The combination of flux-cored arc welding (FCAW-G) and arc welding (SMAW) or gas arc welding (GMAW) in the root pass produces reliable girth welds in X80 class pipes and produces clear results. The practical benefits are welding reliability in all positions and the productivity of a semi-automated mechanized process – without significantly increasing costs. Flexibility in connection selection and chamfer preparation is another important aspect. These properties make it very attractive for girth welding of pipes in the field.
The steel materials available in the pipe industry fall into two broad categories. Steel up to grade X70 is considered ordinary steel. Most pipelines built around the world fall into this category. However, the trend is shifting towards higher valuations. Higher grades of steel such as X80 were developed in the 1980s and provide greater mechanical strength. Using higher grades can also reduce logistics costs and welding levels. The grade of material used in a pipe accounts for approximately 40% of its total cost.
In recent years, even higher quality materials such as X100 and X120 have been considered for very large pipes. However, in practice their use is very limited. However, it is recognized that efforts must be made to persuade gas network operators to use these modern X100 or X120 steels, even if they are not included in current gas industry practice. However, despite intensive efforts to introduce these new materials due to their superior properties, the qualification and fabrication of these pipes poses significant challenges to the industry.
The properties of high-quality steel are primarily related to how it is made. Currently, accelerated cooling thermomechanical processing (ATMT) is used to produce higher grades of steel with improved microstructure compared to traditional steel grades. In particular, X80 material has a fine-grained bainite microstructure that provides higher strength than X60 or X70, while still being strong enough to meet international standards. However, X80 is more susceptible to weld defects than ferritic-based materials.
SMAW and GMAW are widely used in the pipeline industry for field welding. Common processes—GMAW, submerged arc welding (SAW), and flux-cored arc welding (FCAW)—have all been investigated as potential candidates for producing high-grade steel pipes. In addition, combinations of technologies such as GMAW and SMAW or GMAW and FCAW in gas protection (FCAW-G) have been considered for the construction of large pipelines. Highly innovative welding processes are also being explored, such as hybrid laser arc welding and friction stir welding (although mainly for the production of pipes themselves), as well as high-performance submerged arc welding methods.
Standards and tests set by international standards such as API 1104, ASME IX and EN-12732 are commonly used to qualify gas pipeline welding. When considering the use of advanced materials such as X80 for the construction of natural gas transportation infrastructure, three fundamental properties must be considered: mechanical strength, toughness and weldability.
The X80 material is stronger than the more widely used X70 material. But the increased strength must compensate for potentially poor weldability and deformability, which can make the structure more susceptible to failure. The main criteria used to assess the quality of circumferential welds are summarized in international standards.
In this article, the weldability of X80 girth welds produced by FCAW-G in combination with other processes such as SMAW and GMAW is compared with those commonly used in the pipeline industry such as SMAW and GMAW. Welds have been tested to API 1104 and EN-12732 with ASME Part IX specifications. The quality of circumferential welds was assessed by comparing the microstructural and mechanical properties of welded joints.
The main objective was to test whether the welding process in combination with FCAW-G could maintain sufficient mechanical properties to prevent pipe failure while being practical and cost-effective for use with X80 grade pipes. Welding parameters and joint preparation, including preheating required to achieve the expected mechanical properties of the weld, are taken into account.
The research in this paper was conducted on API 5L X80 PSL2 (hereinafter referred to as X80) 406.4 mm (16 in.) diameter and 14.38 mm diameter pipes manufactured using resistance welding (ERW) and high frequency induction (HFI). A welding process commonly used in gas pipeline construction was chosen. When testing welded samples, the mechanical properties obtained as a result of each process, as well as the weldability of the material, were measured. These tests ultimately provide first-party data and information that allows you to select the most appropriate processes when building future infrastructure.
Metallographic studies of welds were carried out on cross sections of processed prismatic samples with a diameter of 80 mm, located in the center of the weld. The specimens were prepared by conventional grinding and polishing and etched using 2% nitric acid and ethanol reagent to reveal the different weld areas and their microstructure. Grain size and crystal orientation were determined by optical microscopy and electron backscatter diffraction (EBSD), respectively. An Oxford Instruments EBSD system coupled to a Jeol JSM 6500F field emission scanning electron microscope was used to characterize the crystallographic orientation of the weld material and heat-affected zone (HAZ). Increments used for EBSD plots are 0.3 µm.
The C and S contents are determined by combustion in an induction furnace and detecting carbon (C) and sulfur (S) by infrared absorption. The remaining elements were determined by spark optical emission spectroscopy (IGDOES).
