2021 Volume 61 Issue 3 Pages 902-910
The obvious feature of local dry underwater welding is the local drainage of the welding area. The choice of drainage method is one of the important factors affecting the quality of underwater welds. Based on hydrodynamics and thermodynamics, the internal gas flow characteristics and arc combustion of the micro-drainage cover with dual-gas curtains is used to simulate which could not be directly observed in the micro-drainage cover with FLUENT, and then the influence of micro-drainage cover on the weld formation of local dry underwater welding is studied. The simulation results show that under the action of intake mode and channel structure, the drainage gas forms a spiral gas wall with a certain “stiffness” around the arc combustion area, which can both protect the stability of the internal dry environment, and cut down the interference of the low drainage pressure on the flow of the shielding gas. However, the low drainage pressure can not effectively achieve the internal drainage. That is, when the pressure of drainage gas exceeds 0.3 MPa, due to the Bernoulli effect, the turbulence situation is intensified owing to the influence of shielding gas, and the arc column fluctuates greatly, which is consistent with the results of underwater weld formation, and provide theoretical basis and basic data for further study of local dry underwater welding.
Local dry underwater welding is a relatively new type of underwater welding. It not only has the flexibility and low cost similar to underwater wet welding, but also has the same effect of weld formation as underwater dry welding, and it has excellent versatility and environmental adaptability, and has obvious advantages in the application of underwater automatic welding.1)
The important feature of local dry underwater welding is to realize the local drainage in the arc combustion area. The drainage device is the main part of the drainage work. Hamasaki2,3) et al. first used the water curtain structure to undertake the drainage task of local dry underwater welding work in the 1970s, and carried out the welding experiment at a depth of 0.3 m under water. Then, in order to solving the problem of low “stiffness” of the water curtain, the external water curtain was changed into a steel brush. Kielczynski et al. putted forward the scheme of moving gas chamber in 1986, which meant sealing the area to be welded and draining the water and then welding the area in turn.4) Due to the high temperature of the drainage device, its structure is basically made of opaque metal, which is impossible to obtain the internal image directly. Researchers generally use the simulation mode to analyze the gas flow form and multi physical field coupling environment under the internal structure of the drainage device. Li et al. believed that the drainage effect and fluid state in the drainage device have a crucial effect to the stability of the welding process.5) Through the Fluent,6) the drainage process of the circular drainage device model with different air intake and exhaust methods is simulated. The distribution of phases and the fluid state of the welding area were analyzed. Finally, the fluid state of the annular outlet-uniform inlet drain hood model in the welding area was considered to be the most stable. Gao et al. developed a micro-drainage cover with a porous media partition, and used Fluent to study the temperature, velocity, and pressure of the arc plasma under the influence of high pressure gas.7) Later, the micro-drainage cover had been redesigned so that it can be used for underwater local dry horizontal welding, and the characteristics of the droplet transfer under the effect of lateral wind were studied using high-speed photography technology.8)
However, the problem lies in the single-layer air curtain of the drainage device. Although this kind of structure can cater for the drainage effect, the actual results of underwater welding needs to be improved for the failure of effective isolation of arc combustion. As the upgrade of single-layer drainage cover, the dual-gas curtain drainage cover is generally divided into two layers of internal and external structure. Guo et al. applied a micro-drainage cover with a dual-gas curtain in the underwater local dry laser welding, and studied relationship between the gas flow rate and the protective conditions of drainage cover by characterizing the formation effect of the welds, the pores in the weld, and the mechanical properties of the joint. Finally the reliability and enhanced isolation effect of the dual-gas curtain drainage cover has been proven.9)
The drainage device used in this study also uses a circular dual-gas curtain structure which refers to the Laval pipe principle in order to more effectively improve the protection and isolation effect of the drainage cover. By using a 3D software, a 3D model of the internal flow channel of the micro-drainage cover was drawn. Meanwhile, Fluent was used to simulate the gas flow pattern of the dual-gas curtain, and to investigate the arc combustion under different gas pressure parameters. In addition, drainage experiments and local dry underwater welding experiments under various gas pressure parameters were performed to analyze the underwater welds formation mechanism under the action of gas pressure in the micro-drainage cover.
