Ultrasonic-assisted drilling (UAD) is a machining process that is known to improve the hole quality and reduce cutting forces. Previous studies focused on optimizing cutting parameters to improve the hole quality in conventional drilling (CD) and UAD, as well as to finding the optimum vibration parameters (frequency and amplitude) that will increase the effectiveness of the UAD process. However, the influence of cutting tool type during UAD has been largely overlooked. This research aims to address this gap by analyzing the effect of cutting tool type during UAD on the cutting forces and hole quality in GLARE (Glass Laminate Aluminum-Reinforced Epoxy) laminates. Four types of drills, namely, twist drill (TD), double cone drill (DCD), a step drill type 1 (SD1), and step drill type 2 (SD2) with different step length, were selected for this study. The lowest thrust force (47.04 N) and torque (0.079 Nm) were achieved using twist drill, while DCD, SD1, and SD2 exhibited higher thrust forces (12.81%, 20.69%, 41.3%) and torques (94%, 92%, 91%), respectively. In addition, TD produced high-quality holes with lowest surface roughness (Ra 1.66 μm, Rz 10.58 μm) and minimal burr formation (entry burr height 152.3 μm, exit burr height 69.22 μm). Conversely, DCD, SD1, and SD2 showed higher surface roughness Ra (23%, 16%, 24%) and Rz (16%, 37%, 29%), respectively, compared to the TD. Holes drilled using SD1 and SD2 generally had smaller burr height. Overall, UAD system effectively reduced cutting forces at low spindle speed and feed rate. To achieve higher drilling quality, specifically to reduce the surface roughness and exit burr height, a medium spindle speed of 3000 rpm, a feed rate of 225 mm/min is recommended. Drilling at higher cutting parameters using UAD resulted in a decline in hole quality, except for entry burr height.
Notes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
The high-tech industry makes extensive use of cutting-edge materials, such as superalloys, composites, fiber metal laminates (FMLs), and a broad range of nonmetallic materials. When machining these materials using traditional drilling methods, it might be difficult to avoid defects such as edge chipping or fracturing the surface layer, delamination, uncut fiber, burr formation, or even the complete component [1, 2]. Given the numerous dimensions and complicated geometries of the structural components, one-shot drilling of the multi-layer material is required to reduce assembly mistakes and attain close tolerances. More than 30% of all machining tasks are performed using hole processing technology, making it an indispensable tool. Aircraft models like the Boeing 787 and Airbus 380 need more than 10,000 precision-machined holes [3]. GLARE has seen increased application in various areas because of its high specific strength, specific stiffness, and low weight in recent years [4]. Since FMLs have diverse thermal and mechanical characteristics and processing procedures, drilling them presents a significant challenge. Delamination and fiber-matrix damage are common when drilling glass fiber composite [5, 6]. In addition, owing to its low elastic modulus and melting temperature, aluminum alloy is prone to developing an adhering layer, built-up edge, and burrs during the drilling process. Furthermore, the quality of the hole is further compromised when glass fiber layers are placed between the aluminum sheets due to multi-material chip evacuation and variation in thermo-mechanical properties [7, 8].
Ultrasonic vibration machining has been employed by numerous researchers to improve surface integrity during conventional machining [9, 10]. By integrating ultrasonic vibration processing with turning technology, Zhang et al. and other researchers were able to break the critical separation speed of ultrasonic vibration-assisted processing, significantly increasing the speed of ultrasonic-assisted vibration process. By defining optimal cutting parameters, namely, feed rate and phase shift, the researchers managed to orchestrate disengagement of the tool from the workpiece along the feed direction. This precision in parameterization facilitated the observation of distinctive chip formations during various cutting states ranging from broken chips to continuous and partially broken wrinkled chips providing empirical confirmation of the anticipated separation mode [11, 12]. Surface texturing experiments were performed on monocrystalline silicon using a tool developed by Wang et al. [13, 14] that combines ultrasonic vibration-assisted cutting with rapid tool servo technology. Researchers have successfully investigated methods to control and maintain the stability of ultrasonic amplitude during the processing stage. This is crucial to ensure high-quality processing results, considering factors such as temperature, load, electrical compensation, and automatic frequency tracking [15, 16]. Several researchers have successfully used ultrasonic vibration cutting in the drilling industry to achieve high-quality machined surfaces [17, 18]. Parameter optimization is the most common method used by many researchers for improving the quality of the drilling process [19]. A comparative study was done by Giasin et al. [20] for CD and UAD of GLARE laminates, and UAD successfully reduced the thrust force by 70% and burr formation by 80%. Ultrasonic vibration technology