Statistics and Computational Fluid Dynamics Analyses of the Experimentally Confirmed Thermal Behavior of Self-designed Internally Cooled Smart Cutting Tool


 When compared with dry machining, using traditional cutting fluids has some weaknesses such as environmental pollution, high machining costs and harmful effects on human health. Internally cooled cutting tools (ICCT) have been a promising, sustainable, health-friendly and green technologies for turning applications. However, the effects of different types of internal coolant fluids on insert tip temperature (Ttip) have not been investigated for ICCTs. Within effective cooling, machining quality of metallic materials and tool life can improve. Therefore, a conjugate heat transfer (CHT) model for a self-designed internally cooled smart cutting tool (ICSCT) was set. The CHT simulation was experimentally confirmed using pure water. After that, the effects of flow velocity (Vf), inlet temperature of the coolant fluid (Tinlet) alongside different types of glycol-based heat transfer fluids (including pure water) on Ttip were statistically evaluated by the Taguchi method and analysis of the variance (ANOVA). It was found that the most effective factor was the Tinlet at a contribution ratio level of 88.32%. Additionally, Vf and the type of heat transfer fluid were found to be significant according to statistics. Hence, since no external coolant is used, the designed smart tool can be counted as being environmentally friendly and health friendly. In conclusion, the glycol-based fluids can be a better choice for internally cooled tool designs owing to their superior features, e.g., corrosion prevention, nontoxicity and stable heat transfer capability at lower temperatures compared to pure water although pure water has better thermal properties than the glycol-based fluids.


Introduction
Cooling is a significant requirement for the machining of metallic materials in order to prevent or minimize tool wear, thermal erosion, dimensional inaccuracy and surface roughness of workpieces. Turning is a leading machining application for which the cutting region can reach high temperatures, especially for the case of turning hard to machine materials made of nickelbased superalloys, titanium and composites [1,2]. To decrease the devastating effect of the high heat generated in cutting regions to cutting inserts and workpieces, different cooling technologies have been developed and used in manufacturing processes. These recent technologies can be listed as cryogenic cooling, minimum quantity lubrication (MQL), high pressure coolants, solid lubricants, gas/air-based coolants, vegetable-based cutting fluids, conventional cutting fluids and internally cooled cutting tools (ICCT). Insert tips are protected against failure and wear by cooling of the cutting region. Numerous cutting fluids can be used to provide quality machining performance. However, cutting fluids can cause environmental pollution and health problems for laborers. In addition, the cost of the cutting fluid can be 15-25% of the total manufacturing cost in Germany [3]. In this framework, researchers have investigated the production of different types of cutting fluids or researched the usage of some natural oils. Some studies have raised the concern of needing more natural and environmentally friendly alternatives, especially vegetable oils and biobased vegetable oils [4][5][6][7][8][9]. Biobased-type oils are generally used for MQL applications in machining. They have been used to realize some benefits such as improved machining performance, decreased machining cost and environmental friendliness. Conversely, bacterial contamination risks and poor performance under heavy machining conditions have led to biobased-type oils being considered as a less preferable choice for cutting fluids for industrial applications [4]. Some studies have focused on the use of nanofluids to provide high machining efficiency, high surface quality and good cooling-lubrication in machining applications. These types of studies have produced different types of nanofluids that are strengthened with nanoparticles such as graphene oxide and alumina [10,11]. In spite of the benefits for machining, cutting fluids enhanced with nanoparticles still show some doubts due to their high manufacturing cost and lack of sufficient scientific studies for direct use in manufacturing.
Recently, researchers have tended towards ICCTs because of their promising low cost, adequate environment-friendly cooling and no hazardous health effects compared to conventional cutting fluids [12][13][14][15]The general design concept for ICCTs creates a closed cooling loop in the cutting tool owing to the provision of a dry cutting process to restrict or eliminate usage of cutting fluids. Hence, there is no external cooling. When investigating studies for ICCTs, it appears that most researchers select pure water as the coolant fluid due to its superior cooling features compared to most coolant fluids for dry machining. Additionally, low cost and accessibility of pure water are significant reasons for its preference [1,[16][17][18][19][20][21][22][23][24][25]. Some researchers tent towards ICCTs do not choose pure water as a coolant fluid because of the phase change temperatures.
These studies claimed that internal coolant fluids are able to change the phase to transfer much more of the heat generated from the cutting contact region. As such a phase change can occur as liquid to gas during the machining, heat removal from the cutting region can be provided more effectively. They also used a condenser to turn the evaporated phase of the coolant fluid into the liquid phase in order to flood the inside of the cutting tool again. For this reason, they preferred to use R-123 (hydrochlorofluorocarbon) as a coolant fluid owing to its low boiling point at 28 ⁰C [26,27]. Nevertheless, the fluid is slightly aggressive towards condition changes and leaking problems can occur in the ICCT system. An interesting study has focused on ICCT design and manufacturing by utilizing an antifreeze (Ethylene-Glycol) used in automotive engine coolant. They reported that their design achieved acceptable and efficient cooling to qualify machining [2].
It can be claimed that the literature has still not concentrated enough on the usage of different types of heat transfer fluids used in engineering applications for use as cutting fluid in ICCTs.
Heat transfer fluid selection can be a complex and multifunctional decision where parameters such as thermal conductivity, pumpability and working temperature ranges are involved.
Additionally, it can affect the realization of efficient performance and economy in ICCTs.
However, the selection of the right heat transfer fluid can be narrowed by determining whether or not the maximum-use temperature requirement is above approximately 175 ⁰C in engineering applications. If the need is above 175 ⁰C, one can choose synthetic, organic or silicone fluids.
Conversely, one needs to select inhibited glycol-based fluids, which are used for freezeprotection applications [28]. As we have observed from our previous study, the inserts can be exposed to high temperatures on their tip, but the coolant fluid cannot reach high temperatures due to the poor heat transfer capacity of the inserts [29]. For this reason, inhibited glycol-based fluids, e.g., ethylene glycol, propylene glycol and bioglycol, can be chosen for ICCT applications. The glycol-based fluids show corrosion prevention, nontoxicity and stable heat transfer features. While microbial growth is not an issue because of the biocide feature of ethylene glycol and propylene glycol, bioglycol offers its users an environmentally safer product [30].
A new internally cooled smart cutting tool (ICSCT) was designed and produced for the turning of metallic materials in our previous study. This designed ICSCT was calibrated by computational fluid dynamic (CFD) and a statistic-based method. For checking CFD simulation, an experimental setup consisting of thermocouples and a soldering machine was used [29]. After confirming the CFD simulations for pure water, the thermal behavior of the

