Study on the Effect of Milling Surface Plastic Deformation on Fatigue Performance of 20Cr and TC17 Specimens

Abstract

In addition to the micro-topography of machined surfaces, plastic deformation is an important factor affecting the machined surface integrity and the fatigue performance of specimens. For 20Cr and TC17 materials, the effect law of milling surface plastic deformation on high- and low-cycle fatigue performance of specimens was studied. Experimental results show that the effect law of milling surface plastic deformation on high- and low-cycle fatigue performance of specimens is different. When the high-cycle fatigue life is about 2 × 10⁵ cycles, severe surface plastic deformation can increase the maximum stress suffered by TC17 specimens from 1060 MPa to 1080 MPa; when the low-cycle fatigue life is about 6 × 10⁴ cycles, severe surface plastic deformation can reduce the maximum stress suffered by 20Cr specimens from 680 MPa to 660 MPa. Therefore, severe surface plastic deformation can improve the high-cycle fatigue performance and, on the contrary, it has a negative effect on the low-cycle fatigue performance. Combined with the analysis of the surface integrity index and the fatigue fracture topography of specimens, it is concluded that the machined surface plastic deformation significantly reduces the plastic deformation capacity of surface layer material and enhances the stress concentration phenomenon caused by the surface micro-topography, so that when a specimen is subjected to a large load, it is very easy to make microcracks appear on the machined surface and form the “over-plastic deformation” phenomenon, resulting in a sharp decrease in the low-cycle fatigue performance of specimens. This suggests that the plastic deformation degree of machined surfaces needs to be optimized depending on the magnitude of the working load and the micro-topography.

1. Introduction

Besides the material properties and structures of specimens, machined surface integrity has a great effect on the fatigue performance. Surface micro-topography, one of the most basic indicators of surface integrity, can form the stress concentration phenomenon on a machined surface, thus influencing the fatigue performance of specimens [1]. Generally, the degree of stress concentration on a machined surface can be expressed by stress concentration factor K_t, as follows [2].

$$K_{\mathrm{t}}=1+2\sqrt{\lambda \frac{R_z}{\rho }},$$

(1)

where λ refers to the ratio between the spacing and depth of asperities, R_z is the 10-point surface roughness, and ρ is the effective profile valley radius of surface texture.

According to the basic definition of K_t, many scholars have studied the effect mechanism of K_t on the fatigue behavior of specimens. The research results of Arola and Williams [3] showed that the high-cycle fatigue life of specimens decreased with the increase of surface roughness. Ås et al. [4] studied the relationship between the surface crack initiation location and the surface stress concentration by combining with specific test methods. These results show that the machined surface stress concentration formed by surface micro-topography has a great effect on the fatigue performance of specimens, and the more severe surface microscopic stress concentration, the greater the stress concentration factor K_t, and the worse the fatigue performance of specimens.

Nevertheless, some studies show that when the microscopic stress concentration phenomenon formed by surface micro-topography is controlled within a certain range, surface hardening, surface residual compressive stress and surface material structure refinement formed by plastic deformation on the machined surface can effectively inhibit the initiation of surface cracks, thus positively influencing the fatigue performance of specimens. Li [5] studied the comparison between conventional cutting and ultrasonic vibration cutting for Ti-6Al-4V, and the fatigue life of specimens machined by ultrasonic vibration cutting is 11.4 times of that by conventional cutting. Pramanik et al. [6] further found that even if the surface roughness of specimens is high, the fatigue performance can still be enhanced by the residual compressive stress caused by machined surface plastic deformation. Therefore, in order to pursue better fatigue performance, the surface strengthening methods, such as shot peening, rolling or optimization of the machining parameters, are often utilized, so that a plastic deformation layer is formed on the machined surface of parts.

On the other hand, the experimental results of some studies show that plastic deformation has different effects on the high- and low-cycle fatigue of specimens. Villegas [7] and Shaw [8] studied the effect of shot peening on the fatigue performance of specimens. The results showed that severe plastic deformation can significantly improve the high-cycle fatigue performance and reduce the low-cycle fatigue performance of specimens. Ding [9] and Kim [10] obtained approximate results that the fatigue load exceeds the threshold stress of specimens, the surface residual compressive stress is released, accelerating the crack initiation and propagation process and decreasing the fatigue performance of specimens. Klotz [11] concluded that using a large shot peening strength to obtain large surface plastic deformation will lead to an increase in surface roughness, resulting in a decrease in the low-cycle fatigue performance of specimens.

