Assessment of the impact of shielding gas flow rate on residual stresses in GMAW-deposited weld beads using LCR wave technique

doi:10.21203/rs.3.rs-4871551/v1

Download PDF

Research Article

Assessment of the impact of shielding gas flow rate on residual stresses in GMAW-deposited weld beads using LCR wave technique

https://doi.org/10.21203/rs.3.rs-4871551/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Welding is a widely employed manufacturing process in the industry for permanently joining pieces. Particularly in fusion welding processes, the high temperatures generate residual stresses upon process completion, rendering the welded and thermally affected region highly susceptible to failures. This occurs because such residual stresses overlay externally applied stresses. Therefore, precise determination of residual stresses in the welded joint becomes fundamental to assessing the actual forces borne by the component and ensuring its operational safety, thus averting unforeseen failures. It is pertinent to point out that welding parameters influence not only the quality and properties of the weld beads but also the magnitude of residual stresses. Among the various process-influencing parameters, the shielding gas flow rate has received minimal attention in the literature, remaining unexplored in terms of its overall impact. To address this gap, this study evaluated the impact of shielding gas flow rate on residual stresses in AWS ER70S-6 weld beads deposited on DIN EN 10025-2 S275JR steel plates. The investigation utilized Gas Metal Arc Welding (GMAW) and the non-destructive ultrasonic method of Longitudinal Critically Refracted (L_CR) waves for stress measurement. Weld beads were deposited via robotic operation, using shielding gas flow rates of 12, 15, and 20 l/min, while keeping other welding parameters constant. Longitudinal profiles of residual stress distribution were obtained for each specimen. The results revealed a significant impact of gas flow rate on the residual stress profile in the weld beads, with the highest values observed for the specimen welded with a gas flow rate of 15 l/min. This study strongly emphasizes the critical importance of evaluating the influence of operational parameters in the welding process on residual stresses in the welded joint, considering their significant impact on the structural integrity of the joined components.

non-destructive testing

LCR waves

ultrasound

residual stress

welding

The welding process is a crucial manufacturing technique used to skillfully join parts made of similar or different materials. Among the numerous welding methods, Gas Metal Arc Welding (GMAW) is one of the most prevalent in industrial applications. Companies continually strive to enhance the efficiency of the GMAW welding process by achieving optimal joints while minimizing financial and time expenditures, as reduced costs lead to increased competitiveness and higher profits. The costs and quality of welds are intrinsically linked to welding parameters, which are variables that can modify the characteristics of the weld bead. Each welding parameter directly influences the final quality of the welded joint, emphasizing the critical importance of their proper selection [1].

While cost reduction is desirable, there is also a focus on improving weld quality to enhance the joints’ ability to withstand external forces. Welded joints with insufficient strength can lead to significant economic losses due to unexpected failures of the welded component [2]. To assess a material's ability to resist external forces, it is essential to know the residual stresses (RS) present within it, thereby determining the actual forces acting on the material [3].

Several studies [4–20] have demonstrated the impact of welding parameters on the mechanical properties and microstructure of steel plate joints welded via the GMAW process. Arc voltage, welding current, welding speed, wire feed rate, welding power, gas flow rate, wire dimensions, and other parameters impact weld quality and resistance. Variation in these parameters affects welding penetration [4, 5, 7, 10, 11, 18], hardness [5, 8, 12, 14, 17], RS [6, 16, 20], mechanical strength [7, 8, 13–15, 17, 18, 20], bead geometry [10, 18] and microstructure [5, 11, 18, 19]. Additionally, the welding method also affects weld quality and strength [14, 21, 22]. Some studies employ numerical models to validate experimental results or try to predict welding process behavior [6, 20–22].

Shielding gas plays a crucial role in the GMAW process, impacting both the metallurgical and mechanical properties of the weld, as well as the overall welding cost. Therefore, selecting the shielding gas and its appropriate application conditions is essential for optimizing welding efficiency. Ideally, the cost of shielding gas should constitute 25% of the total welding cost, considering gas waste [23]. Implementing optimal shielding gas flow conditions can significantly reduce costs, thereby increasing operational efficiency and profitability.

Some researchers aim to optimize this flow parameter, with studies proposing methods such as reducing welding speed to enhance shielding gas delivery efficiency [24]. Other approaches include implementing valves to reduce gas consumption, thereby increasing welding speed and productivity [25]. Additionally, research suggests changes in nozzle geometry [26] and standoff distance [27] to enhance efficiency and productivity. The shielding gas flow rate significantly influences several weld characteristics, including penetration, weld bead dimensions, distortion, and peak temperature [28]. Research indicates that the shielding gas parameters significantly affect the key thermal properties of the weld metal, consequently influencing RS. Lower shielding gas flow rates have shown beneficial results in thermal properties [29]. Besides, shielding gas directly impacts weld geometry, appearance, metal transfer, arc stability, and metallurgical and mechanical properties. Therefore, selecting the appropriate shielding gas and flow rate is crucial for improving productivity, reducing gas consumption, and improving weld properties [30]. Since shielding gas flow rate affects weld thermal properties, it is understood to also influence RS due to their direct relationship. However, there is a lack of research concerning the impact of the shielding gas flow rate directly on the weld beads RS.

Residual stresses in weld beads can be measured using several techniques, such as X-ray diffraction [31–34], neutral diffraction [21, 35], blind-hole method [36], high-density current pulses and speckle-interferometry [38], acoustic birefringence [32], mathematical and numerical methods [38, 39] or ultrasonic method using critically refracted longitudinal waves [40–49] and theoretical approaches [50, 51].

This study utilized the ultrasonic method with Longitudinal Critically Refracted (L_CR) waves to measure RS in AWS ER70S-6 weld beads deposited on DIN EN 10025-2 S275JR steel plates using the robotized GMAW process. The RS in the material was determined by measuring the Time of Flight (TOF) of the L_CR wave. This technique, relatively novel in comparison to others, holds significant promise for RS measuring. Factors affecting wave propagation, such as coupling force, temperature, and the amount of coupling gel, were experimentally determined.

