Steel Conical Hopper Design

[Acoona]

I am designing a concrete silo with a steel conical hopper in a seizmic zone. Using formulas for Meridional, Tangential and forces perpendicular to hopper by hand and comparing them to results in a RISA model, the forces do not correlate.
I am using a wall friction of zero and a vertical design pressure of (gammaXvert depth).
In my Risa model I am combining all the forces (vert & perp to hopper wall), is this correct?
The total load matches but the forces do not and RISA results are 1.72 times higher for Meridional forces (Fm), 1.67 times higher for normal forces(Fn), and close for tangential (Ft).

 

[jhardy1]

How are you modelling the support at the top of the hopper? You can get very different local plate stresses predicted depending on whether you model:

a) Fully fixed – no vertical or radial movement, all rotations fixed.
b) Anchored and pinned (no vertical or radial movements, but plate free to rotate around the circumferential direction)
c) Vertical support only – radially free to slide, and free to rotate about the circumferential direction.
d) Vertical support and moment-restrained – free to slide radially, but plate restrained against rotation about the circumferential direction.

You need to check that your modelled constraints are equivalent to the restraint assumed in your design formulas, to correlate your FEA results with your design formulas. Then you need to ask yourself whether this restraint is actually the best representation of the REAL support condition. For example, is there a stiff ring at the top of the hopper, which is not taken into account in the design formula?

In other words, you may find that properly modelled restraints may actually give different results to the design formula, if the restraint in the design formula is not appropriate to the real-world situation.

Hope this helps.

 

[JoeTank]

Acoona,

I'm not familiar with the RISA program, but am very familiar with the design of bins and silos. You say that you used a zero friction coefficient and that pressure = density x depth. Does the RISA program calculate pressures within the mass of a granular bin or hopper? Why use a zero frictional coefficient? Why not just use the correct value?

If the RISA program is using hydrostatic pressures, then you are comparing apples and oranges. You will not get agreement between hand calcs and FEA results.

Steve Braune
Tank Industry Consultants

http://www.tankindustry.com