Original Article

Proc IMechE Part C:

J Mechanical Engineering Science
A numerical investigation on the hybrid 2022, Vol. 236(1) 3 54­–369
! IMechE 2020
spur gears: Stress and dynamic analysis Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/0954406220982007
journals.sagepub.com/home/pic

Tufan G Yılmaz1 , O� gan2 and Fatih Karpat1

guz Do�

Abstract
In this study, the effect of rim thickness of hybrid gears on the root stress, joint stress, tooth stiffness, natural frequency,
and dynamic behavior are examined numerically. Hybrid gears consist of two materials, which are steel for the teeth-rim
and hub regions of gear, carbon fiber reinforced plastic (CFRP) for the web region. Adhesive bonding is assumed for the
joining of steel and composite materials. FE method is used to evaluate tooth root stress, joint stress, tooth deforma-
tion, and the natural frequency of hybrid gears. The adhesive is defined by cohesive zone modeling (CZM). Moreover,
2-DOF dynamic analyses are implemented to obtain dynamic factors and static transmission error. According to results,
hybrid gears have substantial potential to reduce the mass of gear transmission systems with no adverse effect on root
stress and dynamic factor if the design parameters are appropriately selected. Besides, rim thickness is found as a critical
parameter for the hybrid gears since when its value changes from 0.5xm to 3xm, the root stress decreases 10% while the
tooth stiffness and torque capacity increase 20% and 65%, respectively.

Keywords
Hybrid spur gears, finite element stress analysis, cohesive zone modeling, dynamic analysis
Date received: 18 August 2020; accepted: 25 November 2020

Introduction
joining method that ensures the structural integrity of
According to an international report, the energy the gear under the loading conditions is found.5
demand will grow by 56% up to 2040 due to the Involute spur gears are decent instances for this
increasing population and, consequently, the type of machine elements. During the running time,
number of vehicles.1 Based on this critical point, involute spur gears are subjected to Hertzian stress on
researchers have focused on different ways to the contact region while the tensile and compressive
decrease fuel usage and CO2 emission rates. Among stresses on the root region of the tooth.6 Except for
these ways, reducing the structural mass is a highly these areas, the stress values are quite low, as illus-
preferred way since the weight of the vehicle is trated in Figure 1.
responsible for 75% of the total fuel consumption.2 Based on this situation, different materials can be
The studies reported that 6-8% of fuel savings could used for high and low-stress regions in the designing
be ensured a 10% reduction from the weight of the of the spur gear. Carbon fiber reinforced plastics
vehicle.3 Replacing the materials of components with (CFRP) and aluminum alloys are the candidate light-
a higher specific strength (have a high strength to weight materials for the low-stress region with their
density ratio) is the most convenient way to reduce adequate mechanical properties. Stiffness is another
total weight. This material shifting can be totally or significant parameter for spur gears as it affects the
partially for a component. As a total change instance, static transmission errors, which are the primary
source of gear whine and noise. CFRP materials are
the aluminum alloys that a type of low-density mate-
rials can be used in the manufacturing of body panels
or gearbox casings instead of steel since these compo- 1
Department of Mechanical Engineering, Bursa Uludag University,
nents are subjected to low external forces/stresses.4 Bursa, Turkey
Regarding power transmission machine elements,
2
Department of Mechanical Engineering, Kahramanmaraş Sütçü _Imam
University, Kahramanmaraş, Turkey
the high strength steel alloys are still the best option
Corresponding author:
with their mechanical, physical, and manufacturabil- Fatih Karpat, Department of Mechanical Engineering, Bursa Uludag
ity properties. However, partial material change can University, Bursa, Turkey.
be possible for this type of component provided that a Email: [email protected]
Yılmaz
2 et al. 355
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

debris found on the gear teeth.13 Catera et al. pro-

posed a modeling method for composite material to
obtain natural frequencies of hybrid gear by finite
element analysis (FEA). The method was validated
with previous experimental values from the litera-
ture.14 Catera et al. conducted a non-linear FEA
study to obtain stress values on gear teeth and adhe-
sive for a special designed hybrid gear under the
heavy loading condition. Besides, mesh stiffness and
STE are determined for ply to ply and homogeneous
composite representations. According to the results,
the obtained values of these two approaches are very
close. The proposed hybrid gear design ensures a
lower peak to the peak value of the STE when com-
paring lightweight steel gear for the same mass.15 In
this study differently from Catera’s research, the rim
thickness effect on root and adhesive stress is taken
into consideration. Contartese et al. proposed a novel
quick method to specify the effect of steel teeth and
Figure 1. Stress regions of a spur gear. composite body on the mesh stiffness of hybrid gears
by FEA. The spring elements were used instead of
gear teeth and body. The results show a good agree-
superior to aluminum alloys with their higher stiffness
ment with the full gear model.16 Catera et al. con-
as well as lower density and well damping character-
ducted a comparative numerical and experimental
istics. CFRP materials have been used in aerospace,
study to observe the effect of two different joining
automotive, and several industries as the material of
methods on the frequency response of hybrid gears.
structural components with its tailorable and custom-
According to results, adhesive bonding has better
izable mechanical properties for decades.7,8 They are
damping properties, while tooth stiffness is higher in
preferred to reduce noise and vibration in machine the interference fitting method. The numerical results
elements as well.9,10 Recently, CFRP materials have were verified with experimental ones.17 Karpat et al.
found application areas in involute gears as a hybrid compared the bimetallic gear and hybrid gears in
design material with steel alloys. However, there are terms of root stress, mesh stiffness, and weight for
still very few studies on hybrid gears. Initial impor- same ring rim thicknesses.18 Gauntt and Campell
tant studies were initiated within NASA Glenn investigated the effect of composite type and lay up
Research Center. Handschuh et al. generated and sequence on the natural frequency and mode shapes
tested a prototype lightweight hybrid gear. Adhesive of hybrid gear. According to results, the natural fre-
bonding was used to join steel and composite. They quency is highly dependent on elasticity modulus.19
conducted numerical and experimental studies to Gauntt et al. conducted an optimization study to
obtain free vibration modes of full steel and hybrid decrease weight and increase the joint strength of
gear for comparison aim. Besides, dynamics tests hybrid spur gears with a special algorithm. They pro-
were conducted for several torque and rpm values posed a sinusoidal geometry for the composite part as
to measure vibration and sound level for different it ensures the highest shear strength for adhesive
driven-driving gear configurations. An endurance bonding.20 Kim et al. tried to find the optimum
test was carried out at the level of 10,000 rpm and radial distance of the composite part between steel
300 Nm torque for 300000 cycles. According to the hub and steel teeth in terms of vibration and sound.
results, there was no visible fatigue damage. The FEM and BEM are used to force reaction and modal
hybrid gear has lower natural frequency values than analysis. According to results, the vibration levels
the full steel gear. In addition to a 20% weight reduc- reduce 11.9-33.1% while noise levels reduce 4.6-
tion, low vibration values were obtained at only high 43.2% when it is compared to full steel gear between
load and speed values for hybrid driving hybrid con- range 2000 to 3000 rpm.21 These studies are mostly
figuration.11,12 Based on these studies, It is thought focused on natural frequency, vibration, and sound
that if the rim thickness is selected properly, the stress situation of hybrid gears. There are no studies about
and fatigue damage will not be an important problem the effect of gear parameters on the performance of
for hybrid gears. LaBerge et al. investigated the the hybrid gear. Among these gear parameters, the
hybrid gear performance under no lubrication condi- rim thickness stands out as the most significant
tion. They stated that there is a potential risk factor parameter for weight reduction. For this reason, the
due to the excessive temperature increase, which root stress, joint stress, tooth deflection, and natural
results in damage interface of steel and composite. frequency are determined with FEA for different rim
With this increase, plastic flow occurred, and black thicknesses in this study. In addition, torque capacity
356 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
3

