Wind Girders for API 650 floating roof tanks

[carthago]

What is the theory behind finding the Z (= modulus od resistance of the ring including tank shell contribution ...in in3 or cm3!!).

I mean if there is a (Z) then there is a bending moment.
How to find this Moment and the corresponding Shear Force ??

The reason I am asking this is: we want to use a wind girder with no web , instead we will use a sort of bracing or lattice to link one side of the Ring girder to to the tank wall...
And the question is how to size and check the lattice ?

I mean one has to reckon also with shear load and lnk structurally the two ring flanges !!

Thanks for your answer

 

[JoeTank]

The API load model is quite approximate and is used to calculate bending loads only. The intent is limit the flexing of the shell under wind loading. The stresses are limited to about 15,000-20,000 psi as a means to limit displacements of the windgirder. Shear loads are pretty low, therefore an equation for shear is not given my API-650. I derived the API-650 equation many years ago, but I have not seen the derivation given in open literature.

Joe Tank

 

[TankDude]

It seems to me that what you are proposing could be evaluated much like a lattice tower structure to determine the load transfer into the lattice members. My thought regarding evaluating this per the API-650 requirements is that you would look at the combined moment of inertia of the contributing upper shell and the flange plate to determine if you have enough section. The lattice members would simply serve to transfer the loads and I'm not sure I would contribute any of the cross-sectional area of these members to the overall value.
Just my opinion. I'm curious what others have to say about it.

 

[JStephen]

You can make various assumptions about the wind pressure distribution on the shell, and the way this is transferred to the wind girder. But trying this with different assumptions has never given me the API formula. In the oldest standards I have, this formula is given, and I think it is even referenced in the technical paper introducing the wind stability equations, but with never a mention of derivation or source. It could even have been empirical, for that matter. API will not clarify these kinds of things on a formal basis, although if you can get in touch with the right people, they may give you informal explanation.

A similar question comes up when superimposing dome roof loads onto the wind girder.

I would suggest to take the required section modulus, multiply by your estimation of an allowable stress, and take that as the designated API wind moment.

 

[JoeTank]

TankDude,
Lattice type windgirders were used (though not extensively) on open top tanks many years ago. The few that I have seen were on riveted tanks. Cost-wise, I have to believe that they are far more expensive thanplate type windgirders, primarlity due to the labor content.

Joe Tank