Characterization specimen and associated mechanical testing method

Abstract

The invention relates to a characterization test specimen. According to the invention, the specimen is externally shaped into an oblong form and is internally hollowed out by an oblong orifice (4). The invention also relates to a mechanical testing method using such a test specimen. FIGURE IN ABRIDGED DIAGRAM: Fig. 1

Description

Title of the invention: Characterization test piece and associated mechanical testing method

[0001] The present invention relates to a characterization test piece.

[0002] The invention also relates to a method of mechanical testing on such a test piece.

[0003] BACKGROUND OF THE INVENTION

[0004] Composite materials are used in many applications and for example in the aeronautical field (blades, casings, connecting rods, etc.).

[0005] Such composite parts are generally subjected to significant loads and stresses. Consequently, they must be sized to ensure their proper functioning over time and in the intended application. Such sizing requires in-depth knowledge of the material used.

[0006] For this purpose, methods are known for characterizing the behavior of composite materials.

[0007] Usually such methods consist of carrying out multiple tests on multiple characterization specimens, the specimens being in a material whose behavior is to be validated. The results of the different tests are then compared and analyzed to deduce one or more mechanical characteristics of the material.

[0008] This type of method therefore turns out to be relatively tedious.

[0009] SUBJECT OF THE INVENTION

[0010] The invention aims in particular to at least partially overcome the aforementioned drawback. Summary of the invention

[0011] For this purpose, a characterization test piece is provided.

[0012] According to the invention, the test piece is shaped externally into an oblong shape and is hollowed out internally by an oblong-shaped orifice.

[0013] Surprisingly, the inventor was able to observe that the particular shape of the test piece made it possible to obtain characterization results more efficiently. In particular, with such a test piece shape, the inventors were able to observe that it was possible to characterize several mechanical characteristics with a single type of mechanical test.

[0014] Other characteristics and advantages of the invention will emerge from reading the following description of a particular and non-limiting embodiment of the invention. Brief description of the drawings

[0015] Reference will be made to the accompanying drawings, among which:

[0016] [Fig-1] [Fig.l] is a three-dimensional view of a characterization test piece according to a particular embodiment of the invention;

[0017] [Fig.2] [Fig.2] is a cross-sectional view of the test piece shown in [Fig.l];

[0018] [Fig.3] [Fig.3] is a diagram symbolizing the different stages of characterization of a material from the test piece represented in [Fig.l];

[0019] [Fig.4] [Fig.4] is a schematic view of the test piece shown in [Fig.l] during a mechanical test of said test piece. DETAILED DESCRIPTION OF THE INVENTION

[0020] With reference to [Fig.l] and [Fig.2], the characterization test piece 1 according to a particular embodiment of the invention is in a material that it is desired to characterize.

[0021] For example, the material is a composite material. For example, said composite material is isotropic or anisotropic. For example, said composite material is transverse isotropic or orthotropic isotropic.

[0022] The material is for example a composite incorporating continuous fibers. Preferably, the fibers are all oriented parallel to each other in the raw material, i.e. before the test piece 1 is formed.

[0023] The test piece 1 also extends axially along a first axis X.

[0024] The test piece 1 has a cross section 2 (i.e. a section extending in a plane with normal to the first X axis) which is oblong in shape.

[0025] More precisely, the cross-section 2 has an oblong external contour 3. The external contour 3 thus has an upper line 3a parallel to a lower line 3b, the two lines 3a, 3b being connected at each of their ends by a portion 3c in the shape of an arc of a circle and for example by a portion 3c in the shape of a half-arc of a circle. The external contour 3 is closed.

[0026] The cross-section 2 is also hollow. It thus has an opening 4 drawing an internal contour 5 of the cross-section 2. The opening 4 is itself oblong in shape. The internal contour 5 thus has an upper line 5a parallel to a lower line 5b, the two lines 5a, 5b being connected at each of their ends by a portion 5c in the shape of an arc of a circle and for example by a portion 5c in the shape of a half-arc of a circle. The internal contour 5 is closed.

