Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Contents lists available at ScienceDirect

Tribology International
journal homepage: www.elsevier.com/locate/triboint

Role of cross-grooved type texturing in acceleration of initial

running-in under lubricated fretting
Miki Okamoto a,n, Tatsuhiro Jibiki b, Satoshi Ito c, Tomohiro Motoda d
a
Graduate School, Tokyo University of Marine Science and Technology 2-1-6, Etchujima, Koto-ku, Tokyo 135-8533, Japan
b
Tokyo University of Marine Science and Technology, Japan
c
Tokyo Metropolitan College of Industrial Technology, Japan
d
Basic Technology Research Center, Corporate Research and Development Center, NSK Ltd., Japan

art ic l e i nf o a b s t r a c t

Article history: Continuing efforts have been made towards fretting minimization. Simultaneously, texture studies have
Received 30 July 2015 also been actively pursued as ways toward solving tribological problems. However, few studies focusing
Received in revised form on both fretting and texture have been conducted. Therefore, we conducted fretting tests under lubri-
9 December 2015
cation for cross-grooved type textured, dimple textured and mirror-finished surfaces and evaluated their
Accepted 12 December 2015
initial running-in periods. Our results showed that cross-grooved type texturing is the most effective for
acceleration of the initial running-in period. Furthermore, our numerical analysis of contact pressure on
Keywords: these surfaces showed that the highest local contact pressure occurs when cross-grooved type texturing
Fretting is used. This indicates that high contact pressure promotes surface plastic deformation that leads to an
Running-in
acceleration of the initial running-in period.
Texture
& 2015 Elsevier Ltd. All rights reserved.
Lubricated

1. Introduction
An effective measure for retarding fretting is lubricant appli-
cation [1]. Additionally, texturing is effective for improving tribo-
logical characteristics. In recent years, studies of textured surfaces
under lubrication conditions have been actively pursued [2–5].
However, while numerous studies have examined issues such as
one-way slip, only a few studies focusing on fretting under various
texturing conditions have been conducted [6–8]. At the minimum
displacement amplitude in oil-lubricated fretting situations, it is
not easy to supply lubricating oil to sliding surfaces. Furthermore,
the influence of wear particles on surface conditions show differ-
ent amplitudes [9,10]. Under the fretting, the abrasive effects of
wear particles interposed at contact surfaces can also impose
difficulties. In such conditions, texturing provides relief in the form
of oil grooves and pockets to facilitate even lubricant distribution
and to collect wear particle away from the surface [5]. Thus, tex-
turing can be considered as a useful method for minimizing
fretting.
In this study, we focus on the initial running-in period, where a
significant reduction in the coefficient of friction is observed in the
early stage of fretting, in order to assess lubricating conditions.
While in our previous study we investigated dimple texturing [11],
this study focuses on cross-grooved type texturing. Fretting tests
n
Corresponding author.
are conducted and compared with dimple textured and mirror-
finished surfaces. The tests are conducted under constant load and
speed conditions, during which the influence of the fretting stroke
is investigated. The results are then discussed in light of the results
of numerical analyses of the contact pressure of these surfaces.
2. Experimental
2.1. Apparatus
Fig. 1 shows a schematic of the fretting test setup. As can be
seen in the figure, the driven specimen is attached to a cantilever.
Fretting stroke A is then supplied by a motor and a crank chain to a
cantilever (A is a setting stroke, not a relative stroke). The fixed
specimen is attached to an upper holder and loaded against the
driven specimen, after which the relative stroke between both
specimens is measured by an eddy current pick up. The frictional
force is measured by strain gauges mounted on upper holder.
2.2. Specimens and experimental conditions
Specimens used for the fretting test were HV760 steel ball
bearings (9.525 mm in diameter) for the side driven surface, and
HV760 flat bearing steel for side fixed surface. Before texturing, the
surfaces of the flat specimens were first machined to a mirror
finish process (Rz¼ 0.23 70.04 μm), after which cross-groove type

http://dx.doi.org/10.1016/j.triboint.2015.12.012
0301-679X/& 2015 Elsevier Ltd. All rights reserved.

Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
2 M. Okamoto et al. / Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 1. Schematic illustration of micro reciprocating test apparatus.

Fig. 3. Micrograph of dimple texturing.

