Artigo Dos Italianos

EFA-03032; No of Pages 17
Engineering Failure Analysis xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Engineering Failure Analysis

journal homepage: www.elsevier.com/locate/engfailanal
Experimental analysis of contact fatigue damage in case hardened gears for

off-highway axles
A. Terrin a,b, C. Dengo b, G. Meneghetti a,⁎
a
Department of Industrial Engineering, University of Padova, Via Venezia, 1, 35131 Padova, Italy
b
Engineering and Innovation Department, Carraro S.p.A., Via Olmo 37, 35010 Campodarsego, Padova, Italy
a r t i c l e i n f o a b s t r a c t
Article history: Pitting is one of the main causes of failure in planetary gear sets of axles designed for agricul-
Received 28 August 2016 tural vehicles. The high torque and the low wheel speed typical of such machines result in poor
Received in revised form 29 December 2016 lubrication and promote the onset of contact fatigue failure by pitting, which generally occurs
Accepted 3 January 2017
earlier in the sun gear than in both the planets and the ring of wheel hub planetary drives. In
Available online xxxx
fact, sun gears are subjected to the highest contact pressure and, at the same time,
unfavourable rolling-sliding working conditions. In this paper, six case hardened sun gears
Keywords: damaged by pitting during endurance tests were analysed. The aim of the analysis was to high-
Gear contact fatigue
light the key aspects of the morphology and the evolution of pitting damage on the case hard-
Pitting
ened sun gears.
Micropitting
Macropitting © 2017 Elsevier Ltd. All rights reserved.
Shot peening
1. Introduction
Pitting is a fatigue phenomenon that affects the mating surfaces of mechanical elements characterized by non-conformal con-
tact. In components such as cams, bearings and gears for instance, contacts occur between curved surfaces with different curva-
ture radii, thus the load is carried by a very small contact area, resulting in high pressure. The damaging process starts with the
formation of cracks caused by the repeated stress cycles generated by contact between the mating parts. Cracks grow at small
angles to the surface and, eventually, curve up causing the detachment of material debris and leaving craters. Fatigue cracks
may initiate either at the surface or at a small depth below the surface, where shear stresses are high enough to promote their
propagation from defects or inclusions. Subsurface cracks are more common in well-lubricated pure rolling contacts between
smooth surfaces and in presence of non-metallic inclusions [1–5]. On the contrary, when both rolling and sliding occur, surface
cracks are more prone to form because friction forces contribute to increase the shear stress and to reduce the depth at which
it achieves the maximum value [6]. Moreover, if sliding occurs, energy dissipated by friction is converted to heat, which promotes
pitting damage since it reduces the lubricant viscosity and thus the load carrying capacity of the film [7,8].
Gears are typical components operating in rolling-sliding contact conditions. Contact pressure on gears is usually calculated by
assuming that gear teeth can be treated as cylinders with parallel axes (Fig. 1a). According to Hertz's theory of contact between
elastic bodies [9,10], all normal stress components are compressive and therefore they are not likely to promote crack propaga-
tion. Conversely, the damaging process is driven by the cyclic shear stress [11,12], which achieves the maximum value at a certain
depth under the surface (Fig. 1b, c).
⁎ Corresponding author.
E-mail address: giovanni.meneghetti@unipd.it (G. Meneghetti).
http://dx.doi.org/10.1016/j.engfailanal.2017.01.019
1350-6307/© 2017 Elsevier Ltd. All rights reserved.
Please cite this article as: A. Terrin, et al., Experimental analysis of contact fatigue damage in case hardened gears for off-highway
axles, Engineering Failure Analysis (2017), http://dx.doi.org/10.1016/j.engfailanal.2017.01.019
2 A. Terrin et al. / Engineering Failure Analysis xxx (2017) xxx–xxx
Fig. 1. Parallel cylinders resembling tooth contact (a). Contour plot of shear stress τzx below the surface arising from cylinder-cylinder contact (b). Stress field at a
depth 0.5b, where b indicates the half-width of the contact strip under load (c). Stress values have been normalized by the maximum contact pressure p0.
Hertzian contact fatigue phenomena can be classified into two major failure modes: pitting and micropitting. Although the
morphology of craters is essentially the same for the two modes, they differ for the size. Pitting produces craters from about
0.5 to 1 mm, that are also called macropits. Conversely, micropits' size is on the order of tens of microns, therefore micropitting
is visible to naked eye only when several pits coalesce forming a wide matt surface [13].
Referring to case hardened gears, pitting may be originated by subsurface cracks only in presence of high rolling speeds, very
smooth surfaces and optimal lubricating conditions [4,14–16]. In Such cases, contact between asperities is prevented by the lubri-
cant and the formation of subsurface cracks in correspondence of defects or inclusions [5] may occur earlier than surface originat-
ed damages. However, the lubrication film is not often thick enough to keep roughness asperities separated, so that surface cracks
form as a result of plastic strains in the surface due to friction. The propagation of surface cracks is promoted by the presence of
the lubricant, which is entrapped and pressurized within the crack edges by the contact with the mating surface [16–21]. Surface
originated pits can be distinguished in two categories: (i) Point Surface Origin (PSO) macropitting and (ii) Geometric Stress Con-
centration (GSC) macropitting [22]. Some examples of damage mechanisms are shown in Figs. 2 to 5. Generally, PSO macropits
originate from surface defects such as nicks, dents, grinding furrows, debris bruises and also previous-formed micropits [23,24].
Fig. 2. PSO macropitting.
A. Terrin et al. / Engineering Failure Analysis xxx (2017) xxx–xxx 3
Fig. 3. GSC macropitting.
Typically, a crack nucleates from the surface defect and then propagates with a shallow angle (b30° to the surface) towards the
interior of the tooth. When the crack reaches a critical length, a flake of material detaches leaving a crater with a steep wall,
where the ultimate fracture occurs. Fig. 2 reports an example of PSO macropitting. GSC pitting forms in areas where the non-con-
formity of the mating surfaces is particularly pronounced and therefore the load is carried by a small contact area resulting in ex-
tremely high pressures. In such cases several pits form in the overloaded areas, leading to a widespread damage as reported in Fig.
3. Any geometrical deviation from the reference model consisting of two equivalent cylinders with parallel axis of Fig. 1a may
cause a raise of the contact pressure. Typical causes of GSC macropitting on gears are misalignments due to deflection of shafts
and supports under load, corners of tip and root relieves, crowning, and tip-to-root interference.
Micropitting (Fig. 4) is particularly common in case hardened gears subjected to high loads and poor lubricating conditions
[9,25,26]. During running-in, friction and metal to metal contacts cause cumulative plastic deformation with flattening of the as-
perities. As a consequence, a mirror-like surface with low frequency waviness is obtained [11]. Further application of cyclic stress-
es leads to the initiation of cracks at the surface (surface distress) [20]. Since the damaged surface appears dull, micropitting has
been indicated in the past also with the name “frosting” or “grey staining”. Generally, cracks form in areas subjected to stress con-
centration due to high non conformity of the mating surfaces [22] and may represent favourable initiation spots for subsequent
macropits formation [14,16]. However, the main parameter affecting micropitting is the rolling speed, which strongly influences
the ratio between the lubricant film thickness and the surface roughness. The Elasto-Hydro-Dynamic (EHD) film indeed collapses
at low rolling speed, making slow gears more apt to fail by surface fatigue phenomena [7,13].
Fig. 5 shows a sun gear where the three mechanisms of damage described above occurred in different areas of the tooth flank.
GSC macropitting manifested near the root of the tooth probably due to tip-to-root interference. Micropitting is spread in the
dedendum, particularly at the centre of the tooth where the curvature due to crowning concentrates the contact pressure. Finally
PSO macropitting manifested in the addendum, probably because of the presence of debris originated by the above damages.
In this work, contact fatigue failures of six sun gears subject to Four Square endurance Test (FST) have been investigated. All
analysed gears belong to planetary final drives placed on a wheel hub of axles for off-highway vehicles. The study is part of a
more extensive project concerning the fatigue characterization of gears with reference to pitting as well as bending failures
[27]. The aim of the analysis was to highlight the key aspects of the morphology and of the evolution of pitting on case hardened
Fig. 4. Micropitting in the dedendum of a gear tooth.
Fig. 5. Different damaging modes on the tooth of a case hardened 17NiCroMo6-4 sun gear of a planetary gear set mounted in an axle for off-highway applications.
sun gears. The paper is organised as follows: first the sun gears under analysis are described; after that, the FST set-up and the
relevant experimental results are reported; finally, the observed pitting damage on the sun gears is presented.
2. Material and components under analysis
In a typical off-highway axle reported in Fig. 6, pitting generally occurs first on the sun gears of the planetary gear sets
mounted in the wheel hubs, causing noise, vibrations, loss of efficiency and, eventually, acting as initiation spot for tooth breakage.
The analysed product was a steering axle used in vehicles for agriculture with gross engine power up to 63 kW. Pitting is a fre-
quent failure mode in transmissions for agricultural applications because of the high operating torque and low speed, which result
in poor lubricating conditions. Despite the low speed of rotation, sun gears are characterized by a high rate of contacts due to the
meshing with three or four planet gears and therefore are particularly critical with respect to contact fatigue phenomena; more-
over, the maximum contact pressure along the line of action occurs when the dedendum of the sun gear meshes with the adden-
dum of the planet gear, the former being subjected to unfavourable sliding conditions, as explained later in Section 4.
The main geometrical and material data of the analysed planetary gear set are summarized in Table 1. Six sun gears made of
case hardened 17NiCrMo6-4 steel were analysed; three of them were subjected to shot peening after grinding with an Almen in-
tensity of 20A and shot diameter of 0.7 mm, while the remaining three ones were tested in the un-peened condition. In order to
improve the gear performances, teeth were crowned and relieves were applied to both tip and root of the teeth. The results of
hardness measurements and the microhardness profiles are reported in Table 2 and Fig. 7 respectively. All the sun gears were
characterized by similar microhardness profiles, regardless their surface treatment (shot peened or un-peened). No substantial
differences of the microstructure were noted by means of optical microscope observations (see Fig. 8): all of them presented a
fine martensite structure near the tooth surface, with traces of retained austenite and upper bainite (plate form) and some
lower bainite formations (acicular form). At the core, the microstructure was composed of low-carbon martensite with ferritic
islands.
Fig. 6. Sketch of a steering axle for off-highway vehicles.
Table 1
Data of the analysed planetary gear set.
Sun Planet Ring
Pressure angle, α [deg] 25

