Content uploaded by Xinmin Li
Author content
All content in this area was uploaded by Xinmin Li on Jun 22, 2016
Content may be subject to copyright.
A pin-on-disc study of the tribology characteristics of sintered versus
standard steel gear materials
Xinmin Li
n
, Mario Sosa, Ulf Olofsson
Department of Machine Design, Royal Institute of Technology (KTH), SE 100 44 Stockholm, Sweden
article info
Article history:
Received 21 July 2014
Received in revised form
11 December 2014
Accepted 20 January 2015
Available online 4 February 2015
Keywords:
Gear material
Powder metallurgy
Wear
Friction
Damage mechanism
abstract
Though powder metallurgy (PM) allows manufacturing of complex components, including gears, we lack
knowledge of the tribological performance of PM versus standard steel gear materials. Using a pin-on-
disc machine, we simulate the sliding part of gear tooth contact in boundary and mixed lubricated
regions, comparing the tribological characteristics of two sintered gear materials with those of a
standard gear material. The comparison considered damage mechanisms, wear, and friction between
these materials in different configurations (i.e., standard versus standard, sintered versus sintered, and
sintered versus standard). The results indicate that, for pairings of the same gear materials, i.e., RS–RS
(16MnCr5), AQ–AQ (Distaloy AQþ0.2% C), and Mo–Mo (Astaloy 85Mo þ0.2% C), RS has a lower friction
coefficient. For PM and RS combinations, both PM pins have lower friction coefficients with RS disc
material than do RS pins with PM disc materials. For the wear coefficient, at low and high speeds, RS pins
always display better wear resistance than do AQ or Mo pins because of their high hardness and
compacted microstructure. For RS–PM combinations, Mo pins display higher wear resistance than do AQ
pins because their larger and more numerous pores enable good lubrication. Pins in the Mo–RS
combination displayed the highest wear resistance, mainly because the pores in Mo discs hold lubricant,
lubricating the contact surface and preventing adhesive wear. For the RS pin in the Mo–RS combination
and the AQ pin in RS–AQ, the damage mechanism is slight adhesive wear and scuffing. For pins in the
PM–PM, RS–PM, AQ–RS, and RS–RS combinations, the damage mechanism is a heavier scuffing-type
adhesive wear.
&2015 Elsevier B.V. All rights reserved.
1. Introduction
Powder metallurgy (PM) is usually used in manufacturing parts
with complex geometries, such as gears and structural parts.
The main attraction of PM is the high rate of material utilization,
environmentally friendly production, economically especially for
complex geometries and possibility of obtaining lighter compo-
nents. PM steel gears do not behave like wrought steel gears in
terms of fatigue and other mechanical properties, mainly reason
lies in the pores remaining in the final product after PM gears are
compacted, sintered, and heat treated. Porosity greatly influences
the mechanical properties of sintered steels [1–3]. In addition,
porosity at or near the surface influences the coefficients of friction
and wear. Sonsino [4] found that the fatigue properties of PM
components are inferior to those of wrought or cast parts.
Cristofolin [5] focused on the effect of material variables (i.e.,
porosity and micro-hardness) and geometric parameters (i.e.,
contact length) on wear resistance in dry rolling–sliding contact.
He built a knowledge base usable in the design step to take
account of the effects of material variables (porosity and micro-
hardness) and geometric parameters on wear resistance. Porosity
influences PM materials by reducing the load-bearing surface and
increasing the local plastic deformation, generating debris, and
entrapping wear debris [6–11 ], all of which influence the wear
behavior of PM materials. Govindarajan [12] investigated the
lubricated rolling–sliding contact fatigue damage mechanisms of
sintered steel material and found surface cracks opening up at
higher rolling–sliding contact ratios. Straffelini [13], who claimed
that the damage mechanisms of PM materials were controlled by
large- and small-scale plastic deformations, also used a model to
identify the damage mechanisms of PM materials under rolling–
sliding contact. Cristofolini [14] used dry rolling–sliding contact to
analyze the wear mechanisms of PM materials, characterizing
porosity under different manufacturing parameters to propose a
design methodology for determining the mechanical properties of
PM steel materials. Hoffman [15] found that changing from open
to closed porosity in sintered steel significantly influences the
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/wear
Wear
http://dx.doi.org/10.1016/j.wear.2015.01.032
0043-1648/&2015 Elsevier B.V. All rights reserved.
n
Corresponding author. Tel: þ46 701918177; fax: þ46 87909137.
E-mail address: xinmin@kth.se (X. Li).
Wear 340-341 (2015) 31–40
endurance limit, which affects the development of new, high-load
structural parts such as planetary gears and camshaft lobes.
