HAL Id: hal-02454746
https://hal.archives-ouvertes.fr/hal-02454746
Submitted on 24 Jan 2020
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
Influence of the residual stresses in reshaping operations of large aeronautical parts
Ramiro Mena, Dominique Deloison, Jose Aguado, Antonio Huerta
To cite this version:
Ramiro Mena, Dominique Deloison, Jose Aguado, Antonio Huerta. Influence of the residual stresses in reshaping operations of large aeronautical parts. Proceedings of the International Conference on Adaptative Modeling and Simulation ADMOS 2017, Jun 2017, Verbania, Italy. �hal-02454746�
Influence of the residual stresses in reshaping operations of large aeronautical parts
Ramiro Mena1*, Dominique Deloison1, José V. Aguado2 and Antonio Huerta3
1 Airbus Group Innovations
5 Quai Marcel Dassault, 92150 Suresnes, France
Email: [email protected], [email protected] Web page: http://www.airbus.com/
2 ICI-HPC Institute at Ecole Centrale de Nantes 1, Rue de la Noë, 44321 Nantes, France
Email: [email protected] - Web page: http://ici.ec-nantes.fr
3 Laboratori de Calcul Numeric (LaCaN). Departament de Matematica Aplicada III. E.T.S. de Ingenieros de Caminos, Canales y Puertos, Universitat Politecnica de Catalunya, BarcelonaTech,
08034 Barcelona, Spain.
Email: [email protected] - Web http://www.lacan.upc.edu
Key words: Reshaping, Residual Stresses, Distortion
Summary: Large aeronautical parts present important distortions at the end of its manufacturing process. To solve this problem, a reshaping operation is required. This step is highly manual and involves skilled workers. The traditional analysis of reshaping requires to simulate before all the manufacturing steps, in order to determine the residual stresses that generate distortion. Here, a new strategy to study reshaping is presented, which consist in using the initial distortion as a starting point considering the part free of residual stresses.
1 INTRODUCTION
In the aeronautical industry, large monolithic structures are made of aluminum forgings.
These components present residual stresses (RS) as a consequence of non-uniform plastic strains generated during previous manufacturing steps
1. When parts are machined, important distortions arise and reshaping is needed. This step is highly manual, depends exclusively on worker skills and contributes to increment manufacturing costs.
The final goal of this work is to develop a Reduced Order Model
2(ROM) for the reshaping
operation. However, as a first step, the following question has arisen: What is the influence of
RS for reshaping? For simple geometries, as a rectangular plate, the range of RS is ± 30 MPa
3and it is expected that straightening operations will modify them. Nevertheless, aeronautical
parts have a complex geometry. Therefore, a T shaped beam is defined and the four points
bending is selected to be one of the most used reshaping techniques. The goal is to induce a
localized plastic strain with an extra displacement to include the spring back effect. To
perform this study two cases are compared: a distorted part with and without RS. The selected
material is an AA7010.
R. Mena, D. Deloison, J.V. Aguado.and A. Huerta.
2 2 PROBLEM SETUP
2.1 Geometry definition
The T shaped beam is represented in Figure 1. The component can be defined by 4 dimensions; those are the thickness, rib height, wing span and length. The origin of our reference system is placed at the intersection of the two available symmetry planes (X and Z).
This characteristic is used during this report, where only ¼ of the geometry is depicted.
2.2 Simulation procedure
A sequential approach is used to simulate the manufacturing chain. Quenching is the starting point. All the previous material information is supposed to be restarted, as the initial temperature is close to the melting point (e.g. stresses generated during forging). The RS field is obtained in two stages. First, the temperature evolution is calculated through a transient thermal analysis. Then, this information is included in a mechanical step, producing the stress evolution along the heat treatment. All process parameters and material properties were taken from Jeanmart et al
1.
After quenching, a stress relief operation is followed. However, this step is not included in this study as the main idea is to potentiate the initial distortion caused by RS. Then, yield stress is increased via a second heat treatment called ageing. To perform this step, a thermo- mechanical-metallurgical model
4is required. For our problem, this change of material property is included but not simulated.
The next step is machining. Here, two different machining offset M
Oare applied by changing the position of the final part inside the quenched bulk. Those values are 10 and 5 mm, respectively and will be referred as case 1 and 2. At this point, the hypothesis is that any distortion takes place inside the elastic region. The material properties correspond to an AA7010 previously characterized.
Finally, based on the distortion level, the reshaping operation is performed. The vertical stroke Ys is defined as the given displacement to run the bending test. The process parameters are summarized in Table 1. The rollers in contact with the rib and the wing are named as top and bottom, respectively. At a material level, a Chaboche constitutive law is employed thanks to its ability to represent a non-linear isotropic/kinematic hardening behavior. This realistic model is useful to simulate the spring back phenomena together with the Bauschinger effect, both present during the reshaping operation. Material parameters are protected under a non- disclosure agreement and are not provided in this report.