Module, m [mm] 2.75
Number of teeth, z 12 17 48
Centre distance [mm] 41.5
Case hardening depth [mm]a 0.6–1.1 0.6–1.1 0.6–0.8
Surface hardness H.R.C. 58–63 58–63 57–61
Core hardness H.R.C. 35–45 36–44 31–42
Residual austenite (% max) 20
a
Distance from surface to point where hardness is 550 HV.
3. Four Square Test set-up and results
The Four Square Test (FST) is useful to investigate the structural durability of gears and bearings of axles; a sketch of the test
set-up is reported in Fig. 9. Two axles are positioned with their input shafts facing each other, and mechanically connected by
means of a cardan shaft driven by an electric motor. Two chains connect a pair of sprockets fixed to the wheel hubs of the
axles. The system is statically preloaded using a wrench to impose a given applied torque. Torque, speed and lubricant tempera-
ture are measured during the test, lubricant cooling being provided by fans when the temperature set point is exceeded. Only one
axle is investigated (the one subjected to power input through the pinion), while the auxiliary axle closes the kinematic chain by
recovering the power exiting from the axle under test. Accordingly, FST test benches are addressed as “power-recirculating” test
rigs. To run the system, the electric motor needs only to compensate for friction losses, therefore small size motors can be
adopted, resulting into limited energy consumption.
Three different load cases can be applied during the test, as reported in Table 3. Every 100 h (or every time a failure is detect-
ed) the two axles are disassembled and each part is carefully inspected to ascertain possible damages. Depending on the percent-
age of worn surface observed, a pitting level is assigned to the sun gears by visual inspection, according to Table 4.
Table 5 shows the FST results relevant to pitting observed on sun gears of fourteen different axles for off-highway vehicles
[28]. All gears were case hardened and some of them were subjected to an additional shot peening treatment. Applied torque
was different for each test, since different axle sizes were included in the analyses. For each sun gear the maximum contact pres-
sure acting during the contact path was calculated by means of a dedicated contact analysis performed with the commercial soft-
ware KISSsoft. The analysis accounted for the effects of tooth microgeometry as well as for system compliances on the pressure
distribution over the tooth flank.
For each sun gear characterized by a pitting level 0 or 1 (see Table 4) and which was tested under constant torque in forward
direction, the relevant stress-life point was imported in a S-N chart. Therefore, sun gears belonging to FST 3976 of Table 5 were
not included in the chart. A similar failure criterion is adopted also by ISO 6336 to define the design fatigue curves for case hard-
ened steels [29], where gears are considered to be failed when the damaged area is at least 0.5% of the total working area or 4% of
the working area of a single tooth. Fig. 10 compares all available experimental data with the fatigue curves reported in ISO stan-
dard for 99% survival probability [30].
Some Authors claim a possible improvement in contact fatigue as a consequence of compressive residual stresses induced by
shot peening [12,31–34]. On the other hand, ISO 6336 Standard suggests that the benefits are effective only on bending fatigue at
the tooth root. By considering the sun gears analysed in the present paper, no significant improvement in contact fatigue strength
was noted due to shot peening. However, a more detailed analysis of the results reported in Fig. 10 and Table 5 indicated that a
7% increment in load carrying capacity can be estimated for case hardened gears after shot peening [28].
In the present work, the result of FST #3976 were analysed and the testing conditions they were subjected to are reported in
Table 6. Three shot peened and three un-peened gears were tested using two different lubricants, respectively, having the follow-
ing kinematic viscosities: concerning lubricant 1, ν40 = 60 mm2/s and ν100 = 9.1 mm2/s; concerning to lubricant 2, ν40 =
135 mm2/s and ν100 = 14.5 mm2/s . The FST was run with the following settings: continuous torque: 4320 Nm, peak torque:
9000 Nm and maximum permitted temperature during the FST: 82 °C ± 2 °C. Since the tests were terminated at different pitting
Table 2
Hardness measurements on the considered sun gears.
Sun gear # Core hardness Case hardening depth Surface hardness

