Method for manufacturing a casting core for production of a part by lost-wax casting

Abstract

A method for manufacturing a casting core (12) for production of a part by lost-wax casting, this method comprising the following steps: a) preparing a mixture of ceramic fillers and a thermoplastic binder, and shaping the core (12), b) debinding and sintering the core (12) in a sandbox (14), the core (12) being entirely covered with sand (16) and being subjected to a thermal curing cycle, characterized in that: during step a) metal particles (22) are introduced into the mixture, and during step b) at least one tomographic inspection is carried out through the sand (16) to detect the positions of the metal particles (22) in the core (12).

Description

[0001] DESCRIPTION

[0002] TITLE: METHOD FOR MANUFACTURING A FOUNDRY CORE FOR THE PRODUCTION OF A PART BY LOST-WAX CASTING

[0003] Technical field of the invention

[0004] The present invention relates to a method for manufacturing a foundry core for the production of a part by lost-wax casting.

[0005] Technical background

[0006] The technical background includes the document by Sundaram D. et Al.: “Evaluation of Permeability Models for Foundry Molds and Cores in Sand Casting Processes” - Archives of Foundry Engineering, March 19, 2024, XP093268924, as well as documents FR-A1 -3 033 509, US-A1-2010 / 122791, FR-A-768 122, US-B2-11,548,060, DE-A1-10,210,038455 and US-B2-10,041,890.

[0007] The lost-wax casting process involves pouring molten metal into a ceramic mold that has been previously created around a wax model. This manufacturing process is used, for example, to produce parts for aircraft turbomachinery, such as turbine blades.

[0008] Because they are used in extreme temperature conditions, turbine blades are equipped with internal cooling circuits. To create these hollow cavities inside the part, the lost-wax casting process uses the fabrication of ceramic cores, which perfectly mold these internal cavities of the blade.

[0009] Currently, the conventional manufacturing process for a turbine blade core involves high- or medium-pressure ceramic injection (HP-CIM or MP-CIM) of ceramic fillers in powder form. This process comprises three main steps: injection, debinding, and sintering. Debinding is the phase that removes the thermoplastic binder mixed with the fillers before the core is shaped. In granular debinding (in a sand bath), two thermal debinding phases occur: a first debinding phase by capillary action, where the binder liquefies and flows out of the core through capillary forces, and a second debinding phase that degrades the remaining binder in gaseous form, occurring when the decomposition threshold of the materials is reached.The temperature then increases to consolidate and densify the ceramic fillers; this is called sintering. This densification is accompanied by a volumetric shrinkage that varies depending on the composition of the mixture.

[0010] In addition to facilitating the removal of a large portion of the binder via liquid during debinding, the sand that constitutes the granular medium also serves to maintain the integrity of the cores, which are weakened by debinding at the beginning of the sintering phase. Therefore, in a granular firing cycle, experimental conditions dictate that the core must remain embedded in the sand from the beginning to the end of the thermal cycle to maintain its integrity, thus linking the debinding and sintering stages without allowing them to be separated. Consequently, handling the cores outside the sand before the end of sintering is nearly impossible; the cores are then held in place only by the pressure forces exerted by the sand.

[0011] As part of the continuous improvement of the lost-wax casting process and in view of the next generations of turbine blades, with more complex core geometries allowing for more efficient cooling, the debinding-sintering cooking stage is of particular interest.

[0012] Currently, the various thermal cycles used depending on the geometries are not optimized, primarily because the granular firing stage constitutes a true black box in terms of understanding the physical phenomena involved. Thus, in addition to being a very time-consuming step that can last several days, the thermomechanical behavior of the core is not controlled, and the post-firing state is more a matter of chance than control.

[0013] One of the main challenges is therefore to control and understand the phenomena involved between the core and its environment during cooking in order to predict its thermomechanical behavior.