Figure 2. 3D model of hybrid gear for FEA.

of hybrid gears is specified for different rim thick-

Table 1. Gear properties.
nesses. Tooth stiffnesses and weights are obtained to
calculate mesh stiffness and rotational inertia. Based Design parameters Values
on these calculated values, the 2-DOF dynamic anal-
Module m (mm) 3
yses are conducted to find dynamic factors and STE. Number of teeth z 20
Pressure angle an 20�
Evaluation of tooth root and joint region Addendum ha (mm) 1�m
Dedendum hf (mm) 1.25 � m
stress of hybrid gears Tip radius of cutter pfp (mm) 0.47 � m
During the torque transmission, involute spur gears Profile shifting x 0
are subjected to tooth root stress and tooth deforma- Facewidth b (mm) 24
tion. Tooth root stress in the tensile side of the tooth Rim thickness (xm) 0.5-1-1.5-2-2.5-3-Solid
Hub thickness (xm) 1
is one of the primary and most significant reasons for
Shaft hole diameter (mm) 10
the tooth root crack.22 A 30% reduction in gear
Gear ratio i 1
stresses can lead to a 46% increase in the crack initi- Teeth-rim and hub material 16MnCr5
ation cycle.23 For this reason, tensile root stress Web material CFRP
should be examined in hybrid gears, as in standard
gears with a single material. There are a few analytical
methods to calculate the root stress of gear tooth, between consecutive lamina is determined as 30� in
such as AGMA, DIN3990.24 However, adequate, the following symmetric lay up: [0/30/60/90/-60/-
accurate results are not obtained when they applied 30]s. The mechanical constants of UD laminas15 are
on non-standard gears such as asymmetric, asymmet- presented in Table 2.
ric trochoid, or bimetallic gears. In literature, the Analytical equations are used to obtain the
finite element method (FEM) is generally preferred homogenous orthotropic material properties (as dem-
as a robust tool for stress analysis of this kind of onstrated in Appendix A). Based on this analytical
spur gears.25–28 Based on these points, FEM is used procedure,31 which is programmed in MATLAB,
with ANSYS 16.0 to obtain root tensile stress and the CFRP material constants, which are illustrated
joint stress of hybrid spur gears. 3 D Finite element in Table 3, are calculated.
hybrid gear model is designed in CATIA V5 with the The detail views of composite laminas with the
points of gear tooth, which are imported from a gen- fiber angles are presented in Figure 3.
erated MATLAB code based on Litvin’s Adhesive bonding is assumed for the joining
approach.29,30 FE model used in the study is pre- method of steel parts and CFRP. To evaluate the
sented in Figure 2. The design parameters of the FE joint stress of hybrid gear, the contact region between
model are illustrated in Table 1. steel teeth-rim and CFRP is defined with cohesive
Stress analyses (for root and joint regions) are con- zone modeling (CZM). CZM can predict interface
ducted in the same FE model. Steel teeth and hub damage for adhesive bonding with different materials
parts are defined as isotropic material. Its elasticity without an initial defect.32,33 It is characterized by a
modulus and Poisson’s ratio is selected as cohesive law based on tractions-displacements on the
E ¼ 210 GPa and t ¼ 0.3, respectively. As to CFRP normal (Mode I) or tangential (Mode II) directions in
for the web region, it is assigned as a homogenous the contact region. If these tractions-displacements
orthotropic material. It is assumed that the CFRP occur in both directions, then it is called mixed
material consists of the joining of 12 uni-directional mode. Detail view of cohesive law for mixed-mode
(UD) laminas with 2 mm thicknesses. The angle (Ø) presented in Figure 4.33
Yılmaz
4 et al. 357
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Table 2. Material constants of UD lamina.

E2 E3 G12 G13 G23

E1(GPa) (GPa) (GPa) t1 t2 t3 (GPa) (GPa) (GPa)

230.6 6.23 6.23 0.38 0.31 0.38 3.29 3.29 2.38

Table 3. Material constants of CFRP laminate.

