Research on Process Characteristics and Properties in Deep-Penetration Variable-Polarity Tungsten Inert Gas Welding of AA7075 Aluminum Alloy

Abstract

In this study, a new deep-penetration variable-polarity tungsten inert gas (DP-VPTIG) welding process, which is performed by a triple-frequency-modulated pulse, was employed in the welding fabrication of 8 mm AA7075 aluminum plates. The electric signal, arc shape, and weld pool morphology of the welding process were obtained by means of high-speed photography and an electric signal acquisition system under varying parameters of the intermediate frequency (IF) pulse current. The principle of the arc characteristics and the dynamic mechanism of the weld melting during the process are explained. In addition, the macroforming, microstructure, and microhardness of the welded joints were investigated. The results indicate that, with an intermediate frequency pulse of 750 Hz, the arc displayed a higher energy density and a more effective arc contraction, which improved weld appearance and penetration. Moreover, the impact and stirring action of the arc refined the microstructure grains of the weld center. Therefore, this new welding method is feasible for welding medium-thickness aluminum alloy plates without a groove.

1. Introduction

Aluminum alloy 7075 is favored in rail transportation, aerospace, and machinery manufacturing due to its excellent performance and cost-effectiveness [1,2,3]. The stability and cathode cleaning effect of variable polarity tungsten inert gas (VPTIG) welding make it a popular choice for welding aluminum alloys [4,5]. Nevertheless, the unconstrained welding arc of VPTIG is prone to divergence due to the current-carrying capacity limitation of non-consumable electrodes. The low energy density of the arc results in a shallow welding depth, a slow welding speed, and low production efficiency, significantly limiting its further application [6]. Many researchers have undertaken significant efforts to enhance the quality of TIG welding. Annamalai et al. [7,8], by incorporating ultrasonic-assisted TIG welding, successfully reduced the crack sensitivity of AA7075 welded joints and increased their microhardness. Wang et al. [9] employed mixed gas protection to enhance arc stability and achieve welds of superior appearance. Baskoro et al. [10,11] utilized the active agent TIG (A-TIG) method to study the welding process of 304 stainless steel, improving tensile strength and weld penetration. Ma et al. [12,13] used a laser arc composite action for welding aluminum alloys and employed strong convection stirring to homogenize welded joints. This could reduce Zn loss and improve the weld’s mechanical properties. Dong et al. [14,15] employed high-frequency pulse keyhole TIG (K-TIG) to weld 6 mm thick AISI 444 ferritic stainless steel, achieving complete penetration in a single pass without filler material. However, the structure of the hybrid welding torch becomes more complicated when using laser ultrasonic composite TIG welding and mixed gas-shielded TIG welding. The application of active flux in A-TIG welding is time consuming and compromises both uniformity and accuracy. In contrast, K-TIG is particularly suitable for welding steel materials. When welding complex structural parts that require multiple degrees of freedom and high sensitivity, these composite methods become impractical. In aluminum alloy welding, compared to the aforementioned composite processes, the welding process can be effectively controlled by adjusting the internal current waveform of the power supply. Wang et al. [16,17] demonstrated that, compared to traditional TIG welding, a pulsed current effectively stabilized the welding of 2219 aluminum alloy with a wider process parameter window. Additionally, it improved arc characteristics, refined grains, and enhanced tensile strength.

To further improve the applicability of VP-TIG welding for thick aluminum alloy plates, a new control mode for the electrical signal output waveform of the welding machine is independently designed herein. This mode retains the independent and flexible characteristics of the original operation of the welding torch, while altering the output mode of the composite power supply by changing the circuit structure, thus further improving the arc’s energy state. By controlling the current waveform under three pulse frequencies (low-frequency (1–100 Hz), intermediate-frequency (100–1000 Hz), and high-frequency (20–80 kHz)) [18], three patterns of pulse superposition output and multiple characteristic arc outputs can be achieved, thereby improving arc energy density. This new welding method is called three-pulse deep-penetration VPTIG (DP-VPTIG) and is especially suitable for the welding of medium and thick aluminum alloy plates. Compared to traditional TIG welding, it can achieve a greater welding depth and higher width-to-depth ratios of the welded joints to satisfy diverse welding requirements. Moreover, it can achieve high-efficiency, low-cost, and high-quality arc manufacturing. By using a single-factor experiment, this research aims to investigate the interaction between the pulse frequency-dependent dynamic mechanism of the arc characteristics during the weld melting process and the performance of the welded joint. This paper will provide theoretical support for this new welding method.

