Queries on modelling hydrostatic force acting on tank

[ptc-4960854]

I have a rectangular atmospheric tank (open at the top) that is supported on 4 legs that are fully constrained. The tank is about 1 meter high and is able to hold 20 bbls of water.The atmospheric tank is filled with water and i would like to find analyze the stress acting on the 4 legs.How do i input the hydrostatic forces acting on the tank due to the water. Any help would be very much appreciated.

[jsarkar]

If the cross section does not varies with height then pressure at bottom surface would be same. for calculation of stresses in legs I feel, pressure at vertical walls of tank is not necessarry. So it would solve your problem with proper contact surface definition or full assembly can be treated as a single part alos work fine.

If you are talking about the amount of pressure, it should be 0.1 kg/cm2 to be taken for 1 m of water column. Since the cross section is not varying so paradox need not to be considered. You may need to increase the design pressure more than operating as per codes in which you are working on.

[SylvainA.]

Why don't you simply use the weight of the water as a vertical load ?

Is your tank deformable ?

[JonathanHodgson]

If you do need the load on the walls of the tank, I think you can define pressure as an equation in terms of x, y and z... see whether this can do what you want.

[integrisbrad]

Hello,

I agree with others that your approach is going to be dependent on how the legs are attached and if the tank wall where the legs are attached (if they are welded to the tank, for example) is deformable enough that effects of pressure will come into play. As Jonathan mentions, you certainly can simulate hydrostatic pressure with the ability to input pressure as a function of coordinates that Mechanica provides, if you find that it is necessary to your analysis. Also, if the tank is just sitting on the floor, fully constrained may not be the most appropriate representation of the constraint. It could artificially inflate your stresses, but it might be a good conservative estimate. So much of your setup is dependent on the specifics of your model and your analysis intent.

Brad Green