[0014] To achieve this, measuring deformations and using process simulation are necessary. This allows us to model the various physical phenomena and determine the influence of the different material / process parameters involved. Simulation thus makes it possible to understand not only thermal phenomena but also mechanical phenomena potentially responsible for core rejection after firing.

[0015] Thus, one of the main challenges is understanding the thermomechanical phenomena involved in the debinding-sintering firing cycle in order to adapt this cycle to different future core geometries and optimize the cycle time. However, modifying the thermal cycle can only be validated if there is no regression in the post-firing core quality; in other words, it must be possible to ensure that a change in the cycle will not degrade the material integrity or negatively impact the dimensions. Beyond the thermal aspects, which can be monitored relatively easily using thermocouples and instrumented sand baths, it is therefore essential to be able to precisely track the mechanical properties and deformations of the core during firing, i.e., during the debinding and sintering phases.

[0016] In the scientific literature, closed-cell firing is rarely addressed. Furthermore, debinding and sintering processes are often treated separately when studying thermomechanical behavior, meaning that the influence of debinding on deformation is frequently neglected in favor of that of sintering. However, no method is applicable to the debinding-sintering firing stage of foundry cores. Indeed, the fact that the cores are immersed in sand for the entire firing time makes it impossible to observe their state during debinding or sintering. For example, the use of a TMA-type analyzer is not feasible because it is impossible to accurately reproduce experimental conditions with a probe in sand.

[0017] The present invention offers a simple, effective and economical solution to this problem.

[0018] Summary of the invention

[0019] The invention relates to a method for manufacturing a foundry core for the production of a part by lost-wax casting, this method comprising the following steps: a) preparation of a mixture of ceramic fillers and a thermoplastic binder, and shaping of the core, b) debinding and sintering of the core in a sand bath, the core being entirely covered with sand and undergoing a thermal firing cycle, characterized in that: during step a), metallic particles are introduced into the mixture, and during step b), at least one tomographic inspection is carried out through the sand, or even the bath, to detect the positions of the metallic particles in the core.

[0020] The invention thus proposes to monitor the core during its debinding and sintering. This monitoring is performed with high sensitivity using tomography. This monitoring can, in particular, allow for the experimental validation of numerical models / simulations of core deformations during debinding and sintering, as will be described in more detail below. The invention makes it possible to measure the geometric deformation of a ceramic foundry core with unprecedented sensitivity and at novel manufacturing stages.

[0021] The process according to the invention may comprise one or more of the following features or steps, taken individually or in combination with each other:

[0022] - the metallic particles are beads;

[0023] - the metallic particles have an average size or diameter less than or equal to 2mm, preferably less than or equal to 1 mm, and for example on the order of 50pm;

[0024] - the metallic particles are nickel-based alloy;

[0025] - a first check by tomography is carried out through the sand, or even the tank, before the start of said cycle, then a second check by tomography is carried out through the sand, or even the tank, during said cycle;

[0026] - a third check by tomography is carried out through the sand, or even the tank, after the said cycle;

[0027] - at least one tomographic scan is taken during the check(s);

[0028] - the tomographic images taken during the checks are analyzed digitally in order to determine the displacement vectors of the metallic particles.

[0029] - The core shaping in step a) is carried out using a press. The present invention also relates to a foundry core for producing a part by lost-wax casting, this core being obtained by a process as described above and comprising sintered ceramic fillers and metallic particles.

[0030] The present invention further relates to a lost-wax casting process for the production of an aircraft turbomachine part, such as a blade, by means of a casting core as described above.

[0031] Brief description of the figures. Other features and advantages will become apparent from the following description of a non-limiting embodiment of the invention with reference to the accompanying drawings in which:

[0032] [Fig. 1] Figure 1 schematically represents an installation for implementing a process according to the invention,

[0033] [Fig.2] Figure 2 is an example of a tomographic image obtained by tomography of a nucleus shaped with metallic beads inside a sand bath,

[0034] [Fig. 3a-3b] Figures 3a and 3b are tomographic images taken successively during the same firing cycle, and

[0035] [Fig.4] Figure 4 represents the result of a numerical analysis of the two images in Figures 3a and 3b and shows displacement vectors of metallic particles.

