Mech. Sci.

, 12, 305–319, 2021

https://doi.org/10.5194/ms-12-305-2021
© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Design optimization analysis of an anti-backlash geared

servo system using a mechanical resonance simulation
and experiment
Lianchao Zhang1 , Hongbo Liao1,2 , Dapeng Fan1 , Shixun Fan1 , and Jigui Zheng3
1 College of Intelligence Science and Technology, National University of Defense Technology, Changsha,
410073, People’s Republic of China
2 Northwest Institute of Nuclear Technology, Xi’an, 710024, People’s Republic of China
3 Beijing Institute of Precise Mechatronics and Controls, Beijing, 100076, People’s Republic of China

Correspondence: Lianchao Zhang ([email protected])

Received: 30 May 2020 – Revised: 8 December 2020 – Accepted: 31 January 2021 – Published: 17 March 2021

Abstract. In many mechatronic systems, gear transmission chains are often used to transmit motion and power
between motors and loads, especially for light, small but large torque output systems. Gear transmission chains
will inevitably bring backlash as well as elasticity of shafts and meshing teeth. All of these nonlinear factors
will affect the performance of mechatronic systems. Anti-backlash gear systems can reduce the transmission
error, but elasticity has to be considered too. The aim of this paper is to find the key parameters affecting the
resonance and anti-resonance frequencies of anti-backlash gear systems and then to give the design optimization
methods of improving performance, both from element parameters and mechanical designing. The anti-backlash
geared servo system is modeled using a two-inertia approximate model; a method of computing the equivalent
stiffness of anti-backlash gear train is proposed, which comprehensively considers the total backlash of trans-
mission chain, gear mesh stiffness, gear shaft stiffness and torsional spring stiffness. With the s-domain block
diagram model of the anti-backlash geared servo system, the influences of four main factors on the resonance and
anti-resonance frequencies of system are analyzed by simulation according to the frequency response, and the
simulation analysis results dependent on torsional spring stiffness of anti-backlash gear pair and load moment
of inertia variation are verified by the experiment. The errors between simulation and experimental results are
less than 10 Hz. With these simulation and experiment results, the design optimization methods of improving the
resonance and anti-resonance frequencies such as designing the center distance adjusting mechanism to reduce
the initial total backlash, increasing the stiffness of torsional spring and lightweight design of load are proposed
in engineering applications.

1 Introduction bility, and they are seriously restricted in the application of

high-precision servo systems.
As a primary method of mechanical transmission, gear trans- To solve these problems, researchers and engineers pro-
mission is widely used in the precision servo mechanism posed several solutions, one of which is to choose the princi-
field, such as robot, seeker and inertially stabilized platforms, ple of the transmission mechanism. Using the harmonic gear
which have the requirements of fast response, high position- can eliminate backlash effectively, but friction and flexibil-
ing accuracy and good stability (Chung et al., 2010; Hilkert, ity of the transmission chain are increased (Vassileva et al.,
2008; Slamani and Bonev, 2013; Baek et al., 2003a). How- 2011; Masoumi and Alimohammadi, 2013), and using direct
ever, the traditional gear transmission mechanisms have the drive principle can omit the transmission chain. The motor
disadvantages of backlash, friction and elasticity, which limit and load are connected directly. Hence, the friction is re-
the performance of response, positioning accuracy and sta- duced, backlash is eliminated, and the mechanical stiffness

Published by Copernicus Publications.

