Analysis of the Influence of Drill Tip Geometry on the Dry Drilling Process in CFRP Thermoset Laminate

Abstract

Carbon fibre reinforced composite (CFRP) laminates are widely used in high-tech industries. However, their assembly often requires a drilling process that can create defects. Therefore, studies on the drill tip angle have sought to minimize the surface area affected by these defects and improve the internal hole quality. In this work, drilling was carried out under dry conditions at a constant cutting speed for four different feed rates in the epoxy–carbon-based thermosetting laminate (EPX-C). Two carbide drills with point angles of 118° and 140° were used. The results showed the occurrence of chipping-type delaminations on both the hole entry and exit surfaces, with the latter being more severely affected. The delamination factor values obtained indicated that the 118° drill performed better than the 140° drill. The results were also compared with those obtained in a previous study using drills with angles of 60° and 130°. Although the values were higher, they followed the same trend of reduction with increasing feed. In terms of surface finish, the average roughness (Ra) values obtained with the 140° drill were better at the lowest feed rate.

1. Introduction

Carbon fibre reinforced composite laminates (CFRPs) are widely used in high-tech sectors such as aerospace and automotive industries due to their excellent mechanical properties and lightweight nature. However, assembling these materials often involves drilling processes, which can introduce defects, most notably delamination. Studies investigating the effects of drilling have highlighted the relationship between the laminate’s composition and arrangement, the machining parameters, and the cooling conditions [1].

Despite advances in research and the publication of academic reviews over recent decades, achieving damage-free holes with tight tolerances remains challenging, primarily due to the need for rigorous quality control [2]. Although a vast amount of experimental data is available, general guidelines for drilling CFRP laminates still have limitations, especially in preventing defects such as fibre pull-out, matrix cracks, burrs, and delamination at the hole entry and exit [3,4]. Preventing delamination has become a central focus of manufacturing research because of its significant impact on the structural integrity and final performance of parts [5].

One of the main challenges in manufacturing CFRP components is delamination, which is influenced by factors related to both the material’s characteristics and the drilling process parameters [6]. The most critical factors include the type of matrix and fibre orientation in the composite, as well as the drill geometry and feed rate.

Previous studies have shown that delamination is related to the feed rate [7,8,9,10]; however, the severity of damage is more strongly associated with the drill’s geometry [11]. Using twist drills with smaller point angles and lower feed rates reduced feed forces and, consequently, smaller maximum delamination diameters [12,13].

Drilling analysis in composite materials involves experimental procedures that consider variations in drill geometry to minimize defects resulting from the process, which can cause damage to the assembly and/or structure. Thus, reducing the helix angle leads to a decrease in the drill exit angle, as well as in the shear angle (ϕ), which can intensify the occurrence of delamination at the hole exit [14]. In CFRP laminates, it was found that higher helix angles perform better regarding the influence on delamination at the exit of the hole [15].

Tool geometry is a fundamental factor that influences process performance and the final quality of the composite, especially concerning delamination at both the entry and exit of the hole [16]. It is important to highlight tool wear, which compromises the machinability of composite materials and increases the likelihood of defects. Tools specifically designed for machining composites tend to have greater wear resistance, leading to better quality in the drilling process [17].

However, standard twist drills, which are widely available from tool suppliers for practical reasons, are often used despite wearing out quickly [18]. Twist drills are employed in various industrial processes due to their cost-effectiveness and the ability to re-sharpen the cutting edges after wear, since the cutting edge is the main factor influencing feed force and hole quality in CFRP drilling [19]. It has been observed that the tip angle of the twist drill has a clear impact on the feed force: increasing the tip angle results in higher feed force [20].

Another important aspect is drilling coating, which ensures less wear depending on the number of holes produced, compared to uncoated drills. However, experimental studies on tool wear in CFRP drilling have concluded that wear mechanisms and failure modes vary according to the machining process and tool geometry [21]. In this context, the use of conventional uncoated drills to replace components is an option to be evaluated, since its current cost is approximately five times lower than that of a specific drill for drilling in composite materials.

