HOPKINSON BAR FOR ABRADABLE MATERIALS

Abstract

HOPKINSON BAR FOR ABRADABLE MATERIALS One aspect of the invention relates to an assembly comprising a pressure bar device (200) including: an impactor (1), an incident bar (2) including a first flange (25) opposite the impactor, a first tube (21) mounted on the incident bar, a second tube (22) mounted on the incident bar, a second flange (26) fixed to the first or second tube, a receiving bar (3) located opposite the incident bar, the assembly includes a sample (5) located between the incident bar and the receiving bar, the assembly includes: a furnace (4) opposite the sample, a first insulator (51a) located between the incident bar and the sample, a second insulator (51b) located between the sample and the receiving bar, the insulators having a thermal conductivity lower than the thermal conductivity of the incident bars (2) and receiver (3). Figure to be published with the abbreviation: Figure 1

Description

Title of the invention: HOPKINSON BAR FOR ABRADABLE MATERIALS TECHNICAL FIELD OF THE INVENTION

[0001] The technical field of the invention is that of test benches, in particular an experimental bench for bars known as Hopkinson. TECHNOLOGICAL BACKGROUND OF THE INVENTION

[0002] The Hopkinson bar experimental bench is frequently used to characterize the dynamic compression behavior of specimens.

[0003] Conventionally, an experimental bench of Hopkinson bars is composed mainly of an impactor, an incident bar, a transmission bar instrumented using strain gauges and a sample placed between these two bars. All these elements are coaxial with each other. The impactor strikes the incident bar which compresses the sample between the incident bar and the receiving bar.

[0004] In order to constrain the sample under different temperature conditions, various techniques for heating a sample on a Hopkinson experimental bench are known.

[0005] A first technique for heating the sample is that of Kajberg et al. in Material characterisation using high-temperature Split Hopkinson pressure bar, J. Mater. Process. Technol. 213 (2013) 522-531 using a method of heating the sample by induction coil without putting the bars (receiving and inducing) in contact with the sample. This method allows heating of the sample without heating the tips of the bars.

[0006] A second technique for heating the sample is that of Seo et al. in Constitutive equation for Ti-6A1-4V at high temperatures measured using the SHPB technique, Int. J. Impact Eng. 31 (2005) 735-754. Presenting a method for heating the sample by combining halogen lamps and reflectors heating the sample by radiation phenomenon.

[0007] However, when the sample is brought into contact with the receiving and inductor bars before a test, a temperature gradient appears within the sample due to a heat diffusion phenomenon. The sample is therefore no longer at the desired temperature when it is mounted on the Hopkinson experimental bench.

[0008] Another problem with this experimental method is that it is not possible to carry out post-mortem analyses of a sample with fragile behaviour or an abradable sample because the integrity of the sample is seriously compromised following the test.

[0009] The succession of compressions of the sample does not allow a link to be made between the behavior of the sample (deduced from the first compression) and the post-mortem analyses of the sample.

[0010] Several methods are known for stressing the sample under interrupted uniaxial compression stress, i.e. compression in a direction such that there is no reflection of waves on the experimental bench, thus ensuring the integrity of the sample following the test.

[0011] For example, Nemat-Nasser et al. in Hopkinson technique for dynamic recovery experiments, Proceedings of the royal society A, No. 435, pp. 371-391, 1991 discloses an experimental bench of Hopkinson bars allowing interrupted tests through the bench and using a rigid mass, an incident tube as well as an incident bar comprising a flange and all coaxial with each other. The inner and outer diameters of this tube are mainly directed by a condition of mechanical iso-impedance required between all the elements constituting the experimental bench. The rigid mass via the flange makes it possible to attenuate the wave reflections of an incident wave present in the incident tube in order to have an interrupted compression in the experimental bench. The major disadvantage of this bench is the generation of a new tensile wave through the incident bar which can translate into a wave overlap at the level of the deformation measurement.

[0012] In order to overcome this problem, Song et al. in and quasi-static propagation of compaction waves in a low-density epoxy foam, Experimental Mechanics, n°46, pp. 127-136, 2006 replaced the incident tube of Nemat-Nasser et al. with a pre-setting distance corresponding to the distance "d" between the flange of the incident bar and the rigid mass. This method allows the interruption of a test without generating a new wave in the incident bar. However, this method requires carrying out preliminary tests in order to recalculate the distance "d" for each test under different firing conditions. It is therefore a very time-consuming solution due to the number of tests necessary to determine this distance "d" which is difficult to control due to friction and non-linear effects.