2. Experimental Procedure

In the experiment, AA7075 aluminum alloy was selected, and butt welding was performed without a filler material or groove preparation.

Table 1. Chemical composition of AA7075 (wt.%).

Material	Si	Fe	Cu	Mn	Mg	Cr	Zn	Ti	Al
AA7075	0.18	0.34	1.7	0.19	2.6	0.20	5.6	0.03	Bal.

presents the chemical composition of the AA7075 aluminum alloy.

The dimensions of the welding plate were 250 mm × 130 mm × 8 mm. Before welding, the plates were polished with a steel brush, washed with an alcohol solution, and dried. The shielding gas used was 99.99% pure argon. The diameter of the tungsten electrode was 4 mm, and the angle of the tungsten tip was 90°. The corrosion solution used Keller’s Reagent (95 mL H₂O, 2.5 mL HNO₃, 1.5 mL HCL, 1.0 mL HF). In the experiments, the welded joints were subjected to a four-week stabilization period prior to microhardness testing. This time frame allows for some degree of natural aging, as the properties of 7075 aluminum alloy can stabilize in a range of 4 to 12 weeks, depending on the hardness of the material. The hardness and microhardness measurements were conducted following the ISO 6507-1:2018 standard [19]. The microstructure was observed by optical microscope (OM). Hardness tests were performed with a Vickers microhardness tester (MVS-1000D1, Jiemin, Shanghai, China) at a load of 200 gf for 15 s. A total of eight samples were prepared for hardness testing, with two samples taken for each of the four different intermediate frequency (IF) pulse settings (250 Hz, 500 Hz, 750 Hz, and (wt.%) 1000 Hz). For each frequency, hardness measurements were performed on two separate samples to ensure the repeatability and consistency of the data. The hardness value for each frequency was calculated by averaging the results from the two samples (wt.%). For each individual sample, hardness tests were conducted at three locations along the weld: the upper section, the middle section, and the bottom section. The overall hardness for each sample was then determined by averaging the hardness values from the upper, middle, and bottom sections. As shown in Figure 1, test points were taken along the centerline of the joint cross section, with a 0.5 mm interval near the partial melting zone (PMZ) and a 1 mm interval in the weld zone (WZ), heat-affected zone (HAZ), and base metal (BM). The hardness of the base material was tested at five points along the cross section, and the average value was taken as the test result. The main process parameters are shown in

Table 2. The main parameters of the welding process.

Parameter (Unit)	Value
Weld speed (mm/min)	350
Peak-value current (A)	430
Base-value current (A)	258
Peak-value voltage (V)	42
Base-value voltage (V)	25
Inert gas flow rate (L/min)	20
High frequency (kHz)	15
High-frequency current amplitude (A)	10
Negative/positive phase duration ratio (TEN: TEP)	4:1
Intermediate-frequency duty cycle (%)	65
Variable-polarity frequency (Hz)	200
Intermediate-frequency duty cycle (tb:tc)	13:7
High-frequency duty cycle (%)	50

.

2.1. Experimental Equipment

The welding method used in the experiment was three-frequency pulsed current compound square wave VP-TIG welding, also known as deep-penetration TIG welding (DP-VPTIG). The welding diagram is shown in Figure 2. The test setup comprised four sections:

The welding system consisted of a DP-VPTIG welding machine (Huazhi Welding & Measuring Co., Suzhou, China), a welding torch, a water cooling system, a gas shielding setup, a workbench, and a travel control component.

Electrical signal acquisition system: current and voltage signals were collected using a welding current sensor, an arc voltage sensor, and an electrical parameter data acquisition card.

