PROCESS FOR MANUFACTURING AN ALUMINUM ALLOY PART

Abstract

The present invention relates to a process for manufacturing a part (20) comprising a formation of successive metal layers (20.sub.1 . . . 20.sub.n), superimposed on one another, each layer describing a pattern defined from a numerical model, each layer being formed by the deposition of a metal (15, 25), referred to as a filling metal, the filling metal being subjected, at a pressure greater than 0.5 times the atmospheric pressure, to an input of energy so as to melt and constitute said layer, the process being characterized in that the filling metal is an aluminium alloy of the 2xxx series, comprising the following alloying elements: Cu, in a weight fraction of between 3% and 7%; Mg, in a weight fraction of between 0.1% and 0.8%; at least one element, or at least two elements or even at least three elements chosen from: Mn, in a weight fraction of between 0.1% and 2%, preferably of at most 1% and in a preferred manner of at most 0.8%; Ti, in a weight fraction of between 0.01% and 2%, preferably of at most 1% and in a preferred manner of at most 0.3%; V, in a weight fraction of between 0.05% and 2%, preferably of at most 1% and in the preferred manner of at most 0.3%; Zr, in a weight fraction of between 0.05% and 2%, preferably of at most 1% and in a preferred manner of at most 0.3%; Cr, in a weight fraction of between 0.05% and 2%, preferably of at most 1% and in the preferred manner of at most 0.3%; and optionally at least one element, or at least two elements or even at least three elements chosen from: Ag, in a weight fraction of between 0.1% and 0.8%; Li, in a weight fraction of between 0.1% and 2%, preferably 0.5% and 1.5%; Zn, in a weight fraction of between 0.1% and 0.8%.

Claims

1. Method for manufacturing a part including a formation of successive solid metal layers, superimposed on one another, each layer describing a pattern defined from a numerical model (M), each layer being formed by the deposition of a metal, referred to as a filler metal, the filler metal being subjected to an input of energy so as to melt and constitute, by solidifying, said layer, the process being implemented at a pressure greater than 0.5 times the atmospheric pressure, wherein the filler metal is an aluminium alloy of the 2xxx group, comprising at least the following alloying elements: Cu, the weight fraction whereof lies in the range 3 wt. % to 7 wt. %; Mg, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %; at least one element, or at least two elements or even at least three elements chosen from: Mn, the weight fraction whereof lies in the range 0.1 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.8 wt. %; Ti, the weight fraction whereof lies in the range 0.01 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; V, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; Zr, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; Cr, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; and optionally at least one element, or at least two elements or even at least three elements chosen from: Ag, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %; Li, the weight fraction whereof lies in the range 0.1 wt. % to 2 wt. %, optionally in the range 0.5 wt. % to 1.5 wt. %; Zn, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %.

2. Method according to claim 1, wherein the aluminium alloy further includes at least one of the following elements: Si, the weight fraction whereof is at most 1 wt. %; Fe, the weight fraction whereof is at most 0.8 wt. %.

3. Method according to claim 1, wherein the 2xxx group alloy is chosen from AA2022, AA2050, AA2055, AA2065, AA2075, AA2094, AA2095, AA2195, AA2295, AA2395, AA2098, AA2039, and AA2139, and is optionally chosen from AA2075, AA2094, AA2095, AA2195, AA2295, AA2395, AA2039, and AA2139.

4. Method according to claim 1, wherein the weight fraction of Cu lies in the range 4 wt. % to 6 wt. %.

5. Method according to claim 1, including, after formation of the layers, solution heat treatment followed by quenching and aging.

6. Method according to claim 5 including, between the quenching and aging, cold working.

7. Method according to claim 1, after formation of the layers, hot isostatic compression.

8. Method according to claim 1, wherein the filler metal takes on the form of a wire, exposure whereof to an electric arc results in localized melting followed by solidification, so as to form a solid layer.

9. Method according to claim 1, wherein the filler metal takes on the form of a powder, exposure whereof to a laser beam results in localized melting followed by solidification, so as to form a solid layer.

