Improvement of current-carrying tribological behaviour via laser surface texture

Current-carrying tribometer

Current-carrying tribological tests are conducted in a customized tribometer, which is a kind of ball-on-flat contact mode, as shown in Figure 1. A ball sample is mounted on the bottom of the ball fixture, which can move along horizontal and vertical directions. A flat sample is mounted on the lower sample table. A 3-D accelerator (Donghua 1A342E) is used to detect the vibration responses generated by the surface. Four-wires method is used to test the variation of contact resistance. The tribometer can simultaneously collect and analyse the contact force, voltage and vibration signals generated during friction process. More information about the tribometer can be seen in [19].

OPEN IN VIEWER

Figure 1

Image of test set-up (a) and the schematic diagram of test set-up (b): (1) Moving stage, (2) Force sensor, (3) Linear spring, (4) Cage, (5) Locking device, (6) Ball holder, (7) 3-D accelerator, (8) Ball sample, (9) Flat sample, (10) Reciprocating motion device, (11) Fixed table.

Sample preparation and test condition

Brass (H65) is used as the rub material in the tests, considering that the H65 is widely used in sliding electrical contact devices, such as terminals, circuit breakers, conductive rings, and some military electromagnetic equipment. The material parameters and sizes of ball and flat samples are listed in Table 1. Different mesh sandpapers are respectively grounded the flat samples to a roughness of 0.4 μm R_a. Three kinds of LST are fabricated on the flat sample surfaces, as shown in Figure 2. The laser process parameters are wavelength = 1064 nm, power = 10 W, pulse frequency = 10 kHz, scan speed = 5 mm/s. Before testing, both the ball and flat samples are cleaned in acetone (5 min) and ethanol (5 min) by using ultrasonic cleaning machine. The tests are performed under strictly controlled conditions, as shown in Table 2. The experimental tests are repeated at least three times in each group to ensure the reproducibility of the results.

OPEN IN VIEWER

Figure 2

The morphologies of the round RPTS (a), SPTS (b) and GTS (c) measured by SEM.

OPEN IN VIEWER

Table 1

Material parameters and sizes of ball and flat samples.

Sample	Density	Elasticity modulus	Hardness	Size
Ball	8500 kg/m3	90 GPa∼100 GPa	90 HV∼100 HV	Diameter = 4 mm
Flat				Length × width ×height = 25 mm × 25 mm × 4 mm

OPEN IN VIEWER

Table 2

Test conditions of the experiments.

Condition	Normal load	Sliding displacement	Period	Test time	Current
Values	2 N	3 mm	3 s	1800 s	3 A

Coefficient of friction (COF)

The COF curves of different surfaces are illustrated in Figure 3. For both the smooth surfaces and SPTS, the COFs gradually increase with the time and reach about 0.65 at about 800 s. The reason is lie in that at the beginning of the test, due to the lubricating effect of oxide film on the surface, the COF is relatively low. With the destruction of the lubricant film and the direct contact between the substrate metals, the COF starts to increase. As the intensification of the surface wear level and the variation of dynamic behaviours of wear debris, such as the dispersion, agglomeration and comminution, the COF fluctuates significantly and reaches about 0.7 after the test. For RPTS, although the COF decreases to some extent compared with the above mentioned two surfaces, it still remains at a large value and reaches about 0.6 at the end of the test. In contrast, the COF of GTS keeps at a low value (less than 0.2) during the whole test. In addition, the fluctuation frequency of COF of GTS is visible lower than that of other surfaces. Therefore, in the current-carrying state, the GTS possesses the best potential in reducing the COF among all the textured surfaces.

OPEN IN VIEWER

Figure 3

The variation of COF of the three surfaces (a) and the average SDs of COFs of all the surfaces (b).

Considering that the stability of the COF has a crucial influence on the performances and service life of contact system, the standard deviations (SD) of the COF are calculated to evaluate its stability [20], as shown in Figure 3(b). Visibly, the SD of COF of the smooth surface is 0.225, and the SD of COF of SPTS is in the similar level, which is 0.211. By a contrast, the SD of COF of RPTS is greatly reduced to 0.17, indicating superior friction reduction. While for the GTS, the SD values of COF is further reduced to about 0.05, suggesting that the GTS has the best ability in reducing COF and contributes to maintain the contact stability.