The performance of the tool tip geometry is evaluated based on the internal surface roughness of the hole and the extent of delamination, which is quantified using delamination factors reported in the literature. Most of these factors are two-dimensional, based on area measurements. The conventional one-dimensional delamination factor (F_d) simplifies the evaluation of delamination in composite laminates [22], and is calculated as shown in Equation (1):

$${\mathrm{F}}_{\mathrm{d}}={\displaystyle \frac{{\mathrm{D}}_{\mathrm{m}\mathrm{a}\mathrm{x}}}{{\mathrm{D}}_0}},$$

(1)

where D_max is the maximum diameter of delamination or cracks, and D₀ is the nominal hole diameter.

Since there is no standard method to measure the delamination factor, different researchers have used different methods to quantify delamination, and the adjusted delamination factor (F_da) is the most accepted; however, it does not consider the impact of the feed force on delamination [23]. The adjusted delamination factor (F_da) aims to correct the conventional delamination factor (F_d), since it does not quantify the delaminated area (A_d), since the delamination has an irregular shape, containing damaged areas and cracks on the hole entry and exit surface [24]. Therefore, F_da considers the contribution of the crack (F_d) in the first term and the damaged area (A_d) in the second term of Equation (2):

$${\mathrm{F}}_{\mathrm{d}\mathrm{a}}={\mathrm{F}}_{\mathrm{d}}+{\displaystyle \frac{{\mathrm{A}}_{\mathrm{d}}}{{\mathrm{A}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-{\mathrm{A}}_0}}·\left({\mathrm{F}}_{\mathrm{d}}^2-{\mathrm{F}}_{\mathrm{d}}\right),$$

(2)

where A_max is the area associated with the maximum delamination diameter D_max, and A₀ is the area of the nominal hole D₀.

The present study aimed to evaluate the influence of 118° and 140° tip angles used in metal drilling on the generation of defects such as delamination, torn fibres, burrs on the hole entry and exit surfaces, as well as on the quality of the hole’s internal surface. The results of the F_da factor and average roughness (Ra) were compared with the results obtained with a specific drill for drilling in composite laminates, with 60°/130° tip angles, in a previous study [25].

2. Experimental Procedures

2.1. Material

The laminate used in the drilling test was the EPX-C aeronautical-grade continuous carbon fibre reinforced composite (Hexply 8552^® reinforced with AGP193-PW [26] supplied by Hexcel^®, Seguin, TX, USA), consisting of thermosetting epoxy resin reinforced with continuous carbon fibres, with 3000 filaments per bundle. The ideal fibre volume fraction in the composite is between 50% and 65%. The laminate is made by stacking 24 layers of bidirectional 0/90° plain wave fabric, resulting in a nominal thickness of 5 mm. The laminate was consolidated in an autoclave using a vacuum bag at a temperature close to 180 °C. The final laminate had a final thickness of 5 mm with a stacking sequence of [(0/90), (+45/−45)₂, (0/90)]₆. The specimens were extracted from the base laminate according to the orientation of the weft (0°) and warp (90°), with 14 mm × 140 mm dimensions.

2.2. Drilling Process and Tool

The test was performed under dry conditions on the CNC machine tool Emco Concept Mill 55 (EMCO, Shenzhen, China), with a tool turret with 8 positions, rotation from 150 to 3500 rpm, feed speed from 0 to 2000 mm/min, with simulated programming and executed by CAM (computer assisting manufacturing) in the Advanced Machining Laboratory from Technological College of Sorocaba. The cutting speed (V_c) was set at 60 m/min, with four different feeds (f) per revolution: 0.045, 0.060, 0.075, and 0.090 mm/revolution.

Two ∅ 6 mm carbide drills without an internal cooling channel were used, one with a tip angle of 118° (Figure 1a) according to DIN 6537 [27] and the other with 140° (Figure 1b) according to DIN 8037 [28]. A drill result from a previous study was used as a reference for this study [29] has tip geometry with two angles of 60°/130° (Figure 1c), with diamond coating code A 1163-6 (by Seco Tools), specific for drilling in composites, used at the same cutting speed, only at feed rates of 0.045 mm/rev and 0.090 mm/rev. For each feed condition, six holes were drilled in the same test specimen, three holes with the 118° drill bit and three holes with the 140° drill bit.

The images of hole entry and exit surfaces were obtained by microscopic analysis. This analysis was performed using an Olympus brand optical microscope, model CX31, equipped with an Evolution LC Colour camera and Image-Pro Plus 6.2 capture software. Then, these image surfaces were analyzed two-dimensionally in orthogonal projection, to obtain the values of the variables used in the evaluation of the delamination factors. For this, the public domain ImageJ software 1.54p, developed by the National Institutes of Health and the CAD Autodesk Inventor® educational version 2024 software were used. The images from optical microscopy were saved in the *.ipt extension of the native Inventor software file for tracing and obtaining the values of the dimensions that were used in the evaluation factors of the delamination defect.

