Thermal modelling of steel welds
The metal adjacent to a weld is exposed to severe thermal events. As a result, complex changes in metallurgical structure occur in this region. By varying certain welding parameters, the weld joint strength can be optimized. Two conditions necessary for hydrogen (cold) cracking of welds to occur are...
Main Author: | |
---|---|
Language: | en |
Published: |
University of Canterbury. Mechanical Engineering
2012
|
Online Access: | http://hdl.handle.net/10092/6466 |
id |
ndltd-canterbury.ac.nz-oai-ir.canterbury.ac.nz-10092-6466 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-canterbury.ac.nz-oai-ir.canterbury.ac.nz-10092-64662015-03-30T15:29:21ZThermal modelling of steel weldsStephen, RogerThe metal adjacent to a weld is exposed to severe thermal events. As a result, complex changes in metallurgical structure occur in this region. By varying certain welding parameters, the weld joint strength can be optimized. Two conditions necessary for hydrogen (cold) cracking of welds to occur are a susceptible microstructure (i.e. hard), and the presence of sufficient diffusible hydrogen. Both of these conditions are strongly influenced by how the weld cools. A finite element model was developed, suitable for predicting the thermal history of welds. The model incorporates temperature dependent thermal properties and latent heats. Arc energy is added as a heat flux density over a double ellipsoidal model of the weld pool. The model was used to obtain heat affected zone cooling times, allowing prediction of hydrogen cracking susceptibility. The predictions were verified with previous finite element and experimental studies. A parametric analysis was performed to find the influence of a number of weld conditions on cooling behaviour. Initial results from three-dimensional models indicated the relative safety, from cracking, of welded bar joints compared to plate joints.University of Canterbury. Mechanical Engineering2012-03-22T02:46:14Z2012-03-22T02:46:14Z1996Electronic thesis or dissertationTexthttp://hdl.handle.net/10092/6466enNZCUCopyright Roger Stephenhttp://library.canterbury.ac.nz/thesis/etheses_copyright.shtml |
collection |
NDLTD |
language |
en |
sources |
NDLTD |
description |
The metal adjacent to a weld is exposed to severe thermal events. As a result, complex changes in metallurgical structure occur in this region. By varying certain welding parameters, the weld joint strength can be optimized. Two conditions necessary for hydrogen (cold) cracking of welds to occur are a susceptible microstructure (i.e. hard), and the presence of sufficient diffusible hydrogen. Both of these conditions are strongly influenced by how the weld cools. A finite element model was developed, suitable for predicting the thermal history of welds. The model incorporates temperature dependent thermal properties and latent heats. Arc energy is added as a heat flux density over a double ellipsoidal model of the weld pool. The model was used to obtain heat affected zone cooling times, allowing prediction of hydrogen cracking susceptibility. The predictions were verified with previous finite element and experimental studies. A parametric analysis was performed to find the influence of a number of weld conditions on cooling behaviour. Initial results from three-dimensional models indicated the relative safety, from cracking, of welded bar joints compared to plate joints. |
author |
Stephen, Roger |
spellingShingle |
Stephen, Roger Thermal modelling of steel welds |
author_facet |
Stephen, Roger |
author_sort |
Stephen, Roger |
title |
Thermal modelling of steel welds |
title_short |
Thermal modelling of steel welds |
title_full |
Thermal modelling of steel welds |
title_fullStr |
Thermal modelling of steel welds |
title_full_unstemmed |
Thermal modelling of steel welds |
title_sort |
thermal modelling of steel welds |
publisher |
University of Canterbury. Mechanical Engineering |
publishDate |
2012 |
url |
http://hdl.handle.net/10092/6466 |
work_keys_str_mv |
AT stephenroger thermalmodellingofsteelwelds |
_version_ |
1716798864694968321 |