Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
about dynamic stability 1
- Thread starter ENGLEOSUN
- Start date
- Thread starter
- #4
I'm sorry not to express my question clearly,what i'm really concerned about is the dynamic response of skeletal structures(such as space truss,latticed shell)subject to seismic loads.
The most accurate and common method for seismic analysis is time-history method,and almost every large FEA software is capable of doing this. During the time history of response,i think one or more members of the structure maybe fail due to dynamic instability,can every software simulate the dynamic instability phenomenon accurately or just neglect it?
The most accurate and common method for seismic analysis is time-history method,and almost every large FEA software is capable of doing this. During the time history of response,i think one or more members of the structure maybe fail due to dynamic instability,can every software simulate the dynamic instability phenomenon accurately or just neglect it?
-
1
- #5
Engleosun:
There are two types of instability: structural instability and dynamic instability. The first is usually the condition when a member of a structure bifurcates ( a sudden change in the position to a lower energy state). The load can either be static or dynamic. The second is a more sharply focused on the dynamic energy levels of a structure (Motion of a structure is suddenly changed to another response. If no damping is present, unlimited growth can result).
Dynamic stability is best described using Chaos or other comparable theories. The classical dynamic problem is Mathieu instability. Most studies have been directed towards the response of shells of revolution. Unless the analysis program measures the dynamic response to a specific theory, it is difficult to determine that the observed response from the computer output can establish dynamic instability.
There are very mathematical conditions that can be expressed to determine dynamic instability-- Thompson and Hunt, "A General Theory of Elastic Stability", 1973. Others use the term "Catastrophe Theory" or "Chaos Theory."
Unless the FEM computer program allows geometric non-linear response and the introduction of imperfection geometry, one can not establish dynamic instability. On the other hand, if they do not have a criterion to determine dynamic instability, dynamic instability can not be determined.
The observed numerical response can have a condition known as numerical instability. This condition results from inaccurate time step size for the dynamic terms used in the acceleration algorithm being used. This can be detected by observing the response of one state being the negative of the next adjacent state. Things can be confusing when talking about dynamic instability.
There are two types of instability: structural instability and dynamic instability. The first is usually the condition when a member of a structure bifurcates ( a sudden change in the position to a lower energy state). The load can either be static or dynamic. The second is a more sharply focused on the dynamic energy levels of a structure (Motion of a structure is suddenly changed to another response. If no damping is present, unlimited growth can result).
Dynamic stability is best described using Chaos or other comparable theories. The classical dynamic problem is Mathieu instability. Most studies have been directed towards the response of shells of revolution. Unless the analysis program measures the dynamic response to a specific theory, it is difficult to determine that the observed response from the computer output can establish dynamic instability.
There are very mathematical conditions that can be expressed to determine dynamic instability-- Thompson and Hunt, "A General Theory of Elastic Stability", 1973. Others use the term "Catastrophe Theory" or "Chaos Theory."
Unless the FEM computer program allows geometric non-linear response and the introduction of imperfection geometry, one can not establish dynamic instability. On the other hand, if they do not have a criterion to determine dynamic instability, dynamic instability can not be determined.
The observed numerical response can have a condition known as numerical instability. This condition results from inaccurate time step size for the dynamic terms used in the acceleration algorithm being used. This can be detected by observing the response of one state being the negative of the next adjacent state. Things can be confusing when talking about dynamic instability.
MaxRatio
Mechanical
- Jul 23, 2008
- 2
Englesun,
It seems this problem may be solved using a Dyna3D approach (explict FE formulations) and nonlinear contact. When the strut breaks (loses contact) the program may still continue depending on certain rigid body factors that might be controlled through contact modeling.
Check out and their product called VPG (Virtual Prototype) uses a Dyna3D base and offers a good GUI. They also can assist with support since they are a Dyna VAR.
Best/
Max
It seems this problem may be solved using a Dyna3D approach (explict FE formulations) and nonlinear contact. When the strut breaks (loses contact) the program may still continue depending on certain rigid body factors that might be controlled through contact modeling.
Check out and their product called VPG (Virtual Prototype) uses a Dyna3D base and offers a good GUI. They also can assist with support since they are a Dyna VAR.
Best/
Max
Transient1
Mechanical
I know ANSYS can handle the dynamic response of a structure under the influence of a seismic load. This type of analysis is called a spectrum analysis in ANSYS. I've not used the feature extensively, but it can also be used to determine structural instabilities (buckling).
mtnengr, great post.
mtnengr, great post.
Hi,
afaik, in order to determine non-linear buckling collapse load (critical load), you have to run a transient analysis where load is proportional to time (if inertial effects are important), or a static multi-step analysis with loads ramping from zero to a maximum (if you can neglect inertial effects). The last converged substep will give the critical load.
Regards
afaik, in order to determine non-linear buckling collapse load (critical load), you have to run a transient analysis where load is proportional to time (if inertial effects are important), or a static multi-step analysis with loads ramping from zero to a maximum (if you can neglect inertial effects). The last converged substep will give the critical load.
Regards
- Status
Similar threads
- Question
- Locked
- Question
- Locked
- Question
- Locked
- Question