The objective of this study is to explore the influence of shielding gas flow rate on RS generation. This research is motivated by the limited existing literature linking shielding gas flow rate to the RS profile of weld bead, highlighting the significance of this investigation. The goal is to improve the welding process by identifying optimal parameters that minimize costs and yield weld beads with adequate strength under prevailing conditions. Industries aiming to reduce energy consumption and consumable materials in the welding process will find this research valuable.

When sound waves travel from one solid medium to another, a longitudinal wave that hits the surface at an angle equal to the first critical angle (θ_CR1) creates a longitudinally critically refracted (L_CR) wave. This L_CR wave then travels parallel to the boundary between the two media. The critical angle (θ_CR1) required to generate such parallel waves (0°) is calculated using Snell's law, Eq. 1, based on the speed of sound in media 1 (V₁) and 2 (V₂). The L_CR wave's stress measurement principle is based on the nearly linear change in wave speed when stress is applied.

$$\:\frac{sen\:{\theta\:}_{CR1}}{sen\:0^\circ\:}=\frac{{V}_{1}}{{V}_{2}}$$

The L_CR wave's stress measurement principle is based on the nearly linear change in wave speed when stress is applied. Acoustoelasticity studies how this speed varies due to the presence of stresses in the material. Hughes and Kelly [51] developed the basis of acoustoelasticity, deriving expressions for wave velocities in solids under stress using Murnaghan's theory of finite deformation. According to the authors, the variation in ultrasonic wave speed is related to material strain, which, in turn, is related to stress according to elasticity principles.

Additionally, these authors derived expressions for wave velocities in three directions when propagating along the direction of an applied load in an initially isotropic body subject to a homogeneous triaxial stress field. These equations can be rewritten to describe how the ultrasonic wave velocity varies with stress [52]. By using these equations and Hooke's law, Eq. 2 was derived and utilized in this study to establish the relationship between TOF and RS [50].

$$\:d\sigma\:=\frac{E}{{L}_{11}{t}_{0}}dt$$

In Eq. 2, dσ represents the stress change from a reference of null stresses. E stands for Young’s modulus, t₀ is the initial time of flight (TOF) in the absence of stress, dt is the change in TOF due to stress change, and L₁₁ denotes the acoustoelastic constant, which relates the variation in wave velocity to stress, with values obtainable experimentally or from literature. The L₁₁ constant represents velocity variation in the same direction as the applied stress and is most suitable for ultrasonic tests due to its greater sensitivity compared to velocities perpendicular to the loading.

Nine plates of DIN EN 10025-2 S275JR steel (base metal) were machined, each one with 6.35 × 130 × 330 mm. A square groove was machined in the center of each plate to simulate a standardized bevel joint, onto which weld beads were subsequently deposited using AWS ER70S-6 filler metal. The welds are shown in Fig. 1. The mechanical properties and chemical composition of the base material are presented in Table 1.

The GMAW process was employed to weld the nine samples using a Yaskawa MH6 welding robot, a Yaskawa DX100 controller, and a Fronius TransPuls Synergic welding source. The equipment is shown in Fig. 2. AWS ER70S-6 welding wire with a diameter of 1 mm was employed. The only adjustable welding parameter was the flow rate of the shielding gas, which was modified at three different levels (Table 2).

Table 1

– Mechanical properties and chemical composition of DIN EN 10025-2 S275JR steel [49].
S275JR steel
Chemical composition (maximum % of element)		Physical properties
Carbon (C)	0.210	Density	7.85 g/cm³
Manganese (Mn)	1.500	Melting point	1420–1460°C
Phosphorus (P)	0.035	Mechanical properties
Sulfur (S)	0.035	Yield strength	275 MPa
Nitrogen (N)	0.012	Tensile strength	410–560 MPa
Copper (Cu)	0.550	Minimum percentage of elongation before fracture	23%

Table 2

– Shielding gas flow rate in samples.
SAMPLE	GAS FLOW (l/min.)
01	12
02
03
04	15
05
06
07	20
08
09

The gas mixture was 75% Ar and 25% CO₂. Other welding parameters remained constant, including a welding current of 186 A, voltage of 19.1 V, arc length of 10 mm, wire feed rate of 11.8 m/min, travel speed of 800 mm/min, and stick-out of 20 mm.

To perform L_CR measurements with the ultrasonic devices, a flat surface is necessary to ensure proper coupling of the wedges on the samples. Consequently, a machining operation, such as milling, would be necessary on the weld bead. However, this process would modify the RS state within the material and increase the costs associated with sample preparation. For this reason, it was decided to take measurements on the opposite side from where the weld bead was deposited. This approach is feasible because the penetration depth of L_CR waves is approximately one wavelength [50, 52, 53]. Using 1 MHz transducers, the L_CR wave penetrates to a depth of 5.85 mm, which nearly encompasses the entire thickness of the samples (6.35 mm).

Additionally, the ultrasonic method has the advantage that the surface roughness of the material minimally affects the TOF [44]. This occurs because the L_CR wave propagates just a few millimeters below the surface, depending on the frequency of the transducer. Nevertheless, it is crucial to clean the surface to remove any residues to ensure proper wedge coupling. Consequently, all samples were cleaned to eliminate dust and then hand-sanded to remove coarse imperfections and dirt. Different grades of sandpaper (80, 120, and 240) were used for this task.

Another important step is to verify the transducers to ensure accurate measurements. This verification was conducted by comparing the TOF among the three 1 MHz transducers using a standard block (steel IIW block, type 1, model SN 17-3108) and the pulse-echo method. The verification process entails applying 0.5 ml of coupling gel, attaching the transducer to the standard block, recording ten TOF values, removing the transducer, cleaning the gel, and repeating this process three times for each transducer.