As can be seen from the above achievements, most of these studies attribute the decreased low-cycle fatigue performance of specimens to the redistribution of surface residual stress and large surface roughness. However, these studies are not comprehensive and complete enough. The effect mechanism of surface plastic deformation on the fatigue performance of specimens with controlled surface micro-topography and cutting temperature is not clear.

In this study, a typical elastic-plastic metal material 20Cr and a typical material TC17 for aerospace manufacturing were used. Based on the above previous research results, this paper referred to relevant studies on the optimization of milling parameters [12,13], and the effect of milling parameters on the degree of machined surface plastic deformation [14,15], different milling machining conditions were used to form different surface plastic deformation of 20Cr and TC17 specimens. The high- and low-cycle fatigue performances were tested on the premise that the surface micro-topography and surface residual stress were basically the same. Moreover, the experimental results were analyzed in-depth to study the effect and mechanism of machined surface plastic deformation on the high- and low-cycle fatigue performances of specimens, which provides an important reference for the optimization of the machined surface plastic deformation in practical engineering.

This paper is organized as follows: Section 2 describes the materials, specimens, measurement methods and equipment for surface integrity used in the study. Section 3 presents the measurement results of surface integrity, the test methods and results of fatigue performance, and fracture morphology. The test results are discussed in Section 4 and are summarized in Section 5.

2. Experiments

2.1. Materials and Specimens

The experimental materials were 20Cr and TC17. As per GB/T 228.1–2010, the main mechanical properties at room temperature were tested on standard specimens by a material tensile testing machine (Instron 8801, Boston, MA, USA) at a tensile rate of 2 mm/min, as shown in

Table 1. Mechanical properties of experimental materials.

Material	Young’s Modulus E (GPa)	Yield Strength σs (MPa)	Tensile Strength σb (MPa)
20Cr	192	473	897
TC17	110	1158	1324

.

Table 2. Chemical composition of 20Cr (wt.%).

C	Cr	Mn	Fe
0.2	1.21	1.01	Bal.

and

Table 3. Chemical composition of TC17 (wt.%).

Al	Cr	Zr	Mo	Sn	Fe	C
4.9	4	1.93	4.18	2.16	0.03	0.01

list the materials’ chemical composition.

TiAlN coated carbide end milling cutters were used for side and down milling of specimens. Water-based cutting fluid was used for cooling and lubrication. The typical dimensions and mounting of specimens are shown in Figure 1. The length of the gauge section was 6 mm, the diameter after milling was 5 mm, and the specimens had a 60-sided regular polygon cross-section [16]. Characteristics of cutters are listed in

Table 4. Cutter conditions.

Substrate Material	Cemented Carbide
Coating	TiAlN
Number of flutes	4
Diameter (mm)	10
Helix angle (°)	45
Photograph

and the specific experimental parameters are shown in

Table 5. Experimental parameters.

Experiment No.	Material	Cutting Speed vs (m/min)	Feed per Tooth fz (mm/z)	Depth of Cut ae (mm)
Exp. 1	20Cr	20	0.06	0.1
Exp. 2		20	0.06	0.7
Exp. 3	TC17	80	0.18	0.2
Exp. 4		30	0.18	0.7

.

2.2. Surface Integrity Measurement

Measurements of surface integrity indexes were performed on milled specimens, and each value was tested five times and averaged. Surface roughnesses were measured by a stylus-type surface roughness measuring instrument (TIME 3220, Time Group Inc., Beijing, China). The sampling length was set to 0.8 mm, the evaluation length was set to 4 mm, and the measurement direction was parallel to the feed direction. R_a and R_z were used to evaluate the microscopic topography of the machined surface. Surface residual stresses along the axial direction were measured using X-ray diffraction, which is a simple and efficient method, and an X-ray diffractometer (AutoMATE II, Nippon Rigaku Co., Tokyo, Japan) was used. For 20Cr specimens, the diffraction angle was 156.4°, the exposure time was 20 s, and the aperture diameter was 2 mm. For TC17 specimens, the diffraction angle was 140°, the exposure time was 80 s, and the aperture diameter was 1 mm. Surface hardness measurements were performed on a nano-indentation tester (TTX-NHT3, Anton Paar, Graz, Austria). The maximum load used was 100 mN, corresponding to a retention time of 30 s for the maximum load and the loading and unloading rates were both 10 mN/s. Cut sections were tested in the same way to obtain the hardness of substrate materials, and cut sections were ground and polished. Micro-hardness HVIT of 20Cr and TC17 substrate materials is about 273 and 513 HV, separately. Microstructures of cut sections were observed with a Cam-Scan CS3400 (Cambridge shire, UK) scanning electron microscope (SEM).