Materials and Methods
ICCTs have significant advantages such as decreased surface roughness, tool wear, and machining cost; fewer harmful effects on the environment and human health; and an increased tool life, dimensional accuracy, and machining quality. The designed and manufactured ICSCT promised some capabilities such as a decrease in the tip temperature for the insert (Ttip) of approximately 107 ⁰C and a fixing of the Ttip in a range of temperatures determined by the operator [29]. However, the thermal behavior of the ICSCT has not been compared with different types of internal coolant fluids. Thus, the comparison has been carried out with CFD simulations and the Taguchi method is discussed in this section.

Internally Cooled Smart Cutting Tool CFD Model Design
The CAD design for the ICSCT manufactured in our previous study is shown in Figure 1.a.
The ICSCT has a self-design seat that has a special geometry to flood coolant fluid through the insert tip. After the manufacturing of the ICSCT, a conjugate heat transfer model was built by CFD, as shown in  k-Ɛ turbulence model uses the scalable wall-function to make better accuracy [31]. In this study, a CHT model was selected to simulate more realistic and accurate results. The CFD simulation was set in ANSYS. The designed CHT model was set using a similar geometry, mesh type and boundary conditions as used in the previous study [29]. For clarification, the heat transfer model and boundary conditions are shown in Figure 2 and Tables 1-2.       [32,33].