We lack knowledge of certain basic tribological characteristics of
PM steel materials in gear applications, especially for standard gear
steel in contact with PM gear steel. In this paper, we systematically
compare tribological aspects (i.e., coefficient of friction, wear coeffi-
cient, and damage mechanisms) of both wrought and PM gear steels
using pin-on-disc experiments.
2. Experimental set-up
2.1. Pin-on-disc machine
A pin-on-disc machine was used to simulate the sliding part of
the gear contact. A schematic of the pin-on-disc machine is shown
in Fig. 1. The pin-on-disc machine comprised a horizontal rotating
disc and a calibrated dead-weight-loaded pin. In the present testing,
the tip of the pin was in the form of a non-rotating half sphere.
The ambient environment was nearly the same for each test
(see Table 1 for the test conditions). The friction force and vertical
displacement of pin were automatically measured in the pin-on-disc
machine using a load cell and a linear variable differential transfor-
mer (LVDT).
The experimental conditions are presented in Table 1. For all
experiments, the normal load on the pin was the same and was set
to 7 N. Before the experiment, all specimens were cleaned in an
ultrasonic bath, first rinsed in heptane, then rinsed in methanol,
and finally dried in an oven. A syringe with a brush continuously
applied the filtered ambient temperature lubricant with the aid of
a pneumatic system.
2.2. The test specimens
The pin and disc specimens were manufactured from three
gear materials: one standard gear steel and two PM materials. The
chemical compositions of the materials are presented in Table 2.
RS, AQ, and Mo will be used in the remainder of this paper to identify
the three tested materials: 16MnCr5 (RS), Distaloy AQþ0.2% C (AQ),
and Astaloy 85Moþ0.2% C (Mo). Both AQ and Mo specimens with
density of 7200 kg=m
3
were sintered for 45 min at 1120 1Cin90%
nitrogen and 10% hydrogen atmosphere. The mechanical parameters
of the three materials are presented in Table 3. The standard gear
steel, 16MnCr5, is a well-known wrought steel with easily identified
elastic properties. For the sintered steels, the situation is more
complicated. These materials are porous, and how they are manu-
factured determines their porosity and thereby their density and
mechanical properties. According to Beiss [16],theelasticproperties
of PM steel materials can be calculated using the following formulas:
E¼E
o
ðρ=ρ
o
Þ
3:4
ð1Þ
ν¼ðρ=ρ
o
Þ
1:6
1ν
o
ðÞ1ð2Þ
List of symbols
PM Powder metallurgy
RS 16MnCr5
AQ Distaloy AQþ0.2% C
Mo Astaloy 85Moþ0.2% C
A–B The former is the disc material, the latter is the pin
material
μ
Friction coefficient
KWear coefficient m
3
/N m
rRadius of the pin tip mm
P
max
Maximum contact pressure Gpa
E
1;2
Young's modulus Pa
ν
1;2
Poisson's ratios
ρ;ρ
o
Density kg/m
3
η
Dynamic viscosity Pa
h
o
Minimum film thickness m
R
0
Reduced radius of curvature m
E
0
Reduced Young's modulus Pa
NNormal load N
vSliding speed m/s
λ
Film parameter
R
qB
;R
qA
RMS roughness m
VVolume loss of the pin m
3
hHeight of the worn part of the pin tip m
dDiameter of the wear scar m
SSliding distance m
Fig. 1. Schematic of the pin-on-disc machine.
Table 1
Experimental conditions.
Temperature 19721C
Relative humidity 4075%
Normal load 7 N
Radius of the pin tip (r)5 mm
Disc diameter 200 mm
Table 2
Chemical composition (mass %) of RS, AQ, and Mo gear materials.
Ni Mn Fe C S P Si Cr Mo
RS –1–1.3 96.95–98.78 0.14–0.19 r0.035 r0.025 0.4 1.1 –
AQ 0.5 0.5 98.8 0.2 –––––
Mo –– 98.95 0.2 ––––0.85
Table 3
Characteristics of the specimens.
Specimen materials RS AQ Mo
Young's modulus (GPa) 210 154 154
Poisson's ratio 0.3 0.28 0.28
Surface hardness (pin) (HV) 945 782 804
RMS surface roughness (disc) (mm) 0.0598 0.0606 0.0685
RMS surface roughness (pin) (mm) 0.097 0.0897 0.0923
X. Li et al. / Wear 340-341 (2015) 31–4032
where ρ
o
,E
o
,andv
o
are the density, Young's modulus, and Poisson's
ratio, respectively, of solid steel. Fig. 2 presents a photograph of the
disc and pin samples used in the present experiments. All the
specimens were case hardened.
The lubricant used was BP Castrol Syntrans 75W-80. Table 4
presents detailed specifications of the lubricant. To simulate the
sliding part of the contact correctly, the surface roughness of the
disc and pin was close to that prescribed for gears in DIN quality
level 5 [17].
