INTRODUCTION
The use of composite material is being increased in recent years, because of
their special mechanical, thermal and structural properties. Most of the researchers
have used Glass Fiber Reinforced Polymers for their studies. GFRP material is
being inexpensive material compared to other composite material and having wider
industrial applications. The drilling operation is an important machining process
required for fastening the components in an assembly. To establish the better
quality of hole made by drilling operation, the hole must be free from any damages
like delamination etc. It is quiet difficult to generate hole without any damage,
however changing the influencing machining parameters can minimize the same.
The hole surface quality mainly depends on cutting parameters, like tool material,
tool geometry, cutting forces etc. (Hocheng and Puw, 1992;
Chen, 1997; Lin and Chen, 1996;
Piquet et al., 2000). The influence of the cutting
parameters in damage around the drilled hole with different tool geometry has
been studied on discs made of GFRP (Polyester matrix reinforced with 65% of
glass fiber) produced by hand layup (Davim et al.,
2004). Enemuoh et al. (2004) have studied
the application of Taguchi and a multiobjective optimization criterion and
reported that it is possible to achieve cutting parameters that allow the absence
of damage in drilling of fiber reinforced plastics. Using High Speed Steel (HSS)
drill, a series of vibratory drilling experiments have been conducted to assess
the delamination factor and it was found that vibratory drilling is a promising
machining technique (Arul et al., 2006).
Delamination is being the most critical defect, among the defects caused by
drilling operations. Delamination lowers the bearing strength thereby reducing
the structural integrity of the material (Won and Dharan,
2002; Caprino and Tagliaferri, 1995). Davim
and Reis (2003) have studied the drilling of laminated composite materials
by conventional tools and concluded that the quality of cut surfaces depend
on the cutting parameters, tool geometry and tool material. A comparative study
aiming to evaluate the influence of the drill geometry on unidirectional laminate
glass reinforced plastics (Bhatnagar et al., 2004;
Singh and Bhatnagar, 2006). Khashaba
(2004) has investigated the machining of GFRP composites produced by different
matrix materials and different reinforcing shapes and concluded that the composite
with cross winding had a surface without delamination and the woven composite
with different matrix materials had a negligible effect on thrust. Hocheng
and Tsao (2006) have studied the delamination using specially designed drills
and established a relationship between feed rate, cutting speed and drill diameter.
Kurt et al. (2009) have applied Taguchi methods
and optimized the cutting parameters for surface finish and hole diameter accuracy
in dry drilling operation.
The delamination of FRP composites under machining plays a vital role in determining the machining quality. In this research, an attempt is made to study the internal delamination that cause poor surface quality of hole during drilling of CSMat GFRP apart from entry and exit delamination. The work is carried out based on a statistical based approach namely, Design of Experiments (DOE). The standard HSS twist drills are used in this experimental work to investigate the internal delamination of CSMat GFRP material under various influencing parameters such as drill diameter, spindle speed and feed rate.
MATERIALS AND METHOD
Material and machine used: The specimen material made from the Chopped Strand Mat Glass Fiber Reinforced Plastic of size of a slab of 300x300x23 mm was used for the experiments. The physical properties of CSMat GFRP materials are highlighted in the Table 1. The drilling experiments were conducted in dry condition and at room temperature on MCV400 with a workspace of 600x415x460 mm and the machine has a speed range of 606000 rpm.
Experimental design: The design of experiments is an effective tool to optimize the various machining parameters. In this study, a three level L_{9} orthogonal main effect design was used. This design has an advantage of reducing the number of experiments. The identified parameters were drill diameter, spindle speed and feed rate. The factors and the levels of factors used are listed in the Table 2.
Determining delamination value: The damage around the holes was measured with 3D Coordinate Measuring Machine (SPECTRA Model) by the procedure: the diameter of the hole in the damage zone is measured for four times and the maximum value was recorded as D_{max} value. The delamination factor was determined by the ratio of maximum diameter (D_{max}) of the damage zone to the hole diameter (D). Therefore, the equation used to determine the delamination factor is:
where, F_{d} is the delamination factor, D_{max} is the maximum diameter of the damage zone in mm and D is the diameter of the hole in mm.
Table 1: 
Properties of CSMat GFRP material 

