EP4366897A1

EP4366897A1 - Method for additive manufacturing of turbomachinery parts

Info

Publication number: EP4366897A1
Application number: EP22741341.6A
Authority: EP
Inventors: Hugo Jean-Louis SISTACH
Original assignee: Safran Aircraft Engines SAS
Current assignee: Safran Aircraft Engines SAS
Priority date: 2021-07-05
Filing date: 2022-06-14
Publication date: 2024-05-15
Also published as: US20240216992A1; FR3124748B1; CN117794666A; FR3124748A1; WO2023281176A1

Abstract

The present document relates to a method (200) for additive manufacturing of a turbomachinery part, said part having a primary axis and at least one inclined portion extending in a secondary direction forming a non-zero angle with the primary axis, comprising the steps of: a) for each inclined portion: a1) providing (202) a target roughness of an outer surface of said inclined portion, a2) providing (202) a mechanical lowering of said inclined portion, a3) determining (204) a maximum roughness of the outer surface of said inclined portion depending on the mechanical lowering of said inclined portion; b) determining (206) an overall maximum roughness depending on the maximum roughness of the outer surface of each inclined portion; c) determining (208), depending on the overall maximum roughness, an orientation of the primary axis of the part to be manufactured with respect to a plane of a manufacturing platen of an additive manufacturing device; and d) producing (212) the part by additive manufacturing.

Description

DESCRIPTION

TITLE: Additive manufacturing process for turbomachine parts

Technical field of the invention

This document relates to an additive manufacturing process by fusion on a powder bed, in particular for the manufacture of turbomachine parts.

State of the prior art

It is common today to use additive manufacturing techniques to easily and quickly produce complex parts. When it comes to the manufacture of metal alloy or ceramic parts, the process of selective melting or selective powder sintering makes it possible to obtain complex parts, which are difficult to produce or not possible with conventional methods such as foundry, forge or machining. In particular, it is possible to produce parts with cavities that are difficult to access. The aeronautical field lends itself particularly well to the use of this process.

In addition, such an additive manufacturing process has the advantage of being fast and not requiring specific tooling unlike most conventional processes, which considerably reduces the costs and the manufacturing cycles of the parts.

Such a method generally comprises a step during which is deposited, on a manufacturing plate, a first layer of powder of a metal, a metal alloy or a ceramic of controlled thickness, then a step consisting in heating with a means heating (for example a laser beam or an electron beam) a predefined zone of the layer of powder, and to proceed by repeating these steps for each additional layer, until obtaining, slice by slice, the part final. Such a process can be a process called “laser beam melting” in English or “selective laser melting”. Some turbomachine parts have complex shapes and include parts that are inclined relative to each other, which implies that certain parts of the part end up inclined relative to the manufacturing plate of the additive manufacturing device. Figure 1 shows such an inclined part 10 as arranged on the manufacturing plate 5, with an angle α. The successive melting of the layers 2 can induce a walking effect at the level of the outer surface 3 of the inclined part 10. Indeed, the melting means, for example the laser beams are directed vertically along Z, which induces the walking effect between the layers which have a fixed thickness, and not necessarily consistent with the desired geometry. Furthermore, remaining powder grains can merge with the lower surface 4 of the inclined part 10. These walking effects depend on the angle α. The part thus produced has a relatively high roughness, which can be detrimental to the mechanical strength and to the service life of the part.

In addition, areas of the part which are in the air stream, when the part is arranged in a high-pressure part of the turbine, require a low first maximum roughness to limit pressure drops and boundary layer effects. While the same areas, when the part is arranged in a low pressure part of the turbine, can have a second maximum roughness greater than the first maximum roughness, and do not need such a low roughness.

There is a need to control the roughness of parts produced by additive manufacturing.

For this, some methods consist in measuring the roughness of the parts after their manufacture and in resuming the machining of the surfaces of the parts to obtain a surface condition that complies with the necessary mechanical properties of the part. These additional operations are often expensive, complex and sometimes redundant.

There is a need to improve the roughness control of parts made by additive manufacturing.

