LT5733B - MINERALINES VATOS GAMINIU FIZIKINIU IR MECHANINIU SAVYBIU NUSTATYMO BuDAS - Google Patents

Abstract

The invention relates to analysis of mineral wool as fibrous insulating material by not destroyed methods and to prognosis of primary characteristics of mineral wool products. Method for determining of physical and mechanical characteristics of mineral wool products comprises a scan of all û over or part of cross û section over a thickness of sample, a processing of scan picture with software, new is that after determining moving direction of fibre layer scan two cross û sections of sample: a parallel to moving direction of layer (L) and a perpendicular to moving direction of layer ( C ), scan pictures are adjudicated with software, values of single angles and averages are calculated, results of experiment are formulated by indicator S. Mentioned indicator determinates in cross û section of sample ratio of orientation angles of dominant fibres. Mentioned angles are calculated with regard to primary axes and to perpendicular to that. S is calculated separate for sections L and C, respectively SL and

Description

Technical field

The present invention relates to the study of the structure of mineral wool as a fibrous insulating material by non-destructive methods and to the prediction of the basic properties of mineral wool products.

State of the art

Structurally, mineral wool can be described as a three-dimensional system consisting of a plurality of fibers arranged in a certain order relative to one another and bound together (usually at the fiber contact points) by a binder. The individual elemental fibers, the nature of their spatial orientation in the product volume, the presence of binders determine the structure of the insulating materials and the main characteristics of the products. Thus, mineral wool products are investigated at micro- and macro-structural levels.

Microstructure is usually studied using microscopic examination methods. Usually microscopic investigations determine the linear (thickness, length) dimensions of mineral wool single fibers, their surface area, peculiarities of contacts (bonding area, angles, etc.), and observe the fiber surfaces and record changes in various tests (eg by examining their durability). years. The microstructure of mineral wool mainly influences the thermal properties of the products.

However, research with a variety of microscopic devices has a problem with the limited area of the structure being analyzed. The observations made allow the identification and characterization of only the finest features of the microstructure of the fibrous product and, at best, covering a dozen fibers or an area of 3-5 mm ² .

As a result, the distribution of fibers or the predominant part of the fiber in the mineral wool structure cannot be discerned. Other research methods and devices are required to analyze the macrostructures of fibrous materials.

The macrostructure of a material is a structure with elements or pore sizes of about 10 * ⁵ -10 ' ³ m, so it can be studied with or without low magnification optical means. Visual inspection allows to distinguish mineral wool products according to the fiber orientation in the structure. The directionality of the fibers or groups of fibers, the heterogeneity of the structure (areas with little or no organic matter) and the hardened larger inclusions of unfiltered particles are clearly visible. Digital and computer tools are used to capture and analyze the macrostructure of fibrous materials. Macrostructure (larger areas) analysis can be done by surface scanning and image analysis techniques. For scanning using flatbed scanners, and for image analysis using developed computer programs. Different types of scanners allow you to transfer scanned images of various sizes (up to 420x210 mm), which solves the problem of insufficient area to be analyzed.

International application WO 2003/054270 describes a method for manufacturing mineral fiber insulating two-density panels. Tau (T) and Kappa (K) indices are used to describe the dominant orientation of the fiber. Indicators are obtained by scanning portions of each relevant cross-section over the entire thickness of the layer and performing a fast Fourier transform of the data. The production process is adjusted to the selected sizes of Tau and Kappa.

The indicators for You and Kappa and how they are calculated and determined are not accurate enough.

International Application WO 1999/056112 describes a method for determining the fiber structure of mineral wool mats. In this way, the fiber structure is illuminated diagonally by the determination of the fiber structure, and at least one image of this area is recorded on a video camera perpendicular to its surface. Each pixel is assigned a digital signal corresponding to its luminous intensity, digitized either directly in the camcorder or in a subsequent digitization step. In this way, the percentage and direction of the vertical fibers are determined using an image processing system based on digital signals of light density. This method is used in manufacturing to obtain data on the relationship between mechanical and / or thermal properties of a mineral wool mat and the determined percentage and directional values of vertical fibers and to automate the production of mineral wool mats and automatically adjust the longitudinal compression speed of the reordering machine. .

And this method of determining fiber structure is not accurate enough.

