• 대한기계학회논문집A
• /
• 제24권5호
• /
• pp.1203-1214
• /
• 2000

In this paper, a failure model for co-cured steel-composite tubular single lap joints has been proposed incorporating the nonlinear mechanical behavior of steel adherends and different failure mode s such as steel adherend failure and composite adherend failure. The characteristics of the co-cured steel-composite tubular single lap joint were investigated with respect to the test temperature, the stacking sequence of composite adherend, the thickness ratio of steel adherend to composite adherend, and the scarf ratio of steel adherend. Thus, the optimum design method for the co-cured steel-composite tubular single lap joint was suggested.

• Steel and Composite Structures
• /
• 제24권4호
• /
• pp.425-439
• /
• 2017

As a thin-walled structure, local joint flexibility (LJF) in a tubular structure is prominent, and it may produce significant effect on the static performance for the overall structure. This study presents a simplified analytical model to analyze the static behavior for a steel tubular structure with LJF. The presented model simplifies a tubular structure into a frame model consisted of beam elements with considering the LJFs at the connections between any two elements. Theoretical equations of the simplified analytical model are deduced. Through comparison with 3-D finite element results of two typical planar tubular structures consisted of T- and Y-joints respectively, the presented method is proved to be accurate. Furthermore, the effect of LJF on the overall performance of the two tubular structures (including the deflection and the internal forces) is also investigated, and it is found from analyses of internal forces and deformation that a rigid connection assumption in a frame model by using beam elements in finite element analysis can provide unsafe and inaccurate estimation.

• 국제강구조저널
• /
• 제18권5호
• /
• pp.1525-1540
• /
• 2018

A test rig with multi-functional purposes was specifically designed and manufactured to study the behavior of multi-planar welded tubular joints subjected to multi-planar concurrent axial loading. An experimental investigation was conducted on full-scale welded tubular joints with each consisting of one chord and eight braces under monotonic loading conditions. Two pairs or four representative specimens (two specimens for each joint type) were tested, in which each pair was reinforced with two kinds of different internal stiffeners at the intersections between the chords using welded rectangular hollow steel sections (RHSSs) and the braces using rolled circular hollow steel sections (CHSSs) and welded RHSSs. The effects of different internal stiffeners at the chord-brace intersection on the load capacity of joints under concurrent multi-planar axial compression/tension are discussed. The test results of joint strengths, failure modes, and load-stress curves are presented. Finite element analyses were performed to verify the experimental results. The study results show that the two different joint types with the internal stiffeners at the chord-brace intersection under axial compression/tension significantly increase the corresponding ultimate strength to far exceed the usual design strength. The load carrying capacity of welded tubular joints decreases with a higher degree of the manufacturing imperfection in individual braces at the tubular joints. Furthermore, the interaction effect of the concurrent axial loading applied at the welded tubular joint on member stress is apparent.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2006년도 정기 학술대회 논문집
• /
• pp.655-663
• /
• 2006

Large-scale model tests of welded tubular K-joints were carried out to observe the fatigue behavior of API 2W Gr.50 steel produced by POSCO. The fatigue crack behaviors for various loading conditions were measured and investigated around the critical joint sections. The experimental results have been verified with numerical approaches and also compared with the IIW, DnV RP-C203 and API RP 2A-WSD design curves. The hot spot strss method was applied in the study. The SCF factor for tubular K-joint was also obtained.

• Structural Engineering and Mechanics
• /
• 제10권2호
• /
• pp.125-138
• /
• 2000

The paper analyzes the problem of torsion in an adhesive non-tubular bonded single-lap joint. The joint considered consists of two thin rectangular section beams bonded together along a side surface. Assuming the materials involved to be governed by linear elastic laws, equilibrium and compatibility equations were used to arrive at an integro-differential relation whose solution makes it possible to determine torsional moment section by section in the bonded joint between the two beams. This is then used to determine the predominant stress and strain field at the beam-adhesive interface (stress field along the direction perpendicular to the interface plane, equivalent to the applied torsional moment and the corresponding strain field) and the joint's elastic strain (absolute and relative rotations of the bonded beam cross sections). All the relations presented were obtained in closed form. Results obtained theoretically are compared with those given by a three dimensional finite element numerical model. Theoretical and numerical analysis agree satisfactorily.