HRC [mm] HRC
1 41.0 0.7 58.2

2 41.6 0.7 60.1
3 41.8 0.6 62.9
4 43.1 0.6 61.9
5 43.1 0.7 60.7
6 43.6 0.8 60.2
Fig. 7. Microhardness profiles of the analysed sun gears.
levels, this group of results is the most meaningful to investigate the morphology and evolution of contact fatigue in sun gears of
off highway axles. Therefore FST 3976 was extracted from Table 5 for further analysis.
4. Damage analysis
A summary of the damaged surfaces of the sun gears observed at the end of the FST 3976 are reported in Fig. 11 (the reader
should also refer to Table 6 to identify the gears). Some common features of the failure mechanisms were observed:
1. The initial effect of cyclic contacts is the formation of mirror-like surfaces;

2. Pitting was found to be more severe at the centre of the contact band (Fig. 12) because crowning is applied on tooth flanks in
order to avoid edge contacts as a consequence of gear and shaft deflections or misalignments.
3. With the exception of gear #4, pitting damage is more severe in the dedendum area of the tooth;
4. Pitting damage occurs first in the sun gears of the tested planetary gear sets;
5. Areas where sliding speed is small or null (near to or in correspondence of the pitch line) exhibit a lower amount of damage.
Such behaviours are interpreted according to the following considerations.