Figure 1: T shaped beam geometry definition, bulk and machined parts.
Machining Offset (mm) 10 5
Position ID P1 P2 P3 P4 P5 P6
Top (mm) 150 300 150 425 425 300 Bottom (mm) 425 425 300 150 300 150 Table 1: Reshaping setup. Longitudinal positioning for top and bottom rollers.
3 RESULTS AND DISCUSSION
3.1 Effect of placing the machined part in the bulk
After quenching, the surface is subjected to compression stresses, which are equilibrated by tension in the core. Its values overcome the as-quenched yield stress (162 MPa) and its distribution is not homogeneous across the part. Then, after ageing, σ
yhas evolved to 390 MPa. Machining breaks the previous equilibrium state and a reconfiguration at stress and geometrical level is produced. As a consequence, distortion is arisen longitudinally and across the wings. Additionally, two different values are obtained for each machining case. In this study, only the longitudinal distortion will be considered for reshaping. For that case, distortion δ is measured along the Z direction, in the line generated between the intersection of the wing bottom surface and the symmetry plane X=0. δ is defined as:
δ = ǀ ymax – yminǀ (1)
where y
maxand y
minare the maximum and minimum vertical displacement, respectively. For the first machining case, the vertical distortion has an inverted U shape with a value of 1.9913 mm. On the other hand, the second offset produces a distortion of 5.6337 mm.
3.2 Effect of reshaping configuration:
Once
δ is generated, six reshaping configurations are tested. To validate them, ageometrical tolerance of ± 0.25 mm is defined. Figure 2 shows that each position has its own optimum stroke where
δ is reduced until a certain limit. However, if this point is overcome,the final distortion will increase. As a consequence, distortion can be minimized but not totally removed. For the machining case 1, all configurations are valid but, by using as criteria the minimum force required to perform the operation, P1 is selected. This is not possible for the warp caused by case 2, where only P5 setup is valid.
3.3 Effect of Residual Stresses during reshaping
There is an offset in the evolution of distortion as a function of the given stroke between
the system with and without RS (see Figure 2). It can be positive, if the stresses generated
during reshaping oppose the initial RS; otherwise, the offset will be negative. To explain this
behavior, P5 is selected. Here, as the initial distortion has a U shape, the bending generated
during reshaping induces tension along the rib. For the RS case, the starting point in the
stress-strain diagram is the compression RS at the rib surface. As there is an inversion at
stress level, more force needs to be applied in order to reach the yield surface by tension. For
the Residual Stress Free (RSF) case, it provides a curve with a positive offset, which is
beneficial when the simplified results are translated to the RS system during a calibration step.
R. Mena, D. Deloison, J.V. Aguado.and A. Huerta.
4
Figure 2: Distortion evolution as a function of the given vertical stroke: case 1 (left) and case 2 (right)
4 CONCLUSIONS
Traditionally, the reshaping problem is divided in two marked stages; first, the residual stress generated during quenching need to be determined and after machining, distortion will arise. Then, the reshaping operation totally depends on the level of initial distortion, where the process parameters, as the rollers positioning and the given stroke, need to be defined and optimized. However, for forged aluminum parts, with the available state-of-art simulation software, the real distortion after machining does not match with the results obtained numerically
5. As an alternative, here we have proposed to study the influence of residual stresses during the reshaping operation by comparing a distorted system with and without residual stresses. It was found that there is an offset in the curve that describes the distortion evolution between both systems. Under this hypothesis, it is possible to study the simplified system to know in advance the response of the real part. Further work will be focus on developing a Reduced Order Model for the reshaping problem.
REFERENCES
[1] P. Jeanmart, J. Bouvaist, Finite element calculation and measurement of thermal stresses in quenched plates of high strength 7075 aluminium alloy, Mat. Sci. Tech. 1 (1985) 765–769.
[2] F. Chinesta, A. Leygue, F. Bordeu, J.V. Aguado, E. Cueto, D. Gonzalez, I. Alfaro, A. Ammar, and A. Huerta. PGD-based Computational Vademecum for efficient design, optimization and control. Arch. Comput. Methods Eng., 20 (2013) 31–59.
[3] J.S. Robinson, D.A. Tanner, C.E. Truman, 50th Anniversary Article: The Origin and Management of Residual Stress in Heat-treatable Aluminium Alloys, Strain 50 (2014) 185–207.
[4] G. Fribourg, Precipitation and plasticity couplings in a 7xxx aluminium alloy: application to thermomechanical treatments for distortion correction of aerospace component, PhD thesis, Institut National Polytechnique de Grenoble - INPG, (2009)
[5] X. Cerutti, Numerical modelling and mechanical analysis of the machining of large aeronautical parts: Machining quality improvement, PhD thesis, Mines Paris Tech, (2014).
