A process model for the distortion induced by the electron-beam welding of a nickel-based superalloy
- PDF / 2,110,527 Bytes
- 13 Pages / 612 x 792 pts (letter) Page_size
- 45 Downloads / 207 Views
I. INTRODUCTION
ELECTRON-BEAM welding remains an important manufacturing process, particularly in the automobile, aerospace, construction, and power-generation sectors. This is because it offers a significant number of advantages compared to other fusion welding techniques, despite its high cost. First, weld profiles of high penetration can be obtained, and this enables sections of significant thickness to be joined.[4] Second, as the processing is typically performed under a high vacuum, the likelihood of weld contamination by oxide and nitride inclusions is small, so that the welds are suitable for structural applications in which resistance to fracture and fatigue is particularly important.[5,6] Third, as less power per unit length of weld is required than for the arc-welding processes, the distortion induced in the component is often correspondingly smaller, even when thick sections are joined.[7,8] This helps to minimize the possibility of cracking and to reduce the degradation of the material properties within the weld and heat-affected zone (HAZ). Finally, the process has proved amenable to automation, and this has allowed autogeneous welds of extremely high quality to be reproduced in manufacturing environments.[9,10] However, when electron-beam welding is the preferred joining technique, stringent limits can be placed on the levels of distortions and residual stresses that can be tolerated, often because of the intended use of the structure being fabricated. Where these are exceeded, significant extra costs can be incurred during the manufacturing cycle, since components need to be reworked or scrapped. One way of dealing with this eventuality is to build process models for the various steps in the manufacturing sequence, so that optimal sets of processing parameters can be identified before resorting to [1,2,3]
H.J. STONE, Graduate Student, S.M. ROBERTS, Rolls-Royce Research Fellow, and R.C. REED, Assistant Director of Research, are with the Department of Materials Science and Metallurgy, University of Cambridge/ Rolls-Royce University Technology Centre, Cambridge, United Kingdom CB2 3QZ. Manuscript submitted August 31, 1999.
METALLURGICAL AND MATERIALS TRANSACTIONS A
tests on the factory floor. With the arrival of fast low-cost computers and advanced computer software, the construction and application of this kind of virtual manufacturing capability is rapidly becoming possible. However, when designing such models, the desire to incorporate as much of the relevant process physics as possible must be balanced against the costs of implementation and validation and the limitations of existing computer hardware. For example, if the desired product of a welding model is to generate predictions of the shape of the fusion boundary, then consideration of the fluid flow in the melt pool alone may suffice (e.g., Reference 11). On the other hand, if predictions of the residual stresses and distortion are required, then accurate predictions can be made with a more simplified description of the heat transfer to t
Data Loading...