Due to the low operating speed of off-highway vehicles lubricant does not form an EHD film thick enough to prevent interac-
tion between asperities. In fact, the pitch line velocity of the sun gears during the four square test is 0.27 m/s, which is fairly
below the minimum speed of 2 m/s required to apply the methods for EHD film thickness calculation provided by ISO technical
report [8]. Therefore, during running-in the crests of roughness of all the observed teeth were flattened creating a polished surface
(Fig. 13).
Friction forces due to the sliding of the counter surface cause plastic deformation of the external layers, leading to the initiation
of surface micro cracks, which propagate towards the pitch line for driver gears, while they propagate away from the pitch line for
driven gears, as shown in Fig. 14 [22]. The direction of friction forces is referred in the literature as “sliding direction”, while the
“rolling direction” indicates the direction of motion of the contact point along the tooth profile, as reported in Fig. 14. The ratio
between sliding and rolling speed is called Slide-to-Roll Ratio (SRR), which can be better explained with reference to a pair of
cylinders in sliding/rolling contact (Fig. 15). The rolling velocity is given by the average between the tangential speeds of the sur-
faces, and its direction is defined by the motion of the point of contact on the surfaces. Sliding and rolling speeds have the same
direction for the faster surface (positive SRR) and opposite direction for the slower surface (negative SRR). Considering again the
gear meshing problem, Fig. 14 reports the sliding speed directions of both the driver and the driven gears. Regions with negative
SRR are more apt to fail by pitting because friction forces stretch the top layers before the passage of the contact. Therefore, cracks
Fig. 8. Microstructure near the tooth flank of sun gear #1 (a) and at the core of sun gear #6 (b). Etchant: Nital 4%.
Fig. 9. Four square test (FST) setup.
are opened allowing some oil to be trapped by the subsequent transit of the mating surface. The hydraulic pressure exerted by the
oil trapped inside the crack promotes crack propagation by mode I. Moreover, since friction between crack faces is reduced by the
presence of lubricant, also mode II growth driven by the cyclic variation of shear stresses can occur. Conversely, in the areas
where SRR is positive, cracks are closed by compression of the top layers before the passage of the contact and thus have a
lower tendency to propagate assisted by the hydraulic pressure cycles [7,17,35–37]. In the dedendum of both driven and driver
gears, sliding speed is in opposite direction with respect to rolling speed (see Fig. 14). Nevertheless, the dedendum area of is
more critical in sun gears than in planet gears, since the rate of meshing is higher in the former case and the maximum contact
pressure along the line of action occurs when the dedendum of the sun gear meshes whit the addendum of the planet gear.
Concerning the latter phenomenon, it should be recalled that according to the Hertz's theory of contact between elastic solids,
the nominal contact pressure between two mating teeth can be calculated by:
sffiffiffiffiffiffiffiffiffiffiffiffiffiffi
E0 F
σ H0 ¼ ð1Þ
πρred b
Fig. 10. Comparison between experimental data (pitting level 0 or 1 according to Table 4) and design curves according to ISO Standard 6336. Pressure values were
calculated by a tooth contact analysis with the commercial software KISSsoft.
Fig. 11. Damaged surfaces of the analysed sun gears.
Fig. 12-. GSC Macropitting observed on sun gear #3 of Fig. 10. The damage is due to both the small curvature radius at the dedendum of the tooth and the
crowning, which reduces the contact area to an ellipse narrower than the actual face width.
where F is the normal force acting on the tooth flanks, b is the face width, E′ is the reduced Young's modulus (E′ ≈ 113,190 MPa
when both gears are made of steel) and ρred is the reduced curvature radius, which is function of the curvature radii ρ1 ,2 of the
mating flanks in the point of contact:
1
ρred ¼ ð2Þ
1 1
þ
ρ1 ρ2
The contact pressure along the tooth profile of the analysed gears is reported in Fig. 16. Start of Active Profile (SAP) is the
point (near the root circle) where the contact starts, LPSTC (Lowest Point of Single Tooth pair Contact) is the point where the
force starts to be transmitted by only one pair of teeth, PP is the Pitch Point, HPSTC (Highest Point of Single Tooth pair Contact)
is the last point where the force is transmitted by only one pair of teeth and End of Active Profile (EAP) is the point where the
contact ends. Since both sun and planets gears have tip and root relieves the pressure was calculated assuming a linear increase of
the force between SAP and LPSTC, according to ISO-TR 15144 [6]. Under such hypothesis, the abrupt change that the nominal con-
tact pressure would exhibit in correspondence of the LPSTC is prevented, according to Fig. 16. When only a couple of teeth carries
the whole load (i.e. between the LPSTC and the HPSTC), Fig. 16 shows that the point corresponding to the LPSTC of the sun gear
experiences the highest value of pressure, because there exists the lowest value of ρred according to expression (2). Bearing in
mind the mechanisms described above, Fig. 16 shows that the dedendum area of sun gear #1 is more damaged than the
Fig. 13. Polished area due to flattening of asperities. The detail shows the transition between the mirror-like surface and the part of the tooth that shows the orig-
inal grinding furrows in sun gear #6.
Fig. 14. Direction of rolling and friction forces in a gear meshing.
addendum and that the crack starting from the surface of the dedendum propagated in the opposite direction of sliding speed.
Planet gears of the same final drive showed no signs of pitting.
Contact conditions near the tooth tip may also be severe, as illustrated in the following. Sun gear #4 exceptionally showed a
line of micropitting near the tooth tip (see Fig. 11), which may have been caused by a geometric stress concentration due to the
reduction of the curvature radii at the edge of the tip relief. Pitting in this region was observed in sun gears #3 and #6 (see Fig.
17) and it was probably originated from scuffing scratches [22], which are generated in areas characterized by high sliding speed
such as the tooth tips, particularly when low viscosity lubricants are used [7,13,38]. In fact, scuffing was more severe in the gears
tested with lubricant 1, which is characterized by a lower viscosity as compared to lubricant 2. Micropitting at tooth tips may also
occur because of tip-to-root interferences illustrated in Fig. 18. Due to teeth deflections under load, contact between two incoming
teeth starts before the tips of driven gear intersect the line of action, and thus the tip corner tends to furrow the dedendum of
driver gear. Due to the edge contact with the tip of the mating gear, GSC micropitting may occur in the form of a damaged
line on the tooth dedendum (usually first on the sun gear because of the higher rate of contacts) [23]. Despite the application
of tip and root relieves in the design of the tooth form in order to reduce the risk of interferences under load, an example of
such damage mechanism can be seen in sun gear #6 (see Figs.11 and 19).
In the region between the pitch line and the HPSTC, pitting is generally less pronounced and often starts as PSO macropitting,
as it can be seen in sun gears #3 and #6 (Figs. 11 and 20). Such craters are thought to origin in correspondence of dents caused
by debris, as reported in Fig. 21. PSO craters are typically on the order of 50–100 μm deep with diameters of 0.5–1 mm.
Fig. 22a shows a section of a macropit along with a magnification of the surface of the intergranular fracture that led to the
formation of the crater. In areas where several pits coalesce to form a damaged area, the depth of the craters may be higher,
with secondary branches of the cracks extending hundreds of microns below the surface, as shown by Fig. 22b.
Fig. 15. Definition of SRR for two cylinders in sliding/rolling contact.
Fig. 16. Contact pressure under continuous torque calculated according to ISO-TR 15144 [6] and Slide-to-Roll Ratio along the tooth profile. The microscopic obser-
vation of surface cracks are relevant to sun gear #1 (Driver).
The most common modifications observed in the microstructure as a consequence of contact fatigue are the White Etching
Areas (WEAs) reported in Fig. 23, which were found in gears #3, 4, 5 and 6. The name derives from the white appearance of
the altered microstructure after attack with metallographic etching reagent such as nital (nitric acid in ethanol) [39]. The structure
and formation mechanisms of WEAs are still debated. Some authors suggested that WEAs consist of highly deformed martensite
generated in correspondence of cracks or hard inclusions. During contact, the compression stress field forces the edges of subsur-
face cracks to close. The cyclic shear stresses would cause severe plastic deformation of the edges with the development of high
temperature, which may eventually lead to local re-austenitization of the material and carbide dissolution. The subsequent cooling
would lead to the formation of martensite supersaturated in carbon [40]. According to different authors, heating up to high tem-
peratures is not likely to occur during real in-service contact fatigue and they reasoned that severe deformation of the carbides
would be the cause of their dissolution [11]. More recent works postulate that WEAs are composed of equiaxed nano ferrite grains
generated as a result of recrystallization of the highly deformed steel matrix at the edge of cracks [39,41–45].
5. Residual stress measurements
Residual stresses acting parallel to the involute profile were measured in depth close to the pitch diameter of sun gear #5
(shot peened) and #6 (un-peened) by means of X-ray diffraction. The material was removed by electropolishing in a restricted
area to minimize the perturbation effects on the measurement, which were performed on both the drive and non-driving flanks
Fig. 17. Scuffing on the tooth tip of sun gear #3.
Fig. 18. Schematic representation of tip-to-root interference. Continuous lines show the theoretical position of gears, while dashed lines show the actual position of
gears due to teeth deflections under load.
Fig. 19. SEM image of micropitting observed in sun gear #6.
of the teeth, in order to investigate to which extent contact fatigue phenomena may alter original residual stresses. Results are
reported in Fig. 24. At the surface of the non-driving flank, the un-peened gear showed compressive residual stresses deriving
from carburizing process up to −430 MPa. At depths N 50 μm the residual stress field stabilized at approximately −200 MPa.
The shot peening treatment introduced higher compressive residual stresses, which achieves the maximum value of
Fig. 20. SEM image of a macropit observed in sun gear #6.
Fig. 21. Debris dent on sun gear #1 (a) and scratches in sun gear #3 (b) between the pitch line and the HPSTC.
Fig. 22. Section of a macropit (a) and crack branching (b) in sun gear #3.
Fig. 23. White etching areas near the surface of sun gear #3 (a) and in correspondence of an inclusion in sun gear #4 (b). Etchant: Nital 4%.
Fig. 24. Residual stresses under the surface of gear #5 (shot peened) and #6 (un-peened).
Table 3
Duty cycle of the FST.
Load case Wheel torque Wheel speed [rpm] Duration [h] Running direction
1 Tca 30 80–400c Forward

2 Tca 30 80 Reverse
3 Tpb 10 2 Forward
a
Continuous torque: Tc is the design value of wheel torque for the tested axle.
b
Peak torque: Tp is the maximum expected torque.
c
Set on the basis of the endurance requirements of the specific application.
−750 MPa at about 30 μm below the surface and then decreases with depth. As compared to the un-peened residual stress pro-
file, Fig. 24 shows that shot-peening resulted particularly effective up to a depth of approximately 150 μm , as reported also in
[31]. According to refs [31,32], after cyclic loading a relaxation of residual stresses would be expected, possibly because of mar-
tensite decay due to a stress induced phase transformation [46,47]. Fig. 24 shows that for un-peened gears, cyclic contact stresses
increase the residual stresses significantly in a 150-μm-deep layer below the surface. However, in the shot peened sun gear resid-
ual compressive stresses increased approximately 25% in the external layer, while no significant variations at depths N 30 μm were
found after cyclic contact loading. The increase of residual stresses after cyclic contact loading may be due to the increase in vol-
ume associated with the transformation of retained austenite into martensite [46,47] or to inhomogeneous plastic deformations
[46].
6. Conclusions
Six sun gears damaged by contact fatigue during a four square test were analysed in order to investigate the morphology and
evolution of pitting phenomena in case hardened gears. The gears belonged to planetary final drives of off highway axles, which
are particularly critical because of the high contact pressure and poor lubricating conditions they are subjected to during in-ser-
vice life. The outcome of the analysis suggests that:
Table 4
Pitting levels (determined by visual inspection).
Pitting levela Percentage of damaged area
0 0–5%
1 5%
2 10%
3 25%
4 50%
5 100%
a
When also spalling occurs, letter “s” is added to the pitting level.
Table 5
Summary of FST results.
FST Sun gear Contact stressa Pitting level Load cycles Elapsed timeb Shot peened Normal module Number of teeth Lubricant viscosityc
6
# # [MPa] [×10 ] [h] [mm] Sun gear Planets [mm2/s]
3060 1 2615 0 3.20 100 Yes 3 13 31 144