The essence of the invention

The invention solves the problem of determining and calculating the distribution of mineral wool fibers. The present invention enables the non-destructive method of predicting the compressive strength properties and the macrostructural characteristics of mineral wool products to express the degree of anisotropy of the mineral wool structure.

The proposed method for determining the physical and mechanical properties of mineral wool products, comprising scanning the whole or part of the cross-section through the thickness of the specimen, software scanning the scanned image, is that two specimens are scanned once parallel to the direction of motion (L) and perpendicular to the direction of motion of the web (C), the scanned images are analyzed by software for the analysis of images, objects and backgrounds, the individual angles values and their mean are calculated, and the test results are expressed as macrostructure index S. axis and perpendicular to it. S is calculated separately for the sections L and C, Sl and S _c , respectively, and the functional relationship between the physical and mechanical properties and the macrostructural characteristics Sl and S _{c of} mineral wool products is given by:

σ _θ = -107.77 + 0.51 ps + 5.93 M + 54.81 S _c + 7.02 S _L where σ _β is the compressive stress, ps-density,

M - organic matter content.

As the angles of the dominant fibers differ in some samples in the L and C sections, the calculated macrostructure indices are also different in each section. In the case where no way to the conveyor direction and correctly marked incision edges may be calculated macrostructure indicators Sl and S _c Average SL-C and the mineral wool products of physical and mechanical properties and macrostructure indicator S _L -c functional dependence is determined by the equation :

σ _θ = −140.66 + 0.55 p _s + 8.06 M + 82.35 · S _{L. C.}

The magnitudes of the macrostructure index Sl-c indicate the distribution of the fibers of interest in the structure of the product relative to the main axis (coinciding with the direction of movement of the conveyor): when S _L -c 0.75, then we have an apparently horizontal fiber orientation; when S _{L. C} = 0.76 - 1.09, then we have the supposedly chaotic fiber orientation, and when S _L -c £ 1.10, we have the supposedly vertical fiber orientation.

Brief description of the drawings

The invention is explained in the drawings in which;

Fig. FIG. Figure 1 is a schematic of marking and scanning sample surfaces;

Fig. FIG. 2 - scanning the specimen surface with a scanner;

Fig. FIG. 3 - Image Tool window view: 1 - Analyzed objects (characteristic fibers and their groups) marked on the scanned surface; 2 10 calculation results; calculated angles and their average size;

Fig. FIG. 4 - Schematic diagram of the L section of the specimen: a - direction of movement of the fiber web; b is the direction of the determination of the inclination of the fibers in relation to the x-axis; c a direction for determining the inclination of the fibers with respect to the x 'axis;

Fig. FIG. 5 - computed macrostructure index S _L. Distribution of _C values 15 for all types of specimens;

Fig. FIG. 6 - diagram of macrostructure dependence and density of compressive stress limit;

Fig. FIG. 7 - graph of load-deflection curves depending on direction of compressive load: 3 - along the direction of movement of the fiber blanket (sample MVV-1.4 / 8), 4 - perpendicular to the direction of motion of the fiber blanket (sample MVV-1.3 / 7), 5 surface (sample MVV-1.2 / 5).

Description of Implementation of the Invention

The dominance of the orientation of fibers and their groups in the structure was determined by the principle of 25 digital capture and analysis of macrostructure. Fragments of the surface of mineral wool slabs were scanned and the captured images transferred to a computer. The location of the fibers is then investigated by computerized digital image analysis.

The specimens are cut from the plates with smooth edges and in a way that minimizes the "surface fouling". Before scanning, it is desirable to determine the direction of movement of the fiber web by the conveyor and mark the surface accordingly: section L - parallel to the direction of movement, section C - perpendicular to the direction of movement (Fig. 1). The direction of movement of the fiber is indicated by an arrow.

The edges of the specimen were scanned with a flatbed scanner (used in the Hevvlett Packard ScanJet 7400C), transferring the surface image to a digital image. Place the specimen on the scanner in such a way that the x-axis of the test specimen coincides with the scanning direction of the scanner and the other side of the specimen is perpendicular to the direction of the light emitted by the scanner (Figure 2). Scanning is done by scanning all or part of the cross-section through the thickness of the specimen. The analyzed surfaces are scanned at 120-400 dpi in black and white, and the image is saved in a BMP, TIF, GIF, or JPEG image size of 100 * 500 pixels or other. UTHSCSA Image Tool 3.0 for Windows is used for image processing, which is used for analysis of images, objects and background in microscopy, medicine, metallurgy and many other fields. This program can be found online at http://ddsdx.uthscsa.edu/dig/download.htm/.