Summary of the invention

To this end, this document relates to a process for the additive manufacturing of a turbomachine part, said part having a primary axis and at least one inclined part extending in a secondary direction forming a non-zero angle with the primary axis, comprising the steps: a) for each inclined part: a1) providing a target roughness of an outer surface of said inclined part, a2) providing a mechanical reduction of said inclined part, a3) determining a maximum roughness of the outer surface of said inclined part as a function of the mechanical reduction of said inclined part, b) determining an overall maximum roughness as a function of the maximum roughness of the outer surface of each inclined part, c) determining, as a function of the maximum overall roughness, an orientation of the primary axis of the part to be manufactured with respect to a plane of a manufacturing plate of an additive manufacturing device, and d) producing the part by ad manufacturing dative.

The process makes it possible to obtain a part with an acceptable surface condition and with the mechanical strength necessary for the operation of the part.

The primary axis can be an axis of revolution, an axis of symmetry or an axis along a longitudinal direction of the part. The secondary direction can be along a longitudinal axis, an axis of revolution or an axis of symmetry of the inclined part. The roughness can be the arithmetic mean roughness of the outer surface profile or the maximum roughness of the outer surface profile.

The roughness can be measured by a profilometer with or without contact, for example by a laser or visual profilometer.

Step a1) may include: providing a target roughness of the upper outer surface of the inclined part and a target roughness of the lower outer surface of the inclined part. The upper outer surface may oppose the lower outer surface with respect to a longitudinal plane of the inclined portion.

The mechanical reduction, in fatigue, determined by mechanical fatigue tests at the operating temperature and the operating conditions of the part, can be a function of the roughness, in particular of the roughness of the inclined part when it is subjected to predetermined mechanical stresses. For example, the mechanical reduction may be a mechanical reduction called LCF (for “low cycle fatigue”), which corresponds to low cycle fatigue with respect to known reference curves of the part. The LCF mechanical reduction can be determined by mechanical fatigue tests with stress cycles placed on specimens at a low test frequency. This mechanical reduction LCF can be associated with thermal expansion and shrinkage phenomena due to the temperature to which the part is subjected.

The mechanical damping can be a mechanical damping called HCF (for "High Cycle Fatigue") which corresponds to a vibratory fatigue of the part due to the vibration of the turbomachine. The mechanical reduction HCF can be determined by fatigue tests but with a high test frequency.

The mechanical reduction can be a percentage between a fatigue curve of the part compared to a reference curve of the part when it is not subjected to thermomechanical constraints.

The target roughness may depend on, and/or may be inferred from, the function of the inclined part or an area of said inclined part. The target roughness can be based on an aerodynamic need. For example, areas of the part which are intended to be arranged in an air stream, when the part is a turbomachine high pressure compressor rectifier, may have a target arithmetic mean roughness of less than 1.6 pm. For these same areas, when the part is a low-pressure turbine nozzle of the turbomachine, may have a target arithmetic mean roughness of less than 3.2 pm.

Furthermore, when the inclined part is a connection part between two parts of the part, the target arithmetic average roughness can be less than 3.2 μm. The plane of the build plate can be substantially perpendicular to a direction of the laser rays used for melting the layers of the part, to improve the surface condition. The maximum roughness to be maintained, during step b), can be determined according to the target roughness and the tenable roughness according to the inclination of the part being manufactured. The method may comprise the determination of a first experimental law, said determination comprising the steps:

- providing reference specimens, each reference specimen having a primary axis and being produced by additive manufacturing, each reference specimen comprising a lower surface facing the manufacturing device manufacturing plate forming a first angle with the primary axis and an upper surface opposite said lower surface and forming a second angle with the primary axis,

- for each reference specimen, measure the roughness of the upper surface and the roughness of the lower surface,

- Obtaining said first experimental law by interpolation of the roughnesses of the upper surface and of the lower surface as a function of the first angles and the second angles.

The first angle and the second angle may be manufacturing angles.

The first experimental law may depend on the material of the reference specimen, the thickness of the layers deposited by additive manufacturing, the temperature of the part, for example during use of the part in an engine comprising it, the power laser beams and/or the speed of the laser beams.

The first experimental law can be obtained by a polynomial interpolation or any other suitable function.

