infrmFBPH

[Doudine]

Hello,
I just noticed that in a model with infrmFBPH elements, the free vibration results depend on the length of the plastic hinge?
I'm intrigued !!!!!
can there be an explanation ?

[seismosoft]

Indeed, this is the case. The length of the plastic hinge affects the behavior of the frame element, due to kinematic reasons. It is strongly advised to keep the default plastic hinge length, unless you have a good reason to change it.
Seismosoft Support

[Doudine]

To the best of my knowledge, the length of the plastic hinge is only defined to link the curvature to the rotation, and the free vibrations are calculated with the elastic proriety ?

In this case, it would not be possible to calibrate the length of the plastic hinge (the extent of inelasticity), to remedy localization problems, as discussed in Calabrese et al ?

[seismosoft]

Thanks for the suggestion.
Indeed the tweak to fix the localization problems has been is already included in our to-do list and it will be programmed in a future release of the program.
Thanks
Seismosoft Support

[Doudine]

It is I who thank you

[Doudine]

Taken from:
https://wiki.csiamerica.com/display /kb/Elastic+softening+due+to+fiber+hinges+in+older+versions+of+SAP2000%2C+CSiBridge%2C+and+ETABS

Is it possible to implement fiber hinges without introducing softening?

Extended Question: Fiber hinges seem to introduce artificial flexibility to the model, though hinges are a plastic mechanism which should not influence elastic behavior. Is it possible to implement fiber hinges without introducing softening?

Answer: This softening behavior may be attributed to calculation of the fiber-hinge response curve, which proceeds directly from the axial response of individual section fibers. During linear response, fiber elasticity is added to element elasticity, causing softening. To avoid elastic softening, users should divide frame members at hinge locations such that a smaller segment is positioned along the hinge length. A large area property modifier should then be assigned to small sections, using the Assign > Frame > Property Modifiers menu.

[Doudine]

Below is a detailed description of how the program determines the flexibility of the element with hinge:

The issue here is that the flexibility of the fiber hinge is added to the elastic flexibility of the frame member containing the hinge. It may be better viewed as the modulus of elasticity being modified rather than the moment of inertia. Over the length of the hinge, the effective modulus Eeff can be computed from 1/Eeff = 1/Ef + 1/Eh, where Ef is the modulus of the frame element, and Eh is the modulus of the hinge. For a fully-steel hinge at zero stress, Eh = Ef, so that Eeff = Ef/2 = Eh/2. For reinforced concrete, the averaged value of Eh may be different from Ef.

By setting the axial and bending section property modifiers to a large value over the length of the hinge, The Ef becomes very large and Eff = Eh, or nearly so. This is the recommended approach.

[Doudine]

The value of Eh used for modal analysis, or any linear analysis, depends upon whether the linear load case starts from zero initial conditions or uses the stiffness from a previous nonlinear case. From zero, the initial elastic stiffness of each fiber is used. If the fiber has different initial tension and compression stiffness, such as for concrete, the average stiffness is used. From a previous nonlinear load case, the actual stiffness E of each fiber at the end of the load case is used. The approach most often appropriate is to run a nonlinear gravity load case and use the stiffness at the end of that load case to compute the modes.

[seismosoft]

Hello Doudine and thanks for the information.
As I explained above, fixing such localization problems is in our to-do list and it will be programmed in a future release of the program.
Seismosoft Support

[Doudine]

Thank you very much for your dedication