stiffness multiplier for rotational spring

[esi_opensees]

Hello,

Changing the value of stiffness multiplier for rotational spring (n), used in modeling concentrated plasticity, affects significantly to results of nonlinear time history analysis. In the OpenSees Example:
http://opensees.berkeley.edu/wiki/index ... ment_Frame

"n" has been selected as 10.0. but why 10? is it for this especial problem? How can we determine an appropriate value for any problem?

Thanks.

[kamehdikhani]

Hello Dear esi
i think your answer is in this paper ::::
Ibarra LF, Krawinkler H. Global collapse of frame structures under seismic excitations. Report no. 152. John A. Blume Earthquake Engineering Center. Department of Civil and Environmental Engineering. Stanford University.2005
i think it is better to read this part : Appendix B :properties of Generic Frame Models
specially page 299
i hope it helps u

[shjelvan]

hello Mr mehdikhani
would you mind helping me to find this book
i cant find it any sources

[kamehdikhani]

hello please leave me your mail adress so I can send u the paper . my Mail is : k.mehdikhani@gmail.com
please send your phone number to my mail so I can call u.
tnx

[fmk]

https://blume.stanford.edu/blume-tech-reports

[andy0071]

In this book, there are some explanations of how to calculate the modified parameters for plastic hinge.
However, there is no explanation that why n is selected as 10.
In this book, it mentioned that "a value of n = 10 was used throughout this study." on P 299.

I do a test by using OpenSees wiki example. Define n= 10, 100 and 1000 respectively, and check the base shear vs. roof displacement relationship.
http://opensees.berkeley.edu/wiki/index ... ment_Frame

the base shear force vs. top displacement relations curves of three different n are overlapping, i.e. n does not affect the nonlinearity of the structure.

My conclusion is as long as n >>1, the result has little difference.
Because no matter the value of n, the member stiffness of member can always be separated into stiffness of elastic beam column element and that of plastic hinge. The difference is the proportion of each component, however, the total effects are identical. As long as n>>1.
The only reason to use n is to avoid the damping concentrate at the end of elements, moment gradient and the numerical instability. it will not dramatical affect the final result.

As the time limit, only the base shear vs. roof displacement of static pushover analysis is checked here. You may want to check the cyclic deterioration, ductility capacity or the damping properties to verify my conclusion.