CHARACTERIZATION OF DLC PVD COATINGS PROPERTIES
ON WAVE BEARINGS OF AERO-TRANSMISSIONS
K.-D. Bouzakis1, A. Asimakopoulos1, M. Batsiolas1, N. Ene2, F. Dimofte3,
R. Handschuh4, T. Krantz4
1. Aristoteles University of Thessaloniki, 54124, Greece
2. MIME Department of the University of Toledo, Toledo, Ohio, USA
3. University of Toledo at NASA Glenn Research Center in Cleveland, Ohio, USA
4. U.S. Army Research Laboratory,NASA Glenn Research Center in Cleveland, Ohio, USA
A wave bearing has a wave profile superposed on the surface of the bearing sta-
tionary part. The sleeve (outer ring) and the rotor (inner ring) of this bearing are fab-
ricated from hard steel alloy. The rotor’s external and the sleeve’s internal cylindrical
surfaces were coated with the same DLC coating and the same deposition process.
The scope of the present investigation was to reveal potential differences in the im-
pact test performance of the two DLC coated parts, the sleeve and the rotor. Addi-
tionally, to determine the coatings’ and substrates’ strength properties and to assess
the films’ adhesion quality.
KEYWORDS: Wave bearing, coatings characterization, impact test, fatigue, adhe-
The impact test is used for the quantitative assessment of various properties of thin hard coat-
ings, deposited on machine elements, tools etc. This test is mainly applied on coated speci-
mens with simple geometries such as of cutting inserts and coated plates. In the described in-
vestigations, perpendicular and inclined impact tests were conducted directly on TiC coated
bearing rings. The tests were performed with the aid of appropriate fixtures on the internal cylin-
drical surface of the outer bearing ring, as well as on the external one of the inner ring. The
possibility to test the coatings as deposited on the bearing ring surfaces, i.e. directly on the final
product, is pivotal, because the PVD process significantly affects the attained film properties
In the described investigations, nanoindentations were firstly conducted on coated bearing
rings, to determine the film and the substrate mechanical strength properties with the aid of a
finite elements method (FEM) supported results evaluation /1/. Furthermore, using developed
jigs and fixtures, perpendicular and inclined impact tests were conducted directly on coated
bearing rings. Finally, based on FEM simulations of the experimental procedures and appropri-
ate calculations, the coating’s fatigue and adhesion were quantitatively characterized. Addition-
ally, the film thickness of the sleeves were determined according to results on impact test im-
prints profile measurements.
2. COATINGS’ THICKNESS AND MATERIALS MECHANICAL STRENGTH PROPERTIES
The ball-cratering test was applied to determine the coating thickness on the external coated
rotor surfaces. This test could not be applied on the internal coated sleeve surface, due to
Proceedings of the 7th International Conference Coatings in Manufacturing Engineering, 1-3 October 2008, Chalkidiki, Greece
Edited by: K.-D. Bouzakis, Fr.-W. Bach, B. Denkena, M. Geiger,
Published by: Laboratory for Machine Tools and Manufacturing Engineering (ΕΕΔΜ),
Aristoteles University of Thessaloniki and of the Fraunhofer Project Center Coatings in Manufacturing (PCCM),
a joint initiative by Fraunhofer-Gesellschaft and Centre for Research and Technology Hellas
evaluation accuracy reasons, as it will be further explained in the report. Characteristic ball-
cratering test micrographs on the bearing rotors are illustrated in figure 1a. The DLC coating
has a thickness of 1.6 µm on the rotor ring. According to results based on impact imprints profile
measurements, the DLC coatings are approximately of the same thickness both on the rotor
and on the sleeve rings as well.
The films’ and substrates’ strength properties were determined by nanoindentations. During the
nanoindentation, the course of the applied indentation load versus the occurring penetration
depth is continuously monitored. The precision of this procedure is affected by the specimen’s
surface integrity. In order to improve the results evaluation accuracy, 30 measurements were
conducted on each specimen, under a peak load of 15 mN. The determined nanoindentation
mean values curves of the examined coatings are exhibited in figure 1b. The extracted stress-
strain curves, by means of the ‘SSCUBONI’ algorithm /1,2,3/ are also illustrated in this figure.
