1. Research Background and Purpose
As offshore energy development has been recently conducted in an extreme environment, conditions required for the installment of offshore structures have been reinforced1). Moreover, since most energy resources buried in coastal waters have been extracted, a range of energy resource mining has been extended to waters less than 1,000 m deep. Such extension of the mining range in waters has intensified environmental conditions for the installment of offshore structures2), and these structures have been constantly enlarged to mine a great amount of energy. The establishment of such enlarged offshore structures requires development of high-strength steel for offshore structures and appropriate welding technology3,4). As offshore structures tend to be built in a fixed way unlike vessels, steel for offshore structures has stricter quality requirements than that for vessels. Particularly, quality management of a welding zone is regarded as a significantly crucial parameter5). It is essential to control welding residual stress generated in the welding process among various parameters. High strength steel shows a wide distribution of welding residual stress, which is known as a factor affecting the occurrence and spread paths of unstable fractures6,7) and is significantly considered in numerous reliability tests8-11). It is crucial to control distortion and residual stress generated in the welding process in structures. Welding distortion affects the appearance of structures negatively and has a more negative effect on the safety of these structures. Welding residual stress must be considered to evaluate brittle fracture strength of structures12,13), and structures with a risk of fatigue fracture should be evaluated based on welding residual stress according to components14-16). A distribution pattern of welding residual stress might vary according to constraints in the welding processes and structure manufacturing processes. For this reason, the optimal welding conditions should be identified to minimize welding residual stress. This study analyzed distribution properties of welding residual stress in high-strength steel for offshore structures according to the effect of constraints by adjusting welding constraints17,18). To this end, this study produced and used butt joint specimens which applied different constraints. Specifically, this study evaluated distribution properties of welding residual stress by manufacturing ① a fully constrained specimen and ② a fully unconstrained specimen and welding both specimens under the same conditions. The cutting method, which is well-known as an experimental method that shows comparatively high reliability, was used to evaluate distribution properties of welding residual stress according to the effect of con- straints. Based on the experimental result, this study conducted fundamental research on analyzing different patterns of residual stress generated in both specimens from the perspective of welding mechanics.
2. Applied Steel and Experimental Method
2.1 Applied steel and specimen production
This study evaluated distribution properties of residual stress in welding zones according to the effect of constraints. In the evaluation process, it manufactured and applied butt joints, which are used the most widely for structure welding, made of high strength steel with a yield stress of 500MPa (E500) for offshore structures. Tables 1 and 2 describe the chemical and mechanical properties of the E500 high strength steel for offshore structures, which was used in this study, respectively. This type of steel shows a comparatively high yield strength of 529 MPa compared to other types of steel used for offshore structures. It also shows the average impact toughness of 249 J at -40 ℃, thereby exhibiting excellent performance for impact toughness at low temperature. Mechanical behaviors in the welding zone vary depending on the effect of constraints. In this regard, this study assumed that welding residual stress and distortion will show different behaviors according to the effect of constraints. Thus, this study represented fully constrained and unconstrained conditions in experiments and investigated these conditions on residual stress and angular distortion. This study produced specimens by using a commercial AWS A5.29 E91T1 welding material, which is generally used in E500 steel for offshore structures. Tables 3 and 4 indicate the chemical and mechanical properties of the welding material respectively. The welding material produced based on overmatching joints showed the yield strength of 580 MPa, which was higher by approx. 50 MPa than that of the base metal. The tensile strength and elongation of the welding material were 650 MPa and 20% respectively. Fig. 1 shows the shape and size of specimens manufactured. Specifically, the specimens were made of 25 mm thick steel and applied the following conditions: the root gap of 4 mm and the V-groove shape which has 45° groove angle and welding length was 400 mm. The entire fully constrained and unconstrained specimens were produced under the same welding conditions, which are indicated in Table 5.