2.3. Contact conditions and test procedure
The tests aimed to simulate the sliding tooth contact conditions
of an FZG gear test. In this case, the normal load (7 N) was set to
produce a moderate load that created maximum contact pressures
of 900 MPa, 800 MPa, and 720 MPa for the RS–RS, RS–PM/PM–RS,
and PM–PM material combinations, respectively. The maximum
contact pressure can be calculated according to Stachowiak [18]:
P
max
¼3N
2πa
2
ð3Þ
where ais radius of the contact area:
a¼3NR
0
E
0
1
3
ð4Þ
where R
0
and E
0
are the reduced radius of curvature and the
reduced Young's modulus, respectively. R
0
and E
0
are defined as:
1
R
0
¼1
R
X
þ1
R
Y
ð5Þ
1
E
0
¼1
2
1v
2
1
E
1
þ1v
2
2
E
2
ð6Þ
where E
1;2
is the Young's modulus and v
1;2
is Poisson's ratio for the
pin and disc, respectively. R
X
and R
Y
are the radii of curvature in X
and Ydirection, as for this pin on disc experiment R
1X
¼R
1Y
¼
510
3
m; R
2X
¼R
2Y
¼1m.
1
R
X
¼1
R
1X
þ1
R
2X
ð7Þ
1
R
Y
¼1
R
1Y
þ1
R
2Y
ð8Þ
The material characteristics of the specimens are shown in Table 3.
Using the formula for calculating the minimum film thickness
in elastohydrodynamic contacts derived by Hamrock and Dowson
[19], the minimum film thickness ðh
o
Þfor a circular contact can be
calculated:
h
0
R
0
¼3:63 vη
E
0
R
0
0:68
αE
0
0:49
N
E
0
R
02
0:073
ð1e
0:68k
Þð9Þ
where vis the sliding speed and ɳis the dynamic viscosity of the
lubricant at 40 1C. kis the elliptical parameter and in circular
contact it is 1. The detailed characteristics of the test lubricant are
presented in Table 4. By varying the sliding velocity, we can obtain
different minimum film thicknesses. The general relationship between
the friction coefficient and the film parameter (
λ
, i.e., the ratio of the
film thickness and the RMS value of the roughness amplitude) is
represented by the Stribeck curve (Fig. 3). The film parameter is
calculated as:
λ¼h
0
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
R
qA2
þR
qB2
qð10Þ
where h
o
is the lubricant film thickness and R
qB
and R
qA
are the RMS
roughness of the two surfaces in contact, respectively.
The A–Bformulation will be used in the remainder of this paper to
identify the materials of disc and pin, the former letter identifying the
disc material and the latter the pin material. The test conditions are
shown in Tab le 5.
The various test conditions are also indicated in Fig. 3.
2.4. Test procedure
A total of 6 discs and 36 pins, as shown in Fig. 2, were used for
the experiments in the test matrix presented in Table 6.
Fig. 2. Pin and disc specimens.
Table 4
Test lubricant.
Lubricant name BP Castrol Syntrans 75W-80
SAE viscosity grade 64
Kinematic viscosity at 40 1Cðmm
2
=sÞ52.3
Kinematic viscosity at 100 1Cðmm
2
=sÞ9.2
Density at 15 1C (kg=m
3
)837
Pressure–viscosity coefficient, αðmm
2
=NÞ1:79 10
8
Dynamic viscosity, η(Pa) 42:710
3
Fig. 3. Coefficient of friction, μ, versus film parameter, λ, in the lubricated sliding of
metals. BL¼boundary lubrication, ML ¼mixed lubrication, (E)HL¼elastohydrody-
namic lubrication. Blue and red dots indicate test conditions. (For interpretation of
the references to color in this figure legend, the reader is referred to the web
version of this article.)
Table 5
The test conditions.
Test
conditions
Maximum contact pressure
P
max
(MPa)
Film
parameter λ
Sliding velocity
(m/s)
1 (RS–RS) 900 0.5242 0.5
2 (RS–RS) 900 2.1558 4
3 (RS–PM) 800 0.5331 0.5
4 (RS–PM) 800 2.1922 4
5 (PM–PM) 720 0.5436 0.5
6 (PM–PM) 720 2.2354 4
X. Li et al. / Wear 340-341 (2015) 31–40 33
Each test was performed twice to ensure repeatability. Each
test comprises three steps: the sliding distance was 600 m in the
first step, 1800 m in the second, and 360 0 m in the third. The
surface topography of the test disc was measured between each
step using a stylus instrument. The stylus measurement results
will be used in a separate running-in study and will not be further
discussed here.
2.5. Calculation of friction and wear coefficients
The friction coefficient was calculated from the measured
friction force divided by the normal load exerted on the pin by
the testing machine's dead weight.
The wear scar diameter was measured using optical microscopy
in the ordinate and abscissa directions at the end of each test. The
volume removed from the pin surface can then be calculated using
the following method. The main parameters used in calculating
the volume loss are shown in Fig. 4.
