digilib@itb.ac.id +62 812 2508 8800

Abstrak - Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

BAB 1 Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

BAB 2 Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

BAB 3 Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

BAB 4 Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

BAB 5 Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

DAFTAR PUSTAKA Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

LAMPIRAN Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

COVER Adila Rildova
Terbatas  Irwan Sofiyan
» Gedung UPT Perpustakaan

High-strain rate design requirements for a product require the consideration of materials' properties under said loading condition. Therefore, reliable material testing under a high-strain rate is needed. The Split-Hopkinson Pressure Bar (SHPB) is a common apparatus for dynamic stress-strain response under high-strain rates. It can be modified for shear stress, becoming a Split- Hopkinson Shear Bar (SHSB) using a hat-shaped specimen as the test object. The manufacturing of a hat-shaped specimen requires minimal tolerance, considering specimen geometry errors might alter the test results. However, less tolerance also means more difficult production and higher costs. This research aims to study the effect of geometric error of hat-shaped specimen for the SHSB by using finite element analysis. This research involves literature study, data gathering, 3D modelling of split-Hopkinson pressure bars and specimens, numerical simulation using Abaqus CAE, analysis of results, and conclusion generation for improved high-shear strain rate testing, with recommendations for better testing. Three types of geometric error are studied. The geometric errors studied are axis perpendicularity, surface parallelism, and surface flatness. This research concludes a geometric tolerance recommendation for axis perpendicularity of 0.260 mm, surface parallelism of 0.087 mm, and surface flatness of 0.130 mm based on the characteristic curve error limit.