Horizontal Tank Supports

[agfroese]

Hi,
I understand the typical support sytem for horizontal tanks is two saddles and have seen many of the references on it, but is anyone aware of any references that discuss the "rail support" design as seen in the attached pdf? For those who can't view the pdf for some reason, the "rail supports" are two "feet" supports along the length of the cylindrical shell.

I've been asked to review this design for some larger diameter horizontal tanks that exceed the sizes specified in UL 142. Before I dig too deep into FEA's and such, just curious if anyone knows of any reference material or if anyone has any experience with this type of support system. Obviously the foundation would need to be rigid to avoid issues with differential settlement.

Thanks!

 

[dhengr]

Agfroese:
Well... you don’t give us the shell thickness, but in looking at the 4000gal. tank, I’ll guess the shell thickness is .25" stl. And, you’ve got about 2kips/ft. on the two rails when the tank is full, that’s 1k/ft. on each rail. This support scheme eliminates the need to worry about tank bending along its length, and those combined longitudinal stresses in the shell, and the concentrated reactions at the two saddles. While the foundation should be stiff to avoid differential settlement, as you suggest, the more important thing might be that it be very flat on top, or that you consider some means of shimming and/or grouting the rails so that the loading into the shell is relatively uniform. These considerations will dictate the height and thickness of the rail shapes, and their stiffness, to bridge any hard spots. You might want the actual foot width to be fairly narrow, a function of conc. brg. strength, to facilitate grouting or good shimming. And, as a starting point, I would want the two legs on the rail to intersect the shell at about the same angle w.r.t. the shell, so that the force components introduced into the shell were about equal. This will impart two equal tensile forces into the shell in the immediate region of the rail, but not much unbalanced buckling force beyond the region of the rail, and also two radial punching components on the shell. Then there is also the question of how you tie these rails down to the foundation.

I don’t have UL 142, so you would have to post some of the relevant parts of that. I would start with my std. ref. texts on plates and shells, which I haven’t done yet. You’ve got internal liquid pressures and the above concentrated tangential and radial forces all producing hoop stresses and punching stresses in the shell. Combine the hoop stresses and study the punching/shell buckling situation, and your home free. Watch how you terminate the rails at the end of the tanks. They will have a nasty punching/plate bending affect on the shell very near the stiff end plates. You might consider running the rails out beyond the tank end plates an inch, so that you can weld an end closure plate up onto the tank end plate. This would act primarily in shear, on the end plate, and eliminate the concentrated bending/punching, and it keeps the critters out. Provide some drain holes in these rails. I might also consider eliminating the two short bent weld flanges on the rails, they are two more brake forming operations, and fairly tough to do on such short grasp length. Furthermore, if not bent and fit-up well, they might cause prying on the welds. Why not form the three sides of the rails, and then nibble a 1/16th or 1/8th inch off the inside corner edges; then press the rail to the tank and weld the remaining groove.

The 2kips/ft. on a 5' wide slab foundation should be no problem, just remember the top rebar 4.5' long, across the slab.