E1 E2 E3 G12 G13 G23

(GPa) (GPa) (GPa) t1 t2 t3 (GPa) (GPa) (GPa)

81.65 81.65 6.81 0.32 0.27 0.27 30.76 2.76 2.76

Figure 4. Cohesive law for mixed-mode.

Table 4. Properties of cohesive zone material.

Property XNR6823

Young Modulus E (MPa) 2600

Tensile failure strength rn (MPa) 57
Shear Modulus G (MPa) 1000
Shear Failure Strength sn (MPa) 32.9
Toughness in tension GIC (J/m2) 1180
Toughness in shear GIIC (J/m2) 1500

20 node SOLID 186 hexahedral solid elements are

used with an element size 0.3 for meshing steel and
CFRP parts, while TARGE170 and CONTA174
Figure 3. Detail views of composite laminas: (a) Left view. mesh elements are applied to the joint region for
(b) Front view. CZM. The 3 D full gear model is used for analyses.
The shaft hole is fixed that no rotation or displace-
ment is allowed. Mesh and boundary conditions are
According to Figure 3, the stress (rn) has a linear presented for root stress and joint stress analyses in
relationship with displacement until dns for the adhe- Figure 5.
sive. Until this value, this region is accepted as safe, To obtain the torque capacity of hybrid gear as
and there is no debonding. After this value, damage well as root stress, first, starting from 100 Nm, 50
begins, and full separation occurs when displacement Nm increments up to 400 Nm, torque equivalent
reaches the value of dnc. The same is true in Mode II. normal force is calculated, then this normal force
As to mixed-mode, it is a combination of Mode I and divided into the facewidth (b). In the analyses, this
II. Details of the debonding equations can be found in unit force is applied to HPSTC. As the spur gears
a previous study.33 It is well known that the transmit- are subjected to planar loading, facewidth has a
ted load during the meshing is composed of two com- linear effect on the root stress. Based on this fact,
ponents, which are tangential and radial load in the Facewidth is taken as 1 mm to decrease CPU and
spur gears. The tangential load is responsible for analysis time in FEA. The relation between normal
tensile-compression and shear stress in the tooth force and torque is presented in Appendix B.
root and joint region, while radial force is the The root stress results of hybrid gear with different
reason for only the compressive stress. Based on rim thicknesses for 100 Nm torque value are illustrat-
these facts, the mixed debonding mode, which is a ed in Figure 6.
combination of Mode I (Tensile-Compressive) and According to the results of the analyses, root ten-
Mode II (Shear), is selected to define contact region sile stress decreases as the rim thickness increases. The
behavior under the loading for strength analyses. In root tensile stress of the hybrid gear with a rim thick-
the analyses, a cohesive zone material32 with zero ness of 0.5xm is 11% higher than that of full steel
thickness is defined based on Fracture-Energies gear. This ratio decreases to 4.6% and 1.9% in
debonding in ANSYS. In Table 4, the properties of hybrid gears with 1xm rim thickness and 1.5xm rim
cohesive zone material that are used in the analyses thickness, respectively. After 1.5xm rim thickness,
are presented. there is almost no difference in root stress between
358 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
5

Figure 5. Mesh and boundary conditions for root stress and joint stress analyses.

steel gear and hybrid gear. The root stress of full steel for damage initiation indicates that the stress value
gear is 117.39 MPa. To validate of FEA, DIN3990 reaches the failure strength pointed out in Table 4.
Method B (details given in Appendix C) is used. The limit torque for 0.5xm rim thickness is
The root stress is calculated as 122.62 MPa with the 199 MPa. When the rim thickness is increased from
analytical standard. The difference between the FEA 0.5xm to 1xm, the torque limit value is increased by
and DIN (4.2%) is found very little that FEA has 50%, and when it is increased to 1.5xm, it becomes
adequate accuracy. The other stress results are pre- double. For other rim thickness values, the torque
sented in graphical data (Figure 7) as the relation limit exceeds 400 Nm. The hybrid gear with 0.5xm
between torque and stress is linear. rim thickness is found as the weakest option in
The normal stress values of hybrid gears on the terms of joint stress. When Figures 10 and 11 are
contact region for 100 Nm torque are illustrated in examined, a linear relationship between torque and
Figure 8. joint stresses can be understood. Based on this, the
In Figure 8, the stress values on the tensile side torque capacity of the hybrid gears for different rim
(minimum values) are taken into consideration as thicknesses is presented in Figure 12.
these stresses cause separation in the normal direction On the other hand, According to the previous
for joint. According to results, as the distance experimental and numerical study, the temperature
between the root and joint regions increases, the of the contact region does not exceed 100  C under
normal stresses in joint decrease. Rim thickness is a normal torque and rotational speeds. In addition, the
highly influenced parameter on normal stress. When temperature value is lower than the contact region on
the rim thickness is increased from 0.5xm to 3xm, the the internal regions.34 So if the higher rim thickness is
normal stress decreases 97% in the joint region. This selected, the temperature effect on the joint decreases.
ratio is nearly 46%, 72%, 82% and 87% in 1xm, For these reasons, it is assumed that joint stress due
1.5xm, 2xm and 2.5xm rim thicknesses, respectively. to the temperature remains low that it could be
The shear stresses on the contact region for 100 Nm neglected. Nevertheless, for higher torque and rota-
torque are illustrated in Figure 9. tional speeds, the temperature effect on joint stress
The maximum values (compressive side) in Figure could be a future study. A paper from NASA
9 are the shear stresses that occur in the joint. researchers also proved that usage of hybrid gears
According to the results, as the distance between the does not increase the gearbox temperature when com-
root stress region and the joint region increases, shear paring steel gears for under six different speeds and
stresses on the joint decrease. Shear stress for 0.5xm five different levels of applied torque.35
rim thickness is 16.404 MPa. When the rim thickness
is increased 0.5xm to 3xm, the shear stress decreases
Evaluation of tooth stiffness of hybrid
by approximately 65%. After 2xm rim thickness, the
shear stress is extremely low. For other torque values,
gears
the stress results are illustrated in Figures 10 and 11. As the involute spur gears are subjected to vibration
When Stress results in Figures 10 and 11 are due to the higher rotational speeds during the opera-
observed, it can be concluded that shear stress is tion, they have to be investigated in terms of dynamic
more significant than normal stress in terms of performance. Tooth stiffness is a highly significant
damage initiation. In these figures, the torque value parameter for dynamic analysis. It can be calculated
Yılmaz
6 et al. 359
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Figure 6. Root stress results for 100 Nm torque: rim thicknesses of (a) 0.5xm, (b) 1xm, (c) 1.5xm, (d) 2xm, (e) 2.5xm, (f) 3xm,
(g) full steel.