The volume loss, V, of the pin, which is half-spherical in shape,
is calculated by:
V¼πh
2
rh
3
ð11Þ
where dis the diameter of the wear scar, ris the radius of the half-
sphere, and his the height of the wear volume, which can be
calculated as:
h¼rr
2
d
2
2
!
1
2
ð12Þ
Archard's wear model is used here:
V
S¼KN ð13Þ
where Vis the volume loss in m
3
on the contact surface, Sis the
sliding distance in m, Kis the specific wear coefficient ðm
2
=NÞ, and
Nis the normal load in N. Reformulating this gives:
K¼V
NS ð14Þ
Note that the wear coefficient used in this study is the mean of
that determined in each of three steps of the test procedure.
2.6. Surface measurement and material studies
After the test, the tested specimens were measured using a
Taylor Hobson Form Talysurf to determine the surface roughness
of the contacting part of the pin specimen. The specimens were
also examined using SEM to obtain micrographs for studies of the
damage mechanisms. An optical measuring microscope (Nikon
model MM-60/L3) was used to acquire optical micrographs. An
optical microscope was also used to measure the diameter of the
worn surfaces of the pins and for porosity studies of the PM
materials. A micro-hardness tester (Matsuzawa MMT-7) was used
to determine hardness profiles of the pin specimens.
3. Results
3.1. Friction coefficient results
To obtain an overview of the test results, we first present the
mean values of all the tests and the deviation will be shown in
Section 4. Mean values of two repeats were calculated at each step
of the test procedure, that is, for the first 60 0 m, then 1800 m, and
finally 3600 m of sliding. The friction coefficient results of tests 1–
13, 13 –24, and 25–36 are shown in Fig. 5. The disc material is
regular steel (RS), the PM gear steel referred to as AQ and the PM
gear steel referred to as Mo. The test material combinations and
speeds are presented in Table 6. From Fig. 5 it is obvious that the
friction coefficients during the high-speed (i.e., 4 m/s) tests were
consistently lower than those during the low-speed tests (i.e.,
0.5 m/s) regardless of the material combination. Note the very low
friction coefficient for the AQ pin in contact with the RS or Mo disc.
3.2. Wear coefficient results
The wear coefficient results are presented in Fig. 6, which shows
the mean value and standard deviation for each test condition. Most
of the low-sliding-speed tests indicate a higher wear coefficient
than do the tests conducted at the higher sliding speed. Obviously,
regardless of the disc material, the RS material always has the
lowest wear coefficient at both sliding speeds. Fig. 8 (right) shows
that the Mo–RS combination has the lowest wear coefficient at both
sliding speeds.
3.3. Surface analysis
Figs. 7–9present 3D surface roughness results for the contact-
ing part of all pin specimens after testing; the area of the image is
0.10.1 mm
2
. We can identify several phenomena here, such as
pores in the PM pin shown in Fig. 7b, c, and f and in Fig. 8b and c. In
addition, parallel traces found in most images in Figs. 7–9indicate
an adhesive wear mechanism such as scuffing.
Fig. 10 shows SEM micrographs of RS pins. Note that it is
problematic to analyze PM pins in an SEM because the lubricant
residueinthepinporescancontaminateanddestroytheequipment;
Table 6
Pin-on-disc test matrix.
Test no. Disc material Pin material Sliding speed (m/s) Maximum contact
pressure (MPa)
1–2 RS RS 0.5 900
3–4 RS RS 4 900
5–6 RS AQ 0.5 800
7–8 RS AQ 4 800
9–10 RS Mo 0.5 800
11–12 RS Mo 4 800
13–14 AQ RS 0.5 800
15–16 AQ RS 4 720
17–18 AQ AQ 0.5 720
19–20 AQ AQ 4 72 0
21–22 AQ Mo 0.5 720
23–24 AQ Mo 4 720
25–26 Mo RS 0.5 800
27–28 Mo RS 4 800
29–30 Mo AQ 0.5 720
31–32 Mo AQ 4 720
33–34 Mo Mo 0.5 720
35–36 Mo Mo 4 720
Fig. 4. The wear scar.
X. Li et al. / Wear 340-341 (2015) 31–4034
nevertheless, one PM pin of AQ material was analyzed in an SEM as
shown in Fig. 11. The parallel traces visible in the SEM micrographs
indicatethesamedamagemechanism,i.e.,scuffing, as shown by the
3D surface topography measurements. Also visible in the SEM pictures
are pieces of material transferred to the surface, indicating adhesive
wear, as shown in Figs. 10cand11 .
