Residual Stress Analysis of Welded Structures

Welding is the backbone of modern manufacturing, enabling the assembly of large‑scale and highly complex structures. Yet the intense, localized heating and rapid cooling inherent to most welding processes generate residual stresses—self‑equilibrating internal stresses that remain after the joint has cooled to ambient temperature. If left unchecked, these stresses can drive distortion, promote stress‑corrosion cracking, shorten fatigue life, or even trigger brittle fracture. Consequently, accurate prediction, measurement, and control of welding‑induced residual stress are essential tasks for any solid‑mechanics engineer involved in design, certification, or life‑extension of welded components.

How Residual Stresses Develop During Welding

The origin of welding residual stress lies in the incompatibility between thermal expansion of the heated zone and the mechanical restraint offered by the surrounding cooler material. The evolution can be divided into four overlapping stages:

1. Localized heating – The welding heat source (arc, laser, electron beam, etc.) raises the temperature of the weld pool and the adjacent heat‑affected zone (HAZ) to several hundred or even thousands of degrees Celsius, causing pronounced thermal expansion.

2. Mechanical constraint – The bulk of the workpiece, still at near‑room temperature, possesses a much higher stiffness. This surrounding “cold” material limits the free expansion of the hot region, imposing compressive strains on the heated material.

3. Cooling and solidification – As the torch moves on, the molten metal solidifies and the temperature drops. The solidified metal now attempts to contract, but the surrounding material resists this shrinkage. The result is a tensile residual stress in the weld centre and a compressive residual stress at the outer edges of the HAZ.

4. Plastic strain accumulation – During the heating‑cooling cycle the material transitions from elastic to plastic behaviour. When the temperature finally returns to ambient, the irreversible plastic strain field is locked in as a macroscopic residual‑stress field.

Understanding these mechanisms is the first step toward building reliable predictive models and selecting appropriate mitigation techniques.

Experimental Determination of Residual Stress

Because residual stresses are internal and often hidden, engineers rely on a suite of experimental methods that fall into two broad categories.

Non‑Destructive Techniques

These techniques are indispensable for validating numerical models and for routine inspection of critical components where destructive testing is impractical.

Destructive Techniques

Choosing the right method depends on the required depth, accuracy, accessibility, and budget.

Numerical Simulation of Welding Residual Stress

Finite‑element analysis (FEA) has become the workhorse for predicting residual stresses in full‑scale welded structures, where experimental campaigns would be prohibitive. A typical thermo‑mechanical workflow consists of two tightly coupled stages.

1. Thermal Analysis

• Heat‑source representation – The moving heat source is often idealised using the Goldak double‑ellipsoid model, which captures the asymmetric distribution of heat in arc welding.
• Boundary conditions – Convective heat loss to the surrounding gas and radiative emission from hot surfaces are applied, together with conduction into the bulk material.
• Output – The solution yields a transient temperature field (T(x,y,z,t)) that serves as the driving input for the subsequent mechanical step.

2. Mechanical Analysis

• Temperature‑dependent material model – Elastic modulus, yield stress, thermal expansion coefficient (\alpha(T)), and hardening behaviour must be defined as functions of temperature because the weld experiences a wide thermal span.
• Strain decomposition

[
\varepsilon_{\text{total}} = \varepsilon_e + \varepsilon_p + \varepsilon_{\text{th}}
]

where (\varepsilon_{\text{th}} = \alpha(T),\Delta T).

• Plasticity criterion – A von Mises yield surface with isotropic hardening (or more advanced models such as the Johnson‑Cook law for high‑temperature deformation) is commonly employed.
• Coupling – The temperature increments from the thermal step are converted into thermal strains, and the equilibrium equations are solved to obtain the stress field (\sigma(x,y,z)) at each time increment.

The final residual‑stress state is extracted after the model has cooled to ambient temperature.

Case Study: Residual Stress in a 10 mm Thick Steel Lap Joint

To illustrate the workflow, consider a single‑lap joint fabricated from 10 mm thick Q235 steel plates.

Modelling Setup

1. Geometry – Two overlapping plates are modelled, with a weld bead defined along the overlap region.
2. Material data – Temperature‑dependent properties are supplied, e.g., yield strength decreasing from 245 MPa at room temperature to zero at the melting point.
3. Heat source – Goldak parameters (a, b, c) are calibrated against measured bead geometry and welding current/voltage.
4. Mesh strategy – A fine mesh (≈ 0.5 mm) is placed in the weld and HAZ, while a coarser mesh (≈ 5 mm) is used farther away to keep computational cost reasonable.

Results Overview

• Temperature distribution – The peak temperature at the centre of the weld pool reaches the melting point, then decays exponentially outward, forming a classic “heat‑affected zone” of roughly 3–4 mm width.
• Residual stress pattern –
  • Tensile stresses dominate along the weld centreline, often approaching the material’s yield strength.
  • Compressive stresses appear at the outer edge of the HAZ, balancing the tensile core.
  • Out‑of‑plane warpage is observed because the lap joint is geometrically asymmetric; the simulated deformation matches measured distortion in physical specimens.

These insights guide designers in selecting weld sequences, pre‑heat temperatures, or post‑weld treatments to keep stresses within acceptable limits.

Strategies for Mitigating Welding Residual Stress

When analysis reveals stress levels that threaten service life, engineers can intervene through several proven techniques:

• Post‑weld heat treatment (PWHT) – Uniform heating to a prescribed temperature followed by controlled cooling encourages recovery and recrystallisation, thereby relaxing the internal stress field.
• Mechanical surface treatments –
  • Shot peening introduces a beneficial compressive layer that counteracts tensile stresses at the surface, improving fatigue resistance.
  • Laser shock peening offers deeper compressive zones with minimal surface roughness.
• Process optimisation –
  • Symmetric welding sequences (e.g., welding from both sides in a balanced order) reduce net distortion.
  • Pre‑heating the base material lowers the temperature gradient, limiting the magnitude of induced stresses.
  • Controlling heat input (adjusting current, voltage, travel speed) directly influences the size of the HAZ and the resulting stress distribution.

A combination of these measures, selected based on the specific geometry, material, and service conditions, can bring residual stresses to a safe level while preserving structural integrity.

Concluding Remarks

Residual stresses are an inevitable by‑product of welding, but they need not be a fatal flaw. By grasping the physical mechanisms of stress generation, employing reliable measurement techniques, and leveraging advanced thermo‑mechanical finite‑element simulations, engineers can predict where and how large stresses will develop. Coupled with targeted mitigation strategies—thermal, mechanical, or procedural—this knowledge enables the design of welded structures that meet stringent performance and safety requirements. In the era of high‑performance alloys and complex additive‑manufacturing‑inspired welds, a rigorous residual‑stress workflow is no longer optional; it is a cornerstone of modern structural engineering.