Nanoindentations were also applied on the steel substrate (EX-53) of the DLC coated rings. The
nanoindentation mean values curves of the steel substrates, at a peak load of 15 mN are exhib-
Figure 1 (a) : The ball-cratering test film thickness evaluation method and characteristic imprint
on the DLC coated bearing rotor. (b) Nanohardness mean value curve and
stress-strain curve of the examined coating.
Figure 2: (a) Nanoindentation mean values curves of the substrate of the DLC and the TiC#10
coated rings under a peak load of 15 mN. (b) Stress-strain curves of the bearing
340 7th Coatings – 2008
ited in figure 2. The substrate hardness of the DLC coated rings amounts approximately to 550
HV, whereas the substrate hardness of other bearing rings coated with a TiC film amounts to
830 HV. The hardness decrease of the DLC steel substrate might be associated with annealing
effects during the deposition process. The low hardness of this substrate is expected to lead to
premature film damages and moreover to substrate material forging and abrasive wear effects
in the perpendicular and inclined impact tests correspondingly.
The determined substrate stress-strain curves are also shown in this figure. The substrate of the
TiC#10 coated ring possesses enhanced yield and rupture stress compared to the correspond-
ing one of the DLC coated ring.
3. THE APPLIED IMPACT TEST FIXTURES AND MEASUREMENTS’ CONDUCT
The impact test was applied on the inner and outer cylindrical surfaces of bearing rings with a
tungsten carbide ball of 5 mm diameter. The arrangement enabling the conduct of impact tests
on bearing rings is exhibited in figure 3a. This coated rings’ fixture consists of a metallic plate
which is securely fastened on the base of the impact tester. On top of this plate, a fixture device
(vise) is installed, supporting the V-shaped rings’ fixture. The specimen seats vertically on the
V-shaped fixture. In the case of the impact test on the bearing sleeve, the ball indenter is held
on the impact tester spindle by a steel frame. The desired impact angle during the inclined im-
pact test can be accurately defined and adjusted, considering the displacement L of the speci-
men in the horizontal direction, as illustrated in figure 3b.
The developed impact imprints on the coated bearing rings were monitored by a stereo-
microscope. These have an oval shape, in the case of the bearing sleeves (see figure 3c)
whereas the corresponding ones on bearing rotors micrographs a more circular contour. The
reason for this is the positioning of the sleeves at an oblique direction under the microscope,
Figure 3: (a): Applied fixtures in the impact tests on bearing rings. (b) Coated bearing ring po-
sitioning to conduct perpendicular or oblique impact tests at various inclination an-
gles. (c): Positioning of the rotor and sleeve rings under the stereo-microscope and
characteristic impact test imprints micrographs.
Coatings on Machine Elements and Machine Components 341
concerning the imprint surface, due to the fact that the imprints on the sleeve internal cylindrical
coated surface are not visible at a perpendicular direction. On the other hand, the imprints on
the rotor external cylindrical coated surface can always be observed at a perpendicular direc-
4. PERPENDICULAR IMPACT TESTS AND FATIGUE ENDURANCE STRESS DETERMI-
Due to the substrate plastic deformation during the impact test, a remaining imprint develops
during the impact relaxation stage. The dimensions of this remaining imprint depend on the ap-
plied impact load. To detect whether a potential film damage and a consequent coating removal
took place, appropriate FEM supported calculations to determine the imprint shape of the plasti-
cally deformed substrate were conducted. By this FEM model, the developed imprint profile ge-
ometry during the loading and the relaxation stage can be calculated in the case of an applied
impact load of 100 N and is displayed in figure 4a. The maximum developed stress in the coat-
ing during the impact loading stage, in the present case, amounts to approximately 2.2 GPa in
the crater vicinity, as exhibited in the left part of this figure. During the relaxation stage, due to
the permanent substrate plastic deformation, a remaining stress of 0.9 GPa occurs.
The developed imprint profiles’ maximum depth at various impact loads, due to the remaining
substrate plastic deformation, were calculated and are displayed in the diagram of figure 4b.
