Abstract

Micro-structured surfaces, fabricated by ultra-precision diamond turning (UPDT) with nanometric surface texture, have been widely used due to their wide range of functionalities. Their functional performance is closely related to their surface texture. To characterise the surface texture of the micro-structured surfaces, boundary effect would happen in the boundary area when surface filtering and yield a crucial impact upon their characterisation. However, the boundary effect has received relatively little attention for the surface texture characterisation. Hence, this study focuses on discussing the boundary effect in characterising the surface texture of micro-structured surfaces. Firstly, discrete wavelet transform (DWT) is used to extract the surface texture from the primary surface. Secondly, a probability-based approach is applied to remove the points with large amplitude at the boundary area of the micro-structures in order to reduce the boundary effect. Finally, the UPDT experiment is conducted for the fabrication of micro-structured surfaces and their surface texture is evaluated by the height parameters Sa, Sq, Sz, Ssk, and Sku of ISO 25178-2. The results show that the boundary effect can be reduced efficiently by the probability-based approach during surface filtering. Moreover, Sz, Ssk, and Sku could be more sensitive to the boundary effect than Sa and Sq due to their different statistical characteristics. Significantly, the study provides an effective method to reduce boundary effect in characterising the surface texture of micro-structured surfaces.