Modeling and offline simulation of thermal spray coating process for gas turbine applications

摘要

Modern thermal spray coatings for complex hot gas components such as turbine blades and vanes have very strict requirements for the distribution of thickness. Meeting the requirements is critical for the performance and the lifetime of components. In particular, thickness of the thermal barrier coatings (TBC) determines a temperature gradient through the coating, which provides a thermal protection of cooled substrate and influences thermo-mechanical properties. The majority of high temperature protective coatings are applied with thermal spray techniques. Although there are many scientific and technical studies, all peculiarities of the thermal spray process and their relation to final coating properties are not completely understood. In practice, the parameters of the deposition process to produce coatings with required characteristics are established by a “trials and errors” approach with involvement of various process control techniques. These trials are usually done in production booths with an industrial robot, which makes this approach in most cases very expensive and time-consuming. In order to simplify and speed up the coating development, various models and software tools were developed to simulate the deposition process and predict specific coating properties. These models are predominantly focused on some selected aspects of the coating deposition. At the same time, there is no available model to provide a reliable prediction of thickness of the final coating layer on a particular substrate. In this paper, a self consistent model based on physical principles is developed to simulate a thickness distribution of thermally sprayed coatings. In particular, the mass conservation principle with the application of geometric considerations was applied to model the spray jet and the resulting coating pattern. An influence of the process conditions on the basic spray patterns represented by spray spot and spray profile is theoretically investigated. An analytic relationship between thickness distribution in the spray spot, spray profile and thickness of the corresponding coating layer produced by a motion of the spray gun over a flat and cylindrical substrate was established, and corresponding results were discussed. The model results and assumptions were verified in the corresponding experiments. The application of the model was outlined for an arbitrary free-form substrate surface. Some aspects of coating process modeling, related to development of the off-line programming (OLP) software tools to perform robot programming with simultaneous numerical simulation of the resulting coating thickness are discussed. An example of implementation of the developed model into RobCad software to predict thickness distribution of a TBC coating on a gas turbine blade is presented and discussed. Application of the coating deposition model in combination with the OLP technique enables a full-cycle digital coating program development, considerably optimizing development time and costs