[0027] The orifice 4 is preferably centered in the cross section 2. In this way, the curved portions 3c of the external contour 3 and the curved portions 5c of the internal contour 5 are concentric two by two.

[0028] The thickness e (i.e. the distance separating the external contour 3 from the internal contour 5) of the cross-section 2 is preferably the same over the entire circumference of the cross-section 2.

[0029] The cross-section 2 of the test piece 1 is preferably identical over the entire width of the test piece 1 (depth considered along the first axis X).

[0030] Therefore, the test piece 1 is hollow and its opening 4 extends over the entire width of the test piece 1 in order to open onto a first lateral face 6 of the test piece 1 and onto a second lateral face 7 of the test piece 1. The two faces 6, 7 are opposite each other. The two faces 6, 7 are here flat and parallel to each other. Each of said faces 6, 7 is therefore identical to the cross-section 2 previously described.

[0031] Furthermore, the two lateral faces 6, 7 are connected internally by a continuous internal face 8 (forming the internal contour 5 of the different cross-sections of the test piece 1). Furthermore, the two lateral faces 6, 7 are connected externally by a continuous external face 9 (forming the external contour 3 of the different cross-sections of the test piece 1).

[0032] The test piece 1 thus described therefore comprises an upper spar 10 extending longitudinally along a second axis Y orthogonal to the first axis X and a lower spar 11 extending longitudinally along the second axis Y, parallel to the first axis X. The test piece 1 further comprises a first curved elbow 12 (or bent part 12) connecting first longitudinal ends of the lower 11 and upper 10 spars and a second curved elbow 13 (or bent part 13) connecting second longitudinal ends of the lower 11 and upper 10 spars. The test piece 1 is preferably in one piece.

[0033] The fibers of the composite material thus follow the geometry of the test piece 1. In particular, at the level of the lower 11 and upper 10 side members, the fibers preferably extend parallel to the second axis Y. In particular, at the level of the bent parts 12 and 13 the fibers extend so as to follow the curvature of said bent parts.

[0034] Therefore, the test piece 1 is oblong in shape in the sense that it is longer (along the second axis Y) than it is tall (along a third axis Z orthogonal to the first axis X and to the second axis Y) and that it has two rounded longitudinal ends connected to each other at the top by the upper spar 10 and at the bottom by the lower spar 11.

[0035] The test piece 1 thus forms a closed ring (of oblong section).

[0036] The test piece 1 thus described is symmetrical along at least one axis of symmetry and preferably along at least two axes of symmetry. The test piece 1 thus described is for example symmetrical along the second axis Y. The test piece thus described is for example symmetrical along the third axis Z.

[0037] Thus when the test piece 1 rests on a flat surface: - its two side members extend horizontally, and / or - its second Y axis extends horizontally, and / or - its third Z axis extends vertically.

[0038] According to another aspect, the thickness of the test piece 1 (i.e. the distance separating the continuous internal face 8 from the continuous external face 9) is the same over the entire circumference of the test piece 1.

[0039] The test piece 1 is preferably a thick test piece 1. Indeed, its thickness 1 is not negligible compared to the other dimensions of the test piece 1 unlike “thin” test pieces of the prior art.

[0040] A thick specimen is, for example, a specimen that is not thin. A thin specimen is, for example, defined by the thin plate theory developed by Love-Kirchhoff.

[0041] The thickness 1 is for example identical to plus or minus 30% to the width (considered along the first axis X) of the test piece 1 and is for example identical to plus or minus 20% to the width of the test piece 1 and is for example identical to plus or minus 10% to the width of the test piece 1.

[0042] The thickness of the test piece 1 is preferably greater than 5 millimeters, and preferably greater than 8 millimeters and preferably greater than 10 millimeters.

[0043] With reference to figures 3 and 4, we will now describe a method for characterizing the material forming such a test piece 1.

[0044] For example, it is desired to characterize the material by at least one independent elasticity constant and preferably at least two independent elasticity constants and preferably at least three independent elasticity constants of such a material and preferably at least four independent elasticity constants of such a material and preferably at least five independent elasticity constants of such a material.