Image capturing system: To capture the arc profiles, a high-speed camera (MV-XG51GM-T) (Yantai Zhirui Image Technology Co., Yantai, China) with a 50 mm fixed-focus macro lens was employed in the vision sensing system. The camera was positioned in line with the welding torch’s movement, with its lens focused horizontally near the electrode tip, and captured images at 1000 frames per second with an exposure time of 100 microseconds. It was observed experimentally that during the variable polarity period, when the current switched from negative polarity to positive polarity, the arc profile gradually stabilized [20]. To ensure the consistency of the results, arc images were collected at the same instant after stabilization. The Nikon CMOS camera (Nikon Image Instruments Co., Shanghai, China), along with a DIGIC X digital image processor (Canon, Beijing, China) and a VSTVS-H1-type industrial lens (Shenzhen Ruishi Automation Co., Shenzhen, China), captured the visual display of the weld pool. Simultaneously, the 1064 nm narrow-band filter removed the interference of the strong arc light. Positioned at a downward angle, the camera was affixed to the welding torch with a bracket to observe the area below the tungsten tip. It followed the movements of the welding torch during the operation.

Central control system: this system controlled the electrical parameters for welding, acquired and processed the images simultaneously, and controlled the trigger for the welding power switch.

2.2. Extraction and Measurement of Arc Profile

As shown in Figure 3a–f, image processing software (Adobe Photoshop.2021) was used to extract and measure the arc profile in order to study the arc shape in detail, including stability and contractility analysis. The arc extraction and measurement process were mainly divided into the following steps:

Altered gray level: the original image was processed using image processing software and segmented into 256 levels based on light intensity.

Noise filtering: image processing software was used to denoise and eliminate interference from noise on the arc during the experiment.

By analyzing the color level of the image, the arc profile was identified. After multiple measurements and comparisons, a gray value of 36 was determined as the threshold for isolating the arc profile.

The resulting profile aligned with the actual arc. Definition of feature size in the arc profile: the arc profile size was defined by the arc length (L), the diameter of the arc’s working end (D_W), the diameter of the arc’s electric extreme (D_E), and the diameter of the tungsten electrode (D_T).

To determine the actual measurements of L, D_W, and D_E, a reference size of a 4 mm electrode (D_T) was chosen. The actual size was determined by measuring the distance in pixels from the tungsten electrode edge and applying the formula X-actual = D_T X-pixel/D_T-pixel (X = D_W, D_E, L) to convert it into a physical measurement.

2.3. Extraction of Energy Level Diagram

The brightness of the arc plasma reflected the energy of the arc and the brightness value was positively correlated with the arc energy [21]. To better analyze the arc, image processing software was used to process the arc image through several steps to obtain the arc energy diagram, as shown in Figure 4a–c:

Image processing software generated the arc’s gray value image.

The core region of the arc’s energy was defined based on the energy core theory, where the intensity of color level 216 determined the core region and the edge value of the region was fixed at 36.

The energy region’s area S was calculated (the size of the area was determined by the number of pixels in the area).

The energy distribution coefficient P was calculated as P = E₁/E (E was the overall arc area, E₁ was the core area).

3. Results and Discussion

3.1. Electrical Signal Characteristics

The power supply for the pulsed DP-VPTIG welder was developed in-house. Figure 5 presents the typical current waveform. The standard variable-polarity current waveform is converted into a low-frequency pulse between 50 and 250 Hz. Additionally, the positive phase is combined with an intermediate-frequency pulse between 100 and 1000 Hz, followed by a high-frequency pulse between (20–80 kHz) superposed in the positive polarity stage.

When the intermediate frequency is set at 250 Hz during a variable-polarity phase, the pulse current transitions from 258 A to 430 A and then back to 258 A. As the frequency increases, both the number of intermediate-frequency current pulses and the current shifts from the base peak increase. During welding, the heat generated is directly proportional to the current’s effective value. A higher effective current value results in greater heat input. Figure 6a–d illustrates that, as the intermediate frequency increases, the number of current rising and falling edges also increases. The current takes time to rise and fall along the edge, which results in a decrease in the effective current value and consequently reduces the heat input.

