CFD Heat Transfer Validation of Turbine Structure Heat Treatment Process
Examensarbete för masterexamen
Volvo Aero Corporation is a subcontractor that mainly delivers structural components to both civil and military jet-engines for several large jet-engine manufacturers such as General Electrics, Pratt & Whitney and Rolls-Royce. One of these components is the Turbine Rear Structure (TRS) which has the main purpose of supporting the engine core. The TRS is fabricated from a combination of cast metal, sheet metal and machined metal parts. During the manufacturing process it undergoes a precipitation hardening process which is a two stage process where the first stage is a so called solution heat treatment and the second stage is the precipitation hardening. During solution heat treatment, the material temperature is slowly increased and held there until the material temperature is suddenly dropped resulting in a very aggressive cooling. This aggressive cooling gives a very non-uniform cooling rate of the component which in turn creates large thermal stresses and finally damage the component. The purpose of this thesis is therefore to, by CFD simulations, obtain a deeper understanding of the heat treatment process and its problems with large thermal stresses. The thesis goal is to establish a methodology for how to tackle similar problems in the future. The thesis methodology was first to investigate how well heat transfer can be predicted for different reference cases and then try to simulate the precipitation hardening process using CFD. A prestudy was made for three cases, ow past at plate, ow through a ribbed channel and ow past a 2D cylinder. The main results from these simulations show that the heat transfer rate can be predicted quite well as long as the turbulence intensity is not too great. The main study was performed in two steps. First a 2D axisymmetric model of the process was simulated where the results showed that radiation is crucial for the overall heat transfer meanwhile the convective heat transfer mainly contributes to local temperature variations. In reality however the ow pattern is very complex and cannot be simplified into a 2D axi-symmetric case. 3D simulations are therefore needed. However due to lack of time and knowledge about mesh generation 3D simulations was never carried out. The final conclusions is that despite that 3D simulations was never performed, the heat treatment process can be simulated. However it will be a very large simulation including all heat transfer phenomena's.
Strömningsmekanik , Energi , Innovation och entreprenörskap (nyttiggörande) , Transport , Fluid mechanics , Energy , Innovation & Entrepreneurship , Transport