[0045] For example, we seek to characterize the material by at least five independent elasticity constants such as the following constants: El, E2, G12, v 12 and v23.

[0046] During an initial step, a finite element modeling of the test piece 1 is carried out.

[0047] The modeling is for example formed from a mesh of tetrahedral meshes. A diameter of a mesh (i.e. the largest dimension of a mesh) is preferably between 0.1 and 0.2 millimeters.

[0048] The modeling preferably has the same shape as test piece 1.

[0049] The modeling preferably has the same dimensions as test piece 1.

[0050] The modeling will allow the implementation of a digital simulation aimed at reproducing a mechanical test which will be carried out on test piece 1. Thus, the modeling must be as close as possible to test piece 1.

[0051] The modeling is also defined by an isotropic-transverse elastic behavior law in order to be able to behave like the test piece. The modeling therefore has five independent elastic constants: E1, E2, G12, v12 and v23. In this initial step, arbitrary values ​​are defined for these independent elastic constants and / or search domains for said constants (a search domain being an interval having a maximum limit and a minimum limit, the corresponding independent elastic constant must imperatively remain in this interval).

[0052] The modeling is also defined to symbolize that its fibers follow the contour of the modeling.

[0053] The modeling is thus conformed and / or behaves like test piece 1 as previously described.

[0054] During the initial step, one or more boundary conditions are preferably defined for the rest of the process. The modeling and the test piece 1 preferably share at least one boundary condition and preferably all the boundary conditions.

[0055] The simulation (respectively the modeling) is thus required to have the same characteristics as the mechanical test (respectively test piece 1).

[0056] For example, the boundary condition(s) are chosen from the list below: - one or more planes of symmetry of the modeling and of the test piece, - one or more planes of symmetry of the forces applied to the modeling and to the test tube, - one or more characteristics of at least one force applied to the modeling and to the test piece (value, direction, etc.), - one or more characteristics of a contact between the model and the test piece and at least one shaft through which a force is applied to said model and said test piece, - etc.

[0057] During a first step of the method, an iterative loop 300 is implemented.

[0058] This iterative loop 300 comprises a first phase consisting of carrying out 400 a mechanical test on at least one test piece such as that previously described, a second phase consisting of carrying out 500 a numerical simulation on the modeling previously described and a third phase of comparison 600 of the results obtained during the first phase and the second phase of this first step.

[0059] First phase

[0060] Preferably, the mechanical test is a bending test. Preferably, the mechanical test is a four-point bending test.

[0061] Preferably, the mechanical test is a four-point bending test along an axis parallel to the third axis Z (i.e. a vertical axis when the test piece 1 is arranged on a 1000 test bench).

[0062] In the present case, the test bench 1000 comprises a lower plate 1001 and an upper plate 1002.

[0063] The lower plate comprises a lower jaw having a plurality of grooves 1003 (only a part of which is referenced here). The different grooves extend parallel to each other and to a longitudinal axis L. The axis L is parallel to the first axis X when the test piece 1 is in place in the test bench 1000.

[0064] The lower plate 1001 further comprises a first shaft 1004 and a second shaft 1005. The two shafts 1004, 1005 are identical to each other. The two shafts 1004, 1005 are shaped as a solid cylinder of revolution. The cross-section of each of the shafts 1004, 1005 is a disc.

[0065] Each shaft 1004, 1005 is capable of being temporarily received in one of the grooves 1003.

[0066] In this way, each shaft 1004, 1005 extends longitudinally in one of the grooves 1003, the different shafts 1004, 1005 extending parallel to each other and to the axis L. The shafts 1004, 1005 can therefore be more or less distant from each other depending on the grooves in which they are received.

[0067] Furthermore, the upper plate 1002 comprises an upper jaw having a plurality of grooves 1006 (only a part of which is referenced here). The different grooves 1006 extend parallel to each other and to the longitudinal axis L.