The ultrasonic system for measuring RS consists of several components, each serving one or more functions to enable the measurement of the TOF of the L_CR wave. Figure 3 illustrates the system divided into seven sets, and Table 3 describes the system's elements.

Table 3

– Description of the components of the ultrasonic measurement system illustrated in Fig. 3.
SET	ELEMENT	BRAND	MODEL	FUNCTION
1	Chassis	National Instruments	PXI-1071	Provides a platform for hosting and interconnecting multiple PXI module
	Embedded Controller		PXIe-8840	Controls the overall operation of the PXI system
	Oscilloscope Module		PXIe-5114	Acquires electrical signals emitted by a transducer
	Multifunction Module		PXI-6221	Allows the control of the electric actuator
2	Pulser/ Receiver	JSR Ultrasonics	DPR300	Amplifies the signals sent and received to the emitting and receiving transducers
3	Temperature module	National Instruments	PXI-9213	Enables thermocouple couplings and temperature measurement
4	Monitor	HP	L200hx	Transmits processed information from the computer through images
4	Mouse and keyboard	Lenovo	Unknown	Inputs data into the computer
5	Connector Block	National Instruments	SCB-68A	Allows for signal connections
6	Driver	Unknown	L298N	Amplifies the electrical signal
7	Transducers	Technisonic	ABS-0X0408-GP	Transform electric pulses into ultrasonic waves or vice versa
	Electric actuator	Yuyao Huayi Linear Drive Technology	HY01H-12V	Applies the coupling force
	Magnetic base	Digimess	270.240	Allows the fixation of the device on the component to be measured
	Thermocouple	HUA LON	TP01	Measures temperature

The seventh component, referred to as the ultrasonic device, is designed for swift and precise adjustment of the measurement position when affixed to the samples. Apart from the elements described in Table 3, this device (set 7) features a structure comprising aluminum 6063-T5 structural profiles, connectors (corners, screws, nuts, and washers), ballscrew, bearing housings, linear guide, and acrylic wedges angled at 23.8° for transducers coupling, facilitating the generation and capture of L_CR waves. Figure 4 depicts the ultrasonic device.

The time of flight of the ultrasonic wave was measured using the second zero crossing method. This method consists of detecting the first positive peak of the digitized signal from the emitting transducer in the time domain and recording the second crossing of the signal with the horizontal axis (time) after that peak as the TOF of the wave. Signal digitization was done using the PXIe-5114 oscilloscope module with an acquisition rate of 250 MS/s.

In addition, a program was created using the NI LabVIEW® graphical programming platform. This program connects with the oscilloscope board to calculate, display, and save the Time of Flight (TOF) and signal amplitude. It also connects with the external temperature module to display and save temperatures measured by thermocouples, and with the electric actuator controller (multifunction module) to enable control.

An investigation was carried out to evaluate the impact of the amount of coupling gel on TOF and signal amplitude. The objective is to identify the optimal amount of gel required for the measurements. This quantity should be minimized to reduce costs while ensuring effective coupling and transmission of LCR waves.

Three selected amounts of coupling gel were tested between the wedges and the sample: 0.25 ml, 0.5 ml, and 0.75 ml. The tests were conducted on sample 01 at the same measurement position, away from the weld bead. Initially, one of the three gel quantities was applied under each wedge. Subsequently, the coupling force was applied and TOF and signal amplitude recordings were taken every 5 seconds for 1 minute, resulting in 12 measurements. Then, the wedges are lifted, and the gel is removed. This process is repeated three times for each gel quantity, totaling 36 measurements for each amount (108 measurements overall).

The influence of coupling force was determined by measuring the time of flight (TOF) for various electric voltages supplied to the actuator. Initially, 0.5 ml of coupling gel was applied between the transducer and the wedge. Then, the voltage was supplied to the electric actuator, and ten TOF measurements were taken on sample 01 in a fixed measurement position, away from the weld bead. Subsequently, the coupling gel was cleaned. This process was repeated three times for each supplied voltage.

Temperature influence was assessed by conducting TOF measurements ranging from 18°C to 25°C and from 25°C to 18°C, with 1°C increments. Three measurements were taken at each temperature. Temperature variations were achieved using a split air conditioner. Other parameters, such as measurement position (on sample 01), coupling gel volume (0.5 ml), and coupling force (146.7 N, achieved by supplying 5.1 V to the electric actuator), were kept constant to isolate the effect of temperature on TOF.

The TOF correction was conducted using a correction factor obtained from the study of temperature influence. This factor compensates for TOF variations caused by a 1°C temperature change. The correction factor was calculated by subtracting the product of the correction factor and the difference between the measured temperature and the reference temperature (set at 20°C) from the measured TOF. Temperature was measured using three thermocouples positioned at sample 01 and the wedge connecting bar.

The RS measurement method was devised to mitigate the plate distortions arising from welding. As a result, scanning is conducted individually on each side of the weld bead. Additionally, measurements are taken from the opposite side of the weld bead deposition, eliminating the necessity for milling, thereby preserving the original RS state and reducing sample preparation costs. This method, depicted schematically in Fig. 5, enables the acquisition of the longitudinal RS distribution profile across the transverse direction to the weld bead.

The method consists of:

1) Select a sample.

2) Select one side on the bottom of the sample and fix the magnetic bases of the ultrasonic device in the position indicated in Fig. 6.

3) Move the transducers to the initial position, as shown in Fig. .

4) Apply coupling gel and coupling force.

5) Take a measurement of TOF and signal amplitude.

6) Clean the gel and move the transducers 2 mm towards the weld bead.

7) Return to step 4 until 25 measurements are taken. After 25 measurements, fix the ultrasonic device on the opposite side of the weld bead of the selected sample.

8) Complete two scans of 25 measurements on each side of the sample.

Upon completing the procedure, data processing is required. This entails converting TOF measurements into RS, correcting TOF for temperature fluctuations, and organizing the measurements of each sample to obtain the RS distribution profile.