has been studied extensively and shown to considerably increase tool life. It was also reported that it increases the hole roundness, decreasing position offset and exit burr height [21]. Tool design is another method that can be used to improve the quality of the machining process and surface integrity. Many researchers conducted different studies to demonstrate the impact of tool design on machining CFRP (Carbon Fiber Reinforced Plastic). Hocheng et al. [22‐24] developed a theoretical model for five different tool types to predict the critical thrust force at the onset of delamination. Drilling experiments were carried out to confirm the theoretical model, and it was concluded that the tool type has a major impact on the delamination. The core drill proved to be the most efficient tool in reducing delamination, whereas the TD demonstrated the least effectiveness. Shyha et al. [25] studied the effect of tool design parameters (point angle, helix angle) on the drilling quality of CFRP with twist and step drills. The uncoated step drill resulted in lower tool wear at a high feed rate but also resulted in higher delamination, whereas the point angle did not show any significant effect on the tool life, thrust force and delamination. Jallageas et al. [26] investigated the effect of drill geometry on machining quality during conventional and low vibration-assisted drilling (LVAD) using a step drill with a helix angle and a step drill without a helix angle. The tool geometries performed differently depending on the drilling process and workpiece material. During the CFRP/Al drilling, the step drill without helix angle performed better regarding hole dimension accuracy for both CD and LVAD. During CD drilling of the aluminum sheet, the step drill with a helix angle performed better in terms of hole diameter and surface roughness Ra, whereas for the LVAD step, drill without a helix angle performed better in terms of surface roughness Ra. The opposite trend was seen in CFRP when analyzing surface roughness. Liang et al. [27] studied the effect of point angle on UAD effectiveness, but the point angle had the same effect on the drilling quality during CD and UAD. Previous literature has shown that with the increase in cutting speed during drilling of composite materials, the effectiveness of UAD decreases, as seen in literature [28‐38]. The reduction in thrust force due to UAD in comparison to CD is minimized at higher cutting speeds, due to the decrease in vibration frequency per cutting length. Other factors also affect the effectiveness of the UAD process, such as tool diameter, feed rate, vibration amplitude, and workpiece material [33, 35, 38], but overall, a decreasing trend was observed with the increase in the cutting speed. Therefore, it can be seen that higher cutting speed is not favorable for UAD, as it will diminish the effectiveness of the ultrasonic process in reducing thrust force compared to CD. It was also observed that increasing the feed rate resulted in a substantial reduction in cutting forces [20].
Advertisement
Numerous parameters influencing the effectiveness of UAD have been extensively studied in the literature, including feed rate, spindle speed, vibration amplitude, and frequency. However, the impact of tool type on UAD effectiveness remains largely unexplored. This research aims to address this gap by conducting a comprehensive investigation into the effect of tool type on thrust force, drilling quality, and the overall effectiveness of UAD. The study utilized four commonly used drill types, namely, TD, DCD, SD1, and SD2, to determine their influence on cutting forces, hole surface roughness, and burr height under CD and UAD. Notably, the work aims to also investigate how different drill designs and types impact the effectiveness of the UAD process. To gain deeper insights into the influence of tool type on the efficacy of UAD, a deliberate choice was made to operate at lower spindle speeds while employing medium to high feed rates for the experimentation. Three levels were selected for spindle speed and feed rate to find the trend when changing cutting parameters using statistical analysis and ANOVA (analysis of variance).
2 Materials and methods
2.1 Workpiece and cutting tools details
A CNC machine tool type 850 made by Zhengdai (Guangdong) Machinery Technology Co., Ltd containing BT40 electric spindle with maximum speed of 10000 rpm was used to conduct the drilling tests. TSINGDING ultrasonic tool holder BT40-ER11-EL was used for ultrasonic drilling experiments as showed in Fig. 1a. A GLARE 2B 11/10-0.4 fiber metal laminate was used as the workpiece material, as depicted in Fig. 1b. The workpiece comprised of eleven aluminum (Al2024-T3) sheets, each with a thickness of 0.4 mm, interspersed with ten fiber layers, each measuring 0.266 mm in thickness. Each fiber layer was composed of two unidirectional S2 glass fiber prepregs embedded with FM94 adhesive oriented at a 90° angle to the rolling direction of the aluminum sheets. Four commonly used tools coated with titanium aluminum nitride (TiAlN) including a TD (standard drill), a DCD, a SD1, and a SD2 were employed. Table 1 listed the drills’ geometrical details. The cutting parameters are summarized in Table 2. To ensure the repeatability of the results, each experiment was performed three times, resulting in a total of 216 experiments.
