Apparatus and method for measuring suspension flow in a tube separator

Abstract

The device for measuring the fluid suspension includes a radiation source that induces waves into the fluid suspension, such that the flow of the fluid suspension causes particles in the fluid suspension to be sorted based on their size. The measurement arrangement measures a first value of a first parameter of waves interacting with the fluid suspension in a first wavelength band of the waves. The measurement arrangement measures at least one second value of: a second parameter of the waves interacting with the fluid suspension, and a first parameter of the waves interacting with the fluid suspension in a second wavelength band of the waves. The measurement of the first value and the measurement of the second value are synchronized with each other. The measurement arrangement is configured to form at least one comparison, each comparison being associated with one of the first values ​​and one of the second values. The measurement arrangement forms a distribution having at least one of the first values ​​as a function of one of the comparisons.

Description

[Technical Field]

[0001] This application relates to an apparatus and method for measuring suspension flow in a tube separator. [Background technology]

[0002] Pipe flow fractionation is based on a flow velocity gradient caused by the flow resistance of the pipe wall in the cross-sectional direction of the pipe. The flow velocity is higher in the center of the pipe compared to the end regions of the pipe closer to the wall. Some particles move along the periphery of the pipe due to turbulence, while other particles remain in the center of the pipe with faster flow. Larger particles are more likely to be swept away by the rapid flow in the center of the pipe. Particles moving in the center of the flow reach the end of the pipe first because they overtake slower moving particles in the end regions. Pipe flow fractionation can therefore separate particles into various size classes. Fractions of a sample can be directed to separate containers based on different instantaneous flows or based on flow measured as a function of time.

[0003] Fractionation devices are based on this principle of pipe flow fractionation. Such fractionation utilizes an optical measurement module and optical measurements to measure the fractions of the flow. Based on currently used calculation methods, the signal of the optical measurement module is calibrated to a Bauer-McNett distribution.

[0004] There are several patents regarding pipe flow fractionation. For example, FI20095381 and WO2010116030A1 describe the analysis of samples taken before the headbox based on two instants. Based on these signal values at the two instants, particle size can be modeled. Conventional fractionator measurements are sensitive to temperature and flow rate fluctuations, which increase the error of the measurement results.

[0005] Therefore, there is a need for improved measurement accuracy. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] FI20095381 publication [Patent Document 2] International Publication No. 2010 / 116030 Summary of the Invention [Problem to be solved by the invention]

[0007] The present invention aims to provide improved measurements. [Means for solving the problem]

[0008] The invention is defined by the independent claims. Embodiments are defined in the dependent claims.

[0009] An embodiment of the present invention will now be described, by way of example, with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of the principle of separation. [Figure 2] FIG. 10 is a diagram illustrating an example of signal magnitude as a function of time. [Figure 3] FIG. 10 shows an example of the effect of changing flow rate on fraction calculations. [Figure 4] FIG. 10 shows an example of the effect of separation water temperature on the attenuation signal of softwood pulp. [Figure 5] FIG. 1 is a diagram showing an example of the effect of the temperature of the separation water on eucalyptus pulp. [Figure 6] FIG. 1 shows an example of the change in fraction content of eucalyptus pulp due to temperature change. [Figure 7] FIG. 1 shows an example of the effect of the separation water temperature on pine pulp. [Figure 8] FIG. 1 shows an example of the change in fraction content of pine pulp due to temperature change. [Figure 9] FIG. 1 shows an example of intensity attenuation caused by scattering. [Figure 10]FIG. 10 shows an example of ANir attenuation coefficient response for eucalyptus pulp with three different kaolin contents. [Figure 11] FIG. 1 is a diagram illustrating an example of the effect of multiple scattering on the movement of light. [Figure 12A] FIG. 1 shows an example of a fractogram illustrating particle properties and quantities as a function of flow rate. [Figure 12B] FIG. 1 shows an example of a fractogram illustrating particle properties and quantities as a function of flow rate. [Figure 13] FIG. 10 shows an example of a mapping of quantities corresponding to a first property class in a fiber distribution. [Figure 14] FIG. 10 is a diagram showing an example of mapping of quantities corresponding to a second property class in fiber distribution. [Figure 15] FIG. 10 shows an example of mapping of quantities corresponding to a third property class in a fibrous distribution. [Figure 16] FIG. 10 is a diagram illustrating an example of a mapping of quantitative values corresponding to characteristic classes of fibrous distribution. [Figure 17] FIG. 1 shows an example of the fractionation results of a pine pulp sample using the new calculation method. [Figure 18] 10A-10D show an example of fiber index distribution for a pine pulp sample using AVis or DVis signals for calculation, respectively. [Figure 19] FIG. 1 shows an example of a fiber index distribution as a primary measurement and a size index distribution as a secondary measurement. [Figure 20] FIG. 1 shows an example of fiber distribution in pine pulp when the water temperature changes from 15°C to 40°C. [Figure 21] FIG. 1 shows an example of the change in fraction content as a function of temperature for a pine pulp sample using the prior art liter-based calculation method and the new calculation method. [Figure 22] FIG. 1 shows an example of the fiber index distribution of eucalyptus pulp when the water temperature is changed from 15°C to 40°C. [Figure 23]FIG. 1 shows an example of the change in fraction content as a function of temperature for a eucalyptus pulp sample using the prior art calculation method and the new calculation method on a liter basis. [Figure 24] FIG. 1 shows an example of the distribution of a sample pulp at a consistency of 0.7%. [Figure 25] FIG. 1 shows an example of sample pulp fine fraction response as a function of total consistency. [Figure 26] FIG. 1 shows an example of the response of sample pulp fiber fraction FR2 as a function of total consistency. [Figure 27] FIG. 1 shows an example of the total combined response of a sample pulp fines and fiber fraction as a function of total consistency, and an example of the consistency response of the total combined fiber and fines response. [Figure 28] FIG. 1 shows an example of a fractogram and distribution of a pulp containing only long fibers. [Figure 29] FIG. 1 shows an example of a fractogram and distribution of pulp containing only short fibers. [Figure 30] FIG. 1 shows an example of a fractogram and distribution of a pulp consisting of 90% short fiber pulp and 10% calcium carbonate. [Figure 31] FIG. 1 is a diagram showing an example of size index distribution when calcium carbonate is added to long fiber pulp at various ratios. [Figure 32] FIG. 1 shows an example of a fractogram and distribution of a pulp mixture containing 60% TMP and 40% eucalyptus pulp at a total consistency of 0.3% Cs. [Figure 33] FIG. 1 shows an example of a fractogram and distribution of a pulp mixture containing 60% TMP and 40% half eucalyptus pulp and half kaolin at a total consistency of 0.3% Cs. [Figure 34A] FIG. 10 shows an example of the effect of black liqueur on an attenuation coefficient fractogram. [Figure 34B]FIG. 10 shows an example of the effect of black liqueur on an attenuation coefficient fractogram. [Figure 35] FIG. 1 shows an example of the variation of the size index of the fine fraction as a function of the fiber index. [Figure 36] FIG. 1 shows an example of fiber indicators at different refining intensities. [Figure 37] FIG. 1 shows an example of size index distribution at different purification intensities. [Figure 38] FIG. 1 shows an example of the influence of refining intensity on size index and fiber index. [Figure 39] FIG. 1 shows an example of a fractogram and distribution of a well-mixed dissolving pulp sample. [Figure 40] FIG. 10 shows an example of insufficient mixing of the same sample. [Figure 41] FIG. 1 shows an example of a fractogram and distribution of a nanocrystal sample. [Figure 42] FIG. 1 shows an example of fiber index and size index of refined chemical pulp at different fractionation rates. [Figure 43] FIG. 1 shows an example of the positions of multiple samples in a fractionation tube at a given moment during fractionation. [Figure 44] FIG. 10 is a diagram showing an example of separation of flocs and short fibers based on the order of arrival. [Figure 45] FIG. 1 is a diagram showing an example of a flow chart for a fluid suspension. [Figure 46] FIG. 1 is a diagram showing an example of an instrument for measuring a fluid suspension. [Figure 47] FIG. 2 illustrates an example of a data processing unit. [Figure 48] FIG. 1 is a diagram illustrating an example of a general measurement configuration of a measurement device. [Figure 49] FIG. 2 is a diagram illustrating an example of a control device. DETAILED DESCRIPTION OF THE INVENTION

[0011] The following embodiments are merely examples. Although the specification may refer to "one example" or "one embodiment" in several places, this does not necessarily mean that each such reference refers to the same example or embodiment, nor does it mean that a feature applies only to a single example or embodiment. Single features of different examples or embodiments may be combined to provide other embodiments. Furthermore, the terms "having" and "including" should be understood as not limiting a described example or embodiment to only those features mentioned. Also, such an example or embodiment may have features / structures not specifically mentioned. All combinations of examples or embodiments are considered operable if no structural or logical contradictions arise in the combination.

[0012] Although various examples or embodiments are shown in the drawings, it should be noted that they are simplified diagrams and only some structures and / or functional entities are shown. The connections shown in the drawings may represent logical connections or physical connections. It is also clear to those skilled in the art that the described devices may have functions and structures other than those shown in the drawings and text. It should be understood that details of some functions, structures, and signals used for measurement and / or control are irrelevant to the actual invention. Therefore, there is no need to further describe them here.

[0013] This application describes new computational equipment and methods that make wave motion measurements more widely available and simplify the technical solutions for fractionators.

[0014] First, we will examine currently used fraction calculations. Fraction measurement involves three steps: sample administration, sample separation, and sample measurement. The principle of fractionation is illustrated in Figure 1. In the administration step, a sample is administered to a fractionation tube 10. In the separation step, the sample is forced by a fluid flow through the fractionation tube 10, which divides the sample into different fractions. In the measurement step, the sample divided during separation is measured as a function of time and / or flow rate. However, other types of fractionation may also be used.

[0015] Currently used fraction calculations are based on analyzing changes in measurement signals related to particle amounts at different moments in the flow. The measurement signal is based, for example, on the attenuation, scattering, or depolarization of electromagnetic radiation. In fraction analysis, a time limit or equivalent flow rate value is set, and different fractions are defined into their own calculation classes based on it. Figure 2 shows an example of this. In Figure 2, the sample is defined into three different classes, FR1, FR2, and FR3, as a function of time. Of these, FR1 represents the amount of large fiber particles, FR2 represents the amount of medium-sized particles, and FR3 represents the amount of small particles. These definitions apply throughout this application. If fractionation depends on variables other than time or flow rate, the variables may be treated in a similar way to how time or flow rate is treated in the measurement methods described herein. The other variable for fractionation may be, for example, location.

[0016] Figure 2 shows an example of a quantity signal as a function of time, which depends on the flow rate of the fractionation tube 10, or has a known relationship to the flow rate. The vertical axis is the quantity, the decay coefficient, and the horizontal axis is the instant at which the sample is measured based on the flow rate.

[0017] The quantity is calculated as the sum of the signal values or the integral of the signal values over its own limited flow area. The signal quantity may be an attenuation signal, a depolarization signal, or a scattering signal. The signal may also be a combination of the above, or any other signal proportional to the quantity. In the illustrated example, the measurement representing the quantity on the y-axis is the attenuation signal measured at visible wavelengths. The calculation of attenuation is explained in more detail after Figure 8, where the currently used technique is described.

