International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)

OPTIMIZATION OF PROCESS PARAMETER LEVELS DURING DRILLING HIGH FIBER VOLUME FRACTION NONLAMINATED GFRP POLYMERIC COMPOSITES
B.Ramesh Department of Mechanical Engineering, St.Josephs College of Engineering, Chennai-119, India A.Elayaperumal Department of Mechanical Engineering, CEG, Anna University, Chennai-25, India

Abstract: In this investigation a nonlaminated Glass Fiber Reinforced Plastic (GFRP) composite manufactured by pultrusion process was drilled with a coated cemented carbide drill. The thrust force and torque during drilling were acquired through piezoelectric dynamometer. Taguchis orthogonal array and analysis of variance (ANOVA) were employed to study the influence of process parameters such as feed and spindle speed on thrust force and torque. The optimum level of process parameters towards minimum thrust force and minimum torque were obtained to achieve defect controlled drilling of GFRP composites. Correlations for thrust force and torque with process parameters were established using a statistical software MINITAB 15. Among the process parameters examined, feed significantly influences both the thrust force and torque with 88.52% and 92.83% respectively whereas the influence of spindle speed on the above was relatively insignificant. Keywords: GFRP composites; drilling; thrust; torque; dynamometer; Taguchi method; S/N analysis

1. INTRODUCTION
Glass fiber reinforced plastic (GFRP) composites are most widely used in aerospace, automobile, marine, electronic industries, construction of military vehicles, machine tools, robots and sports equipment owing to their multi potential properties such as high strength to weight ratio, high specific stiffness, high damping, high fracture toughness, good dimensional stability and excellent corrosion resistance [1,2]. Accordingly, the need for accurate machining of composites has increased enormously. Intricacy in the product design necessitates development of the composite product in parts, which are finally assembled. Hole making thus becomes an integral part of the product development cycle. During drilling of fiber reinforced composites, the work material experiences delamination, fiber breakage, matrix crazing, etc. [3,4]. Among these defects caused by drilling, delamination which occurs both at the entrance and exit planes of the workpiece is most critical, since it can result in lowering of bearing strength and thereby reduces service life of the component [3,5,6]. Hence, utmost care is to be exercised to attain defect controlled drilling performance. The fastening efficiency is largely dependent on the bearing strength which defines the quality of machined holes. Many researchers have proposed that the quality of machined holes is strongly dependent on process parameters such as feed and spindle speed [2,3,7-10] which have great influence on the thrust force and torque. Many attempts have been made by various researchers in drilling laminated GFRP composites. They are briefly presented here. Malhotra [1] and Arul et al. [2] stated that carbide tipped drill performs much better than HSS drill with GFRP. Caprino et al. [4] stated that the type of damage induced in glass/polyester composite materials during drilling is strongly dependent on the feed rate. Di Ilio et al. [7] characterized the interaction between the twist drill and the composite during machining. Paulo Davim et al. [8] employed a statistical technique, using orthogonal arrays and analysis of variance, to investigate the influence of cutting parameters in machining Glass Fiber Reinforced Plastics (GFRPs) manufactured by hand lay-up. Mohan et al. [9] analyzed the influence of machining parameters on thrust force and torque

during drilling glass fiber reinforced polyester composite materials. Knig et al. [11] stated that cutting forces such as thrust force and torque which depend on cutting parameters affects the surface quality. Wen-Chou Chen [12] correlated the delamination factor with the average thrust force for the drilling of uni-directional and multidirectional composite materials. Enemuoh et al. [13] proposed an approach combining Taguchis technique and multi-objective optimization criterion to select global optimum drilling parameters for damage-free drilling in carbon fiber reinforced epoxy composite materials. Tsao et al. [14] predicted and evaluated delamination factor in drilling carbon fiber reinforced plastics (CFRPs) based on Taguchis method and the analysis of variance (ANOVA). Ogawa et al. [15] stated that the mean value (static component) of the thrust force influences on a cutting phenomenon occurring at the chisel edge of the drill and the magnitude of variation (dynamic component) of the thrust force influences on a cutting phenomenon occurring at the major cutting edge of the drill. However, literature on the drilling of nonlaminated composite materials is scarce. This work differs from the previous works by machining a higher order fiber volume fraction (68 %) glass/epoxy nonlaminated composite, whose fiber orientation is parallel to the drill axis, with multicoated (TiN/TiAlN) cemented carbide drill. In ballistic applications mostly composites with higher percentage of fiber volume fraction is used to ensure higher order energy absorption. This paper presents the application of Taguchis orthogonal array and analysis of variance (ANOVA) to study the relative influence of process parameters such as feed and spindle speed on thrust force and torque. The optimum level of process parameters to minimize thrust force and torque at drilling of nonlaminated GFRP composites was obtained by using ANOVA and analysis of Signal-to-Noise ratio. Minimization of thrust force and torque during drilling can lead to minimal damage of the hole. If delamination/damage problem in drilling of composites be solved their applications can be substantially widened. The operator or a dedicated software algorithm can control these process parameters and thereby defect controlled

www.ijsea.com

120

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
drilling can be achieved on an industrial mass production system.