The comprehensive welding electrical signal cycle (5 ms) of the variable-polarity pulse at 750 Hz is depicted in Figure 7a–e, alongside the evolution of the arc profile throughout the variable-polarity pulse cycle. It can be observed from the arc that, during the positive polarity stage, as the current increases, the arc becomes increasingly brighter. When the base peak is reached, the arc profile reaches its maximum, the brightness is at its highest, the arc column area is straight, and the arc assumes a conical shape. As the current decreases from peak to base, the arc darkens, the arc profile continues to shrink, and the arc stiffness diminishes. During the negative polarity stage, attaching the workpiece to the cathode allows high-energy positive ions from the arc plasma to strip away the oxide layer on the surface of the aluminum alloy, while the voltage drop at the cathode enhances oxide layer removal [22]. The arc edge region in the image in Figure 7f shows the presence of cathode spots. Once the oxide layer beneath the cathode spot is removed, the cathode spot naturally moves to the next oxide layer near the weld. Consequently, the mid-arc profile slightly enlarges during the negative polarity process and takes on a bell-shaped form.

3.2. Base-Value and Peak-Value Arc Characteristic

Figure 8 details the arc projection size D_W, electric extreme size D_E, and arc length L corresponding to the arc image. As the current reaches its peak value, the arc profile significantly expands due to the intermediate frequency. The dimensions of D_W, D_E, and L are 12.41 mm, 2.53 mm, and 5.81 mm. When the welding current is set to t_b, the D_W, D_E, and L decrease to 8.57 mm, 2.17 mm, and 5.55 mm. The weld pool surface shows the greatest size alteration (3.84 mm) directly from D_W, with D_E and L showing smaller changes (0.36 mm and 0.26 mm). Figure 9a–c demonstrates that the intensity of the arc’s light energy is a result of the interaction force between arc plasma particles, which can be seen from the arc’s shape and optical properties.

According to optics and image processing theory, the light intensity of a region is determined by multiplying the gray value by the area. The arc energy can be denoted as E = G·S (E = arc energy, G = gray value, S = regional area). Simultaneously, the arc energy is converted from the electrical energy supplied by the welding power source. Thus, the electrical characteristics of the DP-VPTIG process explain the variation of the arc profile between the base and peak values. The current and voltage fluctuates in energy during the base–peak transition. The traditional formula for the conversion of electrical energy to light energy is adopted, which can be deduced using the following Equations (1) and (2):

$$\begin{array}{l}{Q}_p=N\cdot \xi \cdot UIT=N\cdot \xi \cdot {\displaystyle \frac{f_{T_a}}{f_{T_L}}}{\displaystyle {\int }_0^{t_a}{U}_t{I}_t{d}_t}\\ =N\cdot \xi \cdot {\displaystyle \frac{f_{T_a}}{f_{T_L}}}\left[I_{pb}+{\displaystyle \frac{I_{pp}-I_{pb}}{4}}\right]\cdot U_{pb}\end{array}$$

(1)

$$\begin{array}{l}{Q}_b=N\cdot (1-\xi )\cdot UIT=N\cdot (1-\xi )\cdot {\displaystyle \frac{f_{T_a}}{f_{T_L}}}{\displaystyle {\int }_0^{t_a}{U}_t{I}_t{d}_t}\\ =N\cdot (1-\xi )\cdot {\displaystyle \frac{f_{T_a}}{f_{T_L}}}\left[I_{pa}+{\displaystyle \frac{I_{pp}-I_{pb}}{4}}\right]\cdot U_{pa}\end{array}$$

(2)

(Q_p = peak arc energy, Q_b = base arc energy, ξ = carrier duty cycle, N = efficiency of converting electrical energy into light energy, U_t = transient voltage, I_t = transient current, (I_pp − I_pb)/4 = average value of high-frequency current, U_pb = positive voltage in t_b, U_pa = positive voltage in t_c.) The peak and base arc energy are primarily influenced by the parameters of the IF pulse, assuming the current parameters of the high-frequency pulse and low-frequency pulse remain constant. Under experimental conditions, the amplitude of I_p (430 A) is significantly greater than that of I_b (258 A), indicating that the arc energy at t_p is more than four times that at tb. Consequently, the variations in the arc between t_p and t_b are mainly due to the substantial difference in arc energy.

