TY - JOUR
T1 - A computational analysis of thermal and mechanical conditions for weld metal solidification cracking
AU - Feng, Zhili
PY - 1996/1
Y1 - 1996/1
N2 - Weld metal solidification cracking (hot cracking) has been a major, persistent problem in a variety of engineering alloys. Like many other cracking phenomena, solidification cracking is a result of the competition between the material's resistance to cracking and the mechanical driving force for cracking. For weld metal solidification cracking, the fundamental parameter representing the material's resistance to cracking is the ductility in the solidification brittle temperature range (BTR), and the mechanical driving force is the mechanical strains accumulated in the BTR as weld metal solidifies from the liquidus temperature. This paper presents the development of a finite element analysis procedure and the results regarding the mechanical driving force (mechanical strain evolutions) in the solidification brittle temperature range in which solidification cracking occurs. Autogenous gas tungsten arc welding (GTAW) is simulated because of its popular use in weldability tests. The FEA models are developed for full penetration bead-on-plate welds on thin plates of aluminium alloys 2024-T4 and 5052-O. A commercial general-purpose FEA ccode, ABAQUS, is used in this study. Two new modelling techniques are developed to simulate the effects of solidification which are universal to fusion welding but which have been so far neglected in most heat transfer and stress/distortion analyses of weldments. The first new technique aims at incorporating the microscopic solidification kinetics into the heat transfer analysis for improved modelling of the latent heat of fusion. By using the second technique, an element rebirth scheme, the following solidification related phenomena could be effectively modelled in the stress/strain analysis: (1) The reinitiation of the strain history for the resolidified weld metal. In other words, for elements experiencing melting and solidifying processes, it is necessary to eliminate at the moment of solidification, the plastic strains improperly calculated according to continuum solid mechanics when the elements are in the molten state. (2) The change of reference (initial) temperature for the thermal strain calculation in weld metal. The reference temperature needs to be changed from the ambient temperature to the melting temperature after an element enters the weld pool so that it will expand before it experiences the solidification (during heating), and contract afterwards (during solidification and cooling). (3) The 6.6% volumetric change during the course of solidification of aluminium alloys. The development of these two modelling techniques is discussed at length, followed by model verification based on the measurements of Matsuda and Johnson. The mechanical strains obtained from the FEA models are then used to explain reported experimental observations regarding the cracking tendency in weld metal. The magnitude of the predicted mechanical strains in the solidification temperature ranges is discussed. Finally, the paper indicates that the prediction of solidification cracking will become possible if the methodology developed in the paper can be expanded to (1) quantitatively measure the material resistance in a properly designed laboratory test and (2) quantitatively evaluate the mechanical driving force in a particular welded structure.
AB - Weld metal solidification cracking (hot cracking) has been a major, persistent problem in a variety of engineering alloys. Like many other cracking phenomena, solidification cracking is a result of the competition between the material's resistance to cracking and the mechanical driving force for cracking. For weld metal solidification cracking, the fundamental parameter representing the material's resistance to cracking is the ductility in the solidification brittle temperature range (BTR), and the mechanical driving force is the mechanical strains accumulated in the BTR as weld metal solidifies from the liquidus temperature. This paper presents the development of a finite element analysis procedure and the results regarding the mechanical driving force (mechanical strain evolutions) in the solidification brittle temperature range in which solidification cracking occurs. Autogenous gas tungsten arc welding (GTAW) is simulated because of its popular use in weldability tests. The FEA models are developed for full penetration bead-on-plate welds on thin plates of aluminium alloys 2024-T4 and 5052-O. A commercial general-purpose FEA ccode, ABAQUS, is used in this study. Two new modelling techniques are developed to simulate the effects of solidification which are universal to fusion welding but which have been so far neglected in most heat transfer and stress/distortion analyses of weldments. The first new technique aims at incorporating the microscopic solidification kinetics into the heat transfer analysis for improved modelling of the latent heat of fusion. By using the second technique, an element rebirth scheme, the following solidification related phenomena could be effectively modelled in the stress/strain analysis: (1) The reinitiation of the strain history for the resolidified weld metal. In other words, for elements experiencing melting and solidifying processes, it is necessary to eliminate at the moment of solidification, the plastic strains improperly calculated according to continuum solid mechanics when the elements are in the molten state. (2) The change of reference (initial) temperature for the thermal strain calculation in weld metal. The reference temperature needs to be changed from the ambient temperature to the melting temperature after an element enters the weld pool so that it will expand before it experiences the solidification (during heating), and contract afterwards (during solidification and cooling). (3) The 6.6% volumetric change during the course of solidification of aluminium alloys. The development of these two modelling techniques is discussed at length, followed by model verification based on the measurements of Matsuda and Johnson. The mechanical strains obtained from the FEA models are then used to explain reported experimental observations regarding the cracking tendency in weld metal. The magnitude of the predicted mechanical strains in the solidification temperature ranges is discussed. Finally, the paper indicates that the prediction of solidification cracking will become possible if the methodology developed in the paper can be expanded to (1) quantitatively measure the material resistance in a properly designed laboratory test and (2) quantitatively evaluate the mechanical driving force in a particular welded structure.
UR - http://www.scopus.com/inward/record.url?scp=0029734029&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:0029734029
SN - 0043-2318
VL - 42
SP - 34
EP - 41
JO - Welding Research Abroad
JF - Welding Research Abroad
IS - 1
ER -