According to these FEM calculations, the substrate is plastically deformed at impact loads larger
than approximately 100 N. Moreover, the coating can be considered as still lying on the ring
surface, if the measured imprint profile maximum depth does not exceed than the calculated
one of the relaxation stage. On the other hand, a coating removal has occurred, if the measured
imprint profile maximum depth is larger than the calculated one plus the coating thickness. The
FEM calculated imprint profile maximum depths at various impact loads, as well as the corre-
sponding measured ones, are monitored in the table of figure 4b. The FEM simulation of the
perpendicular impact test was also used to determine the stress distribution in an assumed
case of a substrate having the enhanced material properties of the TiC coated bearing rings’
one. The maximum developed stress at an applied impact load of 100 N amounts in this sub-
strate case to approximately 0.9 GPa i.e. ca. 35% lower compared to the real substrate case, as
shown in figure 4c. Moreover, in the relaxation stage no plastic deformation develops and con-
sequently no residual stresses. Therefore it is recommended to avoid annealing effects during
the deposition of the DLC coatings on the steel substrates.
Perpendicular impact tests were performed on the DLC coated sleeve and rotor with a ceramic
ball of 5 mm diameter. The perpendicular impact tests were conducted at various impact loads
for 106 impacts and the corresponding imprint profile measurements on the DLC bearing rings
are shown in figure 5. The maximum measured imprint depths and the imprint diameters are
very high depending mainly on the substrate mechanical strength properties the measured im-
print profile maximum depths at various impact loads, are monitored in the table of figure 5. As it
can be observed, coating removal occurred in all the impact load cases. The further increase of
the occurring imprint maximum depth can be explained through forging effects in the substrate
due to its low hardness. The higher imprint depth indicates that the substrate in the DLC coated
rings cases, due to the deteriorated strength properties (see figure 2), deforms more than in the
TiC coating cases.
In order to determine the fatigue strength, an impact load of 50 N was assumed to be associ-
ated with the film removal after 106 impacts, since at a load of 100 N, a film failure occurred (see
figure 5). The DLC coating’s fatigue properties, are presented in form of Smith and Woehler
diagrams, determined by the ITEC+ software /4/. The established Smith and Woehler diagrams
are exhibited in figure 6. In this coating case, the impact load of 50 N, considered as fatigue
critical for 106 impacts, corresponds to an endurance critical stress of 1 GPa. This stress is con-
sidered as very small.
342 7th Coatings – 2008
Figure 4: (a) Von Mises stress distribution of the examined DLC coated rotor and sleeve dur-
ing the loading and relaxation stage. (b) FEM calculated maximum remaining depth
at various impact loads. (c) FEM calculated imprint geometry during the loading
stage applying enhanced substrate material properties
5. FILM ADHESION INVESTIGATION BY THE INCLINED IMPACT TEST ON THE DLC
The adhesion strength of the DLC coated rotor and sleeve, were quantified by the method de-
scribed in /5/. Coatings on poor-adherent substrates are stressed more intensely, in comparison
to well-adherent films, when tangential loads are exercised in the film-substrate interface. By
FEM supported calculations, the developed contact loads between the ball and the coated sur-
face can be determined and these data be further applied in a similar FEM model, with suitable
Coatings on Machine Elements and Machine Components 343
Figure 5: Profiles of the occurred imprints on the DLC coated bearing rings. and measured
maximum imprint remaining depth at various impact loads.
Figure 6: Smith and Woehler diagrams of the examined DLC coatings (rotor and sleeve).
Figure 7: Maximum von Mises stress increase versus the contact stiffness ratio.
344 7th Coatings – 2008
contact elements between the coating and the substrate surfaces. The used contact elements
are characterized by the normal csn and the tangential cst contact stiffness. The ratio CSR of the
tangential contact stiffness cst, to the normal one csn, characterizes the adhesion strength in the
coating-substrate region /5/.
The maximum von Mises equivalent stress increase versus the contact stiffness ratio, devel-
oped during the inclined impact test, at various impact loads, is shown in the left part of figure 7.
