As shown in Figure 1, the textured pin is pressed vertically onto the rotating disc. The angle α between the textured and rotating surfaces is adjusted using the lever with a micrometer gauge. By applying the lubricant in front of the contact zone, a sufficient lubricant supply is ensured (Figure 2). The rotation speed of the disc is set to 1,136.8 min^-1. The eccentric arrangement (r_ex = 42 mm) between the pin and disc results in a relative speed of 5 m/s.

For the experimental tests, first, the lubricant supply is activated, and then the disc is set into rotation. At the beginning, a minimum normal force F_N (∼0.5 N) is set on the tribometer. Due to the hydrodynamic pressure build-up, resulting from the relative speed, the lubrication wedge geometry, and the lubricant viscosity, a certain lubrication gap height h₀ is obtained. This ensures that the two contact surfaces are completely separated from each other and that the normal force on the pin is fully absorbed by the hydrodynamic force in the lubrication gap. Then, the normal force F_N is successively increased on the tribometer, while the relative speed remains constant, which results in a different minimum lubrication gap height h₀. The normal force is increased up to a level of 26 N, which represents a nominal contact pressure of 0.26 MPa. At no point during the experiment did asperity contact or wear occur, and full-film lubrication was always present. The clear height h₀ is measured using the eddy current sensor for the different normal forces F_N (Figure 2). The experiments were conducted at room temperature, which was approximately 25°C.

The pins are made of aluminum alloy EN AW-6082 (AlSi1MgMn). In addition to a smooth version, two further textures with two different texture heights (h_t = 10 μm and 50 µm) (Figure 3) are tested in the experiments. High-precision milling tools are used to produce the textures on the pin surface. The second contact surface, the rotating disc, is made from steel 42CrMo4. Roughness values are provided in Table 2. Lambda values vary between 35 (the maximum occurring lubrication gap height of 45 µm) and 15 (the minimal occurring lubrication gap height of 20 µm).

The details of the manufactured textures are shown in Figure 4. It needs to be pointed out that these textures are not optimized in a preceding simulation process. Therefore, it was initially unclear whether these textures would cause an enhancement or deterioration. The main goal was to provide strongly varying geometry to check the test method’s separation properties.

The engine oil, Shell Rimula R5 10 W-30, with a dynamic viscosity of 0.294 Pas and a density of 857.2 kg/m³ at 25 C, is used as the lubricant.

Table 3 provides an overview of the tested textures and the boundary conditions.

The hydrodynamic forces F_L and F_D are induced by the flow and the pressure in the lubrication gap. The vertical normal force F_L can be evaluated directly using the load cell integrated into the tribometer. In addition, the friction force F_D is determined from the multichannel force of the Optimol Rotational Module (Optimol Instruments Prüftechnik GmbH, 2024). The temperature in the contact environment is recorded by the integrated measuring unit of the test machine. A specific combination of surface texture, the rotational speed of the disc, used lubricant, and applied vertical load results in a corresponding clearance height h₀. This is determined using the eddy current sensor ES-S1 (Micro-Epsilon, Ortenburg, Germany) and transferred by an OPC-UA interface option to SRV software.

At the beginning of a test, the specimen first has to be aligned parallel to the disc. This is achieved by pressing the pin onto the stationary disc. For qualitative evaluation of the parallelism of the pin and the disc surface, pressure-sensitive measurement films (Fujifilm Prescale, Tokyo, Japan) are placed between these two surfaces, and the parallelism can be adjusted using a micrometer screw. The alignment perpendicular to the rotational axis of the micrometer gauge is achieved by inserting feeler gauges into the mounting area of the adapter. For the measurement of the lubrication gap height, the measurement of the initial displacement of the eddy current sensor is recorded at a standstill (h₀ = 0).

Finally, the disc is accelerated under nearly load-free conditions, and the lubrication gap reaches its maximum value at the desired relative speed (5 m/s). Then, the vertical load FL is successively increased, and the change in the lubrication gap height h₀ can be measured using the eddy current sensor with a theoretical accuracy of 0.02 µm. The sensor has been calibrated to the disc material (42CrMo4) by the manufacturer to ensure the corresponding measurement accuracy. Furthermore, field calibrations were performed with the installed test rig.

To check the plausibility of the method, an analytical calculation is used. Based on the Reynolds equation, the approaches detailed by Lang and Steinhilper (2014), Fuller et al. (1960), and Schiebel and Körner (1933) can be used to calculate the normal force of a smooth, convergent, lubricated wedge. In addition, using the approach described by Vogelpohl (1967), the friction force can be calculated analytically as follows: F f i n i t e = 5 6 · F i n f i n i t e 1 + a L B 2 , F i n f i n i t e = B   6 · η · U · L ² h 1 − h 0 ² · ln h 1 h 0 − 2 h 1 − h 0 h 1 + h 0 , h 1 = h 0 + l ∗ ⁡ tan   α , a = 10 1 + 2 m ² m + m 2 2 + 1 − 2 · m + m 2 12   1 + 2 m ln 1 + m m − 2 , m = h 0 h 1 − h 0 = 1 m ′ , µ = 4 5 · 1 + 2 m   ln 1 + m m − 1 , 5 L h 0 · m 1 + a L B 2 1 + 2 m ln 1 + m m − 2 .

