4 Replies Latest reply on Mar 20, 2011 3:45 PM by Dennis Bowman

    Modeling Water - not flow related

    Dennis Bowman

      I have a tank full of water. How do i model the water for a frequency analysis? The water has no tensile properties. I am not interested in stress values in the water just the mass effects on the natural frequencies. As a first guess I simply increased the density of the metal until the weight of the unfilled vessel matched the weight of the water filled vessel but it really does not match the weight distribution of a water filled vessel (it is not a simple cylinder).

       

      Thanks in advance, Dennis

          • Re: Modeling Water - not flow related
            Dennis Bowman

            Well close. I think I can model the tensile strength as the vapor pressure... cavitation. Shear strength will have to be some function of viscosity... a tiny number in any case.

             

            Thanks Tony, I appreciate the quick response.

             

            Dennis

              • Re: Modeling Water - not flow related
                Jerry Steiger

                Dennis,

                 

                How did your analysis work out?

                 

                I assume that your full tank means that there is no free water surface, as it seems like a free surface would really make the analysis difficult. If that is the case, then you can't use simple mass loading due to the water, since one area moving in pushes the water over to cause the rest of the areas to move out.

                 

                I figure that the Modulus of Elasticity for water must mean the bulk modulus. The value seems low, but that probably just means that I don't have a good feeling for how compressible water is.

                 

                Jerry Steiger

                  • Re: Modeling Water - not flow related
                    Dennis Bowman

                    Hi Jerry,

                     

                    Thanks for the interest. Fortunately it is a closed vessel with no free surface. In the end I just increased the density of the metal to get close. This type of analysis is just is not exact anyway because of all the complicated bearings supporting a pump rotor inside a vertical pump column. I sent the recommendations for modification based on the simplification to our shop and they implemented them and installed the unit and all seems to be well... at least the vibration amplitude is about 5% of what it was before the modification.

                     

                    I usually use a different CFD tool for real flow simulation... CFDesign by Blue Ridge Numerics. It is pretty good at rotating regions like impellers or fully submerged objects but does not handle open channel flow. I have read about several open channel flow software solutions but I rarely get involved with them (a couple canal projects for the Corps of Engrs in New Orleans) so I just develop the models then sub the analyses to Mechanical Solutions (www.mechsol.com).

                     

                    By the way, yes, you are correct, bulk modulas is the correct term... hot water is pretty "springy" compared to most engineering materials.. the bulk modulas of steel is 70 to 75 times greater than room temperature water.

                     

                    Again, thanks

                     

                    Dennis Bowman