3. Results

3.1. Surface Integrity Test Results

The measurement results of surface roughness, surface hardness and surface residual stress of 20Cr and TC17 specimens are shown in

Table 6. Results of surface integrity measurements.

Experiment No.	Material	Roughness Ra (μm)	Roughness Rz (μm)	Nano-Hardness HIT (GPa)	Micro-Hardness HVIT (HV)	Residual Stress σH (MPa)
Exp. 1	20Cr	0.32	1.59	4.06	375.54	−157.1
Exp. 2		0.34	1.74	4.79	443.48	−159.0
Exp. 3	TC17	0.22	0.94	5.71	529.19	−96.9
Exp. 4		0.22	0.98	6.21	574.71	−125.0

.

As shown in Table 6, for 20Cr specimens under two milling conditions (Exp. 1 and Exp. 2), the surface roughness and surface residual stress are almost the same when the milling depth a_e is changed. In contrast, the surface nano-hardnesses and surface micro-hardnesses of Exp. 2 specimens are significantly greater than those of Exp. 1 specimens, which is because the larger depth of cut (a_e = 0.7 mm) of Exp. 2 specimens compared to the 0.1 mm depth of cut of Exp. 1 specimens causes a significant extrusion effect on the surface, making cutting force increases, thus intensifying the hardening of the surface layer material.

A comparative analysis of Exp. 3 and Exp. 4 shows that the degree of surface plastic deformation of specimens increases with the decrease of milling speed and the increase of milling depth, which is manifested by the increase of surface hardness and surface residual compressive stress, as shown in Table 3. For TC17 specimens, the milling parameters of Exp. 3 and Exp. 4 do not significantly affect the surface roughness of specimens.

In terms of the machining mechanism, the surface layer material will inevitably generate plastic deformation during the milling process, which directly impacts its mechanical properties. Therefore, compared to the substrate material, the material properties of the machined surface layer will have a significantly higher yield strength and a significantly reduced range of plastic deformation available [17], as shown in Figure 2. Fatigue cracks generally occur on machined surfaces or subsurfaces, and high yield strength and plastic deformation on the surface can effectively inhibit fatigue fracture initiation and improve the fatigue performance of specimens [18].

Meanwhile, the surface microstructure of 20Cr and TC17 specimens with different machining parameters were examined by SEM, as shown in Figure 3. Due to the shear strain caused by the mutual extrusion between the tool and the specimen surface, fibrous deformation of microstructures along the tool feed direction can be observed of all Exp. 1–4 specimens. Additionally, the type of structure of the machined surface does not change due to the reasonable choice of milling parameters [19]. Exp. 2 or Exp. 4 specimens have more severe surface plastic deformation than Exp. 1 or Exp. 3 specimens, respectively, which is consistent with the above analysis of the surface integrity.

In order to obtain the yield strength of the surface material, a nano-indentation tester (TTX-NHT3, Anton Paar, Graz, Austria) was used to examine the nano-indentation hardness HIT, and the yield strength σ_s can be obtained from the empirical formula HIT = aσ_s + b [20]. Yield strength of the substrate material in Table 1 was used for the base validation calculation, where a = 5.23 and b = 5.63 GPa for 20Cr specimens and a = 4.46 and b = 4.24 GPa for TC17 specimens. The calculated results are shown in

Table 7. Calculated yield strength of 20Cr and TC17 specimens.

Materials	Test Location	Nano-Hardness HIT (GPa)	Yield Strength σs (MPa)
20Cr	Substrate material	2.95	456
	Surface material of Exp. 1	4.06	669
	Surface material of Exp. 2	4.79	808
TC17	Substrate material	5.54	1147
	Surface material of Exp. 3	5.71	1186
	Surface material of Exp. 4	6.21	1296

, corresponding to the actual properties. It can be seen that, as the degree of machined surface plastic deformation of specimens increases, so does the yield strength of the surface material.

3.2. Fatigue Performance Test and Fracture Analysis of Specimens

The fatigue performance of 20Cr and TC17 specimens was tested on a high-frequency digital fatigue testing machine (GPS-100, Sino test Equipment Co., Ltd., Changchun, China). In order to study the effect of milling surface plastic deformation on the high- and low-cycle fatigue performance of specimens, respectively, the maximum stresses were 600 MPa, 640 MPa, 660 MPa and 680 MPa for 20Cr specimens and 1060 MPa, 1080 MPa, 1090 MPa, 1100 MPa, 1120 MPa and 1140 MPa for TC17 specimens. The cyclic stress ratio R is 0.1 and the test frequency is about 110 Hz.