Table 2: 
Levels of factors 

Table 3: 
Experimental results for delamination factor and S/N ratio 

The experiments were replicated for two times and the average values of the delamination factor obtained were tabulated in Table 3. The results were analyzed using S/N ratio and ANOVA analysis.
Analysis of drilling parameters
Analysis of S/N ratio: S/N ratio is applied to measure the quality characteristic
deviating from the desired value. The term signal represents the desirable value
of the response variable whereas the term noise represents the undesirable value
of the response variable. The S/N ratio η is defined as:
where, MSD is the mean square deviation for the response characteristics.
To obtain optimal drilling performance, thelowerthebetter quality characteristic
for delamination was taken. The MSD for thelowerthebetter quality characteristic
can be given as:
MSD = (1/n ) ΣF_{di}^{2 } ; i
= 1,…,n 
where, n is the number of repeated experimental run and F_{di }is the value of delamination factor for the ith test.
The experimental results for delamination factor and the corresponding S/N ratio are shown in the following Table 3.
Table 4: 
S/N response table for delamination factor 


Fig. 1: 
S/N response (Drill dia.) for delamination factor 

Fig. 2: 
S/N response (Spindle speed) for delamination factor 

Fig. 3: 
S/N response (Feed rate) for delamination factor 
The S/N response table and S/N response chart for delamination factor are given in Table 4 and in Fig. 1 to Fig. 3.
It is observed that higher S/N ratio for drill diameter is at the first level (Fig. 1), higher S/N ratio for spindle speed is at the third level (Fig. 2) and higher S/N ratio for feed rate is at the first level (Fig. 1).
Table 5: 
Results of ANOVA for delamination factor 

Based on the analysis of S/N ratio and from S/N Response Table, the optimal drilling performance which will minimize the delamination factor was found as 8 mm drill size (level 1), 3000 rpm spindle speed (level 3) and 50 mm min^{1} feed rate (level 1).
Analysis of variance: The analysis of variance is extensively used to
investigate the design parameters, which significantly affect the quality characteristics.
For this, the total variability of the S/N ratios, which is measured by the
sum of the squared deviations from the total mean S/N ratio, is partitioned
into contribution by each of the design parameter and the error, i.e.,
SS_{T} = SS_{drilldia} + SS_{spindle
speed} + SS_{feed rate} + SS_{error} 
To perform the Ftest, the mean of squared deviations is calculated from the sum of squared deviations of design parameter divided by the degrees of freedom associated with the design parameter. Then, the Fvalue for each design parameter is the ratio between the mean squared deviations and the mean squared error. The Fratio corresponding to 95% confidence level from statistical table is used to compare the Fvalue calculated for each design parameter. The percentage contribution by each of the design parameters in the total sum of squared deviations SS_{T} is a ratio between the sum of squared deviations of each design parameter and the total sum of squared deviations SS_{T}.
Table 5 shows the results of ANOVA for delamination factor. Drill diameter has a significant effect on the delamination value. The percentage contribution of each parameter on total variation is 88.19, 2.62, 7.64%, respectively.
Confirmation test: Using the aforementioned data, one can predict the minimum delamination value using the optimal level of the design parameters selected. To verify the predicted results, a confirmation test was conducted at the first level of drill diameter (A1), the third level of spindle speed (B3) and the first level of feed rate (C1). Table 6 shows the comparison of predicted value with experimental value of delamination along with S/N ratio, good agreement being observed.
Table 6: 
Confirmation results 

CONCLUSIONS
The out come of research work are listed below:
• 
It is found that the Taguchi method of approach using L9 orthogonal
arrays could be used to analyze the delamination of CSMat GFRP material
under drilling operations 
• 
Signal to Noise ratio analysis has been used to find the optimal drilling
parameters suitable for minimal delamination which in turn improves the
hole quality in drilling 
• 
It is experimentally found that the optimum parameters for the drilling
are identified from the S/N response table as 8 mm drill size, 3000 rpm
spindle speed and 50 mm/min feed rate 
• 
It is observed from the ANOVA table that the drill size plays a significant
role in determining the delamination value rather than the other parameters
considered in this work 
• 
ANOVA results also show the percentage contribution of the drill diameter,
spindle speed and feed rate on total variation by 88.19, 2.62 and 7.64%,
respectively 
• 
The experimental results have good agreement with the predicted results
obtained from the confirmation tests 
ACKNOWLEDGMENTS
The authors sincerely thank the Management of SASTRA University and Shanmugha Precision Forgings Ltd. (a unit of SASTRA)Thanjavur for the extended support and facilities provided to complete this work successfully.