Friction force in different stages

Figure 4 exhibits the evolution characteristics of the friction forces signals for all the surfaces at different stages. In the initial stage (100 s∼107 s), the friction forces curves of both smooth surface and RPTS fluctuate strongly and the corresponding values are relatively large, which correspond to the higher COF of the two surfaces at this stage (Figure 4(a)). In the middle stage (1100 s∼1107 s), the smooth surface, SPTS and RPTS all exhibit continuous oscillation with high amplitude. In contrast, the friction force value measured from GTS is very small, and nine visible sawtooth waves can be observed clearly. It is speculated that when the ball sample slides through the groove, the changes of the contact state will result in a transient sudden change of friction force (Figure 4(b)). Furthermore, in the late stage (1700 s∼1706 s), the fluctuations of the friction signals from smooth surface, SPTS and RPTS further increase, while the friction force signal of GTS remains at a low level, and no visible and significant high-amplitude fluctuation can be observed. Noteworthy, nine visible sawtooth waves can be still seen in the signal curve (Figure 4(c)).

OPEN IN VIEWER

Figure 4

The friction forces of the different surfaces in 100 s∼107 s (a), 1100 s∼1107 s (b), 1700 s∼1706 s (c) and contact states when the ball slides across the GTS surface (d).

Figure 4(d) illustrates the dynamic characteristics of the ball sample slides through the GTS in the middle stage of the test. It is seen that when the ball slides across the groove area, the friction force decreases rapidly. Once the ball sample is in contact with the groove edge, the impact behaviour occurs between the ball and groove will cause the rapidly increase of friction force, which results in the formation of friction force signal with sawtooth waves. With regard to the sliding amplitude of 3 mm, i.e. nine grooves are on the GTS in this sliding distance, thus the friction force signal fluctuates nine times in a single stroke.

Vibration acceleration analysis

Considering that FIV has an extremely negative effect on the stability of contact system, thus the vibration acceleration signals are extracted, as illustrated in Figure 5. During 1200 s∼1206 s, the RPTS exhibits continuous high-amplitude FIV phenomenon, and the vibration is presented in every stroke of reciprocating motion. In contrast, the vibration signals from smooth surface, SPTS and GTS are quite weak. Combined with the analysis result of friction force shown in Figure 4, it is seen that the generation of FIV from the RPTS leads to the generation of obvious high amplitude oscillation of friction force signal. In the late stage of 1700 s∼1706 s, the RPTS still generates continuous high-intensity FIV, while the smooth surface starts to generate FIV in one single stroke as well. In contrast, the vibration signals detected from both the SPTS and GTS are still very weak, especially for the GTS, which has the lowest level of vibration intensity among all the surfaces, thus the friction system with GTS remains stable during the test.

OPEN IN VIEWER

Figure 5

Vibration acceleration signals in the middle (a) and late (b) stage.

Noteworthy, many studies have verified that FIV is easy to be triggered at a larger COF, and the intensity and tendency of FIV will become stronger with the increase of the COF [21,22]. However, in this study, although the COF of SPTS is relatively large, no visible FIV phenomenon is observed. Therefore, there seems no direct relation between the COF and the FIV in the state of sliding electrical contact. It can be speculated that this is due to the special contact state caused by the textured surface. In addition, due to the presence of current passing through the interface, the wear property will be varied, which will strongly affect the performance of FIV.

