December 6, 2019

Avoiding Fatigue in Welded Joints

Welding is a common method of joining metal objects together, but welded…
Avatar Clare Scott

Welding is a common method of joining metal objects together, but welded joints, though strong, are susceptible to fatigue. In fact, fatigue is the most common cause of failure in welded structures and components, responsible for over 90 percent of failures. This is no small matter, as fatigue-related failure can result in lost production, high repair cost, and even, in some cases, loss of life, if the failed component is a critical one.

There are multiple reasons why welded joints are prone to fatigue; for example, small defects can initiate cracks, which can then grow and lead to failure. To avoid costly and/or dangerous joint failure, engineers must calculate the fatigue life, accounting for the different possible issues. Several methods exist to do this, and most of them can be implemented using finite element analysis (FEA).

SIMULIA offers a product called fe-safe, which includes two methods of predicting fatigue in welded joints. The first method, British Standard 5400/7608, is based on a nominal stress approach and is offered in fe-safe for no additional cost. With this method, local FEA stresses in complex models can be sensitive to mesh density, element formulation, and solver. Subjective decisions make this method more difficult to apply in some cases.

In this method, a weld classification is selected from the standard that has weld details organized into tables. The standard contains a simple welded specimen of defined shape and size, direction of loading with respect to the weld and location of cracking. For real life structures that are much more complex, selecting a classification can be subjective, and selections can vary from one analyst to another. Additionally, it can be difficult to find the appropriate classification for the joint without access to the standard.

In British Standard 5400/7608 method in fe-safe, coefficients for stress-life curves are provided for each classification letter. Stresses are calculated in FEA, measured using a strain gauge, or calculated in a hand calculation. Selecting the value of nominal stress to use is, again, a subjective decision.

The second method is the Verity module, an add-on module for fe-safe that requires additional licensing. Verity is an equivalent structural stress method that requires no subjective decision-making. It is also insensitive to mesh density, loading direction, and thickness. Verity is based on the Equivalent Structural Stress calculated from nodal forces.

In the Verity method, there is no need for weld classification. Its data has been validated against more than 3500 welded joint tests. It offers consistency in stress determination based on nodal forces at the weld, and it is robust for complex structures.

This mesh insensitive structural stress method can be solved using simultaneous equations. Structural stress is calculated by fe-safe before fatigue analysis is performed with Verity.

Welded joints are major components in many applications, including bridges, pressure vessels, tractors, vehicles, ships and more. Preventing their failure due to fatigue is one of the most important parts of engineering these joints. To learn more about welded joints, fatigue, and the role of fe-safe in preventing joint failure, check out the webinar “Fatigue Analysis of Welded Joints,” available here.

Stay up to date

Receive monthly updates on content you won’t want to miss


Register here to receive a monthly update on our newest content.