Measurement of Residual Stress in as Received and Repaired Clad Components

The application of welded austenitic stainless steel claddings to thick ferritic steel pressure vessels has been used as a means of protection of pressure vessels operating in corrosive environments for decades. Cladding could — in some ways — be considered a predecessor of modern coating techniques that for many reasons is still applied in “modern” times. The process of cladding introduces considerable residual stresses because of the large amount of thermal energy added to the component locally, and because two dissimilar materials are joined with different thermo-mechanical properties. Such residual stress fields around the materials interface are of considerable interest as they could strongly contribute to the generation and propagation of interfacial cracks. The assessment of such stresses by experimental means is particularly challenging because of the size clad components would normally have and because of the sharp stress gradient that would be expected near the materials interface. Clad components containing weld repairs are even more complicated to investigate. Three residual stress measurement methods applied to various clad components are presented in this paper: neutron diffraction, deep hole drilling and the ring core method. Residual strains and stresses have been measured by these methods in a non-repaired clad component and in three components containing weld repairs of different sizes. The measurement methods are described with giving emphasis to their respective strengths and weaknesses for these particular applications. In the case of neutron diffraction three different experimental approaches have been used. The way, in which the measurement techniques have been applied on the different components, is explained and an overview of the experimental results is given. In several cases good agreement between test results has been obtained. For example, all tests show a high tensile stress in the cladding layer of the non-repaired component. In other cases, stress magnitudes found were not in excellent agreement, amongst others due to differences in specimen preparation. The paper is concluded with a critical revue of the applicability of these stress measurement methods to welded claddings based on the results obtained.