[0013] To circumvent this problem, Prot and Cloete in A tandem momentum trap for dynamic specimen recovery during split Hopkinson pressure bar testing of cancellons bone, Journal of Dynamic Behavior of Materials, vol. 2, pp. 50-58, 2016 discloses an experimental bench of Hopkinson bars with a so-called “tandem” configuration consisting of the use of an incident bar with a first flange and two concentric tubes of approximately the same length. In order for wave propagation to be established between all the elements of the bench, a second flange is fixed to the end of one of the concentric tubes having the largest diameter. All these elements are coaxial with each other and are in contact before testing.

[0014] However, this test bench does not allow for mechanical iso-impedance between the elements of the bench because the flanges are made of a different material from that of the bars. Thus, reflections of parasitic waves appear on the test bench and therefore an unwanted multiple loading on the tested sample takes place. This multiple loading does not allow recovery of the sample for post-mortem analysis of the sample because no link can be established between the potentially induced damage and the first loading of the test piece. Summary of the invention

[0015] The invention provides a solution to the problems mentioned above, by making it possible to bring a sample to a desired temperature when it is mounted on a Hopkinson pressure bar device between an incident bar and a receiving bar.

[0016] A first aspect of the invention relates to an assembly comprising a Hopkinson pressure bar device comprising: • an impactor generating a wave or a pulse, • an incident bar comprising a first flange fixed on a first end of the incident bar and opposite the impactor, • a first tube concentric with the incident bar and mounted on the incident bar, • a second tube concentric to the incident bar and mounted on the incident bar, • a second flange fixed to one end of the first tube or to one end of the second tube, • a receiving bar coaxial with the incident bar and located opposite a second end of the incident bar, the assembly further comprises a sample located between the second end of the incident bar and the receiving bar, the assembly comprises: • an oven, • a first insulator located between the incident bar and the sample, and • a second insulator located between the sample and the receiving bar, • the insulators have a thermal conductivity lower than the thermal conductivity of the incident and receiving bars.

[0017] Thus, the insulators make it possible to keep the part at temperature without a thermal gradient appearing within the sample by a phenomenon of heat diffusion within the incident bar and the receiving bar.

[0018] Advantageously, the sample is an abradable material. This type of device The Hopkinson pressure bar is particularly suitable for the study of abradable materials because it allows for a post-mortem analysis of the sample and for placing the sample in the same temperature conditions as in a turbomachine compressor. During a test, the abradable sample reaches deformation speeds equivalent to an interaction between a blade and an abradable material in a turbomachine. The device therefore also makes it possible to properly characterize an interaction between a blade and an abradable material in a turbomachine.

[0019] Advantageously, the incident bar and the receiving bar are made of aluminum. Aluminum is a material with high thermal conductivity, so it is necessary to thermally insulate the sample using insulators. For example, the incident bar and the receiving bar are made of Al 7075-T6 aluminum alloy with a thermal conductivity of the order of 130 watts per meter-kelvin.

[0020] Advantageously, the insulators are cylindrical, the first insulator has a diameter equal to the diameter of the incident bar, the second insulator has a diameter equal to the diameter of the receiving bar.

[0021] Advantageously, the thermal conductivity of the insulators is between 3 and 20 watts per meter-kelvin. Thus the thermal conductivity of the insulators is less than that of the inductor and receiver bars.

[0022] Advantageously, the mechanical impedance of the insulators is between 6000 and 8000 newton-seconds per meter. Preferably, the mechanical impedance of the insulators is between 6800 and 7000 newton-seconds per meter.

[0023] Advantageously, the mechanical impedance of the incident bars and the receiving bars is between 3000 and 6000 newton-seconds per meter. Preferably, the mechanical impedance of the incident bars and the receiving bars is between 4000 and 5000 newton-seconds per meter.

[0024] Advantageously, the assembly comprises a steel clamp holding the sample and the insulators. The clamp holds the sample and the insulators and also introduces them into the furnace. The clamp is made of 310S steel and withstands the high temperatures of the furnace, by high temperatures we mean temperatures ranging from 50°C to 900°C.