[0068] The upper plate 1002 further comprises a first shaft 1007 and a second shaft 1008. The two shafts 1007, 1008 are identical to each other. The two shafts 1007, 1008 are shaped as a solid cylinder of revolution. The cross-section of each of the shafts 1007, 1008 is a disc.

[0069] Preferably, the shafts 1007, 1008 of the upper plate 1002 are identical to the shafts 1004, 1005 of the lower plate 1001.

[0070] Each shaft 1007, 1008 is capable of being temporarily received in one of the grooves of the jaw.

[0071] In this way, each shaft 1007, 1008 extends longitudinally in one of the grooves 1006 parallel to each other and to the axis L. The shafts 1007, 1008 can therefore be more or less distant from each other depending on the grooves 1006 in which they are received.

[0072] The shafts 1007, 1008 are positioned at the same time as the test piece 1 on the test bench.

[0073] Optionally, the two plates 1001, 1002 are identical to each other.

[0074] The upper plate 1002 is arranged above the lower plate 1001 of so that the lower and upper jaws are opposite each other. The rest of the 1000 test bench is similar to what currently exists and will not be detailed further here.

[0075] It is noted that the shafts 1004, 1005, 1007, 1008 of the lower 1001 and upper 1002 plates have a length (along the L axis) greater than the width (along the X axis) of the test piece.

[0076] In operation, an operator arranges the test piece 1 on the lower plate 1001. The two shafts 1004, 1005 of the lower plate thus form two support “points” on which the lower spar 11 rests in this case.

[0077] The operator then controls the test bench 1000 so that the upper plate 1002 presses on the test piece 1 by its two shafts 1007, 1008 and thus deforms the test piece 1 in flexion. The two shafts 1007, 1008 of the upper plate 1002 therefore form two “points” on the upper spar 10.

[0078] Thus the lower 1001 and upper 1002 plates frame the test piece 1 and allow its deformation in bending.

[0079] Although we speak of support points and pressure points, we understand that the contacts between the test piece 1 and the different shafts 1004, 1005, 1007, 1008 are in reality linear or surface contact and pressure zones.

[0080] Preferably, the shafts 1004, 1005 of the lower plate 1001 are arranged on either side of a vertical central axis of symmetry separating the test piece 1 in two and / or the shafts 1006, 1007 of the upper plate 1002 are arranged on either side of said vertical central axis of symmetry. It is thus understood that the shafts 1004, 1005 of the lower plate 1001 bear symmetrically on the lower spar 11 and / or that the shafts 1007, 1008 of the upper plate 1002 bear symmetrically on the upper spar 10.

[0081] Preferably, the shafts 1004, 1005 of the lower plate 1001 are closer to the center of the lower spar 11 than to its longitudinal ends and / or the shafts 1007, 1008 of the upper plate 1002 are closer to the center of the upper spar 10 than to its longitudinal ends. It is thus understood that the shafts 1004, 1005 of the lower plate 1001 press on the center of the lower spar 11 and / or that the shafts 1007, 1008 of the upper plate 1002 press on the center of the upper spar 10.

[0082] Preferably, the shafts 1004, 1005 of the lower plate 1001 are aligned in a direction parallel to the second Y axis (and / or parallel to the horizontal). Preferably, the shafts 1007, 1008 of the upper plate 1002 are aligned in a direction parallel to the second Y axis (and / or parallel to the horizontal).

[0083] Preferably, a first shaft 1004 of the lower plate 1001 and a first shaft 1007 of the upper plate 1002 are also aligned with each other in a direction parallel to the third axis Z (and / or parallel to the vertical). Preferably, a second shaft 1005 of the lower plate 1001 and a second shaft 1008 of the upper plate 1002 are also aligned with each other in a direction parallel to the third Z axis (and / or parallel to the vertical).

[0084] Thus, each support point is aligned with one of the pressure points.

[0085] At the end of the first phase, one or more characteristic parameters of the mechanical test carried out are obtained, called “measured parameters”. For example, between 6 and 20 parameters are measured during this first phase and for example between 10 and 15 parameters and for example 12 parameters. The estimation of these parameters makes it possible to deduce independent measured elastic stresses (El, E2, G12, vl2 and v23).