2 2615 1 3.20 100
3 2615 1 3.20 100
3098 1 2495 0 5.39 169 Yes 3 13 31 66
3268 1 2073 1 3.65 100 No 2.9295 15 31 55
2 2139 0 8.33 228
3 2116 0 7.31 200
4 2139 0 9.94 272
5 2161 1 16.5 452
6 2116 0 8.04 220
7 2161 0 11.0 300
3218 1 2153 1 5.83 200 No 2.9295 15 31 100
2 2113 1 2.92 100
3 2115 0 2.92 100
3319 1 2153 1 8.75 300 No 2.9295 15 31 100
2 2198 0 11.7 400
3 2198 0 2.33 80
4 2198 0 2.33 80
5 2108 0 2.92 100
6 2153 0 5.83 200
7 2153 0 8.16 280
3353 1 2037 0 7.20 200 Yes 3 12 23 66
2 2080 0 14.4 400
3 2236 1 3.60 100
3436 1 2236 0 6.39 200 Yes 3 13 31 66
2 2236 0 6.39 200
3879 1 2211 1 8.64 400 No 2.75 12 17 56
3671 1 2518 0 7.20 200 Yes 3 12 23 66
2 2518 0 7.20 200
2057 1 2199 0 6.30 250 No 3 12 23 59
2 2222 1 9.37 372
2091 1 2225 0 6.63 263 Yes 3 12 23 59
2 2225 0 6.63 263
3820 1 2394 0 5.83 200 No 2.9295 15 31 56
2 2394 0 5.83 200
3568 1 2401 0 2.16 100 No 2.75 12 17 66
2 2401 0 4.32 200
3976 1 n.d.d 1 3.93 182 Yes 2.75 12 17 135
2 2042 5 4.32 200 Yes 60
3 2042 5 6.48 300 No 60
4 1926d 1 8.64 400 No 135
5 n.d.e 2–3 4.32 200 Yes 60
6 n.d.e 1 6.48 300 No 60
a
Calculated by KISSoft tooth contact analysis.
b
Unless otherwise specified the value is intended as elapsed time in forward direction with continuous torque.
c
At the temperature of 40 °C.
d
The gear experienced both continuous and peak torque.
e
Sun gear tested in the auxiliary axle. The contact stresses are lower than in the tested axle due to power losses in power recirculation.
- During running-in the tooth surface was polished by flattening of the asperities and formation of a mirror-like area on the
flank. This phase was accompanied by generation of surface micro-cracks which propagated at small angles to the surface,
in opposite direction with respect to friction forces. Crack growth occurred primarily in regions with negative slide-to-roll
(SRR) ratios, according to the relevant literature.
- Geometric Stress Concentration (GSC) macropitting manifested firstly at the Lowest Point of Single Tooth pair Contact (LPSTC)
due to the negative SRR and the small value of the reduced curvature radius, the latter reducing the size of the contact area
and thus increasing the contact pressure. Moreover, the most damaged areas were found in the centre of the tooth facewidth
because of crowning, which is a source of GSC macropitting.
- Point Surface Origin (PSO) macropits were also observed, which were probably originated from defects or dents due to the
presence of metal debris between the meshing gears.
- GSC micropitting was observed at edges of tip relieves and may be addressed also to tip-to-root interference.
- Scuffing occurred at tooth tips, these regions being characterized by high SRR. Scuffing was more evident in gears tested with
the lubricant characterized by the lower viscosity.
Table 6
Test specifications of the analysed gears.
Sun gear # Surface finish Oil Durationa [h] Pitting level Axle typeb
Load case 1 Load case 2 Load case 3
1 Grinding + shot peening Lubricant 2 100 80 2 1 Tested