[0025] Advantageously: • the incident bar and the incident flange, • the first tube, • the second tube and the second flange of the second tube, and • the receiving bar, are made of the same material. Thus, there is an iso-impedance between all the elements of the assembly to ensure a single loading (or interrupted uniaxial compression) of the sample. The iso-impedance between all the elements allows for full transmission of waves between all the elements.

[0026] Advantageously, the oven is at a temperature between 50°C and 900°C. Thus, the oven makes it possible to heat the sample under the same thermal conditions as a low pressure compressor or a high pressure compressor.

[0027] A second aspect of the invention relates to a compression method using an assembly having at least one of the preceding characteristics, the method comprises: • a first step of inserting the sample and the insulators into the furnace, • a second step of heating the sample, • a third step of positioning the sample and the insulators between the second end of the incident bar and the receiving bar, • a fourth stage of generation of a wave or a pulse by the impactor on the incident flange.

[0028] The invention and its various applications will be better understood upon reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES

[0029] The figures are presented for information purposes only and in no way limit the invention.

[0030] [Fig.l] is a schematic front view of an assembly according to the invention.

[0031] [Fig.2] is an enlarged view according to the magnifying glass A of [Fig.l] of a first end of an incident bar of the assembly and of a first tube according to the invention.

[0032] [Fig.3] is an enlarged view through magnifying glass B of [Fig.l] of the incident bar of the assembly according to the invention at one end of a second tube concentric with the incident bar.

[0033] [Fig.4] is a perspective view of the second flange of the assembly according to the invention.

[0034] [Fig.5] is a perspective view of a sample between an incident bar and a receiving bar of the assembly according to the invention.

[0035] [Fig.6] is a perspective view of a sample between two insulators of a assembly according to the invention.

[0036] [Fig.7] is a schematic and sectional view along an axis E representing the sample, the clamp and the insulators.

[0037] [Fig.8] is a perspective view of the assembly according to the invention.

[0038] [Fig.9] is a view of an assembly according to the invention representing different sub- steps after a step of generating a wave on the whole. DETAILED DESCRIPTION

[0039] Unless otherwise specified, the same element appearing in different figures has a single reference.

[0040] As shown in [Fig.l], an assembly 100 comprises a bar device 200 Hopkinson pressure test allowing interrupted uniaxial compression tests to be carried out on a sample.

[0041] The device 200 comprises a table 7 having a flat surface 71.

[0042] The device 200 comprises an impactor 1 comprising a cylindrical body 12. The impactor 1 comprises an actuator 11 located at a first axial end of the cylindrical body 12 along an axis E making it possible to start the impactor 1 in order to generate a wave or a pulse. A second end of the cylindrical body 12 is opposite an incident bar 2. The axis E is parallel to the flat surface 71, by means of two first feet 13. The first feet 13 are in the shape of a rectangular parallelepiped. One face of the rectangular parallelepiped is in surface contact with the flat surface 71 of the table. The first feet 13 each comprise a through hole through which the impactor 1 passes. The through holes comprise a plain bearing allowing the translation of the impactor 1 in the through holes of the first feet 13.The surfaces of the rectangular parallelepiped are perpendicular to the flat surface 71 or perpendicular to the face of the rectangular parallelepiped. The cylindrical body 12 of the impactor 1 passes through the holes emerging from each first foot 13, thus allowing the horizontal maintenance of the impactor 1.

[0043] The cylindrical body 12 of the impactor 1 may have a length of between 200mm and 600mm.

[0044] The incident bar 2 is a solid cylinder. The incident bar 2 is coaxial with the impactor 1. The incident bar 2 has: • A first end opposite the second end of the cylindrical body 12 of the impactor 1, • A second end opposite the first end of the incident bar 2.

[0045] The incident bar 2 is held by second feet 24 fixed to the flat surface 71 of the table 7. The second feet 24 each comprise a through hole crossed by the incident bar 2 and comprise a smooth bearing allowing the translation of the incident bar 2 in the holes of each second foot 24 in an axial direction relative to the axis E.

[0046] The incident bar 2 comprises a first tube 21 concentric with the incident bar 2 and mounted on the incident bar 2. The first tube 21 is held by a third foot 212. The third foot 212 is fixed to the flat surface 71. The third foot 212 comprises a through hole crossed by the first tube 21 as well as a plain bearing allowing the translation of the first tube 21 in an axial direction relative to the axis E. A length of the incident bar 2 in the axial direction is greater than a length of the first tube 21. For example, the incident bar 2 has a length of 2000 mm and the first tube 21 has a length of 800 mm.