as the proportion of normal force to the total deflec- as asymmetric or gears with multi-material, the finite
tion of the gear tooth.36–38 Total deflection consists of element method is more convenient. Finite element
five components, which are bending, compressive, analysis was used to obtain total deflection, conse-
shear, rim, and Hertzian deformation. It can be cal- quently tooth and mesh stiffness for bimetallic spur
culated by an analytical method for standard gears.39 gears,40 asymmetric gear,41 or hybrid gear.15 Based on
However, for gears with non-standard geometry such these, FEA is preferred to obtain tooth deformations.
360 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
7

Figure 7. Root stress results of hybrid spur gears for different torques.

The mesh and boundary conditions of deformation F

kg;1 ¼ (2)
analysis are presented in Figure 13. xg;1
In deformation analyses, the normal force (100 N)
is implemented from five lines of active flank, which F
kp;2 ¼ (3)
are on radii of 33, 31.875, 30.7266, 29.625, 28.5 mm, xp;2
respectively. The shaft hole is fixed as in stress anal-
yses. Mesh size is reduced to 0.15 mm to include F
Hertzian deformation, as proposed by Coy and kg;2 ¼ (4)
xg;2
Chao.42 All contacts between components of hybrid
gear are defined as purely bonded. The adhesive bond
Where F is the applied load on the tooth profile, x
is modeled as a physical layer to include its elasticity.
is the total deformation of the tooth geometry, (mm),
The material of adhesive is assigned as isotropic with
and kp and kg are the single tooth stiffnesses of the
the Poisson’s ratio of 0.3 and 2.6 GPa Young modu-
pinion and gear, respectively (N/mm). 1 and 2 indi-
lus. Adhesive young modulus is relatively low when cate that the first and second tooth pair. Finally, high
comparing it to CFRP and steel Young modulus. For precision (eave<0.002) curves are fitted to obtain the
this reason, if the cohesive zone is enhanced, then the tooth stiffness curves. The tooth stiffness equations
tooth stiffness decreases because of the deformation are used to calculate mesh stiffnesses, which are
of the tooth increases. The thickness of the adhesive inputs for dynamic analyses. In Figure 14, the tooth
layer for steel/CFRP joints generally varies from stiffness curves are illustrated for hybrid gear with
0.25 mm to 1.5 mm in literature. In analyses, different rim thickness and full steel gear.
0.25 mm is selected as thickness of adhesive bond, According to results, the tooth stiffness increases
as used in a previous study.15 Since stiffness is one from tip to the bottom of the tooth for all gears. The
of the most significant parameters for especially stiffness values of hybrid gears are lower than full
transmission error and noise, the bond thickness steel gear since the elasticity modulus of composite
should be kept as minimum as possible with taken material is a third of the steel approximately. When
into consideration bond/slip condition and fracture the rim thickness is decreased, the tooth stiffness
energy in practice. decreases. The average stiffness value decreases by
Steel and CFRP properties are identical to stress nearly 48% for hybrid gear with 0.5xm rim thickness
analyses. Total deformation values are recorded for compared to steel gear. This decrease rate is found
each rim thickness, then the normal force (100 N) is 37%, 31%, 28%, 27% and 26% for 1xm, 1.5xm,
divided into the total deformation to calculate tooth 2xm, 2.5xm, 3xm, rim thicknesses respectively. The
stiffness for five radii. The single tooth stiffness values effect of the rim thickness on stiffness decreases con-
for each point are calculated by equations (1) to (4). siderably after 2xm. Weights of hybrid gear are also
obtained for different rim thicknesses to compare full
F steel gear and to calculate rotational inertia for
kp;1 ¼ (1)
xp;1 dynamic analyses. The density of steel is taken as
Yılmaz
8 et al. 361
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Figure 8. Normal stresses on joint region for 100 Nm torque: rim thicknesses of (a) 0.5xm, (b) 1xm, (c) 1.5xm, (d) 2xm, (e) 2.5xm,
(f) 3xm.

7.86 g/cm3 while it is 1.53 g/cm3 for CFRP. The weight weight reduction percentage divided by the stress
of the adhesive is neglected. The weight status of increase percentage for stress and stiffness decrease
gears is presented in Table 5. percentage for stiffness. In Table 6, the relation
While the rim thickness is decreased, the weight between weight benefit factor and rim thickness is
decreases as well, since the portion of the steel illustrated.
part reduces. It turns out that the weight changes Hybrid gear with 3xm rim thickness is found as the
inversely with the stress while it is in direct proportion best option in terms of weight benefit factor for stress
with stiffness. For this reason, a weight benefit while 1.5xm rim thickness in terms of weight benefit
factor should be defined to obtain optimum rim factor for stiffness. Indeed, the difference between
thickness. The weight benefit factor is calculated by hybrid gear with 1.5xm rim thickness and solid gear
362 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
9

Figure 9. Shear stresses on joint region for 100 Nm torque: rim thicknesses of (a) 0.5xm, (b) 1xm, (c) 1.5xm, (d) 2xm, (e) 2.5xm,
(f) 3xm.