Fig. 1
(a) Experimental setup, (b) Workpiece GLARE 2B 11/10 0.4. (c)Tools used for the experiments
Table 1
Geometrical details of tools used for experiments
Table 2
Experimental plan to check the effectiveness of UAD on different tool type
Parameters
Level 1
Level 2
Level 3
Level 4
Tool type
TD
DCD
SD1
SD2
Feed rate (mm/min)
150
225
300
-
Spindle speed (rpm)
1000
3000
4500
-
Vibration amplitude (μm)
0 (peak to peak)
20 (peak to peak)
Frequency (kHz)
18.7
×
2.2 Cutting forces and hole metrics measurement
The thrust force and torque were measured using a KISTLER 9272 dynamometer connected to a data collecting system (Type 5697A1) as seen previously in Fig. 1a. The cutting force data was obtained using a sampling frequency of 8 kHz. Thrust force and torque were analyzed using the dynoware software version 3.2.5.0. The thrust force reported in this study is the average reading from the instant the drill is in contact with the workpiece until it fully exits from the other side. The percentage reduction in thrust force and torque are calculated using Eq. (1) [39] where X is the measured value for the thrust force and torque.
The hole surface quality was measured by calculating the surface roughness parameter Ra (arithmetical mean roughness value) and Rz (maximum peak to valley height of the profile). These two surface roughness metrics were measured using a Mitutoyo S-3000 roughness profilometer, as shown in Fig. 2a. The surface roughness assessment adhered to the ISO 4287 standards, utilizing a measured length of 6.4 mm and a traverse speed of 0.20 mm/sec. Finally, the burr height was measured at four 90°-apart locations around the hole by using a 3 µm-accuracy KC-X1000 surface profilometer, as shown in Fig. 2b.
Figure 3 presents the thrust force profiles of the cutting tools for both CD and UAD tests. The results show that the lowest maximum thrust force occurs using TDs. Figure 4 also demonstrates that the patterns of the thrust force profiles for CD and UAD are not similar, indicating that the use of vibration has effect on the thrust force profile pattern. In CD, the thrust force increases steadily until the tool is in full contact with the workpiece, then maintains a consistent level throughout the majority of the drilling process, and then diminishes towards the end. Conversely, in UAD, the initial thrust force is notably higher but decreases upon full tool engagement with the workpiece, stabilizing until just before the end of the drilling. At this point, the remaining uncut thickness becomes smaller, and the effect of vibration on reducing thrust force decreases, and a noticeable increase in thrust force is observed. This trend is consistently observed in both the TD and DCD. However, an alternate pattern can be observed with the SD1 and SD2. For SD1, it can be seen that initially, the thrust force is considerably reduced under UAD but then gradually becomes similar to that under CD when center part exits the workpiece. In contrast, the SD2 shows a significant reduction in thrust force. This can be attributed to its smaller step size and a vertical cutting edge, which enhance its cutting performance in the UAD process.
Fig. 3
Thrust force profile for the different tools at 1500 rpm and 300 mm/min
Fig. 4
Effect of cutting parameters and tool type on thrust force and percentage reduction using UAD compared to CD
×
×
Figure 4 shows the average thrust force for CD and UAD tests under different cutting parameters and drill types. The results show that at a spindle speed of 1500 rpm, the SD1 and TD’s generated the lowest thrust force, followed by the DCD and SD2, respectively. The trend is consistent regardless of the level of the feed rate used. It can also be seen that the step drill outperforms all other drills when drilling at a feed rate of 150 mm/min regardless of the spindle speed level used. This could be attributed to the TD optimized geometry (point angle, chisel edge length) reducing the built-up of material and consequently the thrust force required for drilling [40]. The step drill’s design, with a transition from a smaller to a larger diameter, allows for gradual material engagement with higher step length, which can lead to reduced thrust force compared to other drills [41]. Furthermore, the thrust force generated by DCD’s is comparable to that of TD’s when drilling at spindle speeds of 3000 and 4500 rpm, particularly at lower feed rates of 150 mm/min. The smaller point angle reduces the thrust force, while the larger chisel length increase the thrust force resulting in thrust force similar to that produced by the twist and DCD [42]. The lowest thrust force was recorded when using DCD at a spindle speed of 4500 rpm and a feed rate of 150 mm/min during UAD. Similarly, the highest thrust force was recorded when using SD2 at a spindle speed of 1500 rpm and a feed rate of 300 mm/min.