Contact voltage

Figure 6 gives the contact voltages as a function of time and the corresponding SD values. It is found that the contact voltages of different surfaces fluctuate to different degrees. For the smooth surface, the voltage has a small value of fluctuation, and the corresponding SD value is about 0.095 V. This is due to that the contact state changes with the interface wear in real time, thus the voltage signal occurs instantaneous mutation. While for the RPTS, the fluctuation intensity of the contact voltage is further increased. Visibly, the maximum contact voltage can reach 4 V, and the corresponding SD value is of 0.302 V. This is due to the strong FIV generated from the RPTS, which leads to more visible variation in the microscopic contact characteristics of the contact area. In contrast, the contact voltages measured from SPTS and GTS do not show significantly fluctuation during the whole test process, and the corresponding SD values are 0.0546 and 0.0540 V, respectively, indicating that both of the SPTS and GTS have good electrical conductivity, which are conducive to maintaining the contact stability.

OPEN IN VIEWER

Figure 6

The contact voltages (a) and the corresponding SD values (b).

Noteworthy, although the electric voltage signal of the RPTS has a certain degree of fluctuation, no instantaneous circuit break is found for all the surfaces during the whole test. On the other hand, although the LST reduces the nominal contact area of the interface, the reduction of contact area does not cause instantaneous circuit breaking phenomenon during the whole test. Therefore, as long as there are sufficient conductive areas for the current to pass, no high intensity fluctuation of voltage signal will be generated. Similar results were confirmed by Grandin et al. [10] as well, who found that although oxidation of the local contact area increased the local contact resistance, the contact voltage would not change significantly as long as there were sufficient conductive contact surfaces for the contact pair. The above analysis shows the feasibility of the application of surface texture in the electrical contact interface, and proves that the LST with specified pattern has a good effect on improving the interface friction and electrical contact stability.

Wear scars analysis

Figure 7 shows the wear morphologies of smooth surface and textured surfaces, which are observed by using SEM (JSM-6610LV). The smooth surface presents serious wear phenomenon. To be specific, a lot of furrows caused by abrasive wear and material transfer caused by adhesive wear are exhibited on the wear area, and obvious material delamination appears on the wear track. In addition, the weld holes and dispersed droplets caused by arc erosion is also found in the wear area. The RPTS also shows severe wear situation with significant furrow and material transfer on the surface. In addition, as the round pits are filled with wear debris, the sizes of some round pits are seriously reduced and part of round pits are even disappeared, which accordingly triggers the generation of FIV phenomenon [23,24].

OPEN IN VIEWER

Figure 7

The worn surface characteristics of smooth surface (a), RPTS (b), SPTS (c) and GTS (d) observed by SEM.

In contrast, the wear level of SPTS is relatively slighter. Although the material transformation caused by adhesive wear can be seen in some areas of the wear track, the furrows caused by abrasive wear are shallower. In addition, it is seen that the pits on the SPTS can effectively capture part of the wear debris generated from the interface. However, due to the friction and collision between the edges of the square pits and the ball sample, the edges of the pits are damaged as well. Moreover, the edges of the pits are deformed during the repeated impact of the ball sample. It is worth noting that the square pits of SPTS appear more intact than the round pits of RPTS, suggesting that the texture effect of SPTS during friction process is more significant than that of RPTS.

On the contrary, the size of the wear track of GTS is significantly reduced, and the surface wear level is the weakest among all the surfaces. In addition, it is seen that the wear debris mainly appear at the edges of the grooves, and no visible wear debris scattered and accumulated on the bulges of GTS. This is because when the ball slides through the groove, the groove edge will impact and ‘chip’ the ball sample, which causes the damage and delamination of edges areas, and part of wear debris existing on the ball surface in turn remains at the groove edge, whilst the other part of the wear debris falls into the grooves. This performance is conducive to reducing the FIV intensity. Overall, when the electric current passes the contact area, the surface texture can change the tribological properties, and in this work, the GTS exhibits good ability in improving the wear performance and FIV of electrical contact system.