To convert TOF into RS, Eq. 2 was used. The material's modulus of elasticity (E) was set at 210 GPa [54]. The acoustoelastic constant (L₁₁) of 2.58 was adopted based on experimental findings from previous studies [45]. The travel time in the stress-free material (t₀) was calculated by averaging the TOF readings taken at five points furthest from the weld bead on both sides of the base metal. Hence, for each sample, the stress-free material’s TOF was established by averaging the TOF readings from the ten furthest points of the weld bead. These parameters allowed the conversion of TOF (t) into stress according to Eq. 3.

$$\:\sigma\:=\frac{210.{10}^{9}}{\text{2,58}.{t}_{0}}\left(t-{t}_{0}\right)=\frac{\text{8,14}.{10}^{10}}{{t}_{0}}\left(t-{t}_{0}\right)$$

Figure 7 illustrates the surface opposite to the weld bead of one of the samples, before and after preparation for measurements. Visual inspection reveals that surface residues were removed, resulting in a clean surface suitable for transducer coupling. This cleaning process is crucial to avoid potential interference in TOF measurements with the ultrasound device. Any change in the path traveled by the wave could lead to more time and it will influence the results.

Regarding the verification of the transducers, Table 4 presents the TOF values and their standard deviation (SD) with the coefficient of variation (CV) obtained for each transducer.

Table 4

– Results of transducer verification.
TRANSDUCER	TOF (µs)	SD (µs)		CV (%)
N° 1	36.61549	0.00391	0.00704	0.01068	0.01924
N° 2	36.61343	0.00612		0.01671
N° 3	36.61474	0.00977		0.02666

Table 4 shows that the highest SD in measurements from the same transducer was 9.77 ns with a CV of 0.02666%, indicating relatively low and satisfactory variability. Additionally, the SD for the average of the measurements for 1 MHz transducers was 7.04 ns with a CV of 0.01924%, suggesting low dispersion and satisfactory repeatability. The SD of 7.04 ns means about 17.07 MPa, which is quite low when compared with the Yielding Limit of this steel (275 MPa). Therefore, the TOF values obtained using these transducers are quite consistent. Transducers 1 and 2 were selected for further measurements as they demonstrate an even smaller standard deviation and coefficient of variation.

Figure 8 illustrates the impact of gel quantity on both the TOF and signal amplitude. The data shows that TOF increases with the amount of gel applied. This occurs because a thicker layer of gel between the wedge and the sample extends the distance that the L_CR wave must travel, thereby increasing the transit time through the wedges, gel, and material under inspection. On the other hand, regarding the signal amplitude, the data indicates that it increases as the amount of gel applied is reduced. This may be because a thinner layer of gel provides a better coupling between the wedge and the sample, allowing more efficient L_CR wave transmission. Given these results, for all future measurements, 0.25 ml of gel amount was used. This quantity produces satisfactory results comparable to other quantities tested and optimizes the use of coupling gel.

Figure 9 presents the TOF values corresponding to each coupling force. The clamping force of the magnetic bases and the actuator's capacity restrict forces to the tested range. It can be seen in Fig. 9a linear relationship between TOF and applied force, with a correlation factor R² of 0.9861. Furthermore, a ratio factor between TOF and coupling force, in Newtons (N), of -0.2 ns/N was obtained. That is, for each Newton of applied force, the TOF decreases by 0.2 ns. This reduction in TOF is attributed to the decrease in the thickness of the coupling gel layer due to the thinning of the coupling gel layer as the force increases, which shortens the wave path and consequently reduces the travel time. Because the dispersion at the worst case is 18 ns (SD), which means 43.6 MPa, any force level adopted can be adequate. So, we used 146.7 N, which has an SD of approximately 6.8 ns, meaning and stress deviation of about ± 16.5 MPa.

Figure 10 shows the results of TOF measurements at different temperatures, highlighting the temperature’s influence on TOF. Equation y, represented by the dotted blue line, provides a satisfactory model for describing the relationship between TOF and temperature, with a correlation factor R² of 0.9896. Values above 0.95 indicate that the model accurately represents the data behavior. From the slope of the line, a correction factor of 10.1 ns/°C was determined. This factor was used to adjust the TOF based on temperature difference during measurements. Other studies have discovered comparable correction factors with different devices, procedures, and value ranges. Some authors obtained correction factors such as 19.62 ns/°C [55], 11.12 ns/°C [56], and values between 14 and 18 ns/°C [57], using different measurement systems. The larger SD for the whole data set is 20 ns, which means a stress SD of about 48.5 MPa. Because there is no reason to believe that the deviation will be higher than that around 20°C, the reference adopted for all measurements, we used this number in the reliability analysis.

Having determined the main factors influencing TOF, RS measurements on samples were conducted. As mentioned before, 0.25 ml of coupling gel was applied between the wedge and the sample for each measurement, along with an additional 0.25 ml of gel between the transducers and the wedge at the beginning of each scan. A coupling force of 146.7 N was established by supplying an electrical voltage of 5.1 V to the control system.

The results of both RS and signal amplitude (AMP) measurements along the weld are shown in Fig. 11. The graphs reveal an increase in RS as the measurement position moves towards the center of the weld bead, aligning with expectations from the literature [6, 16, 20], which suggests that the maximum tensile RS is theoretically located in the fusion zone at the center of the weld bead. However, compressive regions with negative stress values were not observed. We might have missed detecting these compression areas with the transducers because of the piezoelectric crystal's width. Nevertheless, we took stress measurements from positions on the sides of the samples, which is the main reason for this discrepancy. As a result, the absolute value of the stresses might be slightly different from what the graphs indicate. However, this difference will not affect the search for the best flow rate, as it is similar for every sample.

