CIDECT 1A: Welding in cold- formed areas of rectangular hollow sections

 1. Aim of the investigation



1.1 Evaluation of the effect of welding in cold-formed areas with regard to brittle fracture

On the basis of tests on angle sections described in [15, 22, 30, 31, 32, 33, 34] further investigations on the influence of welding in cold-formed areas require to be carried out.

In the course of the present research programme only the influence of strain ageing beside the weld seam is to be investigated. Notch effects and geometrical influences such as single-layer welding are to be excluded or minimized in order to prevent any dominating influence of such conditions on the test process. A multi-layer weld would e.g. cause a normalizing effect, that would cover the ageing. On this basis it is planned to remove the bead-on-plate weld, afterwards, so that only the influence of the thermal effect of the weld bead in the cold-formed area of the specimen – the strain ageing caused by welding – remains as an influencing parameter. The bead-on-plate weld is only regarded as an “expedient” for the generation of the ageing area typical for a welding seam.

The heat input caused by welding with practice-relevant welding parameters corresponds partially to the conditions of artificial ageing, which leads to critical conditions under service load by increased tendency to embrittlement through the preceding cold-forming (see “Stahlbau-Handbuch”, Part 1, Page 346 – Stahlbau-Verlags-GmbH, Köln 1982).

The present research programme will investigate the shift of the transition temperature caused by welding (ageing effects) using Charpy V notch impact bend specimens and angle sections similar to real components. For the angle sections, the condition “removed weld seam” is mainly to be checked. A real weld would not lead to a stricter condition for the determination of the ageing effect. Another part of the test specimens will be comparatively tested in the conditions “as delivered” and “artificially aged”.

The tests are meant to clarify, whether the region of strain ageing is a preferred location of cracks. The loading is to be chosen in such a way, that in the longitudinal direction of the cold-formed curvature, the ageing zone beside the weld seam is included. This whole area is subjected to high tensile stresses during the impact bend loading.

This research programme is related to the BAM/AVIF-Project A60 (S24/8/91). In this project a low-alloyed high-strength fine-grained structural steel (QStE 380TM), used for instance in vehicle construction, was investigated for the first time.

Because of the increasing influence of weldable high-strength steels in the steel manufacturing industry, an extension of the investigation on 4 other grades will be carried out.


1.2 Evaluation of the effects of welding in cold-formed areas with regard to the alternating fatigue strength

Welding in cold-formed areas is not only an open question for statically loaded sections (brittle fracture) but also of dynamically loaded structures (fatigue resistance).

Especially for structures made of hollow sections with large b-ratios, it is often necessary to weld in the areas close to the edges. A lot of these structures are subjected to dynamic loadings. Because there are nearly no information given about the influence of the weld in cold-formed areas on the fatigue resistance, welding in this areas is not acceptable.

Therefore, systematically gained knowledge about the fatigue behaviour of such structures is of great importance. First tests on this problem have been carried out at the University of Karlsruhe in the 1980’s. During these tests, flattened CHS braces have been welded to the corners of cold-formed RHS chords, forming a K-joint. The results led to the conclusion, that in case of high material quality, welding in cold-formed areas does not necessarily have a bad influence on the fatigue resistance. In the scope of this research project, the fatigue behaviour of welded (notched) and non-welded (unnotched) test specimens made of different steel grades and wall thicknesses will be compared directly with each other. The tests will be carried out using 4-point-bending. Statistical analysis will be carried out to determine, whether a weld notch will influence the fatigue behaviour in cold-formed regions of hollow sections and if so, to what extend.