4 Replies Latest reply on Jun 15, 2012 10:46 AM by Joe Pesci

    Rubber simulation

    Joe Pesci

      I need to do a study involving rubber sealants. The catch here is that the seal is rotated by an external component, that is why we need to do a study on how the rubber reacts to the movement. If Solidworks is capable of estimating the movement and buckling it would be more cost effective than 3D printing a couple of rubber models.

       

      I'm familiar with FEA's on rigid constructions but the rubber simulation is a new area. My question therefore is some pointers or help on how to start with the simulation set-up. The attached file shows the front view with the needed rotation. The bottom cylinder rotates to the point on the right (this is just the first step of the total movement).

       

      What I started with was a non-linear simulation (as some other videos on rubber simulations), then fixed the upper part (which is going to be attached to a metal component). I followed with putting a remote load on the bottom part (which is moved by another component of this system), meshed it and ran the analysis, but it gave an error after the first step.

       

      If you have any questions or need more information please ask.

        • Re: Rubber simulation
          Jerry Steiger

          Joe,

           

          If I understand what you are trying to simulate, this is going to be tough to do, because you have such large deformations.

           

          First, do I understand? I believe that the top face of the upper cylindrical part of your seal is attached to one part and the bottom face of the lower cylindrical part is attached to another. The lower part rotates to the position shown by the line on the right.

           

          If that is correct, then you just need to fix the top face, as you did, and apply a deformation to the bottom face. Simple...but, you are going to have a lot of deformation and the part will probably buckle. You will have severely distorted elements and, in the case of buckling, probably self-contact.

           

          I'm not familiar with how SolidWorks Simulation handles this type of problem, but it can be a real pain in ANSYS, the FEA that I have more experience with.

           

          Jerry Steiger

            • Re: Rubber simulation
              Joe Pesci

              "[...] have a lot of deformation [...]" and this is just the first half of the movement. The second half is rotating it 90 degrees, followed by previous mentioned movement.

               

              We're especially curious in the buckling of the model (due to its movement of the blue highlighted part).Sealingpath.jpg

              Our design intent is developing a sealshape that can twist and bend and at same time doesn't undergo too much stress.

            • Re: Rubber simulation
              Bill McEachern

              Success - not sure but maybe. Suggested steps:

              1. Get the material characterized - you need good materail data to get this stuff to work - likely a mooney rivlin model.
              2. Might want to use shell elements for this.
              3. you will probably need the arc length method to get through the buckling point - if this is indeed the case then any option for contact self or otherwise is out as contact and the arc length method is not supported. It might go with force control if the stiffness doesn't change really fast which it might not.
              4. SWX Sim doesn't support self contact so if you get far enough and self contact happens - you will be able to see where and then you can split the body into two and you can set contact between those, if force control is used. If this gets all complicated as the thing deforms with more and more bodies being required then the fun drops out of this approach real quick.
              5. SWX doesn't support in solution remeshing so if the distrotions get really severe you can try putting in smaller elements but this will loose its appeal quickly as well.
              6. Not the programs best game but with good material data you can probably get further than you might think.