The roughness may be an average of roughnesses. For example, several reference specimens, having the same material, the same first angle and the same second angle, can be used for measuring the average roughness. The first experimental law may comprise a first curve obtained for roughness data at standard deviations of approximately +2 from the average roughness, and a second curve obtained for roughness data at standard deviations of approximately -2 from average roughness. Thus, it is possible to predict the variability of the roughness for the same angle.

The first experimental law can be stored in a database.

The first angle and the second angle can be complementary.

Step c) can comprise the determination of the orientation of the primary axis of the part by using the first experimental law. For example, the inverse of the first experimental law can be used as a function of the global maximum roughness to calculate an angle between the outer surface and the plane of the build plate. The orientation of the primary axis can be obtained as a function of said calculated angle and the angle between one, or each, inclined part and the primary axis.

Step c) may include the determination of the orientation of the primary axis of the part by using a law linking the roughness and the manufacturing angle obtained by simulation.

The method may comprise a step for validating the first experimental law, comprising the steps:

- produce by additive manufacturing at least one specimen having a primary axis and comprising an outer surface forming a test angle with the primary axis,

- measure the roughness of the outer surface,

- compare the roughness measured with a roughness calculated according to the test angle and the first experimental law.

When the measured roughness is different from the calculated roughness, the method may include a step of recalibrating the first experimental law.

The method may comprise the determination of a second experimental law, said determination comprising the steps:

- for each reference specimen, measure the mechanical reduction and the roughness of the lower surface and/or the upper surface,

- obtain said second experimental law by interpolation of the mechanical reductions as a function of the roughnesses.

The mechanical allowance may be an average of mechanical allowances. For example, several reference specimens, having the same material, the same first angle and the same second angle, can be used to measure the average of the mechanical reductions.

The method may include, prior to the measurement of the mechanical reduction, the manufacture of the reference specimen by additive manufacturing.

The reference specimens used to determine the first experimental law may be different from the reference specimens used to determine the second experimental law. For example, the reference specimens for determining the second experimental law may include an outer surface which may be either a lower surface or an upper surface and the roughness is measured for the outer surface.

The second experimental law can be determined for different room operating temperatures.

Step a3) may include the determination of the overall maximum roughness using the experimental second law.

Step a3) may include the determination of the overall maximum roughness using a law linking the roughness and the manufacturing angle obtained by simulation.

The second experimental law can be stored in the database. The method may include, for each inclined part, the steps:

- compare the target roughness and the overall maximum roughness,

- when the overall maximum roughness is greater than the target roughness of said inclined part, machine said inclined part to obtain the target roughness, after step d).

The method may include polishing the tapered portion or milling the tapered portion. This step allows to obtain the target roughness.

Polishing can be chemical polishing, tribofinishing, abrasive paste polishing, sandblasting, etc.

The method may include, for each inclined part, the steps:

- compare the target roughness and the overall maximum roughness,

- when the overall maximum roughness is greater than the target roughness of said inclined part, modify the rating of the inclined part by adding a thickness to it, prior to step d).

The thickness can be a function of the machining method of the part to obtain the target roughness which can be polishing or milling of the part.

The method may include, for each inclined part, the steps:

- compare the target roughness and the overall maximum roughness,

- when the overall maximum roughness is greater than the target roughness of said inclined part, modify the dimension of the inclined part, by modifying the angle between the inclined part and the primary axis or by modifying another dimension of the inclined part which can be its length, thickness or width.

The part can be produced by additive manufacturing by depositing a succession of layers of a powder of the material of the part with a thickness between 20 and 60 microns, for example equal to 40 microns.

The part can be a bearing support of the turbomachine or a turbine blade of the turbomachine or a compressor blade of the turbomachine.

This document also relates to a device comprising means for implementing the method as mentioned above.

Brief description of figures

[Fig. 1] Figure 1, already described, shows a section of a part produced by additive manufacturing,

[Fig. 2] Figure 2 shows a perspective view of a bearing support of a turbomachine,

[Fig. 3] figure 3a shows a perspective view of a section of the bearing support of figure 2 and figure 3b a front view of the section of figure 3a, [Fig. 4] figure 4 represents a block diagram of an example of the method of manufacturing the bearing support of figure 2,

[Fig. 5] figure 5 represents a block diagram of an example of determination of laws for characterizing the surface condition of parts obtained by additive manufacturing,

[Fig. 6] figure 6 represents reference specimens used in the process of figure 5,

[Fig. 7] figure 7 represents curves connecting the roughness and the manufacturing angle,

[Fig. 8] Figure 8 shows curves linking mechanical reduction and roughness,

[Fig. 9] figure 9 represents a perspective view of a section of the bearing support after its resizing,

[Fig. 10] Figure 10 shows the arrangement of the bearing support on the build plate of the additive manufacturing device.

