METHOD FOR DETERMINING THE DEFORMATION ANISOTROPY OF A TEST SPECIMEN

Abstract

The invention relates to a method (101) for determining the deformation anisotropy of a cylindrical sample, comprising positioning (103) the cylindrical sample on a goniometer, rotating (105) a movable part of the goniometer by a first determined angle, in steps of a second determined angle, acquiring (107), for each step, by a digital camera, an image (109) of the cylindrical sample, and calculating (111), from the image (109), a diameter of the cylindrical sample, as well as generating (113), from the calculated diameter, a curve (115) of angular evolution of the diameter and calculating (117), from the curve (115), the value of the Lankford coefficient. Figure for the abstract: Fig. 1

Description

Title of the invention: METHOD FOR DETERMINING THE DEFORMATION ANISOTROPY OF A TEST SPECIMEN technical field

[0001] The invention relates to the field of characterizing the mechanical properties of materials used for manufacturing aircraft turbomachine parts. It relates in particular to a method for determining the deformation anisotropy of a cylindrical sample, for example a metallic specimen such as a titanium alloy specimen. Previous technique

[0002] The certification of certain parts of an aircraft turbomachine requires precise knowledge of the mechanical behavior of the material used in their manufacture. In this context, it is particularly necessary to be able to determine the deformation anisotropy of such a material, such as a titanium alloy, which is a material from which certain discs, such as compressor discs or turbine discs, are made.

[0003] A key indicator of the deformation anisotropy of a material is the Lankford coefficient. This coefficient corresponds to the ratio of the transverse plastic deformations (i.e., the deformations along the width and thickness of a flat specimen or the radial deformations in two orthogonal directions for a cylindrical specimen) of a specimen that is tested in tension along its longitudinal extension direction. Determining this coefficient for a specimen therefore makes it possible to characterize its deformation anisotropy.

[0004] Current techniques for determining strain anisotropy (via the determination of the Lankford coefficient) rely on the use of one or two cameras to monitor the evolution of transverse strain during a tensile test. These techniques are limited when the specimen is tested in a furnace, as the furnace prevents the camera(s) from having a clear view of the specimen. Furthermore, the cameras struggle to operate in a high-temperature environment such as inside a furnace. It then becomes impossible to determine radial strains in situ. Summary of the invention

[0005] The present invention proposes a solution to these drawbacks.

[0006] Thus, an objective of the invention is to enable the characterization of the deformation anisotropy of a specimen subjected to a tensile test, via a simple device, including when the tensile test takes place at high temperature (for example, a test in traction requiring the use of an oven to reach a temperature above 25°C).

[0007] To this end, the invention, according to a first aspect, relates to a method for determining the deformation anisotropy of a cylindrical sample comprising the following steps:

[0008] - positioning of the cylindrical sample on a goniometer, so that an axis the longitudinal extent of said cylindrical sample coincides with an axis of rotation of a moving part of said goniometer;

[0009] - rotation of the moving part of the goniometer by a first determined angle, by steps of a second determined angle;

[0010] - acquisition, for each step, by a digital camera, of an image of the cylindrical sample;

[0011] - calculation, for each step, from the image, of a diameter of the sample cylindrical;

[0012] - generation, from the calculated diameter, of an angular evolution curve of the diameter as a function of the angular position of the cylindrical sample; and,

[0013] - calculation, from the curve, of the value of the Lankford coefficient for the cylindrical sample.

[0014] The method according to the invention may comprise one or more of the following features, taken individually or in combination with each other:

[0015] - the process further comprises, prior to the other steps, a test step in tension of a cylindrical test sample, until the said cylindrical test sample breaks, and the cylindrical sample, whose deformation anisotropy is determined, is a broken part of said cylindrical test sample.

[0016] - the calculation of the diameter of the cylindrical sample from the image includes the next sub-steps:

[0017] - determination, in the image, of a number of pixels corresponding to the dimension of the cylindrical sample along a direction orthogonal to its longitudinal axis; and,

[0018] - conversion, via a determined conversion scale, of the number of pixels into one distance.

[0019] - the determination and conversion substeps are repeated for several lines of the image in such a way as to obtain a plurality of intermediate diameters,

[0020] and the calculation of the diameter of the cylindrical sample further includes the following sub-step:

[0021] - calculation of the average value of the intermediate diameters so that the diameter is taken to be equal to said average value.

[0022] - the conversion scale is obtained by acquiring a plurality of images of less a cylindrical calibration sample, whose dimensions are known.