306 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

is very high (Asada et al., 1989; Su et al., 2002). However, tal backlash on the resonance and anti-resonance frequencies
the direct drive is not suitable for small, light but large output of system were analyzed. Especially the contribution rate of
torque systems. each stage backlash to the resonance and anti-resonance fre-
The elasticity of transmission chain is also an important quencies were obtained. Zhou et al. (2009) built the model of
problem which influences the bandwidth of servo systems. multistage general spur-gear train based on two-inertia ap-
There are several mechanical components that have elasticity proximate model, which contained the backlash, gear shaft
which can influence the transmission error and performance stiffness and gear mesh stiffness. The influences of back-
of the gear transmission chain: transmission shaft elastic- lash and reduction ratio on frequency response characteris-
ity (Dwivedula and Pagilla, 2012), support bearing elasticity tics of system were analyzed. Based on the modeling method
(Walha et al., 2009; Wang et al., 2011, 2016), gear meshing of multistage general spur-gear train, some researchers stud-
elasticity caused by gear tooth stiffness (Jia et al., 2003; Li et ied the model of multistage anti-backlash gear train. Kwon
al., 2020) and even the gear meshing stiffness vary with time et al. (2004) built a two-inertia model of three-stage anti-
(Shi et al., 2020). All these factors make the transmission backlash geared servo system. A normalized describing func-
chain not ideally rigid, but flexible, under specific dynamic tion was used to describe the backlash; the formulas of res-
conditions, the systems will generate undesirable vibration. onance and anti-resonance frequencies were deduced. How-
Moreover, the elasticity and backlash will have combined in- ever, the model did not consider the torsional spring. The re-
fluence on the servo system. lation between the main gears and freewheel gears was not
Some researchers tried to solve these problems directly clear, and only the motor output torque influence on the fre-
from the control aspect without changing any mechanical de- quency response characteristics of system was analyzed.
sign. Kolnik and Agranovich (2012) estimated the equivalent In the anti-backlash geared servo systems, the resonance
backlash and torsional torque disturbances using disturbance and anti-resonance frequencies of system may be close to
observer (DOB). The torque is then compensated in the inner the servo bandwidth, which can limit the performance of re-
feedback loop, which decreased the vibrations caused by the sponse and stability of the servo system (Hale and Slocum,
backlash and the transmission shaft elasticity, but this com- 1994; Shim et al., 2008; Jesper, 2004). Acceleration feed-
pensation algorithm is just simulated using a transmission back loop and notch filter are used in some research to sup-
system model with a very small backlash. Yang et al. (2016) press the mechanism resonance (Lee et al., 2012; Szabat and
proposed a robust shaft torque compensator in a two-inertia Orlowska-Kowalska, 2007; George and Gao, 2001; George,
elastic system with backlash; this compensator can make the 2004; Hoogendijk et al., 2014). However, If the notch fil-
servo system behave like a rigid control system with single ter’s center frequency is different from the actual resonant
moment of inertia by adjusting the feedback coefficient, and frequency, the closed-loop system could exhibit increased os-
this control method can suppress mechanical resonance sig- cillations or even become unstable (Bahn et al., 2016). The
nificantly. center frequency of the notch filter must be the same as the
Another method to solve the backlash problem is us- resonance frequency accurately, which is difficult when the
ing anti-backlash mechanisms. Different anti-backlash gear resonance frequency is approximately equal to the crossover
transmissions were designed to decrease the influence of frequency of the speed loop; the acceleration feedback also
backlash in some research. Most of them mainly focus on increases the cost of system. Therefore, the problem needs
mechanism structure design, transmission precision analysis to be solved by other methods. The best one is to make the
and application. Several anti-backlash mechanisms were in- resonance and anti-resonance frequencies of the system far
vestigated; the advantages and disadvantages of these meth- from the bandwidth of the speed loop. The literature shows
ods were analyzed (Hale and Slocum, 1994). Shi and Yang that current research mainly focuses on the multistage gen-
(2000) enumerated several types of the anti-backlash gear eral spur-gear servo system, and frequency response charac-
train mechanisms, and the preload of torsional spring of dou- teristic analyses are concentrating on the total backlash of
ble gear mechanism was calculated. Allan and Levy (1980) gear train, gear shaft stiffness, gear mesh stiffness and motor
deduced the formula of minimum preload torque in the de- input voltage. However, the literature about the anti-backlash
sign of spring-loaded anti-backlash gear. Yang et al. (2013) gear train model is rare; the key factors affecting frequency
deduced the dynamic model of anti-backlash gear consid- response characteristics are not completely clear.
ering friction and time-varying mesh stiffness. These men- In this paper, a multistage torsional spring-loaded anti-
tioned studies provide theories for us about the calculation of backlash gear transmission chain is designed to eliminate
the comprehensive equivalent stiffness of the anti-backlash the backlash and to achieve a large output torque from a
gear train. small input torque. Based on this transmission chain, this pa-
Some researchers modeled the multistage general spur- per aims to find the key parameters affecting the resonance
gear trains and analyzed the frequency response characteris- and anti-resonance frequencies of anti-backlash gear systems
tics of them. Baek et al. (2003b) considered the dynamics of and to propose design optimization methods of improving
a two-stage general spur-gear train as a two-inertia approx- performance, both from element parameters and mechanical
imate model; the influences of motor input voltage and to- designing. The two-inertia model of torsional spring-loaded