The results shows that the thrust force increases with the feed rate in CD is higher compared to UAD, which is expected due to increased uncut chip thickness [43]. The results also show that the effectiveness of the UAD process in reducing the thrust force becomes less with the increase of the spindle speed. This was also observed in the only available studies on UAD of GLARE laminates [20, 44]. The results also show that under UAD, the thrust force was reduced—compared to CD tests—by up to 40.70%, 47.27%, 41.63%, and 50.29% using TD, DCD, SD1, and SD2, respectively. The percentage reduction in thrust force using TD ranged between 0.96% and 40.7%; for DCD, it ranged between 0.73% and 47.27%, for SD1 between 2.6% and 41.63%, and for SD2 between 7.19% and 50.29%. From this, it can be also seen that under UAD tests, the SD2 generally had the highest PRTF among the other drill types, especially at highest feed rate and lowest spindle speed. The results in Fig. 5 also indicate that in order to maintain effectiveness of the ultrasonic system, increasing the spindle speed should be accompanied with an increase in the feed rate. Nevertheless, further increase will reduce the effectiveness of the UAD system in reducing the thrust force, and therefore, other hole metrics should be checked against the effect of cutting parameters levels. Therefore, as the effectiveness of the UAD system becomes less with the increase in cutting parameters, the optimal level of cutting parameters should also take into account the energy consumption level to determine whether the improvement in hole metrics can justify the increased energy consumption levels.
Fig. 5
Effect of spindle speed on the effectiveness of UAD based on drill design at feed rate of 300mm/min, (a) TD, (b) DCD, (c) SD1, (d) SD2
×
During drilling, the thrust force due to chisel edge has an influence on the material damage and is mainly affected by the feed rate and to a lesser extent by the spindle speed. The material removal mechanism of the drill bit’s chisel edge involves extrusion subsequent to wedging, resembling a cutting action at a right angle characterized by a negative rake angle. This process leads to a concentration of approximately 40–70% of the total thrust force on the chisel edge. In addition, a shorter chisel edge length is a known effective method in reducing thrust force [45, 46]. The SD2 exhibited a higher maximum thrust force compared to SD1, attributable to an increased chisel edge length, as seen in Fig. 3 and Fig. 4. The center part of the SD1 and SD2 with a point angle of 90° exhibited a higher reduction in thrust force, followed by the TD (140°), while the outer edge of the SD1 (180°) demonstrated a minimal reduction in thrust force, consistent with existing literature [27].
Figure 5 presents the influence of spindle speed on the effectiveness of UAD at a feed rate of 300 mm/min for different drill types. The periodic variation in cutting depth within the UAD process leads to intermittent separation and contact between the drill bit and the workpiece. Simultaneously, the ultrasonic impact generated by the drill cutting edges, combined with the impulsive ultrasonic-separated cutting process between the tool and workpiece, amplifies the cutting efficiency of drill bits [47]. This enhancement contributes to improved chip evacuation, effectively mitigating the friction effect. As spindle speed increases, the effectiveness of UAD diminishes due the decrease in vibration frequency per unit cutting length [39]. Moreover, because of the distinct material properties and chip removal mechanisms between the glass fiber (involving brittle fracture) and aluminum (involving elastic-plastic deformation), the PTFR during UAD of glass fiber layers was observed to be less than that during UAD of aluminum sheets similar to a previous study [20]. The UAD’s effectiveness is high at the beginning of the drilling process because of the support by the uncut material beneath which increases the stiffness of the workpiece.
During ANOVA analysis, all parameters and their interactions were found significant. Appendix 1 presents an interaction plot for thrust force, indicating that SD2 exhibited a higher PTFR compared to SD1 when utilizing the same tool type. This increase is attributed to the higher thrust force at the chisel edge. The literature shows that UAD is particularly effective in mitigating this increased thrust force by chisel edge [48]. The SD2 having smaller step length increases contact area possibly causing increase in cutting temperature which helps further reduce thrust force during UAD, while the SD1 was the least effective. For cutting parameters, spindle speed and feed rate exhibit a linear relationship with thrust force. Low spindle speed effectively reduces thrust force by an average of 37%, whereas, for high spindle speed, the average reduction is below 10%. By doubling the feed rate from 150 to 300 mm/min, the PTFR also doubles, increasing from 15 to 30%. Selecting appropriate cutting parameters and tool type is crucial for maximizing the effectiveness of the UAD system.
