From an ultimate strength standpoint, research to date on tubular joints has concentrated on testing work. However, the high costs associated with testing has directed recent efforts towards non-linear analysis techniques to generate data and information. This paper gives a background to these efforts and describes the current status. The manner of application of numerical techniques is described and discussed by reference to various separate investigations which have been carried out, or are currently underway. New strength data generated using the non-linear finite-element program FINAS are presented and discussed. These data, together with others generated elsewhere, are compared where appropriate with existing experimental data and API RP2A recommendations. Areas where future effort needs to be directed are identified and presented.
The non-linear techniques (and the results) discussed herein demonstrate their applicability to generate, new information in a much more cost-effective manner than the high-cost and difficult experimentation that has been undertaken to date. Further, the flexibility offered by such techniques can be applied in practical areas where experimentation becomes prohibitively expensive and difficult, ego for complex joints and joints with realistic load scenarios (hydrostatic loading for example). Notwithstanding this aspect, it is recognized that non-linear numerical techniques represent a developing technology, and the need for experimental data will remain for a variety of reasons, not 1east the need to have relevant test results to assist in the all-important calibration of the numerical techniques.
Most offshore structures are three-dimensional steel frames composed of cylindrical steel members. These give the best compromise in satisfying the requirements of low drag coefficient, high buoyancy, high strength-to weight ratio, and equal bending strength in all directions. The space inside the hollow sections can be used for transport, storage or to obtain additional strength. Tubular joints, which are formed by welding the profiled end of one tubular onto the undisturbed outside of the other, continue to remain a source of difficulty and high cost in the design, construction and maintenance of steel offshore structures.
A number of 'codes and guidance documents provide recommendations which relate to the design and construction of tubular joints. In the case of static strength, which is the subject of this paper, these recommendations are derived from interpretations of test data. In recent times, there has been an increasing awareness that the use of non-linear numerical techniques potentially represents a much more economic tool to generate data on static strength. The need for further knowledge in this area remains high as the majority of the available data (and hence guidance) only cover simple joints subjected to simplistic unidirectional brace loading.
This paper concentrates on recent numerical efforts in the area of ultimate strength of tubular joints. Previous investigations are described and discussed. Characteristics of the non-linear computer program FINAS are described, and the program is used to generate new data and stiffness curves on axially loaded T, DT and K joints.