is only 1.9%. For this reason, 1.5xm rim thickness can (Steel) and hybrid gears. The first six modes are relat-
be preferred to gain more weight reduction. ed to rigid body movements, so these modes are not
taken into consideration. Natural frequencies of steel
gear and hybrid gears for Mode 7, 8, and 9 are pre-
Modal analysis of hybrid gears
sented in Table 7.
Finite element analyses are also conducted to under- According to results, the frequency variation
stand the effect of rim thickness on the natural between hybrid gears with different rim thickness is
frequencies and mode shapes of hybrid gears under rather low. When the rim thickness increases from
free-free boundary conditions. These analyses are 0.5xm to 3xm, the natural frequency increases by
conducted for only comparison aim between Solid nearly 5% for Mode 7 and 8 while it is 13% for
Yılmaz
10 et al. 363
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Evaluation of dynamic behavior of hybrid

gears
Calculation of time-varying mesh stiffness
Dynamic analyses are performed for the determina-
tion of the dynamic loads and resonance regions of
the steel and hybrid gear mechanisms. The time-
varying mesh stiffness (TVMS) is one of the most
critical parameters for the dynamic models. Thus
the (TVMS) directly affects the dynamic loads and
the dynamic response of the gear mechanisms.
Hence, the TVMS should be calculated carefully. In
Figure 10. Normal stress results of hybrid spur gears on the the literature, there are several methods proposed for
joint for different torques. the calculation of the TVMS. These methods can be
classified as analytical, numerical, and experimental
methods. Liang et al. calculated the TVMS for plan-
etary gears by using the potential energy method.43
Munro et al. calculated TVMS by using back to back
test rig experimentally for lower speeds.44 In this
study, the TVMS is calculated by using the finite ele-
ment method, which was developed in the litera-
ture.34–38 During the power transmission of the spur
gears, one and two tooth pairs contact and leave the
meshing, respectively. In a gear mechanism with a
normal contact ratio, at least one and at most two
tooth pairs transmit the torque. The contact starts at
the point of A for tooth pair p,1/g,1. At this time,
another tooth pair (p,2/g,2) is in contact at the
Figure 11. Shear stress results of hybrid spur gears on the point of D (Figure 16). When the first tooth pair
joint for different torques. comes to the point B, and the second tooth pair
leaves the contact. Thus, the regions A–B and D–E
are defined as double tooth contact region. Between
B–D, only the first tooth pair is in contact, so the B–
D region is defined as the single tooth region. For
calculation of the TVMS, the equivalent stiffness of
gear pairs is calculated as a serial-connected spring in
itself. On the other side, in the double tooth pair
region, the equivalent mesh stiffness is calculated by
the sum (assumed as parallel-connected spring) of
each equivalent stiffness of the pairs.
The equivalent stiffness of the first pair of teeth is:

kp;1 kg;1
K1 ¼ (5)
kp;1 þ kg;1
Figure 12. Torque capacity of hybrid gears with different rim
thicknesses. The equivalent stiffness of the second pair of teeth
is:

Mode 9. Solid (Steel) gear has the highest natural kp;2 kg;2
K2 ¼ (6)
frequencies for all modes as expected. The reason kp;2 þ kg;2
why steel has the highest elasticity modulus. As
the volume of steel increases in hybrid gears with If there is contact between B–D in the single tooth
increasing rim thickness, natural frequency values region:
increase. Mode shapes are similar for all gears. For K1 6¼ 0 and K2 ¼ 0
this reason, mode shapes of hybrid gear with only If there is contact between A–B or D–E in the
0.5xm rim thickness are also presented in Figure 15 double tooth contact region:
as a sample. K16¼0 and K26¼0
364 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
11

Figure 13. Mesh and boundary conditions for deformation analyses.

Table 5. Weight of solid gear and hybrid gears with different

rim thicknesses.

Rim thickness (xm) Weight (kg/b) Reduction (%)

0.5 0.009221 52
1 0.010655 45
1.5 0.012000 38
2 0.013228 32
2.5 0.014400 26
3 0.015483 20
Solid (Steel) 0.019442 –

Figure 14. Tooth stiffness curves for hybrid gears and steel Where Jp and Jg are polar mass moments of inertia
gear (solid). of pinion and gear, hp and hg are angular displace-
ments of pinion and gear, T is transmitted torque, rbp
and rbg are the radii of base circles, F is dynamic load,
The TVMS results for the steel and hybrid gears qp and qg radius of curvatures, m is coefficient of fric-
with different rim thicknesses are given in Figure 17. tion. The sign of friction force depends on the linear
It is seen that the TVMS takes the highest value when velocity of the pinion and gear throughout the mesh-
it is made of all steel. The TVMS values are decreased ing. If the linear velocity of the pinion is higher than
with the increase in the volume of composite material the velocity of gear, the sign of the friction force is
in the gear structure. The TVMS nearly two-times positive; otherwise, it is negative. In the analyses, the
decreases for the minimum rim thickness case, contact points on the line of action corresponding
which is 0.5xm. Moreover, after the 2xm rim thickness angular rotation angle of pinion and gear are used
value, there is no remarkable difference in terms of based on equations (9) and (10).
mesh stiffness.
xp ¼ rbp hp (9)
The dynamic model of hybrid spur gear
In this study, to calculate dynamic loads and the xg ¼ rbg hg (10)
static transmission error of the steel and hybrid gear
mechanisms, a 2-DOF dynamic model is developed. Equivalent masses of pinion and gear can be writ-
The equations of motion can be defined by using the ten as;
free body diagram seen in Figure 18.
Jp
mp ¼ (11)
Jp €
h p ¼ Tp � rbp ðF1 þ F2 Þ�qp;1 l1 F1 �qp;2 l2 F2 (7) r2bp

Jp €
h g ¼ rbg ðF1 þ F2 Þ � TG �qg;1 l1 F1 �qg;2 l2 F2 (8) Jg
mg ¼ (12)
r2bg
Yılmaz
12 et al. 365
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

Table 6. Weight benefit factors in terms of stress and stiffness.