[0086] The measured parameters are for example unit deformations e (i.e. deformations per unit of length, the unit deformations having no unit [m / m]).

[0087] The measured parameters are for example chosen so as to present one or more of the characteristics below: - a unit longitudinal deformation, a unit transverse deformation, a unit shear deformation, etc. - unit deformation at the level of the continuous external face 9 of test piece 1, unit deformation at the level of the continuous internal face 8 of test piece 1, etc. - unit deformation at the level of the upper spar 10, the lower spar 11, the longitudinal ends, etc. - unit deformation oriented along an axis parallel to one of the X, Y or Z axes or unit deformation oriented along an axis inclined with respect to one of said X, Y and Z axes.

[0088] For example, at least one of the parameters is a unit deformation measured in one of the bent parts 12, 13 of the test piece 1. Preferably, said parameter is a unit deformation at the center of said bent part and / or oriented in a direction radial to said bent part.

[0089] For example, at least one of the parameters is a unit deformation measured in one of the side members of the test piece 1. Preferably, said parameter is a unit deformation at the center of the side member and / or oriented parallel to said side member and / or oriented transversely to said side member.

[0090] To measure these parameters during the mechanical test, one or more strain gauges are arranged on the test piece 1. The orientation and position of each of the gauges on the test piece 1 depends on the parameter that one wishes to measure. Thus, if the strain gauge makes it possible to measure a unit strain parameter in a zone Z of the test piece and according to a given orientation D, the strain gauge will correspondingly preferably be arranged in the zone Z of the test piece 1 and according to the targeted orientation D. Note that the same strain gauge can allow the measurement of two distinct parameters, for example if it is mono-lateral and not bilateral.

[0091] Second phase

[0092] The second phase is carried out simultaneously or offset with the first phase.

[0093] During this second phase, the same type of test as the mechanical test applied to specimen 1 is applied to the modeling. A four-point bending test is thus reproduced along an axis parallel to the third Z axis on the modeling. In particular, the support points and the pressure points are applied to the same areas of the modeling as for specimen 1. Preferably, it is assumed that the contact between each pressure point and the modeling is sliding (friction being neglected).

[0094] Similarly, the calculated parameters are calculated at the same location and with the same direction as the parameters measured during the first phase.

[0095] At the end of the simulation, calculated (or digital) parameters are thus obtained which are the same as the parameters measured during the first phase, the values ​​of the calculated parameters however being potentially different from the values ​​of the parameters measured during the first phase.

[0096] Third phase

[0097] During a third phase, the measured parameters and the calculated parameters are compared.

[0098] If the comparison between at least one of the measured parameters and the corresponding calculated parameter does not positively meet a convergence criterion (for example if the difference between one of the measured parameters and the corresponding calculated parameter is greater than a convergence threshold), then at least the second phase is repeated once again.

[0099] At least the values ​​of the measured parameters and the calculated parameters are provided as input to the new second phase 900.

[0100] At the end of the new second phase, a third new phase is implemented.

[0101] The iterative loop 300 continues in this way as long as the comparison between at least one of the measured parameters and the corresponding calculated parameter does not positively meet the convergence criterion.

[0102] During a fourth step, if the comparison between the measured parameters and the corresponding calculated parameters positively meets the convergence criterion, then the iterative loop 300 is terminated.

[0103] The iterative optimization loop 300 thus described can be implemented for example by a multi-objective genetic algorithm (or “Multi-Objective Genetic Algorithm”) or even an adaptive multi-objective algorithm (or “Adaptive Multiple-Objective Optimization”).

[0104] We therefore understand that the first stage of the process mixes two different approaches: - a first approach which is a method of carrying out mechanical tests - a second approach which is an inverse method and for example a finite element recalibration method.

[0105] More specifically, the method consists at least in part of comparing and varying the input conditions of the modeling in an iterative manner so that the output parameters of this modeling correspond to experimental data obtained with one or more mechanical tests.