[0036] Detailed description of the invention

[0037] Figure 1 schematically represents an installation 10 for implementing a process according to the invention for manufacturing one or more foundry core(s) 12 for producing a part by lost-wax casting.

[0038] During the lost-wax casting process, the dimensions of the core 12 are a crucial, if not the most important, characteristic and must be precisely controlled throughout the various stages of the process. Therefore, each core undergoes a non-destructive testing measurement at the end of each casting stage to determine if its dimensions still meet expectations. This is particularly true at the end of the debinding-sintering stage, which, in addition to the expected shrinkage induced by sintering, can introduce deformations and defects such as cracks or blisters with varying frequency depending on the core geometry, the thermal cycle used, and the material parameters. Consequently, this stage is currently responsible for the rejection of a very significant proportion of the manufactured cores.

[0039] Process simulation is the ideal tool for understanding the phenomena at play during firing and determining the reasons for the appearance of defects and deformations. Using one or more numerical models, it allows for multiple iterations by varying material and process parameters to evaluate their influence on the final dimensions. This makes it possible to modify the process according to expectations without having to perform every experimental test, except for those necessary to validate the predictability of the chosen models.

[0040] In order to control the variability of these steps and with the aim of verifying the results of the modeling and simulations of the manufacturing steps of the ceramic cores, it is therefore necessary to have a control method which allows us to measure the geometric evolution of the cores at each of the manufacturing steps, with not just a general deformation, but with a precise knowledge of the deformation of all the zones.

[0041] The process according to the invention essentially comprises two steps, a first step of preparation and shaping of the core 12, and a second step of debinding and sintering of the core.

[0042] The first step consists of preparing a mixture of ceramic fillers and a thermoplastic binder, and shaping the core 12. This first step is well known to those skilled in the art and will not be described in detail here.

[0043] Ceramic fillers, for example, are made of silica.

[0044] The thermoplastic binder, for example, is based on polymer wax.

[0045] The core can be shaped using a press.

[0046] The second debinding and sintering step is also well known to those skilled in the art. It is described, for example, in document FR-A1-3 033 509. Figure 1 shows a sand bath, i.e., a bath 14 filled with sand 16, which forms a granular material. The core(s) 12 is / are intended to be immersed in the sand 16. Figure 1 shows the case where several cores 12 are immersed in the sand 12.

[0047] The core debinding and sintering step is carried out by subjecting the core to a thermal firing cycle. This cycle is performed through the sand 16, and also through the tank, meaning that the sand tank 14 containing the cores 12 is subjected to this cycle.

[0048] Taking experimental measurements directly on the cores 12 is not possible throughout the entire cycle without risking skewing the results. The invention allows for monitoring the deformation of the cores during the firing cycle without needing to remove them from the sand bath, and therefore without risking unintentional deformation outside the process.

[0049] According to the invention, at least one tomographic check is carried out through the sand 14 to detect any displacements or deformations of the core during the firing cycle.

[0050] To achieve this, metallic particles are introduced into the aforementioned mixture and will be monitored by tomography to analyze their possible movements.

[0051] Tomography is an imaging technique that reconstructs the volume of an object from a series of images taken from outside the object. Tomography is performed using a tomograph.

[0052] Tomography control is carried out through the sand 16 to detect the positions of the metallic particles in the core 12 and thus to be able to determine if parts of the core 23 undergo displacements or deformations during the firing cycle.

[0053] The metallic particles are preferably balls or ball-shaped (spherical).