[0047] The incident bar 2 also comprises a second tube 22 concentric with the incident bar 2 and mounted on the first tube 21. The second tube 22 is held by two fourth feet 222. The fourth feet 222 are fixed to the flat surface 71. The fourth feet 222 each comprise a through hole crossed by the second tube 22 as well as a plain bearing allowing the translation of the second tube 22 in the holes in the axial direction. The length of the incident bar 2 is greater than a length of the second tube 22 in the axial direction. For example, the incident bar has a length of 2000mm and the second tube 22 has a length of 700mm.

[0048] The second tube 22 is mounted on the first tube 21 such that an inner diameter of the second tube 22 is greater than an outer diameter of the first tube 21. The length of the first tube 21 may be greater than or equal to the length of the second tube 22. For example, the first tube 21 may measure 800 mm and the second tube 22 may measure 700 mm.

[0049] As shown in [Fig.2], a first flange 25 is fixed on the first end of the incident bar 2 such that an inner diameter of the first flange 25 is equal to an outer diameter of the incident bar 2. A flat surface of the first flange 25 is opposite one end of the first tube 21. The first flange 25 is in the form of a hollow cylinder. The first flange 25 can also be called an incident flange.

[0050] The first flange 25 is for example screwed onto the first end of the incident bar 2 such that the first end of the incident bar 2 is threaded and the inner diameter of the first flange 25 is tapped. In view of the forces which are involved during high-speed testing, the first flange 25 is subject to high concentrations of bending stresses. This is why during high-speed testing, it is preferable to screw the first flange 25 onto the incident bar 2.

[0051] As shown in [Fig.3], one or more first rings 23 are placed at the interface between the incident bar 2 and the first tube 21 allowing the first tube 21 to be able to translate axially freely relative to the incident bar 2.

[0052] Also, one or more second rings 24 are placed at the interface between the first tube 21 and the second tube 22 allowing the first tube 21 to be able to translate axially freely relative to the second tube 22.

[0053] The rings 23, 24 are for example made of bronze. The rings 23, 24 do not disturb the propagation of the waves in the incident bar 2 because the contact between the rings 23, 24 and the incident bar 2 is radial relative to the axis E and not longitudinal relative to the incident bar 2.

[0054] A second flange 26 is fixed on one end of the second tube 22 which is the end closest to the second end of the incident bar 2.

[0055] As shown in [Fig.4], the second flange 26 has: • a radially inner surface 261 • a first radially outer surface 262 having a diameter equal to an inner diameter of the second tube 22 and in contact with an inner surface of the second tube 22 • a second radially outer surface 263 having a diameter equal to the outer diameter of the second tube 22.

[0056] As shown in [Fig. 1], the device 200 also comprises a receiving bar 3 coaxial with the incident bar 2. A first end of the receiving bar 3 is opposite the incident bar 2.

[0057] A second end of the receiving bar 3 is opposite a spring 8. The spring 8 is coaxial with the receiving bar 3. One of the ends of the spring 8 is fixed to a spring support 81. The spring support 81 is fixed to the flat surface 71 of the table 7. The spring 8 makes it possible to dampen the translation of the receiving bar 3 during a compression test of the device 200.

[0058] The receiving bar 3 is held by fifth feet 31 fixed to the flat surface 71 of the table 7. The fifth feet 31 each comprise a through hole crossed by the receiving bar 3 and comprise a smooth bearing allowing the translation of the receiving bar 3 in the holes of each fifth foot 31 in the axial direction. The receiving bar 3 may have a length equal to the length of the incident bar 2 in an axial direction.

[0059] The incident bar 2, the first flange 25, the first tube 21, the second tube 22, the second flange 26 as well as the receiving bar 3 are made of the same material. Thus, the device 200 has a mechanical iso-impedance. For example, the material may be aluminum, in particular 7075-T6 aluminum.

[0060] The incident bar 2 and the receiving bar 3 each have a strain gauge 9 making it possible to acquire the deformation stresses undergone by a sample 5 during a compression test.

[0061] The device 200 comprises a furnace 4 for heating the sample 5 between the second end of the incident bar 2 and the first end of the receiving bar 3.