From Fig. 11, the notably high maximum RS values, ranging from 439 MPa to 592 MPa, warrant attention. These stresses surpass the yield strength of both sheet steel (275 MPa) and welding wire (400 MPa), approaching or exceeding the tensile strength limits of 410 MPa to 560 MPa and 480 MPa, respectively. It is worth remembering that the welding process modifies the material’s microstructure, thereby affecting its mechanical properties. Furthermore, in converting the TOF into RS, the same TOF t₀ was used in the stress-free material, as mentioned before. Besides, the same acoustoelastic constant L₁₁ (2.58) was employed, but it is known that this constant is different in the weld region [45] because the deposited material is different from the base metal and due to the microstructural changes that occur in the welding process. It should also be noted that the influence of grain size on TOF was not considered, and such influence was already noted [55].

Regarding the signal amplitude obtained along the RS distribution profiles, it was noticed that there was a decrease in amplitude in the regions near the weld bead in all cases. This amplitude drop may be related to distortions in the samples. If the contact surface between the wedge and the sample is not flat, only a certain area of the wedge comes into contact with the sample. This decreases the signal intensity, causing a drop in amplitude.

Table 5 shows the maximum RS value found in each sample, in addition to the mean and standard deviation (SD) of the samples welded with the same welding parameters. Uncertainty was combined by taking the square root of the sum of the squares of each previously calculated uncertainty. The method may not account for all existing uncertainties, but it provides a reasonable approximation for values for which their magnitudes can be quantified. Observing the data in Table 5, it is possible to verify that the highest SD and CV occurred in the samples produced with a shielding gas flow of 12 l/min, resulting in values of 16.56 MPa and 3.43%. However, the other SD and CV were close to 9.19 MPa and 2.00%, respectively.

Table 5

– Shielding gas flow rate influence on RS.
SAMPLE	GAS FLOW (l/min.)	RS_MÁX. (MPa)	MEAN RS (MPa)	SD (MPa)	CV (%)
01	12	548.42	483.27	16.56	3.43
02		450.66
03		450.74
04	15	534.45	561.02	16.93	3.02
05		591.99
06		556.62
07	20	438.98	458.91	9.19	2.00
08		440.95
09		496.81

In Fig. 12, the graph of the maximum RS versus the gas flow in welding is displayed. Based on Table 5 and Fig. 12, it is evident that the highest RS was achieved with a gas flow of 15 l/min, while the smallest RS was obtained with a gas flow of 20 l/min. Intermediate RS values were observed for the lowest gas flow rate of 12 l/min. Therefore, within the measured range, no direct or inverse relationship between the maximum RS in the material and the gas flow used in welding was identified.

It can be difficult to compare these results with those obtained by other researchers because there are no studies that directly link RS to shielding gas flow rate. It is important to note that very low flow rates can cause porosity and other issues due to insufficient protection, while high flow rates can create depressions in the weld pool, irregularities in the weld bead, and increased operational costs [58]. Besides, a study was carried out to investigate how shielding gas composition and flow affect weld bead profiles in the Gas Tungsten Arc Welding (GTAW) process. The results showed that changing the gas flow did not have a significant impact on the final dimensions of the weld bead [59].

Moreover, research indicates that the shielding gas flow rate impacts the heat transfer efficiency to the base metal. It has been observed that a reduction in gas flow leads to an increase in peak temperature, greater penetration, and increased distortions [25]. This suggests that heat transfer efficiency improves as the gas flow rate decreases. Another study investigated the influence of gas flow on the thermal properties of weld metal in the GMAW process [29]. The findings indicated that reducing gas flow resulted in a decrease in the coefficient of thermal expansion and an increase in thermal conductivity. Consequently, reducing gas flow leads to reduced distortion and residual stress (RS), as there is less volumetric thermal contraction in the material and faster heat dissipation to regions adjacent to the weld bead. Furthermore, if the flow of shielding gas disrupts the heat transfer, it will ultimately impact the heated area at a specific temperature and the rate at which a specific region cools. These are crucial welding parameters that determine the final appearance of the weld [60].

The higher residual stress measured for the welded samples under the shielding gas flow rate of 15 l/min compared to those welded under flow rates of 12 l/min and 20 l/min can be attributed to the following reasons:

Non-uniform heat distribution: a shielding gas flow rate of 15 l/min may have caused non-uniform heat distribution during the welding process. The process can lead to uneven cooling rates along the weld bead, resulting in greater residual stress formation in certain regions.

Rapid cooling rate: employing a gas flow rate of 15 l/min may have resulted in a faster cooling rate of the weld metal, leading to quicker solidification and potentially higher development of residual stress due to more significant contraction during solidification.

Interference in gas shielding: using a gas flow rate of 15 l/min may have interfered with the effectiveness of gas shielding over the weld pool. This could result in increased oxidation or the creation of inclusions in the welding bead, which could in turn lead to an increase in residual stresses.

Variation in gas composition: changes in the gas flow rate can affect the composition of the shielding gas around the welding arc. Variations in gas composition, such as the argon to carbon dioxide ratio, can influence the metallurgical properties of the weld bead and, consequently, the residual stresses.

In this study, residual stress profiles were obtained from nine samples welded with three different gas flow rates, allowing the evaluation of the effect of this parameter on the residual stress to be determined and achieving the planned objectives. Previous experiments had prepared the samples and assessed the influence of measurement temperature, coupling force, and amount of gel. This work confirms the potential of the ultrasonic technique in measuring residual stress, highlighting its importance as a tool for selecting the best welding parameters. The experimental results also show that:

The gas flow rate of 15 l/min generates the highest RS (average of 561 MPa), while the flow of 12 l/min generates intermediate RS (average of 483 MPa), and the flow of 20 l/min generates the smallest RS (average of 458 MPa).
The time of flight (TOF) decreases linearly as the applied force increases, at a rate of 0.2 nanoseconds per Newton (ns/N).
For the amounts of gel assessed (0.25 ml, 0.5 ml, and 0.75 ml), the higher the amount applied, the greater the TOF and the smaller the signal's amplitude.
Regarding the impact of temperature on TOF, we discovered a correction factor of 10.1 ns/°C using the proposed measurement system.
Regarding the distribution of RS profiles, it was observed that the RS increases in central positions for all welded parts, reaching its maximum in the weld bead, as expected.
Regarding the signal amplitude in different positions, a decrease in amplitude is observed with increasing RS in all cases.

