Minimum Shell thickness calculation for Tank

API is an American standard. t_min will be in inches, and maybe H and D also, although one or both might be in feet.

Trying inches first:

9m ≈ 354in; 10m ≈ 394in; S =? 9000psi for SA36 plate (not sure of this last);

t_min = (2.6)(354)(394)(1.0)/(9000)(1.0) ≈ 40.3in; which is ridiculous.

Trying again with H and D in feet:

9m ≈ 29.5ft; 10m ≈ 32.1ft; S as before;

t_min = (2.6)(29.5)(32.1)/(9000)(1.0) ≈ 0.274in; which is at least reasonable.

Beware that I am somewhat guessing here, but this may help get "into the ballpark." All of this should be checked by a professional structural or vessel engineer.

Thanks, this one is acceptable as I was getting unrealistic values. Still trying to equalize units on both side of equation so I could comeup with inch units of thickness.

I found that doing the normal calculations come up with an very thin wall plate thickness. So instead of going straight for that thickness I tend to add the loss of material due to rust and paint prep and so on at 1mm per 8 years.

Also I add wall strength if the tank is open. Also I prefer to stick to standard plate height and to install ring stiffeners at those joints. So going purely on the spec calculation is a good indicator but add all the other factors I normally end up with an 10mm thk. for the first 2.4 m height 8mm thk. for the next 2.4 m and 6 for the rest . ( hope this makes sense ) A ring beam at 4.8 m and along the top rim.

I think I've sussed this out. Basic formula for t is t = P*D/(2*S) in consistent units (ignoring G and E, which are 1 anyway). If you want t in inch, P and S are in psi, D in inch. But for convenience API653 uses H and D in foot, S in psi. P (psi) = H/2.3, and D has to be x 12. So t = H*D*12/(2*2.3*S) = 2.6*H*D/S.

This gives t = 0.075in, less than Tornado in #1 as he used S = 9000psi, but 230MPa = 230*145 = 33350psi. But I wouldn't use 230MPa either, it's much too high. I'd say 170MPa max, and then add at least 3mm corrosion allowance, though that depends on the tank contents and what the full spec says. With 3mm ca, still come to ¼ inch plate.

Codey

Totally agree with you on that point, yes ¼ inch will do, but have you ever build one on-site.

At a radius of 5m these plates are rather wobbly and then there is the welding to the toe plate and shape stability during construction. A slight wind will flop your shape as especially with the height before the first ring stiffener. That's why I prefer using a rolled 10 plate at the bottom on a 8 to 12 thick toe plate and a 3 to 4 thick floor plate on a 50 to 80 thick bitumen sand layer. Builed a few and they lasted till now. (20 years ago) I must say we used only 250 WA (the old British 43a steel) and now we can only get from 350 WA steel. So I guess you are totally right with you recommendations / calculations.

Anybody else for a comment .

You've got me there, no, I've never built one of these myself, just answering the theoretical question . OP probably best to leave thickness up to the supplier/builder, using this thread to make sure they don't come up with thickness below calculated minimum.

Can't tell whether his intention is steel bottom on bitumen sand, or set a ring beam into the concrete base and weld the sidewalls to that, similar to how glass-coated bolted steel sectional tanks are bolted to an angle ring set into the concrete.

You are using API 653, its for inspection purpose. Try using API 650, its for designing purpose and according to it the minimum thickness is 6mm + corrosion allowance. Hope this will help!