Detailed description of the invention

With reference to FIGS. 2 and 3, the bearing support 100 of a turbomachine which can be arranged between a rotor of the turbomachine and a rotor shaft of the coaxial turbomachine. For example, bearing support 100 includes an inner shroud 102 in which the rotor shaft is mounted and an outer shroud 104 on which the rotor is mounted.

The bearing support 100 can be mounted at the level of a fan, a compressor or a turbine of the turbomachine.

The bearing support 100 comprises a connecting wall 106 connecting the inner shroud 102 to the outer shroud 104, which is inclined with respect to the axis X of the inner and outer shrouds. The inner shroud 102 is connected to the connecting wall 106 by an inner rib 108 which also has an inclination with respect to the axis X. The bearing support 100 can comprise other intermediate shrouds 110 having as axis X and connected to the connecting wall 106 by intermediate ribs 112. The intermediate shells 110 may have axes inclined with respect to the axis X. the turbine that attaches to the outer areas. Oil can circulate between the shaft and the bearing support.

Each of the internal rib 108 and the intermediate ribs 112 presents a distinct angle with the X axis. This increases the complexity of producing the bearing support 100 by conventional machining machines. To remedy this, additive manufacturing processes are used to facilitate the production of parts such as the bearing support 100. However, such processes do not make it possible to control the roughness of the external surfaces, in particular of the intermediate ribs 112 and of the internal rib 108 .

An additive manufacturing process 200, represented in FIG. 4, makes it possible to obtain turbomachine parts with controlled surface states. In the following, the method 200 is described in conjunction with the bearing support 100 but can be applied to any other turbomachine part, for example a turbine blade of the turbomachine or a compressor blade of the turbomachine.

The method 200 includes a step 202 of supplying characteristics of the bearing support. For example, target roughnesses and mechanical reliefs for each surface S1, S6, S8, S10 and S11 of the inner rib 108 and the intermediate ribs 112 are provided.

Target roughnesses and mechanical reductions are also provided for each of the surfaces S2, S5, S9 and S12 of the connecting wall 106.

The target roughnesses depend on the layout of the surface in question and the aerodynamic need. For example, surfaces S1, S6, S8, S10 and S11 of inner rib 108 and intermediate ribs 112 may have target roughnesses less than 3.2 µm.

The roughness is an arithmetic average roughness of the profile measured by a profilometer with or without contact.

Alternatively, the roughness may be a maximum roughness of the profile determined by a profilometer with or without contact.

The mechanical reduction is a percentage between a fatigue curve compared to a reference curve in the absence of thermomechanical stresses.

The mechanical reduction can be a mechanical reduction called LCF (for "low cycle fatigue" in English) and/or a mechanical reduction called HCF (for "High Cycle Fatigue"). For example, the mechanical reduction of surfaces S1 to S10 can have a mechanical reduction between 40% and 50%.

The mechanical reduction of the connection surfaces R1 and R3 can also have a mechanical reduction between 40% and 50%.

The surface S12 and the other surfaces of the connecting wall opposite the surfaces S1 to S10 can have a mechanical reduction of less than 20%, in particular less than 15%.

The method 200 includes a step 204 to determine a maximum roughness for each of the surfaces S1-S11 and R1-R3 as a function of the mechanical reduction provided in the previous step 202.

The method 200 then comprises:

- a step 206 to determine an overall maximum roughness as a function of the maximum roughnesses determined in step 204 and possibly of the target roughnesses provided, and

- a step 208 to determine the orientation of the X axis with respect to the manufacturing plate of the additive manufacturing device according to the maximum overall roughness determined in step 206. The orientation of the X axis on the build plate is determined in step 208 using a first experimental law 500 relating the roughness of an exterior surface of a part and a build angle of the part relative to the manufacturing platform.

