I have faced two problems with the dynamic time history:
1- The obtained displacements of the joints when using force based elements are not equal to those obtained using displacement based elements. Although I tried to divide each element to six elements.
2- I checked the axial force in the frame element, I found it is not equal to the base reaction when trying to model shear wall with frame elements.
Two difficulties with dynamic time history analysi
Re: Two difficulties with dynamic time history analysi
Hi amtarabia.
1. I don't think this should be a surprise. As mentioned in the Help System under contents topic "Theoretical Background ad Modeling" -> "Material Inelasticity", FB elements require no discretization to increase accuracy while DB elements do require discretisation. It is certainly possible that the degree of nonlinearity in your model is such that many more than 6 elements would be required to make the DB approach converge to the FB approach.
2. Could you clarify how you modeled the shear walls? As frame infills? If so, then don't forget that this element is an equivalent strut which would need to be included in the free-body diagram of base reaction, column force, and strut force. So I would say, make certain that you have included all necessary objects in summing your forces at the base as a check (which, by the way, is always a good idea).
Best of luck amtarabia.
1. I don't think this should be a surprise. As mentioned in the Help System under contents topic "Theoretical Background ad Modeling" -> "Material Inelasticity", FB elements require no discretization to increase accuracy while DB elements do require discretisation. It is certainly possible that the degree of nonlinearity in your model is such that many more than 6 elements would be required to make the DB approach converge to the FB approach.
2. Could you clarify how you modeled the shear walls? As frame infills? If so, then don't forget that this element is an equivalent strut which would need to be included in the free-body diagram of base reaction, column force, and strut force. So I would say, make certain that you have included all necessary objects in summing your forces at the base as a check (which, by the way, is always a good idea).
Best of luck amtarabia.
Tim Huff
-
seismosoft
- Posts: 1316
- Joined: 06 Jul 2007, 04:55
Re: Two difficulties with dynamic time history analysi
amtarabia,
1) as correctly noted by huffte, differences between force-based and displacement-based formulations are not necessarily surprising. Please refer to the publications indicated in the Help System/User Manual (section indicated by huffte) and/or to the discussions you may find in this Forum (e.g. carry out a search on "force-based displacement-based integration points")
2) carry out a search in this Forum on "support reaction"
Seismosoft Support
1) as correctly noted by huffte, differences between force-based and displacement-based formulations are not necessarily surprising. Please refer to the publications indicated in the Help System/User Manual (section indicated by huffte) and/or to the discussions you may find in this Forum (e.g. carry out a search on "force-based displacement-based integration points")
2) carry out a search in this Forum on "support reaction"
Seismosoft Support