• Proceedings of the KSME Conference
• /
• 2004.04a
• /
• pp.108-113
• /
• 2004

According to our fatigue tests carried out at 20 Hz, R=0.1 on transversely butt~welded joints, fatigue strengths of as-welded specimens, that is, specimens having residual stress are higher than those of annealed specimens in short life range, but vice verse in long life range. This behavior seems to be concerned mainly with residual stress relaxation by applied loading. After analyzing the welding process, we conducted finite element analysis to quantify the degree of residual stress relaxation. By taking into account residual stress relaxation, modified Goodman diagram, and nominal stress, we evaluated the fatigue life of the welded joint from the S-N curve for the parent material. The estimated results are in a good agreement with the experimental results.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.3 s.70
• /
• pp.387-396
• /
• 2004

In this paper, stress analysis anda 3-point bending fatigue test were conducted on the full-scale specimen to investigate the effects of misalignment at the U-rib welded joint due to misfitting in the steel deck bridge. In addition, the researchers investigated the direction and starting point of fatigue cracks by SEM (Scanning Electron Microscope) and beach mark. The results of the stress analysis show that maximum stress occurred at the bottom corner of the U-rib, and that the stress was large when the magnitude of the misalignment was large. On the other hand, the results of the static loading test of the full-scale specimen show that stress was large at the bottom corner of the U-rib. In addition, fatigue life was short when the misalignment was large and fatigue life was short when the misalignment was large and fatigue life was short when the misalignment was large and fatigue life was large when the misalignment was small, as indicated by the results of both the static loading test and the fatigue test. From the observation of the failure surface, fatigue cracks began manifesting at the root of the base metal and proceeded to the bead surface (weld toe).

• Journal of Ocean Engineering and Technology
• /
• v.29 no.1
• /
• pp.45-55
• /
• 2015

Because of the high stress concentration near the toe of a welded joint, the calculation of local stress using the finite element method which is relevant to the fatigue strength of the weld toe crack, is a challenging task. This is mainly caused by the sensitivity of finite element analysis, which usually occurs near the area of a dramatically changing stress field. This paper presents a novel numerical method through which a less mesh-sensitive local stress calculation can be achieved based on the 3D solid finite element, strictly sticking to the original definition of hot spot stress. In order to achieve the goal, a traction stress, defined at 0.5t and 1.5t away from the weld toe, was calculated using either a force-equivalent or work-equivalent approach, both of which are based on the internal nodal forces on the imaginary cut planes. In the force-equivalent approach, the traction stress on the imaginary cut plane was calculated using the simple force and moment equilibrium, whereas the equivalence of the work done by both the nodal forces and linearized traction stress was employed in the work-equivalent approach. In order to confirm the validity of the proposed method, five typical welded joints widely used in ships and offshore structures were analyzed using five different solid element types and four different mesh sizes. Finally, the performance of the proposed method was compared with that of the traditionally used surface stress extrapolation method. It turned out that the sensitivity of the hot spot stress for the analyzed typical welded joints obtained from the proposed method outperformed the traditional extrapolation method by far.

• Journal of the Society of Naval Architects of Korea
• /
• v.45 no.6
• /
• pp.679-687
• /
• 2008

The fatigue life of ship structure under cyclic loading condition is made up of crack initiation and propagation stages. For a welding member in ship structure, the fatigue crack propagation life is more important than the fatigue crack initiation life. To calculate precisely the fatigue crack propagation life at the critical welding location, the knowledge of the residual stress sensitivity on the fatigue strength is necessary. In this study, thermo elastic-plastic analysis was conducted in order to examine the effect of residual stress on the fatigue crack propagation life. Also the fatigue crack propagation lives considering residual stress were calculated using fatigue crack growth code, AFGROW, on the basis of fracture mechanics. AFGROW is widely used for fatigue crack growth predictions under constant and variable amplitude loading. The reliability of AFGROW on the fatigue of ship structure was confirmed by the comparison of the estimated results with the fatigue propagation test results.

• Journal of Welding and Joining
• /
• v.6 no.4
• /
• pp.44-53
• /
• 1988

The mechanical strengths of buried lamellar tears located near the weld toe in the welded tee joints were evaluated in terms of the loss of load carrying capacity as a function of tear area. In static loading, the load carrying capacity was significantly reduced when tear intercepted over 10% of the cross-sectional area of the welded joints. However, the welded joints containing buried tears still failed at stresses over the yield strength of the base metal in the through-thickness direction in spite of the presence of tears up to 20-25% of the area. Fatigue strength of welded joints containing tears markedly reduced with increasing tear areas. Lehigh lamellar tearing test used in this study to produce speicmens was described in detail. The load carrying cpapacity in static loading was influenced by the reduction of supporting area whereas that in fatigue loading was influenced by the stress-concentration effects of lamellar tears and the reduction of supporting area. In bend tests, the pre-existing lamellar tears always grew up toward the weld toe. However, in fatigue loading, cracks grew up and down simultaneously form both the weld toe and the top of lamellar tears because of stress concentration. In fatigue loading, delaminations and decohesion of inclusion/matrix interface generated in multipass welds provided crack propagation paths and enhanced crack propagation because the tips of delaminations and deconhesios acted as stress raisers.