[0106] It is also noted that the mechanical test(s) of the first phase and the numerical calculation of the second phase share the same imposed boundary conditions.

[0107] During the fourth step, the independent elastic constants El, E2, G12, vl2 and v23 are identified from the calculated parameters and the measured parameters. This identification is carried out by an optimization method (like one of those proposed previously) which makes it possible to determine the best possible values ​​for these independent elastic constants El, E2, G12, vl2 and v23 in view of the calculated parameters and the measured parameters which have made it possible to validate the convergence criterion.

[0108] During the first iteration of the loop 300, the convergence criterion has for example a predetermined value. Preferably, the convergence criterion is not fixed and will here evolve according to the values ​​of the parameters measured at each new iteration of the loop 300.

[0109] Preferably, the measured parameters are also used in the calculation of the convergence criterion. The value of the measured parameters is however fixed here between at least two successive iterations. In this way, the variation of the convergence criterion depends mainly on the calculated parameters.

[0110] Advantageously, by identifying these five elastic constants, it is also possible to calculate a sixth independent elastic constant from said five identified elastic constants. For example, the sixth constant is the independent elastic constant G23.

[0111] The method thus described allows an intrinsic mechanical characterization of a material. In particular, the method thus described allows an intrinsic mechanical characterization of a material of massive test specimens.

[0112] The method makes it possible in particular to characterize a material by five to six independent elasticity constants. Advantageously, several independent elasticity constants are determined by using a single type of mechanical test and corresponding simulation (here four-point bending).

[0113] The characterization is thus of good quality.

[0114] Of course, the invention is not limited to the embodiment described but encompasses any variant falling within the scope of the invention as defined by the claims.

[0115] Although here the strain gauges are arranged around the entire circumference of the test piece during the mechanical test, the strain gauges may only be arranged on a portion of this circumference. In particular, the strain gauges may only be arranged on half of the test piece, said half being symmetrical to the other half of the test piece along at least one axis of symmetry. In particular, the strain gauges may only be arranged on a quarter of the test piece, said quarter being symmetrical to the other three quarters of the test piece along two distinct axes of symmetry. Indeed, due to the symmetry of the test piece, the parameter(s) obtained during the mechanical test for a portion of the test piece can be transposed by symmetry to the other portion of the test piece.

[0116] Similarly, although here the calculations carried out via the modeling are carried out on the entire perimeter of the modeling, the calculations may be carried out for only a part of this perimeter. In particular, the calculations may only be carried out on half of the modeling, said half being the symmetrical of the other half of my modeling along at least one axis of symmetry. In particular, the calculations may only be carried out on a quarter of the modeling, said quarter being the symmetrical of the other three quarters of the modeling along two distinct axes of symmetry. Indeed, due to the symmetry of the modeling, the parameter(s) obtained during the calculation are transposable by symmetry to the other part of the modeling

[0117] If several first phases are carried out, the test piece used during a first phase n may be identical to a test piece used during a first phase n+1 or may be different. In the latter case, the shape of the test piece n+1 will be identical to the shape of the test piece n but its dimensions will vary. In the latter case, care should preferably be taken to adapt the modeling accordingly so that the modeling n is preferably identical to the test piece n and the modeling n+1 is preferably identical to the test piece n+1.

[0118] Although here the independent elastic constants determined are the constants El, E2, G12, vl2, v23 and G23, other independent elastic constants may be determined and preferably the constants associated with the isotropic, orthotropic or anisotropic models.

[0119] The test piece may be manufactured by additive method, using one or more strands of pre-impregnated wires which will then be heated to be permanently joined together, using one or more strands of wires which will then be soldered together...

[0120] Although here the parameters are measured via strain gauges, at least one parameter may be measured in another way, for example by a position sensor, by an optical sensor (such as a camera, image processing making it possible to detect, for example, a unit strain)... It will thus be possible to obtain parameters other than unit strain parameters such as at least one elongation value, at least one jaw displacement value, etc.