• 한국해양공학회지
• /
• 제18권6호
• /
• pp.70-78
• /
• 2004

In order to increase the load carrying capacity of tubular structures, the joints of tubular members are usually reinforced with various reinforcement system. A stiffening method with internal ring stiffeners is effectively used for the steel tubular joint with a large diameter. In this study, the behavior of internally ring-stiffened tubular T-joints subjected to axial loading is assessed. For the parametric study, nonlinear finite element analyses are used to compute the static strength on non-stiffened and ring-stiffened T-joints. Based on the numerical results, an internal ring stiffener is found to be efficient in improving the static strength. The influence of geometric parameters has been determined, and the reinforcement effect are evaluated. Based on the FE results, regression analises are performed considering the practical size of ring stiffener. Finally strength estimation formulas for ring-stiffened tubular T-joints are proposed.

• Steel and Composite Structures
• /
• 제21권5호
• /
• pp.1017-1029
• /
• 2016

When a welded circular hollow section (CHS) tubular joint is subjected to brace axial loading, failure position is located usually at the weld toe on the chord surface due to the weak flexural stiffness of the thin-walled chord. The failure mode is local yielding or buckling in most cases for a tubular joint subjected to axial load at the brace end. Especially when a cyclic axial load is applied, fracture failure at the weld toe may occur because both high stress concentration and welding residual stress along the brace/chord intersection cause the material in this region to become brittle. To improve the ductility as well as to increase the static strength, a tubular joint can be reinforced by increasing the chord thickness locally near the brace/chord intersection. Both experimental investigation and finite element analysis have been carried out to study the hysteretic behaviour of the reinforced tubular joint. In the experimental study, the hysteretic performance of two full-scale circular tubular T-joints subjected to cyclic load in the axial direction of the brace was investigated. The two specimens include a reinforced specimen by increasing the wall thickness of the chord locally at the brace/chord intersection and a corresponding un-reinforced specimen. The hysteretic loops are obtained from the measured load-displacement curves. Based on the hysteretic curves, it is found that the reinforced specimen is more ductile than the un-reinforced one because no fracture failure is observed after experiencing similar loading cycles. The area enclosed by the hysteretic curves of the reinforced specimen is much bigger, which shows that more energy can be dissipated by the reinforced specimen to indicate the advantage of the reinforcing method in resisting seismic action. Additionally, finite element analysis is carried out to study the effect of the thickness and the length of the reinforced chord segment on the hysteretic behaviour of CHS tubular T-joints. The optimized reinforcing method is recommended for design purposes.

• 한국해양공학회지
• /
• 제17권2호
• /
• pp.34-39
• /
• 2003

Recent deepwater offshore structures in the Gulf of Mexico utilize butt welded tubular joints. Application of a welded tubular joint includes tendons, production risers, and steel catenary risers. Fatigue life assessment of these joints becomes more critical, as the structures to which they are attached are allowed to undergo cyclic and sometimes large displacements around an anchored position. Estimation of the fatigue behavior of these tubular members in the design stage is generally condrcted by using S-N curves, as specified in the codeds and standards. Applying the stress concentration factor of the welded structure to the S-N approach often results in a very conservative assessment, because the stress field acting on the tubular has a non-uniform distribution through the thickness. Fatigue life analysis using fracture mechanics has been applied in the design of the catenary risers. This technology enables the engineer to establish proper requirements on weld quality and inspection acceptance criteria to assure satisfactory structural integrity during its design life. It also provides guidance on proper design curves and a methodology for accounting for the effects of non-uniform stress distribution through the wall thickness. Still, there is inconsistency when designing tubular joints using a conventional S-N approach and when specifying weld flaw acceptance criteria using fracture mechanics approach. This study developed fatigue curves that are consistent with both the S-N approach and the fracture mechanics approach. Accounting for non-uniform stress distribution and threshold stress intensity factor were key parameters in relating both approaches. A series of S-N curves, generated from the fracture mechanics approach, were compared to the existing S-N curves. For flat plate butt joint, the S-N curve generated from fracture mechanics matches with the IIW class 100 curve when initial crack depth was 0.5 mm (0.02 ). The new curves for tubular joint agree very well with the experimental results. The comparison also indicated the degree of conservatism built into the API X design curve.

