Published 2006 | Version v1
Miscellaneous

New developments for the ultrasonic inspection of austenitic stainless steel welds

Description

EDF R and D undertakes studies in non destructive testing (NDT) for better understanding the influence of various parameters (material, type of defect, geometry) on the 'controllability' of the critical components for nuclear safety. In the field of ultrasonic testing, one of the principal research orientations is devoted to the study of the austenitic stainless steel welds of the primary cooling system. Indeed, the structure of these welds present characteristics making difficult their examination, for example: - a strong anisotropy of the properties of elasticity which, coupled with the heterogeneity of the grain orientations, can involve phenomena of skewing, division and distortion of the beam; - a significant scattering of the waves by the grains involving an high attenuation and sometimes backscattered signals. For several years, actions have been launched to improve comprehension of these disturbing phenomena and to evaluate the controllability of those welds. This work is based on the one hand on experimental analyses on representative mock-ups and on the other hand on the developments of modelling codes taking into account the characteristics of the materials. We present in this document a synthesis of this work by developing the following points in particular: - a description of the phenomena of propagation; - the works undertaken to characterize the structure of the welds; - an example of study coupling experimental and modelling analyses for a butt weld achieved by manual arc welding with coated electrodes. The paper has the following contents: 1. Context; 2. Presentation of the problem; 3. Characterization of austenitic welds; 4. From comprehension to industrial application; 5. Conclusion and perspectives; 5. Conclusion and perspectives. This synthesis shows that each austenitic stainless steel weld is a particular case for the ultrasonic testing. This work allowed to better apprehend the disturbances of the ultrasonic propagation in the welds and thus to propose technical solutions. In particular one highlighted the need for carrying out coupled analyses between experiment and modelling, the latter being very useful for parametric studies and comprehension of the phenomena. To approach as well as possible in modelling the experimental results, realistic models of weld description were developed. However, complementary works are necessary to take into account the following points: - the analysis of welds manufactured with different processes (TIG process in particular); - the continuation of the developments in modelling to obtain increasingly realistic results and to study increasingly complex cases. Work will be focused in particular on: - an evaluation of the attenuation and the backscattering in the welds in order to take into account these parameters in the modelling codes. The complexity of the problem is related to the anisotropy of the weld which involves, contrary to isotropic materials, variations of the attenuation according to the orientation of the grains. This work is the subject of a collaboration with INSA Lyon and IUT Aix en Provence; - the 3D study of the influence of the material, the geometry and the characteristics of the defect. To this aim, actions are undertaken for the coupling of two codes: the CHAMPSONS code (developed by CEA) which would simulate the propagation of the ultrasonic waves in a mock-up with complex materials and irregular geometries and the ATHENA code which would simulate the interaction between the beam and a complex defect

Availability note (English)

Available from: SFEN, 67, rue Blomet, 75015 Paris (France)

Additional details

Publishing Information

Imprint Pagination
9 p.
Report number
INIS-FR--4510

Conference

Title
European nuclear conference. Nuclear power for the 21. century: from basic research to high-tech industry
Acronym
ENC 2005
Dates
11-14 Dec 2005
Place
Versailles (France)

Optional Information

Notes
6 refs., 12 figs., 1 tab.