The average roughness (Ra) values were obtained by the Mitutoyo SJ-210 (Jundiaí, Brazil) roughness metre with a cut-off value of 0.08 mm; the surface roughness values of the hole surface were obtained after the drilling process. The Analysis of Variance (ANOVA) was used to compare the average values of F_d, F_da and Ra, with a significance level of 5% (CI of 5%), which is a common practice in scientific research, allowing to state with 95% confidence that the ranges of these values are significantly different.

3. Results

3.1. Delamination Factor

To determine the delamination factor for each feed condition, six holes were drilled in the same test specimen, three holes with the 118° drill bit and three holes with the 140° drill bit, with the shape shown in Figure 2.

Figure 3 presents the values obtained for the conventional (F_d) and adjusted (F_da) delamination factors on the hole entry surface. As shown in Figure 3a, the F_d and F_da values resulting from drilling with the 118° angle drill are similar and do not exhibit a significant increase with higher feed rates. Figure 3b displays the F_d and F_da values obtained from drilling with the 140° angle drill. It can be observed that the F_da values remain close to those of F_d, even as the feed rate increases, demonstrating a behaviour similar to that observed with the 118° drill. When comparing the results obtained with both drills, it is noted that only at a feed rate of 0.075 mm/rev are the F_d and F_da values of the 140° drill slightly lower.

Therefore, the results indicate that the 22° variation in the drill tip angle did not significantly influence the increase in delamination called peel-up, which can be caused when the fibre tips are pulled up and twisted due to the tool’s helix [30]. In addition, the feed rate increase range used also did not influence the increase in delamination. These results agree with a previous study on delamination defects, which indicate that it is possible to avoid delamination at entry by using a low to moderate feed rate [31].

Figure 4 presents the values obtained for the F_d and F_da factors at the exit surface of the hole. In Figure 4a, it is observed that the values obtained with the drill with a 118° angle are close at a feed rate of 0.045 mm/rev. As the feed rate increases to 0.060 mm/rev, there is a slight reduction in the values, followed by a subsequent increase, reaching maximum values at a feed rate of 0.075 mm/rev. However, at a feed rate of 0.090 mm/rev, a sharp decrease occurs, indicating that this parameter represents the optimal feed rate for minimizing the delamination defect compared to the other feed rates.

Figure 4b shows the values obtained with the drill with a 140° angle. It is observed that the F_da value is higher than that of F_d at a feed rate of 0.045 mm/rev, with an increasing trend at feed rates of 0.060 mm/rev and 0.075 mm/rev, where they reach their maximum values. However, at 0.090 mm/rev, a significant reduction occurs, resulting in the lowest values relative to the other feed rates. This discrepancy between the F_da and F_d values can be attributed to the increase in the damaged surface area (Ad), whose contribution predominates in the second part of Equation (2) for F_da. In other words, the two-dimensional factor prevails over the one-dimensional factor, regardless of whether A_d is concentrated or dispersed around the hole.

These results corroborate the findings in the specialized literature, which indicates that delamination at the hole exit is one of the main challenges in machining CFRP laminates, being considered a highly critical phenomenon [32,33]. In the comparative analysis between the values of the factors obtained with the two drills, it is verified that both reached their maximum values at a feed rate of 0.075 mm/rev and minimum values at a feed rate of 0.090 mm/rev, and that the Fda values obtained with the 118° angle drill are lower than those obtained with the 140° angle drill, indicating better performance in reducing the delamination defect.

Figure 5 shows the comparison of the F_da values obtained with the 118° angle drill and with the specific drill for CFRP laminate, with angles of 60°/130°. It can be observed that the F_da value obtained with the 118° drill is significantly higher than that obtained with the 60°/130° drills, especially at the feed rate of 0.045 mm/rev. At the feed rate of 0.090 mm/rev, the value obtained with the 118° drill remains higher, although closer to the values of the 60°/130° drills. Even so, the values followed the same downward trend as the feed rate increased. These results confirm that the feed of 0.090 mm/rev, with a cutting speed of 60 m/min, is within the parameter recommended by the manufacturer for drills with angles of 60° and 130° and can be applied to conventional drills. Furthermore, they indicate that, for small-scale holes for component replacement, the 118° drill may be a viable option to be considered due to the cost, since it is five times smaller compared to the drill with angles of 60° and 130°.