Because black-and-white images are valued, their color gamut in gray color ranges from black (value 0) to white (255).

In a computer image, all objects in a gray scale are marked in the exploratory image, and the areas occupied by them represent macrostructure components (skeleton and pores). White represents fibers and / or groups of fibers and pores and voids are black. Usually white fibers that come in direct contact with the scanner surface, the varying coloration indicates the distance of objects from the surface or the transition from one structural component to another.

Before starting work, it is necessary to select the appropriate settings in the Image Tool (Settings -> Preferences).

The images were processed in the following order:

a) When you run the "Image Tool" installed on your computer, a window opens to load the image to be analyzed.

(b) automatic brightness and contrast adjustment of the image;

c) in the figure objects are found and marked;

d) In the Find Objects window that opens, check Automatic;

e) after the program has numerically marked each object (Fig. 3, object 1 is analyzed), the analysis is continued by counting the angles;

f) Once the calculations have been completed, the results of the individual measured angles and their mean value appear in the window (Fig. 3, angle values 2).

We then proceed to analyze the specimen rotated at 90 ° according to (a) to (f) above.

In the traditional method of manufacturing mineral wool, most fibers are placed horizontally on a conveyor belt. However, due to the specific nature of the production process, a certain proportion of the fibers are oriented "randomly". The properties of mineral wool and other fibrous materials are a function of the orientation of the fibers, since their orientation determines the size of the physical-mechanical properties of the products.

Fiber Structure Management - By controlling the fiber orientation in the fiber web stage of the manufacturing process, it allows the production of mineral wool products with desired thermal, strength and deformation properties.

The determination of the dominant fiber orientation in the product structure using the 10 macrostructure indices allows quantification of mineral wool products according to the fiber orientation.

As the orientation of all single fibers in a relatively large area is almost impossible to measure, it is assumed that the average tilt angle of the macrostructure-dominated fiber groups relative to the axis under investigation is measured.

According to the authors (Οκπμ3κοβ 1966; 5o6poB 1987), the fibrous structure can be represented as a complex lattice system consisting of many elementary systems stacked on top of one another (graphically represented as a triangle on the two coordinate axes) with angles a and β ranging from 0-180 °. . Each such system consists of structural elements: fiber, small particles of organic matter and non-fibrous melt inclusions.

In the paper (Pourdeyhimi and Kim 2002) it is stated that the orientation of fibers in nonwovens can be defined by the orientation distribution function ψ, therefore ψ is also a function of angle a. The integral of this function ψ between the angles ai and a ₂ is equal to the probability that the fiber is located between the angles a ₁ and a ₂ .

In the source (Rao at ei. 1991) we find that the orientation of single fibers can be expressed by the direction vector p. This vector corresponds to the geometry of the fiber with respect to its axis. Since the fiber orientation is analyzed in the plane (2D), the direction vector consists of the components pi and p ₂ corresponding to the tilt angle <j>

with respect to the x-axis:

P> cos </> J> 2. sin </>

(1)

Thus, as the fiber orientation angle changes, the positions of the vector components with respect to the axes immediately change.

The macrostructure index (S) is used to express the dominant fiber orientation and directionality. It expresses the ratio of the angles of orientation of the dominant fibers and their groups measured in the flat section (through thickness) of the sample to the principal axis (x) and perpendicular to it (x '), in which case the x' axis is obtained by The major axis x is assumed to coincide with the direction of the fiber web movement on the conveyor, so that the orientation of the fibers in the product is described in this context (Fig. 4 (a)).

The macrostructure index (S) is calculated using the formula:

S = - ^ «xx: a _x is the mean angle of the x-axis distribution of the dominant fibers;

σ _χ · - size of the middle angle of the distribution of the dominant fibers perpendicular to the x axis.