3.2 Torque analysis
The analysis of the torque produced by the different tool types under CD and UAD shows that UAD generally results in a reduction of the mean torque across all tools. Specifically, SD2 showed a mean torque of 0.132 N-m (SD = 0.053 N-m) in CD and a slightly increased mean of 0.152 N-m (SD = 0.080 N-m) in UAD. DCD’s produced a mean torque of 0.183 N-m (SD = 0.092 N-m) in CD, which decreases to 0.154 N-m (SD = 0.107 N-m) under UAD. SD1 exhibited a mean torque of 0.177 N-m (SD = 0.066 N-m) in CD, reducing to 0.153 N-m (SD = 0.036 N-m) with UAD. The most notable reduction is observed with TD’s, where the mean torque drops from 0.129 N-m (SD = 0.061 N-m) in CD to 0.079 N-m (SD = 0.023 N-m) in UAD. These findings indicate that while UAD is effective in lowering torque, the degree of effectiveness and variability in torque reduction is highly tool-type dependent, with TD’s showing least produced torque under UAD.
Figure 6 illustrates the impact of cutting parameters and tool type on torque for both CD and UAD. It was observed that spindle speed and feed rate influenced torque in the same manner as in thrust force. As the spindle speed increased, the torque decreased, while an increase in feed rate led to an increase in torque for both CD and UAD. The TD demonstrated exceptional performance in terms of torque efficiency for all drilling parameters compared to other tools. Literature show that increasing the step angle from 70° to 90° increases the thrust force and decreases the torque which agrees with findings in the current study [49]. The torque values for DCD and SD2 become closer to those of TD at higher spindle speeds (3000 and 4500 rpm), particularly at the 150 mm/min feed rate. This might be influenced by the similarity in design and the incremental increase in point angle for DCD, which could affect torque requirements. The effectiveness of UAD in reducing torque seems to diminish at higher spindle speeds, which is consistent with the thrust force observations. This could be due to the decreased influence of ultrasonic vibrations at higher spindle speeds. For the DCD, UAD had a negative impact on torque at low spindle speed (1500 rpm), while at high spindle speed (4500 rpm), the SD1 exhibited an increase in torque. With a smaller step size, the SD2 negatively impacted torque more than the SD1 with a larger step size, indicating that a smaller step size is beneficial for reducing thrust force but detrimental to torque.
Fig. 6
Effect of tool type on UAD effectiveness in reducing torque at different cutting parameters
×
Table 3 presents the percentage contribution of input parameters on torque as obtained from ANOVA analysis. Spindle speed (25.6%) was the most contributing factor affecting torque, while tool type (8.66%) and feed rate (8.42%) were found to be the second most contributing factor. This suggests that controlling the spindle speed, tool type, and feed rate is the most effective method for reducing torque. Appendix 2 presents the interaction plot for the input parameters on torque. The plot reveals that a medium spindle speed of 3000 rpm and a high feed rate of 300 mm/min were the most effective in achieving higher UAD effectiveness in reducing torque. From these findings, it can be concluded that longitudinal vibration was not the most effective factor in reducing torque. Instead, controlling cutting parameters and selecting the optimal tool type are the most effective strategies for minimizing torque in drilling GLARE.
Table 3
ANOVA analysis for torque to check the contribution of cutting parameter and tool type
Thrust force
Torque
Source
% Contribution
Significance
% Contribution
Significance
Tool type
10.61%
Significant
8.66%
Significant
Spindle speed
38.38%
Significant
25.60%
Significant
Feed rate
15.09%
Significant
8.42%
Significant
Vibration
12.64%
Significant
1.42%
Significant
2-Way interactions
Tool type*feed rate
0.63%
Significant
0.50%
Not significant
Tool type*spindle speed
4.13%
Significant
11.02%
Significant
Tool type*vibration
1.94%
Significant
2.14%
Significant
Vibration*spindle speed
6.73%
Significant
0.54%
Not significant
Vibration*feed rate
2.27%
Significant
0.54%
Not significant
3.3 Surface roughness analysis
Figure 7a presents Ra data for CD and UAD on GLARE laminate using various tool types, spindle speeds, and feed rates. Ra values for 2D roughness ranged from 1.05 to 3.76 μm for CD and from 1.54 to 4.32 μm for UAD, as depicted in Fig. 7. Different tool types resulted in varying Ra levels, depending on the cutting parameters. The SD1 produced lower Ra for both CD and UAD. In most cases, it exhibited a reduction in Ra for UAD compared to CD, although the reduction was minimal. The SD2 also displayed a reduction in Ra at high feed rates, while Ra decreased as spindle speed increased. The TD and DCD, with nearly identical designs, exhibited increased Ra when using vibration. Spindle speed had a consistent effect on different tool types for both CD and UAD; as spindle speed increased, Ra decreased due to reduced uncut chip thickness, which in turn diminished the cutting and increased the deformation process. The feed rate did not significantly impact Ra values during CD, but for UAD, Ra increased with increasing feed rate. Overall, the TD performed better during CD, while the SD1 produced the lowest Ra values and demonstrated somewhat better performance during UAD.