3.2.1. 20Cr Specimens

From the number of cycles to failure N_f of 20Cr specimens, as shown in Figure 4, the effect of surface plastic deformation on the fatigue performance under different load conditions is entirely different. N_f of Exp. 2 specimens is greater than that of Exp. 1 specimens when σ_max is less than 630 MPa; when σ_max is greater than 630 MPa, the results are just the opposite. So, there is a crossover point at the threshold stress σ_max = 630 MPa in σ_max—N_f curves of 20Cr specimens.

The fatigue fracture surface of 20Cr specimens was observed by SEM, as shown in

Table 8. Fracture topography and microscopic surface of 20Cr specimens.

Load Conditions σmax/MPa	Exp. No.	Cutting Conditions	Fracture Topography and Surface Microcrack
600	Exp. 1	vs = 20 m/min fz = 0.06 mm/z ae = 0.1 mm
	Exp. 2	vs = 20 m/min fz = 0.06 mm/z ae = 0.7 mm
660	Exp. 1	vs = 20 m/min fz = 0.06 mm/z ae = 0.1 mm
	Exp. 2	vs = 20 m/min fz = 0.06 mm/z ae = 0.7 mm
680	Exp. 1	vs = 20 m/min fz = 0.06 mm/z ae = 0.1 mm
	Exp. 2	vs = 20 m/min fz = 0.06 mm/z ae = 0.7 mm

. When σ_max is less than the threshold stress (630 MPa), i.e., when specimens are subjected to small alternating loads, there is no microcrack on the surfaces of specimens under two milling conditions and the fatigue source of the fracture surface is obvious and clear. Besides, the fatigue source initiation of the Exp. 2 specimen occurs on the subsurface. Combined with the surface integrity of Exp. 1 and Exp. 2 specimens, the surface plastic deformation of Exp. 2 specimens is more severe than that of Exp. 1 specimens, and the surface roughness and surface residual stress are almost same. Therefore, the relatively severe surface plastic deformation is the main reason for the increase of N_f.

When σ_max is greater than the threshold stress (630 MPa), microcracks appear on the surface of both specimens, and the fatigue sources are not obvious. Moreover, with the increase of alternating load, the surface of Exp. 2 specimens with severe plastic deformation is more prone to microcracks, indicating that plastic deformation has a negative effect on the fatigue performance of specimens.

3.2.2. TC17 Specimens

The fatigue life of TC17 specimens is shown in Figure 5, the variation trend of N_f with the plastic deformation of the machined surface is consistent with that of Figure 4. There is also a crossover point σ_max =1120 MPa in σ_max—N_f curves. When σ_max is less than 1120 MPa, N_f of Exp. 4 specimens is greater than that of Exp. 3 specimens, and when σ_max is greater than 1120 MPa, the result is the opposite.

The fatigue fracture topography and the microscopic surface of TC17 specimens are shown in

Table 9. Fracture topography and microscopic surface of TC17 specimens.

Load Conditions σmax/MPa	Exp. No.	Cutting Conditions	Fracture Topography and Surface Microcrack
1060	Exp. 3	vs = 80 m/min fz = 0.18 mm/z ae = 0.2 mm
	Exp. 4	vs = 30 m/min fz = 0.18 mm/z ae = 0.7 mm
1090	Exp. 3	vs = 80 m/min fz = 0.18 mm/z ae = 0.2 mm
	Exp. 4	vs = 30 m/min fz = 0.18 mm/z ae = 0.7 mm
1140	Exp. 3	vs = 80 m/min fz = 0.18 mm/z ae = 0.2 mm
	Exp. 4	vs = 30 m/min fz = 0.18 mm/z ae = 0.7 mm

. When specimens are subjected to small alternating load (1060 MPa), the fatigue source of the Exp. 4 specimen with severe plastic deformation occurs at a deeper subsurface layer than the Exp. 3 specimen, indicating that plastic deformation has a positive effect on fatigue performance. However, microcracks already appear on the surface of TC17 specimens at σ_max = 1090 MPa.

4. Discussion

From the above fatigue performance of 20Cr and TC17 specimens, it can be seen that, regardless of the mechanical properties of materials, when the specimen is subjected to a small alternating load, the machined surface plastic deformation has a positive effect on the fatigue performance, and the greater the degree of machined surface plastic deformation, the higher the fatigue life of the specimen; when the specimen is subjected to an alternating load greater than the threshold stress, the greater the degree of machined surface plastic deformation, the lower the fatigue life of the specimen. The experimental phenomenon is also prevalent in the results of other researchers [7,8,9,10,21]. Combining the above experimental results, N_f corresponding to the threshold stress is generally between 3 × 10⁴ and 5 × 10⁵, regardless of the material used and the processing method. Therefore, it can also be concluded that the machined surface plastic deformation of specimens has a positive effect on the high-cycle fatigue performance, while on the contrary, it has a negative effect on the low-cycle fatigue performance of specimens.

