| Course | H (m) | td (mm) | tt (mm) | Selected (mm) | Weight (t) |
|---|
5 Critical Engineering Traps in API 650 Shell Sizing
1. The Hydrotest Overpressure Pitfall on Light Hydrocarbons
Designers concentrating on operating conditions often assume the design thickness t_d governs because it includes corrosion allowance. However, for light hydrocarbon storage (gasoline, naphtha, condensate with G = 0.70 to 0.78), hydrostatic pre-commissioning testing requires filling the tank to the brim with water (G = 1.0). The resulting hydrotest thickness t_t often exceeds t_d by 15% to 25%. Fabricating shell plates based on t_d alone will rupture the bottom shell course during hydrotest filling.
2. Violating API 650 Table 5.2 Minimum Plate Rigidity Limits
In upper courses (Courses 4, 5, 6), hydrostatic liquid head drops toward zero, yielding calculated membrane thicknesses as thin as 1.5 mm. Sizing plates strictly by calculation causes catastrophic tank failure; thin steel lacks buckling resistance under atmospheric vacuum or lateral wind loads. API 650 Table 5.2 mandates strict absolute minimums (5 mm for D < 15m, 6 mm for 15-36m, 8 mm for 36-60m). Overlooking these minimums leads to shell ovalization and wind girder buckling during construction.
3. Spot Radiography (E = 0.85) vs Full Radiography Steel Penalty
Contractors attempting to minimize NDT inspection costs by specifying spot radiography (E = 0.85) trigger a mandatory 17.6% increase in plate thickness across every single course of the tank. For a 30-meter diameter storage tank, the additional steel weight costs upwards of $60,000, dwarfing the trivial $4,000 cost of full 100% X-ray weld examination (E = 1.0).
4. Floor-to-Shell Bottom Restraint Stress Concentration
The 1-Foot Method evaluates hoop stress at 0.3 m (1 ft) above the bottom weld. However, at the exact corner weld where the vertical shell joins the horizontal annular floor plate, radial expansion is clamped to zero. This creates severe local rotational bending stresses in the bottom 150 mm. If the annular floor plate thickness (typically 8 to 12 mm per API 650 Table 5.1a) or corner weld fillet geometry is undersized, low-cycle fatigue cracking initiates at the bottom corner weld.
5. Using One-Foot Method on Tanks Exceeding 60m Diameter
API 650 Section 5.6.3 explicitly prohibits the 1-Foot Method for tanks exceeding 60 meters (200 feet) in diameter. In large crude oil tanks (e.g. 80m to 100m diameter, 100,000 m³ capacity), the shell stiffness ratio shifts the point of peak bending stress higher up the shell. Tanks over 60 m must be calculated using the Variable Design Point Method (API 650 5.6.4) or Finite Element Analysis (FEA) to ensure structural integrity.
For cylindrical flat-bottom storage tanks, hydrostatic pressure increases linearly with liquid depth. In SI Metric units, the nominal shell plate thickness for each course is calculated as follows:
t_d = [ 4.9 · D · (H - 0.3) · G ] / [ S_d · E ] + CA (mm)
2. Hydrostatic Test Shell Thickness (tt):
t_t = [ 4.9 · D · (H - 0.3) ] / [ S_t · E ] (mm)
3. Governing Thickness for Course i:
t_gov = max( t_d, t_t, t_min_API )
4. Minimum Nominal Plate Thickness per API Table 5.2:
- D < 15 m: t_min = 5.0 mm
- 15 m ≤ D < 36 m: t_min = 6.0 mm
- 36 m ≤ D ≤ 60 m: t_min = 8.0 mm
- D > 60 m: t_min = 10.0 mm
5. Selected Commercial Plate Thickness:
Round t_gov up to nearest commercial standard plate (e.g. 6, 8, 10, 12, 14, 16, 18, 20, 22, 25 mm).
6. Course Steel Mass: m_course = π · D · W_course · (t_selected · 10^-3) · ρ_steel
Where ( D ) is tank diameter (m), ( H ) is liquid head from bottom of course to design liquid level (m), ( G ) is liquid specific gravity, ( S_d ) and ( S_t ) are allowable stresses (MPa), ( E ) is joint efficiency, and ( CA ) is corrosion allowance (mm).