Fig. 7
Surface roughness values at different cutting parameters for different tool types (a) Ra, and (b) Rz
×
Figure 7b illustrates the Rz values for various cutting tools under different cutting parameters. The Rz values for CD and UAD ranged from 10.24 to 17.15 μm and 5.63 to 17.19 μm, respectively. A clear trend is observed for CD with respect to tool types. The lowest Rz was achieved using a TD, followed by the DCD, which had a similar design but with a slightly smaller point angle. This observation indicates that reducing the point angle from 140° to 90° leads to increased Rz as observed in the literature [50]. Conversely, increasing the point angle from 140° to 180° also results in higher Rz due to the rise in thrust force, which in turn causes greater deformations around and throughout the hole vicinity. The SD2 outperformed the SD1, despite having an identical point angle. This can be attributed to the vertical cutting action of the SD2, which facilitates the effective cutting of fiber/matrix layers and metal sheets. For UAD, minimal variation was observed among different tool types, yielding nearly identical Rz values. The TD, which produced lower Rz, exhibited higher Rz when vibration was applied. In contrast, the SD1, which initially resulted in higher Rz, demonstrated a reduction when subjected to vibration, as depicted in Appendices 3 and 4.
Considering the influence of spindle speeds, low Rz values were obtained at high spindle speeds due to reduced uncut thickness and elevated temperature, consistent with Ra for both CD and UAD. Conversely, an increase in feed rate led to a rise in uncut chip thickness, resulting in higher surface roughness (Rz, Ra). Table 4 presents the results of an analysis of variance (ANOVA) for surface roughness, encompassing both Ra and Rz. In the case of Ra, the factors “spindle speed” and “tool type*vibration” were found to be statistically significant, contributing 4.65% and 7.06%, respectively, to the observed variance. For Rz, “vibration” and “tool type*vibration” were identified as significant factors, contributing 6.75% and 5.29% to the variance, respectively.
Table 4
Analysis of variance showing parameters with significant percentage contribution on surface roughness
Ra
Rz
Source
% Contribution
Significance
% Contribution
Significance
Tool type
0.63%
Not significant
2.93%
Not significant
Spindle speed
4.65%
Significant
2.96%
Not significant
Feed rate
1.19%
Not significant
2.25%
Not significant
Vibration
2.00%
Not significant
6.75%
Significant
2-Way interactions
Tool type*feed rate
1.18%
Not significant
0.66%
Not significant
Tool type*spindle speed
3.76%
Not significant
2.75%
Not significant
Tool type*vibration
7.06%
Significant
5.29%
Significant
Vibration*spindle speed
0.21%
Not significant
0.49%
Not significant
Vibration*feed rate
0.42%
Not significant
0.73%
Not significant
In summary, the findings indicate that—regardless of the drill type—increasing the spindle speed from 3000 rpm to 4500 rpm did not yield any significant improvement in Ra or Rz. In addition, according to thrust force results shown previously in Fig. 5, drilling at 4500 rpm somewhat diminished the effectiveness of UAD system. Therefore, consideration of the optimal cutting parameters in this case should consider the effect of cutting parameters and UAD on the energy consumption across the whole range of parameters used in the current study.
3.4 Burr formation analysis
Burr is an undesired extra material rollover on the edge of the hole, which compromises the integrity of the machined structure [51]. Therefore, minimizing or eliminating burr formation during the drilling of GLARE is important. Many factors influence the burr formation, which includes the cutting parameters (spindle speed, feed rate), drill geometry, the workpiece properties (number of layers, thickness of metal sheet, ductility, fiber orientation, etc.), and the cutting conditions (coolant, lubricant, etc.) [52]. Excessive deformation at the hole exit, attributed to increased cutting forces, predominantly contributed to burr formation. The DCD exhibited the most severe burr formation, owing to its smaller secondary point angle and extended cutting edge, which generated thin and wide chips prone to deformation as seen in Appendix 8. The increased horizontal thrust force generated by the lower point angle further facilitated material deformation. Almost negligible burr formation was seen for TD, SD2, and SD2 compared to DCD as observed in Appendices 7, 8, 9, and 10. Among different drilling parameters, higher burr formation was seen at low feed rate due to high deformation of smaller uncut chip thickness. The burr formation observed during UAD surpassed that in CD due to the vibrational effect of the tool, intensifying material deformation. TD, SD1, and SD2 resulted in lower burr formation with thin heightened burrs. Due to the presence of thin heightened burrs, only burr height was extensively analyzed.
