A colourful stress plot is a familiar output from Finite Element Analysis (FEA). But producing a plot and producing a credible engineering assessment require very different levels of understanding.
FEA describes a broad range of analysis methods. It does not, by itself, define the capability of the software, the suitability of the approach or the reliability of the result.
At Rota Design, we believe the value lies in selecting the right analysis method, representing the real application and interpreting the results through practical engineering experience.
The analysis must suit the problem
A linear static analysis can be appropriate when deformation is small and the material response is approximately linear. However, many products operate outside those assumptions.
A component may bend enough to change how it carries load. Parts may come into contact, separate or slide. A material may yield, or its deformation may increase over time under a sustained load.
These situations can require different modelling techniques and solver capabilities, including:
- Material nonlinearity to represent behaviour such as plastic yielding or time-dependent deformation.
- Geometric nonlinearity where movement or deformation changes the structural response.
- Contact analysis to investigate how components transfer load as they touch, slide or separate.
- Dynamic analysis where inertia and the rate or duration of loading influence the response.
- Stability analysis to assess susceptibility to buckling.
Using a simpler method without understanding its limitations can produce convincing-looking results that do not adequately represent the product.
Assumptions shape the answer
Every analysis is a model of reality. Decisions about supports, loads, connections, material properties and mesh all influence the outcome.
Is a mounting genuinely rigid? Does a load act at a point or over an area? Can a joint rotate? Are the material properties representative of the operating temperature and loading duration?
For moulded plastic products, wall thickness variation, creep and temperature sensitivity may be particularly important. For fabricated assemblies, joint stiffness and the load path through the structure can materially affect the result.
A more detailed mesh cannot compensate for an incorrect assumption.
Software capability matters—and so does the analyst
FEA packages differ in their element formulations, material models, contact algorithms and solver capabilities. An analysis feature being available does not automatically make it suitable for every application.
The analyst must understand which capabilities are needed, how to apply them and how to check that the solution is credible. That includes examining convergence, checking reactions and deformation, recognising numerical artefacts and questioning unexpected results.
A solver completing successfully is only one part of that assessment.
Practical engineering turns results into decisions
A high stress value needs interpretation. It may indicate a genuine weakness, a local modelling effect or a mathematical singularity. Equally, an apparently low stress result may conceal unrealistic supports or an incomplete loading scenario.
Practical knowledge of materials, manufacturing, assembly and service conditions helps distinguish between these possibilities—and turn findings into useful design changes.
At Rota Design, we combine analysis with practical engineering experience to assess the modelling approach, explain the assumptions and interpret what the results mean for the product. Where appropriate, physical testing provides further evidence to support and refine the assessment.
The question is more than “Has it been analysed?” It is “Does the analysis represent how the product will actually behave?”
What types of analysis sit behind an FEA study?
“FEA” is an umbrella term. Within it are different analysis types and solvers, each addressing different aspects of a product’s behaviour. Choosing between them requires an understanding of the material, loading, connections and operating conditions.
A solver is the numerical engine that calculates the response of the model. Its formulation determines which physical effects it can represent, and the assumptions under which its results are valid.
Linear static analysis — strength and stiffness under steady loads
Linear static analysis calculates stresses, strains, deflections and support reactions where inertia effects are negligible, material behaviour is linear elastic and deformation does not significantly change how the structure carries load.
Typical applications include brackets, frames, supports and fabricated assemblies under defined service loads.
It can be an efficient and appropriate engineering tool. However, it does not automatically account for yielding, large deformation, changing contact conditions or deformation that develops over time.
Nonlinear static analysis — when the response changes as load increases
Nonlinear analysis is needed when the relationship between load and displacement is no longer proportional. This can arise from three main sources:
- Geometric nonlinearity: deformation changes the stiffness or load path, as can occur in flexible panels, thin shells and components undergoing large movement.
- Material nonlinearity: the material response changes with stress or strain, for example when a metal yields or a polymer behaves nonlinearly.
- Contact nonlinearity: components touch, separate or slide, changing how forces pass through the assembly.
Loads are typically applied progressively, with the solver iterating towards equilibrium at each stage. The load sequence, step size and convergence checks therefore become important parts of the assessment.
For example, a flexible container wall may change shape sufficiently under pressure that its deformed geometry must be considered when calculating stresses and displacement.