[0023] - the calculation of the value of the Lankford coefficient includes the following sub-steps following:

[0024] - adjustment by a semi-elliptic curve, of the curve;

[0025] - determination, from the semi-elliptic curve, of a first diameter according a first radial deformation direction and a second diameter along a second radial deformation direction, orthogonal to the first radial deformation direction;

[0026] - calculation, starting from the first diameter of a first radial deformation and, from of the second diameter of a second radial deformation; and,

[0027] - determination of the Lankford coefficient from the ratio of the first radial deformation on the second radial deformation.

[0028] - the first angle is between 170° and 190° and the second angle is constant, of The preference is between 2° and 10°. Angular positions therefore vary, for example, from 0 to 180° for a step of 5°.

[0029] - the first angle is equal to 180° and the second angle is variable.

[0030] - during the acquisition by the digital camera of at least one image of the cylindrical sample, said cylindrical sample is illuminated by a light source, placed opposite the digital camera with respect to the cylindrical sample.

[0031] - the cylindrical sample is made of a metal alloy suitable for the manufacturing of a mechanical part for an aircraft turbomachine, preferably made of a titanium alloy.

[0032] The invention according to a second aspect also relates to a system comprising a goniometer, a digital camera and a processing unit, said system being configured to implement the steps of the process according to the first aspect of the invention. Brief description of the drawings

[0033] The invention will be better understood with the aid of the following description, given solely by way of example and made with reference to the accompanying drawings in which:

[0034] [Fig.1] is a flowchart describing a method for determining the deformation anisotropy of a cylindrical sample according to an embodiment of the invention;

[0035] [Fig.2] is a schematic representation of a test specimen subjected to a tensile test as carried out in one embodiment of the invention;

[0036] [Fig.3] is an image of a system used for implementing the method of determining the deformation anisotropy of a test specimen according to an example of the invention;

[0037] [Fig.4] is an image of a test tube as acquired by a digital camera during the implementation of the process according to an example of the invention;

[0038] [Fig.5] is a curve showing the evolution of the diameter of a test specimen as a function of the angular position of the test specimen as obtained during the implementation of the process according to an example of the invention;

[0039] [Fig. 6] is a curve showing the evolution of the ratio of radial deformations of a test specimen as obtained during the implementation of the process according to an example of the invention; and,

[0040] [Fig.7] is a schematic representation of a disk in which a test specimen is taken and subjected to a tensile test as carried out in one embodiment of the invention. Description of the implementation methods

[0041] With reference to [Fig. 1], [Fig. 2], and [Fig. 3], we will now describe an embodiment of the method 101 for determining the strain anisotropy of a cylindrical sample 301 according to the invention. In the non-limiting example described below, the cylindrical sample 301 is a test specimen, and therefore only a test specimen is referred to. However, in general, the invention applies to a cylindrical sample (which can be subjected to a tensile test) whose strain anisotropy is to be determined.

[0042] In the non-limiting example described in [Fig.1], the process 101 includes a first step, carried out prior to the other steps of the process 101 (described below), of tensile testing 103 of a test specimen 201, until its breakage, so that the specimen 301 (visible in [Fig.3]) whose deformation anisotropy is determined in the continuation of the process is a broken part of the original test specimen 201.

[0043] Figure 2 shows an example of a cylindrical test specimen 201. This test specimen 201 is placed in a device (such as between the jaws of a tensile testing machine) in which it is subjected to a tensile force along its longitudinal axis R (which is also its axis of rotational symmetry). The tensile force to which it is subjected during this test is sufficient to cause it to break. Thus, the determination of the deformation anisotropy using the method according to this embodiment of the invention can be described as post-mortem in the sense that it is carried out after the test specimen has broken at a point on its surface.

[0044] Fig. 2 also shows axes A and T which extend respectively in two radial directions, orthogonal to each other, of the test specimen 201. These axes are the principal axes of anisotropy of the test specimen 201 and correspond to positions on the test specimen 201 where the radial contraction is minimal for one and where the radial contraction is maximal for the other.

[0045] As illustrated more specifically in [Fig. 7], these axes are known beforehand since the specimens are taken from a disk 701 such that their axis A is oriented along the axis (of revolution) of the disk, while their longitudinal axis R is oriented along the radius of the disk and the axis T is oriented along a tangent to the radius of the disk. An arrow marking on the head of the specimen thus allows these directions to be identified.

