2021 Volume 62 Issue 7 Pages 995-1000
Twin-electrode GTAW is a novel welding technology in recent years and attracts lots of attention to researchers. However, stainless steel cladding with twin-electrode GTAW has been scarcely reported. This paper investigates the microstructure and mechanical performances of austenite stainless steel cladding by twin-electrode GTAW, and particularly the heat input is concerned. Experimental results of hardness tests, bending tests, and corrosion resistance tests show that both single GTAW and twin-electrode GTAW produce defect-free weld beads which meet engineering standards. Compared to single GTAW, twin-electrode GTAW improves the welding productivity at a lower heat input because of its higher welding speed and melting rate. Oscillation twin-electrode GTAW cladding also produces fine weld bead formation, but causes excessive heat input due to its very low welding speed and relative large welding currents. During oscillation twin-electrode GTAW, Fe–Cr(–Mo) intermetallic compound (σ phase) tends to precipitate in the weld bead, which leads to undesirable ferrite content results.
Stainless steel cladding is usually applied in petro-chemical pressure vessels manufacturing. The traditional GTAW has low welding efficiency. Therefore twin-electrode GTAW was utilized to clad the stainless steel. Experimental results show excellent weld bead formation by twin electrode GTAW with and without oscillation. The mechanical performance of the obtained weld beads meet engineer requirements. Twin electrode GTAW improves welding efficiency profoundly.
Petro-chemical pressure vessel usually works at or in high temperatures, high pressures, and hydrogen-rich environments. Therefore, austenite stainless steel is clad on the main body of the petro-chemical pressure vessel usually made of low alloy steels to meet the requirement. Researchers have studied stainless steel cladding with many welding methods, such as shielded metal arc welding (SMAW), gas tungsten arc welding (GTAW), flux core arc welding (FCAW), submerge arc strip cladding (SAC), etc.1–5) The SAC and GTAW are the most practically utilized methods for cladding stainless steel in petro-chemical pressure vessel manufacturing. The SAC offers the highest melting rate, but easily generates welding defects, and it is sensitive to the welding conditions. The GTAW can produce the best welding quality, but the melting rate is very low. In recent years, hot wire GTAW, narrow-gap GTAW, and twin-electrode GTAW were developed by researchers to overcome the disadvantages resulting from the low welding efficiency of the traditional GTAW.6–8) The twin-electrode GTAW was introduced by Japanese researchers in 1998, and successfully applied in the manufacture of liquid gas tanks, since then the twin-electrode welding has been paid great attention broadly.9–11)
During the twin-electrode GTAW, the coupling state of the arc is a key issue. Studies have shown that, it is most likely to form coupling arc when the distance between the twin electrodes is below 2 mm, which means the two single arcs merge into one arc. When the distance between the twin electrodes decreases, the arc pressure and arc temperature increase, but the arc pressure of the twin-electrode GTAW is still much lower compared to the single GTAW.12) Much higher currents can be applied for the twin-electrode GTAW due to the low arc pressure to produce fine weld bead, therefore the twin-electrode GTAW improves welding productivity dramatically. According to experiments, the single GTAW produces discontinuous and bump weld beads at a current of 300 A and a welding speed of 550 mm/min, but smooth and continuous weld beads at 400 A.13) When the melting rate of single GTAW is 1.83 Kg/h, the melting rate of twin-electrode is 4.32 Kg/h, which improves by 136%. Most researchers have utilized the twin-electrode GTAW in low carbon steel welding,14) and recently researchers from Austria have applied twin-electrode GTAW to clad Ni based alloys.15) So far, stainless steel cladding with twin-electrode GTAW has been scarcely reported. This paper studies the welding process of stainless steel with the twin-electrode GTAW, it may provide good reference for applications of the twin-electrode GTAW.
The schematic diagram of the twin-electrode GTAW welding process is shown in Fig. 1. There are two GTAW power supplies and two tungsten electrodes insulated from each other were integrated in one weld torch. The diameter of the tungsten electrode was 4 mm, and the distance between the tips of the two tungsten electrodes was 1.5∼2 mm. Each tungsten electrode linked to one GTAW power supply, and a hot wire power supply was utilized to preheat the feeding wire.
Schematic diagram of twin-electrode GTAW cladding.