[0054] The metallic particles may have an average size or diameter less than or equal to 2 mm, preferably less than or equal to 1 mm, and for example on the order of 50 µm. Determining the particle size is an important factor because they must be small enough to achieve homogeneous mixing with the fillers and binder during the preparation of the core 12, and large enough to be detected by the tomograph 18.

[0055] For example, the metallic particles are in nickel-based alloy, which prevents contamination of the mixture during the casting of the alloy.

[0056] An initial tomographic inspection can be performed through the sand before the start of the firing cycle, followed by a second tomographic inspection through the sand during the cycle. A third tomographic inspection can also be performed through the sand after the cycle.

[0057] At least one tomographic image is preferably taken during the examination(s). The image(s) can be in three dimensions (3D).

[0058] Figure 2 shows an example of image 20. In this image 20, we see the nucleus 12 in cross-section. This nucleus has several parts that appear gray amidst the sand 16, which appears black. The magnetic particles 22 appear very bright and clear, therefore mostly white. They can be easily seen, particularly on the edges of the nucleus 12.

[0059] The advantage of the invention lies in the fact that normally, once the firing cycle has begun, access to the cores 12 is only possible once the thermal cycle is complete, thus limiting the understanding of thermomechanical phenomena to a state before and after firing. Here, the sand tray 14 can be removed from the furnace, and the cores 12 can be examined by the tomograph 18 without having to be removed from the tray 14. For example, the same sand tray 14 described previously was removed at the end of the thermal cycle and examined directly by the tomograph 18.

[0060] These tomographic images taken during the checks are advantageously analyzed digitally in order to determine the displacement vectors of the metallic particles.

[0061] The images 20 can thus be compared to images 20 taken before firing. This would make it possible to observe deformations through the displacement of the metallic particles 22 at any point during the thermal cycle, such as between the debinding and sintering phases. This would allow us to differentiate the respective influences of the debinding phase from those of the sintering phase, influences which are currently inseparable. Figures 3a and 3b show a core 12 with images 20 taken at two different times during the firing cycle. Figure 4 represents the results of a numerical analysis of the images 20 in Figures 3a and 3b, which are presented as vectors V1, V2, Vn representing the displacement of the metallic particles 22.These displacement vectors V1, V2, Vn can then serve as validation data for a modeling / simulation of the thermomechanical behavior of the nucleus 12 because we know at any time the deformation of the nucleus 12 as a function of temperature.

Claims

DEMANDS 1. A method for manufacturing a foundry core (12) for producing a part by lost-wax casting, this method comprising the following steps: a) preparation of a mixture of ceramic fillers and a thermoplastic binder, and shaping of the core (12), b) debinding and sintering of the core (12) in a sand bath (14), the core (12) being completely covered with sand (16) and undergoing a thermal firing cycle, characterized in that: during step a), metallic particles (22) are introduced into the mixture, and during step b), at least one tomographic inspection is carried out through the sand (16) to detect the positions of the metallic particles (22) in the core (12).

2. A method according to claim 1, wherein the metallic particles (22) are balls.

3. A method according to claim 1 or 2, wherein the metal particles (22) have an average size or diameter less than or equal to 2 mm, preferably less than or equal to 1 mm, and for example on the order of 50 µm.

4. A method according to any one of the preceding claims, wherein the metal particles (22) are nickel-based alloy.

5. A method according to any one of the preceding claims, wherein a first tomographic inspection is carried out through the sand (16), before the start of said cycle, and then a second tomographic inspection is carried out through the sand (16), during said cycle.

6. Method according to the preceding claim, wherein a third tomographic check is carried out through the sand (16), after said cycle.

7. A method according to any one of the preceding claims, wherein at least one tomographic image (20) is taken during the one or each inspection.

8. Method according to the preceding claim, wherein the tomographic images (20) taken during the checks are analyzed digitally in order to determine vectors (V1, V2, Vn) of displacement of the metallic particles (22).

9. A method according to any one of the preceding claims, wherein the shaping of the core (12) in step a) is carried out using a press.

Images