flowglow
Mechanical
- Apr 30, 2014
- 14
Hi, everyone
Currently I am working with rubber, trying to get characterize its behaviour during compression and implement this information to FEM software (ANSYS in my case). So, I made a series of compression tests with[/sup] cylindrical samples and two types of boundary conditions: one series of samples were glued to compression plates, another one were lubricated to decrease a barreling effect caused by friction between sample and compression plates. I know that for better fitting it is very good to use different modes of deformation, but at the moment I don't have facilities for making another types of tests, and, moreover, it maybe not necessary, because in application the rubber part works only in compression.So, after getting results of experiments I went to ANSYS, to make a curve fitting. So, here is my first questions is about fitting:
--- In the ANSYS help it is clearly said that I have to use Nominal (Engineering) stress and engineering strain for input data. It is quiet strange for me, because cross-section of rubber sample changes hardly during the test. W[/pre]as I right, when I used negative data for fitting (my test is compression)?
Fitting process was done, and I made model of my compression test to check how my fitted model works. I've decided to take a quarter of cylinder cross-section (if you watch from the side) and make a 2d axisymmetric modeling, so I took solid PLANE182 element for body, and CONTA175 and TARGE169 for contact and target surfaces respectively. For additional information I used tutorial VM201 from "ANSYS Mechanical APDL Verification Manual". So, my second question, the answer on which isn't clear for me:
--- How to choose the stiffness of surface-to-surface contact and how does it affect on further modelling? Is it necessary to use contact stiffness if I use a rigid target (for making target I made two nodes, and after created an element on these two nodes: TSHAP, LINE /n E,NODE1,NODE2).
Ok, I made my model and solved it. Loading conditions were set in displacement mode: 5mm compression displacement to 10mm in high and 25 mm in diameter cylindrical sample. There were two types of boundary conditions as in experiment: free or fixed bottom surface of the cylinder. In the end I've got very strange stress field: compression stress in X direction. My last question: there is mistake can be? I attached some pictures of fitting,
Currently I am working with rubber, trying to get characterize its behaviour during compression and implement this information to FEM software (ANSYS in my case). So, I made a series of compression tests with[/sup] cylindrical samples and two types of boundary conditions: one series of samples were glued to compression plates, another one were lubricated to decrease a barreling effect caused by friction between sample and compression plates. I know that for better fitting it is very good to use different modes of deformation, but at the moment I don't have facilities for making another types of tests, and, moreover, it maybe not necessary, because in application the rubber part works only in compression.So, after getting results of experiments I went to ANSYS, to make a curve fitting. So, here is my first questions is about fitting:
--- In the ANSYS help it is clearly said that I have to use Nominal (Engineering) stress and engineering strain for input data. It is quiet strange for me, because cross-section of rubber sample changes hardly during the test. W[/pre]as I right, when I used negative data for fitting (my test is compression)?
Fitting process was done, and I made model of my compression test to check how my fitted model works. I've decided to take a quarter of cylinder cross-section (if you watch from the side) and make a 2d axisymmetric modeling, so I took solid PLANE182 element for body, and CONTA175 and TARGE169 for contact and target surfaces respectively. For additional information I used tutorial VM201 from "ANSYS Mechanical APDL Verification Manual". So, my second question, the answer on which isn't clear for me:
--- How to choose the stiffness of surface-to-surface contact and how does it affect on further modelling? Is it necessary to use contact stiffness if I use a rigid target (for making target I made two nodes, and after created an element on these two nodes: TSHAP, LINE /n E,NODE1,NODE2).
Ok, I made my model and solved it. Loading conditions were set in displacement mode: 5mm compression displacement to 10mm in high and 25 mm in diameter cylindrical sample. There were two types of boundary conditions as in experiment: free or fixed bottom surface of the cylinder. In the end I've got very strange stress field: compression stress in X direction. My last question: there is mistake can be? I attached some pictures of fitting,