Based on the existing drainage device, by improving the gas intake method, a micro-drainage cover with dual-gas curtain was developed in this research.10) The inner layer still follows the existing onshore torch structure, and the outer layer is a drainage gas channel based on the Laval pipe structure. The cross-sectional radius is determined by the Vickers equation11)
| (1) |
In the formula: de is the diameter of the nozzle throat, which is determined by the size of the welding torch; d0 is the diameter of the outlet; C is the shrinkage ratio C = (d0/de)2; l is the vertical height of the curve. The top view is shown in Fig. 1.
Schematic cross-section (a) and top view (b) of the micro-drainage cover. (Online version in color.)
As shown in Fig. 1(a), the shielding gas is introduced into the micro-drainage cover from the vertical middle position above, which is the same as the way of welding the shielding gas onshore. The drainage gas flows into the micro-drainage cover through the four inlets, as shown in Fig. 1(b). In order to adjust the parameters of the shielding gas pressure and drainage gas pressure, the supply of shielding gas and the supply of drainage gas are separated.
The direct object of gas is the arc. Before investigating the effect of gas on the arc, a detailed research of the gas flow state in the drainage device must be conducted. According to the change of the gas flow form, the effect on the arc should be studied.12)
3.1. Establishment of Simulation ModelThe inlet boundary of the simulation model was set as the velocity boundary conditions. During the welding process, the pressure gauge is connected in series with the flowmeter to obtain the flow parameters of the gas of the welding torch and at the inlet of the drainage device under the corresponding pressure, namely, the protective gas flow and the drainage gas flow. The pressure gauge and the flowmeter were connected in series to obtain the flow parameters of the gas at the inlet of the welding torch and the drainage inlet at the corresponding pressure, that is, the protective gas flow and the drainage gas flow. Through formula 2, the velocity parameters of shielding gas and drainage gas in the micro-drainage cover were obtained.
| (2) |
Where, v is the gas velocity, Q is the gas flow, and d is the pipe diameter through which the gas flows.
Through experiments, and according to the conversion relation of formula 2, the corresponding relationship between gas pressure and gas velocity is arranged as shown in Table 1.
| Gas pressure (MPa) | Gas flow (L/min) | Inlet velocity (m/s) | ||
|---|---|---|---|---|
| Drainage gas | Shielding gas | Drainage gas | Shielding gas | |
| 0.10 | 42.15 | 30.25 | 6.211 | 3.396 |
| 0.15 | 54.85 | 36.53 | 8.083 | 4.102 |
| 0.20 | 61.15 | 42.20 | 9.011 | 4.738 |
| 0.25 | 65.00 | 44.50 | 9.579 | 4.996 |
| 0.30 | 70.05 | 47.80 | 10.323 | 5.37 |
| 0.35 | 75.12 | 50.54 | 11.070 | 5.67 |
| 0.40 | 79.27 | 54.02 | 11.681 | 6.07 |
A three-dimensional model of the internal flow channel of the drainage device was created by the three-dimensional modeling software Creo 3.0. The model is composed of two parts, internal cavity area (yellow part) and slag area (brown part), as shown in Fig. 2.
Three-dimensional model of internal flow channel of micro-drainage cover. (Online version in color.)
To be specific, firstly inputted the three-dimensional internal flow channel model into the fluid analysis software Fluent 17.0, and then used a VOF (Volume of fluid) model based on the Euler method. As the figure shown, fluid analysis model Inlet 1, having a total area of 113.1e−06 m2, was composed of four pagoda-shaped drainage gas inlets, with the diameter of 6 mm. Inlet 2 of the fluid analysis model was the arc shielding gas outlet nozzle, with a total diameter of 10.5 mm, comprising a conductive tip with a diameter of 6 mm in the middle, which contributed to a total area of 148.440e−06 m2. The outlet, with a standard height of 0.5 mm, was used to release the gas from gap of the soft sleeve and the workpiece. In the experiment, the velocity of the gas at the pressure was used as the velocity conditions for both Inlet 1 and Inlet 2 boundaries. The outlet type was set to Pressure-outlet. Since the experiment was made under the water of 0.2 m in depth, the outlet pressure of all calculation models was set to 2 kPa. Dominated by the tetrahedral meshes, the unstructured meshes were used to make a distinction. In solution methods, the default SIMPLE solver of Pressure-velocity Coupling Scheme was selected, and the Standard model was used in pressure interpolation algorithm. Regarding the momentum equation, Second Order Upwind format was used. In contrast to the small proportion of CO2, argon gas is mainly used in the actual practice. Therefore, only argon was added in the software.
