Bertil Jonsson

Fitness for Service

Assessment of welds not meeting “standard” requirements may sometimes be of interest to investigate. An example is to determine the fatigue life from a found crack-like defect to failure. It is recommended to use local-based methods, such as the effective notch method or the fracture mechanics method with the guidance of well-established recommendations as, e.g., BS 7910 or comparable ones.

Development of Weld Quality Criteria Based on Fatigue Performance

Historically, production technology researchers and structural design researchers have had only limited dialogue and each group has focused on their own narrow field of interest. This has led to inconsistencies in the definition of so-called “weld class systems” which have been primarily developed based on concepts related to good workmanship but have little or no relation to the actual performance of the welded structure. Additionally, some of the quality measures in existing systems are qualitative and, thus, subjective. In recent years, Commission XIII of the International Institute of Welding (IIW) has been promoting research and developing the technical background needed to develop a weld quality guideline which quantitatively relates weld acceptance criteria to the expected structural performance (primarily fatigue strength). A new weld class system with this same objective has recently been developed as a Volvo Group Standard. This system is described in this paper. The new standard has three quality levels for fatigue; as-welded normal quality, as-welded high quality and post-weld treated quality. It contains acceptance limits which are consistent with the expected fatigue strength and which can more objectively handle revisions. This new system will help in the development of new structures with lower weight and increased reliability.

Fatigue Assessment and Lefm Analysis of Cruciform Joints Fabricated with Different Welding Processes

In this study fatigue testing and defect assessment were carried out on specimens welded with robotic and manual welding using flux cored (FCAW) and metal cored (MCAW) filler materials in order to study the effect of the welding method on the fatigue strength and weld quality. Thirteen different batches were investigated of which two was shot peened before fatigue testing. The local weld geometry was measured for all the specimens before testing. The specimens welded with flux cored weld wire showed the best fatigue strength, small defects and low residual stresses. Large scatter in the fatigue data is observed, especially when manual welding is employed. The few largest defects were removed by the shot peening process, although small defects survived. This led to a smaller scatter in fatigue live for the shot peened specimens. Linear elastic fracture mechanics, LEFM, was employed for analysis of the fatigue test results. The fatigue life predictions using a 2D LEFM FE-model for simulating a continuous cold lap defect along the weld toe showed a qualitative agreement with the fatigue test results. The 2D analysis showed that a continuous cold lap defect should be no more than 0.5 mm deep in order to comply with the requirement of fatigue lives for normal weld quality according to the IIW design rules. For larger defects (> 0.8 mm) an increased toe radius will have a small effect on the fatigue strength. A 3D LEFM analysis of crack growth from a spatter-induced cold lap defect was also carried out. This showed similar trends in crack growth compared to the 2D analysis of a continuous cold lap, although the spatter-induced cold lap defect (semi-elliptical) had a longer fatigue life (x2.7), and hence is less dangerous from a fatigue point of view.

Fatigue design of lightweight welded vehicle structures: influence of material and production procedures

Structural details and components in many types of products are continuously subjected to variable amplitude loading during operation. Fatigue loading and fatigue damage is thus the most common failure mode for the mentioned equipment in operation. The influence of the material grade, weld quality and fabrication procedure have a major impact on the structural durability of welded vehicle structures. The comprehensive research work within the Nordic research and development community have contained development of finite element modelling of complex structures, including crack growth in two- and three-dimensional fatigue testing of welded small-scale specimen and full-scale components, investigations of weld defects and flaws (e.g. cold laps) and weld roots. An important part of these projects is related to simulation and measurements of formation and relaxation of residual stresses. Within these projects three new quality systems, for welded and cast components and for cut edges, have been developed based on a scientific ground and a fitness for purpose design philosophy. Volvos new weld class system, which is an open standard, is now a base for the revision of the international weld quality system ISO 5817. In this article the major findings in these research activities are briefly presented and discussed.

Welding procedures for fatigue life improvement of the weld toe

This paper presents the results of an experimental study of gas metal arc welded, GMAW, cruciform joints made of common construction steel S355. The hypothesis is that smooth undercuts in as welded conditions can give enhanced fatigue properties similarly as post treated welds. Undercuts are generally seen as a defect or imperfection. Welders try to avoid these and repair them when they occur, which result in increased production lead time. Post weld improvement methods i.e. grinding or high-frequency-impact treatment (HFMI) as fatigue-enhancing post-treatment methods enforce amongst other effects a certain smooth undercut-shaped groove in the treated weld toe region. The obtained shallower weld toe transition reduces the geometrical notch effect and increases fatigue strength. This paper presents a study whereas welded specimens with a weld toe geometry similar to what is obtained by weld toe grinding or HFMI-treatment, has been produced, fatigue tested and analyzed. The improvement of the fatigue strength is comparable to post-weld treated specimens. It has proven to be an efficient way to achieve high-quality welds without introducing any additional operations in production, thus enabling weight reduction using cost-effective methods.

Root Side Requirements

When designing a weld, it is often desirable that the root side has a greater or at least the same fatigue life as the toe side.

Design for Purpose

Typical fabricated structures may have hundreds or even thousands of meters of weld. Thus, many potential fatigue cracking locations are present that must be considered during design development and production.

Weld Quality Levels for Fatigue Loaded Structures

The imperfections and their classification into quality groups are mostly done by the guidance of introduced codes. One standard for weld quality is ISO 5817. This standard is an adoption of the old German standard DIN 8563, which was established as a standard for communication between the welders and the inspectors. The classification criterion was the difficulties, the expenses, or the efforts to fabricate or to inspect by NDT. So by the nature, ISO 5817 has limits in direct application to fatigue problems, it is inconsistent with respect to fatigue properties and needs application guidance. Most dedicated design codes specify a general quality level according to ISO 5817 and give additional regulations. In this situation, the IIW fatigue design recommendations have extended the scope of usual fatigue design codes by describing the fatigue properties of joints containing weld imperfections on a scientific basis.

Classification of Weld Imperfections and Features

The designation and the classification of weld imperfections and features depend on both the material being joined, e.g., steel or aluminum, and the joining process, e.g., fusion welding and pressure welding, etc.

IIW Fatigue Assessment Procedures

Numerous fatigue assessment methods have been introduced to assess the durability of metal structures under dynamic loading. Finite element (FE) modeling is an integral part of most design and analysis work, and methods have evolved as the analysis possibilities have become more sophisticated and computers have increased in speed and memory capacity. Fatigue assessment places two conflicting demands on the analysts. The fatigue damage process itself is highly local, thus requiring a fine FE mesh. On the other hand, welded structures are frequently large and geometrically complex, they have numerous load input locations, and they have boundary conditions which may be difficult to define. These demands are best satisfied with a large FE model. Because of this conflict, fatigue assessment is frequently the slowest link in the design process of welded structures.