Microstructures and machinability of the Inconel 182 overlays at different overlay thickness deposited by shield metal arc welding

Abstract

Nuclear steam turbines are well-known working in saturated wet-steam environment. The use of anti-corrosion material depositing or coating on the critical components and then machining to the required thickness with specific surface quality can be an effective and economical way to prevent critical components of steam turbines from corrosion and erosion. Inconel 182 can be this anti-corrosion material; however, a few literature has reported its machinability, not to mention its machinability at different overlay thickness. The objective of this study is to investigate the influence of the coated overlay thickness on microstructures and machinability of Inconel 182 overlays. First, the micro-hardness and microstructures of Inconel 182 overlays at different overlay thickness are studied. Afterwards, cutting forces, cutting temperature, surface roughness and the machined surface morphology at different overlay thickness are discussed. Finally, the prediction model for machining Inconel 182 overlays at different overlay thickness is established. The results indicate that the microstructures of Inconel 182 overlays strongly rely on overlays thickness and Inconel 182 electrode diameter (3.2 mm). The prediction model illustrates that the machinability of Inconel 182 overlays is acceptable at overlay thickness from about 0 to 2.1 mm; it is poor at overlay thickness from about 2.1 to 6.4 mm, and it is stable at overlay thickness from about 6.4 to 10 mm.

Keywords

Inconel 182 overlays overlay thickness microstructures machinability

Introduction

Surfacing technology has become a separate field of engineering in recent years. Surfacing technology can be used to renovate worn-out surfaces or deposit overlays on new parts to prevent corrosion and erosion in metallurgy, petro-chemistry, agriculture, nuclear power supply and many other fields. In nuclear power supply field, as nuclear steam turbines working in saturated wet-steam environment, it is essential to avoid the critical components such as rotor from erosion and corrosion.¹ The use of surfacing technology to deposit and coat anti-corrosion material on the steam seal as overlays can be an effective and economical way to prevent rotor supporting plane from erosion, corrosion and stress corrosion cracking.

A key factor in the application of surface technology is to select the suitable material because the use of suitable material as overlays can save material, power and work, and it can also improve the overlay’s service life.² Nickel-based alloys are well known as popular material with superior advantages of high resistance to thermal fatigue, thermal shock, creep and erosion.^3–6 Among the nickel-based alloys, Inconel 182 is a typical filler material with superior high resistance to thermal fatigue, erosion and corrosion, which has been widely used as overlays in nuclear steam turbine remanufacture.^7–9 As Inconel 182 is a kind of typical welding material,¹⁰ many scientific researchers focused on the microstructural and compositional features of Inconel 182 as well as its corresponding intergranular cracking or intergranular stress corrosion cracking properties. Bao et al.¹¹ studied the stress corrosion cracking sealing in Inconel 182 overlays by laser surface melting. They found that the Inconel 182 intergranular cracking or intergranular stress corrosion cracking damage could easily occur in nuclear power plants. He also found the chromium depletion and chromium precipitation of Inconel 182 overlays during the welding process. Dehmolaei et al. used the Inconel 182 as filler materials to weld the dissimilar high-performance (HP) heat-resistant steel and Incoloy 800 joints. It can be concluded that the Inconel weld metal, when deposited from Inconel 182, revealed a higher inclusion content than welds produced by gas-shielded welding, and the migration of grain boundaries in the Inconel 182 weld metal was very extensive.¹² Peng et al. investigated the dendrite-boundary microchemistry of Inconel 182 using auger electron spectroscopy analysis. They found that chromium depletion is dominant in the precipitate-free zone. The majority of the precipitates at the dendrite boundary are mass-precipitated and fine-sized chromium carbides.¹³

In some cases, it is possible to use the remanufactured parts with an overlay without any additional manufacturing; but sometimes, it is necessary to machine the remanufactured overlays to the required thickness and surface quality. The necessity of additional machining of the remanufactured overlays depends on the requirements of the shape and size of the part, and on the surface quality.

Compared to the conventional metals, the overlays have some distinct properties that result in more troublesome in the manufacturing process.¹⁴ They can be described as follows:

Overlay surfaces are uneven that cutting changes in the zone of the interrupted cut, which lead to serve tool wear.

The chemical composition of the overlays is usually different from the chemical composition of the substrates, which will induce chemical loss by burning, mixing or diffusing during this dissimilar metal welding process.

The microstructures of overlay are sensitive to the cooling condition after surface remanufacturing; therefore, it affects the mechanical properties and so there is a significant scatter in their range of values.

So far, several scientific researchers are interested in the machinability of overlays. Brozek^14,15 studied the optimum cutting condition and tool’s long-term life. When turning overlays with cemented carbides, cubic boron nitride and ceramics, he found that the best results were achieved using the Kennametal Hertel WNMG 080412 KC9315 insert, and the surface roughness of turned surface slowly increases that extensive deterioration occurs at the end of the tool life. Wang et al.¹⁶ evaluated the machinability of Fe-based amorphous alloy with respect to cutting forces using Box–Behnken experimental design. He also observed the surface roughness, chip morphology and tool wear behavior in machining Fe-based amorphous alloy overlays.

However, the overlay thickness in these studies is thinner than 1.5 mm, and they are not considered as the overlay thickness will affect the machinability of overlays. A few literature has reported the machinability of Inconel 182 overlays, not to mention its machinability at different overlay thickness; especially, the overlay thickness is much larger than 1.5 mm. In addition, as the Inconel 182 overlays deposited on middle plane of nuclear steam turbine are complex and large (Figure 1), face milling can be an efficient machining solution. However, the face milling cutter is usually large enough that it cannot reveal subtle variations when machining this kind of non-uniformed material Inconel 182.

Figure 1.

Application of Inconel 182 overlays covering on the steam seal in nuclear steam turbine.

Thus, this study investigates the microstructures and evaluates the machinability of Inconel 182 overlays at different overlay thickness by side milling. First, the micro-hardness and microstructures of Inconel 182 overlays at different overlay thickness are presented. Afterwards, the influence of overlay thickness on the cutting forces, cutting temperature, machined surface roughness and the machined surface morphology is discussed in detail. Finally, the prediction model for machining Inconel 182 overlays at different overlay thickness is established.

Experiment

Workpiece preparation

The solution-annealed G17CrMo9-10 mold steel, containing up to 90% Fe, 2.3% Cr, 1.2% C, 1.1% Mo, 0.6% Si and small amount of Mn, Cu and Ti, was used as substrates. They were prepared into rectangular blocks with the dimensions of 80 mm × 40 mm × 27 mm. Before depositing, the substrates were degreased with acetone, dried in air, polished and grinded. Then Inconel 182 electrodes were deposited on the substrates by shield metal arc welding (SMAW). Argon gas was continuously supplied as the shielding gas to prevent oxidation. Aging treatment was performed to eliminate the residual stress generated in SMAW process. The details on welding parameters are listed in Table 1. After SMAW, the nominal percentage composition of the deposited Inconel 182 overlays is 60% Ni, 15% Cr, 12% Fe, 7.6% Mn, 1% Si, 1% Ti and small amount of Nb, Ta and Cu, and its mechanical properties such as tensile strength and yield strength are 550 and 360 MPa, respectively. The cross section of prepared workpiece is shown in Figure 2. Inconel overlays are deposited to 13 mm to fulfill the design requirements of the nuclear steam turbine.

Table 1.

Welding details.

Process	SMAW
Preheat	400 K
Shielding gas	Argon
Electrode diameter	3.2 mm
Arc current	145 A
Arc voltage	17 V
Traveling speed	1.4 mm/s
Total deposited weld depth	13 mm

SMAW: shield metal arc welding.

Figure 2.

Cross section of prepared workpiece.

Experimental condition

The experimental set-up for machining Inconel 182 overlays is illustrated in Figure 3. Side milling test was conducted on computer numerical control (CNC) vertical machining center Hurco VMX 42 with a maximum spindle speed of 12,000 r/min. An L₉ (3³) orthogonal array was employed as the experimental design. Cutting speed Vc, feed rate f and radial depth of cut a_w were selected as three factors, and each factor had three levels, which are illustrated in Table 2. Axial depth of cut a_p remained 10 mm during all the testing. Side milling tests are conducted first at an overlay thickness of 13 mm with radial depth of cut a_w optimized in section “Milling parameters optimization” as 1 mm. Then, the overlay thickness of Inconel 182 reduces to 12 mm. Keep on cutting with a_w, at this moment, the overlay thickness is 11 mm. Step by step, the machinability of Inconel 182 overlays at different overlay thickness evaluation can be achieved. The OSG MG-EML-type uncoated cemented carbide milling cutter with four flutes was used as the cutting tool and its diameter was 6 mm. Before each test, a new tool was employed to avoid the influence of tool wear, and all the tests were conducted with three replicates to eliminate the accidental phenomena. All the milling experiments were conducted as down milling and dry cutting condition were employed to facilitate infrared temperature field capture and also prevent the cutting fluid influence.

Figure 3.

Experimental set-up for milling Inconel 182 overlays.

Table 2.

Cutting parameters and their corresponding levels.

Levels	Factors
	A, a_w (mm)	B, f (mm/tooth)	C, Vc (m/min)
1	0.5	0.1	80
2	1	0.2	120
3	0.5	0.3	160

Prior to the milling tests, the micro-hardness of Inconel 182 overlays at different thickness was measured by HR-150A micro-hardness tester. The microstructures of Inconel 182 overlays at critical thickness were observed by Leica MEF4M metallographic microscope. During the milling tests, a Kistler 9272 stationary dynamometer, a four-channel charge amplifier Kistler 5017B and a high-frequency A/D data acquisition board with LabVIEW software were adopted to measure cutting force components on-line. The cutting forces were all measured as the tool reached the midpoint of the workpiece so as to achieve complete stabilization and avoid the transient state. After milling tests, a Mitutoyo SJ-210 tester was employed to measure the surface roughness, and the cut-off distance was set to 0.25 mm. The values of surface roughness were the averages of three time measurements at different locations along the feed direction. Chip morphology and machined surface morphology were observed by optical microscope Keyence VHX-500FE with the magnification of 50×. In addition, cutting temperature images were captured by FLIR A615 infrared camera, and the infrared emissivity was set according to the manufacturer as 0.6.

Result and discussion

Properties of Inconel 182 overlays

Micro-hardness

Figure 4 illustrates the average micro-hardness of Inconel 182 overlays, fusion zone and the heat-affected zone (HAZ) as a function of overlay thickness. Obviously, the micro-hardness at the fusion zone shows the largest micro-hardness value of approximately 330 HV due to re-cooling during the welding process, followed by approximately 280 HV at the HAZ, and the weld metal Inconel 182 shows the lowest micro-hardness. The micro-hardness of Inconel 182 is found to be depended on the overlay thickness. As the overlay thickness increases, the micro-hardness increases gradually and reaches its maximum value at overlay thickness of about 4 mm. Then, the micro-hardness decreases gradually with the increase in overlay thickness. As the overlay properties are sensitive to the welding times and the corresponding cooling condition, the variation of the micro-hardness can be attributed to the chemical composition loss by burning and by mixing the substrates and the Inconel 182 electrode during the welding process.¹⁵

Figure 4.

Micro-hardness of Inconel 182 overlays at different overlay thickness.

Microstructures

The microstructures of Inconel 182 overlays deposited by SMAW at different overlay thickness are observed to further investigate why the micro-hardness varies. It is obvious from Figure 5 that the microstructure of Inconel 182 overlays is fully austenitic with a dendritic morphology, and its morphology are quite different in three regions (marked A, B and C).

Figure 5.

Microstructures of Inconel 182 overlays at critical overlay thickness: (a) Inconel 182, (b) Region A (1 mm),(c) Region B (4 mm) and (d) Region C (7 mm).

As shown in Figure 5(b), Region A is the area nearest to the fusion zone with the overlay thickness of 1 mm—the overlay thickness is smaller than the electrode diameter (3.2 mm). The Inconel 182 overlays in this region have undergone only one welding process. In this case, the melting point of weld metal Inconel 182 is higher than the substrate G17CrMo9-10, which results in some chromium in Inconel 182 grain boundary which diffuses into the melted zone of substrate during the welding process. The depletion of chromium will, in some extent, reduce the micro-hardness of Inconel 182 overlays.¹⁷ Meanwhile, due to the relative high welding temperature, a small amount of carbon in substrate can in turn diffuse into Inconel 182 overlays and forms small round Si carbide and Ti carbide precipitates around grain boundaries, strengthening the hardness. The Si carbide and Ti carbide precipitation in Inconel 182 overlays depends strongly on the heat treatments of the material and the environmental temperature during the solidification phase in welding process. Thus, it can be inferred that the increase in the micro-hardness with increasing the overlay thickness from 0 to 3.2 mm is mainly attributed to the chromium depletion and Si carbide and Ti carbide precipitation.

Region B is the middle area with the overlay thickness of 4 mm, as shown in Figure 5(c). As the overlay thickness is larger than the diameter of Inconel 182 electrode (3.2 mm), the material in Region B had undergone another welding process, and the microstructure in this region shows the recrystallized features with extensive grain boundary migration where the recrystallized grain boundaries cut across the solidification substructures.¹⁸ During the solidification phase of the welding, redistribution of the solutes occurs, and as a result, two types of precipitates have been distributed on the dendritic grain boundaries during the cooling phase after solidification. One is small round Si carbide and Ti carbide hard particles. The other is polygonal chromium carbides of Cr and C as a major metallic element and small amounts of Fe and Mn. Due to relative long distance from the fusion line, the carbon in substrate G17CrMo9-10 cannot effectively diffuse into the Inconel 182 overlays; thus, the amount of small round Si carbides and Ti carbides are dramatically reduced. As the substrate and weld metal are both Inconel 182 during this secondary welding process, the chromium depletion due to difference in melting temperature between base metal and weld metal will no longer happen. In addition, as the dissolution temperature of chromium carbide is far lower than the melting temperature of the matrix,¹⁹ the chromium carbide precipitates during the cooling phase after solidification of the matrix will significantly increase the micro-hardness. Therefore, the micro-hardness in this region reaches the maximum value.

Region C is the outer area with the overlay thickness of 7 mm, as shown in Figure 5(d). In this region, the overlay thickness is larger than twice the diameter of the Inconel 182 electrode (6.4 mm). It is observed that there exist almost no Si carbides and Ti carbides, which demonstrates that welding heat during third welding pass cannot dissipate to melt the G17CrMo9-10 and the carbon in G17CrMo9-10 cannot diffuse into the Region C. Thus, only chromium carbide precipitates during the cooling phase after solidification of the matrix. Under these circumstances, the micro-hardness is a little bit smaller than that in Region B.

Machinability of Inconel 182 overlays at different overlay thickness

Milling parameter optimization

Surface roughness, closely related with the cutting force, cutting vibration and tool wear, is a direct and important indicator of the machined surface quality. Figure 6 illustrates the main effect plots of mean value for surface roughness Ra. It is observed that Vc has minor effects on Ra values. Ra decreases slightly with the increase in cutting speed. This decrease can benefit to softening effect of workpiece material under the high temperature in the main shear deformation region as the cutting speed is relatively high.²⁰ On the contrary, feed rate f and radial depth of cut a_w have major effects on Ra values. When the factors of f and a_w were in high levels, an increase in f from 0.2 to 0.3 mm/tooth and a_w from 1 to 1.5 mm will increase the cutting load and thereby exacerbate the cutting force and tool wear significantly, as a result, deteriorates the surface finish. However, increasing f from 0.1 to 0.2 mm/tooth and a_w from 0.5 to 1 mm resulted in almost no rise in the surface roughness. This can be attributed to the squeeze and plow phenomenon when machining the Inconel 182 overlays at relative low levels of feed rate and radial depth of cut. Comprehensively, cutting speed Vc = 160 m/min, feed rate f = 0.2 mm/tooth and radial depth of cut a_w = 1 mm can be the optimal combination of the cutting parameters.

Figure 6.

Main effect plots of mean value for surface roughness Ra.

In the following sections, the optimal cutting parameters were employed to investigate the machinability of Inconel 182 overlays.

Cutting forces

Cutting forces in milling process can be divided into three components, which are feed force F_x, radial force F_y and axial force F_z. Due to the characteristics of side milling, axial force F_z usually keeps a relatively small value and it has minor effects on the total cutting forces.^21,22 Thus, we will first focus on the feed force F_x and radial force F_y, respectively, when side milling Inconel 182 overlays.

As shown in Figure 7, feed force F_x and radial force F_y when side milling Inconel 182 overlays at different overlay thickness show quite different variation trend as the overlay thickness increases. It can be seen that maximum feed force F_x fluctuates at approximately 125 N, and they are not sensitive to overlay thickness while radial force F_y are relatively larger,²³ and their values strongly rely on overlay thickness. This implies that radial force F_y is the main cutting force component.²⁴ Furthermore, as the singular cutting force components F_x and F_y are quite different, and they lack the ability of being independent on the tool path or the orientation of the cutter in reference to the workpiece,²⁵ the resultant cutting forces F_xy can be used to evaluate the machinability when side milling Inconel 182 overlays.

Figure 7.

Maximum cutting forces of milling Inconel 182 overlays at different overlay thickness.

It is obvious that the resultant cutting forces F_xy show the same variation trend with F_y. The lowest F_xy lies in the fusion line, and it increases at the fusion zone and then decreases at the HAZ. Fusion zone is the narrow area of the weld metal (Inconel 182) and the base metal (G17CrMo9-10). In this area, the weld metal in welded joints is transited to the HAZ. The base metal melted and resolidified in fusion zone during the welding process will change the mechanical properties.¹⁸ After resolidification, the machinability of the mixed and redistributed material in fusion zone (Figure 2) will change. In addition, the micro-hardness due to recrystallization in HAZ is apparently higher than the weld metal Inconel 182, which demonstrates that it is relatively more difficult to cut. Thus, cutting forces in fusion zone will gradually decrease when machining away from the HAZ. When machining Inconel 182 overlays, the F_xy increases gradually with the increase in overlay thickness from 0 to 4 mm and reaches its maximum value at overlay thickness of approximately 4 mm. Then, they decrease gradually with the increase in overlay thickness approximately from 4 to 7 mm. Finally, it keeps constant at about 200 N. This change can be related to the unique microstructures of Inconel 182 overlays as mentioned above.

When the overlay thickness is smaller than the electrode diameter of 3.2 mm, the Inconel 182 overlays has undergone only one welding process. In this case, the weld metal is Inconel 182, while the substrate is G17CrMo9-10, which belongs to dissimilar metal welding process. Attribute to the melting point of weld metal Inconel 182 is higher than the substrate G17CrMo9-10, some chromium in Inconel 182 grain boundary diffuses into the melted zone of substrate to induce chromium depletion. At the same time, the carbon from the substrates will diffuse into Inconel 182 overlays to form chromium carbide hard particles with the remained chromium element and precipitate at grain boundaries during the cooling phase after solidification. The chromium depletion and chromium carbide hard particle precipitation are strongly related to the distance of the fusion line. That is to say, the thicker the Inconel 182 overlays (ranging from 0 to 4 mm), the more the chromium carbide hard particles will precipitate. The precipitate chromium carbide hard particles will significantly increase the cutting force. Therefore, F_xy increases gradually with the increase in overlay thickness from 0 to 4 mm and reaches its maximum value at overlay thickness of approximately 4 mm.

When the overlay thickness is larger than the electrode diameter of 3.2 mm, the weld metal and substrate are both Inconel 182. At this secondary similar metal welding process, the chromium carbide precipitation due to difference in melting temperature between base metal and weld metal will reduce gradually. As a result, the cutting force decreases gradually.

When the overlay thickness is larger than twice the diameter of the electrode (6.4 mm), the Inconel 182 overlays has undergone the welding process for the third time. At this moment, no Si and C from G17CrMo9-10 can diffuse into weld metal to form Si and Ti carbides. Instead, only chromium carbide precipitates. Therefore, the cutting force decreases slightly.

Cutting temperature

Cutting temperature curve and their corresponding captured infrared images when side milling Inconel 182 overlays at different overlay thickness are illustrated in Figures 8 and 9, respectively. Similarly, it can be seen that cutting temperature shows the highest value of 324 °C when milling at fusion zone, about −2 mm away from the fusion line. It can be also observed that the cutting temperature gradually decreases in the fusion zone, which can be attributed to the mechanical properties variation. When milling Inconel 182 overlays, the influence of the overlay thickness on the cutting temperature variation strongly depends on whether the overlay thickness is larger than 4 mm or not. If the overlay thickness is thinner than 4 mm, the cutting temperature increases significantly with the increase in overlay thickness and reaches its maximum value of 258 °C. This significant increase attributed to the chromium carbides precipitating at grain boundaries during the cooling phase after solidification will scratch the cutting tool and induce severe tool wear.²⁶ As a consequence, cutting temperature due to excessive tool wear increases severely.²⁷ On the contrary, if the overlay thickness is thicker than 4 mm, the cutting temperature is stabilized at about 257 °C, demonstrating that there is almost no obvious influence of overlay thickness on the cutting temperature variation when the overlay thickness keeps on increasing from 4 to 10 mm. The reason why the cutting temperature is stabilized when the overlay thickness is from 4 to 10 mm is that the disappeared small round Si carbides and Ti carbides in this region (Figure 5) are relatively small, and they cannot induce the additional tool wear and cutting heat. Thus, cutting temperature in this region is stable.

Figure 8.

Cutting temperature of milling Inconel 182 overlays at different overlay thickness.

Figure 9.

Infrared images of milling Inconel 182 overlays at critical overlay thickness.

Chip morphology at critical overlay thickness (D = 1 mm and D = 4 mm) when side milling Inconel 182 overlays is also shown in Figure 10. It can be observed that chip shapes are almost unchanged, and they all kept the same curling pattern belonging to little spiral chip type.²⁰ However, chip colors are quite different at different cutting positions. The color of the generated chips at overlay thickness of 1 mm is dark brown, while the chip color at overlay thickness of 4 mm is blue. This phenomenon is closely related to the cutting temperature of 224 °C and 258 °C when machining Inconel 182 overlays.

Figure 10.

Cutting chips when milling Inconel 182 overlays with overlay thickness of (a) D = 1 mm and (b) D = 4 mm.

Tool wear

Tool wear when milling Inconel 182 overlays at critical overlay thickness after cutting 80 mm is shown in Figure 11, and it can be seen that tool wear differs with the overlay thickness. Tool wear performance when milling at overlay thickness of 1 mm is shown in Figure 11(a). As low cutting temperature is generated in Figure 8, there is no adhesion in rake face. Tool flank wear slightly and flank mean wear land is not more than 10 µm. However, micro chipping existing at the cutting edge can be attributed to tool edge which unfortunately encounters with the Si and Ti carbide hard particles, scratching severely, which is just a coincidence. Thus, tool wear performance is good when milling at overlay thickness of 1 mm. As overlay thickness increases to 4 mm, as shown in Figure 11(b), the rake face presents a large area of adhesive wear and crater wear, demonstrating that tool–workpiece interface temperature is high enough (Figure 8) that the Inconel 182 adhere to the tool rake face due to its adhesive properties. Meanwhile, tool flank wear a lot and its mean flank wear land reaches 0.1 mm. In this circumstance, milling at overlay thickness of 4 mm, the tool wear performance is poor. When overlay thickness is 7 mm, as shown in Figure 11(c), although adhesion still exists at tool rake face, tool flank wear is uniform. In this case, tool wear performance is acceptable as long as cutting fluid is introduced. When overlay thickness is 10 mm, as shown in Figure 11(d), the adhesion phenomenon reduces and tool flank wear land is small, indicating that tool wear performance when milling at overlay thickness of 10 mm is relatively good.

Figure 11.

Tool wear when milling Inconel 182 overlays at critical overlay thickness after cutting 80 mm: (a) D = 1 mm, (b) D = 4 mm, (c) D = 7 mm and (d) D = 10 mm.

Surface roughness and morphology

Figure 12 illustrates that the surface roughness Ra is parallel to feed when milling Inconel 182 overlays at different overlay thickness. Similar to cutting forces and cutting temperature, the surface roughness also strongly rely on overlay thickness. The surface roughness when milling at fusion zone and HAZ are much worse than milling Inconel 182 overlays. As the overlay thickness increases from 0 to 4 mm, the surface roughness rapidly increases and reaches its maximum value at overlay thickness of about 4 mm. This rapid increase can be attributed to the gradually precipitated chromium carbides. How-ever, surface roughness decreases dramatically when the overlay thickness keep on increasing from 4 to 7 mm. The reason why it decreases lies in the dramatically disappeared small round Si carbides and Ti carbides in this region, with the overlay thickness increases from 7 to 10 mm. The influence of overlay thickness on surface roughness is no longer significant.

Figure 12.

Surface roughness Ra of milling Inconel 182 overlays at different overlay thickness.

Surface morphology after milling Inconel 182 overlays at critical overlay thickness of 1, 4, 7 and 10 mm is illustrated in Figure 13 to further investigate the machinability of Inconel 182 overlays. As shown in Figure 13(a) and (b), the machined surface at overlay thickness of 10 and 7 mm shows oblique, constant and slight curvy marks. Such curvy marks belong to normal milling curve marks because vibration generated between workpiece and milling cutter in this stable milling process is maintained or sustained, and therefore, the machined surface left constant uniform marks that are evenly spaced.²⁸ As the milling cutter periodically cut into the workpiece and then cut out, the regenerative vibration generates, and as a consequence, chatter marks left on the previous tool revolution as the cutting edge moved to the next revolution when machining Inconel 182 overlays with thickness of 4 mm, as shown in Figure 13(c). The best machined surface morphology with almost no curvy marks is obtained as expected when machining Inconel 182 overlays with overlay thickness of 1 mm as shown in Figure 13(d).

Figure 13.

Surface morphology after milling Inconel 182 overlays at (a) D = 10 mm, (b) D = 7 mm, (c) D = 4 mm and (d) D = 1 mm.

Prediction model for machining Inconel 182 overlays at different overlay thickness

Prior to establishing prediction model, the correlation between cutting force, micro-hardness, cutting temperature and surface roughness when machining Inconel 182 overlays at different overlay thickness should be discussed first. As listed in Table 3, it can be seen that the Ra is highly correlated to the micro-hardness and resultant cutting force F_xy, whose correlation coefficients are 0.907 and 0.8, respectively. The correlation coefficient between cutting force F_xy and micro-hardness is only 0.616, which proves that the cutting force F_xy is not only affected by micro-hardness but also affected by the microstructures discussed above. However, the correlation coefficient between cutting temperature and others is rather small, demonstrating that the cutting temperature is not correlated to others. Therefore, micro-hardness cutting force F_xy and cutting temperature as well as surface roughness should be used together to estimate the machinability of Inconel 182 overlays.

Table 3.

Correlation coefficient between micro-hardness H, F_xy, cutting temperature T and Ra.

	H	F_xy	T
F_xy	0.616
T	0.140	−0.153
Ra	0.907	0.800	0.134

After correlation coefficient calculation, the prediction model of the machinability of Inconel 182 overlays milling at different overlay thickness is proposed based on the multi-objective optimization. The micro-hardness, cutting forces, cutting temperature and surface roughness are the “target variables.” The most important work is to find the best fitting curves of these target variables, take cutting temperature, for example, as shown in Figure 14. The fitting curves employed contain Polynomial Fit, Logarithm Fit, Lorentz Fit and Exponential Fit; the results indicate that polynomial fitting can be the most effective way to fit the non-linear curves of cutting temperature versus the overlay thickness.

Figure 14.

Fitting function selection when milling Inconel 182 overlays at different overlay thickness.

Thus, the polynomial fitting curves of these target variables are employed to establish the prediction model for machining Inconel 182 overlays at different overlay thickness, which are expressed as follows

{\begin{matrix} H = 199.77 - 5.89 x + 6.68 x^{2} - 1.62 x^{3} + 0.15 x^{4} - 0.005 x^{5} \\ F x y = 226.74 - 22.76 x + 31.63 x^{2} - 10.92 x^{3} + 1.61 x^{4} - 0.11 x^{5} + 0.003 x^{6} \\ T = 289.14 - 129.67 x + 86.11 x^{2} - 23.93 x^{3} + 3.30 x^{4} - 0.22 x^{5} + 0.006 x^{6} \\ R a = 0.49 - 0.12 x + 0.12 x^{2} - 0.03 x^{3} + 0.003 x^{4} - 0.0001 x^{5} \end{matrix}

(1)

where x is the overlay thickness, H is the micro-hardness, F_xy is the cutting forces, T is the cutting temperature and Ra is the surface roughness.

The analysis of variance (ANOVA) of the polynomial fitting curves has been illustrated in Table 4, and its function is to check whether their fitting curve is significant or not. The ANOVA was carried out for a significant level of α = 0.05, that is, for a confidence level of 95%. The parameters with p value less than 0.05 are considered to have a statistically significant contribution to the responses. The adjusted R ² values of cutting force F_xy, cutting temperature, micro-hardness and surface roughness Ra are all larger than 0.9, indicating that the total variability can be explained by the fitting curves after considering the significance. The F values in ANOVA indicate that the fitting curves are all significant, and they can be used to predict or evaluate the machinability when milling Inconel 182 overlays at different overlay thickness “x.”

Table 4.

ANOVA table for fitting curves of F_xy, temperature T, micro-hardness H and Ra.

	DF	Sum of squares	Mean square	F	p	Adjusted R²
F_xy
Model	6	1131.844	188.641	104.939	0.000	0.98422
Error	4	7.190	1.798
Total	10	1139.035
T
Model	6	2852.731	475.455	168.525	0.000	0.99015
Error	4	11.285	2.821
Total	10	2864.016
H
Model	5	254.341	50.868	34.000	0.001	0.94286
Error	5	7.481	1.496
Total	10	261.822
Ra
Model	5	0.028	0.006	50.766	0.000	0.96136
Error	5	0.001	0.000
Total	10	0.028

ANOVA: analysis of variance; DF: degree of freedom.

The fitting curves of micro-hardness, cutting force F_xy, cutting temperature and surface roughness obtained according to equation (1) are illustrated in Figure 15. As the electrode diameter is 3.2 mm, the machinability of Inconel 182 overlays with the overlay thickness ranging from 0 to 10 mm can be divided into three regions: marked P, Q and R. Region P is the area with overlay thickness from 0 to 2.1 mm. Although the cutting force, cutting temperature and surface roughness vary with the increase in overlay thickness, machining Inconel 182 overlays in this region is acceptable.¹⁶ Region Q is the area with overlay thickness from about 2.1 to 6.4 mm. Due to the high cutting force and high surface roughness accompanied by chatter marks, machining Inconel 182 overlays in this region is rather difficult. Region R is the area with overlay thickness from about 6.4 to 10 mm, as there are almost no changes in cutting force, cutting temperature and surface roughness, machining Inconel 182 overlays in this region is stable.

Figure 15.

Fitting curves of micro-hardness, cutting force F_xy, cutting temperature and Ra.

Pareto graph is illustrated in Figure 16 to find out what are the most common variables that will induce the Inconel 182 overlays to be difficult to cut. The left vertical axis of the graph is the frequency of the occurrence of the Inconel 182 overlays to be difficult to cut. The right vertical axis is the cumulative frequency of the total number of occurrences. Four variables are included in this graph, they are micro-hardness, cutting forces, cutting temperature and the surface roughness. The critical points of these variables are determined according to the prediction model established in our work, as shown Figure 15. The curve amplitude variation of cutting temperature and micro-hardness is relatively small; thus, the cutting temperature of 256.9 °C and micro-hardness of 211.5 HV, corresponding x is approximately 6.4 mm (Figure 15), are selected as the critical points of difficult to cut. By contrast, the curve amplitude variation of cutting forces and surface roughness is relatively large; thus, the cutting force of 236.2 N and Ra of 0.548 µm, corresponding x is 6.4 and 2.1 mm, respectively (Figure 15), are selected as the critical points of difficult to cut.

Figure 16.

Difficult to cut Pareto graph of milling Inconel 182 overlays.

When judging the Inconel 182 overlays at the different overlay thickness is difficult to cut or not, the results of Pareto analysis indicate that surface roughness is the dominate variable, followed by cutting force, and the cutting force and the micro-hardness are the minor variables.

Conclusion

This article investigates the microstructures and evaluates the machinability of Inconel 182 overlays at different overlay thickness. The conclusions are drawn as follows:

The optimal combination of cutting parameters with respect to surface roughness Ra when side milling Inconel 182 overlays is cutting speed Vc = 160 m/min, feed rate f = 0.2 mm/tooth and radial depth of cut a_w = 1 mm.

The microstructures of Inconel 182 overlays are determined by electrode diameter (3.2 mm) and overlay thickness. When the overlay thickness is smaller than the electrode diameter of 3.2 mm, the Inconel 182 overlays has undergone only one welding process. Some chromium in Inconel 182 grain boundary diffuses into the melted zone of substrate to induce chromium depletion, and in turn the carbon from the substrates will diffuse into Inconel 182 overlays to form chromium carbide hard particles with the remained chromium element and precipitate at grain boundaries during the cooling phase after solidification. When the overlay thickness is larger than the electrode diameter of 3.2 mm, the weld metal and substrate are both Inconel 182. At this secondary similar metal welding process, the chromium carbide precipitation due to difference in melting temperature between base metal and weld metal will reduce gradually. When the overlay thickness is larger than twice the diameter of the electrode (6.4 mm), the Inconel 182 overlays has undergone the welding process for the third time. At this moment, no Si and C from G17CrMo9-10 can diffuse into weld metal to form Si and Ti carbides. Instead, only chromium carbide precipitates.

The prediction model illustrates that when the machinability of Inconel 182 overlays is acceptable at overlay thickness from about 0 to 2.1 mm; it is poor at overlay thickness from about 2.1 to 6.4 mm, and it is stable at overlay thickness from about 6.4 to 10 mm.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work was supported by National Natural Science Foundation of China (No. 51405294) and Postdoctoral Research Foundation of Shanghai Jiao Tong University (14X100030001).

References

Poullikkas

. An overview of current and future sustainable gas turbine technologies. Renew Sust Energ Rev 2005; 9: 409–443.

Min

Binshi

Jiaying

. Experimental observations on surface roughness, chip morphology, and tool wear behavior in machining Fe-based amorphous alloy overlay for remanufacture. Int J Adv Manuf Tech 2013; 67: 1537–1548.

Choudhury

El-Baradie

. Machinability of nickel-base super alloys: a general review. J Mater Process Tech 1998; 77: 278–284.

Soo

Hood

Aspinwall

. Machinability and surface integrity of RR1000 nickel based superalloy. CIRP Ann: Manuf Techn 2011; 60: 89–92.

Ezugwu

. Key improvements in the machining of difficult-to-cut aerospace superalloys. Int J Mach Tool Manu 2005; 45: 1353–1367.

Sharman

ARC

Hughes

Ridgway

. Surface integrity and tool life when turning Inconel 718 using ultra-high pressure and flood coolant systems. Proc IMechE, Part B: J Engineering Manufacture 2008; 222: 653–664.

Ulutan

Ozel

. Machining induced surface integrity in titanium and nickel alloys: a review. Int J Mach Tool Manu 2011; 51: 250–280.

Aspinwall

Soo

Curtis

. Profiled superabrasive grinding wheels for the machining of a nickel based superalloy. CIRP Ann: Manuf Techn 2007; 56: 335–338.

Choudhury

El-Baradie

. Machining nickel base superalloys: Inconel 718. Proc IMechE, Part B: J Engineering Manufacture 1998; 212: 195–206.

10.

Sireesha

Shankar

Albert

. Microstructural features of dissimilar welds between 316LN austenitic stainless steel and alloy 800. Mat Sci Eng A 2000; 292: 74–82.

11.

Bao

Shinozaki

Iguro

. Stress corrosion cracking sealing in overlaying of Inconel 182 by laser surface melting. J Mater Process Tech 2006; 173: 330–336.

12.

Dehmolaei

Shamanian

Kermanpur

. Microstructural characterization of dissimilar welds between alloy 800 and HP heat-resistant steel. Mater Charact 2008; 59: 1447–1454.

13.

Peng

Yamauchi

Shoji

. Investigation of dendrite-boundary microchemistry in alloy 182 using auger electron spectroscopy analysis. Metall Mater Trans A 2003; 34: 1891–1899.

14.

Brozek

. The turning of overlays using sintered carbide tools. Int J Adv Manuf Tech 2009; 40: 438–446.

15.

Brozek

. Cutting conditions optimization when turning overlays. J Mater Process Tech 2005; 168: 488–495.

16.

Wang

Dong

. Experimental investigations of cutting parameters influence on cutting forces in turning of Fe-based amorphous overlay for remanufacture. Int J Adv Manuf Tech 2013; 65: 735–743.

17.

Bao

Yamamoto

Shinozaki

. Precipitation and Cr depletion profiles of Inconel 182 during heat treatments and laser surface melting. J Mater Process Tech 2009; 209: 416–425.

18.

Jang

Lee

Sung Kim

. Mechanical property variation within Inconel 82/182 dissimilar metal weld between low alloy steel and 316 stainless steel. Int J Pres Ves Pip 2008; 85: 635–646.

19.

Lim

Kim

Cho

. Microscopic examination of an alloy 600/182 weld. Mater Charact 2009; 60: 1496–1506.

20.

Wang

. Experimental investigation on hard milling of high strength steel using PVD-AlTiN coated cemented carbide tool. Int J Refract Met H 2014; 43: 94–101.

21.

Kaymakci

Kilic

Altintas

. Unified cutting force model for turning, boring, drilling and milling operations. Int J Mach Tool Manu 2012; 54–55: 34–45.

22.

Wang

. Assessment of cutting forces in high-speed milling of Inconel 718 considering the dynamic effects. Proc IMechE, Part B: J Engineering Manufacture 2013; 227: 1581–1595.

23.

Alauddin

Mazid

El Baradi

. Cutting forces in the end milling of Inconel 718. J Mater Process Tech 1998; 77: 153–159.

24.

Zeng

Chen

. An experimental study of tool wear and cutting force variation in the end milling of Inconel 718 with coated carbide inserts. J Mater Process Tech 2006; 180: 296–304.

25.

Marinescu

Axinte

. A critical analysis of effectiveness of acoustic emission signals to detect tool and workpiece malfunctions in milling operations. Int J Mach Tool Manu 2008; 48: 1148–1160.

26.

Wang

C-D

Chen

Q-L

. Tool wear performance in face milling Inconel 182 using minimum quantity lubrication with different nozzle positions. Int J Precis Eng Man 2014; 15: 557–565.

27.

Yan

Zhu

Zhuang

. Modeling and analysis of coated tool temperature variation in dry milling of Inconel 718 turbine blade considering flank wear effect. J Mater Process Tech 2014; 214: 2985–3001.

28.

Toh

. Vibration analysis in high speed rough and finish milling hardened steel. J Sound Vib 2004; 278: 101–115.