[0018] Such fraction calculations based on flow rate require a controlled, stable flow during separation and measurement. However, flow control is difficult, especially during long-term separations. In addition to flow rate, the fractionation that occurs during flow is affected by, for example, the temperature and viscosity of the fractionation fluid, as well as the properties of the pipe wall. Changes in these variables are considered errors in the fractionation calculation.

[0019] Figure 3 shows an example of this. Here, the flow rate is increased, resulting in the fractogram reaching approximately 5 flow units further forward than in the previous case. The flow on the horizontal axis represents the instant at which the sample is measured. Figure 3 therefore shows an example of the effect that a change in flow rate has on the fraction calculation. This results in large errors in the calculated fraction contents. In this example, the fraction contents are listed as fraction-specific percentages of the total volume.

[0020] Flow conditions affect the Reynolds number according to equation (1). The optimum Reynolds number for pipe flow fractionation is between 1,000 and 10,000.

[0021]

number

[0022] Temperature affects the Reynolds number through medium density and medium viscosity. For water above +4°C, both of these properties can decrease with increasing temperature, but viscosity usually decreases more than density. Thus, as temperature increases, the Reynolds number of a water flow increases. Increasing water temperature from 10°C to 40°C doubles the Reynolds number. More generally, changes in temperature can cause changes in the Reynolds number of a water flow, making fraction measurements inaccurate or impossible.

[0023] In addition to water temperature, increasing the flow rate or pipe diameter also increases the Reynolds number. Therefore, if faster flow is desired to shorten fractionation time while keeping the Reynolds number constant, the increase in flow rate can be compensated for by reducing the diameter of the fractionation pipe 10 or by using cooler water. It can be observed that temperature changes have a stronger effect on the separation of the fiber fraction than on the fine fraction.

[0024] Figure 4 shows an example of the effect of fractional water temperature on the decay signal of softwood pulp. The vertical axis is decay and the horizontal axis is flow rate in liters. The curves that decay first rise first. Thus, they are in sequential order. The effect of fractional water on the decay signal of the softwood pulp sample can be clearly observed.

[0025] The effect of fractionation water temperature was measured in tests where the water temperature was varied between 15°C and 40°C. The test pulps used were eucalyptus pulp and pine pulp. The change in the fractogram of eucalyptus pulp resulting from the temperature change is shown in the example in Figure 5. This 25°C temperature change caused the peak of the fractogram to shift by approximately 4 seconds. In the distribution shown, when the fraction limit was set to 34 seconds, Fraction 1 represented 91% of the total, and the remaining 9% was Fraction 2. This change is also confirmed by the fraction volume graph in the example in Figure 6. The content changes with the temperature change, with Fraction 1 decreasing by 22 percentage points to 69%. The content of Fraction 2 increases by a corresponding factor.

[0026] An example of the effect of fractionation water temperature on eucalyptus pulp is shown in Figure 5. The vertical axis is attenuation and the horizontal axis is flow time in seconds.

[0027] Figure 6 shows an example of the change in fraction content of eucalyptus pulp with temperature, where the x-axis is temperature and the y-axis is fraction content [%].

[0028] An example of the effect of fractional water temperature on pine pulp is shown in Figure 7. The vertical axis is attenuation and the horizontal axis is flow time in seconds.

[0029] Figure 8 shows an example of the change in fraction content of pine pulp with temperature. The x-axis is temperature, and the y-axis is fraction content [%]. The fraction 1 content of pine pulp decreases from 95% to 90%. Similarly, the fraction 2 content increases from 5% to 10%.

[0030] The currently used detection techniques are described below. The measurements and results described in this application utilize optical measurements, which are based on measuring the attenuation of light at two different wavelengths and reversing the polarization level at visible wavelengths. Here, the attenuation of light is represented by the abbreviations AVis and ANir. The attenuation AVis represents the attenuation occurring in the visible wavelength range, and ANir represents the attenuation occurring in the near-infrared wavelength range. The abbreviation DVis represents the polarization reversal occurring in the visible wavelength range. The calculations are explained in the next chapter.

[0031] The attenuation coefficient (Attn Coeff, AVis, ANir) describes the loss of light intensity as it travels through a medium. Attenuation is also known as absorbance. Two mechanisms affect the attenuation of light: absorption and scattering. Here, absorption refers to the absorption of light energy into a material, while scattering refers to the change in direction of light when it encounters a solid object in a fluid.

[0032] FIG. 9 shows an example of the attenuation of intensity caused by scattering. The light intensity is measured using the detector 22, and the attenuation can be determined using the data processing unit 24 (see FIG. 46). Low consistency results in weaker scattering than high consistency. In consistency (i.e., concentration) measurements, attenuation is primarily caused by increased light scattering as a function of consistency. An example of the principle of this effect is shown in FIG. 9. Wood fibers and other particles placed in the light path deflect some of the light from its original direction, resulting in a decrease in the light intensity at the detector. The higher the consistency, the more scattering particles there are in the light path, resulting in a greater decrease in intensity.

[0033] The intensity decreases logarithmically with consistency. Attenuation is calculated as the relationship between the measured intensity and the intensity measured in water. The calculation of the AVis signal is as follows:

[0034]

number

[0035] Similarly, the calculation of the ANir signal is as follows:

[0036]

number

[0037] Dark values caused by dark current in the detector etc. are removed from all measurements.

[0038] Therefore, in this case, the attenuation coefficient is proportional to the intensity measured through water. Therefore, this calculation method assumes that the attenuation coefficient of pure water is zero, making the attenuation coefficient zero-based. This can be seen in Figure 10, which shows examples of the ANir attenuation coefficient response for eucalyptus pulp with three different kaolin contents.

[0039] Figure 10 shows an example of the ANir damping coefficient response for eucalyptus pulp with three different kaolin contents. The horizontal axis is the consistency C in %. In addition to being zero-based, the damping coefficient can be considered largely linear with consistency. However, Figure 10 shows an example of this linearity gradually disappearing with increasing consistency, as the dotted line begins to diverge from the straight line.

[0040] Figure 11 shows an example of the effect of multiple scattering on light transport. The nonlinearity in the attenuation coefficient response is due, at least in part, to multiple scattering, which begins to occur as particle density increases with increasing consistency. With multiple scattering, some of the light diverging from the original path is redirected back in its original direction, which begins to reduce the relative increase in attenuation with consistency.

[0041] Depolarization, or polarization, measurements describe the reversal of polarization when light encounters wood fibers. The principle of the measurement calculation is as follows:

[0042]

number

[0043] The effect of attenuation on depolarization can be corrected by dividing the measured DVis by the AVis intensity value measured through the sample.

[0044] If the water value is removed from the measurement, the depolarization measurement can be zero-based and can be considered linear. However, the linearity assumption may be valid for a limited consistency range. Increasing the measurement distance can reduce the linear range of the measurement. At low consistencies, depolarization is sensitive to the amount of fiber, while at high consistencies, depolarization is sensitive to all scattering-causing materials. However, depolarization measurements are less sensitive to fillers with respect to consistency changes than attenuation measurements.

[0045] Thus, in the example of Figure 2, a measurement variable proportional to the volume is used in known fraction calculation methods to account for the amount of material arriving for measurement at different flow times. The increase in the light attenuation coefficient due to an increase in particle volume can be used as a volume-proportional variable. Alternatively, the depolarization of light, which increases with an increase in the amount of birefringent material, can be used. Other measurement methods based on scattering are also used in currently known solutions.

[0046] In addition to the traditional measurement signal representing the quantity, the new calculation method utilizes a variable representing a property of the sample. Thus, the new calculation method provides a simultaneous measurement of the property and quantity of the sample, which can be expressed as a distribution independent of the flow rate. From the distribution, fraction-specific integral values or average fraction-specific centroid values can be calculated as property values. The centroid can mean a weighted average of the values.

[0047] The new term fiber index can be defined in the following way: In fiber measurements, one potential signal that characterizes a sample is the ratio between the depolarized signal and the attenuation measurement. This ratio represents the fibrous nature of the sample, and in this application this ratio is referred to as the fiber index (FI).

[0048]

number

[0049] Water Value DVis water is subtracted from the depolarized signal to provide a ratio starting from zero.

[0050] Pure, unbroken fibers have a high ratio due to strong depolarization and low attenuation. With fine powders and fillers, the signal behaves in the opposite way, resulting in a low ratio.

[0051] Another new term, size index, can also be defined as follows. Here, the ratio of the extinction coefficients of short and long wavelengths is used as a second index to describe the properties. It represents a particle dimension such as particle size, fiber width, or fiber wall thickness. This extinction coefficient ratio is called the size index (SI).

[0052] The ratio of the short wavelength to long wavelength extinction coefficients indicates the particle size.

[0053]

number

[0054] A variable that inversely represents the size index can be obtained by dividing the short wavelength attenuation coefficient by the equivalent long wavelength attenuation coefficient, or vice versa. Instead of the attenuation coefficient, other measured variables that respond to changes in light intensity can also be used. Typically, a function is formed between the attenuation coefficient of a first wavelength and the attenuation coefficient of a second wavelength, where the first wavelength and the second wavelength are different wavelengths.

[0055] In addition to particle size and shape, the attenuation that occurs at different wavelengths is affected by the spectral refractive index of the particle and the fluid.

[0056] When calculating the fiber index, the depolarized and attenuated signals may be measured at any wavelength of electromagnetic radiation, and they do not have to be the same wavelength.

[0057] A wide range of electromagnetic radiation wavelengths can be used to calculate the size index. Very short wavelengths, for example in the ultraviolet range, are most sensitive to small particles. As with the fiber index, the size index calculation utilizes several different measurement signals, among which the measurement variable representing particle size is generated by a calculation method.

[0058] Additionally, size indicator measurements can be sensitive to the presence of dissolved substances in the fluid, as will be illustrated in the measurement examples below.

[0059] The new method for calculating fractions requires a variable representing the particle characteristics in addition to a variable representing the quantity. In this step, this variable is the ratio between two different signals. The change in the ratio represents the change in particle characteristics during fractionation. The particle characteristics themselves do not change, but the flow of the suspension causes the particles to be separated into various size classes, which move one after the other in the flow. That is, the particles move in fractions in the flow. It is also possible to use more than two basic signals in the calculation. Using these signals, more complex models can be calculated to describe the properties of the flowing material. The model does not have to be proportional, but can be formed by any function describing the desired properties.

[0060] Furthermore, the signal need not be optical, but other measurement techniques may form the variable that describes the characteristic.

[0061] In one embodiment, the ratio-based division between the signals may be selected so that the signals are zero-based and linear. For attenuated signals, this may be done when the consistency is relatively low. Depolarized signals are also at least partially linear at low consistency, although linearity is not an absolute requirement.

[0062] In fractions, the sample may be diluted from the consistency of the original whole sample as it is divided over a long distance within the fractionation tube 10. This dilution may be used to keep the attenuation and depolarization signal in the linear range.