Step 204 is performed using a second experimental law 600 linking roughness and mechanical reduction.

With reference to FIGS. 5 to 8, the method 300 makes it possible to obtain the first experimental law 500 and the method 320 makes it possible to obtain the second experimental law.

The method includes a step 302 of manufacturing a plurality of reference specimens 402, 404 and 406, by additive manufacturing. For example, the reference specimens 402, 404 and 406 are produced in the same material, such as nickel or titanium, by the same additive manufacturing device.

Each of the specimens 402, 404 and 406 is inclined relative to the Z axis which is perpendicular to the build plate 410. Each of the specimens 402, 404 and 406 has, respectively, a lower surface 402D, 404D and 406D and an upper surface 402U, 404U and 406U opposed to the corresponding lower surface 402D, 404D and 406D. Each bottom surface 402D, 404D, and 406D forms a first angle θ with the plane of the build plate 410, of approximately 80°, 70°, and 45°, respectively.

Each top surface 402U, 404U, and 406U forms a second angle θ2 with the plane of the build plate 410, complementary to the first angle θ of the bottom surface 402D, 404D, and 406D, respectively.

The method 300 comprises measuring the roughness R _z of each of the upper and lower surfaces of the reference specimens 402, 404 and 406. The roughness is an arithmetic mean roughness of the profile measured by a profilometer with or without contact.

For each reference specimen 402, 404 and 406, several models are manufactured with the same material and the same angles μ and <¾ and roughness data are measured for each of these models.

The first experimental law 500 is obtained in step 306 by interpolation of these roughness data and connects the roughness R _z and the manufacturing angle a. The first experimental law 500 comprises a curve 514 interpolating the average of the roughness data, the curve 512 interpolating the roughness data at standard deviations of about +2 from the average roughness, and the curve 514 interpolating the roughness data at standard deviations of about -2 from the roughness mean.

The curves 510, 512 and 514 of the first experimental law 500 are obtained by polynomial interpolation.

The manufacturing angle a corresponds to the first angle eu when it is less than 90° and to the second angle 02 when it is greater than 90°.

The first experimental law 500 can be stored in a database. The first experimental law 500 can be validated by a method 310 comprising the production 312 of test specimens by additive manufacturing. For example, these test specimens may comprise an outer surface extending along an inclined plane with the plane of the build plate with a build angle which may be different from the first angle eu and second angle (¾. For example, the manufacturing angle can be equal to 40°, 50°, 70° and 90°.

The method 310 then includes for each test specimen, measuring the roughness of the outer surface. This measured roughness is compared with a roughness calculated by the first experimental law 500 for the manufacturing angle of the test specimen. The first experimental law 500 can be corrected according to the roughnesses measured on the test specimens.

To determine the second experimental law 600, the method 320 comprises measuring the mechanical reduction of the test specimens and/or the reference specimens at different temperatures, for example at 20° and at 750°.

Similar to the construction of the first experimental law 500, for each reference or test specimen, several models are manufactured with the same material and the same angles, mechanical reduction data being measured for each of these models.

The second experimental law 600 is obtained in step 324 by interpolation of these mechanical reduction data and the mechanical reduction and the roughness R _z .

For a temperature of around 20°, the second experimental law 600 includes a curve 604 interpolating the mean of the mechanical reduction data.

For a temperature of approximately 750°, the second experimental law 600 includes a curve 608 interpolating the mean of the mechanical reduction data.

The sampling points represented by squares, triangles and circles in Figure 8 correspond to the measurements of the mechanical reduction.

The method 200 includes a step 210 of validating the orientation of the X axis with respect to the manufacturing plate.

This step 210 may include, for each of the inclined surfaces S1-S11 and R1-R3, the sub-steps:

- compare the target roughness and the maximum roughness,

- when the overall maximum roughness is greater than the target roughness S1-S11 and R1-R3, modifying the rating of said inclined surface S1-S11 and R1-R3 by adding a thickness to it. Step 210 may include, for each of the inclined surfaces S1-S11 and R1-R3, the sub-steps:

- compare the target roughness and the maximum roughness, - when the overall maximum roughness is greater than the target roughness of said inclined surface S1-S11 and R1-R3, modify the rating of the inclined part S1-S11 and R1-R3, by modifying the angle between the inclined surface and the X axis.