3.2. Flow Characteristics of Drainage GasFirst, in order to verify the role of the drainage gas in the micro-drainage cover, the drainage gas was introduced through four symmetrical inlet boundaries Inlet 1, and the model with only drainage gas was calculated. At this time, the flow velocity of the drainage gas was settled to 6.211 m/s, namely, the flow velocity of the drainage gas at a pressure of 0.1 MPa with 1000 iteration cycles were set and the single calculation time of the model was about 4 h, which led to converging calculation results. Set the boundary Inlet 1 as the starting point of the streamline. The streamline of the gas is shown in Fig. 3 and the outline of the inner flow channel is shown. When the drainage gas entered the internal cavity area, the streamline was spirally accelerated. The direction of streamline always has a certain angle with the Y axis (vertical direction), and the velocity reaches the maximum before and after entering the Slag area.
Simulation results of gas flow in the micro-drainage cover. (Online version in color.)
Figure 4(a) shows the gas flow form in the Slag area. The gas flow form in the slag area after removing the 3D structure of the periphery is shown in Fig. 4(b). When the gas entered the Slag area, it filled the area and the water would be drained. At the same time, a spiral-shaped circular “gas wall” with a certain thickness is formed around the arc area. There is almost no streamline inside the circle, that is, the interference of the drainage gas on the internal arc combustion area is small, ensuring the stability of the arc combustion area. The gas flow circulates in the slag area to effectively empty the water in the micro-drainage cover and prevent the external water from entering the drainage device, so as to isolate the arc combustion area from the outside.
Flow pattern (a) and streamline diagram (b) of gas in the slag area. (Online version in color.)
The major interference of the external environment on the interior of the drainage cover is water. In order to prevent water from permeated into the drainage cover, it is necessary to ensure that when the drainage gas flows through the outlet, the pressure difference between the internal and external environment is positive. Therefore, the pressure difference data between the inside and outside of the micro-drainage cover should be obtained to determine whether the pressure data used by the micro-drainage cover meets the requirements. Figure 5 shows the distribution of the pressure difference on the cross section of the xy axis of the flow channel in the micro-drainage cover. It can be seen from this figure that the pressure difference between the arc combustion area and the external environment was the largest, which ensured a better shielding effect during underwater arc combustion. However, the internal and external pressure difference at the outlet position was negative, which meant a failure of a complete isolation to the external water environment. In other words, water would seep into the drainage device and affected the arc combustion. Therefore, after that, the inlet pressure of the drainage gas was continuously increased to 0.2 MPa. When pressure difference in the internal and external at the outlet position was positive, Slag area would be protected from the surrounding water going through the outlet.
Pressure difference between the inside of the micro-drainage cover and the outside. (Online version in color.)
The micro-drainage cover is in a non-transparent closed state and cannot be viewed directly. In order to analyze the flow form of drainage gas in the micro-drainage cover, a set of underwater drainage experiments were designed. The drainage device was placed vertically above a piece of transparent glass, with a distance of 0.5 mm. The underwater camera was positioned below the micro-drainage cover and was in line with its central axis. The camera is equipped with CMOS sensor and has 4 K video shooting function, with a the frame rate of slow-motion shooting of 960 FPS(frames per second). Additionally, it was provided with enough light source and adjustment of the focal length to ensure that clear underwater images can be obtained. The schematic diagram of the experiment platform is shown in Fig. 6.
Schematic diagram of underwater drainage experiment platform. (Online version in color.)
The shooting results of the drainage gas working process are shown in Fig. 7.
The shooting results of the drainage gas working at different times; (a) 0.489 s, (b) 0.506 s, (c) 0.591 s, (d) 0.658 s, (e) 1.157 s, (f) 2.053 s. (Online version in color.)
