Session: CS-19-01 Small Scale Mechanical Testing

Paper Number: 123330

123330 - Improved Analysis of Miniature Impact Toughness Tests 

Charpy impact testing remains a common mechanical testing method for material performance assessment. While more sophisticated methods such as fracture toughness assessment have also become commonplace, impact testing has kept its position in the nuclear industry, among others. This is due to the simplicity of the procedure, continuing presence of specimens in surveillance programs, and abundance of previous results, making comparisons to existing data straight-forward and reliable. Demand to test smaller components and diminishing materials has led to development of miniaturized testing techniques, also in impact testing. On the other hand, instrumentation of the impact hammer and resulting load-displacement curves have extended the material response information available from impact tests greatly.

Since the 1990s, both ISO and ASTM have issued standardized procedures for both instrumented and miniaturized Charpy impact testing. Along with the testing procedures, these standards instruct the user in interpretation of the load-displacement curve. However, while the interpretation of the curve is ideally quite simple, the reality is more complicated. Many details, such as smoothing and fitting of the data, are left to the expertise of the user, potentially leading to irreproducibility and inconsistency in assessment.

In projects AMOS and CHAOS, series of miniature impact tests were conducted. The resulting instrumented impact load-displacement curves were compared to the standard-sized results from previous results. The analysis of instrumented impact curves was studied, and improved procedures are developed to increase the consistency and reproducibility of results, as well as decrease human dependence in analysis.

In order to smoothen noise and oscillations in the load-displacement curve, low-pass filtering of data is suggested. The method has lesser impact on underlying data than previously proposed moving average method. A suitable cutoff frequency can be found by spectral analysis of the signal. Dual-tree complex wavelet transform (DTCWT) based filtering is proposed as an alternative.

In the analysis, different characteristic points are identified from the load-displacement curve. While determining these points, the curve is sectioned into pieces. Fitting Hooke’s line in the elastic region follows previously suggested methods. A polynomial of varying degree is fitted in region following elastic limit. Points of discontinuity reveal unstable fracture initiation and arrest locations, and a model consisting of linear and exponential terms is fitted in the data beyond arrest. Once finalized, the points of interest in the instrumented impact curve are determined in accordance with ASTM and ISO standard guidance by searching the intersections of the fitted curves. A dynamic force adjustment suggested previously is demonstrated to improve the consistency of the results.

The improved methodology of evaluating the instrumented impact load-displacement curves provides a reliable evaluation and identification of the characteristic points, thus increasing the confidence on the material assessment from the data.

Presenting Author: Pentti Arffman VTT Technical Research Centre of Finland

Presenting Author Biography: Pentti Arffman graduated from University of Helsinki as a M.Sc. in physics in 2019. Since then, he has worked in nuclear safety at VTT Technological Research Centre of Finland as a research scientist, studying the mechanical properties of structural materials.

Among the work in several research and customer projects, Mr. Arffman has managed the Nordic project BRUTE, evaluating the properties of the pressure vessel of the decommissioned Barsebäck 2 NPP, as well as work package WP2 of the EURATOM project FRACTESUS, investigating the fracture mechanics of RPV steels with sub-sized specimens.

Authors: