Turning Research of Additive Laser Molten Stainless Steel 316L Obtained by 3D Printing

This paper presents the characteristic of 316L steel turning obtained by 3D printing. The analysis of the influence of turning data on the components of the total cutting force, surface roughness and the maximum temperature values in the cutting zone are presented. The form of chips obtained in the machining process was also analyzed. Statistical analysis of the test results was developed using the Taguchi method.


Introduction
In recent years, an intensive Additive Manufacturing (AM) industry has developed, due to possibilities of manufacturing of very complex structures inside parts and work pieces. This innovative technology is often determined as "3D printing". The technology, in contrast to the conventional (subtractive) method of "top-down" machining processes (SM) which removes material, creates parts via a "bottom-up" process. In AM, the elements are created layer-by-layer by using a computer-controlled laser beam source. This allows fabrication of complex shapes inside of the parts, which are difficult to obtain by using molding or conventional manufacturing techniques [1][2][3]. AM technologies are used in several applications such as turbine blade manufacturing in aerospace engineering, prosthesis and medical implants (in the medical industry), and die casting molds, valves, heat exchangers, manifolds and collectors. One of the main challenges of this technology is the characterization and prediction of manufactured structures and their connection with selected fabrication settings [4][5][6][7].
Both subtractive and additive manufacturing have several advantages and disadvantages. In practice, parts made in additive technology are a replacement for casting technology. The big advantages of the additive manufacturing rely basically on no restrictions to building complex shapes and ability to produce several units at the same time. On the other hand, the disadvantage is undoubtedly the lower surface quality and the dimensional-shape accuracy compared to parts made with the use of machining. The material obtained with 3D printing technology shows porosity and areas with different consistency of material. The characteristics of the parts obtained in the AM technology is the layered structure of the material and necessity to decrease the size of stresses within the material caused by a poor heat distribution [8,9].
We determined the influence of parameters cutting data on the molten laser stainless steel 316L on the surface finish, cutting forces and temperatures values in the cutting zone. A machining analysis of a specially designed sintered turning ring was carried out.

Design of Experiment
The external surface of the test sample was subjected to the longitudinal turning. The 3D samples made of 316L stainless steel (with grain size in range 23-48 µm) were fabricated with SLM/LMF technique using a TRUMPF TruPrint 1000 3D Laser Metal Fusion machine (Ditzingen, Germany). Figure 1 presents a ring-shaped sample and its dimensions. The SLM process parameters are characterized in Table 1.
Materials 2019, 12, 182 3 We determined the influence of parameters cutting data on the molten laser stainless steel 316L on the surface finish, cutting forces and temperatures values in the cutting zone. A machining analysis of a specially designed sintered turning ring was carried out.

Design of Experiment
The external surface of the test sample was subjected to the longitudinal turning. The 3D samples made of 316L stainless steel (with grain size in range 23-48 μm) were fabricated with SLM/LMF technique using a TRUMPF TruPrint 1000 3D Laser Metal Fusion machine (Ditzingen, Germany). Figure 1 presents a ring-shaped sample and its dimensions. The SLM process parameters are characterized in Table 1.  The prepared shape of the test stand enables the installation on a specially prepared attachment bolt. The test position and applied cutting tool, produced by ISCAR company (Tefen, Israel), the cutting insert, DCGT 11T302-AS IC20; and the tool holder, SDJCR2020K11) [32] are shown in Figure 2.    The prepared shape of the test stand enables the installation on a specially prepared attachment bolt. The test position and applied cutting tool, produced by ISCAR company (Tefen, Israel), the cutting insert, DCGT 11T302-AS IC20; and the tool holder, SDJCR2020K11) [32] are shown in Figure 2.
Materials 2019, 12, 182 3 We determined the influence of parameters cutting data on the molten laser stainless steel 316L on the surface finish, cutting forces and temperatures values in the cutting zone. A machining analysis of a specially designed sintered turning ring was carried out.

Design of Experiment
The external surface of the test sample was subjected to the longitudinal turning. The 3D samples made of 316L stainless steel (with grain size in range 23-48 μm) were fabricated with SLM/LMF technique using a TRUMPF TruPrint 1000 3D Laser Metal Fusion machine (Ditzingen, Germany). Figure 1 presents a ring-shaped sample and its dimensions. The SLM process parameters are characterized in Table 1.  The prepared shape of the test stand enables the installation on a specially prepared attachment bolt. The test position and applied cutting tool, produced by ISCAR company (Tefen, Israel), the cutting insert, DCGT 11T302-AS IC20; and the tool holder, SDJCR2020K11) [32] are shown in Figure 2.   During the investigations, measurements of surface roughness, surface topography, cutting forces, micro-hardness and temperature in the cutting zone were performed. Furthermore, the created chips were classified (advantages, disadvantages, acceptable or unacceptable). The turning tests of the sample were carried out on a Masterturn 400 lathe, equipped with a special prepared measurement system that enables the measurement of cutting forces and temperature values in the machining area. The cutting forces were measured and recorded by applying the dynamometer 9257B and the amplifier 5070B produced by Kistler company (Winterthur, Switzerland). The cutting forces waveforms were converted by analog to digital converter and analyzed using DynoWare software (Version 2825A, Kistler Group, Winterthur, Switzerland). Sampling frequency was 1 kHz and measuring time was 10 s. The measurements of surface roughness and topography were performed using Talysurf Intra 50 profilometer produced by Taylor Hobson company (Leicester, UK). The microhardness was measured using Micro-Vickers HM-112 tester produced by Mitutoyo company (Kawasaki, Japan). After the investigations, the average hardness value of the external surface was determined. The measurements were performed along the radius of the workpiece over a distance of 0.5 mm. The analysis of the results did not show any changes in the hardness value, which was HRC mean = 43 (HRC-Rockwell Hardness scale C).

Material
AISI 316L stainless steel is specified as X2CrNiMo17-12-2/1.4404 according to European standard and includes the austenitic structure of stainless steel. The chemical composition and main mechanical properties of this material are presented in Tables 2 and 3, respectively. This stainless steel is used for fabricating parts working in salt water condition; for chemical, paper, and food industries among others; and for architectural elements. Additionally, due to anticorrosive properties, austenitic stainless steels are commonly used. The addition of Molybdenum (Mo) in chemical composition of the stainless steel contributes to increase acetic and sulfuric acid resistance.

Experimental Details
The experimental research was performed according to the Taguchi method [33]. The purpose of the tests was to examine the impact of some machining data, such as feed rate f and cutting speed v c , on: • component values of the total cutting force F, main force F c , feed force F f and thrust force F p ; • surface roughness parameter values Ra and Rz; and • maximum temperature value T max in machining zone.
Cutting depth a p = 0.5 mm and cutting edge radius r ε = 0.2 mm were assumed. The cutting data values of the turning experiment are presented in Table 4. To statistically fit the experimental data, the polynomial was selected.
The strategy of factor analysis S/N (signal to noise) was determined as "smaller-is-better" according to following formula: where: y i is the respective characteristic and n is the number of observations.

Results Analysis of Cutting Forces Measurements
In Table 5, the experimental results of the components of the total cutting force with standard deviations (Std. Dev.) are shown. The cutting forces, such as F c , F f , and F p , are presented as their average values. The impact of feed rate f on the cutting forces values and the total cutting force F, are presented in Figure 3. Table 5. Experimental design and results of the components of the total cutting force. The results show that the feed rate f significantly affects the total cutting force F. In the case of the longitudinal force F c , the increase of the feed rate values causes a stable increase of the force with constant increments. A threefold increase of feed rate, from f = 0.07 mm/rev to f = 0.211 mm/rev, contributes to the cutting force increase F c of about 120-150 N in relation to the cutting speed. It was observed that, when applying the cutting speed v c = 100 mm/min for feed rate f > 0.15 mm/rev, the increment of thrust force F p decreases and the increment of feed force F f increases. This result indicates changes in the direction of the forces (F f and F p ) in the case of using higher cutting speed values. The analysis of the results shows a decrease of about 10% of the cutting force F c when using the higher cutting speed (100 m/min vs. 60 m/min). The obtained results of S/N parameters and its values for the cutting forces (F c , F f , and F p ) are presented in Table 6. In Figure 4, the impact of cutting data on the cutting forces is shown.
The analysis of results confirms that the feed rate mainly affects the values of the cutting forces F c , F f and F p . The cutting speed increase causes a decrease of values for all components of the total cutting force. Tables 7-9 show the ANOVA regression analysis results of the components for the total cutting force (where: DF-degrees of freedom, Seq SS-sums of squares, Adj SS-adjusted sums of squares, Adj MS-adjusted means squares).    The results show that the feed rate f significantly affects the total cutting force F. In the case of the longitudinal force Fc, the increase of the feed rate values causes a stable increase of the force with constant increments. A threefold increase of feed rate, from f = 0.07 mm/rev to f = 0.211 mm/rev, contributes to the cutting force increase Fc of about 120-150 N in relation to the cutting speed. It was observed that, when applying the cutting speed vc = 100 mm/min for feed rate f > 0.15 mm/rev, the increment of thrust force Fp decreases and the increment of feed force Ff increases. This result indicates changes in the direction of the forces (Ff and Fp) in the case of using higher cutting speed values. The analysis of the results shows a decrease of about 10% of the cutting force Fc when using the higher cutting speed (100 m/min vs. 60 m/min). The obtained results of S/N parameters and its values for the cutting forces (Fc, Ff, and Fp) are presented in Table 6. In Figure 4, the impact of cutting data on the cutting forces is shown.     (2)- (4): Table 10 presents the results of measured surface roughness Ra and Rz. The examples of topographies and profiles of the parts surface are shown in Table 11 (Trial 1 for f min and Trial 7 for f max , v c = 60 m/min).                  Presented analysis of the relations in Figure 5 shows that values of surface roughness parameters are proportional to feed rate values. In all cases, higher values of surface roughness Ra and Rz are obtained with higher values of cutting speed, during changes from v c = 60 m/min to v c = 100 m/min. Moreover, it was observed that a higher dispersion of the measured values for v c = 100 m/min is accrued. The graphical representations of the impact of the surface roughness parameters Ra and Rz on the cutting data and S/N factor are presented in Figure 6. The cutting feed increase causes the increase of surface roughness Ra and Rz. Results analysis presented in Figure 6 additively confirms the most significant impact of feed rate on the cutting forces Fc, Ff and Fp. Tables 13 and 14 present the ANOVA regression analysis results for each roughness parameter. Ra (f, vc) and Rz(f, vc) are described by Equations (5) and (6).  Figure 7 shows photographs of the obtained chips for selected compositions of the experimental design (Trial 1 for fmin and Trial 7 for fmax). During the experimental research, the classification of created chips was performed. A three-step scale was adopted: "+", advantageous The cutting feed increase causes the increase of surface roughness Ra and Rz. Results analysis presented in Figure 6 additively confirms the most significant impact of feed rate on the cutting forces F c , F f and F p . Tables 13 and 14 present the ANOVA regression analysis results for each roughness parameter. Ra(f, v c ) and Rz(f, v c ) are described by Equations (5) and (6). Figure 7 shows photographs of the obtained chips for selected compositions of the experimental design (Trial 1 for f min and Trial 7 for f max ). During the experimental research, the classification of created chips was performed. A three-step scale was adopted: "+", advantageous chips; "−", disadvantageous chips; and "0", unacceptable chips. In all experimental tests, unacceptable chips were obtained (long, tangled, and spiral).

Results Analysis of Temperature in Cutting Zone
The temperature measurements were performed using a FLIR SC 620 thermal camera (FLIR Systems, Wilsonville, OR, USA)which was installed above the cutting zone and connected to a computer. ThermaCam Researcher Pro 2.9 (FLIR Systems, Wilsonville, OR, USA) was used for acquisition and analysis of the recorded thermograms. Two-second sequences of a stable phase of machining process (30 frames per second) were recorded and the maximum temperature Tmax that existed in the cutting area was obtained. The main errors during temperature measurements are the emissivity factor and reflections. In our case, the emissivity factor was 0.98. The configuration parameters of the thermal camera are presented in Table 15.

Results Analysis of Temperature in Cutting Zone
The temperature measurements were performed using a FLIR SC 620 thermal camera (FLIR Systems, Wilsonville, OR, USA)which was installed above the cutting zone and connected to a computer. ThermaCam Researcher Pro 2.9 (FLIR Systems, Wilsonville, OR, USA) was used for acquisition and analysis of the recorded thermograms. Two-second sequences of a stable phase of machining process (30 frames per second) were recorded and the maximum temperature Tmax that existed in the cutting area was obtained. The main errors during temperature measurements are the emissivity factor and reflections. In our case, the emissivity factor was 0.98. The configuration parameters of the thermal camera are presented in Table 15.