[0063] This section describes the computational procedures for generating the distribution. Figures 12A and 12B show example fractograms showing particle properties (DVis / AVis) and quantity (AVis) as a function of flow rate. The fractogram representing quantity generated by the decay signal (AVis) is in Figure 12A, and the fractogram representing fiber content (DVis / AVis) is in Figure 12B. In addition to the decay signal representing quantity (AVis) in Figure 12A, a fiber index representing the equivalent fractogram (DVis / AVis) was calculated (see Figure 12B). High fiber index values (DVis / AVis) represent fibers at low flow rates at the beginning of the fractogram. As the flow progresses, the fiber content of the particles arriving at the measurement location decreases, resulting in a decrease in the fiber index (DVis / AVis) value.

[0064] Distribution information is generated by combining quantitative variables corresponding to the property classes. This combination is illustrated by the examples in Figures 13-16. In the illustrated example, a fiber index (DVis / AVis) representing fiber quality generates a fiber distribution, which can be used to represent the amount of material corresponding to each fiber index that comprises the material being measured. The distribution is generated by classifying each measurement point measured for the flow based on the fiber index and calculating the sum of the quantitative signal values corresponding to each fiber class.

[0065] In this example, the quantity graph is the attenuation of visible wavelengths AVis, which is the graph in the middle of the figure. The graph at the bottom is the fiber index distribution created based on the characteristic graph and the quantity graph.

[0066] Figure 13 shows an example of a mapping of quantities corresponding to a first characteristic class in fiber distribution. The x-axis of the top graph is flow velocity, and the y-axis is the ratio of signals DVis / Avis. The x-axis of the middle graph is flow velocity, and the y-axis is the amount and / or attenuation of the optical signal. The x-axis of the bottom graph is fiber index FI, and the y-axis is the sum of signals AVis. In this example, values 0-0.25 correspond to the first class; therefore, the mean value of the class is 0.125. In Figure 13, there are values in this class in both the flow range 0...5 and the flow range 30...42. However, the Avis quantity values corresponding to values in the flow range 0...5 indicate that there is no material in this flow range; therefore, these points do not affect the calculated sum.

[0067] The sum of the quantity values corresponding to this smallest property class is 3.6. This value is located at the point on the distribution graph that corresponds to the mean value of 0.125. Therefore, the (x,y) location is (0.125, 3.6) on the bottom graph. This value therefore represents the amount of most non-fiber particles in the sample.

[0068] FIG. 14 shows an example of mapping a quantity corresponding to a second characteristic class in a fiber distribution. The x-axis of the top graph is flow velocity, and the y-axis is the ratio of the signal DVis / Avis. The x-axis of the middle graph is flow velocity, and the y-axis is the magnitude and / or attenuation of the optical signal. The x-axis of the bottom graph is fiber index FI, and the y-axis is the sum of the signal AVis. A calculation of the quantity of material corresponding to the second characteristic class is performed. In this example, the second characteristic class ranges from 0.25 to 0.5, resulting in a mean value corresponding to the class of 0.375. Material corresponding to this class is measured over a flow rate range of 23...29. The sum of the AVis for these measurement points is 15.0. This sum is located at the point corresponding to the mean value of 0.375 on the distribution graph at (x,y) coordinates (0.375,15).

[0069] FIG. 15 shows an example of a mapping of quantities corresponding to a third characteristic class in a fiber distribution. The x-axis of the top graph is flow velocity, and the y-axis is the ratio of signals DVis / Avis. The x-axis of the middle graph is flow velocity, and the y-axis is the amount and / or attenuation of the optical signal. The x-axis of the bottom graph is fiber index FI, and the y-axis is the sum of signals AVis. In the example of FIG. 15, there is a single measurement point for the third characteristic class, 0.5-0.75, for flow values 22. This AVis value is 1.5, which corresponds to the fiber index value 0.625 in the distribution, with an (x,y) arrangement of (0.625, 1.5).

[0070] FIG. 16 shows an example of a mapping of quantitative values corresponding to characteristic classes in a fiber distribution. The x-axis of the top graph is flow velocity and the y-axis is the ratio of signals DVis / Avis. The x-axis of the middle graph is flow velocity and the y-axis is the amount and / or attenuation of the optical signal. The x-axis of the bottom graph is fiber index FI and the y-axis is the sum of signals AVis (see equation (2)).

[0071] When the entire fractogram is classified in this way, an example of a complete distribution may be obtained. Thus, in this example, low values of the fiber index correspond to the amount of non-fiber particles, and high values correspond to the amount of fiber particles. Materials with a high fiber index are typically pure, non-fibrillated fibers. Fibers in the mid-range may, for example, be materials that have retained some fibrous character but have been pulverized and fibrillated. The lowest fiber class includes materials that do not contain fines, especially polarization-induced pulp.

[0072] The attenuation response in the visible wavelength range used in this example is different for particles with different fiber indices. Thus, fines and fillers will produce a different response in terms of pulp content from pure fiber. Fines and fillers may, for example, produce a stronger response in terms of attenuation in terms of pulp content than pure fiber. This may be taken into account by calibrating the quantity signal so that a smaller weighting factor is given to the quantity measurement in the low fiber index range than in the high fiber index range. Calibration may be performed by measuring particles of known properties and known content.

[0073] Figure 17 shows an example of the fractionation results for a pine pulp sample using the new calculation method. In the graph on the left, the bottom continuous line shows the AVis signal, and the dashed line above the AVis signal shows the DVis signal. The highest line is the fiber index FI. These signals are shown as a function of flow time. The flow time calculation begins after the separation sequence at the start of the measurement sequence, so that the sample reaches the measurement position.

[0074] The sample measured between 10 and 30 seconds is composed of fiber, the amount of which peaks at 18 seconds. The fines in the sample are reached after 30 seconds. The fiber index, which represents the relationship between the DVis and AVis signals, is indicated by the highest line and is read from the right vertical axis. At the 20-second point in the fiber fraction, the fiber value is 6. That is, the DVis signal value is approximately six times the AVis signal value. The higher the fiber value, the more purely fibrous the sample. Moving toward the fines fraction, the fiber content drops to approximately 1, and the DVis and AVis values become comparable. Therefore, a low fiber value indicates that the sample at that point is, on average, highly non-fibrous.

[0075] At moments when the Avis and DVis signals are very low, the fiber index obtained by division will be noisy, but this noise will not interfere too much with the measurement, since the quantity signals at these moments are so low that they do not actually add up to the calculated sum.

[0076] Therefore, the fiber index represents the fibrous nature of the sample but does not respond significantly to the sample volume. The AVis and DVis signals respond to the sample volume and are at significantly lower levels in terms of the fines fraction. From this, it can be directly inferred that the pine pulp samples in question contain very little fines compared to the fibrous material.

[0077] As already explained, Figure 17 shows an example of a fiber index distribution for a pine pulp sample based on the AVis signal. The x-axis of the left graph is time and the y-axis is the value of the AVis signal. The x-axis of the right graph is the fiber index FI and the y-axis is the sum of the AVis signal values. In the example of Figure 17, the fines limit is set at a fiber index value of 3, above which is the fiber fraction and below which is fines.

[0078] On the right side of Figure 17, the fiber index distributions calculated as described in Figures 12 to 16 and their descriptions are shown, where the quantities are expressed as sums calculated from the AVis signals by fiber index class.

[0079] In this example, the fiber distribution is centered approximately at the value 6, i.e. the particles of the sample are relatively fibrous. Here, the sample has a relatively low amount of fines.

[0080] The AVis signal, based on attenuation, is relatively more sensitive to the amount of fines than to the amount of fiber when comparing quantities based on dry matter mass. The situation is opposite for the DVis signal, based on depolarization: the sensitivity to fiber is relatively higher than the sensitivity to fines. In the example of Figure 18, the top two graphs show an example of the fiber index distribution of a pine pulp sample using the AVis or DVis signal for calculation, respectively. The sample is the same as the example of Figure 17.

[0081] These sensitivity stresses necessitate calibration when comparing pulp content fractions. To obtain equivalent amounts in different parts of the distribution, measurements must be performed using a weighting function, in which the amounts of different materials are weighted by the coefficients required for their response. When using the AVis signal, this function assigns a smaller weighting factor to fines than to fibrous materials. On the other hand, when using the DVis signal, the weight of fines must be greater than the weight of fibrous materials. The weighting factor function can be linear or nonlinear, depending on the case. When using measurement methods other than those mentioned above, a suitable weighting factor function can be used based on the response behavior.

[0082] The sensitivity difference between the attenuation coefficient signal and the depolarization signal may be equalized by using a suitable model for them. In the measurements herein, a commonly used model is the Euclidean distance of the AVis and DVis signals from the origin. This is referred to herein as CsDA, which is shown to be similar to a consistency measure calculated using the D and A signals. In CsDA, Cs stands for consistency, D stands for depolarization, and A stands for attenuation. The calculation of the CsDA variable is shown in Equation 7. CsDA may be considered to be based on a quantity. CsDA may be converted to consistency by a suitable calculation or by using a suitable calibrated coefficient.

[0083]

number

[0084] The calculation equalizes the sensitivity differences between the different components, reducing the need for calibration compared to using only the DVis or AVis signals.

[0085] Figure 18 shows an example of the fiber index distribution for the same pine pulp sample using the signals AVis, DVis, and CsDA. The x-axis is the fiber index FI, and the signals AVis, DVis, and CsDA are on the y-axis. The CsDA variable is used as the quantity signal.

[0086] As used herein, the term "total CsDA" refers to either the sum or integral of the CsDA values contained in the fraction of interest, or the sum or integral calculated over all fractions, i.e., it is the sum of the bars in the graph.

[0087] When analyzing samples containing fibrous particles, the fiber index works well as a primary measurement, allowing the sample to be classified. The advantage of the fiber index in pipe flow fractionation is that it is effective in clearly dividing samples in a flow based on their fiber content.

[0088] However, samples can also be analyzed based on size characteristics, which is particularly suitable for micro-fractionation, where fractionation takes place in fractionation tubes 10 with very small diameters, where particle size is typically very small and fibrousness is a less important particle characteristic.

[0089] Alternatively, size index can be used as a secondary measure of fiber index distribution, where size index is calculated fraction-specifically based on fiber index.

[0090] Figure 19 shows an example of a fiber index distribution as a primary measurement and a size index distribution as a secondary measurement. The left graph shows the size index as a function of flow time, similar to the fiber index signal in the left graph of Figure 17. In this case, moments with AVis values less than 0.001 are excluded from the index calculation, removing insignificant noise points from the fiber index and size index values. Typically, values below the noise level may be removed from the calculation.

[0091] In this example, the size index signal at the fiber fraction is approximately 0.8 and decreases to less than 0.6 towards the fines fraction (see Figure 17). This indicates a decrease in particle size as a function of flow. The particle dimension that has the greatest impact on the size index of the fiber fraction appears to be the fiber wall thickness.

[0092] The graph on the right shows the fraction-specific size index distributions. The white bars represent the size index distribution of particles with a fiber index above 3, and the black bars represent the size index distribution of particles with a fiber index below 3. The distribution graph shows the fiber size index, which likely indicates the fiber wall thickness and is significantly higher than the size index representing the particle size of the fine fraction.

[0093] In one embodiment, a size index distribution can be calculated for all fractions generated based on the fiber index, allowing the size distribution data of different particle classes to be distinguished from one another. The width of the classification window used to generate the fiber index is the only factor limiting the number of fractions. Thus, there may be as many fraction classes as there are bars in the distribution.

[0094] In one embodiment, a characteristic diagram may be obtained from a distribution similar to the previous example, showing the fraction-specific amounts and properties of the measured sample.

