1 Reply Latest reply on Jan 7, 2008 7:09 PM by Vince Adams

    CosmosWorks Professional Non-Linear problem.

    Henry Belch
      I have set up a non-linear analysis to simulate a twisting action on one of our srpings. We took an existing spring and applied 130 ft-lbs of torque (it rotated about 90 degrees) When I set up a non-linear anaysis in cosmos and applied the same torque to a model of the part, it only rotated about 1/3 of the angle...

      Can you apply an enforced displacement?

      Why are the results so far off... (the scale is set to 1.0)?
        • CosmosWorks Professional Non-Linear problem.
          Vince Adams
          Without more info, here are a couple of questions to consider:

          1) Did you allow the spring to compress/extend during twisting? Not allowing this would stiffen it.

          2) Are you confident in the properties? I've found that drawn wire for springs may not exhibit the same elastic properties as the base material. It was quite a shock the first time this came up in a project. Tensile testing on your spring wire is the only way to check this.

          3) Are you confident in your restraints on the fixed end? While it probably wouldn't account for all of your rotation, any slippage in the test fixture restraint could exxagerate the empirical measurements.

          After these, I'd need to see some pix of the test set up and your model.

          Good luck!!