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 307

anti-backlash gear train will be built based on the model

of the general spur-gear train; a new method of computing
equivalent stiffness of the anti-backlash gear train will be
proposed, which considers the total backlash of transmis-
sion chain, gear mesh stiffness, gear shaft stiffness and tor-
sional spring stiffness comprehensively. Then the s-domain
block diagram model of the anti-backlash geared servo sys-
tem will be built. With this model, the influences of differ-
ent parameters of the four main factors on the resonance and
anti-resonance frequencies of system are analyzed by simula-
tion. The simulation results will be validated by experiments.
The design optimization methods of improving the resonance
and anti-resonance frequencies are according to the simula-
tion and experiment results.
The structure of this paper is as follows. The model of anti-
backlash geared servo system is established, the proposed
equivalent stiffness computation method of anti-backlash
gear train is analyzed, the formulas of key parameters are de-
duced, and the s-domain block diagram of the anti-backlash
geared servo system is given in Sect. 2. The influences of dif-
ferent parameters of total backlash, torsional spring stiffness,
load moment of inertia, gear mesh stiffness, and gear shaft
stiffness variation on the resonance and anti-resonance fre-
quencies of system are analyzed by simulation in Sect. 3. Figure 1. Schematic of the anti-backlash geared servo system. V is
The simulation results of load moment of inertia and tor- a tachometer, P is a potentiometer as load position sensor, and G is
sional spring stiffness variation are validated by experiments a gyroscope.
in Sect. 4. Finally, conclusions are obtained in Sect. 5.

2 Modeling

2.1 System description

The anti-backlash geared servo system is composed of a DC

motor, four-stage anti-backlash gear train, load, sensors, mo-
tor driver and motion controller, which is shown in Fig. 1. Figure 2. Equivalent model of the anti-backlash geared servo sys-
This mechanism is applied in a two-axis inertially stabilized tem.
platform; the gyroscope closed loop is designed to isolate the
carrier disturbances.
In Fig. 1, the anti-backlash gear train is mainly composed lyzed, which are helpful to the design, assembly and applica-
of driving gear (1), main gears (2, 3, 4, 5, 6, 7, 8), freewheel tion of the anti-backlash gear train.
gears (20 , 30 , 40 , 50 , 60 , 70 , 80 ), torsional spring and bearing.
The driving gear is connected to the motor. Following the 2.2 Model of the anti-backlash geared servo system
motor rotation, the main gears are fixed on the shafts, and
the freewheel gears are nested on the shafts, which rotate In the multistage anti-backlash gear train, the transmission
around the shafts. One of the torsional spring butts is fixed components can be treated as a single, composite, equivalent
on the last stage main gear (8). Another one is fixed on the torsional spring that interconnects with the motor and load,
last stage freewheel gear (80 ). A certain preload is applied which is shown in Fig. 2.
on the torsional spring to guarantee the main and freewheel Where Jm is the equivalent moment of inertia on the mo-
gears contact with the positive or negative teeth surface of tor shaft, it mainly contains the moment of inertia of motor
the driving gear, respectively, so as to eliminate the back- rotor, equivalent moment of inertia of each stage main and
lash. In the anti-backlash geared servo system, the elastic- freewheel gears and shafts (not containing the last stage); θm
ity of transmission chain can not be neglected, which makes is the angular displacement of the motor; Tm is the output
the machine resonance close to the bandwidth of the gyro- torque of the motor, its direction is the left arrow in the fig-
scope closed loop. Therefore, the key factors affecting the ure; Tmf is the total equivalent friction torque; Tsm is the tor-
frequency response characteristics of system need to be ana- sional torque applied on the motor by the transmission train,

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

308 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Tmf and Tsm direction is the arrow in the middle in the figure,
which is the reverse direction to the Tm direction; Tl is the
driving torque of the last stage shaft on the load; its direction
is the right arrow in the figure, the same direction with the
Tm direction; Jl is the equivalent moment of inertia of load;
it mainly contains the moment of inertia of the last stage main
and freewheel gears, the shaft and the load; θl is the angular
displacement of the load; Ka is the equivalent stiffness of the
anti-backlash gear train.
The output torque of the motor is controlled. Namely, the
motor driver is operated in the current closed-loop mode, and
the controller of current loop is a proportional plus integral
(PI). The composite model of motor driver and motor can be
written as
Figure 3. Stiffness model of anti-backlash gear train.
dIa
Ua = Ra Ia + La + Ea (1a)
dt
Ea = Ke θ̇m (1b) 
Tm = Kt Ia (1c) Ka (1θ − b), 1θ > b