[0046] In a particular embodiment, the test specimen 201 (and therefore the test specimen 301) is made of a material suitable for manufacturing a mechanical part of an aircraft turbomachine. For example, the test specimen 201 can be made of titanium alloy such as is used for manufacturing compressor disks or turbine disks of an aircraft turbomachine.

[0047] More generally, the invention applies to a cylindrical test specimen made of all types of metallic materials, such as steel, aluminum, copper or magnesium, or of all types of plastic materials, such as thermoplastic resins or liquid crystals.

[0048] The method 101 then includes a step 105 of positioning the specimen 301 on a goniometer 303, so that the longitudinal extent axis R of the specimen 301 coincides with a rotation axis X of a movable part 303a of the goniometer 303.

[0049] As shown in [Fig. 3], illustrating a system 300 for implementing the method according to the invention, the goniometer 303 is a device comprising a fixed part 303b and a movable part 303a (in this case, a platform), as well as means 303c for varying the angle of rotation, around the X-axis, of the movable part 303a relative to the fixed part 303b. By way of example, for this type of device, the accuracy of adjusting the angular position of the movable part 303a can be on the order of 0.5°. Depending on the type of goniometer, the change in the angular position of the movable part can be carried out manually by an operator or be controlled electronically.

[0050] Positioning the test specimen 301 on the goniometer 303 involves, in this case, fixing the test specimen 301 to the movable part 303a of the goniometer 303. In other words, the test specimen 301 is made fixed to the movable part 303a so that the rotation of the movable part 303a by a certain angle causes the test specimen 301 to rotate by the same angle.

[0051] The method 101 also includes a step 107 of rotating the movable part 303a of the goniometer 303 by a first determined angle Th, by steps of a second determined angle 0 and a step of acquiring 109, for each step, by a digital camera 305, an image 109 of the specimen 301.

[0052] In other words, the angular position (relative to the axis of rotation X) of the specimen varies between an initial position (for example, considered to have a zero angle value) and a final position having a first angle Th known relative to the initial position. Furthermore, the variation of the angular position is achieved in steps of a second angle 0 less than the first angle Th, so that the specimen adopts a plurality of intermediate angular positions between the initial and final positions, and that, for all angular positions, including the initial and final positions, the digital camera 305 acquires an image of the specimen 301.

[0053] In a non-limiting example of implementing the method, the first angle Th is between 170° and 190°, for example, equal to 180°. The second angle 0 can be constant and preferably between 2° and 10°, for example, equal to 5°. In another non-limiting example of implementing the method, the first angle Th is equal to 180° and the second angle 0 is variable (i.e., the pitch varies). In all cases, a person skilled in the art will be able to adapt the angular values ​​and any variation in the pitch to achieve the desired precision and / or simplicity of implementation or calculation.

[0054] For example, a smaller step size can allow for a more precise adjustment of a curve obtained as a function of angular position (as described in more detail later). Furthermore, a variable step size can increase the number of points obtained in an area where greater precision is required.

[0055] The method 101 further includes a calculation step 111, for each step, from the image 109, of a diameter D of the specimen 301. This step can be carried out, for example, via image processing software.

[0056] In the embodiment shown, step 111 of calculating the diameter D of the specimen from the image 109 includes a substep 111_1 of determining, in the image 109, a number px of pixels corresponding to the dimension of the specimen 301 along a direction orthogonal to its longitudinal extent axis R and a substep 111_3 of converting, via a determined conversion scale, the number px of pixels into a distance.

[0057] Indeed, since it is an image acquired by the digital camera 305, the image 109 is made up of pixels whose number and dimensions depend on the resolution of the digital camera 305. Furthermore, in the example shown in [Fig. 3], the test specimen 301 is positioned on the movable part 303a of the goniometer 303, which is a platform horizontal so that the longitudinal extent axis R of specimen 301 is vertical. With the digital camera 305 positioned in a horizontal plane passing through specimen 301, the radial dimension of specimen 301 corresponds to a portion of a (horizontal) line in image 109.

[0058] Furthermore, the conversion scale can be obtained by calibrating the imaging system, in this case the digital camera 305. Thus, in a particular embodiment of the method, the conversion scale can be obtained by acquiring a plurality of images of at least one calibration specimen, the dimensions of which are known. In other words, the conversion scale corresponds to a multiplicative factor associating a pixel in the image 109 with a given distance in the plane of the specimen 301.