10. Metal part obtained by a method as claimed in claim 1.

11. Metal part according to claim 10 having in the T6 or T8 temper, by a Vickers Hardness HV 0.1 of at least 150 and optionally at least 170 or at least 180.

12. Metal powder or wire comprising, optionally consisting of, an aluminium alloy of the 2xxx group, comprising at least the following alloying elements: Cu, the weight fraction whereof lies in the range 3 wt. % to 7 wt. %; Mg, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %; at least one element, or at least two elements or even at least three elements chosen from: Mn, the weight fraction whereof lies in the range 0.1 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.8 wt. %; Ti, the weight fraction whereof lies in the range 0.01 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; V, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; Zr, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; Cr, the weight fraction whereof lies in the range 0.05 wt. % to 2 wt. %, optionally at most 1 wt. % and in a preferred manner at most 0.3 wt. %; and optionally at least one element, or at least two elements or even at least three elements chosen from: Ag, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %; Li, the weight fraction whereof lies in the range 0.1 wt. % to 2 wt. %, optionally in the range 0.5 wt. % to 1.5 wt. %; Zn, the weight fraction whereof lies in the range 0.1 wt. % to 0.8 wt. %.

13. Wire or powder according to claim 12, further comprising a filler metal for additive manufacturing or welding.

Description

FIGURES

[0067] FIG. 1A is a diagram showing an additive manufacturing method of the WAAM type. FIG. 1B is a photograph of a wall produced according to the method shown with reference to FIG. 1A. FIG. 1C is a diagram showing the wall illustrated in FIG. 1B.

[0068] FIG. 2A shows comparative hardness tests conducted on wall-shaped parts manufactured by the WAAM method from different alloys, the parts having undergone different treatments after the additive manufacturing step.

[0069] FIG. 2B illustrates the evolution, along a transverse axis Z, in the hardness of wall-shaped parts obtained by the WAAM method from aluminium 2139 type alloys respectively with and without implementing a heat treatment resulting in the T6 temper.

[0070] FIG. 2C shows the evolution of the yield strength and tensile strength on test pieces derived from wall-shaped parts formed by WAAM from different alloys, the parts having undergone different treatments after the additive manufacturing step.

[0071] FIG. 2D shows the evolution in the elongation at rupture of parts formed by WAAM from different alloys, the parts having undergone different treatments after the additive manufacturing step.

[0072] FIG. 2E shows fatigue strengths determined during fatigue tests on test pieces derived from wall-shaped parts obtained by the WAAM method from different alloys, the parts having undergone different treatments after the additive manufacturing step.

[0073] FIG. 2F shows comparative hardness tests conducted on wall-shaped parts manufactured by the WAAM method from different alloys.

[0074] FIG. 2G illustrates the evolution, along a transverse axis Z, in the hardness of wall-shaped parts obtained by the WAAM method from aluminium 2295 alloys.

[0075] FIG. 2H shows cross-sections of walls produced from aluminium 2295 alloys and having undergone different heat treatments.

[0076] FIG. 3A and 3B show test pieces respectively used in the tensile and fatigue tests.

[0077] FIG. 4A is a diagram showing an additive manufacturing method of the SLM type.

[0078] FIG. 4B shows hardness measurements for different cube-shaped parts produced by SLM, the parts having undergone different heat treatments after the additive manufacturing step.

DETAILED DESCRIPTION OF THE INVENTION

[0079] In the description, unless stated otherwise: [0080] the designation of the aluminium alloys is compliant with the nomenclature laid down by The Aluminum Association; [0081] the designation of the tempers is compliant with standard NF EN 515 in force in April 2017; [0082] The chemical element contents are denoted as a weight percentage and represent weight fractions.

