Slip or permanent displacement during Earthquake Excitation

[IRSHADQUR]

Hello everybody

I am doing dynamic analysis of a rocking wall. In the first step i applied the self weight and post-tensioning and base of the foundation was fixed in all DOF. In the 2nd step i have to apply dynamic loading. So i free the horizontal DOF of foundation base (EQ will be applied in Horizontal DOF) and then acceleration is applied at the wall base in horizontal direction. After running analysis when i check the Horizontal displacement of foundation it shows that there is some slip occured during the dynamic loading. Actually this displacement should be like a sine wave along the zero displacement line but it shows permanent displacement of whole structure (Like slip). At the end of the analysis structure has 250mm of permanent lateral displacement in the lateral direction. You guys have any idea abt this problem? am i applying EQ in the right way?

Regards
Irshad

 

[bkal]

Is your model of the EQ excitation fully linear elastic, or is there any non-linearity, friction or contact? Have you checked if your input motion (acceleration) exhibits any displacement drift?

 

[IRSHADQUR]

Dear Bkal
Contact is established between the wall and the foundation with both normal and tangential behavior. Behavior of the structure is non-linear. There is a chance of slip movement between wall and foundation but it was observed to be zero rather foundation was slipping. For more information kindly take a look at this video.

http://www.youtube.com/watch?v=Xc0zeI2Qtio&feature=youtu.be

Thx for ur reply
Regards
Irshad

 

[bkal]

After watching you video, to me it looks as if you have released horizontsal degrees of freedom of the foundation, and than applied input motion to the base of the wall. If there is any friction between the base of the wall and the foundation, then the friction force will move the foundation, which has no resistance to horizontal motion.

I think that what you want to do is as follows:
- have a proper contact model between the base of the wall and the foundation, including friction.
- apply input motion to the foundation, not the wall.

 

[IRSHADQUR]

I released the horizontal DOF of foundation but i applied the input motion (acceleration) at the base of the foundation (Not the wall). Any idea how to avoid this slip?

 

[bkal]

If that is the case, than your input acceleration has an inherent drift - have you tried to integrate it to get displacement of the input motion.

 

[IRSHADQUR]

yes i get the Displacement time history using a software SeismoSignal. Is it wise to give displacemnet history as an input excitation because this displacement history is not looking very good. Its just like a sine wave. Can u guide me abt this inherent drift thing.
Thx alot

 

[bkal]

I was just trying to say that you need to check what is a displacement at the end of your time history. Could you post a plot of your input acceleration and displacement time histories (in separate graphs). The reason is that it seems that at the end of your acceleration time history, the point to which it is applied is not returned back to its original position, but has drifted away.

 

[IRSHADQUR]

i think i am getting ur point but if u can elaborate more it will be helpful. Following are the links for the histories.

http://www.flickr.com/photos/93814781@N03/8719316101/in/photostream

http://www.flickr.com/photos/93814781@N03/8720439184/in/photostream

First one is the acc, velocity and displacement history of excitation.
Second one is the top displacement history of rocking wall.

 

[bkal]

It seems that your input motion shows about 20mm of permanent drift (displacement at the end of your time history, which is quite short), and the wall 40mm - I am not clear which point on the wall is this for. If it is for the base of the wall, than there is some slippage between the wall and the foundation - you could test it by increasing a coefficint of friction. If your input motion should not have any permanent drift, you either need to generate a new motion or do a base correction on the exsiting one.I assume that you are using a consistent system of units in your model, as I cannot see where 250mm from your original post came from.

 

[IRSHADQUR]

Actually that 250 mm was when i run the analysis for first time but later on i changed the damping so it reduced to present value.
This result is for a node at the top of the wall.
Actually i got this time history from the results of acceleromete in the shake table test of this wall. May be i need to do some kind of base correction for my excitation. Because this one is just raw result from accelerometer. What do u suggest?