ABSTRACT: In this paper the dynamic behaviour of a pipe in a resonant bending fatigue test rig is investigated. This kind of test rig is used for full scale testing of pipeline joints such as girth welds and threaded connections. In order to optimize the test conditions in such a test rig, accurate knowledge of its dynamic behaviour is necessary. This dynamic behaviour is first evaluated theoretically using a semi-analytical model to study the effects of pipe dimensions and test rig properties. Afterwards, the validity of this model is illustrated by experimental test results. To measure the bending deflection of the pipe during a test, a dynamic 3D optical measurement system is used. Using the measurements it is shown that the dynamic behaviour of a pipe in the resonant bending test rig can be accurately predicted by the theoretical model. For a given pipe diameter and wall thickness size, the model prescribes the required pipe length, the necessity to fill the pipe with water, the necessary mass of the endweights, the systems natural frequency and the position of the test rig supports. KEY WORDS: Fatigue; Resonant bending; Pipeline; Experimental Testing; Optical measurement. 1 INTRODUCTION In offshore applications pipes and tubular members are used as structural elements, in transport pipelines, risers etc. In all these applications the pipe joints are generally the weakest parts. A wide variety of pipe joining techniques is commonly used, going from welding to mechanical joints with bolted flanges or threaded couplings. When subjected to dynamic loads, weld defects or geometrical stress raisers can initiate fatigue cracks causing the columns or pipelines to fail prematurely. Due to scale effects, full scale testing is in many cases the most reliable method to characterize the fatigue strength of a pipe joint, e.g. if the effect of misalignment [1] or weld defect size [2] of girth welds is investigated. For other pipe joining techniques such as threaded connections for risers [3], casing [4], structural columns [5] and drill pipes [6-7] full scale tests are the only option. For this reason the Laboratory Soete recently built a resonant bending fatigue test rig, suitable for testing pipes within a wide diameter range [8]. In this paper the dynamic behaviour of a pipe in the test rig is investigated more in detail. This is done theoretically by using a semi-analytical model and experimentally by measurements obtained with a dynamic 3D optical measurement system. 2 TEST RIG DESCRIPTION 2.1 Test rig working principle To apply a fatigue load to a pipe in a resonant bending test rig, a dynamic excitation force is applied with a frequency close to the natural frequency of the pipe. This causes the pipe to come into resonance, which means that it will deform according to a standing wave that rotates at the excitation frequency. When the pipe is supported in the nodes of this wave, the supporting framework will not be subjected to high dynamic forces. The pipe’s natural frequency depends mainly on its mass and bending stiffness, the higher the weight of the pipe, the lower its natural frequency. In order to limit its value the pipe’s natural frequency is lowered by filling it with water and attaching endweights at both ends of the pipe. 2.2 Test rig overview A global overview of the test rig is shown in Figure 1 together with a picture of the completed test rig and the dynamic 3D optical measurement system in Figure 2. The test pipe (1) that should contain a central pipe joint, which can be a threaded connection as well as a girth weld is excitated by the drive unit (2). It contains two excenters that are rotated at a speed close to the natural frequency of the pipe to apply the excitation force. Their relative position can be changed in order to control the resulting eccentricity which can be varied from 0 to 100 %. 4 3 2 6 6 1 7 5 8 4 3 2 6 6 1 7 5 8 Figure 1. Overview of the resonant bending fatigue test rig. Dynamic 3D optical measurement system for the characterisation of the behaviour of a pipe in a resonant bending test rig J. Van Wittenberghe 1 , P. De Baets 1 , W. De Waele 1 , T.T. Bui 2 , G. De Roeck 2 1 Laboratory Soete, Ghent University, Sint-Pietersnieuwstraat 41, B-9000 Gent, Belgium 2 Department of Civil Engineering, K.U.Leuven, Kasteelpark Arenberg 40, B-3001 Leuven, Belgium email: Jeroen.VanWittenberghe@UGent.be, Patrick.DeBaets@UGent.be, Wim.DeWaele@UGent.be, TienThanh.Bui@bwk.kuleuven.be, Guido.DeRoeck@bwk.kuleuven.be Proceedings of the 8th International Conference on Structural Dynamics, EURODYN 2011 Leuven, Belgium, 4-6 July 2011 G. De Roeck, G. Degrande, G. Lombaert, G. M¨ uller (eds.) ISBN 978-90-760-1931-4 3554