Table 1 presents the results of the analysis. These values correspond to the average of three independent determinations of the test sample. These results are consistent with the chemical composition specifications for L555Q or X80M steel grades in API 5L-ISO 3183 and are specific to the elements analyzed.
The prismatic specimen, 55 mm long, was centered on the HFI connection line of the pipeline pipe and had a circumferential weld in its cross section. In traditional metallographic preparation, samples are ground, polished, and etched with 2% nitric acid to a mirror-like finish that demonstrates the various microstructural areas of the weld. Hardness testing was carried out using a Vickers hardness tester at room temperature and a load of 10 kg. Two rows of recesses are made on the sample, corresponding to the internal and external sections of the pipe wall.
Pipe tensile testing and transverse girth weld testing were carried out using standardized round steel specimens. The strain rate during testing was 5×10-4 s-1.
The samples are cylindrical (diameter 6.2 mm, calibrated length 26 mm), all weld metal samples are machined longitudinally along the weld bead, which makes it possible to determine the mechanical characteristics of the weld metal under study.
Charpy and crack tip opening displacement (CTOD) tests were carried out on single sided bending notched (SEBN) specimens cut into the HAZ and weld metal and machined according to the sketch shown in Figure 1. The dimensions of the specimen were 10 × 7 section. 5 mm2, length 55 mm. Charpy samples were machined with V-shaped grooves 2 mm deep. The deep CTOD groove has a width of 0.1 mm and a radius of 60°. Each sample was chemically etched with metallographic reagents (2% nitrogen ethanol) to precisely position the notches at the desired test locations. The test temperature is -18℃. The Colora Ultra-Kriostat KT905 cools the sample.
The Charpy test uses a Charpy impact bending pendulum with a full voltage range of 300 J. CTOD specimens were pre-cracked in force mode according to ASTM-1820 and ISO 12135 standards. CTOD tests were performed on a 100 kilonewton (kN) Microtest universal testing machine in force control mode 20 Newton/second until maximum load is reached or unstable crack propagation occurs.
Manual welding processes are still widely used, requiring less investment in equipment and logistics, and more construction companies are available to carry out these jobs. Therefore, the first weld tests were carried out entirely using the SMAW procedure. Due to the properties of the steel, a cellulose type electrode was selected for the root pass and a low hydrogen electrode was selected for the fill and closing passes.
In the 1980s, following a general trend in other industries, semi-automatic welding processes began to be used in pipeline construction. A shortage of highly skilled welders and increases in welding productivity have prompted contractors to adopt these processes, which are still used on gas pipelines when factors such as distance, complexity or diameter do not justify the use of mechanized processes. Therefore, procedures using the manual SMAW process for the root pass and the semi-automatic FCAW-G process for the remaining weld passes were also included for comparison.
Mechanized processes became more widely used in the late 1990s and were the preferred choice for larger projects, even those involving small pipelines. Taking this into account, three fully automated procedures combining different welding processes were also developed for comparison with the various aspects studied in this paper.
Therefore, two manual processes and three mechanical processes were developed to study the influence of filler metal, process, and welding technology on the microstructure and mechanical properties of joints.
The use of a copper backing is discussed as there is a risk of weld cracks due to copper grain boundary segregation even after the pipe is in service. Companies are working to improve welding parameters and joint configurations to avoid this risk. In addition, welding equipment manufacturers are improving the ability to produce quality root passes without the need for a copper backing.
Special pulse modes, as well as welder programming capabilities, also improve arc control in every part of the circular weld. However, the weld pool is very difficult to control and can result in poor fusion of the welded part to the base metal and subsequent weld passes. This phenomenon is commonly known as incomplete fusion and is difficult to detect using non-destructive testing. Therefore, very strict control of welding parameters is necessary, which, in turn, requires the implementation of production quality control methods and often involves destructive testing of welds. If a fault is discovered, there is a risk that all welds made that day may require repair.
In the GMAW-P+FCAW-G process, the risk of copper contamination during the GMAW root pass must be controlled. However, this combination combines the benefits of a fully automated root pass with the greater reliability of the FCAW-G (and sometimes FCAW-S) in all positions.
Develop welding processes in all positions without the need for advanced welding equipment and complex programmers to control a continuous arc. FCAW requires significantly less investment than a fully automated process.