Expression of the macrostructure index over angles increases the accuracy of measurement (especially in products with a chaotic fiber orientation), since dominant fiber angles are measured in two directions: parallel (Fig. 4 b) and perpendicular (Fig. 4 c) to the x axis. With the directional orientation of the fibers, their distribution with respect to the axes is more evident. Some specimens showed different fiber distributions in both slices (parallel (L) and perpendicular (C) to the direction of motion of the fiber web), and therefore the macrostructure indices were calculated separately with the corresponding slice letters (L or C).

Because calculated and Sc values differ for the same specimen, the average of the indices was calculated instead of the macrostructure indices for single slices to eliminate the probability of a false result if specimens were improperly labeled. Oriental mineral wool products with a density of 33-200 kg / m ³ . Visual inspection of the specimens reveals a random (chaotic) or directional arrangement of the fiber groups in the structure.

The mean values of the dominant fiber tilt angles and macrostructure indices measured and calculated in the experimental studies are shown in Table 1.

table. Average values of fiber tilt angles and macrostructure for all types of specimens

Sample type Harvest L Section C Sl Sc Sl-c a _x ^a a _x ° a _x ° a _x ° MVV-1.1 32.12 52.20 31.96 51.96 0.62 0.62 0.62 MVV-1.2 41.02 45.38 37.68 49.05 0.90 0.77 0.84 MVV-1.3 46.16 39.48 48.12 37.53 1.17 1.28 1.23 MVV-1.4 45.78 40.77 40.44 45.73 1.12 0.88 1.00 MVV-1.5 42.35 43.68 49.16 35.94 0.97 1.37 1.17 MVV-1.6 37.07 46.51 35.23 48.62 0.80 0.73 0.76 MVV-1.7 39.48 44.82 36.84 48.13 0.89 0.77 0.83

From the results in Table 1, we can see that the calculated mean angles for the x and x 'axes differ for different types of specimens. The largest angular differences were measured for specimens which, when viewed visually, exhibit clear fiber directionality. For MVV-1.1 specimens, the difference between the mean angles a _x and σ _χ <was 20.08 °. And for MVV-1.5 specimens, the mean angle difference between a _x and α _χ · was 13.22 °. From all specimens, the maximum angle difference was measured for one of the MVV-1.1 specimens (section C) and reached 29.19 °.

Below we present 7 visual descriptions of one specimen (with calculated angles and macrostructure indices that are close to the average for that type) from each MW-1.1 to MW1.7 specimen type.

Visual inspection of the MVV-1.1 / 1 specimen showed that the structure is dominated by the horizontal orientation of the fibers and that the distribution of the fibers with respect to the x-axis is similar in both sections (L and C). This is confirmed by the measured close angles of a _x (31.3 ° in section L and 33.48 ° in section C). Meanwhile, the angles of a _x · were slightly different: in section L, the average a _x - = 52.54 ° was measured, whereas in section C a _x · was 49.89 °. Calculation of the macrostructure indices in the respective slice yielded similar values: both slices (S _L = 0.6 and S _c = 0.67). In the following, only the macrostructure indices are used to describe the peculiarities of fiber distribution of the different types of specimens, not the numerical angles.

Meanwhile, in the MVV-1.2 / 5 specimen, different fiber distributions were seen in sections L and C: section L is dominated by fibers of a more chaotic orientation, while section C shows a larger amount of horizontally distributed fibers. Therefore, the calculated macrostructure indices (0.9 for slice L and 0.78 for slice C) differ accordingly.

For the MVV-1.3 / 7 sample, the measured angles σ _χ and a _x ′ and the calculated and S _c values were very large (1.16 and 1.32) compared to the samples analyzed previously (MW-1.1 and MW-1.2). calculated macrostructure indices. In the specimen, both slices show the directional dominance of the vertically located fibers in the structure.

The MVV-1.4 / 8 sample exhibits quite different shapes: the L-section clearly shows the orientation of the fibers in the vertical direction (since S _L = 1.11), and the C-section does not clearly show the distribution of dominant fibers on any one axis, so the macrostructure index Sc = 0.89.

The structure of the MW-1.5 / 8 specimen is characterized by a very close distribution of fibers as in the MW-1.3 specimens. The MW-1.5 specimen has a L-section dominated by the structural chaosicity of the fibers (Sc = 0.96) and a C-section showing a very well-ordered vertical orientation of the fibers. This is confirmed by the estimated maximum size of the macrostructure (in that section) (1,4) ·

In the MVV-1.6 / 2 and MVV-1.7 / 2 specimens, a chaotic fiber orientation was seen in both (L and C) slices. Although the estimated small (0.75; 0.77) macrostructure indices show a "horizontal" tendency of fiber distribution.