[0095] When measuring nanoscale particles, the primary interest is the size characteristic of the particle. In some cases, the fiber characteristic is a secondary measurement, and thus the crystallinity of the nanocrystals may be characterized.

[0096] Integrating the entire fiber and / or size distribution gives the overall consistency of the sample. Integrating the distributions by fraction may give fraction-specific consistencies, e.g., long fiber consistency, medium fiber consistency, fines consistency. If greater accuracy is required, the quantity signals used in the consistency definition may be calibrated using a weighting factor function. However, even without calibration, the ratio of values will represent the relative amount changes of fines and fibers to the overall consistency.

[0097] In one embodiment, for example, fraction-specific centroid values, weighted averages, averages, or values based on distribution symmetry can be generated. The fraction-specific weighted averages, which can be considered centroids, represent the average characteristics of the fraction of interest. For example, the centroid of the fiber index for the fine fraction in the example of FIG. 17 is approximately 1, and the centroid for the fiber fraction is approximately 6. If the centroids move toward higher values, it can be determined that the particles of the fraction of interest are, on average, more fibrous. Similarly, if the centroid of the size index distribution moves toward lower values, the average particle size may decrease, and if the centroid moves toward higher values, the average particle size may increase.

[0098] FIG. 20 shows an example of fiber distribution of pine pulp when the water temperature is changed from 15°C to 40°C. The x-axis is the fiber index and the y-axis is the value of the fiber index. In one embodiment, the influence of fractionation water temperature on the fractogram may be compensated for by using the new calculation method described in this application. Here, the quantity signal is CsDA. The reference value of the measurement performed at a temperature of 15°C is shown as a solid line, and the measured value is shown as a bar, i.e., in the graph at the top right of FIG. 20. Even if there is a large change in fractionation at different temperatures, the distribution does not change significantly.

[0099] Even with the same sample pulp and flow rate, the fractions may vary significantly when the fractionation temperature is changed.

[0100] Figure 21 shows an example of the change in fraction content as a function of temperature for a pine pulp sample using the traditional liter-based calculation method and the new calculation method described herein. The fiber fraction content from the traditional method is referred to as Prior Art Fraction 1, and the traditional fiber fraction content from the present method is referred to as Present Method Fraction 1. The fines fraction content from the traditional method is referred to as Prior Art Fraction 2, and the fines fraction content from the present method is referred to as Present Method Fraction 2. With the new calculation method described herein, the content remains more stable compared to the liter-based method.

[0101] As an example, the results of a similar test carried out using eucalyptus pulp are shown in the examples of Figures 22 and 23. An example of the fiber distribution is shown in Figure 22. An example of the fraction content results of the test is shown in Figure 23. Thus, using fraction calculations based on the new calculation method, fractionation errors caused by temperature may be eliminated.

[0102] Figure 22 shows an example of the fiber index distribution of eucalyptus pulp as the temperature changes from 15°C to 40°C. The x-axis is the fiber index and the y-axis is the fiber index value. The reference value for measurements carried out at a temperature of 15°C is shown as a solid line and the measurements are shown as bars (see Figure 20).

[0103] Figure 23 shows an example of the change in fraction content as a function of temperature for a eucalyptus pulp sample using the prior art calculation method on a liter basis and the new calculation method described herein. The fiber fraction content from the prior art method is referred to as Prior Art Fraction 1, and the fiber fraction content from the present method is referred to as Present Method Fraction 1. The fines fraction content from the prior art method is referred to as Prior Art Fraction 2, and the fines fraction content from the present method is referred to as Present Method Fraction 2. As can be seen, the new calculation method is independent of temperature.

[0104] In Figures 20 and 22, the distributions of both pulp types are skewed towards lower fiber values at higher temperatures compared to lower temperatures. This may be due to the fact that at higher temperatures, fractionation divides the sample into shorter time / flow times, which causes these different quality fibers to arrive partially entangled instead of completely separated from each other. For this reason, there may be non-fibrous material measured at the same time as purely long fibers, and the fiber index value may represent the average fiber quality of the particles at the same time point of the measurement.

[0105] The new calculation method described in this application is outlined as follows, and the following may be used as a distribution characteristic diagram: Integral calculated by the fraction or concentration fraction specific to the consistency; the total integral of the distribution, which represents the overall consistency or concentration; fraction-specific calculated center of gravity or other equivalent characteristic value for the characteristic of interest and a characteristic graph showing its expected variation; A center of gravity, or other equivalent characteristic value, calculated for the entire distribution for the characteristic of interest and a characteristic graph showing the change in the center of gravity, or an equivalent characteristic graph showing the change in the characteristic of interest across the sample; symmetry of the distribution with respect to the property graph showing the weighting of material amounts of different properties - this may be carried out either fraction-specifically or for the whole sample; Secondary properties calculated in relation to the primary properties - for example, size index of different fractions based on fiber index.

[0106] To accomplish the foregoing, an instrument for measuring a fluid suspension includes a radiation source configured to induce waves in the fluid suspension, causing the flow of the fluid suspension to separate particles in the fluid suspension based on their size. Size may represent the volume of the particles, but the shape of the particles may also vary, for example. Those skilled in the art are familiar with this type of field flow fractionation. Field flow fractionation can be used to separate particles in a suspension configured to flow through a fractionation tube 10.

[0107] The wave may be, for example, optical radiation or acoustic radiation. Wave excludes particle radiation, typically particles being atomic or subatomic particles. The measurement device may measure a first value of a first parameter of the wave interacting with the flowing suspension in a first wavelength band of the wave. Each of the first values may be a single measurement or an average of multiple measurements. The measurement device may measure at least one second value of: a second parameter of the wave interacting with the flowing suspension; a first parameter of the wave interacting with the flowing suspension in a second wavelength band of the wave. The measurements of the first and second values are synchronized with each other value by value. Like the first values, each of the second values may be a single measurement or a value based on multiple measurements. The measurement device may be configured to form at least one comparison, each comparison relating one of the first values to one of the second values. The measurement device may form a distribution, which may have at least one of the first values as a function of one of the comparisons that substitute a representation of the attenuation of the wave through the sample for consistency.

[0108] In one embodiment, the measurement device may measure a first value of a first parameter that depends on the consistency of the flowing suspension. The measurement device may form at least one comparison that depends on a physical property of the particles of the flowing suspension.

[0109] In one embodiment, the device may comprise a tube flow fractionator through which the suspension flows.

[0110] In one embodiment, an optical radiation source directs optical radiation at the flowing suspension, the radiation having at least one beam that is polarized or multiple beams, at least one of which is polarized and at least one separate beam that is unpolarized. A measurement device may measure the attenuation of the electromagnetic radiation interacting with the flowing suspension. The measurement device may measure a parameter of polarization of the electromagnetic radiation interacting with the suspension, and the attenuation measurements and the polarization parameter may be synchronized. The measurement device may then form a comparison of the polarization and attenuation parameters, each of which corresponds to each of the parameters and the attenuation compared based on the synchronization. The measurement device may finally allocate the attenuation in relation to the comparison between the polarization parameter and the attenuation.

[0111] In one embodiment, the measurement device may form a comparison measured at wavelengths having a known relationship to each other. To form the distribution, the measurement device may calculate a sum of attenuation measured synchronously with the parameter of polarization for at least one of at least two ranges of values of the parameter of polarization, the parameter of polarization being one of at least two ranges of values, the ranges of values not overlapping. The measurement device may correlate each of the sums and each of the ranges of values used to calculate the sum with each other to form a distribution of first sums of values as a function of the range of values.

[0112] In one embodiment, the light source may direct electromagnetic radiation of at least two different wavelengths into the fluid suspension, and the measurement arrangement may form a size indication by comparing the attenuation of the different wavelengths through the same section of the fluid suspension to each other.

[0113] This type of measurement allows the formation of a quality signal, which can extract flow velocity information about the fluid suspension. As a function of the quality and quantity measurements, a distribution can be formed, which can be used to form a characteristic value.

[0114] For particles that are primarily fibers, a fiber indicator may be used. For non-fiber or near-non-fiber particles, a size indicator or the like may be used.

[0115] Based on the fractions formed using a fiber index or a size index, etc., a secondary quality value can be determined. Size indices for the separate fractions may be formed, for example, based on the determined fiber index. Fiber indices for the separate fractions may be formed based on the determined size indices, where the fiber indices then represent, for example, the degree of crystallinity of each of the fractions. Fluorescence indices for the separate fractions may be formed based on the determined size indices, where the fluorescence indices represent, for example, the percentage or amount of fluorescent material in the sample.

[0116] Consistency measurements are important in pulp and paper manufacturing. Consistency is defined as the mass content of dry matter relative to the total volume. According to standardized methods (ISO, Tappi, SCAN), consistency is determined by filtering, drying, and weighing a sample. In many applications, the dry matter may contain components that differ significantly from one another. Components of different size, shape, and specific gravity pose considerable challenges, especially in optical consistency measurements. Measurements often require two separately measured variables: ash consistency, which represents the amount of filler, and fiber consistency, which represents the amount of wood-based material. These are then summed to obtain the total consistency. Separate measurements of ash and fiber consistency are difficult to implement as process measurements.

[0117] Fractionation allows different types of components in a sample to be temporarily separated and measured. A mixture of components with different responses that poses a traditional measurement challenge is then separated into suspensions of components with similar consistency responses, allowing truly component-specific measurements and calibrations. The new calculation method presented herein may be used to simplify the required mechanics of fractionation, thus enabling its use as part of a standalone process consistency concern. In particular, the calculation allows for separate consistency measurements of fines, ash, and fiber fractions, where fractionation allows for the separation of these components into their own fractions. The calculation also provides fraction-specific particle size information.

[0118] Here, we examine the effect of pulp type on fiber distribution. The test equipment used to study the pulp type response is a fractionator capable of measuring using a 950 nm wavelength band, measuring both depolarization and attenuation. Because only one wavelength is used, only a fiber index can be calculated from the results. However, as with other measurements described in this application, measurements can be performed using more than one wavelength band.

[0119] In the example test below, five different samples are included: -TMP waste refinery feed (TMP RJ feed), TMP after waste refinery (TMP RJ After), unsorted TMP (TMP), -pine pulp, and -Eucalyptus pulp (Euca).

[0120] Of these, dilutions of consistency were prepared ranging from 0.1% Cs to 1.6% Cs.

[0121] FIG. 24 shows an example of the distribution of a sample pulp at a consistency of 0.7%. The horizontal axis represents the fiber index, and the vertical axis represents the equivalent quantity value. The quantity value is the fraction-specific sum of the CsDA values according to Equation 7. Here, the distribution is divided into two fractions as an example, and the extreme values of the two fractions are set to a fiber index value of 0.6. This is an example. Thus, in this example, the fine fraction FR1 contains values less than 0.6, and the fiber fraction FR2 contains values greater than 0.6.

[0122] Figure 24 shows that refined mechanical pulp (TMP) contains a significant amount of fines, while chemical pulp contains less fines. The TMP sample in the center of the figure is the original unscreened refined mechanical pulp and has the highest fines content, while the fiber fraction contains a uniform mix of different fiber qualities in the fiber index range of 0.6 to 1.4.