Before the drainage gas reaches the inside of the micro-drainage cover, the slag area was filled with water. As the gas drove the flow of water, the form of water flow could be used as the evaluation standard. After the gas had entered the micro-drainage cover, it could be observed that the water in the slag area began to circulate. But at this time the gas was not enough to drain the water, and no trace of liquid outflow was observed around the drainage device, as shown in Fig. 7 (a). With the arrival of a large amount of gas and the gradual decrease of the fluid in the micro-drainage cover, the water in the internal cavity area immediately gushed into the slag area. However, the arc combustion area was clearly visible, indicating a greater pressure in this area, as shown in Figs. 7(b), 7(c). As the continuous gas flowed from the inlet to outlet, the water inside the drainage cover had been pumped near the outlet and exhausted in sequence. It can be seen that the area filled by the water in Fig. 7(e) had sharply decreased comparing to that in Fig. 7(d). After about l s when the internal balance of the micro-drainage cover had been reached, whereas water seepage occurred in the contact part of the shell and the glass plate, as shown in Fig. 7(f). The experiment results were similar to the simulation results from Fig. 5, of which the accuracy was confirmed.
Therefore, the drainage gas guided by the internal structure of the micro-drainage cover could realize the drainage of the slag area and the isolation of the arc combustion area. Despite the results above, the external water may have chance entering the drainage cover when the gas pressure is not sufficient.
3.4. Simulation of the Influence of Drainage Gas Flow on Shielding Gas FlowThe micro-drainage cover adopts the dual-gas curtain, with a shielding gas of inner layer and a structure equivalent to the nozzle of onshore welding guns. During welding, shielding gas and drainage gas cooperate with each other and do not interfere with each other without interference, so as to achieve arc protection and drainage isolation respectively, which the welding environment is similar to that onshore. Therefore, it is necessary to investigate the flow behavior of the two channels of gas in the micro-drainage cover. Four sets of experiment parameters were designed to study the influence of drainage gas on inner protective gas, namely, the boundary conditions of Inlet 1 and Inlet 2, as shown in Table 2.
| No. | Pressure (MPa) | Inlet1 (m/s) | Inlet2 (m/s) |
|---|---|---|---|
| 1 | 0.1 | 6.211 | 3.396 |
| 2 | 0.2 | 9.011 | 3.396 |
| 3 | 0.3 | 10.323 | 3.396 |
| 4 | 0.4 | 11.681 | 3.396 |
The diameter of the shielding gas flowing through the nozzle remained constant, but the diameter of the drainage gas from the inlet to the outlet changed greatly. According to formula 3, the instantaneous velocity of the drainage airflow into the Slag area can be calculated from the total cross-sectional area of the drainage gas at the inlet and the cross-sectional area of drainage gas instantly flowing into the Slag area.
| (3) |
In the formula, vn represents the flow velocity of the drainage gas at the inlet, s1 represents the total area of the inlet, s1 = 113.1e – 06 m2, vm represents the flow velocity of the gas when flowing into the Slag area, and s2 represents the cross-sectional area when flowing into the Slag area, and s2 = 106.029e – 06 m2. After calculation, the instantaneous velocity of the drainage gas flowing from the internal cavity area into the Slag area is 6.609 m/s, 9.588 m/s, 10.984 m/s and 12.429 m/s respectively when the inlet pressure of the drain gas is 0.1 MPa, 0.2 MPa, 0.3 MPa and 0.4 MPa.
From the parameters in Table 2, the velocity vector simulation results on the longitudinal cross section of the gas velocity under four groups of gas pressure parameters are shown in Fig. 8. Taking into account the factors mentioned above, the velocity displayed by the gas when it entered the slag area from the internal cavity area was literally consistent with the actual calculation result.
Simulation results of cross section of gas inside the micro-drainage cover: Fig. (a), (b), (c) and (d) in sequence are vector diagrams of cross section velocity distribution of the micro-drainage cover under No. 1, 2, 3 and 4 experiment parameters in Table 2. (Online version in color.)
As known, when the fluid flows in the pipeline, the faster it flows, the lower pressure it gets. The relation between the velocity and the pressure is
| (4) |
In the formula, P is the pressure at a point in the fluid, v is the velocity of the fluid at that point, ρ is the density of the fluid, g is the gravitational acceleration, h is the height of the point, and C is a constant. This formula is called Bernoulli formula.13)
During underwater welding process, two channels of gas are required to enter the slag area at the same time. When the velocity of drainage gas had augmented, the difference between the velocity of shielding gas and that of drainage gas gradually increased, which according to formula 4, would cause disturbance to the flow of shielding gas. Moreover, the Bernoulli effect will be intensified when the flow space of two channels of gas is small. So it is necessary to check the streamline diagram of shielding gas. The CFD-post results of shielding gas flow under the action of drainage gas are shown in Fig. 9.