Results Analysis of Temperature in Cutting Zone
The temperature measurements were performed using a FLIR SC 620 thermal camera (FLIR Systems, Wilsonville, OR, USA)which was installed above the cutting zone and connected to a computer. ThermaCam Researcher Pro 2.9 (FLIR Systems, Wilsonville, OR, USA) was used for acquisition and analysis of the recorded thermograms. Two-second sequences of a stable phase of machining process (30 frames per second) were recorded and the maximum temperature Tmax that existed in the cutting area was obtained. The main errors during temperature measurements are the emissivity factor and reflections. In our case, the emissivity factor was 0.98. The configuration parameters of the thermal camera are presented in Table 15.     The polynomial T max (f, v c ) is described by Equation (7).
The impact of the cutting data on the values of the maximum temperature in the cutting zone is shown in Figure 9.
During the experimental research, the camera was installed perpendicular to the cutting zone and recorded the flown chip on the rake face of the cutting insert. It had the most impact on the recorded value of the temperature. The feed rate increase contributes to a decrease of the maximum temperature recorded by the thermal camera. The section of the cutting layer and chip thickness increase with a feed increase. Further, the part of generated heat flux on the junction chip and cutting edge spreads in more material volume. The cutting speed increase causes a decrease of the temperature value in the cutting zone. It can result from the shorter contact time between the chip and the cutting edge, which effects on the decrease of the heat source friction. Figure 10a,b shows the relation between the average maximum temperature and feed rate during the applied cutting speed of v c = 60 m/min and v c = 100 m/min.
The analysis of Figure 10 shows that an applied lower cutting speed causes higher temperature values in the cutting zone. A similar correlation was observed for the components of the cutting forces.
( , ) = 245 − 625 + 0.13 + 2245.59 2 − 2.53 , The impact of the cutting data on the values of the maximum temperature in the cutting zone is shown in Figure 9. During the experimental research, the camera was installed perpendicular to the cutting zone and recorded the flown chip on the rake face of the cutting insert. It had the most impact on the recorded value of the temperature. The feed rate increase contributes to a decrease of the maximum temperature recorded by the thermal camera. The section of the cutting layer and chip thickness increase with a feed increase. Further, the part of generated heat flux on the junction chip and cutting edge spreads in more material volume. The cutting speed increase causes a decrease of the temperature value in the cutting zone. It can result from the shorter contact time between the chip and the cutting edge, which effects on the decrease of the heat source friction. Figure 10a,b shows the relation between the average maximum temperature and feed rate during the applied cutting speed of vc = 60 m/min and vc = 100 m/min. The analysis of Figure 10 shows that an applied lower cutting speed causes higher temperature values in the cutting zone. A similar correlation was observed for the components of the cutting forces.

Conclusions
The following can be concluded from the performed experimental research: 1. Speed rate f has a significant effect on the values of the cutting forces. The speed rate increase causes the linear increase of all components of the cutting forces. The values of the cutting forces can be decreased by the increase of the cutting speed values. During the applied cutting speed of vc = 100 m/min, the total cutting force is about threefold lower than for the applied vc = 60 m/min.

Conclusions
The following can be concluded from the performed experimental research: 1.
Speed rate f has a significant effect on the values of the cutting forces. The speed rate increase causes the linear increase of all components of the cutting forces. The values of the cutting forces can be decreased by the increase of the cutting speed values. During the applied cutting speed of v c = 100 m/min, the total cutting force is about threefold lower than for the applied v c = 60 m/min.

2.
Surface roughness values (Ra and Rz) are connected to the feed rate f and cutting speed v c . For threefold increase of the speed rate f, values of surface roughness parameters Ra increase 2.5-fold, and values of surface roughness parameters Rz increase about 1.5-fold. In addition, higher values of surface roughness parameters (Ra and Rz) were obtained for v c = 100 m/min than v c = 60 m/min.

3.
Values of the average maximum temperature T max in the cutting zone decrease with the increase of the speed rate f and the cutting speed v c ; the correlations are connected to chip thickness and contact time chip between the chip and the cutting edge, respectively. 4.
The applied cutting data have no effect on the chips form-all of them were unacceptable.