3.2. Defects on the Hole Exit Surface

The hole exit surface images analyzed using optical microscopy were converted into black and white images using the CAD Autodesk Inventor® educational version 2024 software as shown in Figure 6. The hole location and the neighbourhood where the damage could be seen were traced and the dimensions values were obtained to evaluate the delamination defect factors.

Figure 7 shows the main defects found at the hole exit, such as the volume of uncut fibres, delaminated area (A_d), and burrs, especially at the feed rate of 0.090 mm/rev, which presented the best results for the F_d and F_da factors. In Figure 7a, a greater concentration of uncut fibres is observed at 0° and 90° orientations when using the 118° drill, compared to the 140° drill, shown in Figure 7b. Fibre orientation is a fundamental factor, as it directly influences the integrity of the machined surface. The 90° orientation is considered a critical angle, which can cause subsurface damage [34]. These results reinforce previous studies, which indicate that the uncut fibre defect is related to the relationship between fibre orientation and cutting angle [35]. In Figure 7c, the presence of uncut fibres in thin volumes can be seen, using a 60°/130° drill bit. These uncut fibres can cause problems such as fibre detachment from the matrix, although they are less of a concern when they present small fraying [36]. Burrs were also observed at the exits of the holes made by the 118° and 140° drills, in Figure 6a,b, respectively. Both the uncut fibres and the burrs on the entry and exit surfaces can be removed by reaming and deburring processes, although these procedures increase the final cost of the process.

As for the delaminated area (A_d), it appears dispersed around the exit of the holes made by the three drills evaluated. The largest extension was observed in Figure 7b, with the 140° drill, followed by the 118° drill (Figure 7a), and the smallest areas in Figure 7c, with the 60°/130° drills. Delamination can also occur within a layer, without interlaminar separation, but with spalling (splintering) in the layer around the hole, which is considered a type of delamination [37].

3.3. Average Roughness

Figure 8 presents the average surface roughness (Ra) values as a function of drill point angles and increasing feed rates. It is observed that the drill with a 140°-point angle yielded lower Ra values compared to those obtained with the 118°, 130°, and 60° drills at a feed rate of 0.045 mm/rev, achieving its minimum roughness at 0.060 mm/rev. This condition is therefore considered to provide the best surface finish among the evaluated feed rates. At a feed rate of 0.075 mm/rev, the Ra values for the 118° and 140° drills were comparable. At 0.090 mm/rev, the 140° drill once again demonstrated superior performance compared to the 118° drill and produced values similar to those obtained with the 60°/130° drills.

These results should be interpreted with caution due to potential measurement uncertainties. In the three distinct regions where roughness was measured, the stylus may have become entangled with protruding fibres on the internal surface of the hole or passed through areas where the matrix or fibres had been dislodged, thereby deepening surface valleys. Protruding fibres frequently interfered with the stylus tip, possibly introducing additional measurement errors and leading to increased Ra values [38].

Furthermore, the choice between thermoplastic and thermoset matrix materials plays a significant role in the type of damage that may occur during drilling. Brittle matrices are more susceptible to microfracture, which can contribute to increased surface roughness [37]. Supporting these findings, previous studies involving thermoset matrices under dry drilling conditions have shown that higher feed rates are associated with increased roughness [38,39].

4. Discussion

The study of the influence of drill tip geometry on the dry drilling process in carbon fibre reinforced polymer (CFRP) thermoset laminates reveals significant insights into the challenges of machining, especially drilling, in minimizing process-related defects. Delamination is a significant concern, predominantly governed by the intrinsic properties of the material (number of reinforcement layers and type of matrix), as well as by specific process parameters, and the component and/or structure may be rejected by quality control. This investigation shows that drill tip geometry, especially the tip angle, helps reduce delamination, depending on the specified feed (f) and cutting speed (Vc), and that conventional drills can be used for repairs, which is the objective of the present work. This was evidenced by the number of holes obtained for each specified feed, in which no wear was observed on the cutting edges of the drills used.