The averages, standard deviations, and minimum-maximum size distributions of the estimated macrostructure indices are shown in Figure 5.

The distribution of the S / .. c values in Figure 5 shows that the direction of the dominant fibers in the structure can be characterized by a macro-structure indicator. For MW-1.1 specimens with a more dominantly oriented fiber structure, the S _L .c index is low (0.5-0.73) and the estimated coefficient of variation is within ± 14.8%. In the MVV-1.3 and MW-1.5 samples, where vertical orientation fibers are predominant, Sl-c is high (1.12-1.31) but the coefficient of variation is rather low ± 3.5-4.4%. From these coefficient of variation coefficients we can say that the structure of the samples with vertical fiber orientation is more homogeneous. The S _L -c index of the remaining, chaotic structure samples varies between the minimum and maximum values of the directional structure (0.76-1.09) due to the absence of a clear fiber orientation.

Thus, we can state that the macrostructure indices S _G S _c , S _L -c are suitable for the quantitative evaluation of fiber orientation in mineral wool structure. Based on the obtained values, the distribution of dominant fibers can be determined.

The classification of products of a homogeneous structure is based on structural differences due to the orientation of the fibers. The numerical values of the macrostructure indicator used to prove this have been calculated from the results of experimental studies after testing products of different orientation and structure. 5 Based on the size of the macrostructure index, we can apply a conditional classification to products of different structure, taking into account the distribution of the fibers dominating the structure with respect to the main axis (Table 2).

table. Division of mineral wool products of different structure into groups according to the size of the macrostructure indicator

"Apparently horizontal orientation" "Allegedly chaotic orientations" "Purportedly vertical orientation" for Sl-c 0.75 when Sl-c = 0.76-1.09 when Sl-c 1.10

The characterization of the fiber position in the product structure was selected based on the position of the dominant fibers with respect to the direction of action of the compressive loads and the direction of fiber web conveyor movement (production line direction). For example, horizontal directional fibers in the product structure are perpendicular to the direction of load application, while the direction of the dominant fibers coincides with the movement of the fiber web, i.e. in the horizontal direction.

It can be seen from Table 2 that products with different structures, with the largest number of fibers arranged parallel to the x-axis, can be referred to as "supposedly horizontal orientations". Products predominantly perpendicular to the x-axis are termed "supposedly vertical". Conversely, products with randomly positioned fibers or fibers with an indeterminate orientation and / or orientation are referred to as "allegedly chaotic orientations". The generally accepted fiber structure partitioning is more conditional, since ideally one or another structure product is practically not produced.

Thus, by scanning the product surfaces and analyzing the resulting image and calculating the average angles of distribution of the dominant fiber structure and calculating the macrostructure index, it is possible to quantify the product type according to the fiber structure orientation.

The classification facilitates the appropriate assessment of mineral wool products of different structures, addressing the problems of increasing the efficiency of traditional thermal insulation materials. Because fiber structure control, fiber orientation change and the introduction of different densities and / or layers in the manufacturing process determine the basic strength and deformation properties of mineral wool products (Table 3).

The mean values of density, compressive stress, organic matter and macrostructure measured during the tests are summarized in Table 3 5. The determination of the compressive stress limit (σ _β ) was chosen because of the different deformation of the specimens of different structure: for some, σ / ο is measured and for others, ū _m .

table. Density of specimens, tensile stress, organic matter content, and

macrostrukl average values of volume indices Samples Density (in compression tests) (Ps), kg / m ³ Compressive stress limit (σ _β ), kPa Organic matter content (M),% Macrostructure indicators S _t Sc Sl-c MVV-1.1 48.9 2.6 4.15 0.62 0.62 0.62 MVV-1.2 95.2 11.5 3.70 0.90 0.77 0.84 MVV-1.3 98.8 46.8 3.66 1.17 1.28 1.23 MVV-1.4 97.1 20.4 3.63 1.12 0.88 1.00 MVV-1.5 108.0 52.6 4.18 0.97 1.37 1.17 MVV-1.6 119.7 21.1 4.26 0.80 0.73 0.76 MVV-1.7 178.8 59.3 4.01 0.89 0.77 0.83