Finite element analysis

Finite element model

To establish the finite element model of the test system in ABAQUS 6.18, the dimensions of each component in the test system are measured in detail. Considering that overly complex models may lead to the increase computation time, and will not further improve the calculation accuracy, thus the finite element model established in this work includes holder bar, connection rod, ball fixture, ball sample and flat sample, as shown in Figure 8. The normal load (2 N) is acted on the top surface of the holder bar. The velocity condition is applied on the bottom surface of the flat sample. The electric current of 3 A is input from a reference node on the ball fixture, whilst the electric potential at the bottom of the flat sample is set as 0 V. The flat sample surface is set as the master surface, and the ball surface is set as the slave surface. In order to ensure the calculation accuracy and the calculation efficiency, the surface textures of the flat samples in the sliding area are fine modelled, while the non-sliding area of textured surface is replaced by the smooth surface, as shown in Figure 8(c). All the components are meshed by using Q3D4 or Q3D8 element, considering that the coupled thermal-electrical-structural analysis will be carried out during numerical analysis. For the more detailed mesh properties of the finite element model can be seen in Ref [19].

OPEN IN VIEWER

Figure 8

Finite element model of the friction system (a), the boundary conditions of the model (b) and the detailed views of the ball slides on the flat samples (c).

In the calculation process, the normal load is firstly applied to establish contact between the flat sample and the ball sample, subsequently the current of 3 A is provided to the contact surface. A velocity signal is then applied to the flat sample to generate reciprocating motion. The implicit dynamics analysis method from ABAQUS/Standard is used in this model the calculate the sliding electrical responses of the system.

Analysis of contact pressure

Figure 9 shows the contact stress on different surfaces under the action of electric-thermo-mechanical multiple physical fields. The smooth surface exhibits visible stress concentration phenomenon. In the initial sliding stage, the maximum contact pressure is concentrated at the leading edge of the contact area. With the rise of temperature and the deformation of the contact area, the maximum stress transfers to the trailing edge. The stress concentration will lead to energy accumulation, serious wear and strong FIV generation, thus the smooth surface exhibits more severely wear situation and generate high-intensity vibration, as the test results shown in Figure 5 and Figure 7(a). While for the textured surfaces, the stress distribution is more uniform in the current-carrying state. Additionally, the maximum contact pressure is mainly concentrated at the edges of the textures, and the maximum pressure region changes continuously with the sliding process. Specially, for the RPTS, the maximum contact pressure is mainly concentrated on the rims of the rounds pits along the sliding path, thus the rims of the rounds pits along the sliding path are seriously damaged, as the test results shown in Figure 7(b). While for the SPTS, due to the sharp edges and right angles of the square pits, the maximum contact pressure sometimes appears on the edges of the pits along the sliding path, and sometimes appears on the edges of the pits on both sides of the sliding path, which reduces the wear degree of the surfaces of the pits to a certain extent, thus the pits remain relatively intact after the test, as the test results shown in Figure 7(c). In contrast, for the GTS, the interfacial stress distribution region is smaller and more uniform. Meanwhile, the contact area is significantly reduced due to the existence of grooves, thus the degree of interfacial wear is reduced. In addition, it can be seen that the maximum stress of GTS is most likely to appear on the edge of the groove, which explains the damage phenomenon of the edges of GTS, as shown in Figure 7(d).

OPEN IN VIEWER

Figure 9

The contact pressure of the different surfaces in the case of sliding electrical contact.

Visibly, the variation characteristics of contact stress can well explain the wear and vibration properties of different surfaces. Noteworthy, the existence of surface texture inevitably causes geometric discontinuity, which enlarge the pressure value of the textured surfaces. For the smooth surface, the maximum contact pressure is 51.55 MPa, while it becomes 91, 79 and 121 MPa for the RPTS, SPTS and GTS, respectively.

Analysis of contact temperature

Figure 10 shows the contact temperature distribution of the ball sample during sliding process. In the case of smooth surface, the temperature distribution on the ball surface is characterized by centrifugal distribution, and the highest temperature reaches 38.09°C, which is 13.09°C higher than that before friction process, as can be seen in Figure 10(a). Figure 10(b)–(d) show the surface temperature of the ball sample in sliding electrical contact with different texture surfaces. Visibly, the highest temperature of the ball sample in the case of RPTS further increases to 41°C, and temperature distribution area on the ball sample is similar with that case of smooth surface. In contrast, for both cases of SPTS and GTS, the temperature distribution areas on the ball surfaces increase visibly, and a visible low temperature area appears, suggesting that the temperature concentration area is interrupted. Additionally, the maximum temperature on the ball surface is 28.36°C and 31.32°C for the cases of SPTS and GTS, respectively, which is decreases by 33% and 25% compared with that of smooth surface.

