**3.5 Taguchi analysis**

The Taguchi L9 (3<sup>3</sup> ) Orthogonal Array (OA) was applied. The OA was generated by Minitab 14 consists of 9 runs with 3 factors at 3 levels. **Table 10** shows the Orthogonal Array (OA) of the coated HSS end mills experiment and the combinations of conditions for each control factor (A-C).

## *3.5.1 Regression equation*

Surface roughness equations were generated using machining parameters such as spindle speed, feed rate, and depth of cut. Eq. (1) outline the main effects of surface

*Characterisation and Application of Nickel Cubic Boron Nitride Coating via Electroless Nickel… DOI: http://dx.doi.org/10.5772/intechopen.105364*

**Figure 16.** *Normal probability plot for Ra response.*

roughness and Ra response. **Figure 16** shows the normal probability plot for Ra response based on Eq. (1).

Ra = 0.654 -0.197spindle speed+0.0793Feed Rate+0.0413Depth of Cut (1)

#### *3.5.2 Analysis of variance (ANOVA)*

The OA L9 (3<sup>3</sup> ) contains nine tests of ANOVA investigation that identify the effects of the different parameters on the response variables. A significance level of 95% was chosen in the ANOVA analysis, and the factor was considered adequate if the P-value was less than 0.05 [53]. In this study, the relation of spindle speed (A), feed rate (B), and depth of cut (C) factors on the surface roughness Ra responses are identified using ANOVA analysis. The model was formulated for a 95% confidence level. The P-value shows that the model is significant and has no influence on noise. The experiment result of surface roughness (Ra) formed the first-order model using the Minitab software.

The ANOVA results depicted in **Table 11** is the estimation for machining parameters, with a selected -level of 0.05. The outcomes show that the spindle speed factor has the lowest p-value. This reveals that the consequence of spindle speed is significant as p-value factors that are above 0.05 are considered as insignificant [57].

#### *3.5.3 Factor level combination and determination of optimum parameter*

Based on the rank in **Table 12**, spindle speed ranks first, followed by the depth of cut and feed rate. This demonstrates spindle speed as the significant factor that affects surface roughness. Spindle speed is the most critical machining parameter affecting surface roughness because it is substantially influenced [56]. The table also represents the Taguchi response to determine the optimal factors affecting surface roughness. According to Signal to Noise (smaller is better), the optimum machining settings are 1860 RPM for spindle speed, 334 mm/min for feed rate, and 2 mm for depth of cut are. The experiment was confirmed through the S/N ratio using the optimum parameter level A1B3C2.

The surface finish was the most important influence on spindle speed and feed rate, as shown in **Figures 17** and **18**. The slope between the horizontal line and spindle speed is more pronounced than the depth of cut and feed. The changes in spindle


#### **Table 11.**

*ANOVA table for Ra response.*


#### **Table 12.**

*Response table for S/N ratio (smaller is better).*

**Figure 17.** *Main effects plot for SN ratios.*

*Characterisation and Application of Nickel Cubic Boron Nitride Coating via Electroless Nickel… DOI: http://dx.doi.org/10.5772/intechopen.105364*

speed significantly affect the surface roughness [56]. The optimum machining settings are determined at spindle speed value of 1860 RPM, feed rate of 334 mm/min, and depth of cut of 2 mm.

#### **Figure 18.**

*Main effects plot for means.*

**Figure 19.** *Interaction plot for Ra.*

**Figure 19** shows the interaction plot for surface roughness, Ra in the machining process. When the lines are more non-parallel, an interaction occurs, resulting in higher strength of the interaction. The factors of spindle speed affect the surface roughness more than other factors for machining Aluminium Alloy 7075 with a Ni-CBN HSS coated end mill.
