Model robustness for films coating hardness

The extensive use of appropriate coatings to improve wear resistance, friction coefficient, electrical properties and protection against corrosion has stimulated a growing interest in their mechanical properties, and especially in hardness measurements. The aim of this study is to compare the robustness of different models tacking into account the indentation size effect on film and substrate proposed in the literature to estimate the hardness of coatings from indentations where both coating and substrate interact. First, experimental data are obtained on a thick (55 mu m) chemical nickel coating chemically deposited on a steel substrate using macrohardness measurements. The high thickness of the coating allows the direct determination of the applied to depict hardness variation with the indentation depth. The best fit is obtained by varying the model parameters, and the model curve is then used as the experimental curve. The experimental data deduced from this experimental curve are then perturbed by a Gaussian random noise with 0 as mean and 1 mu m as standard deviation. A second perturbation is performed by subtracting the results related to the smallest indentation depths and applying the former noise. It is shown that the Jonsson and Hogmark model produced the best prediction. Prediction is less accurate four models which used three or for parameters since these parameters are strongly connected.