[0123] In Figure 24, the defective pulp screened from the TMP is at the top. This pulp was designated the defective refinery feed sample. The fines content of the defective TMP is lower compared to the unscreened TMP sample. The fiber fraction still contains significant fiber particles, as indicated by the distribution value of the Fiber Index value of 1.3.

[0124] The second graph from the top in Figure 24 is an example of a sample taken after reject purification, where the amount of fibrous material has decreased and the amount of fines has increased equally.

[0125] Examples of chemical pulp samples are shown in the bottom of Figure 24. The second-lowest is pine pulp, and the bottom is eucalyptus pulp. Chemical pulp fibers are more uniform and fibrous than mechanical pulp fibers. This is evident in the higher Fiber Index values. The longer fiber length of pine pulp compared to eucalyptus pulp is evident in the weighting of the distribution toward higher fiber values. The fines fraction of the chemical pulp used in the discussion of these results contains a smaller amount of sample material in the 0.3-0.6 range. There are no particles in the range below 0.3, which can be interpreted as indicating that the fines in chemical pulp are of a completely different quality than mechanical pulp.

[0126] Figure 25 shows an example of the response of a sample pulp fines fraction as a function of total consistency. The dilution consistency response is obtained by integrating the fraction-specific distribution. The horizontal axis represents consistency, and the vertical axis represents the equivalent quantity value. The quantity value is the fraction-specific sum of the CsDA values according to Equation 7. The resulting fraction integral is then expressed as a function of laboratory consistency. Figure 25 shows an example of the response of the fines fraction FR1 for a sample pulp measured at different total consistencies. The high fines content of the unsorted TMP is visible as a strong response, and similarly, the low fines content of the pulp is seen as a low response. The effect of reject refining on fines is seen as a higher fines response of the refined pulp. It is noteworthy in this result that the response is zero-based and linear over the entire consistency range. This allows for a one-point calibration to be performed using only one confirmation value.

[0127] Figure 25 shows an example of the response of sample pulp fines fraction as a function of total consistency. The x-axis is consistency [%] and the y-axis is SUM DA FR1. In Figure 25, the abbreviation SUM DA FR1 means SUM CsDA of FR1.

[0128] Figure 26 shows an example of the response of a sample pulp fiber fraction FR2 as a function of total consistency. The X-axis is consistency [%] and the Y-axis is the sum of DA FR2. In Figure 26, the abbreviation SUM DA FR2 means SUM CsDA of FR2. The horizontal axis is consistency and the vertical axis shows the equivalent quantity value. The quantity value is the fraction-specific sum of CsDA values according to Equation 7. The response follows a logical pattern, with fines-free chemical pulp showing the highest response. The TMP response is lower than chemical pulp, with unsorted TMP showing the lowest.

[0129] Figure 27 shows an example of the total pulp consistency response, including the total combined response of sample pulp fines and fiber fractions as a function of total consistency, as well as the total combined fiber and fines response. The x-axis is consistency [%], and the y-axis is the sum of DA FR1 and FR2 on an arbitrary scale. In Figure 26, the abbreviation SUM DA FR1+FR2 means SUM CsDA of FR1 and FR2. The quantity value is the fraction-specific sum of the CsDA values according to Equation 7. The responses are very close to each other and are still linear and zero-based.

[0130] The results are promising considering that the response is not calibrated: the CsDA signal used in the measurement balances the inverse response difference between the fiber and the fines, which is usually a common problem in optical measurements.

[0131] The depolarization and attenuation used as the base measurements may be zero-based, and this characteristic may be transferred to the calculated CsDA signal. The fraction measurement method dilutes the sample, and the response remains at least approximately linear over a wide consistency range. Linearity and zero baseline allow for single-point calibration, allowing for simple, single-confirmation effects. The mutual similarity between the responses of the fines and fiber fractions in the CsDA signal allows for minimal measurement error, even without calibration.

[0132] Below, we will discuss the influence of different fiber types and the effect of calcium carbonate on the reaction.

[0133] In tests investigating the response of fiber type and calcium carbonate, long fiber pulp (LF), short fiber pulp (SF), and ground calcium carbonate (GCC) were mixed in various ratios. The sample names are indicated by the numbers preceding LF and SF, which indicate the relative proportions of the pulp samples to each other. The abbreviation GCC is preceded by a percentage indicating the relative share of filler in the total fines. For example, 0LF 100SF 50GCC means that 100% of the fibers in the sample mixture are short fiber and calcium carbonate makes up 50% of the total fines.

[0134] (Example) The overall consistency of all samples was constant at 0.5% Cs. When analyzing the results, the fines limit was set at a fiber index of 1.

[0135] Figure 28 shows an example of the fractogram and distribution of a pulp containing only long fibers. The majority of the pulp is the fiber fraction. The maximum fiber index is close to 10, and the weight is in the range of 6 to 9. The pulp contains different fibrous materials. The size index of the fiber fraction averages about 0.95, meaning that the fibers have relatively thick walls, or large diameters / widths, for example. The pulp also contains a significant amount of fines, with an average size index of about 0.7. The fines bar is white in the upper right graph and black in the lower right graph. The fiber bar is black in the upper right graph and white in the lower right graph. Here, the time scale from 0 to 0.06 is actually 0 to 60 seconds. The same applies to Figures 29 to 34.

[0136] Figure 29 shows an example fractogram and distribution for a pulp containing only short fibers. The fiber index for the pulp in the example of Figure 28 tends toward smaller values than the long fiber pulp. The size index of the fiber fraction is also slightly smaller for this pulp. The short fiber pulp has significantly more material in the fines fraction than the long fiber pulp, with a minimum size index of 0.45. This indicates that the smallest fines are smaller in size for this pulp compared to the long fiber pulp. The fines bar is white in the top right graph and black in the bottom right graph. The fiber bar is black in the top right graph and white in the bottom right graph.

[0137] Figure 30 shows an example of a fractogram and distribution for a pulp consisting of 90% short fiber pulp and 10% calcium carbonate. In the graph on the right, the fines bars are white and the fibers are black. The fines content, determined based on the fiber index, increased slightly, and the size index shifted toward slightly higher values.

[0138] The increase in the size index is due to the larger particle size of the added calcium carbonate compared to the wood-based fines. The calcium carbonate fraction and the fines fraction arrive overlapping in the flow, so both components are present simultaneously during the measurement. The measurement therefore represents both components on average.

[0139] Here, we consider what happens when calcium carbonate is added to long fiber pulp.

[0140] Figure 31 shows examples of size index distributions when calcium carbonate is added to long-fiber pulp at different ratios. The x-axis is the size index SI, and the y-axis is the SUM CsDA on a common scale. Adding calcium carbonate can change the size index distribution of the fine fraction in long-fiber pulp. Wood-based fines, represented by the white bars in the initial graph on the upper left, form a distribution ranging from 0 to 160. In the final graph on the lower right, at a GCC content of 45%, the distribution shifts to the 100 to 1500 range. The size index SI of the fines appears to rise, so to speak, on the opposite side of the size distribution of the fiber fraction depicted in black, due to the effect of calcium carbonate.

[0141] Below, we will consider TMP with the addition of eucalyptus pulp and kaolin.

[0142] For the test, a 60% TMP / 40% eucalyptus pulp blend was prepared. Kaolin was added to this blend at different ratios, and the consistency of each blend was set to 0.3% Cs. The fractogram and distribution of the pulp blend without kaolin at a consistency of 0.3% Cs are shown in the example of Figure 32. The example of Figure 33 shows the fractogram and distribution of a sample containing 50% TMP eucalyptus pulp and 50% kaolin.

[0143] In the fractogram graph, three different fractions are distinguished. Between 0 and 20 seconds of CsDA variation shown in the example in Figure 32, the fibers are mainly the largest and least processed fibers in the TMP pulp. Between times 20 and 30 seconds, the fibers are eucalyptus fibers and the fibers of the shorter fractions of TMP. Between times 30 and 50 seconds, the material is mostly TMP fines.

[0144] The CsDA variables shown in the example of Figure 33 correspond to otherwise identical fractions, but in the fine fraction, kaolin particles are present along with wood-based fines.

[0145] The size index of the measurements is displayed as separate graphs of the three fractions at the bottom right of the figure.

[0146] An example of the fractogram and distribution of a pulp mixture containing 60% TMP and 40% eucalyptus pulp at a total consistency of 0.3% Cs is shown in Figure 32. The CsDA in Figures 32 and 33 is shown as a function of the size index SI of fractions FR1, FR2, and FR3.

[0147] FIG. 33 shows an example of a fractogram and distribution of a pulp mixture containing 60% TMP and 40% half eucalyptus pulp and half kaolin at a total consistency of 0.3% Cs.

[0148] Next, the black liquid of dissolved lignin is evaluated. In addition to the Kappa number, it is extremely important in chemical pulping to measure the amount of dissolved lignin in the black liquid.

[0149] In a test to measure the lignin content of black liquor, 2.4% black liquor was added to a stock sample containing 0% black liquor.

[0150] The examples in Figures 34A and 34B show the effect of black liquid on the extinction coefficient fractogram. Figure 34A shows the original sample without black liquid, and Figure 34B shows the sample with 2.4% black liquid. ANir, which operates in the infrared range, does not respond as strongly to the addition of black liquid as AVis, which operates at blue wavelengths in the visible range.

[0151] The fines / black liquid content peaks in the attenuation curve for the sample without black liquid are set at the same time at both wavelengths. With the addition of black liquid, the AVis peak shifts to a significantly later time compared to the equivalent ANir peak (see dashed line between the peaks). Based on the sample without black liquid, it can be assessed that the AVis attenuation caused by particles peaks at a time of 40 seconds and decreases to zero at approximately 52 seconds. The sample with a high black liquid content peaks at a time of approximately 42 seconds. As the graph shows, there is a significant overlap between the particulates and the dissolved fluid in these measurements.

[0152] To be able to separately measure particle size, amount, and concentration and amount of dissolved material, it is necessary to separate the attenuation changes caused by the particles, on the one hand, and the medium absorption, on the other hand. This is best done if the particles and dissolved elements can be separated fractionally. Thus, by measuring attenuation, or scattering combined with some other measurement variable indicative of the presence of particles, it may be determined whether the attenuation at a flow point of interest represents the amount of particles or the amount of dissolved material.

[0153] In this case, the properties of the dissolved material may be determined by spectroscopy or other suitable methods of characterizing substances, and these properties may be used to identify the various material components.

[0154] If the fractions of fines and dissolved material after separation overlap, the content of materials that affect attenuation may be classified using known fines size distribution and attenuation at different wavelengths.

[0155] Although the attenuation coefficient values are mutually subtractable, the difference between the black liquid sample and the original sample is believed to significantly represent the attenuation caused by the black liquid.

[0156] By using longer NIR wavelengths instead of the currently used Anir wavelengths, it is possible to detect wavelengths that respond in a limited way or not at all to lignin, but only to particles, for example. Thus, the long-wavelength attenuation signal serves as a measure of solids, while the short-wavelength attenuation responds to both solids and dissolved materials. In such cases, if the effect of fines on the AVis signal relative to the ANir signal is known, the ANir signal allows for the calculation of the share of the AVis signal that comes from fine particles. The remaining AVis signal can then be interpreted as absorption caused by the dark liquid or some other material dissolved in the fluid. In this case, it is useful to know the response of the pure fine fraction, and the fractionation must be stable to the flow. The size distribution of the fine fraction is often quite non-constant in chemical pulping processes, making such calculations feasible.

[0157] It is also possible to determine the fines fraction using other measurements, such as scattering measurements, which indicate the presence of fines. If no scattering is detected in the measurement, it may be assumed that the attenuation in the measurement is significantly formed by absorption caused by dissolved material.

[0158] The benefits of chemical pulping process measurements can be summarized as follows: Fractionation allows washing the sample in a fractionation tube 10 for Kappa number measurement. This can be replaced by a separate washing unit in the measurement and pure fibers in water suspension may be obtained for Kappa number measurement. -It is possible to measure the lignin content of the solution. - Effective fractionation also allows for separate fines kappa number determination.

[0159] The tests related to the measurements at the deinking stage included samples taken from four different points in the flotation process. There were three to four parallel samples from every sample point. The measurements showed that the changes occurred mainly in the fines fraction. -Four parallel samples from the primary flotation receiving location - 3 parallel samples from the primary flotation feed location - Three parallel samples from the secondary flotation receiving position - Three parallel samples of the final pulp The changes in size index of the fine fraction FR1 appear to be more stable than those of the fiber. The size of the fine particles decreases until they undergo secondary flotation, then increases in the final pulp. The changes in the mean centroid value of the size index of the fine fraction are due to the removal of small particles from the fine fraction or material transferred from the fiber fraction to the fine fraction as it is refined. The high fine content of the final pulp may be due to screening before bleaching, which removes the smallest particles from the pulp.

[0160] Figure 35 shows an example of the change in size index of the fine fraction as a function of fiber index. The y-axis is the average size index SI and the x-axis is the fiber index FI. The centroids of the size index and the average fiber index of the fine fraction FR1 at various phases of the process are shown in relation to each other. As flotation progresses, the particle size of the fines decreases and fibrousness increases. The increase in fibrousness can be attributed to flotation removing larger non-fiber ink and dirt particles over smaller cellulose-containing particles. The increase in fibrousness and size index of the final pulp is likely due to sieving before bleaching, where the finest fines are removed from the pulp.

[0161] The refining effect will be examined below. In this test, the influence of refining intensity on particle properties was investigated. Figure 36 shows an example of fiber index at different refining intensities. The y-axis represents SUM CsDA on an arbitrary scale, and the x-axis represents fiber index FI. The fiber index distributions of pulp refined at three different refining intensities and unrefined pulp are shown. The fiber index distribution responds to changes in refining intensity.

[0162] Figure 37 shows examples of size index distributions at different refining intensities. The y-axis represents the SUM CsDA in an arbitrary scale, and the x-axis represents the size index SI. The size index distributions are shown for pulp refined at three different refining intensities and for unrefined pulp. The particle size index varies as expected: the more power used for refining, the more fragmented the particles become.

[0163] Figure 38 shows an example of the effect of refining intensity SRE on size index SI and fiber index FI. SRE stands for specific refining energy. With increasing SRE, both size index and fiber index decreased.

[0164] Fractionation also allows monitoring of fiber refinement based on fiber quality (fibrillation), fines accumulation, and possibly fines quality and fiber wall width.

[0165] In one embodiment, we consider the measurement of nanomaterials. There is a growing need to measure micro- and nano-sized particles. Nano-sized particles cannot be observed using conventional optical microscopes, so particle analysis is typically performed using time-consuming electron microscopes.

[0166] The measurements shown below are carried out with a micro-separator having a separation tube 10 with a diameter of approximately 5 mm.

[0167] Figure 39 shows an example of a fractogram and distribution of a well-mixed dissolving pulp sample. In the upper left of the figure, the x-axis is flow rate and the y-axis on the right is fiber index FI. In the lower left figure, the x-axis is flow rate and the y-axis on the left and right is size index SI. In the upper right figure, the x-axis is fiber index FI and the y-axis is total CsDA. In the upper right figure, the bar represents fraction FR1 because the other fractions FR3 and FR2 are absent or only slightly present. In the lower right figure, the bar represents fines because there are no or only very few fibers present.

[0168] In the fractogram and distribution, the dissolved pulp sample containing pure nanomaterial was thoroughly mixed before measurement. The fiber index was observed in the range of 0.1 to 0.2, i.e., very low. The size index distribution also had very low values, ranging from 0.2 to 0.35. The largest peak was at a value below 0.25, indicating a very small particle size.

[0169] The example in Figure 40 shows the same sample with very little mixing. In the top left image, the x-axis is flow rate and the y-axis on the right is the fiber index FI. In the bottom left image, the x-axis is flow rate and the y-axis on the right is the size index SI. In the top right image, the bar represents fraction FR1 because the other fractions FR3 and FR2 are absent or only slightly present. In the bottom right image, the bar represents fines because fibers are absent or only slightly present.

[0170] Here, the particles are not completely separated from one another and agglomerates are clearly detectable in the size index distribution. The distribution peaks at point 0.25, but the distribution continues down to a size index value of 1. Particles or agglomerates appearing in this range are likely in the micron size range.

[0171] The example in Figure 41 shows an example of a fractogram and distribution of a nanocrystal sample.

[0172] In the top left diagram, the x-axis is time (seconds) and the y-axis is the fiber index FI on the left and right. In the bottom left diagram, the x-axis is flow rate and the y-axis is the size index SI on the left and right. As in Figure 40, the bar in the top right diagram represents fraction FR1 because the other fractions FR3 and FR2 are absent or only present in small amounts. In the bottom right diagram, the bar represents fines because fibers are absent or only present in small amounts.

[0173] The crystallites have widths in the nanometer range and lengths in the hundreds of nanometers. The size distribution of this sample is in the range of less than 0.2. Here, the dimension that primarily influences the measurement is likely the nanocrystal length. If the measurement wavelength is short enough, crystallite width data can also be obtained, but this requires extremely short wavelengths.

[0174] Here, the strong effect of polarized light on increasing the fiber index value is of particular interest, as it allows for the determination of the crystallinity of nanocrystals, which is therefore of great need.

[0175] For nanomaterials, size indicators may be used as primary measurements, in conjunction with which other properties such as crystallinity are assessed by polarized light or by the presence of fluorescent components in the sample, for example by fluorescence measurements. Of course, other measurements that describe the properties of the sample can also be used as primary or secondary measurements.

[0176] Applications of the presented measurements to other industries are considered below. Such applications can be considered as embodiments of the measurements of the present application. In addition to pulp and paper production, measurements showing particle size distributions and particle concentrations are also required in other processes in mechanical process industries, such as mining, minerals, recycling, food, pharmaceuticals, and environmental technology. The pipe flow fractionation and related calculation methods presented herein are applicable to these industries. In addition to size distributions, other property distributions can also be measured using the methods described herein.

[0177] Calculation methods that take into account the properties of particles or fluids can simplify the technical setup and make the measurement results more robust, which allows the application of pipe flow fractionation to measure a wide variety of samples.

[0178] In principle, any sample containing particles can be fractionated. The sample does not have to be a suspension or dispersion; various powders can be dosed into the fractionation fluid. If the dose volume or mass is standardized, it may be used to measure content or concentration. However, these must take into account many practical considerations.

[0179] In addition to fluid media, pipe flow fractionation may be performed using air or other types of gas flows to measure dry samples, and the calculation methods described herein are also applicable to gas flow-based fractionation.

[0180] The presented measurements can be applied to environmental monitoring. The majority of environmental measurements are aqueous. For example, measuring the humus content of natural waters poses significant challenges. Humus can occur as aggregates of different sizes, which poses problems for traditional optical online methods, especially when measuring the entire sample. Measurements using fractionation and quality classification create new scope for measurements via the size distribution of humus aggregates. In addition to classification based on size indicators, the measurement method can include, for example, fluorescence measurements to measure aromatic components.

[0181] Measuring dissolved substances in water often requires a particle-free sample, which is particularly difficult for online measurements. Fractionation creates a particle-free sample, which allows the content of diluted substances to be measured.

[0182] The condition of natural waters and wastewaters is monitored, for example, by the analysis of various zooplankton and phytoplankton species. Such samples containing plankton can be classified into different classes that represent the condition of the water by means of size indicators and reference measurements. Reference measurements can include, for example, fluorescence measurements. Some plankton organisms are too small to be detected by light microscopy. However, this is possible by means of size indicator measurements based on light attenuation in their own fraction.

[0183] Fractionation-based measurements are necessary for the measurement of both industrial and municipal wastewater. Fractionation pretreatment can be used to analyze particle size in the flocculation process, and flocculant dosage may be adjusted based on these particle size data. On the other hand, when considering very small particles, the signal generated by free polymer in the size indicator distribution suggests polymer overdosing.

[0184] Oxygen demand measurements are an important aspect of nutrient loading analysis. These measurements require particle-free samples, which are difficult to obtain under field conditions. Fractionation produces at least a portion of a particle-free fractogram, which allows the proportion of particle-free dissolved material in the fractogram to be measured.

[0185] Fractional measurements can also be used to address the current major global problem of microplastic loads in different waters or sediments. Particles of many size classes are considered microplastics. The greatest measurement challenge is the smallest particles, some of which are submicrometer in size and cannot be observed with an optical microscope. Separating these particles by fractionation and applying measurements based on, for example, size indicators, allows the generation of size distribution data for samples containing microplastics. It is also possible to identify the small plastic fractions, for example, by fluorescence measurements.

[0186] Minerals may also be measured using the teachings described herein. Particle size distribution measurements based on pipe flow fractionation may be used in the crushing, classification, concentration, and granulation of minerals, with possible secondary measurements. In particular, suspensions containing small particles can be examined, providing additional value. Particle size distribution data facilitates process control and evaluation of partial process success or recovery.

[0187] For example, the manufacturing process of precipitated calcium carbonate (PCC) used as a filler can be enhanced using particle size distribution data.

[0188] The food industry can also benefit from these types of measurements: many particle size distribution measurement needs are met by fractional preparation in the food industry.

[0189] In emulsification processes, lipolysis monitoring may be performed, for example, by fractionation measurements. Other measurable property data include the amount and size distribution of casein micelles. Processing milk into cheese is also a process for which particle size distribution data is available. In addition to size distribution, dissolved materials can be measured in a manner similar to measurements in the pulp and paper industry. Fractionation-based measurements of properties and quantities can also be applied to several other areas of the food industry. For example, refractive index measurements to measure sugars are a potential application on their own or in combination with other measurements.

[0190] Furthermore, measurements in the petroleum industry may be carried out with the methods described herein. There are many applications in petroleum industry processes where fraction measurement methods can be applied. Fractionation is used to separate different fractions, the properties and amounts of which may be measured for the most part. The indication and size distribution analysis of particles in oil are possible with fraction measurement.

[0191] Likewise, this method is of interest for the measurement and control of pyrolysis processes, as well as in the pharmaceutical and health industries, where there is a considerable need for size distribution measurements, especially for the detection and analysis of nanomaterials.

[0192] Fractionation can be applied especially on a microscale when measuring various human and animal blood, urine, and other samples.

[0193] Fractionation may be applied to size distribution and color measurement of paint or printing color pigments, as well as many other analyses in the paint and printing industries.

[0194] The advantage of this method is that it provides a quick overview of the particle size distribution of a sample with a small sample size, allowing for real-time monitoring of the production process.

