Improving Efficiency and Surface Quality in Hard Turning of SKD61 Steel Through Cooling and Cutting Parameter Optimization

Table 3 presents a summary of the results from 27 experimental runs utilizing the L27 orthogonal Taguchi array. The experiments yield data on Ra and MRR across different machining conditions, which encompass cooling methods (dry, MQL, nanofluid), cutting speed (v), depth of cut (d), and feed rate (f). The average Ra values function as the response variable for assessing the impact of these input parameters, whereas the MRR values indicate the machining performance. The table facilitates a detailed analysis of the relationship between machining conditions and performance outcomes.

The Eq. (1) presents the calculation of S/N ratio:

$\frac{S}{N}=-10 log \frac{1}{n}\left(\sum_i^n y_i^2\right)$                    (1)

where, y_i represents the observed data, and n is the number of repeated experiments.

The main effects plot for Signal-to-Noise (S/N) ratios was obtained in Figure 2. By using this plot, researchers can determine which factors and their levels contribute most significantly to minimizing the response, thereby optimizing the machining conditions for improved quality and efficiency. In this instance, Ra serves as the response variable, with lower values being preferable. The optimal Ra value attained with Nanofluid, with a cutting speed of 80 m/min, a depth of cut of 0.2mm, and a feed rate of 0.10 mm/rev, is readily identifiable. The optimal parameters for minimizing Ra identified in the study-cutting speed of 80 m/min, cutting depth of 0.2 mm, and feed rate of 0.10 mm/rev-align with the findings of Do and Phan [12], which found that nanofluid MQL produced better outcomes under similar cutting circumstances. Our multi-objective optimization method, however, goes beyond these results by striking a balance between MRR and Ra innovation that hasn't been highlighted in earlier studies. The adoption of nanofluid cooling as a viable and efficient solution for hard turning applications is further supported by this work.

Table 3. S/N table of 27 experiments

2.png

Figure 2. Main effects plot for S/N ratios

Further analysis using the response table for S/N presented in Table 4 provides a comprehensive understanding of how each machining parameter influences the response variable, Ra. The S/N ratios offer a clear metric to evaluate the robustness of the process against variability, with higher S/N ratios indicating better performance.

As shown in the table, the feed rate (f) emerges as the most influential factor, ranking first with the highest delta value of 0.905645. This result highlights its critical role in determining the surface quality and its sensitivity to changes in the machining process. The cooling condition (c) ranks second with a delta value of 0.600360, demonstrating that the choice of cooling and lubrication significantly impacts Ra. Cutting speed (v) and depth of cut (d) follow, with delta values of 0.273305 and 0.212666, respectively, indicating relatively lower, yet notable, contributions to the response.

The rankings highlight the significance of prioritizing feed rate and cooling conditions in the optimization of the machining process to reduce Ra. This insight aligns with the findings from the Analysis of variance (ANOVA) (Table 5), further corroborating the critical influence of these parameters on machining performance.

ANOVA is a crucial statistical method employed to evaluate the impact of several factors on a response variable and to determine the significance of each component in an experimental investigation. Table 5 presents the ANOVA results for the Ra in this experiment. The analysis reveals that feed rate (f) and cooling condition (c) are the most significant factors affecting Ra, as indicated by their respective p-values, which are less than 0.05 [30]. Specifically, the feed rate demonstrates the highest impact, with a p-value of 0.000 and contributing to 49.9% of the total variation in Ra. This finding underscores the critical role of feed rate in achieving the desired surface finish during machining processes.

Table 4. Response table for S/N ratios

Table 5. Analysis of variance

The cooling condition also plays a notable role, with a p-value of 0.003 and accounting for 16.25% of the total variation. This outcome underscores the efficacy of various cooling and lubricating techniques in affecting Ra. Although cutting speed (v) and depth of cut (d) exhibit a modest impact on Ra, their p-values (0.051 and 0.073, respectively) suggest that these variables lack statistical significance at the 95% confidence level. Together, these two parameters contribute 6.96% and 5.97% to the total variation, respectively.

The high contribution percentages of feed rate and cooling condition to Ra underscore their critical roles in machining performance. Feed rate significantly affects the surface texture because higher feed rates increase the distance the tool travels per revolution, leading to deeper tool marks and a rougher surface finish. In contrast, lower feed rates produce finer cuts, resulting in smoother surfaces. This observation aligns with findings by Abouelatta and Madl [27], who demonstrated that lower feed rates significantly improve Ra due to finer tool passes.