Stress Average Tooth Weight benefit factor Weight benefit factor

Rim thickness (xm) Increase (%) Stiffness Decrease (%) for stress for tooth stiffness

0.5 11 47 4.72 1.1

1 4.6 38 9.78 1.18
1.5 1.9 32 20 1.19
2 1.2 28 26.66 1.14
2.5 0.7 27.2 37.14 0.95
3 0.5 26.9 40 0.74
Solid – – – –

Table 7. Natural frequencies of hybrid gears and steel gear.

Rim thickness (xm) 0.5 1 1.5 2 2.5 3 Solid

Modes (Hz) 7 15086 15282 15326 15434 15664 16006 17536

8 15387 15587 15632 15742 15977 16326 17888
9 18213 18878 19870 20626 20653 20984 22198

The static load (FD) that applied on the gears can be defined as;

Tg Tp
FD ¼ ¼ (13)
rbg rbp

The above expressions are added to equations (7) and (8). If the difference of two equations is taken and
required arrangements are conducted on the equations, the static transmission error and general equation of
motion are obtained as in equations (14) and (15), respectively. Derivation of equations is given in Appendix D
section in detail.
The static transmission error can be described as;
� � � �
ðmg mp ÞFD þ K1 e1 fp;1 mg þ fg;1 mp þ K2 e2 fp;2 mg þ fg;2 mg
xs ¼ � � � � (14)
K1 fp;1 mg þ fg;1 mp þ K2 fp;2 mg þ fg;2 mp

Finally, the equation of motion of the system can be written as;

" � � � �#12 � � � �
K1 fp;1 mg þ fg;1 mp þ K2 fp;2 mg þ fg;2 mp K1 fp;1 mg þ fg;1 mp þ K2 fg;2 mg þ fg;2 mp
xr þ 2 nx r þ xr
mg mp m g mp
� � � �
ðmg mp ÞFD þ K1 e1 fp;1 mg þ fg;1 mp þ K2 e2 fp;2 mg þ fg;2 mg
¼ (15)
mg mp

Where n is the damping of the gear pair, and m is the mass of the pinion and gear. The equation of motion is
solved by using the fourth-order Runge – Kutta method and static transmission error results, and maximum
dynamic forces are taken into consideration, and the dynamic factor (DF) is calculated between 1000 and
40,000 rpm pinion speed. The dynamic factor is calculated according to equation (16).

Maximum Dynamic Load

DF ¼ (16)
Static Load

STE and dynamic factor results

Static transmission error is one of the most critical factors for power transmission and gear noise and vibration.
Thus, the static transmission error should be predicted by the gear designers. In this study, static transmission
366 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
13

Figure 18. Free body diagram for 2 – DOF dynamic model.40

errors are investigated for steel gear and hybrid gears

for comparison aim. The static transmission errors of
Figure 15. Mode shapes of gears (a) Mode 7, (b) Mode 8, the gear mechanisms are calculated by using equation
(c) Mode 9. (14) in the dynamic model. The damping of meshing
gears and the coefficient of friction varies throughout
the line of action. These expressions can be calculated
with equations (17) and (18), respectively.41
2 312
K1 þ K2
n ¼ 2nt 4 r2bp r2bg 5 (17)
JP þ Jg

0:05 pffiffiffiffiffi
l¼ þ 0:002 Vs (18)
e0:125Vs

Where nt the damping ratio of meshing gears, J is

the rotational inertia of gears and Vs is the sliding
velocity between pinion and gear. Damping ratio of
meshing gears are taken as 0.1.
The effect of rim thickness on the static transmis-
Figure 16. The meshing of spur gears.18 sion error is illustrated in Figure 19. Static transmis-
sion error can be defined as a reverse function of the
time-varying mesh stiffness. Thus the results of the
TVMS and static transmission error should be com-
patible. Actually, a small fluctuation can be noticed at
the middle of the single tooth contact region as dif-
ferently from TVMS. The reason for this situation is
the change in the coefficient of friction throughout
the line of action. The values of static transmission
error are obtained for 1 rpm pinion angular velocity.
When Figure 19 is examined, it is seen that the min-
imum static transmission error is obtained for solid
gear because it has the highest time-varying mesh
stiffness. When the rim thickness of the hybrid gear
increases, the static transmission error decreases as
the time-varying mesh stiffness increases with the
rim thickness. The effect of rim thickness on STE
reduces drastically after 2xm. The dynamic factor of
the steel and hybrid gears between “1000 and
Figure 17. Time-varying mesh stiffness results for steel and 40,000 rpm” is given in Figure 20. It is seen that the
hybrid gears. differences in the dynamic response for the different
Yılmaz
14 et al. 367
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

modeled as a physical layer to include its elasticity

on deformation. Single tooth stiffness of hybrid gears
was calculated with load divided by deformation value
obtained from analyses results. Finally, 2-DOF
dynamic analyses were performed to understand the
effect of rim thickness on the dynamic response for
hybrid gears. According to results,

• In hybrid gears maximum shear stress on joint

region is higher than maximum normal stress so,
adhesive bonds with high shear strength should be
used in hybrid gear design.
• Rim thickness is the most important parameter for
weight reduction so, it should be defined properly.
According to results 1.5xm rim thickness is found
best option in terms of stress and stiffness. After
Figure 19. Static transmission error results for steel and this value stress and stiffness values change very
hybrid gears. little. For this value, root tensile stress is nearly
equal with root stress of steel gear.
• Rim thickness is more effective on joint normal
stress than shear stress. For hybrid gear with 3xm
rim thickness, the shear stress decreases nearly by
65% while normal stress decreases %87 with
regard to 0.5xm rim thickness.
• When the rim thickness is increased from 0.5xm to
3xm, the torque transmit capacity of hybrid gear
increases 20% nearly.
• Static transmission error decreases approximately
25% when the rim thickness becomes 3xm with
regard to 0.5xm rim thickness.
• For dynamic factor there is no noticeable difference
between hybrid gears for different rim thicknesses.

