Flat End Plates Under Pressure
Bending-dominated response of flat end plates, plate stiffness, boundary conditions, pressure and thickness relationships, and membrane effects at larger deformation.
Technical provenance
Applicable standards / specifications
- ASME BPVC Section VIII Division 1 (2025) — Rules for Construction of Pressure Vessels
- ASME BPVC Section VIII Division 2 (2025) — Alternative Rules for Construction of Pressure Vessels
- EN 13445 — Unfired pressure vessels — Relevant European pressure-vessel code family where specified by the project.
References
- ASME Boiler and Pressure Vessel Code — 2025 edition — Primary code family reference for pressure-vessel design where ASME BPVC is the governing basis.
- Moss, D. R. & Basic, M. — Pressure Vessel Design Manual — Background engineering reference for pressure-vessel load paths, stresses and design checks.
Bending-dominated response
A flat end plate under pressure responds primarily in bending, unlike a domed head which responds in membrane action. The plate bends like a circular plate under uniform load, with the maximum bending moment at the centre or at the edge depending on the boundary condition. The bending stress in the plate is much higher than the membrane stress in an equivalent domed head, which is why flat ends require significantly greater thickness.
Plate stiffness and thickness
The bending stiffness of a flat plate is proportional to t^3, so increasing the thickness has a strong effect on the bending stress. The maximum bending stress in a clamped circular plate under uniform pressure is proportional to p * r^2 / t^2, compared to the membrane stress in a domed head which is proportional to p * r / t. The quadratic dependence on r and the inverse-square dependence on t make flat end plates very sensitive to diameter and thickness.
Boundary conditions
The boundary condition at the plate edge significantly affects the bending moment distribution. A clamped edge (no rotation, no displacement) gives the highest moment at the edge and a lower moment at the centre. A simply supported edge (no displacement, free rotation) gives the highest moment at the centre. The actual boundary condition in a vessel lies between these extremes — the shell provides some rotational restraint but is not perfectly clamped. The assessment should consider both bounding cases or use a realistic boundary stiffness.
Pressure and stress relationship
For a clamped circular plate of radius r and thickness t under uniform pressure p, the maximum bending stress is approximately:
sigma_max approximately = (3 + 3*nu) * p * r^2 / (8 * t^2)
For nu = 0.3:
sigma_max approximately = 0.41 * p * r^2 / t^2
Compare with a hemispherical head:
sigma_hemisphere = p * r / (2 * t)
For r/t = 10:
Flat plate stress / hemisphere stress =
0.41 * 10 / 0.5 = 8.2
The flat plate carries 8 times the stress of a
hemisphere of the same radius and thickness.Membrane effects at larger deformation
At small deflections, the flat plate response is purely bending. As the deflection increases to a significant fraction of the plate thickness, membrane forces develop because the plate edges are restrained from moving inward. These membrane forces stiffen the plate and reduce the rate of stress increase with pressure. This membrane-bending coupling means that a flat end plate at large deformation is stronger than the pure bending prediction. However, relying on membrane effects requires allowing the deflection to occur, which may be unacceptable for sealing or functional reasons.
Plate-theory basis and applicability
Classical circular-plate solutions are useful because they expose the controlling scaling: pressure loading grows with area, bending moment grows strongly with span, and bending stress varies approximately with p a^2 / t^2. The numerical coefficient depends on edge restraint and Poisson ratio, so a formula should not be used without matching its boundary condition to the actual detail. A welded plate connected to a flexible shell is neither perfectly clamped nor simply supported. Analytical solutions are best used as bounds and as independent checks on an FE model rather than as universal design equations.
Large-deflection behaviour
When centre deflection becomes comparable with plate thickness, geometric nonlinearity becomes important. In-plane membrane forces develop and the pressure is carried by a combination of bending and stretching. This can reduce incremental bending stress relative to small-deflection theory, but it also changes edge reactions and may produce permanent deformation, gasket movement or loss of alignment. A linear-static FE model cannot capture this stiffness evolution. If the design relies on membrane action developing in a flat closure, a geometrically nonlinear analysis with realistic support restraint is required and serviceability limits should be checked separately from collapse capacity.
Edge details and attachment loads
The edge of a flat closure is often more important than the plate centre. A welded edge introduces local bending and weld-toe stress; a bolted cover introduces flange rotation, bolt preload and gasket compression; a machined integral closure can have a sharp thickness transition. Pressure reactions should be recovered around the support circumference to confirm that the attachment carries the full resultant. Where the closure forms part of a removable access cover, local bolt-circle bending and prying can govern even when the plate itself is adequate.
Nonlinear collapse and proof behaviour
For high pressure or thin plates, elastic stress may exceed yield locally well before collapse. The relevant question then becomes whether plastic redistribution is stable and whether permanent deformation is acceptable. Elastic-perfectly plastic or realistic material curves can be used to examine load-deflection response, plastic-zone growth and collapse mechanism. Proof pressure should not be interpreted simply by multiplying operating stress by the proof ratio, because yielding and geometric nonlinearity can change the response. The model should distinguish a permitted proof-induced local plasticity from unacceptable gross deformation or progressive damage.
Practical verification checks
Check plate thickness, clear span, actual edge restraint and pressure area independently before reviewing contours. Compare FE centre deflection and bending stress with at least one closed-form circular-plate solution using a bounding support assumption. Confirm that reaction force equals pressure times loaded area. Repeat with a deliberately coarser and finer mesh to separate physical bending from local edge singularities. If contact, bolts or a gasket are present, also check that the pressure load is not being carried accidentally by numerical constraints rather than the intended structural interface.