3.4.1 Entry Burr height analysis
In this dataset, a thorough statistical examination was conducted, focusing on the influence of different tool types and cutting parameters on entry burr heights during CD and UAD. The analysis highlighted a significant spread in CD entry burr heights, ranging from 30 to 815 μm, with an average of 186.87 μm and a substantial standard deviation of 160.74 μm. In comparison, UAD entry burr heights showed a variation from 1.5 to 244.5 μm, with a mean value of 128.70 μm and a standard deviation of 70.01 μm. The spindle speed varied between 1500 and 4500 rpm, and feed rates exhibited a range, yet neither displayed a direct linear correlation with burr heights, pointing to intricate and tool-specific interactions. Furthermore, the percentage reduction between CD and UAD burr heights was notably diverse, extending from − 122% to 813.5%, with an average reduction of 58.17% and a high standard deviation of 201.11%. This diversity in data indicates the complexity and nuanced nature of machining processes, emphasizing that optimal machining strategies are likely to be highly individualized based on tool type and specific operational parameters.
Figure 8 illustrates the entry burr height values for various tools under different cutting parameters. During CD, the tool’s point angle significantly influenced the entry burr height. The smallest point angle (DCD) combined with a low feed rate resulted in a higher burr height due to the workpiece being pushed outwards. As the feed rate increased, the pushing effect diminished and transitioned more into the cutting process, thereby reducing the burr height. The TD with 140°-point angle exhibited a lower burr height compared to the DCD, owing to the reduced horizontal component of the thrust force, which in turn decreased the workpiece material’s deformation towards the outside. In the case of the SD1 and SD2, the horizontal component of the thrust force became zero due to 180°-point angle, leading to comparatively lower burr height as reported in literature [27, 50, 53, 54]. Generally, greater point angles facilitate maximum lip movement, minimizing work hardening and subsequently leading to thinner burrs by altering chip flow direction [55]. The SD2 achieved the minimum burr height while maintaining the same point angle as the SD1, due to its curved outer cutting edge, which smoothed the cutting process at the hole’s periphery. During the UAD process, both the twist and SD2 exhibited an increase in burr height, which could be attributed to the elevated temperature facilitating material deformation. Conversely, a reduction in entry burr height was observed for the SD1 and DCD, indicating that UAD is very effective in reducing the entry burr height for small angles and also effective with larger step sizes, which is essential in reducing the temperature and increase the effectiveness of the UAD.
Fig. 8
Effect of tool type on entry burr height at different cutting parameters
×
The impact of spindle speed on entry burr height exhibited distinct differences when comparing CD and UAD processes. In the case of CD, an increase in spindle speed led to a reduction in entry burr height, which can be attributed to the decrease in cut chip thickness [50, 56]. Conversely, for UAD, the entry burr height increased due to higher cutting temperatures with spindle speed [57]. As seen in Appendix 5, with the increase in feed rate the burr height decreases during CD, whereas not much change was observed for feed rate during UAD. ANOVA shows that the tool type, vibration, and their interaction had a relatively small yet statistically significant impact on entry burr height.
3.4.2 Exit burr height analysis
Most tools produced very thin, heightened burrs, except for the DCD, which led to severe burr formation. The SD2 demonstrates relatively moderate mean exit burr heights for both CD (63.48 μm) and UAD (58.59 μm), with a standard deviation of 26.23 μm and 14.21 μm, respectively, indicating a consistent performance. In contrast, the DCD exhibits the highest variability and mean burr heights among all tools, with a mean of 171.93 μm (CD) and 201.98 μm (UAD), and a notably high standard deviation of 111.68 μm (CD) and 52.77 μm (UAD), suggesting sensitivity to cutting conditions. The SD1 shows the lowest average burr heights for both CD (46.58 μm) and UAD (65.92 μm), coupled with lower standard deviations (12.04 μm for CD and 11.00 μm for UAD), which might indicate more consistent and efficient performance. Lastly, the twist drill records a mean of 69.22 μm for CD and a considerably higher mean of 122.85 μm for UAD, with standard deviations of 22.04 μm and 18.52 μm, respectively, pointing towards a significant difference in burr height creation between CD and UAD. Overall, this analysis underscores the varying efficiency and consistency of different tool types in creating burr heights during the machining of GLARE laminates.
