SouzaMech
Mechanical
- Sep 24, 2019
- 2
Dear all,
My question is about using Abaqus to predict accurate forces within a vibrating structure. I am carrying out a frequency domain vibration analysis on a tensile test, for my PhD project. Currently I am using the Steady-state Dynamics, Direct Solution procedure as this appeared the best compromise between coping with my boundary conditions and solving time. I wish to know the forces within the structure (and not necessarily the stresses) as I want to compare to a measurement made by a force transducer.
The Abaqus post-processor has a Free-Body Cut tool which, when used with the View Cut manager, can find the free-body force at any planar section. (I had to request NFORC in the Field Outputs to ensure it had the data to work with, so I suspect all it is doing is summing the nodal forces over the nodes nearest the defined plane.)
However, when I use this Free-Body Cut tool a warning pops up telling me:
"Free body computation is based on complex magnitude or phase angle in a steady state dynamics analysis. The result is not physically meaningful."
Which obviously is not a good warning to get when I want to use the simulation to verify the force transducer output (and whether this is a sound thing to do is another question for another day).
I find this puzzling as the kinematic parameters (displacement, velocity, acceleration) are presented as real, physically meaningful values, and the dynamic force experienced by the structure is directly related to the acceleration and the boundary conditions. So why would the force amplitude be any less meaningful? (and of course it is an amplitude, as the analysis is in the frequency domain.)
I should say that I have accurate velocity measurements so I can validate the simulated velocities against those.
So, I guess my questions are:
1. Why can't I take the force (amplitude) values from a Steady-state dynamics analysis as meaningful?; and
2. Should I be using a different Abaqus procedure, and which one? (Should I move to the time domain and Standard Implicit?)
Sorry in advance if this is a trivial question, I just haven't managed to get a straight answer from my colleagues or anyone else I've asked so far. Even though I am surrounded by experts, they all work in slightly different fields (eg exclusively time domain) or have never had to worry about my particlar force problem.
Thanks in advance,
Colin
My question is about using Abaqus to predict accurate forces within a vibrating structure. I am carrying out a frequency domain vibration analysis on a tensile test, for my PhD project. Currently I am using the Steady-state Dynamics, Direct Solution procedure as this appeared the best compromise between coping with my boundary conditions and solving time. I wish to know the forces within the structure (and not necessarily the stresses) as I want to compare to a measurement made by a force transducer.
The Abaqus post-processor has a Free-Body Cut tool which, when used with the View Cut manager, can find the free-body force at any planar section. (I had to request NFORC in the Field Outputs to ensure it had the data to work with, so I suspect all it is doing is summing the nodal forces over the nodes nearest the defined plane.)
However, when I use this Free-Body Cut tool a warning pops up telling me:
"Free body computation is based on complex magnitude or phase angle in a steady state dynamics analysis. The result is not physically meaningful."
Which obviously is not a good warning to get when I want to use the simulation to verify the force transducer output (and whether this is a sound thing to do is another question for another day).
I find this puzzling as the kinematic parameters (displacement, velocity, acceleration) are presented as real, physically meaningful values, and the dynamic force experienced by the structure is directly related to the acceleration and the boundary conditions. So why would the force amplitude be any less meaningful? (and of course it is an amplitude, as the analysis is in the frequency domain.)
I should say that I have accurate velocity measurements so I can validate the simulated velocities against those.
So, I guess my questions are:
1. Why can't I take the force (amplitude) values from a Steady-state dynamics analysis as meaningful?; and
2. Should I be using a different Abaqus procedure, and which one? (Should I move to the time domain and Standard Implicit?)
Sorry in advance if this is a trivial question, I just haven't managed to get a straight answer from my colleagues or anyone else I've asked so far. Even though I am surrounded by experts, they all work in slightly different fields (eg exclusively time domain) or have never had to worry about my particlar force problem.
Thanks in advance,
Colin