#TITLE_ALTERNATIVE#

• Main Author: ESTRIYANTO (NIM 23705006); Pembimbing: Dr. Ir. Slameto Wiryolukito, YUYUN
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/13176
• Institution: Institut Teknologi Bandung

Metal injection molding (MIM) is a type of compaction process in powder metallurgy technology to produce green compact. Because of the similarities between plastic injection molding and metal injection molding, it thought to be possible to simulate metal injection molding using plastic injection molding software. The software studied in this research is Moldlow Plastic Insight 5.0 that works by a finite element method.<p>An injection molding experiment using three kinds of Fe-2%Ni feedstocks shows different results. Binder constituents type and composition of the feedstocks are not recognized. From the literature study, FTIR, TGA, and DSC testing, it recognized that the binder system consists of Paraffin Wax and HDPE in a composition of 63.5%:36.5% by volume. It still also recognized that there are any other additive compounds, but their weight fraction couldn't be recognized. It also recognized that the injection molding experiment failed because the operating temperature was higher then its binder decomposition temperature and it did not have deaglomerator compound in its binder system.<p>To perform the simulation, the most important parameter is the feedstock properties because the feedstock properties are not available in Moldflow's database. The mixed feedstock properties mathematically approached using rules of mixture, logarithmic additivity rule, and suspension relative viscosity rule. Moldflow requires feedstock datas such as mechanical properties, thermal properties, Cross-WLF Viscosity Model Coefficients, 2 Domain Tait PVT Model Coefficients, and filler properties.<p>The simulation shows that Moldflow Plastic Insigt 5.0 could be used to simulate metal injection molding process. The resulted output are best injection location suggestion, filling simulation, pressure distribution, defect prediction, clamp force prediction, recommended ram speed, shrinkage, and deflection prediction. It also recognized that not all of the feedstock properties can be input into Moldflow's database due to input range limitation of the material thermal conductivity value shall be under 10 W/mC.