[0083] FIG. 1A shows an additive manufacturing device of the WAAM type, the acronym of Wire +Arc Additive Manufacturing. An energy source 11, in this case a torch, forms an electric arc 12. In this device, the torch 11 is supplied by an inert gas welding power source. The torch 11 is maintained by a welding robot 13. The part 20 to be manufactured is placed on a support 10. In the embodiment described in FIG. 1A, the manufactured part is a wall extending along a transverse axis Z perpendicular to a longitudinal plane XY defined by the support 10. Under the effect of the electric arc 12, a filler wire 15, in this case forming an electrode of the torch 11, melts to form, by solidifying, a weld bead. The welding robot is controlled by a numerical model M, and is displaced so as to form different layers 20.sub.1 . . . 20.sub.n, stacked on top of one another, forming the wall 20, each layer corresponding to a weld bead. Each layer 20.sub.1 . . . 20.sub.n extends in the longitudinal plane XY according to a pattern defined by the numerical model M. FIG. 1B is a photograph of a wall formed in this way. FIG. 1C is a diagrammatic representation of the wall 20 which extends, along the longitudinal plane XY, in thickness e and in length 1, and along the transverse axis Z, in height h relative to the support 10.

[0084] The method according to the invention is implemented at a pressure that is 0.5 times greater than atmospheric pressure. Thus, unlike the method described in Brice 2015, the Mg content remains high and controlled, which explains the high hardness measured on the wall manufactured from the alloy 2139. Moreover, during the implementation of a T6 treatment, the inventors consider that the controlled Mg and Ag contents of the alloy 2139 allows the best mechanical properties to be obtained owing to a precipitation of the Q phase in the dense planes {111}. Moreover, work at a pressure greater than 0.5 times atmospheric pressure, and advantageously at around atmospheric pressure enables parts to be obtained by additive manufacturing, the mechanical properties of which parts are homogeneous. The term around atmospheric pressure is understood according to the present invention to preferably mean between 80% and 120% atmospheric pressure.

[0085] The inventors attribute the remarkable properties, in particular in terms of mechanical strength, elongation and fatigue properties, to the homogeneity of the Mg content. Operations at atmospheric pressure enable the Mg content to be better controlled, as well as the homogeneity thereof in the parts manufactured by additive manufacturing. This is a particularly important point for applications such as those in the aeronautics field.

[0086] Advantageously, the method according to the invention includes, after the formation of the layers, a solution heat treatment followed by quenching and aging, in particular to obtain a T6 temper. The T6 treatment in particular enables the hardness to be significantly increased, this increase being advantageously at least 50% and preferably at least 60%.

[0087] According to one embodiment, the HIP treatment can be carried out before solution heat treatment, or instead of solution heat treatment. HIP treatment in particular enables the elongation properties and fatigue properties to be improved.

[0088] According to one embodiment, the method includes cold working between quenching and aging, cold working including, for example, modification of a dimension of the part that lies in the range 0.5% to 2%, or even 0.5% to 5%. The inventors have estimated that this enables, for example, an increase in hardness after aging treatment, which can in particular correspond to a T8 temper, and/or a reduction in the aging duration.

[0089] FIG. 4A shows another embodiment wherein the additive manufacturing method implemented is an SLM-type method (Selective Laser Melting). According to this method, the filler material 25 is present in the form of a powder. An energy source, in this case a laser source 31, emits a laser beam 32. The laser source is coupled to the filler material by an optical system 33, the movement whereof is determined as a function of a numerical model M. The laser beam 32 follows a movement along the longitudinal plane XY, describing a pattern that is dependent on the numerical model. The interaction of the laser beam 32 with the powder 25 causes selective melting of the latter, followed by solidification, resulting in the formation of a layer 20.sub.1 . . . 20.sub.n. When a layer has been formed, it is coated in powder 25 of the filler metal and another layer is formed, superimposed on the previously formed layer. The thickness of the powder forming a layer can, for example, lie in the range 10 to 100 m.

[0090] The metal parts obtained after application of a method according to the invention advantageously have, in the T6 or T8 temper, a Vickers Hardness HV 0.1 of at least 150 and preferably at least 170 or even at least 180.

[0091] Advantageously, the metal parts obtained after applying a method according to the invention have, in the T6 or T8 temper, a yield strength R.sub.p0.2 of at least 400 MPa, preferably at least 410 MPa and preferably at least 420 Mpa, and/or an ultimate tensile strength R.sub.m of at least 460 MPa and preferably at least 470 MPa and/or an elongation A % of at least 6% and preferably at least 8% and/or a fatigue strength at 10.sup.5 cycles of at least 240 Mpa and preferably at least 290 MPa.