2. EXPERIMENTAL DETAILS
GFRP rod which is the workpiece was made by pultrusion process using E-CR glass directional roving fibers (Figure 1) such that the orientation of glass fibers are parallel to the axis of rod. The GFRP rod has 20 mm length, 25 mm diameter with 68 % fiber volume fraction. The properties of this nonlaminated GFRP composite are presented in Table 1.

(a)

Figure. 1 Photograph of the GFRP rod manufactured by pultrusion process Table 1. Properties of GFRP composite rod Properties Youngs modulus (MPa) Ultimate stress (MPa) Elongation (%) Density (kg/m3) Specific gravity Water absorption (%) Diameter of glass fiber (m) Shore D hardness Value 49881.52 1004.89 36.40 212.52 2.15 0.0001 24.8 74-76 (c) Figure. 2 Composite rod surface: (a) perpendicular to the fibers axes (b) parallel to the fibers axes and (c) close-up view of bonding of glass fiber with epoxy resin (b)

The micrographs of GFRP rod obtained through Scanning Electron Microscope (SEM) are shown in Figure 2. The composition of GFRP rod obtained through Energy Dispersive X-ray Spectrometry (EDX) is found to be Carbon 62.69 %, Oxygen -16.88 %, Magnesium - 0.22 %, Aluminium - 0.73% and Silicon -2.33 %. Cemented carbide twist drill (R840-1000-30-A1A) of grade GC1220 (make Sandvik) was used in this work to produce through holes (Figure 3).

Figure. 3 Photograph of twist drill used in the tests

www.ijsea.com

121

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
The specification of the drill is given in Table 2. CNC machining center (ARIX VMC 100) was used to perform the drilling operations (Figure 4). Table 2. Details of the twist drill Drill standard Shank type Coolant type Drill diameter (mm) No. of flutes Helix angle Point angle Overall length (mm) DIN 6537 Cylindrical Internal 10 2 300 1400 89 Figure 6. Schematic representation of drilling setup. The axial thrust force and torque during drilling were measured using a piezoelectric dynamometer (Kistler make, Model No. 9257B). Due to the orientation of glass fibers parallel to the axis of drill, dynamic component of thrust force is taken into consideration for minimizing the damage of hole [15]. Vacuum cleaner was used to remove powdery chips away from the cutting zone. dynamometer on the table of CNC machining center (Figure 5). The schematic representation of drilling setup is shown in Figure 6.

2.1 Plan of experiments

The experiments were conducted with selected cutting conditions [16] without using coolant. The process parameters and their levels selected for the experiments are presented in Table 3. The orthogonal array L9 was selected for this work as shown in Table 4. Table 3. Process parameters and their levels in drilling Symbol F S Factor Feed (mm/rev) Spindle speed (rpm) Level 1 0.05 750 Level 2 0.10 1000 Level 3 0.15 1250

Table 4. L9(32) orthogonal array with the assigned values Figure. 4 Photograph of experimental setup for drilling GFRP composites Feed Exp. No. F 1 2 3 4 5 6 7 8 9 1 1 1 2 2 2 3 3 3 S 1 2 3 1 2 3 1 2 3 F 0.05 0.05 0.05 0.10 0.10 0.10 0.15 0.15 0.15 S 750 1000 1250 750 1000 1250 750 1000 1250 Speed Factor

Figure. 5 Close-up photograph of the fixture and dynamometer setup Fixture consisting of V-jaws between which the GFRP rod held was specially designed. The GFRP solid round rod was mounted on the fixture which in turn was mounted on a