3.3. Analyzing Peak Value of Arc Profile

By processing data in batches to calculate average values, the arc images at various intermediate pulse frequency peak stages and their corresponding arc sizes are shown in Figure 10 and Figure 11. The most significant change occurs in the size of D_W, which initially decreases steadily and then increases slightly as the intermediate pulse frequency increases. Increasing the intermediate pulse frequency from 250 Hz to 750 Hz results in a higher drop rate of D_W, causing its size to decrease rapidly from 14.77 mm to 11.06 mm. As the intermediate pulse frequency increases, the diameter of D_W slightly augments and settles at around 11.25 mm when reaching 1000 Hz. The extreme arc size D_E exhibits a trend comparable to that of D_W, but the impact of intermediate pulse frequency change is significantly smaller due to the restrictions of the tungsten electrode. Increasing the intermediate pulse frequency from 250 Hz to 750 Hz leads to a drop in D_E from 2.66 mm to 2.32 mm. As the intermediate pulse frequency increases, the diameter of D_E slightly augments and settles at around 2.35 mm when reaching 1000 Hz. An increase in the intermediate pulse frequency results in a longer arc length (L). Increasing the intermediate pulse frequency from 250 Hz to 750 Hz leads to an increase in L from 5.46 mm to 5.82 mm. As the intermediate pulse frequency continues to rise, L eventually decreases to 5.77 mm. The energy distribution coefficient indicates that the arc energy density in the core region initially rises and then falls as the intermediate pulse frequency increases. The arc profile experiences a significant reduction followed by stabilization as the intermediate pulse frequency increases during the peak period of the arc. Once the arc reaches 750 Hz, the size of the arc contracts to its smallest size under four parameters. The arc shows a slight size increase beyond 750 Hz, with a simultaneous decrease in the central energy area. However, the core area of the arc expands at 750 Hz, while the overall size of the arc diminishes.

3.4. Analyzing Base Value of Arc Profile

The arc images at the base stage for different intermediate pulse frequencies and their corresponding feature sizes are illustrated in Figure 12 and Figure 13. Elevating the intermediate pulse frequency from 250 Hz to 750 Hz shows that D_W and D_E at the base stage exhibit opposite variations compared to the peak value of the intermediate pulse frequency, whereas L exhibits a similar trend. D_W, D_E, and L increase from 8.62 mm, 2.11 mm, and 4.96 mm to 9.85 mm, 2.61 mm, and 5.56 mm, respectively. Beyond 750 Hz, D_W, D_E, and L begin to increase slowly and eventually settle at approximately 9.88 mm, 2.59 mm, and 5.58 mm. As the intermediate pulse frequency increases, the arc profile initially expands and then reaches a steady state. The largest arc core area and energy are observed at 750 Hz.

3.5. Weld Formation

Figure 14 shows a typical arc profile, molten pool, and weld bead at peak and base values. As the primary supplier of heat and force in welding, the welding arc directly influences the heat and mass transfer processes within the weld pool, ultimately determining the height and quality of the resulting weld. As illustrated in Figure 14a, at the peak time (t_p) in the DP-VPTIG welding process, the arc expands extensively, leading to a wider weld pool, deeper melting depth, and noticeable concave deformation on the weld pool surface. As depicted in Figure 14b, at the base time (t_b), the arc profile shrinks, causing a narrower pool width and reduced weld depth due to lower thermal and force effects, resulting in the disappearance of the concave deformation on the pool surface. Changes in waveform at different intermediate pulse frequencies lead to variations in arc size and arc energy, ultimately influencing the flow of the weld pool and weld formation.

