Effect of cutting parameters on surface roughness in end milling of AlSi/AlN metal matrix composite
This paper presents the effects of cutting parameters and the corresponding prediction model on the surface roughness in the machining of AlSi/AlN metal matrix composite (MMC). This new composite material was fabricated by reinforcing smaller sizes of AlN particles at volume...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier Ltd
2017
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/73640/ http://irep.iium.edu.my/73640/ http://irep.iium.edu.my/73640/ http://irep.iium.edu.my/73640/1/73640_Effect%20of%20Cutting%20Parameters%20on%20Surface%20Roughness_article.pdf |
| Summary: | This paper presents the effects of cutting parameters and the corresponding prediction model on the surface roughness in the machining of AlSi/AlN metal matrix composite (MMC). This new composite material was fabricated by reinforcing smaller sizes of AlN particles at volume fractions of 10%, 15% and 20%with AlSi alloy. The machining experiments involved of uncoated carbide tool and PVD TiAlN coated carbide and conducted at different cutting parameters of cutting speed (240–400m/min), feed rate (0.3–0.5mm/tooth) and depth of cut (0.3–0.5mm) under dry cutting conditions. Taguchi’s L18 orthogonal arrays approach was performed to determine the optimum cutting parameters using a signal-to-noise (S/N) ratio according to the stipulation of the smaller-the-better. The test results revealed that the type of cutting tool is the most significant factor contributing to the surface roughness of the machined material. A mathematical model of surface roughness has beendeveloped using regression analysis as a function of all parameters with an average error of 10% can be observed between the predicted and experimental values. Furthermore, the optimum cutting parameters was predicted; A1 (uncoated carbide), B2 (cutting speed: 320m/min), C2 (feed rate: 0.4mm/tooth), D2 (axial depth: 0.4mm) and E1 (10% reinforcement) and validation experiment showed the reliable results. |
|---|