Stress concentration caused by the surface topography inevitably appears on the machined surface when the specimen is subjected to axial loads, as shown in Figure 6a. This stress concentration due to the deformation at the bottom of the circular notch of the surface cutting topography is a strain concentration essentially. As shown in Figure 6b, the actual strain ε_a at the bottom of the circular notch is larger than the nominal strain ε_n of the specimen under the effect of strain concentration. Especially when the specimen is subjected to a large applied load, the actual strain ε_a at the strain concentration on the surface will exceed the ultimate tensile strain ε_b of the specimen material, resulting in microcracks on the surface and a sharp decrease in the fatigue performance. However, the surface micro-topography of 20Cr specimens in both Exp. 1 and Exp. 2 is basically the same at the macroscopic scale. Therefore, the degree of stress (strain) concentration on both surfaces does not differ significantly under the same load. Meanwhile, although there is a redistribution of surface residual stress under large loads, the values of surface residual stress for Exp. 1 and Exp. 2 specimens are also essentially the same.

Actually, due to the presence of the machined surface plastic deformation, the stress (strain) concentration is generated as shown in Figure 6c,d. The Exp. 2 specimens with a more severe degree of plastic deformation have the surface material with a higher yield strength and a smaller range of plastic deformation available. Therefore, only smaller applied loads are required to cause microcracks on the surface of Exp. 2 specimens compared to Exp. 1 specimens, as shown by the difference between the two groups of fatigue fractures in Table 5 when σ_max is 660 MPa.

For the test results of TC17 specimens, the change law is also basically the same as that of 20Cr specimens. However, the maximum stress σ_max (1060–1140 MPa) used in the fatigue tests of TC17 specimens did not reach their yield strength (1158 MPa), and microcracks already appear on the surface at a maximum stress σ_max of 1090 MPa. Combining the strength data of TC17 specimen surface material in Table 7 and Figure 6d, it can be concluded that when σ_max is 1090 MPa, the actual strain ε_a at the strain concentration on the surface of the specimens has already exceeded the ultimate tensile strain ε_b′ of the surface material, thus causing microcracks to be generated on the surface.

For Exp. 1–2 or Exp. 3–4, the machined surface plastic deformation has a positive effect on the fatigue performance of specimens under small loads, and the greater the degree of machined surface plastic deformation under the same load, the better the high-cycle fatigue performance. However, the greater the degree of machined surface plastic deformation under large loads, the easier it is to generate microcracks on the surface of specimens, resulting in a sharp decrease in fatigue performance. Specifically, when ε_n increases or the degree of machined surface plastic deformation increases, it is possible that ε_a exceeds ε_b′, resulting in microcracks forming on the machined surface. This phenomenon, called the “over-plastic deformation” phenomenon of the machined surface, needs to be avoided.

5. Conclusions

Under the condition that surface micro-topography and surface temperature were controlled, the law and mechanism of the effect of milling surface plastic deformation on the high- and low-cycle fatigue life of 20Cr and TC17 specimens were studied, and the following conclusions were obtained:

(1) The plastic deformation on a machined surface has different effects on the high- and low-cycle fatigue performances of specimens. When the specimen is subjected to a small alternating load, severe surface plastic deformation can improve the high-cycle fatigue performance; when the specimen is subjected to a large alternating load, severe surface plastic deformation has a negative effect on the low-cycle fatigue performance. Within the experimental conditions and parameters, for TC17, the increase in the degree of surface plastic deformation can improve the high-cycle average fatigue life of specimens by up to 2.43 times; for 20Cr, the increase in the degree of surface plastic deformation reduces the low-cycle average fatigue life of specimens by up to 63.63% of the previous one;

(2) Cutting results in severed deformation on the surface of the specimen, which reduces the plasticity of the surface layer material and enhances the surface stress concentration phenomenon induced by surface micro-topography. So when the specimen is subjected to a large load, microcracks easily appear on the machined surface, and the “over-plastic deformation” phenomenon is formed, which leads to a sharp decrease in the low-cycle fatigue performance of specimens;

(3) For improving the fatigue of specimens, the plastic deformation on machined surface needs to be optimized. As the load and surface roughness rises, the degree of machined surface plastic deformation should be appropriately decreased; conversely, the degree of machined surface plastic deformation should be increased.