2008:05 ULiAS 4 – Experimental validation of asoftware that models ultrasonic wavepropagation through an anisotropic weld

New and stronger demands on reliability of used NDE/ NDT procedures and methods have evolved in Europe during the last decade. In order to elaborate these procedures, efforts have to be taken towards the development of mathematical models of applied NDT methods. Modelling of ultrasonic non-destructive testing is useful for a number of reasons, e.g. physical understanding, parametric studies, and the qualification of procedures and personnel. An important issue regarding all models is the validation, i.e. securing that the results of the model and the corresponding computer programs are correct. This can be accomplished by comparisons with other models, but ultimately by comparisons with experiments. In this study a numerical model and experimental results are compared and the work has been performed in collaboration with SQC Kvalificeringscentrum AB.

Four different welds have been investigated to give basic data to a mathematical model that describes the ultra sonic wave paths through the welds in these materials. The welds are made in austenitic stainless steel (type 18-8) and in Inconel 182. Two cuts outs are made in each weld, one longitudinal and one transversal cut across the welds, in order to determine the material orientation.

In the numerical model the incident field, described by rays, is given by a P wave probe model. The ray tracing technique is based on geometrical optics and a 2D algorithm has been developed. The model of the weld is based on a relatively primitive assumption of the grain structure for a V-butt weld. The columnar structure of austenitic welds is here modelled as a weld where each sub region corresponds to a grain group. The response of the receiver is calculated according to Auld’s reciprocity principle. UT data collection was performed by SQC according to guidelines given from Chalmers. The purpose to collect data from real inspection objects with known material structure is to compare experimental data with theoretically calculated values.