This investigation is carried out to optimize the process parameter levels within the selected range to attain minimum of thrust force and torque and thereby to attain defect controlled drilling of nonlaminated composites. Each test was repeated thrice and an average was taken for analysis. Analysis of variance (ANOVA) were calculated for the experimental data obtained for dynamic component of thrust force and mean torque to study the relative significance of process parameters. In the Taguchi method, S/N ratio is the measure of quality characteristics and deviation from the desired value. In this investigation, smaller the better characteristic has been applied to determine the S/N ratio for

www.ijsea.com

122

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
dynamic component of thrust force and mean torque as they are to be minimized. A higher the value of S/N ratio means the better the fit for the combined objective. polymeric matrix. General variation of thrust force in Figures 8 and 9 shows that it varies significantly with increase in feed and doesnt vary significantly with increase in speed. With higher feed and speed (0.15 mm/rev and 1250 rpm) lower order thrust force was monitored. Only a marginal variation in thrust force occurs with 0.15 mm/rev feed and 1000 rpm of speed. From Table 5, it is observed that experiment no. 9 and experiment no. 4 gives a minimum and maximum thrust force respectively.
600 550 500
Thrust Force (N)

3. EXPERIMENTAL RESULTS AND DISCUSSIONS

The influence of process parameters (feed and speed) on thrust force and torque is discussed below.

3.1 Thrust force

Typical thrust signals generated during experiment no. 5 are shown in Figure 7. At the beginning of the drilling cycle, the thrust force increases rapidly and found to be average during full engagement of the drill in the workpiece and then reduces as the drill emerges out of the workpiece and approaches to null value. This is due to the stiffness provided by the remaining composite material often becomes insufficient to withstand the feed force.
Speed (rpm) 750 1000 1250

450 400 350 300

0.05

0.10 Feed (mm/rev)

0.15

Figure 9. Influence of spindle speed on dynamic component of thrust force. Table 5. Thrust force and torque values as a function of process parameters Process parameters Exp. Figure 7. Typical thrust force obtained during drilling nonlaminated GFRP (Exp.No.5). Figures 8 and 9 indicates that the thrust force decreases and then increases with increasing speed. Also the Figures 8 and 9 indicates that the thrust force increases and then decreases with increasing feed. These increase in thrust force can be attributed to higher order fiber volume fraction of the workpiece.
600 550 500
Thrust Force (N)

Average of three trials Thrust force (N) 391.03 336.10 385.33 584.92 552.37 581.05 355.54 286.75 274.06 Torque (Nm) 0.44 0.37 0.41 0.83 0.76 0.66 0.83 0.88 0.78

No.

Feed (mm/rev) 0.05 0.05 0.05 0.10 0.10 0.10 0.15 0.15 0.15

Speed (rpm) 750 1000 1250 750 1000 1250 750 1000 1250

1 2 3 4

Feed (mm/rev) 0.05 0.10 0.15

5 6 7

450 400 350 300

8 9
750 1000 Speed (rpm) 1250

3.2 Torque
Typical torque signals generated during experiment no. 5 are shown in Figure 10. The torque increases rapidly during the initial engagement of the drill with the workpiece and thereafter increases gradually until major cutting edges start to contact with the workpiece during drilling process. The torque increases up to the maximum value over the drilling cycle followed by a gradual drop up to a certain value due to frictional torque.

Figure 8. Influence of feed on dynamic component of thrust force. The reduction in thrust force observed in the case of higher speed and higher feed conditions can be attributed to possible thermal softening of the polymeric material. This is also observed by [17]. Because of the low co-efficient of thermal conduction and low transition temperature of GFRP, the accumulated heat around drill edge leads to softening of the

www.ijsea.com

123

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
significantly with increase in speed. With lower feed and medium speed (0.05 mm/rev and 1000 rpm) lower order torque was monitored. Only a marginal variation in torque occurs with 0.05 mm/rev feed and 1250 rpm of speed. From Table 5 it is observed that experiment no. 2 and experiment no. 8 gives a minimum and maximum torque respectively.

3.3 Analysis of variance (ANOVA) and Signal-to-Noise (S/N) ratio

Figure 10. Typical torque obtained during drilling nonlaminated GFRP (Exp.No.5). The frictional torque is due to very fine chips and dust from the workpiece being forced upward through the drill flutes. These become wedged between the drill and composite rod and continue to induce a torque even after the drilling action has finished. The frictional torque which is between the drill margin and the hole depends on the constituents and diameter of the composite rod and on the process parameters. Figure 11 indicates that the torque increases with increasing feed. This may be due to the increasing cross-sectional area of the undeformed chip [18].
0.9 0.8 Feed (mm/rev) 0.05 0.10 0.15