Two distinct types of delamination factors were examined: the standard one-dimensional delamination factor (F_d) and the adjusted two-dimensional delamination factor (F_da). The results indicated that the Fda values differed from the F_d values in drilling with the 140° drill angle compared to the 118° drill angle due to the larger damaged area (A_d) on the hole exit surface, confirming why, of all the models proposed in the literature, F_da is the preferred model. In addition, it was possible to quantify the delamination defect using the conventional (F_d) and adjusted (F_da) factors, in addition to characterizing the delamination by chipping. It was observed that the delamination at the hole exit was more severe than at the hole entry, due to the lower resistance of the laminate during the final drilling stage, in all four feed rates evaluated. The 118° drill demonstrated superior performance to the 140° drill at feed rates of 0.060 mm/rev and 0.090 mm/rev, demonstrating the influence of the tool angle on the generation of delamination-type damage. In addition, this tool achieved F_da values close to those of the 60°/130° drill at feed rates of 0.090 mm/rev.

The defect analysis revealed a higher incidence of burrs and uncut fibres at the hole exit compared to the hole entry, especially when using the 118° and 140° drills at all feed rates considered. This indicates that a feed rate of 0.090 mm/rev, combined with a constant cutting speed of 60 m/min and the use of the 118° drill, constitutes the most favourable processing condition to minimize the formation of burrs and uncut fibres.

The comparison of the F_da values obtained with the 118° and 140° drills versus the specific 60°/130° composite drill showed that, although the conventional tools present higher F_da values, both show a tendency to improve with increasing feed. In addition, a more in-depth study should be carried out involving a greater number of holes and analysis of the wear of the cutting edges, with the same process parameters to verify its applicability on an industrial scale regarding the cost/benefit of using the conventional drills evaluated.

The study revealed that the average roughness values varied significantly depending on the geometry of the drill tip and the feed rates used during the drilling process. Specifically, the 140° drill bit provided better surface finish results at lower feed rates compared to the 118° drill bit, indicating that the choice of drill bit geometry plays a vital role in determining the surface quality of the drilled holes. On the other hand, the results of this study showed that increasing the drill bit angle increases delamination but achieves better surface finish, evidencing that achieving the desired quality in the holes is still a difficult task. Furthermore, the roughness measurements were influenced by the presence of defects such as delamination and protruding fibres in the CFRP laminate. These defects can interfere with the measurement process, leading to increased Ra values due to the tip becoming entangled with fibres or passing through areas where the matrix has been debonded.

5. Conclusions

By applying the conventional (F_d) and adjusted (F_da) delamination factors and analyzing the surface finish, it was possible to investigate the influence of the drill tip geometry on the generation of defects resulting from the dry drilling process in CFRP thermoset laminate, and the following results were obtained:

-

The delamination defect was characterized as a spalling defect, being more severe at the hole exit than at the hole entrance, due to the lower resistance of the laminate in the final drilling stage, at the four feed rates evaluated for the 118° and 140° drills.

-

The drill with an angle of 118° showed superior performance compared to the 140º drill at feed rates of 0.060 mm/rev and 0.090 mm/rev, evidencing the influence of the tool angle on the generation of delamination-type damage. Furthermore, this tool obtained F_da values close to those of the 60° and 130° drill at feed rates of 0.090 mm/rev. This suggests that the use of conventional drills may be viable in the machining of CFRP laminates, enabling a reduction in operating costs without significantly compromising the quality of the finish, especially with regard to delamination.

-

The defect analysis revealed a higher incidence of burrs and uncut fibres at the hole exit compared to the entry, especially when drilling with the 118° and 140° drills, at all feed rates considered. This result indicates that the feed rate of 0.090 mm/rev, associated with a constant cutting speed of 60 m/min and the use of the 118° drill, constitutes the most favourable processing condition for minimizing the generation of defects. The comparison of the F_da values obtained with the 118º drills in relation to the specific drill for composites of 60° and 130°, revealed that, although the conventional tools present higher F_da values, both follow a trend of improvement with increasing feed.

-

The evaluation of the surface finish, through the average roughness (Ra), indicated that the drill with an angle of 140° provided better results compared to the one with 118° and, when compared at feed rates of 0.045 mm/rev and 0.090 mm/rev with the results of the drill with an angle of 60° and 130°, they were close. These results can be credited to the relationship between the angle of the drill tip and the nature of the thermosetting matrix due to its high melting point and the thickness of the laminate in drilling under dry conditions.