Table 1
Experimental conditions.

Specimen Fixed Flat bearing steel (HV760)

Driven Ball bearing steel (HV760), ⌀9.525 mm

Configuration Ball on flat point contact

Fretting stroke A 30–200 μm
Normal load 9.8 N
Atmosphere Lubrication condition
Lubricant Poly-α-olefin (Base oil, 75 cSt @ 40 °C)
Test duration 2  105 cycles
Frequency 7.0 Hz
Temperature, Relative humidity 24 7 2 °C, 457 20%
Fig. 2. Micrograph of cross-grooved type texturing.

and dimple texturing were applied to individual specimens. We

used a diamond tipped contact probe profilometer with a tip
radius of 5 μm and a resolution of 2 nm to measure surface
roughness. The trace length was 4.0 mm, the filter was 2CR, and
the cutoff was 0.8 mm. The trace speed was 0.1 mm/s.
Mirror finish processing was applied using emery papers
(#80  #3000) and diamond paste. All three surfaces (mirror,
cross-grooved type, and dimple) were used in our tests. The tex-
ture processing method was the same as used in previous report
[11], and a diamond indenter was used to create textures. Post-
processing for removing pile-ups was performed using emery
paper (#3000) and diamond paste. The created texture pattern
was observed in detail using a confocal laser-scanning microscope
(CLSM). The images are shown in Figs. 2 and 3. Table 1 shows the
detailed experimental conditions. The CLSM specifications are
shown below. The lateral resolution is 0.30 μm, the z-axis reso-
lution is 0.05 μm and the measurement range is 256  192 μm2.
Fig. 4. Change in coefficient of friction (A ¼70 μm).

3. Results
3.1. Coefficient of friction μ stroke A. As A becomes larger, Nr becomes smaller. Arrows at
Fig. 4 shows changes to the coefficient of friction μ during a
fretting test (A¼70 μm). As can be seen in the figure, μ rises
sharply for approximately 500 cycles after the test starts, and then
becomes stable at about 0.5 for each test piece. After approxi-
mately 900 cycles, μ for the cross-grooved type texturing drops
drastically, indicating that the running-in process had been com-
pleted. In contrast, the same μ drop for the dimple textured sur-
face occurs at approximately 4000 cycles. Finally, for the mirror
finished specimen, a μ decline phenomenon is seen at about
30,000 cycles. It should also be noted that the final coefficient of
friction is approximately 0.2 for all test specimens.
Fig. 5 shows the relationship between the number of cycles
needed for completion of the running-in period (Nr) and fretting
approximately 200,000 of Nr show when the running-in period
could not be observed during the tests. In contrast, an Nr of
1 shows that coefficient of friction was low from the starting point.
Regardless of fretting stroke, the running-in process completed
first for the cross-grooved type textured, then the dimple textured,
and finally the mirror finished surfaces.
3.2. Wear scar observations
The width and the cross-sectional form of the wear scars were
measured using a laser microscope. Fig. 6 shows a comparative
laser microscope image of each test specimen for a fretting stroke

Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
 M. Okamoto et al. / Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

of 70 μm. As can be seen in the figure, damage is more intense on The following equation can be established for the apparent
the textured surfaces (a, b) than on the mirror finished surface (c). contact surface, S, on the X–Y plane.
Fig. 7 shows the wear scar width for each fretting stroke A. The
left vertical axis shows the actual measurement value of the wear δ  ðw1 þ w2 Þ ¼ z1 þz2 ð1Þ
scar (solid line). As can be seen in the figure, wear scar widths tend to
increase for all specimens. Furthermore, the right vertical axis shows where δ is the relative approach between specimens, wi is the
the wear scar width ratio of textured surfaces to the mirror-finished elastic deformation quantity and z1 þ z2 is the initial interspace.
surface (dash line). Even though the part is reversed, it can be seen The subscripts represent the two solid specimens.
that the wear scar on the cross-grooved typed textured specimen is Additionally, wi is described by the following equation.
larger than the one seen on the dimple textured specimen.
Z
1  ν2i pdX 0 dY 0
wi ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2  2ffi ði ¼ 1; 2Þ ð2Þ
4. Discussion π Ei s
X  X0 þ Y  Y 0

4.1. Contact pressure numerical analysis method

where νi is Poisson's ratio and Ei is Young’s modulus.
Substituting (2) for (1), following equation is obtained.
Numerical analysis of the contact pressure was performed in
order to investigate the initial running-in period mechanism. The Z
pdX 0 dY 0
calculation method used previously was also adopted for this δ  C qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
    ffi ¼ z1 þ z2 ð3Þ
X  X0
2
þ Y Y 0
S 2
study [11,12].