For all the investigated pitch angles, the analytical solution was applied. Figure 5 shows the result for a pitch angle of 0.3° as blue dots. As can be seen, with an increasing normal force, which was controlled during the experiments, the lubrication gap height decreases. In addition, a decrease in friction force in relation to the normal force can be recognized.

For the reference tests with smooth surfaces, three separate tests were conducted, showing good reproducibility between them. Furthermore, similar trends to those seen in the analytical solution can be observed. However, a discrepancy exists between the analytical and experimental results. Similar mismatches, as shown in Figure 5 for the 0.3° configuration, were found for the other pitch angle configurations. Possible reasons for this mismatch will be analyzed in the “Discussion” section.

Figures 6–8 depict the results of the experiments with textured pins. For each pitch angle, the results of texture specimens are compared with the corresponding results of smooth pin specimens. It can be observed that there is a clear separation between the different texture geometries and texture depths, as well as a significant separation from the reference results of the smooth pin.

All tests were conducted in full hydrodynamic film lubrication. All tests were started with a low normal force, which corresponds to a high gap height. The maximum applied normal force was chosen in such a way that the gap height did not exceed 20 µm. If one considers the roughness values from Table 2, lambda values are between 35 (maximum occurring lubrication gap height of 45 µm) and 15 (minimal occurring lubrication gap height of 20 µm), which is typical for full-film lubrication.

Figure 6 shows the results for a pitch angle of 0.2° and 0.3°. Comparing the lubrication gap heights, it can be recognized that all textures show a smaller gap height at the same normal force as the smooth reference tests. In this way, a higher risk of entering a mixed friction regime can be assigned.

It is the cylinder texture with a depth of 50 µm that shows the highest gap heights for the tests at 0.2°. In addition, the friction force is lower at a similar normal load than for the smooth reference geometry. This results in a beneficial COF behavior for the cylinder size with 50-µm depth.

The chevron textures with a similar depth of 50 µm depict the lowest gap heights. Friction is in the very range of smooth results. For both texture geometries with smaller depths (10 µm), the lubrication gap heights lay between the two deeper geometries. It seems that, at least for the gap height, the geometry of the texture, whether a cylinder or a chevron, is important.

For the test with a pitch angle of 0.3°, the lubrication gap height of all textured geometries is much more in the range of the smooth reference geometries. Chevron and cylinder textures with a depth of 50 µm depict a lower gap height, whereas the cylinder texture with 10-µm depth shows the highest gap height. Regarding the friction force, this exact texture (cylinder with 10-µm depth) shows the highest friction force and, consequently, also the highest COF, which lies in the very range of the smooth results. The other textures show lower friction forces at the corresponding normal forces, which leads to lower COF values. The Chevron geometry with a depth of 10 μm, which provides the lowest COF values, seems very beneficial, but it only creates a moderate lubrication gap height reduction.

Figure 7 shows the results for pitch angles α = 0.4° and 0.5°. The trend continues, where, with an increasing pitch angle, the lubrication gap heights of the textured specimens approach those of the smooth reference tests. For a pitch angle of 0.4°, the shallower textures (chevron and cylinder with 10-µm depth) have the higher gap height, followed by a higher friction force and COF. Cylinder textures provide slightly higher values (gap height, friction force, and COF) than the chevron texture with 10-µm depth. For this pitch angle, again, the chevron geometry with a depth of 10 µm is the one that provides the best combination of friction reduction and moderate gap height reduction.

The situation for a pitch angle of 0.5° seems much more complex. The highest gap height is provided by chevron textures with 10-µm depth. However, for friction force and COF, this texture does not provide such a clear trend and lies more in the middle of the results. This is in agreement with the previous results that the chevron geometry with 10-µm depth is a very good compromise between friction reduction and bearing capacity. The lowest gap height for the investigation at a pitch angle of 0.5° is provided by the 50-µm-deep chevron texture; however, contrary to the trend, this not only provides the lowest friction but also the highest friction force and COF. For friction, the cylinder texture with a depth of 50 µm is the best.

Figure 8 depicts the results of the investigations with pitch angles of 0.75° and 1.0°. For both pitch angles, the cylinder texture with 10-µm depth shows quite similar lubrication gap heights as the smooth reference specimens. In addition, the friction force and COF for this texture are the highest for both pitch angles. Furthermore, it can be observed that for both angles, 0.75° and 1°, the chevron with 10-µm depth provides the lowest COF, so from a frictional point of view, it provides the best enhancement. Particularly, for the investigation at a 0.75° pitch angle, the measured lubrication gap height is quite close to the smooth reference, so the bearing capacity is also good. However, it must be noted that at the higher pitch angles, the overall bearing capacity is significantly reduced, with the maximum reachable normal force being approximately 14 N and the friction forces being less than 1 N for a pitch angle of 1°.

In general, it can be observed that the lubrication gap heights at the same normal forces are lower for the textured specimens than for the smooth specimens. The consequence is a reduction in the friction force and COF, but there is also the risk of operation in the mixed friction regime. Overall, it is difficult to say which texture performs the best. For most cases, the chevron geometry with a depth of 10 µm provides the best compromise between friction reduction and loss of bearing capacity. For some cases (e.g., a pitch angle of 0.75°), the lubrication gap heights of this texture are in the range of the smooth results, indicating a similar bearing capacity but with significant reduction in friction and COF. However, the main finding is that a universal statement is hard to find, and textures need to be designed individually for the specific case of operation.