Statistical analysis of surface roughness during high speed edge trimming of flax-fiber composite based on the Split-Split plot randomized complete block design

Abstract

This paper attempts to highlight some of the most common problems facing the machining of flax fiber composite. The essential aspects of modeling surface roughness of flax-fiber composite are also covered in this work. The most appropriate model for predicting surface roughness is then selected. The trimming experiments were conducted based on the Split-Split Plot randomized complete block design, varying the cutting velocity, feed, cutting tool geometry and fiber orientation as cutting parameters against surface roughness and delamination as responses. Accordingly, two cutting tools were tested and the most suitable tool was selected. Results show that the two-flute uncoated carbide end mill tool with cutting profiles of 70° (tool # 2) provides more accurate and repeatable results than the two-flute polycrystalline Diamond end mill (tool # 1). Results also show that the B-spline model provides more accurate estimates of the surface roughness than the two other proposed models.