The maximum von Mises equivalent stress percentile increase is also shown in the right part of
Inclined impact tests were conducted on the DLC coated specimens at various numbers of im-
pacts at a load of 50 N and an inclination angle of 100. The corresponding inclined impact test
imprints micrographs and profiles are shown in figure 8. At these test conditions no plastic de-
formation occurs in the substrate.
Figure 8: Inclined impact test on the DLC coated bearing rings imprints micrographs and pro-
Coatings on Machine Elements and Machine Components 345
The coating on the sleeve was damaged already after 100x103 impacts, whereas the rotor film
was almost intact after 300x103 impacts. The adhesion strength superiority of the DLC coated
rotor versus the DLC coated sleeve is evident by these results. It is recommended to further
optimize the deposition process on the sleeve rings. Moreover, taking into account the meas-
ured imprint profiles, the DLC coating thickness on the sleeve ring amounts to 1.6 µm, as al-
ready shown in figure 1.
The associated with this impact load contact stiffness ratio i.e. the stress, leading to a film fail-
ure, was determined for the DLC coated rings, as shown in figure 9. The number of impacts as-
sociated with the film damage at an impact load of 50 N was approximately 100x103 impacts for
the DLC sleeve film, as extended coating failure was detected in all the conducted impact test
imprints after this number of impacts. The corresponding number of impacts for the DLC rotor
film was estimated to be approximately 350x103. According to the diagram in the right figure
part, the contact stiffness ratio amounts to 4.5x10-3 for the DLC sleeve film, which is considered
as poor and to 1.5x10-1 for the DLC rotor film, considered as very good /5/. The adhesion of the
DLC rotor film was found to be very sufficient, in contrast to the corresponding one of the
sleeve, justifying the early coating failure on the sleeve and the adequate resistance of the rotor
in the conducted inclined impact tests at an impact load of 50 N.
Figure 9: Determination of the contact stiffness ratio for the DLC coated rings, considering the
number of impacts at the film failure start.
The DLC coated rings film possesses better mechanical strength properties than the corre-
sponding TiC ones, with a higher elasticity modulus and yield stress. However, due to the insuf-
ficient substrate hardness, the film fatigue critical load is lower than 100 N, which is significantly
lower than the maximum fatigue critical load of approximately 800 N, in the case of the TiC
coated rotors and sleeves. The DLC coating fatigue endurance stress is assessed as low.
The adhesion strengths of the DLC coating on both rotor and sleeve were found to be signifi-
cantly improved and enhanced. Better adhesion strength was detected on the DLC coated rotor,
in comparison to the DLC film of the sleeve.
346 7th Coatings – 2008
1. K. -D. Bouzakis, N. Michailidis, G. Erkens, Thin hard coatings stress–strain curve determi-
nation through a FEM supported evaluation of nanoindentation test results, Surface and
Coatings Technology 142-144 (2001) 102-109.
2. K. –D. Bouzakis, N. Michailidis, S. Hadjiyiannis, G. Skordaris, G. Erkens, A continuous
FEM simulation of the nanoindentation to determine actual indenter tip geometries, material
elastoplastic deformation laws and universal hardness, Zeitschrift fuer Metallkunde 93
(2002) 9, 862-869.
3. K. -D. Bouzakis, N. Michailidis, S. Hadjiyiannis, G. Skordaris, G. Erkens, The effect of
specimen roughness and indenter tip geometry on the determination accuracy of thin hard
coatings stress–strain laws by nanoindentation, Materials Characterization 49 (2003) 149–
4. K. -D. Bouzakis, N. Michailidis, A. Lontos, A. Siganos, S. Hadjiyiannis, G. Giannopoulos, G.
Maliaris, G. Erkens, Characterization of Cohesion, Adhesion and Creep-Properties of Dy-
namically Loaded Coatings through the Impact Tester, Zeitschrift fuer Metallkunde 92
(2001) 10, 1180-1186.
5. K.-D. Bouzakis, A. Asimakopoulos, G. Skordaris, E. Pavlidou, G. Erkens, The inclined im-
pact test: A novel method for the quantification of the adhesion properties of PVD films,
Wear 262 (2007) 1471–1478.
Coatings on Machine Elements and Machine Components 347
348 7th Coatings – 2008