Although static transmission errors increase at a

Figure 20. Dynamic response of the steel and hybrid gears specific rate, it can be reduced by changing gear
between 1000 and 40,000 rpm values. parameters. It can be concluded that using hybrid
design in spur gears can ensure 30–50% weight reduc-
tion with regard to rim thickness when comparing to
cases are minimum. The first resonance region is
steel gear without a higher performance drop. Of
nearly 9000 rpm for all cases, and the value of a
course, it is not possible to replace steel for every
dynamic factor is almost 1.43. Similarly, the second
application area, however, it offers a very good alter-
resonance regions are also close, and the value of the native. Hybrid spur gears can be used in several
dynamic factor is approximately 2.23 around nearly industries where the fuel conception is an important
30,000 rpm. The backlash effect is not taken into con- issue, such as the automotive and aerospace indus-
sideration in this study. According to the results, the tries. The fact that it has the potential to dampen
situation does not worsen prominently in terms of the noise and vibration due to its structure, and provides
dynamic factor when using lightweight hybrid gears. design-dependent flexibility in terms of dynamic
properties, can enable hybrid gears to find application
Conclusion areas in these sectors.

In this study, the static and dynamic behavior of Declaration of Conflicting Interests
hybrid gears with different rim thicknesses were inves-
The author(s) declared no potential conflicts of interest with
tigated and compared with the solid steel gear, numer- respect to the research, authorship, and/or publication of
ically. 3 D full gear models for steel and hybrid gears this article.
were created in CATIA. To obtain root tensile and
joint stress, finite element analyses were conducted. Funding
In these analyses, cohesive zone material (CZM) was The author(s) disclosed receipt of the following financial
used to define joint region behavior for hybrid gears. support for the research, authorship, and/or publication
For deformation analyses, the adhesive bonding is of this article: This work was partially supported by the
368 et al.
Yılmaz Proc IMechE Part C: J Mechanical Engineering Science 236(1)
15

Scientific and Technical Research Council of Turkey 14. Catera PG, Gagliardi F, Mundo D, et al. Multi-scale
(TUBITAK) under Grant No. 118 M 810. modeling of triaxial braided composites for FE-based
modal analysis of hybrid metal-composite gears.
ORCID iDs Compos Struct 2017; 182: 116–123.
15. Catera PG, Mundo D, Treviso A, et al. On the design
Tufan G Yılmaz https://orcid.org/0000-0003-3772-7871
and simulation of hybrid metal-composite gears. Appl
Fatih Karpat https://orcid.org/0000-0001-8474-7328
Compos Mater 2019; 26: 817–833.
16. Contartese N, Catera PG and Mundo D. Static mesh
Supplemental material stiffness decomposition in hybrid metal-composite spur
Supplemental material for this article is available online. gears. In: Uhl T (ed.) Advances in mechanism and
machine science. Berlin: Springer International
References Publishing, 2019, pp. 977–985.
17. Catera PG, Mundo D, Gagliardi F, et al. A compara-
1. International Energy Outlook Report. Washington,
tive analysis of adhesive bonding and interference fit-
DC: U.S. Energy Information Administration, 2013.
ting as joining technologies for hybrid metal—
2. Yuce C, Karpat F, Yavuz N, et al. A case study: design-
composite gear manufacturing. Int J Interact Des
ing for sustainability and reliability in an automotive
Manuf 2020; 14: 535–550.
seat structure. Sustainability (Switzerland) 2014; 6:
18. Karpat F, Yılmaz TG, Dogan O, et al. Stress and mesh
4608–4631.
3. Yilmaz TG, Tüfekçi M and Karpat F. A study of light- stiffness evaluation of bimaterial spur gears. In: ASME
weight door hinges of commercial vehicles using alumi- International Mechanical Engineering Congress and
num instead of steel for sustainable transportation. Exposition, Proceedings IMECE), Salt Lake City,
Sustainability (Switzerland) 2017; 9: 1661. Utah, USA, 11-14 November 2019. New York:
4. Lee S, Kim B, Lee H-J, et al. Warm forging of an alu- ASME. DOI: 10.1115/IMECE2019-11554.
minium alloy for the differential case in an automobile 19. Gauntt SM and Campbell RL. Characterization of a
transmission. Proc IMechE, Part D: J Automobile hybrid (steel-composite) gear with various composite
Engineering 2016; 230: 1131–1139. materials and layups. In: American Institute of
5. Wu P, Wang B, Lin J, et al. Investigation on metal flow Aeronautics and Astronautics Scitech 2019 Forum, San
and forming load of bi-metal gear hot forging process. Diego, USA, 7-11 January 2019. Reston: AIAA. DOI:
Int J Adv Manuf Technol 2017; 88: 2835–2847. 10.2514/6.2019-0146.
6. Lisle TJ, Shaw BA and Frazer RC. External spur gear 20. Gauntt S, Campbell R and McIntyre S. Design optimi-
root bending stress: a comparison of ISO 6336:2006, zation of a hybrid spur gear. In: Vertical Flight Society’s
AGMA 2101-D04, ANSYS finite element analysis 75th Annual Forum and Technology Display,
and strain gauge techniques. Mech Mach Theory 2017; Philadelphia, USA, 13-16 May 2019. Fairfax: AHS.
111: 1–9. 21. Kim H, Kim C, Kim S, et al. Novel steel and aramid/
7. Cho SK, Kim HJ and Chang SH. The application of phenol composite gear for a transmission with optimum
polymer composites to the table-top machine tool com- design and FEM vibration analysis. Intj Automot
ponents for higher stiffness and reduced weight. Technol 2019; 20: 749–754.
Compos Struct 2011; 93: 492–501. 22. Wen Q, Du Q and Zhai X. A new analytical model to
8. Kim JH and Chang SH. Design of l-CNC machining calculate the maximum tooth root stress and critical
centre with carbon/epoxy composite-aluminium hybrid section location of spur gear. Mech Mach Theory
structures containing friction layers for high damping 2018; 128: 275–286.
capacity. Compos Struct 2010; 92: 2128–2136. 23. Zou T, Shaker M, Angeles J, et al. An innovative tooth
9. Cho DH, Lee DG and Choi JH. Manufacture of one- root profile for spur gears and its effect on service life.
piece automotive drive shafts with aluminum and com- Meccanica 2017; 52: 1825–1841.
posite materials. Compos Struct 1997; 38: 309–319. 24. DIN 3990-3:1987-12. Calculation of load capacity of
10. Bae JH, Jung KC, Yoo SH, et al. Design and fabrica- cylindrical gears; calculation of tooth strength.
tion of a metal-composite hybrid wheel with a friction 25. Cavdar K, Karpat F and Babalik FC. Computer aided
damping layer for enhancement of ride comfort. analysis of bending strength of involute spur gears with
Compos Struct 2015; 133: 576–584. asymmetric profile. J Mech Des 2005; 127: 477–484.
11. Handschuh R, Roberts GD, Sinnamon R, et al. Hybrid 26. Mo S, Ma S and Jin G. Research on composite bending
gear preliminary results—application of composites to stress of asymmetric gear in consideration of friction.
dynamic mechanical components. In: 68th American Proc IMechE, Part C: J Mechanical Engineering Science
Helicopter Society (AHS) Annual forum and 2018; 233: 2939–2955.
Technology Display, Fort Worth, Texas, 2012, NASA/ 27. Yilmaz TG, Dogan O, Yüce C, et al. Improvement of
TM—2012-217630, pp. 1–18. loading capacity of internal spur gear with using asym-
12. Handschuh RF, Laberge KE, Deluca S, et al. Vibration metric trochoid profile. In: ASME International
and operational characteristics of a composite-steel Mechanical Engineering Congress and Exposition,
(hybrid) gear—NASA/TM—2014-216646, 2014. Proceedings (IMECE), Tampa, Florida, USA, 3-9
13. LaBerge KE, Berkebile SP, Handschuh RF, et al. Hybrid November 2017. New York: ASME. DOI: 10.1115/
gear performance under loss-of-lubrication conditions. IMECE2017-71009.
In: American Helicopter Society International 73rd 28. Politis DJ, Politis NJ, Lin J, et al. An analysis of the
Annual Forum & Technology Display, Fort Worth, tooth stress distribution of forged bi-metallic gears.
Texas, USA, 9-11 May 2017, pp. 2250–2256. Fairfax: Proc IMechE, Part C: J Mechanical Engineering
AHS. Science 2018; 232: 124–139.
Yılmaz
16 et al. 369
Proc IMechE Part C: J Mechanical Engineering Science 0(0)