EXAMPLES

Example 1

[0092] A plurality of filler wires 15 were used in order to manufacture different walls: [0093] alloy 2319 wires corresponding to industrial welding wires; [0094] alloy 2219 and 2139 wires obtained from cast prototype alloys, the wires being obtained by extrusion and wire drawing from billets having a diameter of 55 mm and a length of 150 mm.

[0095] In this example, the filler wire had a diameter of 1.2 mm. An inert gas welding power source available under the reference FK 4000-RFC by Fronius and a Motoman MA210 welding robot by Yaskawa were used.

[0096] The walls had a thickness e in the range 4 mm to 6 mm. The walls had a length l of 10 cm and a height h of 3 cm.

[0097] The parameters for the implementation of the WAAM method were as follows: [0098] torch travel speed: 42 cm/min; [0099] wire feed rate: in the range 5 to 9 m/min; [0100] test conducted at atmospheric pressure.

[0101] The chemical composition of the walls was measured by mass spectrometry of ICP-OES type (inductively coupled plasmaoptical emission spectrometry). The analysis results are provided in Table 1. Each result corresponds to a weight percentage. An analysis was conducted on each wall.

TABLE-US-00001 TABLE 1 Alloy Si Fe Cu Mn Mg Ti Ag V Zr 2319 0.08 0.21 5.7 0.27 <0.01 0.12 <0.01 0.09 0.10 2219 0.04 0.10 6.3 0.29 <0.01 0.03 <0.01 0.12 0.17 2139 0.03 0.05 4.7 0.36 0.42 0.03 0.34 <0.01 <0.01

[0102] The WAAM walls obtained with the different alloys tested did not show any cracks or microcracks.

[0103] Moreover, analyses were also conducted on the filler wires 15. No noteworthy variation was observed as regards the composition between the filler wires and the walls respectively obtained from each filler wire.

[0104] Given that the alloys of the 2xxx group are capable of hardening by heat treatment, a so-called T6 treatment was carried out on the walls 20 so as to obtain a T6 temper. The treatment included a solution heat treatment (duration of 2 htemperatures of 529 C. for 2139 and 542 C. for 2219 and 2319temperature rise in stages of 40 C./h), quenching and aging (duration 25 htemperature of 175 C. for 2219 and 2319duration 15 htemperature of 175 C. for 2139).

[0105] The Vickers Hardness HV 0.1 of the walls 20 was firstly characterised. The measurements were conducted according to standard NF EN ISO 6507-1. The results obtained are shown in FIG. 2A. This figure shows, for each alloy, from left to right, the hardness measured on the filler wire 15 (bdf-1), the wall produced as manufactured (bdf-2), the wall produced after aging (R), and the wall produced after T6 treatment. Each value shown in this figure corresponds to an average of 5 measurements. When the aging was carried out without solution heat treatment and quenching, the parameters (temperature, duration) were identical to those described in the paragraph hereinabove. The hardness obtained using the alloy 2139 is seen to be systematically greater than that of the walls obtained from the other alloys, and in particular alloy 2319, the latter being currently considered to be the alloy of reference for implementing the WAAM method. Moreover, the T6 treatment enables the hardness to be significantly increased, this increase being from about 50% to 60%.

[0106] Moreover, in order to ensure the spatial homogeneity of the hardness of the walls 20 obtained from the alloy 2139, a plurality of measurements of the Vickers Hardness HV 0.1 were carried out at different heights h, along the transverse axis Z. FIG. 2B shows the results obtained on walls that are respectively as manufactured (bdf), i.e. without any post-treatment, and with solution heat treatment, quenching and aging (T6 treatment). The abscissa represents the height h, expressed in mm, whereas the ordinate corresponds to the Vickers hardness measured. The abscissa 5 mm corresponds to the interface between the wall 20 and the support 10 (height equal to 0), materialised by a vertical dashed line. The abscissae less than 5 mm correspond to the support 10. Good homogeneity of the hardness was observed along the transverse axis Z for the two walls analysed. A significant increase in hardness was also observed under the effect of the T6 treatment applied to the wall, the increase being from about 50% to 60%. Obtaining homogeneous mechanical properties is a particularly interesting aspect compared to the method described in Brice 2015, which was implemented at a low pressure.