[0059] Furthermore, also in a particular embodiment, during step 107 of acquisition by the digital camera 305 of the image 109 of the specimen 301, the specimen 301 can be illuminated by a light source 307 (as represented in [Fig.3]), placed opposite the digital camera 305 with respect to the specimen 301.

[0060] The light source used can be, for example, an LED light projector with adjustable intensity to obtain the desired sharpness. In all cases, the source illuminates the test specimen both in front of and behind the camera so as to produce a sharp outline (according to the user's judgment).

[0061] Advantageously, the outline of the test specimen 301 in the image 109 is thus sharper so that the accuracy of the determination of the diameter D is improved.

[0062] Fig. 4 shows in more detail an example of an image 109 acquired by the digital camera 305 in which each line 401 from the window 403 in the image 109 allows a diameter value of the specimen 301 to be determined.

[0063] Thus, to improve the reliability and accuracy of the determination of the diameter D of the specimen 301, in the implementation of the process shown in [Fig.1], the substeps 111_1 of determination and 111_3 of conversion are repeated for several lines of the image 109 (i.e. those of the window 403) so as to obtain a plurality of intermediate diameters Di and the calculation 111 of the diameter D of the specimen 301 further includes the substep 111_5 of calculation of the average value of the intermediate diameters Di so that the diameter D is taken equal to this average value.

[0064] The method 101 also includes a step 113 of generating, from the calculated diameter D, a curve 115 of the angular evolution of the diameter D. This step can be carried out, for example, by a computer program that sorts into a table (or a vector) all the values ​​of diameter D obtained for each step. In effect, the angular evolution of the diameter D means the evolution of the determined value of the diameter D as a function of the angle (i.e., the angular position) of the specimen 301.

[0065] Figure 5 shows in more detail an example of a curve 115 in which the points corresponding to each diameter value determined for each step (in this case, the value of the first angle Th is equal to 180° and the value of the second angle 0 is constant and equal to 5°) are connected by straight lines. Furthermore, Figure 5 also shows a continuous smooth curve, which is described in more detail later.

[0066] The process 101 finally includes a step 117 of calculation, from the curve 115, of the value of the Lankford coefficient Lk for the specimen 301.

[0067] In particular, in the example shown in [Fig.1], step 117 of calculating the value of the Lankford coefficient Lk includes a first substep 117_1 of fitting by a semi-elliptic curve, of the curve 115. The fitting (or "fit" in English) by a semi-elliptic curve (as seen in smooth continuous line in [Fig.5]) makes it possible to take into account the anisotropic character of the material characterized.

[0068] Step 117 also includes a substep 117_3 for determining, from the semi-elliptic curve (with which curve 115 has been fitted), a first diameter DA along a first radial deformation direction A (i.e., along the axis A mentioned above) and a second diameter DT along a second radial deformation direction T (i.e., along the axis T mentioned above), orthogonal to the first radial deformation direction A. In practice, the first diameter DA and the second diameter DT correspond respectively to the minimum and maximum values ​​of the semi-elliptic curve fitted to curve 115.

[0069] Step 117 further includes a substep 117_5 of calculation, from the first diameter DA of a first radial deformation eA and, from the second diameter DT, of a second radial deformation eT.

[0070] In particular, the first radial deformation eA is obtained by the formula:

[0071] [Math.l] Da-Dq £ad„

[0072] where Do is the known theoretical diameter of the test specimen 201.

[0073] While the second radial deformation eT is obtained by the formula:

[0074] [Math.2] £r~ D.

[0075] Step 117 finally includes a substep 117_7 of determining the Lankford coefficient Lk from the ratio of the first radial strain eA to the second radial strain eT.

[0076] In practice, the calculation of the Lankford coefficient Lk involves the use of so-called true strains, obtained from the radial strains eA and eT by applying a natural logarithm to them. More precisely, the true strain along the A direction is equal to ln(l+ eA), which corresponds to -ln(D0 / DA), while the true strain along the T direction is equal to ln(l+ eT), which corresponds to -ln(D0 / Dt).

[0077] The value of the Lankford coefficient Lk is therefore obtained by calculating the formula:

[0078] [Math.3] 111(1+ sr)

[0079] By way of example, [Fig.6] shows, in its lower part, points tracing the evolution of the true strain along the direction T as a function of the true strain along the direction A. The slope of the line that fits these points corresponds to the Lankford coefficient while the line passing through the center (of the figure) shows a theoretical case without anisotropy.

[0080] By way of further example, steps 111, 113 and 117 of the process can be implemented by a processing unit, such as the processing unit 309 shown schematically in [Fig.3], which is also capable of controlling the rotation of the moving part of the goniometer 303 and of processing images acquired by the digital camera 305.