The hardness measurements were made near the tip of the
new pin specimens (HV 0.1). The hardness–depth curve shown in
Fig. 12 indicates that the RS material is harder than either the AQ
or Mo material. In addition, the Mo material is nearly as hard as is
the AQ material above a depth of 0.5 mm
Fig. 13 shows the porosity structure of AQ and Mo pins. Note
that the Mo material has larger and more numerous pores both on
the surface and in the bulk material. Differences in porosity on or
near the surface may influence the friction and wear coefficients.
4. Discussion
4.1. Friction results and mechanisms
The lambda value (
λ
)showninFig. 3 measures the severity of the
asperity contact during lubricated sliding. Between these regimes,
when 1o
λ
o3, is a mixed lubrication regime under which the load
iscarriedpartlybytheasperitiesandpartlybythelubricatingoil.In
the test run at the high sliding speed of 4 m/s, the
λ
value was
approximately 2.2, indicating a mixed lubrication regime. At a low
speed of 0.5 m/s, a boundary lubrication regime dominated at
λ
¼0.5.
This difference in lubrication regime explains the results presented in
Fig. 5,wherethefrictioncoefficient was always lower in the high-
speed tests conducted in the mixed lubrication regime than in the
low-speedtestsconductedinaboundary lubrication regime.
Fig. 5. Friction coefficient results (mean value) of tests 1–36.
Fig. 6. Wear coefficient (mean value and standard deviation) results for different material combinations.
Fig. 7. 3D images of 0.10.1 mm
2
areas of worn surfaces of pins when disc material is RS. (a) RS pin, 0.5 m/s. (b) Mo pin, 0.5 m/s. (c) AQ pin, 0.5 m/s. (d) RS pin, 4 m/s.
(e) Mo pin, 4 m/s. (f) AQ pin, 4 m/s.
X. Li et al. / Wear 340-341 (2015) 31–40 35
Fig. 5 show that the friction coefficient tended to increase in
most of the tests. In boundary lubrication, all the loads are carried
by asperity contact. At first, an oxide layer exists, so the friction
coefficient is somewhat lower; this oxide layer eventually wears
away, leading to more metal-to-metal contact and a more adhesive
friction mechanism, possibly resulting in a higher friction coeffi-
cient. In the meantime, as wear is developing, considerable debris
[6–11] may enter into the pin–disc contact. The presence of debris
between the contact surfaces somewhat enhances the surface
roughness in an indirect way, giving a transiently lower value of
λ
.
Fig. 14 shows the friction coefficient results for RS–RS, AQ–AQ,
and Mo–Mo material combinations. Under boundary lubrication
conditions, the PM material obviously has a higher friction coeffi-
cient. This is because the rougher surfaces of the PM–PM combina-
tion and debris of PM pins can also increase the friction coefficient.
As shown in Fig. 8b, the Mo–Mo combination results in a rougher
surface than does the RS–RS material combination (Fig. 7a). Fig. 9c
shows 3D images of the AQ–AQ combination. The pins in the RS–RS
and AQ–AQ combinations have nearly the same surface roughness;
however, the AQ pin wears more and this may lead to more debris,
contributing to a higher friction coefficient than that of the RS–RS
combination. Asperity contact carries most of the load in boundary
lubrication, so the surfaces of pins of all materials are easily worn in
this regime. In addition, the RS material is harder than the PM
material, and higher hardness means higher wear resistance. The
PM pins are easily worn and may produce more wear particles.
Further research is necessary to quantify the differences in wear
debris between PM and RS material and their influence on the
friction coefficient.
In mixed lubrication, the trend of the friction coefficient is not
so obvious. Both PM and RS pins have nearly the same degree of
surface smoothness.
Fig. 15 shows the friction coefficient results of testing combina-
tions of RS and PM gear materials. Note that both combinations are
Fig. 8. 3D images of 0.1 0.1 mm
2
areas of worn surfaces of pins when disc material is Mo. (a) RS pin, 0.5 m/s. (b) Mo pin, 0.5 m/s. (c) AQ pin, 0.5 m/s. (d) RS pin, 4 m/s.
(e) Mo pin, 4 m/s. (f) AQ pin, 4 m/s.
Fig. 9. 3D images of 0.10.1 mm
2
areas of worn surfaces of pins when disc material is AQ. (a) RS pin, 0.5 m/s. (b) Mo pin, 0.5 m/s. (c) AQ pin, 0.5 m/s. (d) RS pin, 4 m/s.
(e) Mo pin, 4 m/s. (f) AQ pin, 4 m/s.
