varying support conditions?

[rick3186]

Hello,

Does anyone know if it is possible to change the support conditions of a structure part way through an analysis.

For example: say I were to model a bridge and wanted to make the support conditions at the cap pinned for the dead load case. Then I want the analysis to begin a cyclic load case using the final conditions from the dead load case. However, for the cyclic case, I want the support conditions at the cap to behave in a fixed manner.

Any suggestions (or even a confirmation that this is/is not possible) would be greatly appreciated.

Thanks

rick3186

 

[philipckwusa]

Hi,
I am trying to input the P-y curve from LPile (nonlinear soil spring)to SAP 2000.
I used Link/Support Properties and input the value in the Force (lbs) and Displ (in) column in SAP. However, the result is not correct. Units of P-y curve from Liple are lb/in and in, and I convert the unit to lb and in but it is still not working.
Does anyone know how to do it? I DO NOT want the equivalent linear spings.

Thanks!

 

[rick3186]

Philipckwusa,

Can you be more specific as to what is not working correctly within your model. Are you sure you defined the properties/direction of your spring properly? Also, did you use a one-joint spring or a two-joint spring?

Any ideas on my original thread question?

rick3186

 

[masomenos]

rick3186, you should be able to do this through stage construction. Go to the main menu, and under the Define menu, check out Named property sets if have Version 12. When you open the dialogue for Named property sets -> frame releases, press F1 to read the writeup. Even if your office doesn't have access to the stage construction version of SAP, the program lets you run 1 or 2 stage construction stages for "free" without any special license, so you can run a simple test.

philip, if the P-y curve is nonlinear, then make sure that you assigned your force vs deformation to a multilinear plastic type link and also make sure that your load cases, which used to be called 'analysis cases' in version 11, are set to run nonlinear analysis instead of the default linear static. Anytime you want to consider a nonlinear link or nonlinear tension only frame assignment or nonlinear p-delta, you need to set up your load cases to be nonlinear too

 

[philipckwusa]

masomenos,
I have assigned two different ways for force vs deformation in SAP. First, from Lpile, the unit is Force (lb/in) and deformation (in) and I just copy and paste those value to SAP 2000 under force and displacment column. I then compare the result of deflection for SAp and Liple with the same lateral load P. The deflection are different.

Second, I multiply the length of the depth to the force (lb/in) from Liple into forcae (lb) and keeping the displacement (in) and input them into SAP 2000 under force and displacment column. I then compare the result of deflection for SAp and Liple with the same lateral load P. The deflection are different.

Until I got the simial results,; otherwise I cannot perform nonlinear push over analysis to locate the plastic hinge at pile top and in ground.

If you have an example files (LPile nonlinear springs and SAp files) for handling nonlinear springs, please let me know.
Thanks!

 

[rick3186]

philipckwusa,

I do not mean to impose on your conversation with masomenos, however the following link is a good tutorial showing how to perform a nonlinear static pushover analysis to determine hinge locations, etc.

http://www.csiberkeley.com/Watch&Learn/Pushover_V11S.wmv

(csiberkeley.com has a watch and learn series with a lot of other valuable tutorial videos). It is also important to note that the video uses hinges rather than link elements.

It sounds like your second option is the best way to define your link element. Along the lines of masomenos, make sure that you defined your link as a nonlinear element in the correct local degree of freedom. You will also need to make sure your load case is defined as nonlinear.

Finally, how many elements are you using to model your pile? The number of elements used should influence your interpolation functions, which might increase or decrease the accuracy of your results.

Hope some of these ideas might help you.

rick3186

 

[masomenos]

Presumably the LPile nonlinear spring is force vs deformation, so you need to specify whether it's force vs deformation at a joint, or on a per lineal unit or area basis, and you need to make sure your units are consistent between what was calculated with LPile and what you modeled in SAP. If you still have different results, send your model to CSI technical support and ask them to look at it.