[0107] Thus, work carried out at a pressure exceeding 50% atmospheric pressure, and ideally at around atmospheric pressure, enables parts to be obtained by additive manufacturing, the mechanical properties of which parts are homogeneous. The term around atmospheric pressure is understood herein to preferably mean between 80% and 120% atmospheric pressure.

[0108] The results exposed in FIG. 2A and 2B show that the alloy 2139 is promising for the implementation of additive manufacturing techniques carried out at atmospheric pressure. Different walls were produced by WAAM based on this alloy, as well as alloy 2319, which is considered to be the alloy of reference. Test pieces were formed on each wall so as to carry out tensile and fatigue tests. The test pieces were sampled either along the transverse axis Z (test pieces V), or along the longitudinal axis Y parallel to the length l of each wall (test pieces H). The geometrical features of the test pieces depended on the tests conducted and will be described hereafter.

[0109] During these tests, the thickness e, the length l and the height h of each wall 20 were respectively equal to about 5 mm, about 440 mm and about 200 mm.

[0110] The walls were subjected to different heat treatments: [0111] T6 treatment: solution heat treatment, quenching and aging so as to obtain the T6 temper. For 2319, solution heat treatment was carried out for 2 h at 542 C., and was preceded by a period in which the temperature was risen by 40 C./h. For 2319, solution heat treatment was carried out for 2 h at 529 C., and was preceded by a period in which the temperature was risen by 40 C./h. For each alloy, aging was carried out for 15 h at 175 C., and was preceded by a period in which the temperature was risen by 40 C./h. [0112] T6 treatment preceded by hot isostatic compression (HIP). For each alloy, the HIP parameters were a pressure and temperature rise over 2 hours from atmospheric pressure and ambient temperature, followed by a period of 2 hours at 497 C. and 1,000 bar.

[0113] FIG. 2C shows the yield strength Rp0.2 results (also referred to by the acronym YS) and tensile strength Rm results (also referred to by the acronym UTS for Ultimate Tensile Stress). The yield strength Rp0.2 corresponds to a relative elongation of the test piece by 0.2%. The test pieces implemented are TOP C1 test pieces defined as per standard NF EN ISO 6892-1 and shown in

[0114] FIG. 3A. Each measurement corresponds to an average of the results obtained for 3 test pieces. The results obtained for each alloy were compared with measurements conducted on test pieces sampled from an industrial sheet metal made of 2139 alloy having undergone T8 treatment. The abscissa corresponds to the alloys used, the ordinate corresponds to the yield strength or tensile strength, measured in MPa. On each alloy, the left-hand bar quantifies the yield strength R.sub.p0.2 whereas the right-hand bar shows the ultimate tensile strength R.sub.m. Letters H and V denote the axes along which the test pieces were sampled.

[0115] It can be seen that the yield strength and tensile strength are systematically greater when using alloy 2139 than when using alloy 2319, regardless of the treatment performed (T6 or HIP+T6), and in particular as regards the yield strength. The performance levels obtained with alloy 2139 are similar to those obtained using the industrial sheet metal (2139-T8).

[0116] The use of alloy 2139 results in increases to the yield strength and tensile strength respectively of about 40% and 10% relative to the walls formed using alloy 2319.

[0117] The reference 2319 T6 Cranfield corresponds to bibliographic data resulting from the publication by Gu Jianglong et al The strengthening effect of inter-layer coldworking and post-deposition heat treatment on the additively manufactured Al-6.3Cu alloy, Journal of Materials Processing Technology, 2016, 230, 26-34.