The point angle was identified as a critical factor in tool type concerning exit burr height. A parallel trend was observed between exit burr height and entry burr height concerning the point angle. The burr height increased with the use of vibration due to increased deformation in the feed direction and elevated temperature caused by the vibrating tool. The DCD exhibited a reduction in burr height at low spindle speeds, while the SD2 showed a reduction at high feed rates. The burr height increased with the increase in spindle speed, which contributes to the temperature increase making it more sensitive to spindle speed, as seen in Appendix 6. The feed rate also displayed an appropriate trend during UAD when using TD and DCD. At low spindle speeds, the burr height increased with an increase in feed rate due to higher uncut chip thickness resulting in higher deformation assisted by tool vibration during UAD. Conversely, at high spindle speeds, the burr height decreased due to smaller uncut chip thickness (Fig. 9).
Fig. 9
Effect of tool type on exit burr height at different cutting parameters
×
Table 5 presents the percentage contributions of cutting parameters and tool type on exit burr height. The analysis revealed that tool type had the most significant contribution at 35.86%, indicating that it is a critical factor in controlling burr height compared to other parameters. Although vibration and the interaction of tool type*feed rate were found to be statistically significant, their contributions were relatively small. Based on these findings, we can conclude that UAD and changes in cutting parameters are not as effective in controlling burr height when compared to tool type.
Table 5
Analysis of variance showing parameters with significant contribution on entry and exit burr height
Entry burr height
Exit burr height
Source
% Contribution
Significance
% Contribution
Significance
Tool type
35.86%
Significant
7.02%
Significant
Spindle speed
0.43%
Not significant
2.55%
Significant
Feed rate
0.97%
Not significant
0.44%
Not significant
Vibration
1.96%
Significant
0.80%
Not significant
2-Way interactions
Tool type*feed rate
5.14%
Significant
17.01%
Significant
Tool type*spindle speed
2.56%
Not significant
2.41%
Not significant
Tool type*vibration
1.44%
Not significant
2.65%
Not significant
Vibration*spindle speed
1.16%
Not significant
1.86%
Significant
Vibration*feed rate
0.40%
Not significant
1.61%
Significant
4 Conclusion
The primary objective of this research was to investigate the influence of tool type on the effectiveness of UAD in reducing cutting forces and improving hole quality. Four drill types, namely, the TD, DCD, SD1, and SD2, were utilized. The following conclusions were derived from the study:
1.
The TD performed excellent by the lowest thrust force and torque and the DCD, SD1, and SD2 exhibited an increase in thrust force of 12.81%, 20.69%, and 41.3% and torques of 94%, 92%, and 91%.
2.
During CD, despite a little increase in cutting forces, the TD is highlighted by the high-quality holes with low surface roughness of Ra 1.66 μm and, Rz 10.58 μm and improved control on burr formation lowering entry exit burr height compared to DCD. Conversely, DCD, SD1, and SD2 showed an increase of 23%, 16% and 24% in Ra and 16%, 37% and 29% in Rz.
3.
Notably, SD1 and SD2 showed smaller burr height despite producing a poorer hole quality due to higher cutting forces at low drilling parameters. The DCD resulted in higher burr formation to the extent that deburring is required post drilling.
4.
The efficiency of the UAD system decreases with the increase in cutting parameters. To achieve higher drilling quality, specifically in terms of surface roughness and exit burr at a medium spindle speed of 3000 rpm and a feed rate of 225 mm/min are recommended in the investigated range of cutting parameters.
5.
Limitations of this work: The current work only investigates the effect of UAD process using a single vibration frequency and amplitude. A more detailed study could also vary the aforementioned parameters to assess their impact on hole quality and cutting forces. In addition, the range of cutting parameters (i.e., spindle speed and feed rate) can be increased to assess the effectiveness of the UAD process at high machining rates and the resulting effects on hole quality. Future work should also investigate the effect of different drill geometry, coating, and coolants on cutting forces and hole quality. Finally, a more realistic approach would also investigate the impact of UAD system on the overall energy consumption in addition to hole quality. This will ensure that the hole quality is not maintained at the expense of increased carbon emissions and costs.
Declarations
Ethics approval
Not applicable.
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
