Pushover of Wall

I am trying to "push" an unreinforced masonry wall. I am trying to get the base reaction vs. top displacement. The result of the analysis should coincide with the experimental results below: 

Pushover Experimental curve

I have tried to model the material using SOLID65, and material nonlinearity using William-Warnke concrete parameters. But, the model lost its convergence too early without even going into the nonlinear regime. I don't know what am I missing though. Another thing: I can't define nonlinearity using MISO. It displays warning that SOLID65 doesn't support MISO. So, I have simplified nonlinearity using BISO.

I have attached the input file for your reference. What I am trying to do in the input file is to give initial pressure at the top of wall and acceleration due to gravity in the first step of solution and acceleration in x-direction, as in a pushover in the second step. Then, I am getting the displacement and reaction forces in a text file. I am stuck in this for more than two weeks now, but can't figure out the solution. The model loses convergence too early. I can't get to the nonlinear regime as I expected.

Any help in this regard will be highly appreciated. Thanks in advance!

Sujit

Comments

  • LimitLimit Member
    edited February 2019

    Hello!

    Try these solver settings:

    SOLCONTROL,ON,OFF,NOPL,

    NROPT,unsym,,on

    neqit,200

    KBC,0

    cutcontrol,plslimit,0.2

    LNSRCH,OFF

    cutcon,noshape,1

     

    If the solution diverges, leave the control of convergence in force:

    cnvtol,f,,0.3

  • sujitbhandarisujitbhandari Member
    edited February 2019

    Thanks Limit,

    will try that. It's too late here in Nepal, and too cold. Will try first thing in the morning! Will let you know if it works.

  • sujitbhandarisujitbhandari Member
    edited February 2019

    Hi Limit,

    That worked, but not quite as I expected. Unsymmetric Newton-Raphson has done its job.

    However, I think I have to get my parameters in place before I can continue on.

    Thanks again for the help.

    Next question: Is there any way to introduce MISO in SOLID65. I don't know why has it stopped working in ANSYS 19.2?

  • AminAmin Member
    edited February 2019

    I think MISO command is working well with SOLID65 in Ansys 15.0. I am working in the same theme as you. From your attached file it seems that you use the William-Warnke by activating both the crushing and cracking parameters (fc, ft). I suggested to activate only the cracking capability and deactivate the crushing capability, and use one of the nonlinear models available in ANSYS (MISO,etc.) to define the non-linearity in compression 

  • sujitbhandarisujitbhandari Member
    edited February 2019

    Hi Amin,

    Will try that too! I can't get MISO to work with SOLID 65 in ANSYS 19.2. Maybe because I have activated crushing parameter. I will deactivate that and let you know what happens next. Cheers and thanks for the reply!

    Sujit

  • sujitbhandarisujitbhandari Member
    edited February 2019

    Hello again,

    I am trying to push a wall with three distinct layers with friction in between layers. I can't figure out the way to make this pushover work. It seems so simple, yet so difficult. I have attached the input file. Your help will be highly appreciated.

    I am stuck with another problem now! I am literally banging my head against the wall for this problem to get done!  

     

  • AminAmin Member
    edited February 2019
    Hi sujit.
    Since MISO command didn't work in 19.2. Try to use MKIN command to define the nonlinear behavior of brick units. It works well with me with deactivating the crush capability in concr command using solid65. This command is defind by only 5 points.
    Also, I would like to ask you how did you model the contact between the brick units?
  • AminAmin Member
    edited February 2019
    I also see that you used three different mathods to model the wall, one of these methods is the micro-plan. I would like to ask you, did this model capture the nonlinear part of the curve?. I used this technology for concrete but the curve obtained was linear
  • sujitbhandarisujitbhandari Member
    edited February 2019

    Hi Amin,

    Yes, I was able to get the nonlinear capacity of masonry using microplane model. The results seemed quite satisfactory. In fact, it did a good job. However, you have to use the model with damage-plasticity model. And, there are a lot of parameters to define, which are quite hard to obtain. Here's a thread you might find helpful.

    Entering nonlinearity in concrete

    Regards,

Sign In or Register to comment.