A base material of pressure-vessel SA516Gr70N with geometries of 800 × 300 × 80 mm3 was investigated. The chemical composition of the SA516Gr70N is shown in Table 1. Welding wires of ER309LMo and ER316L with a diameter of Φ1.2 mm were used, and their chemical compositions are also included in Table 1. The shielding gas is 99.99% Ar gas. Before welding, the base metal was preheated to 398 K, and the temperature between each layer was controlled at around 443 K during welding. ER309LMo was clad on the base metal as the first layer, then the ER316L was clad for the second layer. The twin-electrode GTAW welding with and without oscillation were separately conducted, and single GTAW cladding was also conducted to make comparisons between the two welding processes. Welding parameters utilized are shown in Table 2. For the oscillation twin-electrode GTAW welding, the oscillation amplitude and frequency were 20 mm and 3 Hz, respectively. Heat input was calculated by the eq. (1), where the U and I are the average voltage and current, and V is the welding speed. η is the heat efficiency, and in this paper it is assigned with 0.9. The calculated heat inputs are 0.68 KJ/mm, 1.56 KJ/mm, 1.86 KJ/mm for the twin-electrode GTAW, single GTAW and oscillation twin-electrode GTAW welding processes, respectively. Melting rate and melting efficiency were determined by wire feeding rate and welding speed, respectively. In Table 2, it shows that despite the heat input is lower than the single GTAW, the twin-electrode GTAW has a higher wire feeding rate, and a higher welding speed, indicating a higher melting rate and a higher welding productivity.
| \begin{equation} \mathit{HI} = \frac{\eta UI}{V} \end{equation} | (1) |
After welding, specimens were cut out by wire-electrode cutting, then were ground and polished. Chemistry reagents of 4% HNO3 of ethyl alcohol and 10% oxalic acid were utilized to corrode the specimen for microstructure observation. In order to indentify the σ phase in the microstructure, Groesbeck reagent was also applied. With corrosion of the Groesbeck reagent, the σ phase presents a color of dark gray in the image. The hardness was tested in a Vickers hardness machine. Bending tests and corrosion resistance were also conducted.
The macro-structures of the weld beads by the twin-electrode GTAW cladding without oscillation are Fig. 2(a) and Fig. 2(b). Figure 2(a) shows the appearance of the weld bead is continuous and smooth, and no obvious weld bead defects can be found. Figure 2(b) is the cross-section of weld bead by twin-electrode GTAW cladding, it shows that complete fusion were achieved between the weld bead and the base material. The penetration depth in the base material is 0.979 mm, implying that a low dilution rate was obtained during the cladding. The thickness of each cladding layer was around 3.5 mm, and total thickness was around 7 mm. Figure 2(c) and Fig. 2(d) are the appearance and the cross-section of the weld bead by oscillation twin-electrode GTAW cladding. It shows the weld beads were also continuous, but wrinkles formed on the surface of each bead due to the oscillation, they are deep and large because of the very large weld pools during the welding process. Figure 2(d) shows complete fusion without obvious weld bead defects. The penetration depth in the base material was around 2 mm, and the thickness of the each cladding layer was also 3.5 mm.
Weld bead macrostructure of twin-electrode GTAW, (a) Bead appearances by twin-electrode GTAW, (b) Bead cross-section by twin-electrode GTAW, (c) Bead appearance by oscillation twin-electrode GTAW, (d) Bead cross-section by oscillation twin-electrode GTAW.
Chemical compositions of the weld beads produced by the three welding processes were examined by an optical emission spectrometer (OES), as shown in Table 3. For the single GTAW welding, the weld bead had the lowest carbon content, indicating the lowest dilution rate of the weld bead.
Figure 3 shows the typical microstructure of the weld bead produced by the twin-electrode GTAW, and the microstructure are observed from base material, the first cladding layer and the second cladding layer, respectively. Figure 3(a) shows the base material (BM), which mainly consists of pearlites and ferrites. Figure 3(b) shows the heat affected zone (HAZ) of the base material and it mainly exhibited granular bainites. Figure 3(c) and (d) show the microstructure of first layer and the second layer, respectively. It shows that both the first layer and the second layer mainly consists of austenite and ferrite, which are the typical microstructure of the austenite stainless steel weld bead.
Microstructure of the weld beads produced by the twin-electrode GTAW, (a) base metal, (b) heat affected zone, (c) first cladding layer, (d) second cladding layer.
Figure 4 shows the typical microstructure of the weld beads produced by single GTAW, and the microstructure are observed from base material, the first cladding layer and the second cladding layer, respectively. The microstructure of the BM mainly consists of ferrite and pearlite, as shown in Fig. 4(a). The HAZ also consists of small size granular bainites, as shown in Fig. 4(b). The microstructures of the first cladding layer and the second layer of the fusion zone are shown in Fig. 4(c) and (d), respectively. The fusion zone mainly consists of austenites and ferrites. When compared with twin-electrode GTAW process, the grain size in the weld bead fusion zone of single GTAW is relatively larger, due to higher heat input.
Microstructure of single GTAW weld beads, (a) base metal, (b) heat affected zone, (c) first cladding layer, (d) second cladding layer.