Table 1
Chemical composition of E500 steel (wt. %)
| Material | C | Si | Mn | P | S |
|---|---|---|---|---|---|
| E500 | ≥0.08 | ≥0.2 | ≥1.6 | ≥0.01 | ≥0.005 |
Table 2
Mechanical properties of E500 steel
| Material | Yield stress (MPa) | Tensile stress (MPa) | Elongation (%) | Charpy impact test, -40°C, (J) |
|---|---|---|---|---|
| E500 | 529 | 646 | 19 | 249J |
Table 3
Chemical composition of welding consumable (AWS A5.29 E91T1) (wt. %)
| Welding consumable | C | Si | Mn | P | S |
|---|---|---|---|---|---|
| AWS A5.29 E91T1 | 0.06 | 0.29 | 1.23 | 0.007 | 0.008 |
2.2 Welding residual stress measurement method
Destructive or non-destructive methods are generally utilized to analyze welding residual stress. This study measured welding residual stress by using the cutting method, which is known as a destructive method that shows high data reliability. Single- and two-axis strain gauges were used in the cutting method, and a distortion rate was sequentially measured based on five densely spaced single-axis strain gauges. Table 6 presents information on the strain gauges used. Fig. 2 shows the constraints and macro sections of the fully constrained and unconstrained specimens. As for the fully constrained specimen, end tabs in the same size as that of specimens were used to fully constrain both ends of the welding start and finish sections. A steel backing agent was applied, and a jig was utilized in the welding process to implement the fully constrained conditions. As for the fully unconstrained specimen, both ends of this specimen were tack-welded with end tabs in the minimum size, which facilitated welding, to satisfy the minimum constrain conditions. A ceramic backing agent was applied to allow welding processes in conditions as free as possible. Fig. 3 shows the cutting location and the attachment location for strain gauges for residual stress measurement at welding zones. The distribution of welding residual stress tends to rapidly change near welding zones including heat-affected zone (HAZ). For this reason, strain gauges were densely attached near the HAZ. When residual stress is measured based on the cutting method, it is required to control the heat generated by cutting velocity. The cutting velocity was maintained as 2 mm/min, and cutting fluid and water were used to control the heat generated in the cutting process. Strain gauges were attached near the area located at 150 mm, the center in the direction of a welding line. In the HAZ which showed considerable stress change, single-axis 5-elements strain gauges were densely attached from the direction of the fusion line to the width direction of the welding line up to the area located at 20 mm. Two-axis strain gauges were attached at the locations at 25 mm, 30 mm, 40 mm, 65 mm, 85 mm, and 130 mm, respectively, from the direction of the fusion line to the width direction of the welding line. Moreover, strain gauges were attached at the same location on the opposite surface in the direction of the welding line to ensure the reliability of distortion rates measured. Fig. 4 shows the photos of specimens taken after they were cut in the vertical direction to the welding line. The specimens were cut in the direction of the welding line and the direction of thickness to release both residual stress in the vertical direction to the welding line and welding residual stresses in the thickness direction. Cutting was conducted in directions including the thickness direction to obtain only the effect on the surface as much as possible.
