fluid are classified such that, as a fluid flows into the classification chambers from the outer circumferential side to the inner circumferential side, particles larger than a separation particle size are moved to the outer circumferential side and particles smaller than the separation particle size are moved to the inner circumferential side, characterized by , that the blades have a constant height in the direction of rotation of the rotor and the thickness in the circumferential direction is formed to be thicker towards the outer circumference, whereby the thickness t(d) of the blades at the position of the diameter d of the classification chambers can be determined by means of the following equation 15: t(d)=1N[πd−QT×1D12×2×894d⋅n2×18η9.8(ρ2−ρ1)] where Q: the flow rate, N: the number of classification chambers in the circumferential direction, D1: the particle size, n: the rotational speed of the rotor, η: the viscosity of the fluid, ρ1: the specific gravity of the fluid, ρ2: the specific gravity of the particles and T is the height (constant) of the blades. [2] Classification device according to claim 1, wherein the thickness t(d) of the blades is equal to 0 at the inner circumference of the blades. [3] Classification device comprising a plurality of blades arranged at suitable intervals in the circumferential direction and arranged radially or eccentrically, and a rotor with classification chambers between the blades, by which fine particles in a fluid are classified such that, as a fluid flows into the classification chambers from the outer circumferential side to the inner circumferential side, particles larger than a separation particle size are moved to the outer circumferential side and particles smaller than the separation particle size are moved to the inner circumferential side, characterized by , that the classification device has a rotor in which the thickness of the blades is fixed in the circumferential direction and the height in the direction of rotation of the rotor is formed higher towards the inner circumference, wherein the height T(d) of the blades at the position of the diameter d of the classification chambers satisfies the following equation 11: T(d)=Qπd−tN⋅1D12×2×894d⋅n2×18η9.8(ρ2−ρ1) where Q: the flow rate, N: the number of classification chambers in the circumferential direction, D1: the particle size, n: the rotational speed of the rotor, η: the viscosity of the fluid, ρ1: the specific gravity of the fluid, ρ2: the specific gravity of the particles and t is the thickness of the blades. [4] Classification device comprising a plurality of blades arranged at suitable intervals in the circumferential direction and arranged radially or eccentrically, and a rotor with classification chambers between the blades, by which fine particles in a fluid are classified such that, as a fluid flows into the classification chambers from the outer circumferential side to the inner circumferential side, particles larger than a separation particle size are moved to the outer circumferential side and particles smaller than the separation particle size are moved to the inner circumferential side, characterized by, that on the one hand the height of the blades is made higher towards the inner circumference in the direction of rotation of the rotor and on the other hand the thickness is made thicker towards the outer circumference in the direction of the circumference, whereby the height T(d) and the thickness t(d) of the blades at the position of the diameter d of the classification chambers can be determined by means of the following equations 18, 19 and 21: T(d)=QE(d)⋅N⋅1D12×2×894d⋅n2×18η9.8(ρ2−ρ1) E(d)=πN⋅{b⋅d2−b⋅d2−a⋅d1d2−d1×(d2−d)} t(d)=πdN−πN⋅{b⋅d2−b⋅d2−a⋅d1d2−d1×(d2−d)} where a: the gap coefficient (πd1 - Nt1) / πd1 between the inner circumferential blades, b: the gap coefficient (πd2-Nt2) / πd2 between the outer circumferential blades, d1: the inner circumferential diameter of the rotor, d2: the outer circumferential diameter of the rotor, t1: the inner circumferential thickness of the blades, t2: the outer circumferential thickness of the blades, Q: the flow rate, N: the number of classification chambers in the circumferential direction, D1: the particle size, n: the rotational speed of the rotor, η: the viscosity of the fluid, ρ1: the specific gravity of the fluid, and ρ2: the specific gravity of the particles.

Images