Future studies will aim to search for a more suitable way of selecting the reference stress-free state, which has been a challenge for applications of acoustoelasticity to measure stresses for a long time. This will give the real magnitude of stresses, allowing the use of the method for other applications, where the correct value of the stress in each position is required, as in strength studies. Besides, since the main component of dispersion in the stress measurement by this technique is the measured stress, a search for ways to reduce this dispersion should be performed, as well as a better understanding of how the uncertainties should be combined in this case. Finally, a more detailed study about the reasons why the gas flow leads to a maximum stress magnitude at 15 l/min will be conducted.

Competing Interests

The authors have no relevant financial or non-financial interests to disclose.

Funding

The authors would like to express their gratitude to Bruning Tecnometal for their support and provision of infrastructure and materials necessary for this research. This work was supported by the National Council for Scientific Research (CNPq) for granting a scholarship to the first author through the MAI/DAI Program (grant number 130058/2021-1) and to sponsor Dr. Santos through an ongoing grant (grant number 303582/2023-5). The authors also thank the Secretariat for Economic Development, Science, and Technology (SDECT) of Rio Grande do Sul (Brazil) for the partial financial support.

Author Contribution

Igor Felipe Grzybowski: Conceptualization; Investigation; Methodology; Resources; Software; Validation; Roles/Writing - original draft. Diego Tolotti de Almeida: Conceptualization; Funding acquisition; Investigation; Project administration; Resources. Cristiano José Scheuer: Investigation; Resources; Validation; Roles/Writing - original draft; Writing - review & editing. Paulo Pereira Junior: Resources; Roles/Writing - original draft; Writing - review & editing. Auteliano Antunes do Santos Junior: Resources; Roles/Writing - original draft; Writing - review & editing. Alexandre Aparecido Buenos: Conceptualization; Funding acquisition; Investigation; Methodology; Project administration; Resources; Supervision; Roles/Writing - original draft; Writing - review & editing.