Three - dimensional streamline diagram of shielding gas inside the micro-drainage cover: (a), (b), (c) and (d) are the streamline of No. 1, 2, 3 and 4 groups of experiment parameters in Table 2 respectively. (Online version in color.)
Figure 8(a) shows that under the double action of shielding gas and drainage gas, the arc combustion area is surrounded by a dual-gas curtain. The flow velocity of shielding gas in the nozzle was almost the same, and the drainage gas was accelerated obviously in the flowing process. When the inlet pressure of the drainage gas was 0.1 MPa, it could be seen from Fig. 9(a) that the flow process of the shielding gas performed a typical laminar flow and almost unaffected by the drainage gas. But, according to section 2.2, when the inlet pressure of the drainage gas is 0.1 MPa, the pressure difference of internal and external at the outlet is negative, which cannot ensure the isolation to the external water environment. It means that the surrounding water environment was likely to affect the arc combustion and underwater weld formation. When the inlet pressure on Inlet 1 had been increased to 0.2 MPa, though isolation to water could be achieved as the pressure difference in internal and external was positive, the flow of shielding gas had showed a slight change. As shown in Figs. 8(b) and 9(b), the shielding gas flowing in the arc combustion area had a trend to spread to the surrounding area, whereas no obvious interference by drainage gas was seen. When the inlet pressure of the drainage gas had been increased to 0.3 MPa, the pressure difference between the micro-drainage cover and the external environment became larger, and both the drainage effect and drainage velocity were improved. Conversely, the simulation results showed that the disturbance caused by the flow of the drainage gas to the flow of the shielding gas had been witnessed an obvious disturbance, which can be seen in Figs. 8(c) and 9(c) that the streamline began to oscillate and thus led to the increase of the amplitude and the generation of the transient flow around the shielding gas. When the pressure of the drainage gas had been increased to 0.4 MPa, the streamline of the shielding gas was no longer clear and was hard to distinguish, and vortices began to appear in the flow field. As shown in Figs. 8(d) and 9(d), the shielding gas could not participate in the arc combustion as usual because of the dominant position of the turbulence caused by the drainage as well as the destruction of the laminar flow. From above, in spite of the dry internal environment of the micro-drainage cover, the quality of weld formation will be affected under the interference from drainage gas for the arc combustion has been interrupted and thus destroyed the transition process of droplet.
The drainage experiment results for each parameter in Table 2 are shown in Fig. 10. In Fig. 10(a), it can be clearly observed that there was water around the rubber sleeve, which indicated that under equilibrium, water may still permeate the interior of micro-drainage cover, and the continuous air flow could not completely evacuate it. With the increase of shielding gas pressure, water seepage was effectively alleviated. As shown in Fig. 10(b), when the pressure of drainage gas had been increased to 0.2 MPa, only the drainage of water could be observed around the rubber sleeve, which meant that no water had penetrated into the micro-drainage cover. But some water droplets remained on the transparent glass surface, as shown by the arrow. When the pressure of the drainage gas had been raised to 0.3 MPa, the gas flow rate was accelerated with a shorter time on drainage and less water droplet left on the transparent glass, which indicated a dryer environment than before. When the pressure of the drainage gas had been increased to 0.4 MPa, the drainage time was the shortest and the drying degree was the highest among the four experiments. However, a long time before the equilibrium state was reached, the “water column” surrounded by the drainage gas could still be observed in the arc burning area. The “water column” rotated and diffused to the surroundings, and then disappeared, which indicated that the moment the drainage gas had entered the Slag area, the faster velocity of gas flow prevented the flow of water that previously occupied the arc combustion area. With the cooperation of the protective gas and the Bernoulli effect, the trapped water was slowly drained until it finally disappears. But the residual water would affect the arcing of the arc.