X. Li et al. / Wear 340-341 (2015) 31–4036
shown in each diagram, that is results for PM materials as pin and
regular steel materials RS as disc and the opposite RS as pin and
PM gear materials as disc is compared in each diagram. Each
diagram presents the mean and standard deviation of the friction
coefficient for each step. The results indicate that the AQ–RS and
Mo–RS material combinations always have higher friction coeffi-
cients than do the RS–AQ and RS–Mo combinations, respectively,
mainly because of the unique microstructure of PM materials. As
the sliding continues, the increased pin temperature may have
several effects, i.e., the pin's mechanical properties will change, its
rate of oxidation will increase, and phase transformation may
occur, all of which will influence the friction behavior. As shown in
Fig. 13, PM specimens contain many pores and have a specific
porosity structure. When the disc is rotating, lubricant constantly
enters and exits the pores on the contact surface of PM pins,
improving the heat dissipation and preventing the mechanical
properties of PM specimens from changing. Furthermore, the
lubricant contained in the pores of PM pins can provide good
lubrication conditions. RS specimens, however, do not dissipate
heat as well as PM specimens do, and the higher temperature
somewhat worsens the friction properties of RS pins. PM pins
therefore always have lower friction coefficients than do RS pins.
Here further research is necessary to generalize these results to
studies on gear test samples for gear efficiency studies.
Fig. 16 compares the friction coefficient results of AQ and Mo
pins, the disc materials being RS and Mo. With RS and Mo discs
(see Fig. 16), AQ pins always have lower friction coefficients than
do Mo material pins at both high and low speeds. Given the
porosity structure shown in Fig. 13, it is obvious that Mo pins have
more and larger pores in the bulk material than do the AQ pins. In
that case, Mo pins can easily have rougher surfaces after running
for some time, leading to a higher coefficient of friction. With the
AQ disc (see Fig. 5 c) there is no obvious trend, AQ and Mo pins
give nearly the same friction results.
As shown in Fig. 5, regardless of the disc material, the AQ pin
specimens have the lowest friction coefficients in most material
combinations in both boundary and mixed lubrication regimes.
This may be because of its porosity structure, which may promote
heat dissipation and good lubrication. This is a meaningful finding
for gear transmission efficiency.
4.2. Wear results and damage mechanism
4.2.1. Wear results
Fig. 17 shows the wear coefficient results for all material combi-
nations. Among pairings of the same materials at both low and high
speeds (see Fig. 17a), pins in the RS–RS combination always display
Fig. 10. SEM micrographs of RS pins: (a) RS–RS 0.5 m/s, (b) AQ–RS 0.5 m/s, (c) Mo–RS 0.5 m/s, (d) RS–RS 4 m/s, (e) AQ–RS 4 m/s, and (f) Mo–RS 4 m/s.
Fig. 11. SEM micrograph of the AQ pin in the RS–AQ combination at 4 m/s.
Fig. 12. Hardness–depth curves for new pin specimens (HV 0.1) of the three
materials.
X. Li et al. / Wear 340-341 (2015) 31–40 37
better wear resistance than do pins in the AQ–AQ and Mo–Mo
combinations. Bergseth et al. [20] studied wear in a pin-on-disc
machine, obtaining wear coefficients of the same orders as those
presented here.
Given the hardness–depth curve shown in Fig. 12, we see that
the RS material is harder than both AQ and Mo materials, which
have similar near-surface hardness profiles. Higher hardness is
beneficial for wear resistance. The compacted microstructure also
contributes to the lower wear coefficient of the RS–RS material
combination. Given the porosity structure shown in Fig. 13, the AQ
and Mo materials clearly contain many pores. The resulting
somewhat more open microstructure of the AQ and Mo materials
make them more easily worn than the RS material. In addition, the
pores in the AQ and Mo materials influence the roughness of the
Fig. 13. Porosity structure of PM pins: (a) Mo and (b) AQ.
Fig. 14. Friction coefficient results (mean value and standard deviation) for same material combinations at different speeds.
Fig. 15. Friction coefficient (mean value and standard deviation) results for RS and PM material combinations.
Fig. 16. Friction coefficient results (mean value and standard deviation) of AQ and Mo pins.
X. Li et al. / Wear 340-341 (2015) 31–4038
contact surfaces, which are case hardened and ground in all speci-
mens. After reasonable sliding distances, the originally ground and
somewhat smooth surfaces become worn. The pores make the
surface rougher (Figs. 7aand8b show that the Mo pin has a rougher
surface than does the RS pin), providing more opportunities for
asperity contact. Direct asperity contact can enhance the wear
coefficient. AQ–AQ and Mo–Mo material combinations have nearly
the same wear coefficient. The higher wear resistance of RS material
comes from its higher hardness and compacted microstructure. The
higher wear coefficient of PM materials results from their more
open microstructure, lower hardness, and rougher contact surfaces
due to the presence of pores. In AQ–AQ and Mo–Mo combinations,
pins of AQ and Mo have nearly the same wear coefficient.
From Fig. 17 we can compare the wear coefficients of RS pins
when the disc materials are RS, AQ, and Mo. The pins in the AQ–RS
and Mo–RS combinations have successively lower wear coeffi-
cients than that of the RS–RS combination. The main reason for
this is that the pores in AQ and Mo discs contain lubricant that
can lubricate the contact surface well, preventing adhesive wear.