[0062] As shown in [Fig.5] and [Fig.7], sample 5 is a cylindrical part held by a gripping clamp 6 and having: • a first flat edge 50a opposite the second end of the incident bar 2. • a second edge 50b opposite the first end of the receiving bar 3.

[0063] Sample 5 is in an abradable material.

[0064] The furnace 4 has an inlet 41 located opposite the sample 5. The clamp 6 can translate in a direction perpendicular to the axis E in order to place the sample 5 in the furnace 4 via the inlet 4L. The clamp 6 is made of a material that can withstand the temperatures of the furnace 4. For example, the clamp 6 is made of 310S steel. The furnace is at a temperature between 300°C and 900°C. This temperature range corresponds to the conditions of use of an abradable material in a turbomachine. More particularly to the temperatures of a low-pressure compressor with a temperature of 400°C or a high-pressure compressor with a temperature of 800°C. The furnace can also reach a temperature of 1000°C.

[0065] As shown in Figures 6 and 7, the assembly 100 comprises a first insulator 51a located between the second end of the incident bar 2 and the first edge 50a of the sample 5 and a second insulator 51b located between the first end of the receiving bar 3 and the second edge 50b of the sample 5.

[0066] The insulators 51a, 51b are cylindrical pellets having a diameter equal to the diameter of the incident bar 2 and the receiving bar 3. The insulators 51a, 51b are identical. The insulators 51a, 51b have a thermal conductivity lower than the thermal conductivity of the sample 5. The insulators 51a, 51b have a thermal conductivity lower than the thermal conductivity of the incident bar 2 and the receiving bar 3. For example, the insulators have a thermal conductivity of between 3 and 20 watts per meter-kelvin. For example, the insulators are made of titanium, more particularly Ti6A14V. Titanium is a metallic material having a low thermal conductivity. The diameter of the insulators 51a, 51b may be greater than the diameter of the sample 5, for example twice the diameter of the sample 5. For example, the insulators 51a, 51b have a diameter equal to 20mm and the sample 5 has a diameter of 10mm.Insulators 51a, 51b are concentric with sample 5.

[0067] [Fig. 7] is a sectional view along an axis E of the sample 5 schematically illustrating the clamp 6, the sample 5 and insulators 51a, 51b. The clamp 6 comprises a first jaw 61c and a second jaw 61d gripping the sample 5. The jaws 61c, 61d also grip the insulators 51a, 51b located on either side of the sample 5. Each jaw 61c, 61d comprises: • a first interior surface 91 matching an exterior surface of the first insulator 51a • a second surface 92 matching an exterior surface of the sample 5 • a third interior surface 93 matching an exterior surface of the second insulator 51b, the third inner surface 93 is identical to the inner surface 91. Each inner surface 91, 92, 93 of the jaws 61c, 61d are half-shaped cylinder.

[0068] The mechanical impedance of the insulators 51a, 51b is for example equal to 6890 newton-seconds per meter. The mechanical impedance of the incident bar 2 and the receiving bar 3 is for example equal to 4460 newton-seconds per meter. For example, Ti6A14V is used as the material of the insulators 51a, 51b because it has a low thermal conductivity allowing the sample 5 to be well thermally insulated as well as a mechanical impedance close to the mechanical impedance of the receiving bar 3 and the incident bar 2. Thus, there is little reflection of parasitic waves during a compression test. The insulators 51a, 51b therefore have a function of insulating the sample 5 as well as a function of non-disturbance of a compression test.

[0069] As shown in [Fig.8], the clamp 6 is fixed to a piston 61 allowing the clamp 6 and therefore the sample 5 and the insulators 51a, 51b to be moved in the furnace 4. The piston 61 moves using a programmable controller, thus protecting the user from the high temperatures of the furnace 4.

[0070] The invention also relates to a method of compressing the sample 5 using the assembly 100. The method comprises: • A first step of inserting or placing the sample 5 and the insulators 51a, 51b into the furnace 4 by the clamp 6. The sample 5 and the insulators 51a, 51b pass through the inlet 41 of the furnace 4. • A second step of heating the sample 5 by the oven 4 to a desired temperature. During this step the clamp 6 still holds the sample 5 and the insulators 51a, 51b. • A third step of positioning the sample 5 and the insulators 51a, 51b by the clamp 6 between the second end of the incident bar 2 and the first end of the receiving bar 3, • A fourth step of generating a wave or a pulse by the impactor 1 on the first flange 25.