Figure 5 shows the typical microstructures of the weld beads produced by the oscillation twin-electrode GTAW. The base material consisted of ferrites and pearlites. The HAZ mainly consists of bainites, ferrites, and cementite as shown in Fig. 5(b). The cooling rate during oscillation twin-electrode GTAW is the lowest because the heat input is the highest among the three types of welding processes. The ferrites and cementite tend to form near the fusion line where the low cooling rate was low and the peak temperature was high. However, the bainites tend to form away from the fusion line due to the increasing cooling rate. Figure 5(c) and (d) shows the first layer and second layer microstructure and they also consists of austenite and ferrite. Compared to Fig. 3(c) and Fig. 4(c), the grain size in the first layer produced by the oscillation twin-electrode GTAW was the largest among the three welding processes because of the largest heat input.
Microstructure of oscillation twin-electrode GTAW weld beads, (a) base metal, (b) heat affected zone, (c) first cladding layer, (d) second cladding layer.
Ferrite content (FN) is very important for austenite stainless steel, because it can prevent hot crack and enhance corrosion resistance of the weld bead, therefore the ferrite contents under the three welding conditions were calculated and measured. During the calculation, the Cr equivalence (Creq) and Ni equivalence (Nieq) were obtained by eq. (2) and (3), and then the ferrite contents were calculated according to Delong diagram. During the measurement, the ferrite contents were measured by an electromagnetic ferrite detector. The calculated and measured ferrite contents are shown in Table 4. The calculated ferrite contents in the weld beads produced by the three welding processes were all above 3%. The measured ferrite contents in the weld beads produced by the single GTAW and the twin-electrode GTAW were above 3%, which meets engineering requirements. However, the measured ferrite content by the oscillation twin-electrode GTAW was below 3%, which fails to meet engineering requirements. The reason is that Fe–Cr(–Mo) intermetallic compound (σ phase) precipitated due to the large heat input during the oscillation twin-electrode welding, which caused the δ ferrites to transfer to austenites. Figure 6 shows the typical microstructure of the cladding zone in the weld beads by oscillation twin-electrode GTAW. It shows Fe–Cr(–Mo) intermetallic compound (σ phase) mainly distributed at the grain boundary of the austenite. Therefore, a lower heat input should be applied in order to eliminate the σ phase.
| \begin{equation} \text{Cr}_{\text{eq}} = \text{% Cr} + \text{% Mo} + 1.5 \times \text{% Si} + 0.5 \times \text{% Nb} \end{equation} | (2) |
| \begin{equation} \text{Ni}_{\text{eq}} = \text{% Ni} + \text{30% C} + 30 \times \text{% N} + 0.5 \times \text{% Mn} \end{equation} | (3) |
Fe–Cr(–Mo) intermetallic compound (σ phase) in fusion zone.
The hardness tests were conducted using a Vickers hardness tester at a testing force of 98.07 N and a lasting time of 10 s. Three sets of tests were conducted on the specimen at there random locations and the hardness value was obtained by averaging the three measurement results. The hardness distributions of the weld beads are shown in Fig. 7. Among the three regions of the BM, HAZ, and FZ, the base metal exhibited the lowest hardness, and the fusion zone exhibited the highest hardness. The single GTAW and twin-electrode GTAW produced similar weld bead hardness at the heat affected zone, while the twin-electrode GTAW produced higher weld bead hardness at the fusion zone due to its lower heat input. In the fusion zone, both weld beads by single and twin-electrode GTAW consist of austenite and ferrite, but the hardness of the twin-electrode GTAW was higher than that of the single GTAW, which was caused by the higher ferrite content and smaller austenite grain size of the twin-electrode GTAW.
Hardness distribution of the weld bead.
The bending specimens including cladding layer, heat affected zone, and base metal with a geometry of 10 × 38 × 120 mm3 were cut perpendicularly to the welding direction. Bending tests were conducted for both single GTAW and twin-electrode GTAW weld beads according to ASME IX QW-160 under an experimental condition of D = 4a, α = 180, p = 2a. Figure 8 shows the experimental results of twin-electrode GTAW weld beads. Seen from Fig. 8, the cladding layer did not peel from the base metal, and cracks and incomplete fusion are not observed, which indicates the excellent toughness of the weld beads.
Bending test specimens.
The corrosion resistance tests were conducted with 10%H2SO4–10% CuSO4·5H2O solution for both the single GTAW and the twin-electrode GTAW weld beads. The size of the specimen was 3 × 20 × 80 mm3. The solution containing the specimen was heated until slightly boiled, then kept for 16 hours. Afterwards the specimen was cleaned, dried, and bent. No macroscopic cracks were observed on the specimens after the corrosion and bending tests, which indicates a good corrosion resistance of the weld beads.
This work was supported by State Key Laboratory of Marine Equipment Made of Metal Material and Application (grant No. HGSKL-USTLN-202010) and the Natural Science Foundation of Liaoning Province of China (grant No. 2019JH3/3010004) and the Science Foundation of USTL.