During underwater welding, there were various physical fields such as water, electricity, liquid, magnetic field and gas in the micro-drainage cover. Because the internal gas flows and the arc combustion state could not be directly observed inside the device, numerical simulation is the common verification for analyzing the combustion state of the arc.14)
The MIG welding arc system is taken as the physical prototype of mathematical modeling, and the model is simplified to make it easily converge and shorten the calculation time for the premise of meeting the simulation requirements. The calculation model is shown in Fig. 11. The application of the governing equations is based on the Navier-Stokes governing equations and Maxwell’s equations which are for incompressible fluids in a three-dimensional rectangular coordinate system;15) Properties of argon material were defined, including density, conductivity, viscosity, specific heat capacity of constant compression, thermal conductivity and its function to temperature.16,17)
Two-dimensional arc model and boundary conditions.
Multi-region quadrilateral meshing mode was used in the two-dimensional model. Inlet 1 was the boundary condition for inlet velocity of drainage gas, and flow velocity parameters were obtained from the fluid simulation results of the micro-drainage cover. Inlet 2 was the boundary condition for inlet velocity of shielding gas, and the velocity parameter was the inlet velocity of shielding gas under corresponding pressure. The outlet boundary was set to a pressure of 0.002 MPa, which was equivalent to a 20 cm water depth pressure. The boundary conditions are shown in Table 3.
| Boundary name | Boundary type | Pressure p (MPa) | Velocity v/(m·s−1) |
|---|---|---|---|
| Inlet 1 | Velocity inlet | – | 3.396, 3.396, 3.396, 3.396 |
| Inlet 2 | Velocity inlet | – | 6.609, 9.588, 10.984, 12.429 |
| Axis | Axis | – | – |
| Positive | Wall | – | – |
| Outlet | Pressure outlet | 0.002 | – |
Firstly open the energy equation in Fluent to import the UDF (user-defined program) into the model and load it on the boundary Positive. Meanwhile, add other boundary conditions, and select the SIMPLEC algorithm for the pressure-speed coupling mode, of whom the number of iterations are set to 20000.18)
Figure 12 shows the arc combustion in the micro-drainage cover under different gas pressure parameters. According to Fig. 12, as the pressure of the drainage gas had increased, the combustion state of the arc and the pressure in the arc combustion region changed.
The cloud diagram of the simulation results of the model, (a), (b), (c), (d) are the simulation results of arc combustion in the drainage device under the No. 1, 2, 3, and 4 groups of parameters in Table 3, (e), (b), (c), and (d) are the simulation results of the internal pressure difference in the drainage device under the underwater welding test parameters of No. 1, 2, 3, and 4 in Table 3 respectively. (Online version in color.)
When the pressure of the shielding gas and the drainage gas in the micro-drainage cover was both 0.1 MPa, the arc combustion was normal as shown in Fig. 12(a). However, pressure difference of the internal and external at the outlet boundary was negative as shown in Fig. 12(e), which made it difficult to prevent the water from affecting the arc, and thus the occurrence of spatters in the molten pool and blowholes in the weld was inevitable as described in section 2.4. Figures 12(b) and 12(f) are the arc combustion temperature and the simulation results of pressure difference under the cooperation of shielding gas of 0.1 MPa and drainage gas of 0.2 MPa. At this time, the shielding gas and the drainage gas performed well in their respective functions, which means the pressure difference between inside and the outside is positive. Although there was a vortex, the overall pressure difference in the drainage device didn’t show a great change, and the effect of the vortex is small. It showed that after using 0.2 MPa drainage gas, the shielding effect inside the micro-drainage cover was indeed intensified and the arc combustion environment was improved. Figures 12(c) and 12(g) are the arc combustion temperature distribution and pressure difference simulation results after the drainage gas pressure had been increased to 0.3 MPa. At the same time, the internal pressure of the micro-drainage cover changed greatly, and vortex are formed near the outlet. Due to a long distance from the arc, however, only a small amplitude fluctuation of the arc column was caused, and the influence on the arc combustion was limited. When the drainage gas was 0.4 MPa, the velocity difference increased further the moment the shielding gas and drainage gas entering the slag area. At this point, although the arc combustion was not disturbed by the external water environment, it generated a large “saw-tooth” amplitude fluctuation in the arc column, as shown in Fig. 12(d). Under Bernoulli effect, a wide range of vortices were formed into the arc column, as shown in Fig. 12(h), which led to unstable factors in the arc combustion process and the divergence of arc energy density.