The analysis of variance of the experimental data was done to statistically analyze the relative significance of the process parameters, feed (F) and speed (S), on thrust force and torque. The last column of the ANOVA table indicates the percentage of contribution () of each parameter to the total variation indicating the influence of each parameter. From the analysis (Table 6), it is observed that the factor F, feed, with contribution of 88.52% has both physical and statistical significance on the thrust force. The factor S, speed, with contribution of 3.12% has no either physical or statistical significance on the thrust force. The parameters are not physically significant if their variance ratio is less than the pooled error. Table 6. ANOVA table for thrust force Source F S (e) DF 2 2 4 8 S 46.33 2.66 2.14 51.14 V 23.17 1.33 0.53 6.39 F 43.35 2.49 _ _ S 45.27 1.60 4.28 _ (%) 88.52 3.12 8.36 100

Mean Torque (Nm)

0.7 0.6 0.5 0.4 0.3 750 1000 Speed (rpm) 1250

Total

Figure 11. Influence of feed on mean torque.

0.9 0.8
Mean Torque (Nm)

DF degree of freedom, S sum of squares, V variance, F variance ratio, S pure variation, percentage of contribution, (e) pooled error From the analysis (Table 7), it is observed that the factor F, feed, with contribution of 92.83% has both physical and statistical significance on the torque. The factor S, speed, has no either physical or statistical significance on the torque. Table 7. ANOVA table for torque

0.7 0.6

Speed (rpm) 750 1000 1250

Source
0.5

DF 2 2 6 8

S 67.13 1.65 3.82 70.95

V 33.57 0.82 0.64 8.87

F 52.79 _ _ _

S 65.86 _ 5.09 _

(%) 92.83 _ 7.17 100

F
0.4 0.3 0.05 0.10 Feed (mm/rev) 0.15

S (e) Total

Figure 12. Influence of spindle speed on mean torque. General variation of torque in Figure 12 shows torque decreases with increase in speed and increases with increase in feed. This may be attributed to higher order fiber volume fraction of the workpiece. The increase in torque observed in the case of high speed and high feed conditions also may be attributed to higher order fiber volume fraction of the workpiece. Also it shows that the rate of change of torque varies significantly with increase in feed and doesnt vary

DF degree of freedom, S sum of squares, V variance, F variance ratio, S pure variation, percentage of contribution, (e) pooled error From the analysis (Tables 6 and 7) it is inferred that the factor feed, has highest degree of influence on torque and then that on thrust force. The factor speed has least degree of influence on thrust force and no influence on torque. In Glass/epoxy

www.ijsea.com

124

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
high fiber volume fraction nonlaminated composites, speed is not so significant as feed for thrust force and torque and this is also observed in Glass/polyester high fiber volume fraction laminates [8,9]. The S/N ratio response graphs for thrust force and torque are shown in Figures 13 and 14 respectively.

4. CONCLUSIONS
The research in the present study was to optimize the process parameter levels within the range examined based on minimum of thrust force and torque and thereby attaining defect controlled drilling of nonlaminated GFRP composites using coated solid carbide twist drill. Based on the experimental results presented, the following conclusions were drawn for drilling high fiber volume fraction E-CR GFRP solid round rods manufactured by pultrusion process: In glass/epoxy high fiber volume fraction nonlaminated composites, feed significantly influences thrust force and torque with 88.52 % and 92.83 % respectively, whereas the influence of speed on thrust force and torque was relatively insignificant. This is similar with glass/polyester high fiber volume fraction laminates.

Figure 13. Response graph of thrust force for process parameters in drilling nonlaminated GFRP.

The optimal process parameter levels for drilling nonlaminated GFRP composites to produce lower order thrust force are 0.15 mm/rev feed and 1000 rpm speed and that to produce lower order torque are 0.05 mm/rev feed and 1250 rpm speed within the range examined. The speed is physically and statistically insignificant in influencing the thrust force and torque. Therefore, the optimal process parameter levels within the range examined to produce high quality holes in drilling high fiber volume fraction nonlaminated GFRP composites can be concluded as that to produce lower order torque (0.05 mm/rev feed and 1250 rpm speed), as these levels produce only a marginal variation in producing lower order thrust force.

Figure 14. Response graph of torque for process parameters in drilling nonlaminated GFRP. Hence, from Figure 13 the optimal combination to obtain lower thrust force was identified as 0.15 mm/rev feed and 1000 rpm speed and from Figure 14 the optimal combination to obtain lower torque was identified as 0.05 mm/rev feed and 1250 rpm speed within the range examined.

5. ACKNOWLEDGEMENTS
The authors are grateful to M/s. Sunrise Fibre Glass Industries, Bangalore, India for supplying the composite materials for experimentation.