• International Journal of Ocean Engineering and Technology Speciallssue:Selected Papers
• /
• 제6권1호
• /
• pp.69-74
• /
• 2003

Recent deepwater offshore structures in the Gulf of Mexico utilize butt welded tubular joints. Application of a welded tubular joint includes tendons, production risers, and steel catenary risers. Fatigue life assessment of these joints becomes more critical, as the structures to which they are attached are allowed to undergo cyclic and sometimes large displacements around an anchored position. Estimation of the fatigue behavior of these tubular members in the design stage is generally conducted by using S-N curves, as specified in the codes and standards. Applying the stress concentration factor of the welded structure to the S-N approach often results in a very conservative assessment, because the stress field acting on the tubular has a non-uniform distribution through the thickness. Fatigue life analysis using fracture mechanics has been applied in the design of the catenary risers. This technology enables the engineer to establish proper requirements on weld quality and inspection acceptance criteria to assure satisfactory structural integrity during its design life. It also provides guidance on proper design curves and a methodology for accounting for the effects of non-uniform stress distribution through the wall thickness. Still, there is inconsistency when designing tubular joints using a conventional S-N approach and when specifying weld flaw acceptance criteria using fracture mechanics approach. This study developed fatigue curves that are consistent with both the S-N approach and the fracture mechanics approach. Accounting for non-uniform stress distribution and threshold stress intensity factor were key parameters in relating both approaches. A series of S-N curves, generated from the fracture mechanics approach, were compared to the existing S-N curves. For flat plate butt joint, the S-N curve generated from fracture mechanics matches with the IIW class 100 curve when initial crack depth was 0.5 mm (0.02). The new curves for tubular joint agree very well with the experimental results. The comparison also indicated the degree of conservatism built into the API X design curve.

• Steel and Composite Structures
• /
• 제21권2호
• /
• pp.323-342
• /
• 2016

To study the effect of collar-plate reinforcement on the static strength of tubular T-joints under axial loading, fundamental research work is carried out from both experimental test and finite element (FE) simulation. Through experimental tests on 7 collar-plate reinforced and 7 corresponding un-reinforced tubular T-joints under axial loading, the reinforcing efficiency is investigated. Thereafter, the static strengths of the above 14 models are analyzed by using FE method, and it is found that the numerical results agree reasonably well with the experimental data to prove the accuracy of the presented FE model. Additionally, a parametric study is conducted to analyze the effect of some geometrical parameters, i.e., the brace-to-chord diameter ratio ${\beta}$, the chord diameter-to-chord wall thickness ratio $2{\gamma}$, collar-plate thickness to chord wall thickness ratio ${\tau}_c$, and collar-plate length to brace diameter ratio $l_c/d_1$, on the static strength of a tubular T-joint. The parametric study shows that the static strength can be greatly improved by increasing the collar-plate thickness to chord wall thickness ratio ${\tau}_c$ and the collar-plate length to brace diameter ratio $l_c/d_1$. Based on the numerical results, parametric equations are obtained from curving fitting technique to estimate the static strength of a tubular T-joint with collar-plate reinforcement under axial loading, and the accuracy of these equations is also evaluated from error analysis.