3. Results and Considerations
3.1 Results of welding residual stress measured based on the cutting method
This study measured welding residual stress at weld metal zone and HAZ by applying the cutting method to analyze the distribution of welding residual stress according to the effect of constraints. Fig. 5(a) shows distribution behaviors in the direction of the welding line. The fully constrained specimen showed tensile residual stress of 379 MPa at weld metal zones and that of 332 MPa at the location of FL+1mm. Tensile stress decreased as the measured part was far from the welding line. The fully constrained specimen showed the typical residual stress distribution of butt joints, which showed compressive residual stress of -106 MPa. In addition, the fully unconstrained specimen showed similar residual stress of 375 MPa at the weld metal zone to that of the fully constrained specimen at the weld metal zone. This result indicated that different external constraints of both specimens did not have a significant constraint effect on welding residual stress at weld metal zones. Moreover, the fully unconstrained specimen showed tensile residual stress of 202 MPa at the location of FL+1mm. Tensile stress decreased as the measured part was far from the welding line. Compressive residual stress of –9 MPa was observed at the location of FL+100mm. The fully unconstrained specimen showed a difference of residual stress of approx. 100 MPa at the same location of FL+1mm from the fully constrained specimen. The fully constrained specimen with significant constraints showed significant welding residual stress at the HAZ. However, as the measured location was moved to the base metal from the HAZ, both specimens showed a similar distribution of welding residual stress. This result was obtained because residual stress was analyzed as stress distributed at the location of base metal rather than residual stress caused by weld heat. Fig. 5(b) shows distribution behaviors of residual stress in the vertical direction to the welding line. The fully constrained specimen showed tensile residual stress of 170 MPa at the weld metal zones and that of 114 MPa at the location of FL+1mm. Tensile stress decreased as the measured part was far from the welding line. The tensile residual stress was measured as 22 MPa at the location of FL+130mm. The fully unconstrained specimen showed tensile residual stress of 135 MPa at the weld metal zones, that of 114 MPa at the location of FL+1mm, and that of 22 MPa at the location of FL+130mm. It showed similar residual stress at the weld metal zones to that of the fully constrained specimen in the direction of the welding line. The fully constrained specimen showed higher tensile residual stress at the HAZ than the fully unconstrained specimen. Furthermore, both specimens showed similar distribution tendencies of welding residual stress. Welding residual stresses were affected by contraction and expansion at welding zones including weld metal zone and HAZ according to the degree of external constraint. When external constraint was applied severely, the measured parts were constrained by contraction and expansion caused by weld heat. As a result, residual stress at the welding zone increased. On the other hand, areas located far from the welding zone were not affected by weld heat or external constraint. Consequently, residual stress was similar in these areas regardless of the effect of constraints.
3.2 A relationship between welding residual stress and distortion according to the effect of constraints
The HAZ showed a greater difference of welding residual stress distribution according to the effect of constraints than the weld metal zone, and the fully unconstrained specimen showed less welding residual stress than the fully constrained specimen. It was found that both specimens showed different welding residual stress according to the effect of constraints under the conditions where the same steel and welding requirements were applied. Accordingly, it was analyzed that welding residual stress reduced in the fully unconstrained specimen appeared in a different form. For this reason, this study measured the amount of distortion in both specimens. Generally, both amounts of in-plane and out-of-plane distortion should be considered. However, this study considered only angular distortion, which is out-of-plane distortion. In the fully constrained specimen, angular distortion did not occur at all due to the effect of constraints. On the other hand, angular distortion of approx. 6° occurred in the fully unconstrained specimen, as shown in Fig. 6. In other words, as the fully unconstrained specimen was not externally constrained, it accompanied free contraction, expansion, and angular distortion caused by weld heat. However, this specimen showed significant welding residual stress at the weld metal zone and the HAZ due to the effect of internal constraints. Yet, the residual stress at the welding zone was reduced in this specimen compared to that in the fully constrained specimen. It was analyzed that the reduced residual stress at the welding zone might appear in the form of in-plane or out-of-plane distortion. This study quantitatively measured only out-of-plane distortion, and it is analyzed that the reduced residual stress energy might have appeared in the form of in- plane distortion as well. This study considered only out-of-plane distortion without in-plane distortion.
4. Conclusions
This study analyzed the distribution of welding residual stress according to the effect of external constraints. To this end, it manufactured fully constrained and unconstrained specimens by using high-strength steel for offshore structures, which was 25 mm thick, and compared distribution properties of welding residual stress of these specimens. Based on the analytic results, it derived the following conclusions.
1) Both fully constrained and unconstrained specimens showed similar residual stress at weld metal zones in the direction of a welding line, which was at a similar level to the yield stress of the base metal. Moreover, both specimens showed similar distribution patterns of welding residual stress to those observed at the existing butt joints.
2) The fully unconstrained specimen showed a distribution of reduced welding residual stress at the HAZ than the fully unconstrained specimen. Both specimens showed similar welding residual stress at base metal zones, which were unlikely to be affected by weld heat.
3) The fully constrained specimen showed a distribution of reduced welding residual stress at the HAZ according to the effect of constraints, which was observed in the form of in-plane and out-of-plane distortion.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
