bmercer
Mechanical
- Dec 17, 2013
- 1
Hello all,
I'm trying to set up a model for a device involving a pre-stressed wire embedded in a substrate. I have a model which is working OK but I'm wondering what if there's anything I could do to make things run a little smoother, so, any advice is greatly appreciated. Here are the main things I'm dealing with:
* The model consists of a wire (beam elements) which is first bent into a particular shape (hence, pre-stressed) and then implanted in a thin substrate (2d plane stress). The substrate-wire assembly is then allowed to relax. In order to get the wire and substrate to "bond" together, I'm first running the wire deformation problem. Separately, I create a mesh of the substrate. Then I have an external code which reads my wire/substrate files and finds the closest wire node for each substrate node, and writes constraint equations that enforce that these nodes move together after the wire is initially deformed. All of this goes into a model containing both the substrate and wire in an assembly, and which performs the following steps:
- Model change - deactivate substrate
- deform wire
- model change - reactivate substrate --> assembly relaxes
--> As you can see, there's a lot of steps here, and it is somewhat inconvenient to have to do so much pre-processing of the input files to write out the constraints which link nodal displacements together. Anytime wire parameters are varied, I have to run all the pre-processing again because the location of the deformed wire nodes changes, hence the constraint equations defining nearest substrate nodes change as well. I looked into embedded elements but these only seem to work if everything is embedded at the outset and stays that way (won't work because I need to deform the wire before bonding to substrate). Is there a better way to link the deformation of the wire and substrate, after the initial wire deformation?
* I would eventually like to "poke some holes" in the substrate by introducing another "model change" which eliminates additional regions of elements after the relaxation occurs. Is there a way I can define the region of material removed after deformation? For example, let's say I want to poke a circular hole. If I define a circular region on the undeformed substrate, by the time I remove the material after relaxation has taken place, that region is no longer circular. I want to define a circular region on the *deformed* substrate configuration, not the reference configuration. Is this possible?
I'm fairly new with Abaqus - my gut tells me that some of this could potentially be accomplished by using restart files, or maybe some intelligent scripting, but I'm not sure how exactly...
Anyhow, thanks for reading, and if anyone has any advice or insight, I'd greatly appreciate it!
I'm trying to set up a model for a device involving a pre-stressed wire embedded in a substrate. I have a model which is working OK but I'm wondering what if there's anything I could do to make things run a little smoother, so, any advice is greatly appreciated. Here are the main things I'm dealing with:
* The model consists of a wire (beam elements) which is first bent into a particular shape (hence, pre-stressed) and then implanted in a thin substrate (2d plane stress). The substrate-wire assembly is then allowed to relax. In order to get the wire and substrate to "bond" together, I'm first running the wire deformation problem. Separately, I create a mesh of the substrate. Then I have an external code which reads my wire/substrate files and finds the closest wire node for each substrate node, and writes constraint equations that enforce that these nodes move together after the wire is initially deformed. All of this goes into a model containing both the substrate and wire in an assembly, and which performs the following steps:
- Model change - deactivate substrate
- deform wire
- model change - reactivate substrate --> assembly relaxes
--> As you can see, there's a lot of steps here, and it is somewhat inconvenient to have to do so much pre-processing of the input files to write out the constraints which link nodal displacements together. Anytime wire parameters are varied, I have to run all the pre-processing again because the location of the deformed wire nodes changes, hence the constraint equations defining nearest substrate nodes change as well. I looked into embedded elements but these only seem to work if everything is embedded at the outset and stays that way (won't work because I need to deform the wire before bonding to substrate). Is there a better way to link the deformation of the wire and substrate, after the initial wire deformation?
* I would eventually like to "poke some holes" in the substrate by introducing another "model change" which eliminates additional regions of elements after the relaxation occurs. Is there a way I can define the region of material removed after deformation? For example, let's say I want to poke a circular hole. If I define a circular region on the undeformed substrate, by the time I remove the material after relaxation has taken place, that region is no longer circular. I want to define a circular region on the *deformed* substrate configuration, not the reference configuration. Is this possible?
I'm fairly new with Abaqus - my gut tells me that some of this could potentially be accomplished by using restart files, or maybe some intelligent scripting, but I'm not sure how exactly...
Anyhow, thanks for reading, and if anyone has any advice or insight, I'd greatly appreciate it!