GMAW-STT + FCAW-G. GMAW Surface Tension Transfer (STT) and similar processes can produce high quality root beads without the need for a copper backing. They are based on the control of the signal shape of pulsed current welding machines. Each machine tool manufacturer develops its own patent for waveform control. Because each machine controls the arc in a unique way, new versions of the International Welding Standards (API, ASME, EN) consider the specific model of welding machine used when qualifying welding procedures used during construction as an “important variable.”
GMAW-STT can be used with semi-automatic or fully automatic technology. Although it requires a skilled welder, semi-automatic technology can produce good welds even with imperfect seam preparation. Mechanized technology requires investment in automated welding and joint preparation equipment (chamfering, clamping) for each joint type, but is less dependent on operator skill.
Table 2-4 provides descriptions of the manual, semi-automatic, and mechanized welding processes used in this article. All welds use a preheat of 100°C and a maximum temperature between passes of 250°C. In the SMAW process using a high hydrogen filler metal (cellulose electrode), dehydrogenation or soaking is performed at 100°C.
Table 5 shows the mechanical properties of the filler metal using actual values. Figure 2 summarizes the heat input for all welding passes studied. In Fig. Figure 3 shows a diagram of the connection of the studied welds.
Figs In Fig. Figures 4-8 show macro photographs of the examined welds. The same applies to microstructures obtained by manual welding and mechanical welding, in Table. 6. The microstructure of the weld metal contains acicular and polygonal ferrite with a low bainite content.
The microstructure corresponds to the direction of solidification. Typically, coarser microstructures are present in the fill passages. The microstructure of the HAZ largely depends on the welding technology and the welding materials used. The grain size of the weld is shown in Figures 1 and 2. 9-10. The average grain size of all welds is about 2 µm. However, grains as small as 10 µm can be found in SMAW welds.
There are large differences in the effective grain size and orientation for EBSD of the sputtered materials studied. In manual SMAW welding, large grain sizes do not have preferential crystallization directions, which should lead to good mechanical properties. Direction is measured using (hkl), which is a family of planes orthogonal to a given direction. In cubic crystal systems such as iron, the crystal orientation is described by {h+k+l}b, where b is the fundamental inverted lattice spacing.
In the case of SMAW+FCAW-G welding (Fig. 9a), the observed effective grain size is significantly smaller than that of SMAW and has a significant orientation gradient. The upper region of the map (Fig. 9b) is dominated by particles (220 hkl). And in the lower region the most pronounced particles (200 ppl) and (111 ppl) were observed. This texture gradient reduces the mechanical properties of SMAW+FCAW-G welds.
Mechanized welding shows a smaller effective grain size, especially when welding GMAW-P, where there are practically no grains oriented along the cleavage direction (220 hkl) (Fig. 9c). In other machine welds, the effective grain size is between the grain sizes of manual welds, but in STT+FCAW-G welds there are more grains oriented along the cleavage direction (Fig. 9e).
For the SMAW+FCAW-G process, the maximum hardness gain is approximately 100 hardness values (HV), taking into account the different welding consumables used in the root and cap passes. When welding SMAW, again using different materials for the root and top welds, the hardness increment is about 50 HV.
With mechanized welding, the difference in hardness is much smaller. Especially in the GMAW-P and GMAW-P+FCAW-G processes, the internal regions are harder than the external regions, despite the reheat associated with multi-pass welding. This can be attributed to the combination of the J-shaped weld configuration and the low number of welding passes. This combination reduces the grain refinement associated with interpass heat treatment and promotes the formation of a needle-shaped grain from the outer surface to the center of the weld (Figures 6b and 7b).
The Rm values obtained from tensile tests in Table 7 are very similar for all welds, but lower than the Rm of the corresponding welding consumables, with the exception of SMAW welds (Table 5). The maximum resistance of the SMAW and SMAW+FCAW-G processes is very similar, even slightly higher than the maximum resistance of the X80 welding wire. In two SMAW + FCAW-G manual welding test specimens, the fracture was located inside the weld. However, the stretch quality of both is good as the criteria for excellence are met. In one of the specimens tested for a GMAW+FCAW-G weld, fracture occurred at the interface between the HAZ and the base metal.
The elongation values found in tensile tests are typically 5-10% lower than the elongation of the welding consumables (specified by the manufacturer) in the manual method and 5-10% higher in the mechanized method. In most cases, the lack of correlation between nominal and measured values is due to liquefaction occurring between different materials during the welding process.
With the exception of the GMAW-P process, the Charpy V-notch (CVN) values absorbed in the HAZ are similar to those of the X80 substrate. In general, there is a large scatter in CVN values due to changes in the microstructure of the weld.