[0118] Moreover, images of cross-sections of walls were produced, for which a surface fraction of porosity was estimated using image processing software. It was seen that the HIP treatment carried out before the T6 treatment enables a low level of porosity, of less than 0.05%, to be obtained. Without HIP treatment, the porosity levels were in the vicinity of 0.5% with alloy 2139 and about 1.5% with alloy 2319, whereby T6 treatment was applied in each case. The T6 treatment was seen to enable the low porosity level obtained by implementing the HIP treatment to be preserved.

[0119] The use of HIP treatment had no significant effect on the yield strengths or tensile strengths observed. However, as shown in FIG. 2D, such a treatment enables the elongation to be increased to about 14.5% for alloy 2319 and about 9% for alloy 2139, regardless of the sampling direction (test pieces H or V). In FIG. 2D, the ordinate represents the relative elongation of the test pieces resulting from the tensile strength tests, expressed as a percentage.

[0120] Fatigue tests were conducted, using FPE 10 A test pieces as shown in FIG. 3B, according to standard NF EN ISO 6072. FIG. 2E shows the fatigue strength at 10.sup.5 cycles for different alloys.

[0121] Each value obtained is an average of 7 test pieces. Without HIP treatment, the average fatigue strength at 10.sup.5 cycles is about 240 Mpa with alloy 2319 and 245 Mpa with alloy 2139. The implementation of HIP treatment enables the average fatigue strength to be significantly increased, this value reaching 310 Mpa for alloy 2319 and 295 Mpa for alloy 2139.

[0122] The tests presented with reference to FIG. 2D and 2E show the relevance of HIP-type treatment applied prior to T6 treatment. FIG. 2C and 2D show significantly greater performance levels, in terms of yield strength or tensile strength, for the parts formed by additive manufacturing, at atmospheric pressure, using a 2139-type alloy compared to a 2319-type alloy.

Example 2

[0123] Another series of tests was conducted using a filler material formed by a 2295 alloy. Walls 20 similar to those described hereinabove were produced again by implementing a WAAM method at atmospheric pressure. The chemical composition, in terms of weight percentage, of each wall was as follows:

TABLE-US-00002 TABLE 2 Li Si Fe Cu Mn Mg Ti Ag V Zr 1.08 0.02 0.04 4.53 0.34 0.18 0.02 0.23 <0.01 0.15

[0124] Measurements performed on the filler wire did not reveal any significant deviations between the composition of the filler wire and the walls formed therefrom.

[0125] The walls 20 then underwent T6 treatment or T6 treatment preceded by a hot isostatic compression (HIP) step. During the T6 treatment, solution heat treatment was carried out for 2 h at a temperature of 529 C. and aging was carried out for 100 h at a temperature of 160 C.

[0126] FIG. 2F shows the Vickers Hardness HV 0.1 values for the walls 20 obtained by implementing different alloys, these measurements having been performed according to standard NF EN ISO 6507-1. An average value of 5 measurements was calculated for each wall. FIG. 3A shows the average values calculated: [0127] using an alloy 2319 as a filler material, with the wall then being subjected to T6 treatment as described hereinabove; [0128] using an alloy 2139 as a filler material, with the wall then being subjected to T6 treatment as described hereinabove; [0129] using an alloy 2295 as a filler material, with the wall then being subjected to T6 treatment according to the parameters stipulated in the previous paragraph; [0130] using an alloy 2295 as a filler material, with the wall then being subjected to hot isostatic compression (2 hours at 497 C.1000 bar) then T6 treatment.

[0131] The hardness of the wall formed from an alloy 2295 was seen to be clearly greater than that obtained with an alloy 2139. It was also seen that hot isostatic compression, before T6 solution heat treatment enables a hardness of 187 Hv to be obtained, that is to say an increase: [0132] of about 20% relative to the hardness of a wall obtained from an alloy 2139 and having undergone T6 treatment; [0133] of about 35% relative to the hardness of a wall obtained from an alloy 2319 and having undergone T6 treatment.

[0134] FIG. 2G shows a profile of the evolution in hardness according to the height of a wall produced with an alloy 2295, the wall having undergone HIP treatment before the T6 treatment. The ordinate represents the hardness, the abscissa represents the height along the Z axis. The hardness is seen to be spatially homogeneous.