3.5 Correlation of thrust force and torque with process parameters

In use of non-linear regression analysis, the correlations for thrust force and torque with process parameters in drilling nonlaminated GFRP composites were obtained using statistical software MINITAB 15. The regression equations are : Thrust force in N = 326 + 18118F - 1.24S 93858F2 + 0.000591S2 , R2 = 0.982 Torque in Nm = - 0.212 + 14.8F + 0.00021S - 52.7F2 0.00000019S2, R2 = 0.965

6. REFERENCES
[1] Malhotra S.K. Some studies on drilling of fibrous composites. J. Mater. process. Technol. 24 (1990), 291300. Arul S, Vijayaraghavan L, Malhotra S.K., Krishnamurthy R. Influence of tool material on dynamics of drilling of GFRP composites. Int. J. Adv. Manuf Technol. 29 (2006), 655662. Tagliaferri V, Caprino G, Diterlizzi A. Effect of drilling parameters on the finish and mechanical properties of GFRP composites. Int. J. Mach. Tools. Manuf. 30 (1) (1990), 7784. Caprino G, Tagliaferri V. Damage development in drilling glass fibre reinforced plastics. Int. J. Mach. Tools. Manuf. 35 (6) (1995), 817829. Bhattacharyya D, Horrigan D.P.W. A study of hole drilling in Kevlar composites. Composites Science and Technology. 58 (1998), 267283. Khashaba U.A. Delamination in drilling GFRthermoset composites. Composite Structures. 63 (2004), 313327.

[2]

(1)

[3]

(2) [4]

where F is feed in mm/rev and S is spindle speed in rpm . In multiple regression analysis, R2, which is called R-sq, is the regression co-efficient, where R2 > 0.85 for the models indicates that the fit of the experimental data is satisfactory. Therefore, the models for thrust force and torque deems to be fit.

[5]

[6]

www.ijsea.com

125

International Journal of Science and Engineering Applications (IJSEA) Volume 1 Issue 2, 2012, ISSN - 2319-7560 (online)
[13] [7] Di Ilio A, Tagliaferri V, Veniali F. Cutting mechanisms in drilling of aramid Composites. Int. J. Mach. Tools. Manuf. 31 (2) (1991), 155165. Paulo Davim J, Pedro Reis, Conceio Antonio C. Experimental study of drilling glass fiber reinforced plastics (GFRP) manufactured by hand lay-up. Composites Science and Technology. 64 (2004), 289297. Mohan N.S., Ramachandra A, Kulkarni S.M. Influence of process parameters on cutting force and torque during drilling of glass-fiber polyester reinforced composites. Composite Structures. 71 (2005), 407413. Dharan C.K.H., Won M.S. Machining parameters for an intelligent machining system for composite laminates. Int. J. Mach. Tools. Manuf. 40 (2000), 415426. Knig W, Grass P. Quality definition and assessment in drilling of fibre reinforced thermosets. Annal CIRP. 38 (1989), 119124. Wen-Chou Chen. Some experimental investigations in the drilling of carbon fiber- reinforced plastic (CFRP) composite laminates. Int. J. Mach. Tools. Manuf. 37 (8) (1997), 10971108. [14] Enemuoh Ugo E., Sherif El-Gizawy A., Chukwujekwu Okafor A. An approach for development of damage-free drilling of carbon fiber reinforced thermosets. Int. J. Mach. Tools. Manuf. 41 (2001), 17951814. Tsao C.C., Hocheng H. Taguchi analysis of delamination associated with various drill bits in drilling of composite material. Int. J. Mach. Tools. Manuf. 44 (2004), 10851090. Ogawa K., Aoyama E., Inoue H., Hirogaki T., Nobe H., Kitahara Y., Katayama T., Gunjima M. Investigation on cutting mechanism in small diameter drilling for GFRP (thrust force and surface roughness at drilled hole wall). Composite Structures. 38 (1-4) (1997), 343350. http://www.sandvik.coromant.com/in Singh I., Bhatnagar N. Drilling of uni-directional glass fiber reinforced plastic (UD-GFRP) composite laminates. Int. J. Adv. Manuf Technol. 27 (2006), 870876. Velayudham A., Krishnamurthy R., Soundarapandian T. Evaluation of drilling characteristics of high volume fraction fibre glass reinforced polymeric composite. Int. J. Mach. Tools. Manuf. 45 (2005), 399406.

[8]

[9]

[15]

[10]

[16] [17]

[11]

[18]

[12]

www.ijsea.com

126