Hiras
Mechanical
- Aug 1, 2008
- 1
Hi all,
I am currently investigating on deposition-induced residual stress in plasma-sprayed ceramic coating (multi-layered) on metallic substrate. I want to numerically evaluate the stress and strain histories during the deposition process exploiting the feature of ANSYS “element birth and death”. I have mainly focused my attention on the setting-up of the numerical methodology (I am not interested in inelastic or temperature dependence yet) because I have not managed to perform a complete simulation so far.
Through a batch command file, I am trying to build a 2D symmetrical study characterized by the activation of one row of elements (the idealization modelling of a single deposited layer) at each timestep of calculation. Besides, in order to calculate single-timestep residual stress, I need to perform a coupled thermo-structural analysis for each single timestep. To do this, I am using PLANE 42 elements for the transient thermal heat transfer step and PLANE 45 elements for structural step (in the structural step the transient temperature distributions, obtained in the previous thermal step, are imposed as loads).
I guess the crucial point is represented by the acquisition, at the beginning of a new timestep, of the correct information from the previous timestep in terms of deformation and stress. So my problem is the further strategy to adopt to tackle this particular problem (the birth of a row of elements means a model with a different database).
If anyone has experience with “Element birth and death” feature, I would like to ask some questions. For example:
1. How should I build the model? Which is the strategy I should adopt (coupled thermo-structural for each timestep)?
2. Which are the main commands I should use? Should I use RESUME/RESTART... command? Should I save a .db file (database) for each timestep of calculation?
3. At the beginning of each timestep, which information should I import from the previous timestep (stress, deformation)? If yes, how should I import them? If not, how should I properly take into account the deformation/stress field of the previous timestep?
I would really appreciate any advice and suggestions.
Best regards,
Hiras
PS: If it can help, I have some references (papers) in which are shown the results achieved using this approach.
I am currently investigating on deposition-induced residual stress in plasma-sprayed ceramic coating (multi-layered) on metallic substrate. I want to numerically evaluate the stress and strain histories during the deposition process exploiting the feature of ANSYS “element birth and death”. I have mainly focused my attention on the setting-up of the numerical methodology (I am not interested in inelastic or temperature dependence yet) because I have not managed to perform a complete simulation so far.
Through a batch command file, I am trying to build a 2D symmetrical study characterized by the activation of one row of elements (the idealization modelling of a single deposited layer) at each timestep of calculation. Besides, in order to calculate single-timestep residual stress, I need to perform a coupled thermo-structural analysis for each single timestep. To do this, I am using PLANE 42 elements for the transient thermal heat transfer step and PLANE 45 elements for structural step (in the structural step the transient temperature distributions, obtained in the previous thermal step, are imposed as loads).
I guess the crucial point is represented by the acquisition, at the beginning of a new timestep, of the correct information from the previous timestep in terms of deformation and stress. So my problem is the further strategy to adopt to tackle this particular problem (the birth of a row of elements means a model with a different database).
If anyone has experience with “Element birth and death” feature, I would like to ask some questions. For example:
1. How should I build the model? Which is the strategy I should adopt (coupled thermo-structural for each timestep)?
2. Which are the main commands I should use? Should I use RESUME/RESTART... command? Should I save a .db file (database) for each timestep of calculation?
3. At the beginning of each timestep, which information should I import from the previous timestep (stress, deformation)? If yes, how should I import them? If not, how should I properly take into account the deformation/stress field of the previous timestep?
I would really appreciate any advice and suggestions.
Best regards,
Hiras
PS: If it can help, I have some references (papers) in which are shown the results achieved using this approach.