[0135] FIG. 2H shows three cross-sections of walls produced so as to assess the porosity level, and more specifically a surface fraction of porosity. FIG. 2H shows, from left to right, cross-sections of a wall obtained from an alloy 2295, the wall being respectively as manufactured (bdf), having undergone HIP treatment and having undergone HIP treatment followed by T6 treatment (solution heat treatment, quenching and aging). On the wall as manufactured, the surface fraction of porosity was assessed to be 7%, which is attributed to a poor surface condition of the wire formed from the filler material. Hot isostatic compression enables the surface fraction of porosity to be reduced to 0.05%. The implementation of T6 treatment after HIP had no noteworthy effect on porosity.

[0136] These tests show that the alloy 2295 is particularly adapted to the manufacture of parts by additive manufacturing, and more particularly by implementing the WAAM method. Combination with HIP treatment and/or T6 treatment enables remarkable mechanical properties to be obtained.

Example 3

[0137] In this example, walls were produced by the SLM method described hereinabove. In the following tests, the laser source 31 is a Nd/Yag laser with a power of 400 MW.

[0138] Cubic parallelepipeds of dimensions 1 cm1 cm1 cm were formed according to this method, by stacking different layers formed, the powder 25 being obtained from aluminium alloy 2139.

[0139] The composition of the powder was determined by ICP-OES and is given as a weight percentage in the following table.

TABLE-US-00003 TABLE 3 Si Fe Cu Mn Mg Ti Ag V Zr 0.04 0.09 4.8 0.29 0.39 0.05 0.34 <0.01 <0.01

[0140] A particle size analysis was conducted according to standard ISO 1332 using a Malvern 2000 particle size analyser. The curve describing the evolution in the volume fraction as a function of the diameter of the particles forming the powder describes a distribution similar to a Gaussian distribution. If d.sub.10, d.sub.50 and d.sub.90 respectively represent the fractiles at 10%, at 50% (median) and at 90% of the distribution obtained, a rate of uniformity

[00001]$=\frac{d_{90}-d_{10}}{d_{50}}$

and a standard deviation

[00002]$\mathrm{.Math.}=\frac{d_{90}}{d_{10}}$

can be defined. For the powder considered, =4.10.1% and =1.50.1% were measured. The values d.sub.10, d.sub.50 and d.sub.90 were respectively 18.9 m, 38.7 m and 78 m.

[0141] Different cubes were produced by UTBM (Universit de Technologie de Belfort Montbliard) while varying the experimental parameters linked to the power of the laser source 31 and the scanning speed of the beam 32 impacting the powder 25. The parameters are shown in Table 4. The first column corresponds to the references of each test. The second and third columns respectively correspond to the volume energy dissipated by the laser beam 32 and the scanning speed of the beam 32 at the surface of the powder.

TABLE-US-00004 TABLE 4 E (J/mm.sup.3) V (m/min) V5-4 167 40 V5-24 194 30 V5-opt 194 25 V8-18 1,600 5 V8-25 255 23

[0142] Measurements were performed for the Vickers Hardness HV 0.1 either on so-called as manufactured walls (Bdf) not having undergone any treatment after the production thereof, or on walls having undergone T6 treatment, including solution heat treatment, quenching and aging, according to the parameters (temperature and duration) described hereinabove.

[0143] FIG. 4B shows the results obtained, with the Vickers Hardness HV 0.1 being shown as the ordinate. Each result is an average of 4 measurements. This figure also shows the Vickers Hardness HV 0.1 measurements respectively measured on walls manufactured by the WAAM method, respectively as manufactured, undergoing aging and undergoing T6 treatment.

[0144] For the as manufactured walls (Bdf), the hardness reached 10010 Hv, which corresponds to the hardness obtained for walls manufactured by the WAAM method, as manufactured, or having undergone aging. The T6 treatment enabled the hardness to be significantly increased by about 60%, which is in accordance with the observation made with reference to FIG. 2B. The hardness obtained by SLM after T6 treatment was of the same order as that obtained by a wall formed by WAAM after T6 treatment.