[0081] Finally, thanks to the invention, it is possible to characterize the deformation anisotropy of the material of a test specimen via the use of a single camera including when the test specimen has been subjected to a tensile test involving the use of a furnace.

[0082] Furthermore, the use of a broken test specimen (i.e., a post-mortem determination) does not require intervention on the system used for the tensile test, since the determination of anisotropy is done completely independently of the tensile test.

[0083] Finally, the system used is simple since it does not require the use of a camera capable of following the deformation of the specimen in real time and the diversity of materials that can be characterized is large.

Claims

Demands

1. A method (101) for determining the deformation anisotropy of a cylindrical sample (301) comprising the following steps: - positioning (103) of the cylindrical sample (301) on a goniometer (303), such that an axis (R) of longitudinal extent of said cylindrical sample (301) coincides with an axis (X) of rotation of a movable part (303a) of said goniometer (303); - rotation (105) of the movable part (303a) of the goniometer (303) by a first determined angle (Th), in steps of a second determined angle (0); - acquisition (107), for each step, by a digital camera (305), of an image (109) of the cylindrical sample (301); - calculation (111), for each step, from the image (109), of a diameter (D) of the cylindrical sample (301); - generation (113), from the calculated diameter (D), of a curve (115) of angular evolution of the diameter (D) as a function of an angular position of the cylindrical sample (301);and, - calculation (117), from curve (115), of the value of the Lankford coefficient (Lk) for the cylindrical sample (301) representative of the deformation anisotropy of the cylindrical sample.;

2. Method (101) according to claim 1, further comprising, prior to the other steps, the following step: - tensile test (103) of a cylindrical test sample (201), until the breakage of said cylindrical test sample (201), and wherein the cylindrical sample (301), whose deformation anisotropy is determined, is a broken part of said cylindrical test sample (201).

3. A method (101) according to claim 1 or claim 2, wherein the calculation (111) of the diameter (D) of the cylindrical sample (301) from the image (109) comprises the following substeps: - determination (111_1), in the image (109), of a number (px) of pixels corresponding to the dimension of the cylindrical sample (301) along a direction orthogonal to its longitudinal extent axis (R); and, - conversion (111_3), via a determined conversion scale, of the number (px) of pixels into a distance.

4. A method (101) according to claim 3, wherein the determination (111_1) and conversion (111_3) substeps are repeated for several lines of the image (119) so as to obtain a plurality of intermediate diameters (Di), and the calculation (111) of the diameter (D) of the cylindrical sample (301) further comprises the following substep: - calculation (111_5) of the average value of the intermediate diameters (Di) so that the diameter (D) is taken to be equal to said average value.

5. Method (101) according to claim 3 or claim 4, wherein the conversion scale is obtained by acquiring a plurality of images of at least one cylindrical calibration sample, the dimensions of which are known.

6. A method (101) according to any one of the preceding claims, wherein the calculation (117) of the value of the Lankford coefficient (Lk) comprises the following substeps: - fitting (117_1) by a semi-elliptic curve, of the curve (115); - determination (117_3), from the semi-elliptic curve, of a first diameter (DA) along a first direction (A) of radial deformation and of a second diameter (DT) along a second direction (T) of radial deformation, orthogonal to the first direction (A) of radial deformation; - calculation (117_5), from the first diameter (DA) of a first radial deformation (eA) and, from the second diameter (DT) of a second radial deformation (eT); and, - determination (117_7) of the Lankford coefficient (Lk) from the ratio of the first radial strain (eA) to the second radial strain (eT).

7. Method (101) according to any one of claims 1 to 6, wherein the first angle (Th) is between 170° and 190° and the second angle (0) is constant, preferably between 2° and 10°.

8. A method (101) according to any one of the preceding claims, wherein, during the acquisition (107) by the digital camera (305) of at least one image (109) of the cylindrical sample (301), said cylindrical sample (301) is illuminated by a light source (307), placed opposite the digital camera (305) with respect to the cylindrical sample (301).

9. A method (101) according to any one of the preceding claims, wherein the cylindrical sample is made of a metal alloy suitable for manufacturing a mechanical part of an aircraft turbomachine, preferably a titanium alloy.

10. System (300) comprising a goniometer (303), a digital camera (305) and a processing unit (309), said system (300) being configured to carry out the steps of the method (101) according to any one of the preceding claims.

Images