5 Replies Latest reply on Mar 29, 2011 3:06 PM by Bill McEachern

    Poisson's Ratio: in or out?

    Kevin De Smet

      Considering the detrimental effect Poisson's Ratio can have on fixed restraints, and the relatively low impact it seems to have in situations where it does model reality appropriately under correct boundary conditions. Is it acceptable to just enter a value of 0.0 and forget about Poisson and his little ratio?

       

      Or does it really matter and should you just be diligent about it?

       

      Poisson's Ratio... in or out?

        • Re: Poisson's Ratio: in or out?
          Bill McEachern

          If it were me I would just find a more realistic restraint system that produces a closer to reality approximation or ignore the high stress at the restraint - if thisis the area of interest thent he former is the way to go and it may involve moving the restraint further away from the area of interest - ie add more to the model so that a better approximation at the area of interest is obtained. Leaving out Poisson's ration is a bad idea in my humble opinion.

          • Re: Poisson's Ratio: in or out?
            Paul Kellner

            It is absolutely necessary, especially in 3D problems. Poisson's ratio is a derived quantity that helps express the nature of the material. Just because it is a little number and frequently overlooked in material testing, don't think it is unnecessary. Poisson't ratio in most elastic materials ranges from .1 to .4. Materials that are incompressible like rubber use .5 which requires special treatment in FEA.

              • Re: Poisson's Ratio: in or out?
                Kevin De Smet

                Definitely for the more exotic materials close to 0 and close to 0.5, poisson's ratio is a must. I'm thinking more along traditional analyses on steels and aluminum. It seems like it has much less of an effect?

                  • Re: Poisson's Ratio: in or out?
                    Paul Kellner

                    Kevin,

                     

                    It still has an effect. An important effect. How would you calculate the correct stress in a plane strain situation?

                     

                    E-nu.png

                    As you can see for a simple uniaxial test bar this all reduces to:

                     

                    e11 = sigma11/E

                     

                    But as soon as you add stresses in other directions e11 gets more complicated. If for example sigma22 and sigma33 equal sigma 11 then e11 would be 40% of that calculated if you disregarded a poisson's ratio of .3.

                     

                    And if you have a loading condition that produces pure shear your answer would be off by 1/(1+nu) or 23%.

                     

                    Assuming you are using FEA because the problem you are solving is more complicated than a simple tensile test bar you really want to include a correct poisson's ratio unless you can live with 20-40% error by setting it to zero. And to make it clear, in the first example above, leaving poisson out would underpredict stress potentially making an unsafe design.

                     

                    Message was edited by: Paul Kellner  Added detail

                     

                    Message was edited by: Paul Kellner Clarified meaning of equation.

                    • Re: Poisson's Ratio: in or out?
                      Bill McEachern

                      When you pull on a metal bar it contracts in the direction normal to the pull - this is the poisson effect. Rubbers are constant volume materials, the volume gain in the elogation is exactly lost by the contractions in the other two. And, yeah most metals are around 0.3 but that doesn't mean you can leave it out and get reliable results.