Fig. 5. Relationship between running-in cycles and fretting stroke for various samples. Fig. 7. Relationship between width of wear and fretting stroke for various samples.

Fig. 6. Observation of fretted wear scars (A ¼70 μm).

Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
4 M. Okamoto et al. / Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Fig. 8. Model used for numerical analysis (Contact pressure ¼0 at dark area).

Fig. 9. Comparison of wear scar with contact area by numerical analysis (A ¼ 30 μm).

where The following conditions are used to obtain a solution:

Xr
1ν 2
1ν 2
Ap ¼P ð6Þ
C¼ 1
þ 2
ð4Þ j¼1 j i
π E1 π E2
Because it is difficult to solve Eq. (3) analytically, we instead use pi Z0ði ¼ 1; 2; ⋯; r Þ ð7Þ
numerical analysis. The contact surface zone is divided into a finite
where P is the contact pressure and A is the area of minute region.
number of minute regions. The modified equation is shown below.
Next, we prepared a calculation model. By utilizing the symmetry
Xr
δ  C j ¼ 1 pj Dij ¼ z1;i þ z2;i ði ¼ 1; 2; ⋯; rÞ ð5Þ of the contact point, calculations were performed on divided the
contact surface zone into four pieces. Fig. 8 shows the model used
where r is the division number, Dij is the influence coefficient and for the numerical analysis. The discretizing pitch was 1 μm. Since
pj is the average contact pressure of the minute region. it was assumed that the textured portion was not in contact, the

Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
 M. Okamoto et al. / Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Fig. 10. Comparison of wear scar with contact area by numerical analysis (A¼ 70 μm).

Fig. 11. Contact pressure contour lines.

contact pressure was set to zero. Young's modulus of the bearing
steel was set to 212 GPa and Poisson’s ratio was set to 0.30.
4.2. Contact pressure of textured surfaces
Fig. 9 shows a comparison of the contact area obtained by a
calculation with the wear scar. Here, the fretting stroke A is 30 μm,
which is very small. However, initial running-in was not observed
under these conditions. The white circles in the figure describe the
Hertzian contact area. Because the actual contact position is dif-
ferent from the calculated contact position, slight differences
occurred in the contact surface shapes. However, the shapes are
very similar. This confirms the effectiveness of our numerical
analysis.
Fig. 10 shows a comparison of the contact area obtained by
calculation with the wear scar at A ¼70 μm. Under these condi-
tions, initial running-in was observed. Looking at the cross-
sectional shape, the texture shape can still be observed.
Fig. 11 shows contour plots of the plane contact pressure for the
different surfaces. It can be clearly seen that the maximum pres-
sure is higher for the textured surfaces than in mirror finished
surface. Fig. 12 shows the line pressure distribution along the y-
axis, which is obtained by displaying the part of the respective
surfaces that receives the maximum contact pressure. However,

Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
6 M. Okamoto et al. / Tribology International ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 12. Contact pressure distribution.
Table 2
Comparison of contact area.
Cross-grooved Dimple Mirror finished
type texturing texturing
Contact area [μm^2] 7600 (53.0) 9352 (65.2) 14,340 (100)
(Ratio of mirror fin-
ished [%])
the second highest pressure area for cross-grooved type texturing
is also displayed.
As can be seen in the figure, the maximum contact pressure for
the cross-grooved type texturing is more than 20 GPa, which is
much higher than that for dimple texturing. This 20 GPa is an edge
stress. However, paying attention to the portion away from the
edge, it can be seen that contact pressures are sufficiently high
(about 1–2 GPa). The order is cross-grooved type texturing, dimple
texturing and mirror finished.
Table 2 shows the contact area (the blue area in Fig. 9). With
regard to high pressure in the textured surfaces, the contact area
that supports the load is the largest for the mirror finished surface,
and smallest for the cross-grooved textured surface.
Now, since the yield point of the bearing steel is approximately
0.4 GPa, it can be presumed that the local high pressure causes
plastic surface deformation, which promotes acceleration of the
initial running-in period. As a result, the initial running-in period
was seen early for the cross-grooved texturing because it had the
highest contact pressure.
4.3. Consideration of initial running-in period mechanism
In the same manner as described above, another study exam-
ined running-in for the case of plastic deformation caused by
micro texturing [13,14]. It was also found that micro texturing
reduced the surface stiffness [11]. This leads us to presume that
the surface deformation effect accelerated the initial running-in
Please cite this article as: Okamoto M, et al. Role of cross-grooved type texturing in acceleration of initial running-in under
lubricated fretting. Tribology International (2015), http://dx.doi.org/10.1016/j.triboint.2015.12.012i
period. In addition, it can also be explained from the sliding dis-
tance effects. Archard claims that wear volume is increased in
proportion to the sliding distance [15]. In our results, initial
running-in occurs earlier as the fretting stroke becomes larger
(Fig. 5). Therefore, it can be said that initial running-in is promoted
by wear-caused deformation.
5. Conclusions
1. The effect of texturing was confirmed under lubricated fretting
conditions.
2. The initial running-in was observed earlier in cross-grooved
type texturing than in species with dimple texturing and
mirror-finishing. Cross-grooved type texturing is the most
effective at the initial running-in.
3. According to the contact pressure calculation, higher contact
pressures are generated on textured surfaces, and was highest
on cross-grooved type textured surfaces where the contact area
is also the smallest. This, in turn, promotes contact surface
deformation, leading to acceleration of the initial running-in
process.
Acknowledgments
This work was supported by a Grant-in-Aid for Scientific
Research (C) (No. 24560985) from the Japan Society for the Pro-
motion of Science (JSPS).
References
[1] Waterhouse RB. Fretting Corrosion. Pergamon Press; 1975.
[2] Hamilton DB, Walowit JA, Allen CM. A theory of lubrication by micro-
irregularities. Trans ASME J. Basic Eng 1966;88(1):177–85.
[3] Hubertus U., et al. Investigation of microstructured cylinder liner surfaces for
friction reduction. International council on combustion engines 2013 Shanhai.
[4] Etsion I, Halperin G, Brizmer V, Kligerman Y. Experimental investigation of
laser surface textured paralle lthrust bearings. Tribol Lett 2004;17(2):295–300.
[5] Umehara N. Improvement of friction Property with surface texture. J Japan Soc
Mech Eng 2009;112(1086):406–9.
[6] Vázquez J, Navarro C, Domínguez J. Analysis of the effect of a textured surface
on fretting fatigue. Wear 2014;305:23–35.
[7] Volchok A, Halperin G, Etsion I. The effect of surface regular microtopography
on fretting fatigue life. Wear 2002;253:509–15.
[8] Varenberg M, Halperin G, Etsion I. Different aspects of the role of wear debris
in fretting wear. Wear 2002;252:902–10.
[9] Li Q, Shima M, Yamamoto T, Sato J. Study on fretting wear of rolling bearing
(part 4) -effects of various factors on fretting wear-. Jpn J Tribol 1995;40
(12):1029–36.
[10] Itoh T., Fuji S., Sugimura J., Yamamoto Y.. Effects of frequency and stroke on the
running-in process in reciprocating friction. In: Proceeding of the 56th JSME
Kyushu Branch Symposium 2003, 56, pp. 243–244.
[11] Jibiki T, Shima M, Motoda T. Shipway P. Role of surface micro-texturing in
acceleration of initial running-in during fretting. Tribol Online 2010;5(1):33–9.
[12] Motoda T, Jibiki T, Shima M. Effect of surface roughness on hertzian contact
deformation. Jpn J Tribol 2004;49(2):163–75.
[13] Kumada Y, Hashizume K, Kimura Y. Performance of plain bearings with cir-
cumferential microgrooves. Tribol Trans 1996;39(1):81–6.
[14] Sugimura J, Kimura Y, Amino K. Analysis of the topographical changes due to
wear -Geometry of the running-inprocess-. J Japan Soc Lubr Eng 1986;31
(11):813–20.
[15] Archard JF. Contact and rubbing of flat surfaces. J Appl Phys 1953;24(8):981–8.