20th International Forgemasters Meeting - IFM 2017

The potential consequences of failure within a nuclear reactor means components are designed with a high level of reliability and large margins for error. In a pressurised water reactor, this design criteria is most relevant to parts contained within the reactor coolant system. Amongst others, this includes a number of pressure vessels, each constructed from several forged sections. To ensure structural integrity and the desired safety margins during operation, mechanical properties such as tensile strength and impact toughness are maximised. These properties are a function of the steel grade and manufacturing processes utilised. As a consequence of the conservatism within the nuclear industry there is little motivation to modify the chemistry of tried and tested alloys. Thus attention must be given to the optimisation and design of manufacturing processes, particularly heat treatments. Due to the size, bespoke nature and cost of each forging, empirical investigations are considered infeasible. Therefore computational modelling is a powerful tool when assessing and optimising industrial practices. The aim of this work was to develop a high fidelity computer model capable of predicting accurate cooling data during quenching. Representing scientific phenomena and the quality of input data is key to the accuracy of any model. The paper explains how the latent heat of transformation was established in the form of an effective specific heat capacity using differential scanning calorimetry. These values were then incorporated by a user defined routine in a commercially available finite element modelling software. The paper also describes how values of heat transfer coefficient were established using inverse analysis. This was based upon a production scale heat treatment trial performed by Sheffield Foragemasters international. Finally the completed model was validated and utilised as part of an investigation into the use of welded thermal buffers during quenching. Simulated data produced by the model reinforced a limited set of empirical results, thus helping to qualify a new welding procedure and build confidence when using welded thermal buffers in production.