[0071] The method may also comprise: • A step of measuring the compressive stresses and the deformation rate undergone by the sample 5 using the strain gauges 9. • A step of recovering the sample 5 after generation of a wave in the device 200. The sample 5 can thus undergo post-mortem analyses following the compression test.

[0072] The method comprises between the third step of positioning the sample 5 by the clamp 6 and the fourth step of generating a wave: • a step of holding the sample 5 and the insulators 5la, 51b by the incident bar 2 and the receiving bar 3 such that the incident bar 2 comes to bear on the first insulator 5la and the receiving bar 3 comes to bear on the second insulator 51b, the movement of the incident bars 2 and receiving bars 3 is carried out by motorized pads; • a step of loosening the jaws 61c, 61d of the clamp 6.

[0073] The fourth step is represented [Fig.9] by a series of sub-steps (“a” to “f”): • a first sub-step (a) such that the impactor 1 strikes the first flange 25 screwed onto the incident bar 2, • a second sub-step (b) such that a wave propagates through the incident bar 2 and through the first tube 21, • a third sub-step (c) such that a first part of Fonde propagating in the first tube 21 is reflected by the second flange 26, a second part of Fonde continues to propagate in the incident bar 2, • a fourth sub-step (d) such that the second tube 22 moves towards the receiving bar 3, the second part of the incident bar 2 is reflected and propagates towards the impactor 1, • a fifth sub-step (e) such that the second part of Fonde is reflected on the first flange 25 through the first tube 21, • a sixth sub-step (f) such that the first tube 21 moves towards the receiving bar 3.

Claims

Claims

1. An assembly (100) comprising a Hopkinson pressure bar device (200) comprising: - an impactor (1) generating a wave or a pulse, - an incident bar (2) comprising a first flange (25) fixed on a first end of the incident bar (2) and facing the impactor (1), - a first tube (21) concentric with the incident bar (2) and mounted on the incident bar (2), - a second tube (22) concentric with the incident bar (2) and mounted on the incident bar (2), - a second flange (26) fixed to one end of the first tube (21) or to one end of the second tube (22), - a receiving bar (3) aligned with the incident bar (2) and located facing a second end of the incident bar (2), - the assembly (100) further comprises a sample (5) located between the second end of the incident bar (2) and the receiving bar (3),characterized in that the assembly (100) comprises: • a furnace (4), • a first insulator (51a) located between the incident bar (2) and the sample (5), and • a second insulator (51b) located between the sample (5) and the receiving bar (3), • a steel clamp (6) holding the sample (5) and the insulators (51a, 51b) • the insulators have a thermal conductivity lower than the thermal conductivity of the incident (2) and receiving (3) bars.,

2. Assembly (100) according to the preceding claim characterized in that the sample (5) is an abradable material.

3. Assembly (100) according to one of the preceding claims, characterized in that the incident bar (2) and the receiving bar (3) are made of aluminum.

4. Assembly (100) according to one of the preceding claims, characterized in that the insulators (51a, 51b) are cylindrical, the first insulator (51a) has a diameter equal to the diameter of the incident bar (2), the second insulator (51b) has a diameter equal to the diameter of the receiving bar (3).

5. Assembly (100) according to one of the preceding claims, characterized in that the thermal conductivity of the insulators (51a, 51b) is between 3 and 20 watts per meter-kelvin.

6. Assembly (100) according to one of the preceding claims, characterized in that the mechanical impedance of the insulators (51a, 51b) is between 6000 and 8000 newton-seconds per meter.

7. Assembly (100) according to one of the preceding claims, characterized in that: - the incident bar (2) and the first flange (25), - the first tube (21), - the second tube (22) and the second flange (26) of the second tube (22), and - the receiving bar (3), are made of the same material.

8. Assembly (100) according to one of the preceding claims, characterized in that the oven (4) is at a temperature between 50°C and 900°C.

9. Compression method, by means of an assembly (100) according to one of the preceding claims, characterized in that it comprises: - a first step of inserting the sample (5) and the insulators (51a, 51b) into the furnace (4), - a second step of heating the sample (5), - a third step of positioning the sample (5) and the insulators (51a, 51b) between the second end of the incident bar (2) and the receiving bar (3), - a fourth step of generating a wave or a pulse by the impactor (1) on the first flange (25).

Images