The cooling condition is the second most significant element, accounting for 16.25% of the difference in Ra. Effective cooling and lubrication reduce tool wear, cutting forces, and friction, which collectively enhance surface quality. Nanofluid cooling demonstrates superior performance by enhancing heat dissipation and lubricating properties compared to traditional MQL and dry-cutting conditions.

The interaction between feed rate and cooling conditions has a notable impact on machining outcomes. At lower feed rates, effective cooling conditions like nanofluid maximize heat dissipation and reduce friction, resulting in superior surface finishes. Conversely, at higher feed rates, the thermal and mechanical stresses on the tool increase, making the cooling condition even more critical to maintain acceptable Ra. These findings are consistent with research by Davim [28], which highlighted the importance of balancing feed rate and lubrication to achieve optimal machining results in hard materials

The error term accounts for 17.7% of the variation, reflecting experimental variability not explained by the tested factors. The overall model has an R-squared value of 82.30%, indicating that the selected factors collectively explain a significant proportion of the variability in Ra. These findings provide valuable insights into optimizing machining parameters to enhance surface quality.

The probability plot of Ra, presented in Figure 3, demonstrates a strong alignment between the actual Ra values and the predicted normal distribution. The data points are distributed closely along the center line, indicating that the experimental results conform well to the normality assumption within a 95% confidence interval. The mean Ra value was calculated as 0.9865µm, with a standard deviation of 0.06085µm. Additionally, the Anderson-Darling (AD) statistic of 0.300 and a corresponding p-value of 0.558 further confirm that the data does not deviate significantly from a normal distribution. This result validates the consistency of the experimental measurements and supports the reliability of the statistical analyses conducted in this study.

The multi-objective optimization of cutting parameters was performed utilizing the desirability function approach to concurrently increase MRR and minimize Ra (Figure 4). The optimization results revealed an overall composite desirability (D) value of 0.7473, indicating an effective trade-off between productivity and surface quality. The predicted optimal values for the responses are an MRR of 8038.3220 mm³/min, with an individual desirability of 0.82254, and a Ra of 0.9558 µm, with an individual desirability of 0.67901. The cutting parameters were established to produce best results: cutting speed (v) of 80 m/min, depth of cut (d) of 0.6 mm, feed rate (f) of 0.171 mm/rev, and Nanofluid cooling condition. The desirability plot shows that higher cutting speeds and depths of cut improve MRR but may increase Ra unless a balanced feed rate is maintained. Moreover, the use of Nanofluid cooling proved to be highly effective in reducing Ra while sustaining high MRR, highlighting its advantage over other cooling methods tested in this study.

Nanofluid cooling exhibits superior performance due to its enhanced thermal conductivity and lubricating properties, which facilitate effective heat dissipation and minimize friction at the tool-workpiece interface. The nano-sized SiO₂ particles (100 nm) penetrate microscopic asperities on the workpiece surface, forming a robust lubricating layer that minimizes wear and improves the surface finish. This finding is consistent with prior studies, such as those by Das et al. [17] and Peng et al. [23], who reported similar improvements in machining performance with nanofluids containing oxide-based nanoparticles

The superior performance of nanofluid cooling compared to MQL and dry cutting can be attributed to several critical mechanisms. Dispersed nanoparticles in the fluid improve thermal conductivity, facilitating enhanced heat dissipation from the cutting zone. This minimizes tool wear and thermal deformation, leading to improved surface finishes. Furthermore, the nanoparticles create a protective tribo-film on the surfaces of the tool and workpiece, thereby reducing friction and wear. The self-repairing effect of nanoparticles, where they fill micro-cracks and restore surface integrity, further contributes to improved machining performance. Lastly, the rolling and polishing effects of nanoparticles help smoothen the workpiece surface during machining [31].

However, implementing nanofluid cooling in industrial settings poses certain challenges. The initial cost of nanoparticle-based cutting fluids is higher than traditional fluids, and specialized equipment is required to maintain stable dispersion and prevent agglomeration. There are also concerns about the long-term environmental impact and potential wear on machine components due to nanoparticle residue. Developing effective filtration systems and establishing proper handling protocols are essential to mitigate these risks. Additionally, further research is required to evaluate the recyclability and economic viability of nanofluids for large-scale production.