ANALYSIS OF TOLERANCE IN ASSEMBLY OF ANG-FIXED MILLING VISE VA-5 AT PT PRESA GENTA ENGINEERING

• Main Author: Althof, Abdurrauf
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/74887
• Institution: Institut Teknologi Bandung

PT Presa Genta Engineering (PT PGE) is a company engaged in the manufacturing of automotive components. In carrying out orders, PT PGE performs manufacturing processes without applying tolerances. Evaluation of the variations in separately produced components with different machine accuracies is also not conducted, leading to longer processing time as it goes through the physical prototype design phase and component adjustments that do not meet specifications. Market competition requires a short product development duration, necessitating the shift from physical prototypes to digital ones. In an effort to prevent difficult-to- control deviations from the early stages of product development, a tolerance analysis system is designed to enforce tolerances and evaluate these determinations. The tolerance analysis system design is conducted on a case study product, the Vertex Ang-Fixed Milling Vise VA-5. The research begins with the measurement of vise components and proceeds with the identification of mechanisms and determination of geometric requirements. These geometric requirements are processed into tolerance specifications, followed by determining the tolerance values using the constant precision factor (CPF) method. Tolerance analysis is performed using two models, namely, vector loop and tolerance stack, with the Solidworks software. Based on the data processing, the results show that the vector loop model produces variations with a mean of 1.1681 for the worst-case method and 0.8258 for the root sum square (RSS) method. On the other hand, the tolerance stack model yields means of 1.0931 for the worst-case method and 0.7988 for RSS. The vector loop model has a tighter tolerance range compared to the tolerance stack model, by 6.24% for the worst-case method and 3.26% for RSS.