• Journal of Welding and Joining
• /
• v.12 no.1
• /
• pp.102-115
• /
• 1994

This paper is to evaluate the fatigue strength of lap joints of materials applied to LNG carrier cargo containment of GAZ-TRANSPORT(GT) type, which was welded by manual and automatic TIG welding process. The thicknesses of lapped members were 1.5mm/1.5mm or 1.5mm/0.7mm in Invar to Invar joint, and 1.5mm/8.0mm in Invar to stainless steel joint, respectively. These lap joints were mainly applied to the membrance fabrication of GT-LNG carrier. Fatigue tests of Invar/Inar lap joints were conducted under the stress ratio R=0 at room temperature. The effect of mean stress and cumulative fatigue damage on the allowable stress of Invar lap joint was evaluated on the basis of test results. Fatigue test was also conducted on Inver/Stainless steel lap joints welded by automatic TIG process without filler metals. The fatigue test of the joint was carried out under the same conditions as those of Invar/invar lap joints. The fatigue strength of the joint welded without filler metal was comparable to those welded with filler metal quoted from reference. The fatigue strength of Invar/stainless steel lap joint was only dependent on the lap throat thickness, and not on the welding process. Based on test results, the applicability of TIG welding process without filler metal in Invar/stainless steel lap joint was reviewed by controlling welding variables to assure the valid throat thickness of lap joints.

• Laser Solutions
• /
• v.12 no.3
• /
• pp.14-17
• /
• 2009

The laser welding was performed for the flange joint of two overlapped coaxial circular pipes which serve as the inlet and connector pipes of STS 316L. The laser welding test finally resulted in a good penetration depth of 1.8 to 2.0 mm. On the way to get the good welding quality, two important parameters were found to be optimized. One is the focal positioning which is the offset of the laser beam focus to the exact welding seam line, which is more critical in the inner flange laser welding. When the beam spot size was deviated more than $200\;{\mu}m$ from the seam line, welding of two pipes is failed. The other is a gap size since a certain amount of gap is inevitable due to fabrication tolerance, or artificial allowances for smooth insertion of a pipe. However, it is required to restrict the gap allowance within 0.2mm to avoid undesirable undercut on a welding bead.

• Journal of Welding and Joining
• /
• v.23 no.6
• /
• pp.87-92
• /
• 2005

It has been well hewn that ductile fracture of steels is accelerated by triaxial stresses. The characteristics of ductile crack initiation in steels are evaluated quantitatively using a two-parameters criterion based on equivalent plastic strain and stress triaxiality. The present study focuses on the effects of strength mismatch, which can elevate plastic constraint due to heterogeneous plastic straining, on the critical condition for ductile fracture initiation usinga two-parameter criterion. Fracture initiation testing has been conducted under static loading using notched round bar specimens which had different notch locations. This study provides the fundamental clarification of the effect of strength mismatching and effect of notch location on the critical condition to ductile crack initiation from notch root using fuite element method and ultrasonic-mechatronics system. The critical condition of ductile crack initiation from notch root of strength mismatched tensile specimens under static loading appeared to be almost the same as those of homogeneous tensile specimens with circumferential sharp notch specimen. Also, the effect of notch location in mismatched specimens was estimated using finite element(FE) analyses.

• Journal of Advanced Marine Engineering and Technology
• /
• v.17 no.4
• /
• pp.100-106
• /
• 1993

A study on optimizing the friction welding of copper(C1100) to aluminium(A1050) for developing the electrical sleeve was experimentally carried out and also on real-time nondestructive evaluation of the friction weld quality (strength) was accomplished by acoustic emission technique. The results obtained are summarized as the following ; 1) The heating upset $U_1$(mm) or total upset U(mm) tends to increase according to the increase of heating time $t_1$(sec). The relations between $U_1$ and $t_1$ or U and $t_1$are computed as follows when n=2000rpm, $P_1$=4, $P_2$=8kgf/$mm^2$, and $t_2$=6sec. U=1.6$e^{0.39t_1}$ $U_1$=3.65$e^{0.25t_1}$. 2) It was notified that the proper welding conditions by considering on both strength with more than 100% joint effieciency and toughness are heating time of 1.5-2.25 sec under n=200rpm, $P_1$=4, $P_2$=8kgf/$mm^2$, $t_2$=6sec. 3) It was confirmed that both AE total counts(N, counts) and the weld tensile strength (${\sigma}$, kgf/$mm^2$) of the welded joints increase as the increase of heating time, respectively, the relations between N and $t_1$, ${\sigma}$ and $t_1$ are computed from data points by regression analysis using the least square method as follows in case of the above proper condition ; N=50108+23917(ln $t_1$)${\sigma}$$=11.85+2.06(ln $t_1$). 4) Both empirical and calcularated equations of relationship between .sigma. and N are very coincident with a high reliability, as the following in case of the above proper welding condition ; Calculated : ${\sigma}$=0.00008N+7.5 Empirical :${\sigma}$= $8.17e^{0.0000072N}$. 5) It was confirmed that the real-time nondestructive weld strength evaluation for friction welding of copper(C1100) to aluminium(A1050) could be possible by acoustic emission technique.

• Journal of the Society of Naval Architects of Korea
• /
• v.49 no.1
• /
• pp.99-108
• /
• 2012

In this paper, fatigue tests are conducted for the specimens with longitudinal and transverse attachment under variable amplitude axial loading based on storm model. Considered loadings include repeated single storm, 6 or 8 storms randomly, and storms including calm sea condition while the mean stress and the maximum stress of loadings are changed. The effect of three variables are investigated; root mean square(RMS) value of stress amplitude, mean stress shift and maximum stress, which can characterize storm loading on fatigue life. In addition, experiments including calm sea loading are also carried out to investigate the effect of calm sea state. Test results are evaluated and compared with DNV-CN2005 and Matsuoka's method for the estimation of crack initiation and propagation life. To verify the validity of the criteria, the measured crack initiation lifes are compared with the specific crack length 15mm, which are calculated with beach marks.