Different material contact can prevent adhesive wear. Mo disc
produced the lowest wear coefficient in an RS pin, largely because
the larger and more numerous pores in the Mo disc held more
lubricant, better lubricating the contact surfaces. PM–RS combina-
tions therefore merit examination in future gear research; in
particular, the Mo–RS combination displays good potential.
Fig. 17 shows that most mixed lubrication conditions result in
higher wear resistance than do boundary lubrication conditions.
As shown in Table 5, low-sliding-speed experiments have a
λ
value
of approximately 0.51 (in the boundary lubrication region) while
high-sliding-speed experiments have a
λ
value of approximately
2.2 (in the mixed lubrication region). All loads are carried by
asperity contact in boundary lubrication, but in the mixed lubrica-
tion region, parts of the load are carried by the lubricant film.
Here, the low sliding speed leads to a boundary lubrication regime
that causes more severe wear.
4.2.2. Damage mechanism
Wear refers to the loss of material from a contact surface.
Depending on the specific wear mechanism, wear is of four types:
abrasive wear, adhesive wear, surface fatigue, and corrosion wear.
Abrasive wear refers to the damage to contact surfaces caused by
hard particles between the contact surfaces. These hard particles
may come from the contact components or the outside environ-
ment. The degree of abrasive wear increases with increasing relative
speed, load, and the hardness of the hard particles. Adhesive wear is
material transfer caused by partial melt of the contact surface at high
temperatures. Fatigue wear refers to micro-cracks caused by contact
stress concentration and plastic deformation due to the long-term
effects of cyclic loads. As these micro-cracks expand and intersect,
peeling happens. Corrosion wear refers to the loss of surface material
due to chemical or electrochemical reactions between the contacting
surface material and the surrounding medium.
From the wear results shown in Fig. 17, it is clear that the Mo–
RS material combination has the highest wear resistance. Fig. 8a
and d show no obvious grooves on the pins' surfaces, so no heavy
scuffing occurred due to adhesive wear. However, 3D images of other
material combinations, for example, Fig. 7d and f and Fig. 8c, show
deep parallel grooves that provide evidence of heavy adhesive wear,
such as scuffing. Carefully examining Fig. 10c and f reveals adhesive
wear occurring under boundary lubrication conditions. The shallow
grooves shown in Fig. 10cindicateslightscuffing-type adhesive wear
under boundary lubrication conditions: under these conditions, the
load is completely carried by asperity contact, which provides great
opportunities for scuffing. In mixed lubrication, as shown in Fig. 10f,
only slight scuffing occurs. Fig. 17 shows that pins in the AQ–RS
combination have somewhat higher wear resistance than do pins in
the RS–RS combination. As for the damage mechanism, the deep
grooves shown in Fig. 9aandd(alsoFig. 10bande)indicatethat
scuffing has occurred. There are several possible explanations why,
when the disc and pin materials are PM and RS, respectively, the RS
pins always display higher wear resistance. First, the RS material is
harder than the PM materials, as shown in Fig. 12. Second, different
material combinations can effectively prevent the occurrence of
adhesive wear. Finally, the microstructure of PM materials includes
pores full of lubricant between pin and disc contact surfaces;
lubricant from these pores can lubricate the contact surfaces, helping
prevent adhesive wear.
As for the mechanism of damage to the AQ pin in the RS–AQ
combination, the SEM micrograph of the AQ pin in Fig. 11 shows
signs of slight scuffing-type adhesive wear.
When it comes to the PM–PM, RS–PM, AQ–RS, and RS–RS
material combinations, the damage mechanism differs completely
from that in the Mo–RS combinations. Figs. 7–9clearly show
heavy scuffing on the surfaces of the pins in the PM–PM, RS–PM,
AQ–RS, and RS–RS combinations. Fig. 10 shows SEM micrographs
of RS pins, in which it is clear that there are scuffing-type adhesive
wear and material transfer.
5. Conclusion
The sliding part of gear tooth contact was simulated in pin-on-
disc experiments. The friction properties, wear resistance, and
damage mechanisms of two PM steel gear materials and a regular
steel gear material were investigated and compared. The following
conclusions could be drawn from the study:
The friction and wear coefficients in the high-speed (4 m/s)
tests are consistently lower than in the low-speed tests (0.5 m/
s) regardless of the material combination. This is probably due
to a mixed lubrication regime in the high-speed tests and a
boundary lubrication regime in the low-speed tests.
The AQ–RS (AQ disc and RS pin) and Mo–RS (Mo disc and RS
pin) combinations always have higher friction coefficients than
do the RS–AQ (RS disc and AQ pin) and RS–Mo (RS disc and Mo
Fig. 17. Wear coefficient results: (a) same materials in pins and discs, (b) RS–AQ, and (c) RS–Mo.