As shown in Figure 9, the angle of the rib 108 is modified because the overall maximum roughness that will be obtained at the end of the additive manufacturing is far from the target roughness of the surfaces S1 and R1.

The method 200 may then include a return to step 202 with new mechanical characteristics of the bearing support 100 i.e. a new angle of inclination of the rib 108.

Finally, the method 200 includes a step 212 of producing the bearing support 100 by additive manufacturing. For this, the bearing support 100 will be oriented so that the X axis is perpendicular to the plane of the build plate 410 and the surface S11 faces the build plate.

To this end, honeycomb supports 712, as shown in Figure 10, are arranged under the remote parts of the build plate 410 to prevent the bearing support 100 from collapsing during manufacture. These supports 712 can be removed manually or by machining.

Figure 9 shows the bearing support 100 as obtained by additive manufacturing.

Process 200 may include machining one or more sloped surfaces to achieve the target roughness as determined in step 210.

For example, the surfaces of area 702 are fitted by machining and then polished by sandblasting.

The surfaces of zones 704, 706 and 710 can only be polished by sandblasting. The machining of the can also be carried out by milling or by polishing such as chemical polishing, tribofinishing, polishing with abrasive paste, sandblasting, etc.

The added thickness determined in step 210 can be a function of the type of machining.

Claims

1. Method (200) for additive manufacturing of a turbomachine part (100), said part having a primary axis (X) and at least one inclined part (106,108,110) extending in a secondary direction forming a non-zero angle with the primary axis, comprising the steps: a) for each inclined part: a1) providing (202) a target roughness of an outer surface (S1-S11,R1-R3) of said inclined part, a2) providing (202) a mechanical reduction of said inclined part, a3) determining (204) a maximum roughness of the outer surface of said inclined part as a function of the mechanical reduction of said inclined part, b) determining (206) an overall maximum roughness as a function of the maximum roughness of the outer surface of each inclined part, c) determining (208), based on the overall maximum roughness, an orientation of the primary axis of the part to be manufactured with respect to a plane of a work plate (410) an additive manufacturing device, and d) producing (212) the part by additive manufacturing.

2. Method (200) according to claim 1, comprising the determination (300) of a first experimental law (500), said determination comprising the steps:

- supplying (302) reference specimens (402,404,406), each reference specimen having a primary axis and being produced by additive manufacturing, each reference specimen comprising a lower surface (402D,404D,406D) facing the manufacturing platform of manufacturing device forming a first angle (a-i) with the primary axis and an upper surface (402U,404U,406U) opposite said lower surface and forming a second angle (02) with the primary axis,

- for each reference specimen, measure (304) the roughness of the upper surface and of the lower surface,

- Obtaining (306) said first experimental law by interpolation of the roughnesses of the upper surface and of the lower surface as a function of the first angles and of the second angles.

3. Method (200) according to claim 2, in which step c) comprises determining the orientation of the primary axis of the part using the first experimental law (500).

4. Method (200) according to one of claims 2 or 3, comprising the determination (320) of a second experimental law (600), said determination comprising the steps:

- for each reference specimen, measure (322) the mechanical reduction and the roughness of the lower surface and/or of the upper surface, - obtaining (324) said second experimental law by interpolation of the mechanical reductions as a function of the roughnesses.

5. A method (200) according to claim 4, wherein step b) comprises determining the overall maximum roughness using the second experimental law (600).

6. Method (200) according to one of the preceding claims, comprising, for each inclined part (106,108,110), the steps:

- compare the target roughness and the overall maximum roughness,

7. Method (200) according to the preceding claim, comprising polishing the inclined part or milling the inclined part (106,108,110).

8. Method (200) according to one of the preceding claims, comprising, for each inclined part (106,108,110), the steps: - comparing the target roughness and the overall maximum roughness,

- when the overall maximum roughness is greater than the target roughness of said inclined part, modify the rating of the inclined part (106,108,110) by adding a thickness to it, prior to step d).

9. Method (200) according to one of the preceding claims, in which the part is a bearing support of the turbomachine or a turbine blade of the turbomachine or a compressor blade of the turbomachine.