Design and Analysis of Simulations
The Taguchi method transforms the experimental results to a signal-to-noise ratio (S/N). The S/N ratio is a performance characteristic of the experimental data and is defined as the ratio of the real experiment values to the undesired random noise values [34].
Three performance characteristics can be selected in the Taguchi Method for experimental designs according to decisions made by experts. They are listed as below [35].
• Smaller is the best situations: • Higher is the best situations: • Nominal is the best situations: where yi are the observed data in the experiments and n is the number of experiments for all the equations above. Additionally, � is the mean for the observed data in the experiments and S 2 is the variance for the observed data in Eqs. 4-5.
In this study, different types of glycol-based heat transfer fluids, Vf and Tinlet were optimized for the Ttip data derived from confirmed CFD simulations. Three factors, namely, the heat transfer fluid, Vf and Tinlet, were selected as being related to the hypothesis of "smaller is the best". The hypothesis was selected "smaller is the best" because the tip temperatures in turning process are need to be small to improve machining quality. Each factor was set in four levels for the design of the experiments. The factors and their levels are shown in Table 3. The L16 orthogonal array for the Taguchi method and the S/N ratios calculated by using Minitab software are shown in Table 4.

Simulation Results and Discussion
The S/N ratios for the Ttip data derived from the CFD model were calculated according to Eq.
1 using Minitab software. The main effects of the S/N ratios for Ttip are presented in Figure 4.
The optimal Ttip value will be reached at A3B1C4, as illustrated by a red point in Figure 4. The optimum Ttip can be obtained from CFD simulation under A3B1C4 conditions as 265.55 ⁰C, with an S/N ratio calculated to be -48.48.  In Taguchi method, optimization is need to be verified by confirmation simulations after the determination of the optimal simulation results. If the optimum value determined with the Taguchi method was among the L16 orthogonal array, then the optimum value would not need to be confirmed. In contrast, the optimum value A3B1C4 was not between the L16, and it had to be confirmed using the equations given below. Thus, the optimum values for Ttip and S/N were calculated to be 266.52 ⁰C and -48.51, respectively. The calculated values were almost the same as the simulation results obtained under optimum conditions.
where is the calculated S/N ratio for the optimum condition, ��� is the average S/N ratio for all factors, 0 ��� , 0 ��� 0 ��� are the average S/N ratios when the factors are at optimum levels, and is the calculated Ttip value [34,36].
After statistical analysis, the effects of the heat transfer fluids and Vf values were found to be smaller than that of Tinlet. All factors had significant effects as determined statistically.
However, in order to understand whether they can affect the Ttip significantly according to the

Conclusions
In this study, the thermal behavior of a self-designed ICSCT was assessed using an experimentally confirmed CHT model for pure water. After that, the effects of different types of glycol-based heat transfer fluids, Vf and Tinlet on Ttip were investigated by using the CHT model. A series of simulations were performed according to the Taguchi method, and the results are summarized below: • The cooling effects on Ttip for four fluid types are reasonable in statistically. However, there might no significant differences found according to the terms of engineering. In other words, it was observed that 50% of the volume of the heat transfer fluids had a similar cooling capacity as that of pure water under the determined boundary conditions.
• It was observed that the Ttip values decreased for increasing Vf for all coolant fluids.
• It was determined that the main factor influencing the Ttip values for all cooling fluids was Tinlet. The percentage contribution ratio was found to be 88.32. Namely, a lower Tinlet leads to a lower Ttip. Thus, it can provide high quality machining, cost saving and delayed tool wear.
• The glycol-based fluids have some advantages compared to pure water such as corrosion prevention, nontoxicity and stable heat transfer features. Additionally, their lower temperature limits for cooling are approximately -50 ⁰C. Therefore, glycol-based fluids can be a better choice for use in internally cooled cutting tool designs.
• While bioglycol and pure water offer users environmentally safer products directly, they can create some microbial growth problems. Conversely, microbial growth is not an issue because of the biocide feature of ethylene glycol and propylene glycol. In addition, internally cooled cutting designs have generally no need for external coolants. Thus, such systems can be counted as being environmentally friendly and health friendly even if a harmful cutting fluid is chosen as a constant small volume internal coolant.
• Glycol-based fluids can be used as a coolant at high temperature limits (up to 175 ⁰C) compared to pure water. Using glycol-based fluids can be a better choice due to their upper temperature limits for the turning of difficult to machine materials, which can generate substantial heat in machining.
Next, CFD simulations and experiments for different fluid types can be compared under the same boundary conditions to improve machining quality and the tool life of the insert.