Buckling and stability — failure without exceeding material strength
A component can become unstable before its material reaches its strength limit. Slender members, thin panels and shells may buckle under compression or external pressure.
A linear eigenvalue buckling study identifies theoretical buckling load factors and associated mode shapes. It provides useful information about susceptibility to instability, but the result should not automatically be treated as the allowable working load.
Where imperfections, large deformation, yielding or changing contact influence stability, a nonlinear assessment may be needed to investigate a more representative response.
Natural frequency and modal analysis — understanding vibration characteristics
A modal study identifies a structure’s natural frequencies and the shapes in which it tends to vibrate.
This helps assess whether operating conditions, such as motor speeds or repeated mechanical excitation, could coincide with a natural frequency and create a risk of resonance.
However, identifying a natural frequency does not establish how large the vibration will be in service. That requires a response analysis incorporating the excitation and appropriate damping assumptions.
Harmonic response — repeated excitation at operating frequencies
Harmonic response analysis predicts the steady vibration response to sinusoidally varying forces or support movements.
It can investigate how displacement, stress and phase vary across a frequency range, helping identify operating conditions where vibration becomes excessive.
Typical applications include machinery supports and assemblies subjected to periodic excitation from rotating equipment. The assumed damping can strongly influence the predicted response near resonance.
Transient dynamic analysis — following an event through time
Transient dynamic analysis calculates how a structure responds as loading changes over time, including the effects of mass, inertia and damping.
Applications can include sudden loading, acceleration, braking, shock and impact events. Results show how displacement, velocity, acceleration and stress develop throughout the event.
A linear transient approach may be suitable where the response remains linear. Nonlinear transient analysis may be required where the event involves large deformation, material nonlinearity or changing contact.
The loading history, time-step size and representation of contact and damping are essential to obtaining a meaningful result.
Spectral response and random vibration — loading described by frequency content
Some dynamic environments are described using a spectrum rather than a single force or a detailed time history.
Response spectrum analysis estimates peak structural responses from a specified spectrum, while power spectral density analysis assesses statistical response to random excitation.
These approaches can be useful for seismic assessments, transport vibration and machinery environments. Their results require careful interpretation: a statistical vibration response or combined modal maximum is different from a result at a particular instant in time.
Quasi-static and creep analysis — behaviour under sustained loading
A product may perform satisfactorily when first loaded but continue to deform while that load remains in place.
Quasi-static analysis follows behaviour over time while neglecting dynamic inertia effects. With an appropriate creep material model, it can investigate deformation and stress redistribution under sustained loading.
This is particularly relevant to polymer tanks, containers, supports and other products carrying loads for extended periods. Reliable assessment requires material data representative of the stress level, temperature and loading duration; short-term stiffness data alone may be insufficient.
Heat transfer and thermal stress — temperature as an engineering load
Thermal analysis investigates temperature distributions and heat flow through a component or assembly.
A steady-state study assesses the equilibrium temperature distribution. A transient study follows heating or cooling over time. Depending on the application, the model may include conduction, convection, radiation and temperature-dependent material properties.
The calculated temperatures can then be applied to a structural model to assess thermal expansion, distortion and stresses caused by restrained movement or uneven heating.
A useful study goes beyond choosing a solver
The analysis type is only part of the specification. A credible study also considers which operating conditions, uncertainties and design variations need to be investigated.
This may include:
- Load cases and combinations: different filling levels, lifting conditions, stacking arrangements, pressure, temperature and support conditions.
- Sensitivity studies: how results change with wall thickness, material stiffness, joint flexibility, friction or manufacturing tolerances.
- Mesh convergence studies: whether further mesh refinement materially changes the engineering quantities being assessed.
- Design comparisons: whether changes to geometry, material or reinforcement improve performance.
- Verification and validation: checking equilibrium and simplified calculations, then comparing predictions with physical testing where appropriate.
A polymer container, for example, may need separate assessments of immediate deformation, long-term creep, local contact at its supports and stability under a particular loading condition. One stress plot cannot answer all of those questions.
At Rota Design, the starting point is the engineering question: what must the product withstand, how could it fail, and what evidence is needed to support the design decision?
The value of FEA lies in choosing and checking an analysis that represents the behaviour that matters.
Contact Rota Design to discuss the analysis approach your product needs.