WATER TANK CONE ROOF DESIGN

[RichRook]

Wanting to retrofit a 36' diameter vertical water tank with a cone roof.

The original tank drawings reference AWWA D-103 'Factory Coated Bolted Carbon Steel tanks for Water Storage' But this standard gives a very vague criteria for roof design...

"5.8.1 Roofs having a slope of 1 in 2.75 or greater shall have a minimum thickness of 0.070 . . . . ."

I have an old AISI standard (steel tanks for liquid storage 1976) that gives D/400/SIN(theta) (theta=roof slope) which would give about 0.25" plate for my situation.

Do one of the other AWWA standards (D-100?) have better written requirements/formula for the calculation of the thickness for unsupported/unstiffened cone roofs? Would like to find a current rational method from a good source. I think the D-100 might be the one I'm looking for, but the description isn't great and I'd rather be sure before I buy.

 

[BAretired]

Not a simple question. You are asking about a thin conical shell spanning 36' in diameter under gravity and wind loading. The following link provides a few factors to consider. Buckling of the shell is not a trivial analysis.

The formula you cite for thickness is not clear. Is it t = D/(400sinθ)? I don't think it can be that simple.

https://books.google.ca/books?id=rv...#v=onepage&q=conical steel shell roof&f=false

BA

 

[JStephen]

Is this a bolted tank (D103) or a welded tank (D100)?
If a bolted tank, the most likely roof retrofit, I'm guessing, would be an aluminum geodesic dome.
If a bolted tank, also take a good look at replacing the whole tank.

For a welded tank, you can use a supported cone, self-supporting cone, or umbrella roof. Size is kind of borderline for a self-supporting cone, but it can be done. It is about that simple in terms of design of the roof itself. The D/400sin(theta) is from API-650, and that design is referenced in the current AWWA D100. There's also a factor to allow for varying total DL & LL in the current standard. Note that you can vary the slope by adjusting thickness or vice versa. There's a separate equation for minimum top-angle area, but it doesn't normally control the design. The thinnest roof will also be the least convenient to access due to slope. For a roof that size, I'd probably look at 5/16" or 3/8" plate. There's a lot of different ways things could be done, and a lot of the details just come down to economics for that particular project.

In either case, it'd be prudent to get in touch with some of the people that might actually do the work for recommendations.

 

[RichRook]

The drawings that I have indicate D-103 (bolted) although the details aren't very specific.

JStephen . . for the Bolted Tank, why would you say replace the tank (it's relatively new)? also, why not use a supported steel shell in this situation if the unsupported shell isn't appropriate?

Thanks

 

[JStephen]

Have you seen the actual tank? If not, is it visible on Google Streetview or satellite pics?
Replacing vs rehab vs new roof or whatever is anticipated would just depend on the relative cost. But worth looking into both ways.

 

[MonsieurR]

I assume this is materialized as "pizza slices", you haven't provided a general sketch of the situation. Why do you want to retrofit it?

Take a look at API 620 and 650 for sanity checks as well.

Hand calculations for buckling would be hard, I would personally use radial rafters to reinforce the shell. Maybe if the tanks are in use one could weld the rafters over and place another conical shell over.