[0121] If strain gauges are present, these may be placed on any face of the test piece (internal contour, external contour, first lateral face or second lateral face) and / or on any area of ​​the test piece (upper spar, lower spar, curved longitudinal ends) and / or in any direction. The strain gauges may also be unidirectional and / or bidirectional and / or tridirectional.

[0122] Although here the mechanical test is a four-point bending test, other types of mechanical tests may be considered depending on the independent elasticity constant(s) that one wishes to obtain. Preferably, in all cases, a single type of mechanical test is used to characterize the material used.

[0123] Although here the material to be characterized is a composite material, the material to be characterized could be another type of material such as for example a metallic material.

[0124] Similarly, although here the test piece (and the associated modeling) is thick, the test piece (and the associated modeling) could be thin.

[0125] Although here the modeling is formed from a mesh of tetrahedral meshes, one or more other mesh shapes may be used. Although here a diameter of a mesh (i.e. the largest dimension of a mesh) is between 0.1 and 0.2 millimeters, the diameter may be larger or smaller than this interval.

[0126] Although here the same convergence criterion is applied to all the parameters, different convergence criteria may be applied for at least two of said parameters.

[0127] Although here we validate all the parameters at once, we can validate the parameters as we go along. Thus, if the comparison between one of the measured parameters and one of the corresponding calculated parameters meets the convergence criterion, we can validate this parameter. For the other parameters, it will then be necessary to carry out a second phase again; however, during this new second phase, the parameter already validated will not be calculated again. Its value will be considered definitive, which will limit the calculation time of this new iteration.

[0128] Although here we carry out only one first phase of mechanical testing, we will be able to carry out the first phase several times. The different values ​​of the measured parameters can then be averaged to form averaged measured parameters which will be used for comparisons with the calculated parameters. Alternatively or in addition, for comparisons with the calculated parameters, it will also be possible to rely on at least one standard deviation value associated with the different measurements of the same parameter.

[0129] Alternatively or in addition, at each new iteration of the second calculation phase, an iteration of the first measurement phase may also be implemented. Advantageously, in all cases, the number of mechanical tests will be much less than in the cases of standardized tests of the prior art.

[0130] Although here the iterative loop stops when the convergence criterion is respected, the method could stop otherwise, for example when a maximum number of iterations has been reached.

[0131] Although here the boundary conditions are not modified between two successive first phases and two successive second phases, the boundary conditions may be modified between two successive first phases and / or two successive second phases.

[0132] Although here the shafts are not fixed to the test bench but only rest by gravity on the lower plate (for the lower shafts) or on the test piece directly (for the upper shafts), the shafts may be fixed to the test bench temporarily (for example by magnetic attraction, by screw-nut type fixing, etc.).

Claims

Claims

1. Characterization test piece, characterized in that the test piece is shaped externally into an oblong shape and is hollowed out internally by an orifice (4) of oblong shape.

2. Test tube according to claim 1, in which the orifice (4) is transverse and extends in an axial direction (X), the test tube being shaped into a closed oblong ring.

3. A test piece according to claim 1 or claim 2, wherein a thickness of the test piece is constant around the entire circumference of the test piece.

4. Test piece according to one of the preceding claims, in which the test piece is symmetrical along at least one axis of symmetry.

5. A test piece according to any preceding claim, wherein the test piece has a thickness greater than 5 millimeters.

6. Test tube according to one of the preceding claims, in which the label is made of composite material.

7. A method of mechanical testing on a test piece according to one of the preceding claims, in which the test piece is subjected to a four-point bending test.

8. A method according to claim 7, wherein four shafts contact the test piece during testing, the test piece resting on two first shafts (1004, 1005) by a lower face of the test piece and two second shafts (1007, 1008) bearing on an upper face of the test piece.

9. The method of claim 8, wherein each shaft pressing on the upper face of the test piece is aligned with one of the shafts pressing on the lower face of the test piece.

10. Method according to one of claims 8 to 9, in which a middle of the lower face of the test piece rests on the two first shafts (1004, 1005) and in which the two second shafts (1007, 1008) press on an upper face of the test piece.