Figure 15a–d illustrate the welding process at different intermediate frequencies and the resulting weld appearance and joint cross section. Observations of the macrostructure show that, at an intermediate frequency of 250 Hz, the surface of the weld is rough and the weld ripples are uneven. The weld surface is heavily oxidized, and element burn loss is evident. Moreover, the weld width is non-uniform, and the weld surface appears to collapse. At 500 Hz, weld ripples begin to appear, showing slight fluctuations. Oxidation is obvious, with more noticeable color changes, and there is a slight undercut. At 750 Hz, the surface of the weld is smooth with an obvious continuous weld ripple, and the weld shows a bright silver appearance. The uniformity of the weld is good with no undercut. At 1000 Hz, the weld is uniform with a good overall appearance, but there is oxidation. None of the weld beads exhibit spatter during the process, the welding process under these conditions stable.

The joints exhibit an obvious depression on the weld pool surface when the intermediate frequency increased, as depicted in Figure 16 and Figure 17. As the frequency increases, the weld depth continuously decreases from 9.51 mm at 250 Hz to 6.33 mm at 1000 Hz. The weld width initially decreases from 14.51 mm at 250 Hz to 12.88 mm at 750 Hz, before further diminishing at higher frequencies. The depth-to-width ratio gradually decreases from 70.5% at 250 Hz to 48.3% at 1000 Hz, with the most significant reduction occurring between 500 Hz and 750 Hz.

According to the above analysis, during welding, the welding arc functions as the primary source of heat and force, directly impacting the heat and mass transfer within the weld pool. The extent and width of the weld pool are directly influenced by the varying thermal effects of the arc at both peak and base intermediate frequencies. Altering the intermediate frequencies leads to concurrent modifications in the peak transformation of the current base value and the thermal effect of the arc, consequently impacting the melting depth and width. As shown in Figure 18, the formation of a welding seam is primarily influenced by the gravitational force P_b of the molten metal, the downward pressure P_c from the arc, and the upward surface tension P_a (surface tension is divided into an inward force (P_a2) and an outward force (P_a1) along the edge of the weld pool). These three forces work together to stabilize the weld pool. The outward flow of molten liquid in TIG welding is caused by the Marangoni effect, which directs the liquid from the center to the periphery of the weld pool. Within the intermediate frequency range of 250–750 Hz, the arc pressure increases, the position where the arc interacts with the weld pool shifts downward, and the axial driving force within the weld pool becomes stronger. This causes the sunken surface of the weld pool to increase in size, thereby decreasing the arc extension at the workpiece and resulting in a reduced weld width. The increase in intermediate frequency leads to a decrease in overall heat input, reducing the effect of gravity and subsequently diminishing the weld depth and width. The reduction in heat input has a greater impact on the depth and width of the weld pool compared to the effect of arc pressure, leading to an overall decrease in weld size. Above 750 Hz, the arc pressure decreases, the heat input decreases, and the radial force of the arc increases. Most of the heat input from the arc plasma is dispersed to the edge of the weld pool, and melts more solid metal near the edge of the weld pool, forming a shallow and wide weld formation.

3.6. Microstructure

Figure 19 depicts the microstructure of the welded joint under low-light conditions at four intermediate frequencies. Multiple datasets are acquired using processing software (Image J 7.0) for grain size measurement. Figure 20 illustrates the evaluation of grain size in the microstructure at each intermediate frequency. This research compares the average grain sizes at 250 Hz, 500 Hz, 750 Hz, and 1000 Hz across the welded joints. It is evident that, as the intermediate frequency increases, the weld grains refine significantly, resulting in smaller grain sizes. At 500 Hz, the grain size decreases by 12.5% compared to 250 Hz and by 28.18% at 750 Hz. Increasing the intermediate frequency to 1000 Hz decreases the grain size of the weld joint by 22.3%. The grain size decreases noticeably with increasing intermediate frequency, but increases again beyond 750 Hz. At 750 Hz, the grain size is the smallest and shows a more uniform distribution. Modifying the intermediate frequency greatly enhances the fluidity of the liquid weld pool and improves convective heat transfer within it, leading to reduced heat loss. The powerful stirring action of intermediate-frequency pulsed current can facilitate the breakdown of the columnar crystals and refine grains, enabling the weld zone to attain an equiaxial structure.