[0195] Future possibilities are discussed below: The measurements described in this application are primarily performed using optical attenuation and depolarization.

[0196] In fiber measurements, fiber indicators based on polarity level inversion are a usable basic signal, supported by the calculation of size indicators based on spectral attenuation. In addition to these, measurements based on the scattered light intensity offer other viable possibilities in detection techniques.

[0197] The scattering intensity can indicate whether changes in attenuation occurring during the measurement are due to scattering changes caused by particle changes or absorption by dissolved materials. On the other hand, spectrally performed measurements of scattering intensity allow the measurement of absorption changes at the surface or internal structure of particles. In pulp and paper samples, changes in particle surface absorption represent, for example, the brightness or modified Kappa number of fibers in pulp digests. Measurements of scattering intensity can be based on diffuse reflectance or direct scattering intensity measurements.

[0198] A potential optical measurement technique may be based on the fluorescence of light. The fluorescence effect can be utilized, for example, to measure the amount of lignin, which supports attenuation and scattering, or to measure itself. Another envisioned application of fluorescence is the indication and measurement of the amount of fluorescent brighteners. Fluorescence can also be used to identify and measure different grades of plastics, especially in the field of microplastic analysis.

[0199] The use of flow cuvettes allows the application of many other measurement techniques. Optical measurements may include, for example, diffraction methods based on optical interference, methods based on the Raman effect, and scanning methods. Methods based on the attenuation, scattering, diffraction, and retardation of microwave and acoustic waves are also possible.

[0200] Other contemplated detection techniques also include laser diffraction and charge measurement.

[0201] In some embodiments, the sensitivity of (current) optical measurements can be improved by utilizing measurement optics that measure as large an area as possible. This reduces measurement noise caused by flow. If slow flows can be used during the measurement, very stable measurements can be obtained even with difficult samples. Furthermore, by performing measurements on different signals, and as simultaneously as possible, noise is significantly reduced.

[0202] The challenge in fractionation is the similar fractionation of fillers and wood-based fines of the same size, and therefore they arrive at the same time during measurement. These particles are distinct from each other even under simultaneous measurement, but are difficult to easily distinguish from each other, so that the measurement results obtained represent the average properties and amounts of these particle classes.

[0203] There is a particularly strong need in the paper industry to measure the consistency of fillers and wood fines separately, and similar needs may exist in other industries as well.

[0204] In some embodiments, differences in particle specific weights can be exploited to solve this problem. For example, the starting section of the sorting tube 10 is placed vertically upward, so that lighter fine particles in the flow move faster in the flow direction than heavier filler particles. If necessary, the tube spirals downward, so that the correlation between the force vector parallel to the flow and the gravity vector controls the heavier particles at the end of the sorting tube 10, which therefore flows slowest. Therefore, the heavy filler particles arrive separately from the fine fraction, so that they can be measured as a separate fraction.

[0205] Another contemplated approach in one embodiment is to install the fractionator coil horizontally, with flow occurring perpendicular to the direction of gravity, so that gravity will pull heavier particles towards the tube wall, thus slowing the flow.

[0206] In some embodiments, a more viscous medium than water, such as ethanol, may be used to accelerate fractionation and achieve better separation. This may be done so that the more viscous medium is present only in the portion of the fractionation tube 10 through which the sample travels, while the remainder of the fractionation tube 10 is filled with water. In this case, the water pushes the ethanol and sample forward. Thus, the actual fractionation occurs in the ethanol environment.

[0207] In one embodiment, refractive index tuning can be used to determine the refractive index of a particle. The particle is immersed in a fluid of a certain refractive index and measured using a scattering compatible measurement. Samples where the refractive index of the fluid is close to that of the particle will produce the lowest scattering, which can be visualized as a low scattering intensity or a low attenuation value. The refractive index of the fluid that minimizes scattering is equal to the refractive index of the particle.

[0208] The immersion method does not work for samples containing particles with different refractive indices, as they will confound the measurement.

[0209] Fractionation can be used to separate particles with different refractive indices according to their different time in the flow. If fractionation is performed in a fluid with a refractive index similar to that of certain particles, these particles can be invisible in scattering measurements. In such cases, the only thing affecting the measurement is the absorption, which is different from that of the fluid. For example, by adjusting the refractive index of the fractionation fluid to correspond to the refractive index of wood-based fines, filler particles arriving at the same time as these fines can be visualized without the simultaneous effect of the fines.

[0210] The immersion method still requires adjusting the refractive index of the water or other fluid in the sample by changing the water or adding a refractive index adjusting component to the fluid. The immersion method is naturally more suited to dry samples than to wet samples.

[0211] For many reasons, particles in a sample often adhere to one another and become agglomerated. In many processes, such as paper machine retention conditioning, deinking of recycled pulp by flotation or wastewater flocculation, these aggregates or flocs can be promoted by various chemicals. In some situations, particles naturally aggregate.

[0212] The strength of the bonds that create these aggregates is often important data for process control or follow-up.

[0213] In tube flow fractionation, the turbulence of the water breaks up particle agglomerates. The stronger the bonds, the greater the force required to break them. This effect may be exploited, for example, by performing separate fractionations of a given sample under different flow conditions. In one embodiment, the flow conditions may be adjusted by varying the flow rate, fluid temperature, or fluid viscosity within the fractionation tube 10. The bond-breaking turbulence may also be adjusted by using, for example, different tube geometries, reducers, flow elbows, or other changes within the fractionation tube 10. The material selection for the inner surface of the fractionation tube 10 also affects the friction coefficient of the tube wall, which in turn affects the turbulence and the fractionation therethrough.

[0214] Figure 42 shows examples of fiber and size indices for refined chemical pulp at different fractionation rates. This is the test result for chemical pulp. The y-axis is SUM CsDA, and the x-axis is fiber index (left) or size index (right). The sample contains micro- and nano-sized materials. The sample was fractionated at three different pump rotation speeds and the same three flow rates. The results show a significant change in particle size, especially in the fine fraction. At a low flow rate of 6 ml / s, the fines have a size index peak value of 0.42 within the size grade. Similarly, the fiber index has a significant peak at a value of 1, below which there is no significant signal. With increased flow, the fine distribution becomes shallower, reaching an extremely low size index value of 0.35, indicating a very small particle size. This suggests that fines tend to agglomerate together, and this agglomeration sensitivity, and thus partial bond strength, can be measured by fractionating parallel samples of the same pulp at different flow rates. Another detectable change is a slight shift in the size index and fiber index distribution of the fiber fraction towards smaller particles. This is likely due to small particles adhering to the fiber surface, which remain within the fiber at low flow rates. However, as the flow rate increases, fines detach from the fiber. Agglomeration appears to have a similar effect on both the size index distribution and the fiber distribution.

[0215] These effects may be accentuated in the distribution maps generated by the new computational methods, whereas they are difficult to distinguish in conventional flow-based fraction distributions.

[0216] An example of fiber index and size index of refined chemical pulp at different fractionation rates is shown in Figure 42. The x-axis is fiber index and the y-axis is SUM CsDA.

[0217] In the embodiment shown in Figure 43, the statistical significance of fractionation measurements is improved by feeding several samples consecutively into the fractionation tube 10, while allowing enough space for samples to be fractionated without mixing with adjacent samples. The statistical significance of the distribution of these samples may be improved by summing them or taking their average. The black area at the top of Figure 43 shows the arrangement of multiple samples in the fractionation tube 10 during fractionation. The particle size distribution of the samples is generated by the flow in the fractionation tube 10.

[0218] To obtain a constant flow rate during the measurement, the sample supply may be synchronized based on the measurement signal. A new sample may be supplied to the beginning of the separator tube 10 after each measurement to identify water between the measured samples. The flow is always stopped at each new sample point, but without interrupting the measurement. To stabilize the flow rate, it is advantageous to prevail at the same flow rate during the measurement.

[0219] Figure 44 shows an example of separation of floc and short fibers based on their arrival order. In the example shown in Figure 44, the sample is divided into three fractions based on fiber index. The fraction ranges are FR1 = 0-2, FR2 = 2-6.5, and FR3 = 6.5-10. Material accumulates in the medium fraction FR2 at different times in the flow. These times are defined by the dashed rectangle 44 in Figure 44. Initially, in the range below 17 liters, the material consists of large fiber bundles. Next, in the range from 16.8 to 17.6 liters, pure fibers, classified as fraction FR3, arrive. After the range from 17.6 to 18.5 liters, the arriving material consists of smaller fibers, some of which may be non-fibrous. The accumulations produced by these two distinct materials, defined by the dashed rectangle 44, appear in the same fiber index range, despite their very different properties. If necessary, these materials may be separated into their own fractions by considering the order of arrival of the fractions, which eliminates the need for liter volume information, although if the measurement is performed with an instrument where the flow rate is stabilized and the flow rate is known, this may be utilized for separate fractions.

[0220] The process control device may be the same as the data processing unit 24 in Figures 46 and 47, or a separate control device for the process may be used to control the process measured using the flowing suspension. The process control device receives process information from a measurement device that measures the suspension flowing through a tube sorter (see Figure 49). The measurement device includes a radiation source that induces a wave motion in the flowing suspension, and the flow of the flowing suspension causes particles in the flowing suspension to be sorted based on size. The measurement device measures a first value of a first parameter of a wave that interacts with the flowing suspension in a first waveband of the wave. The measurement arrangement measures at least one second value: a second parameter of the wave that interacts with the flowing suspension, and a first parameter of the wave that interacts with the flowing suspension in a second waveband of the wave. The measurements of the first and second values are synchronized with each other. The measurement arrangement forms at least one comparison, each comparison relating to one of the first values and one of the second values. The measurement arrangement forms a distribution having one of the first values as a function of the comparison as process information. The process control device then controls the process based on the process information.

[0221] In the method of measuring a fluid suspension shown in FIG. 45, in step 10, waves are induced in the fluid suspension, and the flow of the fluid suspension sorts particles of the fluid suspension based on size. In step 12, a first value of a first parameter of the waves interacting with the fluid suspension in a first waveband of the waves is measured. In step 14, at least one second value is measured: a second parameter of the waves interacting with the fluid suspension, and a first parameter of the waves interacting with the fluid suspension in a second waveband of the waves. The measurements of the first and second values are synchronized with each other. In step 16, at least one comparison is formed, each comparison relating to one of the first values and one of the second values. In step 18, a distribution having at least one of the first values is formed as a function of one of the comparisons, substituting an expression for decay for consistency and avoiding dependence on flow and / or time.

[0222] In one embodiment, a first value of a first parameter dependent on the consistency of the fluid suspension may be measured. At least one comparison may be made dependent on a physical property of the particles of the fluid suspension.

[0223] In one embodiment, a tube flow fractionator may be used through which the flowing suspension flows.

[0224] In one embodiment, electromagnetic radiation may be directed at the fluid suspension and may include at least one beam that is polarized, or multiple beams, at least one of which is polarized and at least one separate beam that is unpolarized. The attenuation of the electromagnetic radiation interacting with the fluid suspension may be measured. A parameter of polarization of the electromagnetic radiation interacting with the suspension may be measured, and the attenuation measurement and the polarization parameter may be synchronized. A comparison of the polarization parameter and the attenuation parameter may be formed, and each of the parameters and each of the attenuations in the comparison may correspond to each other based on the synchronization. The attenuation may be distributed across the comparison.

