Non-linear modelling

[gswarup]

Hi,

I am doing a non linear time history analysis of 2D RC frame building in SAP2000, I've prepared a nonlinear model with NL-Link and analysed for an arbitrary sine function and got some result. However I am confused whether my modelling is correct or not. I am attaching the SAP model. Kindly anyone can guide me about my modelling?

Thanks in advance.....

 

[kostast88]

If what you try to model is rotational springs, I think that Rotation-z, is R3 and not R2, since R2 is connected with M2 which is minor bending moment (and since your problem is 2D, your springs should respond to M3). Not 200% sure though.
Why go with NLlinks and not with hinges?

 

[gswarup]

Thanks kostast88.....I'll go through it as per your suggestion and let you know the results.
Thaks a lot for your response.....

 

[johnbridge231]

kosta, you are right.

Analysis and Design of arbitrary cross sections
Reinforcement design to all major codes
Moment Curvature analysis

http://www.engissol.com/cross-section-analysis-design.html

 

[gswarup]

Hi Kosta....

I obtained good result using AutoM3 hinges, however I have some doubts regarding that

1. How moment-rotation curve is generated in auto hinges (FEMA356 beam) ? is it calculated from section propery automatically?
2. how can I ignore strength degradation part (i.e. beyond point C)? (whenever deformation reaches beyond point C it shows a peculiar force deformation curve.)
3. what type of hysteresis is assumed in auto hinges (in user defined we have options for Isotropic, Kinematic, Takeda or Pivot)?

Please answer me so that i can advance further in my analysis....

Thanks a lot

 

[kostast88]

1. Moment rotation points in the backbone curve are defined as the title implies by the FEMA 356 guidelines (that is θy=(My/EI)*Lp for beams,etc). Refer to the document for further information.
2. This part of the curve cannot be ignored with the auto-hinges.The FEMA 356 curve is very specific and is integrated automatically in the program.If you want to avoid it (I suppose all you want is a bilinear model) then create user defined hinges, where it is you who define all the points in the curve.All the points should be calculated with FEMA relations(or whatever you want),so when you come to define M-θ at point C, just extend the A-B part, so that it doesnt fall down.Same goes for D.
3. I suppose it is isotropic hardening, since it is default hysteretic behaviour for every material in sap.

A comment:
For columns you should use P-M (or PMM for biaxial)and not AutoM3 hinge, to account for axial-flexural interaction, since you're doing an earthquake time history analysis, and you have no idea what kind of axial loads the earthquake will produce.P-M diagram is genereated either automatically,or user defined,it is an option at hinges definition.


You should check the reference as well:
Link

 

[gswarup]

Many many thanks Kosta......I'll check it with user defined hinge......