X. Li et al. / Wear 340-341 (2015) 31–40 39
pin) combinations. The main reason for this is the more open
microstructure of the PM materials, whose pores can contain
lubricant that improves lubrication and heat dissipation.
When the disc material is RS or Mo, AQ pins always have lower
friction coefficients than do Mo pins at both high and low speeds.
The RS–AQ combination is a potential candidate for improving the
gear transmission efficiency.
At both high and low speeds, RS pins always display better
wear resistance than do AQ and Mo pins because of both their
high hardness and compacted microstructure.
In RS–PM combinations, Mo pins display higher wear resis-
tance than do AQ pins because the microstructure of Mo can
provide more lubricant, enabling better lubrication and heat
dissipation than in AQ pins.
Pins in the Mo–RS combination displayed the highest wear
resistance of any pins in all gear material combinations. This
was mainly because the pores in Mo discs contain lubricant
that can lubricate the contact surfaces and prevent adhesive
wear. The Mo–RS combination could usefully be applied in gear
transmissions.
The surfaces of RS pins in the Mo–RS combination and of AQ
pins in the RS–AQ combination show some indications of
adhesive wear. For the pins in the PM–PM, RS–PM, AQ–RS,
and RS–RS combinations, the damage mechanism is more
severe adhesive wear, such as scuffing.
Acknowledgments
Thanks are expressed to Dr. Michael Andersson from Höganäs
AB for his guidance and technological support. This project was
financially supported by Höganäs AB.
References
[1] G. Hoffmann, C.M. Sonsino, K. Michaelis, Rolling Contact Fatigue –Component
Design and Testing for P/M Applications, Society of Automotive Engineers Inc.,
Marquette university, 1999.
[2] N.A. Fleck, R.A. Smith, Powder Metall. 3 (1981) 121–125.
[3] G. Bocchini, Int. J. Powder Metall. 22 (1986) 185–202.
[4] C.M. Sonsino, Powder Metall. 33 (1990) 235–245.
[5] M. Cristofolini, G. Pilla, A. Straffelini, A. Molinari, Design guidelines for PM
parts subject to dry rolling–sliding wear, Powder Metall. 56 (2) (2013)
124–134 .
[6] A.I. Taskinen, M.H. Tikkanen, Mod. Dev. Powder Metall. 17 (1984) 545–567.
[7] H. Khorsand, S.M. Habibi, H. Yoozbashizadea, K. Janghorban, S.M.S. Reihani,
H. Rahmani Seraji, M. Ashtari, Mater. Des. 23 (2002) 667–670.
[8] E.R. Leheup, D. Zhang, J.R. Moon, Wear 176 (1994) 111–119 .
[9] G. Straffelini, A. Molinari, Powder Metall. 44 (4) (20 01) 344–350.
[10] B. Dubrujeaud, M. Vardavoulias, M. Jeandin, Wear 174 (1995) 155–161.
[11] G. Straffelini, A. Molinari, Powder Metall. 44 (3) (2001) 248–252.
[12] N. Govindarajan, R. Gnanamoorthy, Study of damage mechanisms and failure
analysis of sintered and hardened steels under rolling–sliding contact condi-
tions, Mater. Sci. Eng.: A 445–446 (2007) 259–268.
[13] G. Straffelini, T. Marcu Puscas, A. Molinari, Identification of rolling–sliding
damage mechanisms in porous alloys, Metall. Mater. Trans. A 31 (12) (200 0)
3091–3099.
[14] I. Cristofolini, A. Molinari, G. Straffelini, P.V. Muterlle, A systematic approach to
design against wear for Powder Metallurgy (PM) steel parts: the case of dry
rolling–sliding wear, Mater. Des. 32 (4) (2011) 2191–2198.
[15] G. Hoffman, K. Lipp, K. Michaelis, C.M. Sonsino, J.A. Rice, Testing P/M materials
foe high loading gear applications, Int. J. Powder Metall. 35 (6) (1999) 100.
[16] P.,Beiss Mechanische Eigenschaften von Sinterstählen. Tagungsband zum
Symposium für Pulvermetallurgie: Material-Prozess-Anwendung Hagen. Vol.
27. 2003.
[17] ISO 1328-2:1997(E).
[18] Gwidon Stachowiak etc., Engineering Tribology (third ed.), 298–300.
[19] B.J. Hamrock, D. Dowson, Ball Bearing Lubrication, The Elastohydrodynamics
of Elliptical Contacts, John Willey and Sons, Inc., New York, 1981.
[20] E. Bergseth, M. Torbacke, U. Olofsson, Wear in environmentally adapted
lubricants with AW technology, J. Synth. Lubr. 25 (4) (2008) 137–158 .
X. Li et al. / Wear 340-341 (2015) 31–4040