Average measured densities range over a wide range: the lowest 10 mean densities measured for MVV-1.1 samples are 48.9 kg / m ³ and the highest for MVV-1.7 samples is 178.8 kg / m ³ . Because MVV-1.1 type specimens were low density (33.3-67.4 kg / m ³ ) and usually do not have undetectable and undeclared compressive stress and our measurements are very small (1.1-3.5 kPa), these samples will not be considered in these calculations. The estimated PS coefficients of variation for 15 (δ) for each sample type differ between 5.7 and 10.2%. The mean values of M range from 3.63% to 4.26%, and the calculated s = 0.05–0.88%.

From Table 3 we can see that the size of σ _β is most dependent on the density and macrostructure indices. Since the angles of the dominant fibers differ in some samples in the L and C sections, the calculated macrostructure indices are 20 different in each section. In order to determine the functional dependencies between physical and mechanical properties and macrostructure indices, statistical multiple regression calculations were performed and empirical equation was made: a _e = -107.77 + 0.51 p _s + 5.93 M + 54.81 S _c + 7, 02 S _L (2)

The statistical indicators of the empirical equation are presented in Table 4. It can be seen from the data in this table that the coefficient of determination R ² - 0.9161 is much higher than 0.7, so we can say that the chosen mathematical model is correct.

table. Functional dependencies and significance of σ ₀ and exploratory variables (p, M, Sc and SJ)

R R ² s _e The meanings of the student criterion Ps M Sc s _£ 0.9571 0.9161 5.83 13.94 3.35 15.19 2.27

From Table 4 we can see that all investigated parameters: density, organic matter content, L and C section macrostructure indices are significant, because all of them calculated (2,27-15,19) Student's values are higher than critical table size (2,021). However, the macrostructure indices have a different influence on σθ: S _c than S /., Because the size of the Student's criterion of Sc is more than six times larger than that of Sl. This was determined by the fiber arrangement during the manufacturing process, as the edges of the specimens were marked according to the direction of movement of the fiber web by conveyor: section L - parallel to the direction of movement, section C - perpendicular to the direction of movement. However, under realistic conditions and / or having a small fragment of the material to be tested, it can be quite difficult to determine the direction of movement of the conveyor and correctly mark the edges of the sections. Incorrectly marked slices will distort the results of the calculation.

Therefore, in order to avoid inaccuracies in the measurements, we will perform additional statistical calculations (Table 5), which will evaluate the influence of the average size, density and organic matter of both macrostructure indices on the compressive limit stress.

table. Functional dependencies and significance of the influence of S _L -c, Ps and M on σθ by multivariate regression

R R ² That's it Stjuden that criterion rei <shames Ps M Sl-c 0.9492 0.9006 7.22 12.43 3.79 14.03

Comparing the data in Tables 4 and 5, we can see that the magnitudes of multiple correlations, coefficients of determination and mean standard deviation are small, so we can construct an equation that uses a single indicator (their average size) instead of the macrostructure indices. This eliminates the likelihood of a false result if the edges of the specimens are incorrectly marked.

From Table 5, we can see that p _s and Sl-c are most influenced by σ _θ , since the high values of the Student criterion (12.43; 14.03) are calculated. Thus, the dependence of the most significant indicators can be plotted using a surface diagram (Figure 6).

The correlated empirical equation calculated for all relationships in Table 5 and Figure 6 is:

o _e = -140.66 + 0.55 p _s + 8.06 M + 82.35 S _{L. C} (3)

The values of the multiple correlations R, the coefficients of determination R ² , the mean standard deviation s _e, and the Student criterion values for this equation are given in Table 5. The coefficient of determination of the empirical equation is greater than

0.7 (Table 5), therefore, the mathematical linear multiple regression model was chosen correctly.

From the data in Figure 6, equation 3 and table 5, it can be concluded that the compressive fracture stress is most influenced by the macrostructure index St-c (maximum Student's criterion value 14.03) as it describes the directionality of the fiber structure.

