A pressure vessel can pass a simple hoop-stress check and still be poorly understood. Local loads at a nozzle, a thin transition near a head, thermal restraint, support flexibility, cyclic service, or external pressure can control the design long before the shell membrane stress becomes critical. That is why knowing how to analyze pressure vessels means building a defensible engineering workflow, not simply applying a pressure load to a finite element model.
For engineering teams, the objective is clear: demonstrate that the vessel satisfies the governing code, survives its real operating environment, and can be manufactured, inspected, and operated with acceptable margin. Hand calculations, code rules, finite element analysis (FEA), and engineering judgment each have a role. Problems arise when one method is treated as a substitute for the others.
Start With the Design Basis, Not the Mesh
The analysis begins by defining exactly what is being evaluated. Identify the governing construction code and edition, such as ASME Section VIII for many US pressure-retaining applications, along with applicable client specifications, material standards, and jurisdictional requirements. The code determines allowable stresses, required load combinations, weld assumptions, fatigue methods, and acceptance criteria.
Pressure is only one part of the load definition. Establish the design pressure, maximum allowable working pressure if relevant, design temperature, operating temperature range, vacuum condition, hydrotest pressure, fluid density, corrosion allowance, and expected life. Then define external actions: dead weight, piping loads, wind, seismic acceleration, transportation loads, platform loads, support reactions, and thermal expansion restraints.
This stage often reveals the real question. A horizontal vessel may be controlled by saddle stresses during hydrotest. A vertical vessel may be governed by shell compression from wind and seismic overturning. A jacketed vessel may require careful treatment of differential thermal expansion. If the design basis does not represent these conditions, a highly refined model only produces a highly refined wrong answer.
Use Code Calculations to Establish the Baseline
Code-based calculations should establish the initial shell, head, and nozzle geometry before detailed FEA begins. For common geometries under internal pressure, these calculations provide required thicknesses and a transparent baseline for review. They also help identify whether corrosion allowance, mill tolerance, material availability, or fabrication limits will govern the nominal thickness.
For a thin cylindrical shell, the membrane stress relationships provide useful physical checks. Circumferential stress is approximately proportional to pressure times radius divided by wall thickness, while longitudinal stress is roughly half the circumferential value for a closed-end cylinder. These relationships are not a complete design method, but they are valuable for detecting unit errors, incorrect pressure application, or unrealistic FEA results.
The baseline must also address heads, transitions, openings, reinforcement areas, flanges, tubesheets, and external-pressure stability where applicable. Standard code rules are efficient for conventional details within their limits. When geometry falls outside those limits, or when load interaction is complex, a local FEA assessment may provide the needed insight.
The trade-off is not code calculation versus FEA. It is using code methods where they are well suited and using FEA where geometry, loading, or behavior requires a more detailed representation.
How to Analyze Pressure Vessels in an FEA Model
An FEA model should answer a defined engineering question. Modeling every weld bead, flange thread, and attached component is rarely productive. Conversely, an overly simplified shell model may miss the local behavior that prompted the analysis. The appropriate model fidelity depends on the decision it must support.
Select Elements That Represent the Structure
Shell elements are often effective for vessel shells, heads, skirts, and stiffeners when the thickness is small relative to other dimensions. They are computationally efficient and can capture global behavior, including membrane and bending stresses. Solid elements are more appropriate near thick nozzles, forged components, tubesheets, local discontinuities, or regions where through-thickness stress gradients matter.
A mixed shell-solid model can be effective, but transitions require care. Incompatible offsets, poorly connected interfaces, or artificial stiffness at shell-to-solid boundaries can distort results. The model should represent the shell midsurface or physical geometry consistently, including thickness offsets and eccentric load paths.
Apply Loads and Constraints Realistically
Pressure must act normal to the wetted surface and include end-cap effects when required by the model boundaries. A common modeling error is applying internal pressure to a shell while constraining the ends in a way that eliminates the axial load generated by closed ends. The resulting longitudinal stress can be materially incorrect.