2 Grinding + shot peening Lubricant 1 200 / / 5 Tested
3 Grinding Lubricant 1 300 / / 5 Tested
4 Grinding Lubricant 2 400 80 2 1 Tested
5 Grinding + shot peening Lubricant 1 200 / / 2–3 Auxiliary
6 Grinding Lubricant 1 300 / / 1 Auxiliary
a
Load cases defined according to Table 3.
b
See Fig. 9.
- The pitting life resulted improved thanks to the use of a lubricant with higher viscosity, despite the low rolling speeds involved
in off-highway applications, which prevent the application of technical standards to evaluate the elastohydrodynamic film
thickness.
- In the analysed tests, shot-peening did not provide improve significantly the contact fatigue strength of sun gears. Residual
stress measurements highlighted that residual stress profiles of the un-peened compared to the shot peened gears are
much more similar in the drive side of the tooth than in the non-driving side. That was due to the increase of residual stresses
of the un-peened teeth due to cyclic contact stresses.
Acknowledgements
The present work was financed by Carraro S.p.A. (25-2016 / 544) (Italy). Special thanks to Mr. Fabio Scarpa for his fruitful con-
tribution in the analysis with optical and SEM microscopes.
References
[1] G. Niemann, H. Winter, Maschinenelemente, Band 2 (Machine Elements), Springer, Berlin, 1985.
[2] F. Sadeghi, B. Jalalahmadi, T.S. Slack, N. Raje, N.K. Arakere, A review of rolling contact fatigue, J. Tribol. 131 (2009)http://dx.doi.org/10.1115/1.3209132.
[3] D.J. Wulpi, Understanding How Components Fail, third ed. ASM International, 2013.
[4] S. Glodež, J. Flašker, Z. Ren, A new model for the numerical determination of pitting resistance of gear teeth flanks, Fatigue & Fract. Eng. Mater. Struct. 20 (2007)
71–83, http://dx.doi.org/10.1046/j.1460-2695.2003.00711.x.
[5] S. Netpu, P. Srichandr, Failure of a helical gear in a power plant, Eng. Fail. Anal. 32 (2013) 81–90http://dx.doi.org/10.1016/j.engfailanal.2013.03.002.
[6] P.J.L. Fernandes, C. McDuling, Surface contact fatigue failures in gears, Eng. Fail. Anal. 4 (1997) 99–107, http://dx.doi.org/10.1016/S1350-6307(97)00006-X.
[7] G. Stachowiak, A.W. Batchelor, Engineering Tribology, third ed. Butterworth-Heinemann, 2013.
[8] ISO/TR 15144-1:2014 - Calculation of Micropitting Load Capacity of Cylindrical Spur and Helical Gears - Part 1: Introduction and Basic Principles.
[9] K.L. Johnson, Contact Mechanics, Cambridge University Press, 1987.
[10] A.P. Boresi, R.J. Schmidt, Advanced Mechanics of Materials, sixth ed. John Wiely & Sons Inc., 2003
[11] R.S. Hyde, Contact fatigue of hardened steels, ASM Handbook, Vol. 19, 2003 (Ohio, p. 1749).
[12] Y. Shen, S.M. Moghadam, F. Sadeghi, K. Paulson, R.W. Trice, Effect of retained austenite – compressive residual stresses on rolling contact fatigue life of carburized
AISI 8620 steel, Int. J. Fatigue 75 (2015) 135–144.
[13] AGMA 925-A03, Effect of Lubrication on Gear Surface Distress, 2003.
[14] C. Santus, M. Beghini, I. Bartilotta, M. Facchini, Surface and subsurface rolling contact fatigue characteristic depths and proposal of stress indexes, Int. J. Fatigue 45
(2012) 71–81, http://dx.doi.org/10.1016/j.ijfatigue.2012.06.012.
[15] G. Fajdiga, S. Glodež, J. Kramar, Pitting formation due to surface and subsurface initiated fatigue crack growth in contacting mechanical elements, Wear 262
(2007) 1217–1224, http://dx.doi.org/10.1016/j.wear.2006.11.016.
[16] G. Fajdiga, J. Flašker, S. Glodež, T.K. Hellen, Numerical modelling of micro-pitting of gear teeth flanks, Fatigue & Fract. Eng. Mater. Struct. 26 (2003) 1135–1143,
http://dx.doi.org/10.1046/j.1460-2695.2003.00711.x.
[17] S. Way, Pitting due to rolling contact, J. Appl. Mech. (1935) A49–A58.
[18] A.F. Bower, Influence of crack face friction and trapped fluid on surface initiated rolling contact fatigue cracks, J. Tribol. 110 (1988) 704–711.
[19] M. Dallago, M. Benedetti, S. Ancellotti, V. Fontanari, The role of lubricating fluid pressurization and entrapment on the path of inclined edge cracks originated
under rolling–sliding contact fatigue: numerical analyses vs. experimental evidences, Int. J. Fatigue (2016)http://dx.doi.org/10.1016/j.ijfatigue.2016.02.014.
[20] M. Boniardi, F. D'Errico, C. Tagliabue, Influence of carburizing and nitriding on failure of gears – a case study, Eng. Fail. Anal. 13 (2006) 312–339, http://dx.doi.org/
10.1016/j.engfailanal.2005.02.021.
[21] E. Bormetti, G. Donzella, A. Mazzù, Surface and subsurface cracks in rolling contact fatigue of hardened components, Tribol. Trans. 45 (2002) 274–283, http://dx.
doi.org/10.1080/10402000208982550.
[22] R. Errichello, Morphology of Micropitting, AGMA 11FTM17, Am. Gear Manuf. Assoc, 2011.
[23] R.L. Errichello, C. Hewette, R. Eckert, Point-surface-origin macropitting caused by geometric stress concentration, Gear Technology 28 (1) (2011) 54–59.
[24] V.M.M.B. da Mota, P.M.G.P. Moreira, L.A.A. Ferreira, A study on the effects of dented surfaces on rolling contact fatigue, Int. J. Fatigue 30 (2008) 1997–2008, http://
dx.doi.org/10.1016/j.ijfatigue.2008.01.003.
[25] J.R. Davis, Gear Materials, Properties, and ManufactureASM International 2005.
[26] S. Li, A. Kahraman, A micro-pitting model for spur gear contacts, Int. J. Fatigue 59 (2014) 224–233, http://dx.doi.org/10.1016/j.ijfatigue.2013.08.015.
[27] C. Dengo, G. Meneghetti, M. Dabalà, Experimental analysis of bending fatigue strength of plain and notched case-hardened gear steels, Int. J. Fatigue 80 (2015)
145–161, http://dx.doi.org/10.1016/j.ijfatigue.2015.04.015.
[28] C. Dengo, A. Terrin, G. Meneghetti, Design curves against contact fatigue for case hardened gears of off-highway drivelines, Proceedings of the 44th AIAS National
Conference, Messina (Italy), 2015 (in italian).
[29] ISO 6336-5:2016 - Calculation of Load Capacity of Spur and Helical Gears - Part 5: Strength and Quality of Materials.
[30] ISO 6336-2:2006 - Calculation of Load Capacity of Spur and Helical Gears – Part 2: Calculation of Surface Durability (Pitting).
[31] A.C. Batista, A.M. Dias, J.L. Lebrun, J.C. Le Flour, G. Inglebert, Contact fatigue of automotive gears: evolution and effects of residual stresses introduced by surface
treatments, Fatigue & Fract. Eng. Mater. Struct. 23 (2000) 217–228, http://dx.doi.org/10.1046/j.1460-2695.2000.00268.x.
[32] I. Fernández Pariente, M. Guagliano, Contact fatigue damage analysis of shot peened gears by means of X-ray measurements, Eng. Fail. Anal. 16 (2009) 964–971,
http://dx.doi.org/10.1016/j.engfailanal.2008.08.020.
[33] M. Guagliano, E. Riva, M. Guidetti, Contact fatigue failure analysis of shot-peened gears, Eng. Fail. Anal. 9 (2002) 147–158, http://dx.doi.org/10.1016/S1350-
6307(01)00002-4.
[34] Y. Gao, Influence of deep-nitriding and shot peening on rolling contact fatigue performance of 32Cr3MoVA steel, J. Mater. Eng. Perform. 17 (2007) 455–459,
http://dx.doi.org/10.1007/s11665-007-9155-7.
[35] S.P. Radzevich, Dudley's Handbook of Practical Gear Design and Manufacture, second ed. CRC Press, 2012.
[36] M.F. Frolish, D.I. Fletcher, J.H. Beynon, A quantitative model for predicting the morphology of surface initiated rolling contact fatigue cracks in back-up roll steels,
Fatigue & Fract. Eng. Mater. Struct. 25 (2002) 1073–1086, http://dx.doi.org/10.1046/j.1460-2695.2002.00601.x.
[37] A.V. Olver, L.K. Tiew, S. Medina, J.W. Choo, Direct observations of a micropit in an elastohydrodynamic contact, Wear 256 (2004) 168–175, http://dx.doi.org/10.
1016/S0043-1648(03)00374-0.
[38] G.D. Thakre, S.C. Sharma, S.P. Harsha, M.R. Tyagi, Tribological failure analysis of gear contacts of Exciter Sieve gear boxes, Eng. Fail. Anal. 36 (2014) 75–91http://dx.
doi.org/10.1016/j.engfailanal.2013.09.019.
[39] M.-H. Evans, White structure flaking (WSF) in wind turbine gearbox bearings: effects of “butterflies” and white etching cracks (WECs), Mater. Sci. Technol. 28
(2012) 3–22, http://dx.doi.org/10.1179/026708311X13135950699254.
[40] H. Schlicht, E. Schreiber, O. Zwirlein, Effects of material properties on bearing steel fatigue strength, ASTM Spec. Tech. Publ. STP 81–101 (1988).
[41] R. Errichello, R. Budny, R. Eckert, Investigations of bearing failures associated with white etching areas (WEAs) in wind turbine gearboxes, Tribol. Trans. 56 (2013)
1069–1076, http://dx.doi.org/10.1080/10402004.2013.823531.
[42] A. Grabulov, R. Petrov, H.W. Zandbergen, EBSD investigation of the crack initiation and TEM/FIB analyses of the microstructural changes around the cracks formed
under Rolling Contact Fatigue (RCF), Int. J. Fatigue 32 (2010) 576–583, http://dx.doi.org/10.1016/j.ijfatigue.2009.07.002.
[43] M.-H. Evans, White structure flaking (WSF) in wind turbine gearbox bearings: effects of ‘butterflies’ and white etching cracks (WECs), Mater. Sci. Technol. 28
(2012) 3–22, http://dx.doi.org/10.1179/026708311X13135950699254.
[44] T. Bruce, E. Rounding, H. Long, R.S. Dwyer-Joice, Characterisation of white etching crack damage in wind turbine gearbox bearings, Wear 338-339 (2015)
164–177, http://dx.doi.org/10.1016/j.wear.2015.06.008.
[45] M.-H. Evans, An updated review: white etching cracks (WECs) and axial cracks in wind turbine gearbox bearings, Mater. Sci. Technol. 32 (2016) 1133–1169,
http://dx.doi.org/10.1080/02670836.2015.1133022.
[46] R.C. Dommarco, K.J. Kozaczeck, P.C. Bastias, G.T. Hahn, C.A. Rubin, Residual stresses and retained austenite evolution in SAE 52100 steel under non-ideal rolling
contact loading, Wear 257 (2004) 1081–1088, http://dx.doi.org/10.1016/j.wear.2004.01.020.
[47] A. Voskamp, R. Ӧsterlund, P. Becker, O. Vingsbo, Gradual changes in residual stress and microstructure during contact fatigue in ball bearings, Met. Technol. 7
(1980) 14–21.

Artigo Dos Italianos

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Artigo Dos Italianos

Hochgeladen von

Copyright:

Verfügbare Formate

EFA-03032; No of Pages 17

Engineering Failure Analysis xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Engineering Failure Analysis

Experimental analysis of contact fatigue damage in case hardened gears for

Fig. 2. PSO macropitting.

Fig. 3. GSC macropitting.

Fig. 4. Micropitting in the dedendum of a gear tooth.

2. Material and components under analysis

Fig. 6. Sketch of a steering axle for off-highway vehicles.

Sun Planet Ring

Pressure angle, α [deg] 25

3. Four Square Test set-up and results

Sun gear # Core hardness Case hardening depth Surface hardness

1 41.0 0.7 58.2

Fig. 7. Microhardness proﬁles of the analysed sun gears.

1. The initial effect of cyclic contacts is the formation of mirror-like surfaces;

Such behaviours are interpreted according to the following considerations.

Fig. 9. Four square test (FST) setup.

Fig. 11. Damaged surfaces of the analysed sun gears.

Fig. 14. Direction of rolling and friction forces in a gear meshing.

Fig. 15. Deﬁnition of SRR for two cylinders in sliding/rolling contact.

5. Residual stress measurements

Fig. 17. Scufﬁng on the tooth tip of sun gear #3.

Fig. 19. SEM image of micropitting observed in sun gear #6.

Fig. 20. SEM image of a macropit observed in sun gear #6.

1 Tca 30 80–400c Forward

Pitting levela Percentage of damaged area

3060 1 2615 0 3.20 100 Yes 3 13 31 144

Load case 1 Load case 2 Load case 3

1 Grinding + shot peening Lubricant 2 100 80 2 1 Tested

Das könnte Ihnen auch gefallen