The increasing numerical value of this indicator indicates an increase in the amount of vertically located fibers (with the direction coinciding with the direction of application of the external pressure force). In this way, the predominantly vertical fibers in the structure are more resistant to compression! load and low deformation. If S / .. c <0.75, the structure is dominated by horizontally located fibers and their groups, which are easier to deform due to the action of the load, since only a small proportion of fibers distributed vertically or chaotically in the volume of the structure resist compression. A significant effect of density is also observed on the compressive stress (since the Student criterion is 12.43), so that as the density increases, the strength increases. In this case, density has a direct relationship with the macrostructure index, as it measures the fiber content per unit volume. Therefore, as the quantity of fibers (i.e., density) increases and their explicit directionality with respect to the applied load increases, the mechanical properties increase. Significance of organic matter content (because Student's criterion 3.79) indicates the importance of the coupling phase in the fibrous structure as well. Higher amount of binder results in more bonded contacts in the fiber contact areas, which distributes the bonded fiber groups more evenly under the load and operates over the entire area.

The calculated empirical equation (3) gives an estimate of the compressive stress of a mineral wool product (with an error of ± 7.2 kPa) with 90% accuracy, by determining the average size of macrostructure, product density (70-200 kg / m ³ ) and organic matter content . Thus, even in the absence of dedicated compression equipment, the approximate value of σ _θ in each laboratory can be determined quickly enough. This would be the first non-destructive method for determining the compressive strength of mineral wool. Previous attempts to associate the values obtained in destructive (compressive) tests with the non-destructive method (using the method of damping vibrations in the structure of the specimen) have been described in (Jarleva at ei. 1984), but they have not gained practical application and have not been widely used.

As already mentioned, the structure orientation of mineral wool fibers is expressed by other indicators. In WO 2003/054270, the authors used the Tau (T) and Kappa (K) indices to express the orientation of the fiber and the degree of distribution of the dominant fibers.

In order to compare the macrostructure indices (Sc, S /.) With the T and K indices described above and used by the authors, we made comparative calculations. Comparison of the multiple correlation and determination coefficients and mean standard deviations of the macrostructure indices (S / .. c, T _Y : T _X , K _Y : K _X ) calculated by different methods showed that all macrostructure indices correlate with o _e . However, the strength of the functional dependence depends on the macrostructure indicators used. The strongest interconnection is observed between a _e and S _{L. C} (because

R = 0.9492) and the empirical equation (3) can be used to calculate the compressive fracture stress values with an error of ± 7.2 kPa. Statistical calculations show that T _Y : T _X and K _Y : K _X have a weaker correlation with a _e (R = 0.7065), and the values of compressive stress stress calculated using these indicators will be within +14.3 kPa. From the calculated values of the coefficient of determination (Table 5), we find that the compressive stress limit of the same specimen is calculated with 90% accuracy using S _L -c and with 49.9% using T _Y : T _X and K _Y : K _X. Thus, we can argue that the macrostructure indicator (S ^ -c) describes more accurately than other indicators the functional relationship between macrostructure and physico-mechanical properties.

The dominance of fibers in one direction changes the characteristics of different materials (EepecTOBa, 2006), because the spatial properties of the fibrous structure are different in different directions. In mineral wool products manufactured by conveyor (Strazdas and Eidukevicius, 1985), most fibers are arranged horizontally, so that the direction of loading depends on the compressive strength of the specimens. Knowing the directional orientation of the dominant fibers in the structure and linking it to mechanical-deformation characteristics, it is possible to adjust the orientation of the fibers. This allows for the optimization of production processes or the production of products with the desired strength (deformation) properties, using a clearly expressed direction.

In order to determine the anisotropy and deformability of the compressive strength structure of stone wool products, we determined the macrostructure indices and measured the compressive strength depending on the direction of loading in relation to the beam web motion. Due to the differences in fiber distribution in the structure, the values of the compressive limit stress were also measured during these tests in addition to the compressive stress at 10% deflection and compressive strength. In order to objectively compare the compressive strength properties of all specimen types, it is necessary to compare σ _{β by} measuring the measured value at the end of the elastic zone (since exceeding this value causes decomposition). We also performed comparative calculations, which compared the compressive stress stress values measured for the specimens and calculated using the empirical equation (3). For MVV-1.2 samples, the measured mean σ ₀ is 11.6 kPa. Therefore, by adding the values of the relevant parameters into equation (3): density - 95.2 kg / m ³ , content of organic matter - 3.70% and macrostructure index - 0.84, we obtain the calculated compressive stress limit value of 10.7 kPa. Doing the same for MVV-1.3 and MVV-1.4 specimens gave the following values: a _e = 44.5 kPa for MVV-1.3 and σ _θ = 24.4 kPa for MVV-1.4. Comparison of the calculated values with the real ones shows that the difference is not large and does not exceed the average standard deviation (Table 5). Thus, it can be stated that the calculated empirical equation (3) is suitable for predicting the compressive strength properties of mineral wool products.