Giridharan PK, Murugan N (2009) Optimization of pulsed GTA welding process parameters for the welding of AISI 304L stainless steel sheets. Int J Adv Manuf Technol 40:478–489. https://doi.org/10.1007/s00170-008-1373-0
Thekkuden DT, Mourad A-HI, Bouzid A-H (2021) Failures and leak inspection techniques of tube-to-tube sheet joints: A review. Eng Fail Anal 130:105798. https://doi.org/10.1016/j.engfailanal.2021.105798
Chiocca A, Frendo F, Aiello F, Bertini L (2022) Influence of residual stresses on the fatigue life of welded joints. Numerical simulation and experimental tests. Int J Fatigue 162:106901. https://doi.org/10.1016/j.ijfatigue.2022.106901
Karadeniz E, Ozsarac U, Yildiz C (2007) The effect of process parameters on penetration in gas metal arc welding processes. Mater Design 28:649–656. https://doi.org/10.1016/j.matdes.2005.07.014
Ibrahim IA, Mohamat SA, Amir A, Ghalib A (2012) The effect of gas metal arc welding (GMAW) processes on different welding parameters. Procedia Eng 41:1502–1506. https://doi.org/10.1016/j.proeng.2012.07.342
Nuraini AA, Zainal ASM, Azmah Hanim MA (2013) The effect of welding process parameter on temperature and residual stress in butt-joint weld of robotic gas metal arc welding. Aust J Basic Appl Sci 7(7):814–820
Sabiruddin K, Bhattacharya S, Das S (2013) Selection of appropriate process parameters for gas metal arc welding of medium carbon steel specimens. Int J Analytic Hierarchy Process 5(2):252–267. https://doi.org/10.13033/ijahp.v5i2.184
Nuraini AA, Zainal ASM, Azmah Hanim MA (2014) The effects of welding parameters on butt joints using robotic gas metal arc welding. J Mech Eng Sci 6:988–994. http://dx.doi.org/10.15282/jmes.6.2014.26.0096
Kumar A, Khurana MK, Yadav PK (2016) Optimization of gas metal arc welding process parameters. Mater Sci Eng 149:012002. https://iopscience.iop.org/article/10.1088/1757-899X/149/1/012002
Srivastava S, Garg RK (2017) Process parameter optimization of gas metal arc welding on IS:2062 mild steel using response surface methodology. J Manuf Process 25:296–305. http://dx.doi.org/10.1016/j.jmapro.2016.12.016
Tawfeek T (2017) Study the influence of gas metal arc welding parameters on the weld metal and heat affected zone microstructures of low carbon steel. Int J Eng Technol 9:2013–2019. https://dx.doi.org/10.21817/ijet/2017/v9i3/1709030272
Neves FO, Moreira Junior AC, Silva ASC (2018) Evaluation of residual stress on steel parts welded and milled. Adv Mater Sci Eng 1–5. https://doi.org/10.1155/2018/4826020
Om Prakash S, Pathmasharma S, Bharath V, Deivasikamani SM (2019) Parameters optimization of gas metal arc welding process (GMAW) by using design of experiments. Int J Tech Innov Mod Eng Sci 5:1–12
Sharma A, Sharma K, Islam A, Roy D (2020) Effect of welding parameters on automated robotic arc welding process. Materials Today: Proceedings 26(2):2363–2367. https://doi.org/10.1016/j.matpr.2020.02.507
Amiruddin MH, Ismail ME, Sumarwati S, Salleh MRM, Noor NAA (2021) Parameters optimization using factorial analysis method for gas metal arc welding (GMAW) Process. AIP Conference Proceedings 2401, 020005, 1–9. https://doi.org/10.1063/5.0073698
Grzybowski IF, Zientarski RR, Buenos AA, Santos Junior AA (2022) Residual stresses in weld bead of low carbon steel plates welded by GMAW using L_CR waves. J Brazilian Soc Mech Sci Eng 44:290. https://doi.org/10.1007/s40430-022-03587-8
Tesfaye FK (2023) Parameter optimizations of GMAW process for dissimilar steel welding. Int J Adv Manuf Technol 126:4513–4520. https://doi.org/10.1007/s00170-023-11356-7
Tukahirwa G, Wandera C (2023) Influence of process parameters in gas-metal arc welding (GMAW) of carbon steels. In: Kumar S (ed) Welding - Materials, fabrication processes, and industry 4.0. IntechOpen, pp. 1–23. http://dx.doi.org/10.5772/intechopen.1002730
Tabaie S, Greene T, Benoit MJ (2023) Optimization of GMAW process parameters for weld overlay of Inconel 686 superalloy on low-carbon steel. Int J Adv Manuf Technol 127:4769–4788. https://doi.org/10.1007/s00170-023-11798-z
Zhang H, Li R, Yang S, Zhan L, Xiong M, Wang B, Zhang J (2023) Experimental and simulation study on welding characteristics and parameters of gas metal arc welding for Q345qD thick-plate steel. Materials 16(17):5944. https://doi.org/10.3390/ma16175944
Colegrove P, Ikeagu C, Thistlethwaite A, Williams S, Nagy T, Suder W, Steuwer A, Pirling T (2009) Welding process impact on residual stress and distortion. Sci Technol Weld Joining 14(8):717–725. https://doi.org/10.1179/136217109X40693
Zhang Y, Liu X, Zhou Y, Shi Y (2022) Influence of welding method on residual stress and metallography of a mild steel welded butt-joint plate. J Constr Steel Res 199:107640. https://doi.org/10.1016/j.jcsr.2022.107640
Mvola B, Kah P (2017) Effects of shielding gas control: welded joint properties in GMAW process optimization. Int J Adv Manuf Technol 88:2369–2387. https://doi.org/10.1007/s00170-016-8936-2
Campbell SW, Galloway AM, McPherson NA (2011) Techno-economic evaluation on the effects of alternating shielding gases for advanced joining processes. Proc IMechE Part B: J Eng Manuf 225(10):1863–1872. https://doi.org/10.1177/0954405411408353
Campbell SW, Galloway AM, Ramsey GM, McPherson NA (2012) A potential solution to GMAW gas flow optimization. Proceedings of the 9th International Conference on Trends in welding research 453–460
Campbell SW, Galloway AM, McPherson NA (2013) Gmaw shielding gas flow optimisation by refinement of nozzle geometry. In: Marquis, F. (eds) Proceedings of the 8th Pacific Rim International Congress on Advanced Materials and Processing. Springer, Cham, pp. 2131–2137. https://doi.org/10.1007/978-3-319-48764-9_265
Bitharas I, McPherson NA, McGhie W, Roy D, Moore AJ (2018) Visualisation and optimisation of shielding gas coverage during gas metal arc welding. J Mater Process Technol 255:451–462. https://doi.org/10.1016/j.jmatprotec.2017.11.048
Campbell SW, Galloway AM, McPherson NA (2012) Techno-economic evaluation of reducing shielding gas consumption in GMAW whilst maintaining weld quality. Int J Adv Manuf Technol 63:975–985. https://doi.org/10.1007/s00170-012-3961-2
Ley FH, Campbell SW, Galloway AM, McPherson NA (2015) Effect of shielding gas parameters on weld metal thermal properties in gas metal arc welding. Int J Adv Manuf Technol 80:1213–1221. https://doi.org/10.1007/s00170-015-7106-2
Mvola B, Kah P (2017) Effects of shielding gas control: welded joint properties in GMAW process optimization. Int J Adv Manuf Technol 88:2369–2387. https://doi.org/10.1007/s00170-016-8936-2