Support boundary conditions deserve equal attention. Saddles, lugs, skirts, legs, and foundation anchors do not behave as perfectly fixed points. Model the contact area, restraint directions, and stiffness that are physically present. For piping loads, apply forces and moments at realistic nozzle locations and consider whether a local solid submodel is required.
Thermal loads require a temperature field, temperature-dependent material properties where appropriate, and a credible representation of restraint. Uniform heating of an unconstrained vessel creates little thermal stress. A temperature gradient or restrained expansion can create significant stress. Treating thermal loads as a generic load case without understanding the constraint path is a frequent source of misleading results.
Refine the Mesh Where the Stress Changes Rapidly
Mesh refinement should follow stress gradients, not visual preference. Nozzle intersections, knuckles, saddle edges, stiffener terminations, support attachments, and abrupt thickness changes typically require local refinement. Use a mesh convergence study to demonstrate that the stress quantity used for acceptance has stabilized sufficiently.
Do not evaluate a singular stress peak as though it represents a physical failure mechanism. Sharp re-entrant corners, point loads, fixed constraints, and idealized weld toes can produce mathematically unbounded stresses as the mesh is refined. In these regions, engineers must distinguish between membrane stress, bending stress, peak stress, and stress linearization results. The appropriate classification depends on the governing code method and failure mode under review.
Account for Nonlinearities and Stability When They Matter
Linear static analysis is a useful starting point, but it is not always enough. Large deformation effects may matter in slender structures, flexible supports, or components approaching buckling. Material nonlinearity may be needed for limit-load evaluation, and contact nonlinearity may govern flange, saddle, or support behavior.
External pressure deserves particular discipline. Vacuum, jacket pressure, or differential pressure can cause shell instability at stress levels far below material yield. An eigenvalue buckling analysis can identify likely mode shapes and provide a preliminary indication of sensitivity. It should not automatically be treated as the final allowable load because real vessels are affected by geometric imperfections, residual stresses, ovality, fabrication tolerances, and nonlinear response.
For fatigue-sensitive equipment, pressure cycles alone may not define the damaging event. Startup and shutdown thermal transients, mixer loads, piping movement, and batch-process temperature changes can dominate the stress range. Use a transient thermal analysis when the temperature distribution through the wall is significant, then transfer that field to a structural model for stress and fatigue evaluation.
Validate Results Before Accepting Them
A credible vessel analysis includes independent checks. Compare global FEA membrane stresses with hand-calculation expectations. Confirm that applied pressure produces the correct resultant forces. Review reactions against total applied loads. Check displaced shape plots to ensure the vessel deforms as engineering intuition predicts.
Validation also means reviewing assumptions with fabrication and operations in mind. Is the nozzle configuration weldable? Does the reinforcement concept create inspection challenges? Are assumed support conditions consistent with the installation? Is corrosion allowance included in the correct geometry for the service-life case? An analysis can be numerically correct yet unsuitable for the equipment that will actually be built.
When the outcome carries significant safety, cost, or schedule consequences, independent model review is valuable. Another experienced analyst can often identify an omitted load path, inconsistent unit system, inappropriate element formulation, or code interpretation issue before it becomes a late-stage design change.
Document a Decision, Not Just a Color Plot
A useful report explains the design basis, code criteria, geometry, materials, load cases, boundary conditions, element types, mesh strategy, solver settings, and acceptance method. It should clearly separate nominal stress results from local peaks and identify any assumptions that require field confirmation or fabrication controls.
Color contours are communication tools, not proof by themselves. Include free-body checks, deformation plots, stress linearization paths where relevant, convergence evidence, and a concise explanation of what governs the design. Decision-makers need to know whether the vessel is acceptable, what margin exists, what conditions limit that conclusion, and whether a design change is recommended.
The best pressure-vessel analysis creates confidence because its conclusions can be traced from service conditions to load paths, code requirements, model assumptions, and independently checked results. That traceability is what turns simulation into an engineering decision tool rather than an attractive post-processing image.