29. Litvin F and Fuentes A. Gear geometry and applied of spur gears with asymmetric teeth on dynamic
theory. 2nd ed. Cambridge: Cambridge University load analysis. Adv Mech Eng 2017; 9:
Press, 2004. 168781401772185–168781401772112.
30. Fetvaci C and Imrak E. Mathematical model of a spur 38. Karpat F, Yuce C and Dogan O. Experimental mea-
gear with asymmetric involute teeth and its cutting sim- surement and numerical validation of single tooth stiff-
ulation. Mech Based Des Struct Mach 2008; 36: 34–46. ness for involute spur gears. Measurement 2020; 150:
31. Chamis CC. Mechanics of composite materials: past, 107043.
present, and future, NASA/TM-1984-100793, 1984. 39. Liang X, Zhang H, Liu L, et al. The influence of tooth
32. Ribeiro TEA, Campilho RDSG, da Silva LFM, et al. pitting on the mesh stiffness of a pair of external spur
Damage analysis of composite-aluminium adhesively- gears. Mech Mach Theory 2016; 106: 1–15.
bonded single-lap joints. Compos Struct 2016; 136: 40. Yılmaz TG, Dogan O and Karpat F. A comparative
25–33. numerical study of forged bi-metal gears: Bending
33. Shokrian MD, Shelesh-Nezhad K and Soudmand BH. strength and dynamic response. Mech Mach Theory
3D FE analysis of tensile behavior for co-PP/SGF com- 2019; 141: 117–135.
posite by considering interfacial debonding using CZM. 41. Dogan O and Karpat F. Crack detection for spur gears
J Reinf Plast Compos 2016; 35: 365–374. with asymmetric teeth based on the dynamic transmis-
34. Long H, Lord AA, Gethin DT, et al. Operating temper- sion error. Mech Mach Theory 2019; 133: 417–431.
atures of oil-lubricated medium-speed gears: numerical 42. Coy JJ and Chao CH. A method of selecting grid size to
models and experimental results. Proc IMechE, Part G: account for hertz deformation in finite element analysis
J Aerospace Engineering 2003; 217: 87–106. of spur gears. J Mech Des 1982; 104: 759–764.
35. Laberge KE, Handschuh RF, Roberts G, et al. 43. Liang X, Zuo MJ and Patel TH. Evaluating the time-
Performance investigation of a full-scale hybrid com- varying mesh stiffness of a planetary gear set using the
posite bull gear. In: American Helicopter Society potential energy method. Proc IMechE, Part C: J
International 72nd Annual Forum & Technology Mechanical Engineering Science 2014; 228: 535–547.
Display, West Palm Beach, Floria, USA, 17-19 May 44. Munro RG, Palmer D and Morrish L. An experimental
2016. Fairfax: AHS. method to measure gear tooth stiffness throughout and
36. Karpat F, Dogan O, Ekwaro-Osire S, et al. A novel beyond the path of contact. Proc IMechE, Part C: J
method for calculation gear tooth stiffness for dynamic Mechanical Engineering Science 2001; 215: 793–803.
analysis of spur gears with asymmetric teeth. In: ASME 45. Colbourne JR. The geometry of involute Gears-J.R.
International Mechanical Engineering Congress and Colbourne.pdf. 1st ed. New York: Springer Verlag,
Exposition, Proceedings (IMECE), 2014, pp. 1–8. 1987.3
37. Karpat F, Dogan O, Yuce C, et al. An improved
numerical method for the mesh stiffness calculation