OPEN IN VIEWER

Figure 10

The temperature distribution of the ball samples in the cases of smooth surface (a), RPTS (b), SPTS (c) and GTS (d).

Therefore, the SPTS and GTS can effectively increase the heat dissipation area, which is conducive to dissipating the heat generated from contact surface during the friction process. The maximum temperature in the case of SPTS is the lowest, and the low-temperature region in the case of GTS state is the largest. The different of the temperature distribution characteristic can be used to further explain the different in the wear performances of these surfaces, as illustrated in Figure 7. When the ball sample slides on the smooth surface, the stress concentration and surface temperature rise will accelerate the generation of wear debris at contact interface. The wear particles will gradually adhere to the contact surface since they cannot be removed in time. With the sliding process continues and the temperature further increases, the local convex body forms bonded joints. Under the action of friction force, the bonded joints are destroyed, and a large amount of discrete abrasive wear particles are formed again. The above process is repeated over and over again, and eventually the smooth surface is seriously worn, which is characterized by adhesive wear and abrasive wear. The higher temperature of the contact surface can be used to explain the wear characteristics of the smooth surface, and the formation of the wear mechanism to a certain extent. Meanwhile, the test results shown in Figure 7(a) also exhibit the wear behaviour as predicted by simulation.

While for the GTS, the existence of grooves is able to increase the heat dissipation area. When the ball sample passes through the grooves, the uncontacted area at the bottom of the ball sample forms a convective surface with the air, which effectively reduces the interface temperature and cause the formation of low-temperature area. The lower temperature of contact surface is beneficial for reducing the wear level, and suppressing the occurrence of adhesive wear, as the results shown in Figure 7(b). In the meantime, the grooves can effectively collect the wear debris and accordingly avoid the accumulation of wear debris, thus the surface wear is greatly alleviated.

Analysis of contact voltages and displacement

Figure 11 shows the variation of contact voltage and normal displacement of an observation point, which is located at the centre of the bottom surface of ball sample. For the smooth surface and textured surfaces, the contact voltage signals are visibly different due to the different contact situations. Since the effects of surface wear and FIV are not considered, the smooth surface produces the lowest contact voltage, while GTS produces the highest contact voltage. On the other hand, the contact position of the observation point changes when the friction ball slides through the groove, thus the voltage signal detected from the GTS appears significant fluctuation. Although the simulation results of the contact voltage are different from those of the test (as shown in Figure 5), the variation of the contact state of the interface is not taken into account in the calculation process, thus it can only imply that the voltage signal fluctuates and becomes larger in the initial stage due to the reduction of the contact area of the texture surface. However, with the change of interface wear, the characteristics of contact voltage will be modified as well.

OPEN IN VIEWER

Figure 11

The contact voltages (a) and normal displacement (ND) of the observation point on the bottom surface of ball sample (b).

Furthermore, the normal displacement (ND) of the observation point is analysed, as shown in Figure 11(b), which can be used to further explain the possible wear performances of different surfaces. Visibly, the ND signals of the observation point fluctuate to a certain extent in the case of all the textured surfaces. Specially, for both cases of the smooth surface and RPTS, the ND of the observation points are less than zero, suggesting that the observation point is penetrated into the flat sample surfaces during the friction process, and may consequently causes serious wear phenomenon. While for the case of SPTS, the ND of the observation point is always greater than zero, and the ND curve presents a gradual upward trend with the ball sample sliding through the square pits. It can be inferred that the ball sample warped to a certain extent during the friction process, which raises the centre position of the ball bottom, resulting in the formation of uneven contact area and eventually causes serious wear at the interface. In contrast, for the case of GTS, the ND of the observation point fluctuates slightly at zero value, indicating that the ball sample maintains good contact with the GTS, and no visible penetration can be seen between the ball sample and GTS, thus the GTS surface wear is slightest among all the surfaces.