XFEM MODELLING IN MULTI-BOLTED JOINTS USING A UNIFIED BOLT PRELOAD

Abstract

MULTI-BOLTED JOINTS ARE ADOPTED AND DESIGNED TO PROVIDE EFFICIENT LOAD TRANSFER WITHIN ASSEMBLED ENGINEERING PARTS. BEARING FAILURE IS FAVORABLE DURING DESIGN PHASE DUE TO MORE PROGRESSIVE FAILURE MODE, HOWEVER, ABILITY OF BY-PASS STRESS TO BE TRANSFERRED TO ADJACENT BOLTS IN MULTI-BOLTED JOINTS PRONE TO CATASTROPHIC NET-TENSION FAILURE. FORMER APPROACH KNOWN AS EQUIVALENT SPRING STIFFNESS (ESS) WAS PROPOSED BUT IT REQUIRES EXPERIMENTAL SLIDING LOAD VALUE. THIS HAS LED TO SEMI-EMPIRICAL APPROACH TO REQUIRE EXPERIMENTAL SET-UP THAN INCORPORATING A GENERIC BOLT LOAD VALUE. THIS PAPER AIMS TO PROVIDE A UNIFIED BOLT LOAD (UBL) VALUE TO BE IMPLEMENTED IN EACH BOLT INDEPENDENT UPON PLATE PROPERTIES AND BOLTS ARRANGEMENTS. STRENGTH PREDICTION WERE TAKEN PLACE BY 3-D EXTENDED FINITE ELEMENT METHOD (XFEM) FRAMEWORK OF VARIOUS STAGGERED AND NON-STAGGERED ARRANGEMENTS TO INCLUDE VARIOUS LAY-UPS TYPES AND PLATE THICKNESS. THE FAILURE LOADS PREDICTIONS IN EACH TESTING SERIES WERE INVESTIGATED AND THEN VALIDATED AGAINST EXPERIMENTAL DATASETS AND ALSO COMPARED WITH PREVIOUS TECHNIQUE (ESS APPROACH). CRACK PATTERNS AND FAILURE MODES FROM THIS APPROACH WERE CONSISTENT WITH EXPERIMENTAL OBSERVATIONS, WHERE NET-TENSION FAILURES WERE OBSERVED WITHIN ALL TESTING SERIES.  LESS GOOD PREDICTION COMPARED TO FROM ESS TECHNIQUE, PARTLY DUE TO SEMI-EMPIRICAL NATURE IN FORMER APPROACH. NEVERTHELESS, REASONABLE AGREEMENT IN UBL TECHNIQUE WITH EXPERIMENTAL DATASETS WERE OBTAINED (AVERAGE DISCREPANCY OF APPROXIMATELY 20%).