Ranjbarnodeh E, Serajzadeh S, Kokabi AH, Fischer A (2011) Effect of welding parameters on residual stresses in dissimilar joint of stainless steel to carbon steel. J Mater Sci 46:3225–3232. https://doi.org/10.1007/s10853-010-5207-8
Costa ES, Assunção PDC, Dos Santos EBF, Feio LG, Bittencourt MSQ, Braga EM (2017) Residual stresses in cold-wire gas metal arc welding. Sci Technol Weld Joining 22(8):706–713. https://doi.org/10.1080/13621718.2017.1306014
Lai H-H, Wu W (2020) Practical examination of the welding residual stress in view of low-carbon steel welds. J Mater Res Technol 9(3):2717–2726. https://doi.org/10.1016/j.jmrt.2020.01.004
Baltazar-Hernández VH, López-Baltazar EA, Alvarado-Hernández F, Gómez-Jiménez S, Ruiz-Mondragón JJ, Biro E, Zhou N (2022) Surface residual stress analysis in GMAW and LBW of the dissimilar TRIP-DP steels joint: An experimental approach. Metals 12(5):880. https://doi.org/10.3390/met12050880
Paradowska AM, Larkin N, Li H, Pan Z, Shen C, Law M (2014) Neutron diffraction residual stress measurements of welds made with pulsed tandem gas metal arc welding (PT-GMAW). Powder Diffr 29(S1):S24–S27. https://doi.org/10.1017/S0885715614000992
Zhang Y, Liu X, Zhou Y, Shi Y (2022) Influence of welding method on residual stress and metallography of a mild steel welded butt-joint plate. J Constr Steel Res 199:107640. https://doi.org/10.1016/j.jcsr.2022.107640
Lobanov LM, Savitsky VV, Shutkevych OP, Shyian KV, Kyianets IV (2024) Nondestructive method of residual stress determination in welded joints based on application of high-density current pulses and speckle-interferometry. Paton Weld J 1:51–56. https://doi.org/10.37434/tpwj2024.01.08
Gunaraj V, Vel Murugan V (2007) Intelligent welding system for automobile components to optimize angular distortion and longitudinal residual stresses. https://doi.org/10.4271/2007-01-0786. SAE Technical Paper 2007-01-0786
Yang YP (2021) Recent advances in the prediction of weld residual stress and distortion—Part 1. Weld J 100(5):151–170. https://doi.org/10.29391/2021.100.013
Leon-Salamanca T, Bray DF (1996) Residual stress measurement in steel plates and welds using critically refracted longitudinal (L_CR) waves. Res Nondestr Eval 7:169–184. https://doi.org/10.1007/BF01606385
Joseph A, Palanichamy P, Rai SK, Jayakumar T, Raj B (1998) Non-destructive measurement of residual stresses in carbon steel weld joints. Sci Technol Weld Joining 3(6):267–271. https://doi.org/10.1179/stw.1998.3.6.267
Bray DE, Tang W (2001) Subsurface stress evaluation in steel plates and bars using the L_CR ultrasonic wave. Nucl Eng Des 207(2):231–240. https://doi.org/10.1016/S0029-5493(01)00334-X
Belahcene F, Lu J (2002) Determination of residual stress using critically refracted longitudinal waves and immersion mode. J Strain Anal 37(1):13–20. https://doi.org/10.1243/0309324021514790
Chaki S, Ke W, Demouveau H (2013) Numerical and experimental analysis of the critically refracted longitudinal beam. Ultrasonics 53(1):65–69. https://doi.org/10.1016/j.ultras.2012.03.014
Javadi Y, Afzali O, Raeisi MH, Najafabadi MA (2013) Nondestructive evaluation of welding residual stresses in dissimilar welded pipes. J Nondestr Eval 32(2):177–187. https://doi.org/10.1007/s10921-013-0171-2
Joseph A, Palanichamy P, Jayakumar T (2015) Evaluation of residual stresses in carbon steel weld joints by ultrasonic Lcr wave technique. J Nondestr Eval 34:266. http://dx.doi.org/10.1007/s10921-014-0266-4
Geľatko M, Hatala M, Vandžura R, Botko F (2021) Longitudinal critically refracted (L_CR) ultrasonic wave for residual stress measurement. In IOP Conference Series: Materials Science and Engineering 1199(1):012060. https://doi.org/10.1088/1757-899X/1199/1/012060
Grzybowski IF, Zientarski RR, Buenos AA, Santos Junior AA (2022) Residual stresses in weld bead of low carbon steel plates welded by GMAW using L_CR waves. J Brazilian Soc Mech Sci Eng 44(290):1–15. https://doi.org/10.1007/s40430-022-03587-8
Hebaz SE, Agon F, Bouzzit A, Walaszek H, Hodé R, Zhang F, Serfaty S, Wilkie-Chancellier N (2023) Longitudinal critically refracted wave for residual stress assessment on a welded plate using 3D laser vibrometry. 10th Convention of the European Acoustics Association Forum Acusticum 2023, European Acoustics Association, Turin, Italy. 3475–3482. https://dx.doi.org/10.61782/fa.2023.1009
Bray DE, Stanley RK (1997) Non-destructive evaluation: A tool in design, manufacturing, and service. Rev. CRC, Boca Raton
Hughes DS, Kelly JL (1953) Second-order elastic deformation of solids. Phys Rev 92(5):1145–1149. https://doi.org/10.1103/PhysRev.92.1145
Lu J (ed) (1996) Handbook of Measurement of Residual Stresses. Lilburn, Fairmont Press
DIN EN 10025-2 (2004) 2004, Standard Hot Rolled Products of Structural Steels
DIN EN 1993-1-1:2005, Design of Steel Structures, (2005)
Buenos AA, Santos AA, Pereira P, Santos CS (2012) Effect of Mean Grain Size in the Time of Flight for _LCR Waves. Proc ASME 2012 Int Mech Eng Cong and Exp IMECE2012-. 89055:63–69. https://doi.org/10.1115/IMECE2012-89055
Buenos AA, Santos AA Jr, Rodrigues AR, Tokimatsu RC (2013) Application of acoustoelasticity to measure the stress generated by milling in ASTM A36 steel plates. J Braz Soc Mech Sci Eng 35(4):525–536. https://doi.org/10.1007/s40430-013-0049-3
Fraga RS, Santos AA, Andrino MH (2008) Temperature effect on the measurement of stresses in pipelines using ultrasonic Lcr waves. Proc ASME 2008 Int Mech Eng Cong and Exp IMECE2008-. 68720:361–367. https://doi.org/10.1115/IMECE2008-68720
Marques PV, Modenesi PJ, Bracarense AQ (2017) Soldagem: Fundamentos e tecnologia. Elsevier Editora Ltda, 4ª edição, Rio de Janeiro
Anbarasan N, Jerome S (2018) Effect of flow rate and argon-hydrogen shielding gas mixture on weld bead morphology of inconel 718. Materials Today: Proceedings 5(13):26990–26996. https://doi.org/10.1016/j.matpr.2018.09.002
Mishchenko A, Scotti A (2018) Tensões residuais em soldagem a arco: Uma visão holística. Soldagem e Inspeção 23(1):93–112. https://doi.org/10.1590/0104-9224/SI2301.10

Download PDF

Reviewers agreed at journal
15 Aug, 2024
Reviewers invited by journal
14 Aug, 2024
Editor assigned by journal
14 Aug, 2024
First submitted to journal
12 Aug, 2024

You are reading this latest preprint version

Assessment of the impact of shielding gas flow rate on residual stresses in GMAW-deposited weld beads using LCR wave technique

Status:

Version 1

Abstract

Figures

1 Introduction

2 LCR waves and their role in stress measurement

3 Experimental procedures

4 Results and discussion

5 Conclusions

Declarations

Competing Interests

Funding

Author Contribution

References

Status:

Version 1