3.6. Local Dry Underwater Welding Test and AnalysisIn order to verify results of the simulation and drainage experiment, the MIG welding power supply independently developed by the laboratory was used for the local dry underwater welding experiments. The base material of welding test is AISI 304 stainless steel,19) of which the size is 300 × 100 × 5 mm3, and the type of welding wire is 308 Lsi. The experiment parameters are shown in Table 4.
| No. | Base current (A) | Peak current (A) | Duty ratio | Shielding pressure (MPa) | Drainage pressure (MPa) | Welding velocity (mm/s) |
|---|---|---|---|---|---|---|
| 1 | 70 | 300 | 30% | 0.1 | 0.1 | 7 |
| 2 | 70 | 300 | 30% | 0.1 | 0.2 | 7 |
| 3 | 70 | 300 | 30% | 0.1 | 0.3 | 7 |
| 4 | 70 | 300 | 30% | 0.1 | 0.4 | 7 |
The formation effect of the local dry underwater weld under the parameters in Table 4 is shown in Fig. 13.
Figure 13(a) shows the welding surface formation effect when the shielding gas and the drainage gas were both 0.1 MPa. During the welding process, several arc interruptions occurred, resulting in a lot of spatter and the formation of hump on the surface of the welds, which explained that the water in the surrounding environment permeated into the drainage device and caused interference to the underwater welding process. Owing to the lack of self-protection of the wire used for underwater MIG welding, the arc combustion process was affected and thus caused defects to the weld formation, which is consistent with the simulation results. When the working pressure of the drainage gas had been increased to 0.2 MPa, the weld formation was in a uniform width, the formation quality was good without obvious defects, as shown in Fig. 13(b). This showed that a better isolation from the combustion area was achieved and proved no effect of the surrounding water to the arc combustion during welding process. With the increase of drainage pressure, the shielding effect inside the micro-drainage cover will be further improved, but defects such as hump and spatter also appear in the weld formation, as shown in Fig. 13(c). And according to the simulation results of Figs. 8(c) and 9(c), because the shielding gas around the arc combustion area had generated a certain turbulence, the combustion process of the arc was disturbed to cause the defects of weld formation. When the pressure of the drainage gas had been increased by 0.4 MPa, the normal welding process was too short for only had arc combustion within 3 s of the arc starting, along with spatter and arc interruption. The surface of the weld is rough with obvious slag inclusion, and the effect of weld formation is poor, as shown in Fig. 13(d). Figures 8(d) and 9(d) show the flow of shielding gas in the arc combustion area. As we can see, with the increase of the flow rate of the drainage gas, the laminar flow of the shielding gas was almost completely destroyed, and the turbulent flow occupied the arc combustion area, which the function of the shielding gas could no longer work and the arcing would be affected.
From what has been discussed above, with the increase of pressure of drainage gas, although the isolation effect and the drying condition in the micro-drainage cover are improved, the rapid flow of drainage gas gradually destroys the laminar flow of shielding gas, with intensified turbulence and the worse formation effect of the weld due to the existence of the Bernoulli effect.
(1) During the underwater welding, both reasonable design of the structure and appropriate parameters of gas pressure are required for the realization of local area drainage. Only when the shielding gas and drainage gas have fully played the respective roles in the drainage device of the double air curtain structure, the ideal weld formation can be obtained.
(2) The drainage gas flows in the micro-drainage cover, forming a rotary air curtain with certain stiffness in the arc combustion area and separates the internal and external environment to achieve the internal arc combustion similar to the welding environment onshore.
(3) When the pressure in the drainage gas is lower than 0.1 MPa, the pressure difference between internal and external formed at the outlet is not enough to completely isolate the surrounding water environment. However, when the pressure in the drainage gas is over 0.3 MPa, although the shielding effect is improved, the interference between the drainage gas and the internal shielding gas is caused and the combustion of the arc will be affected by the Bernoulli effect.
The authors gratefully acknowledge the financial support for this work from National Natural Science Foundation of China (No. 51875212), Science and Technology Planning Project of Guangdong Province (No. 2018A030313192, No. 2019B090919002), and International Cooperation Project of Guangzhou City (No. 201807010035, No. 2019070110006).