For SMAW, GMAW+FCAW-G and FCAW-G processes, the CVN value of the base material is approximately one-third that of the X80 base material, and the deviation is small. Automatic GMAW-G+FCAW welding produced the highest CVN values among weld metals. In contrast, the SMAW+FCAW-G process has lower CVN values and larger deviations compared to other welds.
Industry CTOD requirements API 1104 and DNV-OS-F101 specify minimum impact strength of 0.1 mm and 0.15 mm, respectively. Table 8 shows the values of Je, Jp, Kc(Jt) and CTOD for SMAW, GMAW-P and GMAW-P+FCAW welds. The load area versus CTOD in Figure 13 was calculated for the maximum load since no overshoot occurred during the test. The average CTOD of the SMAW weld filler metal is 0.37. The average CTOD value of GMAW-P filler metal is 0.30, and the average CTOD value of HAZ is 0.34. The smallest value corresponds to the filler material – 0.16. The average CTOD value at the lowest filler metal dispersion of GMAW-P-FCAW-G weld is 0.30. In all cases, CTOD values were greater than 0.15 mm. The Kc(Jt) values of SMAW, GMAW-P and GMAW-P+FCAW welds are also very similar (Table 8). The good correlation between CVN and CTOD results was striking.
All welds examined met the superiority criteria of the API 1104 standard (Table 7). The mechanical strength of the filler material (Table 5) has a great influence on the transverse strength of the weld and is calculated as the average mechanical strength weighted by the volume of the weld bead.
With the exception of the GMAW-P weld, the joint strength is slightly higher than that of the filler metal (Table 5). Although the failure was located in the weld metal of SMAW+FCAW-G and GMAW-P+FCAW-G welds, the exceedance criterion was still met. The location of weld metal failure is often associated with its lower mechanical strength compared to the base metal. When welding SMAW+FCAW-G, the hardness inside the pipe decreases very significantly in the central region of the filler material (Fig. 4b). This hardness gradient creates metallurgical notches that help explain the failure of the weld metal.
In addition to heat input, the mechanical properties of the weld depend on the number of welding passes used. When welding SMAW+FCAW-G, the number of passes is significantly less than when welding SMAW with similar joint preparation. This also explains the strong stiffness gradient found in SMAW+FCAW-G welds. The hardness value depends on the combination of grain size and chemical composition of the electrode used (Midawi et al., 2015).
For SMAW+FCAW-G welds, the hardness values are similar to SMAW welds. Although the grain size of hand-welded welds is larger, hand-welded seams are 10% harder than mechanical welds (Figure 4).
The materials used in the mechanized process, E-91T1-M21 4G, E-R70S-6 and E-R110S-G, are chemically similar and provide a higher level of alloying than the materials used in manual welding. In particular, the E-91T1-M21 4G material used in the GMAW-P+FCAW-G and STT+FCAW-G processes has a lower manganese content. In addition, it does not contain elements such as niobium, titanium and zirconium, which tend to form fine precipitates that anchor grain boundaries. Therefore, GMAW-P+FCAW-G and STT+FCAW-G welds have a larger grain size (Figs. 9 and 10) and lower mechanical resistance (Table 7) than GMAW-P welds.
The CVN value of the filler meets the criteria defined in the standard. The CVN of all welds was similar except for the high value found for GMAW-P + FCAW (WM) weld metal and the low value found for SMAW + FCAW WM (Table 7). The latter may be caused by the presence of small defects in the weld that are not detected by radiographic analysis.
The CVN value depends on the mechanical properties of the welding consumables used in various processes. In fact, there is a linear correlation between the CVN value and the strain area when considering tensile stress, elongation and hardening values. Consequently, the higher the Rm*E*(Rm-Re0.2), the higher the CVN (Fig. 13). The only welds that do not follow this linear relationship are GMAW-P welds. Its bainitic structure explains the lower CVN of GMAW-P welds despite the higher mechanical properties of the filler metal. CVN depends on the effective grain size, the number of low-angle grain boundaries and the phases present in the microstructure.
The most common phase in X80 high strength welds is acicular ferrite, although it is often accompanied by polygonal ferrite and bainite. At different heat inputs, the nucleation of granular bainite and the growth of the martensitic-austenitic component reduce the fracture toughness of welded partitions. Electrode chemistry, number of passes in a particular weld, and heat input are known to influence the phases present in the X80 material being welded.
Post time: Jun-13-2024