The results obtained from the studies carried out show that the degree of anisotropy can be objectively and quantitatively quantified using a macro-structural index, on which the mechanical properties of stone wool products depend.

Industrial applicability

1. The structure of mineral wool fibers in the structure can be defined by the macro structure characteristics Sl, Sc or Sl-c. their S _{L. C} 0.75. Products predominantly perpendicular to the x axis are referred to as "supposedly vertical" and have a Sl-c z of 1.10. Conversely, products with randomly spaced fibers or fibers with an indeterminate orientation and / or orientation are referred to as "supposedly chaotic orientations" when S _{L. C,} 0.76-1.09.

2. There is a very strong functional relationship between the compressive stress and macrostructure characteristics, density and organic matter content of stone wool products due to the multiple correlation coefficient value of 0.9492. It was found that the proposed macrostructure index S ^ c and its method of determination and calculation are almost twice as accurate as the other method.

3. An empirical equation is presented that can be used to calculate the size of σ _β with an accuracy of ± 7.2 kPa, knowing the average size of macrostructures, product density (within 70-200 kg / m ³ ) and organic matter content. This enables the non-destructive method to predict the compressive strength properties of mineral wool products.

4. Macrostructure indicators may be used to express the degree of anisotropy of mineral wool structures. It has been shown numerically that the direction of load application and the distribution of fibers in the product structure significantly alter the compressive strength properties of stone wool slabs.

Claims (5)

Definition of the Invention 5 1. A method for determining the physical and mechanical properties of mineral wool products, comprising scanning the whole or part of the cross-section through the thickness of the specimen, scanning software by determining the direction of motion of the fiber web, scanning the cross section and perpendicular to the direction of motion of the web. analyzes the direction of motion of the web (C) 10 software for image, object, and background analysis calculates single angles values and their mean size, expresses test results by a macrostructure index, S, which expresses the ratio of angles of fiber orientation dominated by a flat section of the specimen measured relative to the principal axis for sections L and C, Sl and S _o , _respectively , for physical and mechanical mineral wool products The functional relationships between Sl and S _{c for the} 15 properties and macrostructure are determined by the equation: σ _θ = -107.77 + 0.51 p _s + 5.93 · M + 54.81 -S _c + 7.02 S _L where σ _β is the compressive stress, p _s is the density, 20 M- organic matter content. 2. A method for determining the physical and mechanical properties of mineral wool products, comprising scanning the whole or part of the cross-section through the thickness of the specimen, processing the scanned image with software that: 25 scans cross-section (L) and perpendicular cross-section (C), analyzes images with software for image, object and background analysis, calculates values and average size of single angles, expresses test results by macrostructure index S, which defines the fiber orientation of planar sections the ratio of angles measured between the reference axis and perpendicular to it is calculated by S separately For the slices L and C, Sl and S _c , respectively, calculate the average of the macrostructure indices Sl and S _c , and determine the functional relationship between the physical and mechanical properties of the mineral wool products and the macrostructure index Sl-c by: Oe ⁼ -140.66 + 0.55 p $ ⁺ 8.06 · M + 82.35 · Sl-c where a _e is the compressive stress, Ps- density, M- amount of organic matter. Method for determining the physical and mechanical properties of mineral wool products according to claim 1 or 2, characterized in that the macrostructure indicator Sl-c 0.75 defines the apparent horizontal distribution of the fibers dominant in the product structure relative to the main axis. 4. Method for determining the physical and mechanical properties of mineral wool products according to claim 1 or 2, characterized in that the macrostructure indicator S _L -c = 0.76 - 1.09 defines the allegedly chaotic distribution of fibers dominating the product structure relative to the main axis. Method for determining the physical and mechanical properties of mineral wool products according to claim 1 or 2, characterized in that the macrostructure indicator Sl-c £ 1.10 defines the apparent vertical distribution of the fibers of interest in the major axis of the product structure.