[0225] In one embodiment, a size indicator may be formed by comparing the attenuation of different wavelengths interacting with the same section of the fluid suspension with each other.

[0226] In one embodiment, each of the parameters of polarization may be measured by measuring the degree of polarization or the degree of depolarization.

[0227] In one embodiment, optical radiation is directed at the fluid suspension, and the optical radiation may include at least one beam of polarized optical radiation, or multiple beams, at least one of which is polarized and at least one separate beam of which is unpolarized.

[0228] In one embodiment, the sum of the size indices may be corrected by the Euclidean distance of the corresponding attenuation and polarization parameters from each other.

[0229] In one embodiment, a comparison may be made between a polarization parameter and an attenuation parameter measured at the same wavelength. To form a distribution calculation, a sum of attenuation may be calculated for at least one of at least two value ranges of the polarization parameter, the attenuation being measured synchronously with the polarization parameter, the polarization parameter being within one of the at least two value ranges, the value ranges not overlapping. Each sum and each value range are associated with each other, and each value range is used to calculate the sum, forming a distribution of a first value sum as a function of the value range.

[0230] In one embodiment, Euclidean distances for the polarization and attenuation parameters may be formed, and the association is performed by correlating each of the attenuation sums and each of the Euclidean distances, each corresponding to said sums, with each other to form a distribution of the attenuation sums as a function of Euclidean distance.

[0231] In one embodiment, forming the consistency of one or more fractions may be formed by integrating the sum over one or more of the ranges of values within the fraction.

[0232] In one embodiment, an average centroid value of the size index of the fines fraction may be formed to determine the ash content of the fines fraction.

[0233] Figure 46 shows an example of an instrument for measuring a fluid suspension. The instrument has a radiation source 20 that induces waves into the fluid suspension in the sorting tube 10, or into the fluid suspension in the tube after it is ready for fractionation by the sorting tube. The waves may be, for example, electromagnetic radiation or acoustic radiation. The flow of the fluid suspension causes the particles in the fluid suspension to be sorted based on their size.

[0234] Figure 46 shows a wave passing through a fluid suspension. It is also possible to receive waves in a direction that deviates from the direction of propagation or wave motion.

[0235] The measuring device 26 includes a detector 22. The detector 22 may be, for example, a semiconductor detector. In one embodiment, if the wave is optical radiation, the detector 22 may include, for example, at least one optical sensor or spectrometer. In one embodiment, if the wave is acoustic radiation, the detector 22 may include, for example, at least one acoustic transducer. The detector 22 may measure, for example, a first value of a first parameter of the wave in a first wavelength band of the wave and a second value of the wave interacting with the wave suspension in the first wavelength band of the wave. In this case, the parameter is common to both values, since it is a matter of wave attenuation, but the values are based on different wavelengths. Alternatively, the second value may be a second parameter of the wave interacting with the fluid suspension that is different from the first parameter. The parameter may, for example, be related to attenuation and polarization. The measurements of the first and second values are performed in a synchronized manner, or the measurement results are synchronized in the data processing unit 24 of the measurement arrangement 26. Synchronization means that the values being compared are measured from the same sample at the same instant. When measurements of the comparison values are made at different times, synchronization means that the same sample is measured at a first instant, the sample is moved a distance, and then the sample is measured at a second instant. The term "same instant" here means that the measurements of the comparison values are made at different times, but the time difference between the measurements is so small that the measured samples are the same within an acceptable range. That is, the sample does not move significantly, and the effect of the movement on the measurements is below noise or a desired measurement tolerance. Alternatively, the detector 22 may move a distance relative to the sample between measurements of the comparison values.

[0236] For wavelength-based attenuation measurements, the detector 22 may have a filter 34 to allow a suitable wavelength band to pass through the detector 22. In an embodiment, an example of which is shown in FIG. 48, the detector 22 has a detector element 36 for each measurement wavelength band. In one embodiment, the detector 22 has one detector element and a filter temporarily positioned in front of the detector element to measure the first and second values. That is, the filter may move from or towards the detector element.

[0237] In one embodiment, the radiation source 20 may transmit two different wavelength bands simultaneously or sequentially in time. The detector 22 may then be used without the aforementioned filters. The radiation source 20 may have one or more radiation source units 38 that output waves at one or more wavelengths.

[0238] Those skilled in the art are familiar with measuring the attenuation of wavelength bands per se, which is why Figure 48 is an overview of some possibilities.

[0239] In polarimetry, detector 22 may have one or more polarizing filters in front of detector 22. In this example, filter 34 in FIG. 48 can be considered a polarizing filter. The degree of rotation or polarization of polarized light may be measured using one or more detector elements 36 of detector 22. At the moment of measurement, the suspension may be flowing, or the suspension may be stationary or non-flowing.

[0240] At least one sensor 28 may measure, for example, consistency, temperature and / or flow (velocity / speed). The sensor 28 may be a semiconductor sensor. Those skilled in the art will be familiar with various types of sensors 28 as such. A clock may also be present to measure time.

[0241] The data processing unit 24 then forms at least one comparison, each comparison being configured to relate one of the first values and one of the second values, the comparison being based on equations (2)-(6).

[0242] The measurements of the first and second values to be compared may be performed at the same moment or at successive times. Different wavelengths may be detected simultaneously using two or more detector elements, or a common detector element may receive different wavelengths successively through different filters. Each filter is configured to filter the incident wave at successive times. That is, the filters may be changed one by one in front of the detector 22 as a function of time. However, if the radiation source 20 outputs different wavelengths as a function of time, filters are not necessary.

[0243] Finally, the data processing unit 24 is configured to form a distribution having at least one of the first values as a function of one of the comparisons, as shown in Figures 13-16 and described herein.

[0244] 47 shows an example of a data processing unit 24, which may include one or more processors 30 and one or more memories 32 containing suitable computer program code. The one or more memories 32 and computer program code, together with the one or more processors 30, cause the data processing unit 24 to perform the measurement steps described herein.

[0245] An example control arrangement is shown in Figure 49. The process control device 42 may be the data processing unit 24 or a separate controller, which, like the data processing unit 24, may contain one or more processors and one or more memories containing suitable computer program code. The process control device 42 controls the process 40 based on the distribution having at least one of the first values as a function of one of the comparisons. Examples of the process 40 can be found in pulp and paper manufacturing, as well as other processes in the mechanical process industries, such as the mining and minerals industry, the recycling industry, the food industry, the pharmaceutical industry, and also in the environmental technology industry, for example, as previously mentioned.

[0246] The presented measurement method embodiments may be implemented as logic circuit solutions or computer programs. Correspondingly, process control may utilize measurement information for process control and may be implemented as logic circuit solutions or computer programs. The computer program may be located and distributed on a computer program distribution means that is readable by a data processing device and encodes computer program commands to perform measurements and, if necessary, control the process based on the measurements.

[0247] The computer program may be distributed using a distribution medium, which may be any medium readable by a controller. The medium may be a program storage medium, a memory, a software distribution package, or a compressed software package. In some cases, distribution may be performed using at least one of near-field communication signals, short-range signals, and telecommunication signals.

[0248] It is obvious to those skilled in the art that with the advancement of technology, the concept of the present invention can be implemented in various ways. The present invention and its embodiments are not limited to the exemplary embodiments described above, but may vary within the scope of the claims.

Claims

1. An apparatus for measuring particles in a flowing suspension that are separated on the basis of size by flow of the suspension through a pipe flow fractionator, comprising: a radiation source for irradiating particles of the fluid suspension with optical radiation having at least two beams, the at least two beams comprising a first beam, a second beam having a different wavelength band from the first beam, and / or a third beam having a different polarization state from the first beam; A measuring means for calculating a time series change in a size index or a fiber index, measuring a first parameter related to attenuation due to interaction of the first beam with the fluid suspension; and measuring a first parameter related to attenuation upon interaction of the second beam with the fluid suspension and / or measuring a second parameter related to polarization upon interaction of the third beam with the fluid suspension; a measuring means; and the size indicator is calculated based on a first value of the first parameter for the first beam and a second value of the first parameter for the second beam; the fiber index is calculated based on the first value and a third value of the second parameter for the third beam; The apparatus, wherein the measuring means is configured to calculate a distribution of at least one of the size indicator or the fiber indicator.

2. the measuring means is configured to form a comparison measured using wavelengths having a known relationship to one another; 2. The apparatus of claim 1, wherein to perform said distribution, said measuring means is configured to calculate a frequency distribution of said fiber indicators.

3. 1. A process control instrument for controlling a process measured using a fluid suspension configured to sort particles of the fluid suspension based on size, comprising: the process control device receives process information from a measurement device measuring the suspension flowing through the tubular separator; The measuring device is a radiation source for irradiating particles of the fluid suspension with optical radiation having at least two beams, the at least two beams comprising a first beam, a second beam having a different wavelength band from the first beam, and / or a third beam having a different polarization state from the first beam; A measuring means for calculating a time series change in a size index or a fiber index, measuring a first parameter related to attenuation due to interaction of the first beam with the fluid suspension; and measuring a first parameter related to attenuation due to the interaction of the second beam with the fluid suspension; and / or measuring a second parameter related to polarization upon interaction of the third beam with the fluid suspension; a measuring means; and the size indicator is calculated based on a first value of the first parameter for the first beam and a second value of the first parameter for the second beam; the fiber index is calculated based on the first value and a third value of the second parameter for the third beam; the measuring means is configured to form a distribution of at least one of the size indicator or the fiber indicator; The process control device is configured to control a process based on the process information.

4. A method for measuring particles in a flowing suspension that are separated on the basis of size by flow of the suspension through a pipe flow fractionator, comprising: irradiating particles of the fluid suspension with optical radiation having at least two beams, the at least two beams comprising a first beam, a second beam having a different wavelength band than the first beam, and / or a third beam having a different polarization state than the first beam; measuring a first parameter related to attenuation due to interaction of the first beam with the fluid suspension; - measuring a first parameter related to attenuation upon interaction of said second beam with said fluid suspension and / or measuring a second parameter related to polarization upon interaction of said third beam with said fluid suspension; A step of calculating a time series change in a size index or a fiber index, the size indicator is calculated based on a first value of the first parameter for the first beam and a second value of the first parameter for the second beam; the fiber index is calculated based on the first value and a third value of the second parameter of the third beam; calculating a distribution of at least one of the size indicator or the fiber indicator; A method comprising:

5. 5. The method of claim 4, further comprising measuring each of the second parameters related to polarization by measuring a degree of polarization or a degree of depolarization.

6. The method of claim 4 , further comprising correcting the distribution of the fiber indices based on the Euclidean distance of the attenuation and polarization parameters corresponding to each other.

7. and forming a comparison between the polarization parameter and the attenuation parameter measured at the same wavelength; The method of claim 4 , wherein a frequency distribution of the fiber indices is calculated to form the distribution.

8. moreover, forming a Euclidean distance between said second polarization parameter and said first attenuation parameter; and The method of claim 7, wherein a frequency distribution of the fiber indices is formed as a function of the Euclidean distance.

9. 8. The method of claim 7, further comprising forming a consistency of one or more fractions by a frequency distribution of the fiber indicator across one or more fractions.

Images