3.7. Microhardness

Figure 21a–d illustrate the distribution of microhardness in DP-VPTIG welded joints of AA7075 aluminum alloy at different intermediate frequencies. The curve exhibits the characteristic of being “low in the middle and high on both sides”. The microhardness of the weld zone is lower, reaching its minimum value at the fusion line. The microhardness of the base material used in the test ranges from 150 to 160 HV, whereas the microhardness of the welded joints under different intermediate frequencies falls between 112 and 128 HV. Among the frequencies tested, the lowest microhardness of the weld joints was observed at an intermediate frequency of 250 Hz, with an average value of 116.1 HV. The highest microhardness was found at an intermediate frequency of 750 Hz, with an average value of 125.26 HV. The average microhardness values at 500 Hz and 1000 Hz were 121.26 HV and 120.65 HV, respectively. In the HAZ, the hardness decreases to a certain extent at different intermediate frequencies. The microhardness in the coarse-crystalline HAZ is higher, whereas the microhardness in the fine-crystalline HAZ decreases to approximately 122 HV. This phenomenon is attributed to the significant thermal effect during welding, which causes most of the second phase θ (CuAl₂) to dissolve in α (Al). As a result, the solid solution strengthening effect is greatly enhanced, leading to higher microhardness in the coarse-crystalline HAZ. Conversely, the thermal effect in the fine-crystalline HAZ promotes the significant growth of θ (CuAl₂), resulting in decreased microhardness values [23,24]. Finally, the most severely softened region is the PMZ. At an intermediate frequency of 500 Hz, the microhardness decreases to about 100 HV, while at 250 Hz and 1000 Hz, it decreases to 110 HV and 117 HV, respectively. The minimum microhardness at 750 Hz is 122 HV. This reduction in microhardness is mainly due to the non-uniform distribution of the Cu element in the AA7075 aluminum alloy after welding. The Cu element in the AA7075 aluminum alloy primarily enhances strength through solid solution strengthening and precipitation strengthening. As for the behavior of the η′ (MgZn₂) phase, it plays a critical role in strengthening the 7075 aluminum alloy. During the welding process, particularly in the heat-affected zone (HAZ), the heat input can cause partial dissolution of the η′ phase into the aluminum matrix. Over time, the η′ phase may precipitate again, depending on the post-weld cooling rate and aging conditions. In the HAZ, due to the elevated temperatures, over-aging of the η′ phase is likely to occur, which would weaken the mechanical properties in this region. This over-aging leads to coarser precipitates, which are less effective in hindering dislocation movement [25,26].

4. Conclusions

In this study, a new deep-penetration variable-polarity tungsten inert gas (DP-VPTIG) welding process, performed by a triple-frequency-modulated pulse, was successfully achieved in the welding of 8 mm AA7075 aluminum plates without a groove. The welding electric signal, arc profile, weld pool morphology, and joint microstructure and microhardness were investigated. The conclusions from the above analysis are as follows:

Analyzed based on typical electrical signals of the welding process: during an intermediate-frequency cycle, the arc profile appeared dark at the base value and bright at the peak value. A peak intermediate frequency resulted in a more intense thermal effect, heightening the number of base–peak current transitions, and reducing the effective current and heat input.

By contrast, the intermediate frequency of 750 Hz was the optimal frequency, resulting in the lowest contraction of arc contours in the peak value, the maximum base value of the arc profile, and the highest value in the core energy region.

The intermediate frequency influenced weld pool oscillation, enhancing the continuity and smoothness of weld formation. By contrast, adjusting the frequency to 750 Hz improved fluidity, refined grain size, and promoted uniform grain distribution and the decrease in hardness was minimized. Additionally, in the heat-affected zone, the η′ phase (MgZn₂) tends to undergo over-aging due to elevated temperatures, leading to coarser precipitates that reduce the alloy’s mechanical properties.