Zt Tl = 0, |1θ | ≤ b (3a)

Ua = Kp Kpwm (Ic − Ia ) + KI Kpwm (Ic − Ia ) dt, (1d) Ka (1θ + b), 1θ < −b

! !
0 Yn
1θ = θm / Ni − θ l , (3b)
where Ua is the input voltage of the motor; Ra is the re- i=1
sistance of the motor armature; La is the inductance of the
motor armature; Ia is the current of the motor; Ea is the where b is half of total backlash of gear train.
back electromotive force (EMF); Ke is the back electromo- Compared with the general spur-gear train, the motor and
tive force coefficient; Kt is the torque coefficient; Kp is the load always contact in the anti-backlash gear train because of
proportional gain of the current loop controller; KI is the in- the preload torsional spring. The transmission torque is not
tegral gain of the current loop controller; Kpwm is the motor zero in the backlash area, and the equivalent stiffness of the
driver gain; Ic is the reference signal of the current loop. backlash area should be considered, which is different with
In the spur-gear train, if the backlash is not considered, the the spur-gear train. Assuming that the equivalent stiffness of
transmission torque can be calculated by main gear train is Keq1 , the equivalent stiffness of freewheel
gear train is Keq2 , and the stiffness of torsional spring is Kx .
Then, the stiffness model of anti-backlash gear train can be
! !
Y n
Tl = K a θ m / Ni − θ l (2a) shown in Fig. 3. The arrow in the middle is the direction of
i=1
! Tm , and the right arrow is the direction of Tl .
n
Y There are two states in the anti-backlash gear motion pro-
Tsm = Tl / Ni , (2b) cess. One is the motion in the backlash area. The main
i=1 gear train does not transfer the torque directly. The torque
where Ni is the ratio of each stage gear train, Tl is the torque is transferred through the freewheel gear train and the tor-
transmitted to the load from the last stage transmission shaft, sional spring. The equivalent stiffness of the anti-backlash
Ka is the comprehensive stiffness of transmission train, and gear train is the series of Keq2 and Kx , which can be calcu-
n is the stages number of the gear train. lated by
However, the backlash can not be neglected in the system. Keq2 Kx
If the transmission angle error 1θ is small enough, the mo- Kab = . (4)
Keq2 + Kx
tor and load does not contact in the backlash area. The trans-
mission torque of gear train can be rewritten as (Nordin and Another motion process is not in the backlash area. The main
Gutman, 2002) gear train, freewheel gear train and torsional spring all take
part in the torque transmission. The equivalent stiffness con-
tains two parts: one part is the series of Keq2 and Kx ; another
part is Keq1 . The formula can be written as
Keq2 Kx
Kan = Keq1 + . (5)
Keq2 + Kx

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 309

parameters can only be obtained through theoretical analysis.

The formulas of Jm , Jl , Keq1 , Keq2 and Kx will be deduced.

2.3 Formulas of key parameters

In the anti-backlash geared servo system model, the param-

eter values of motor, motor driver and bearing friction can
be obtained from their handbooks. However, for the anti-
backlash gear train, researchers only know the size and ma-
terial of gears and springs. The equivalent moment of inertia
and stiffness can not be obtained directly. Without consider-
ing the backlash, the main gear train and the freewheel gear
Figure 4. Stiffness to transmission angle error. (a) Gear with back- train may be considered as a spur-gear train. The calcula-
lash. (b) Anti-backlash gear. tion methods of equivalent stiffness of the spur-gear train can
be used (Baek et al., 2003b). The anti-backlash gear train is
In the spur-gear train, with the existence of backlash, the re- composed of the main gear train and the freewheel gear train,
lation between the transmission angle error and transmission which is connected together by a torsional spring, as shown
torque is described in Fig. 4a. However, in the anti-backlash in Fig. 6.
gear train, the relation between the transmission angle error In the two-inertia system, the motor is one mass, and the
and transmission torque can be described in Fig. 4b, as the load is another one. The total mass of the system is focused
equivalent stiffness is not zero in the backlash area. on the motor and load. The equivalent moment of inertia of
Compared with the general spur-gear train, the equivalent the motor can be calculated by
stiffness is enhanced by using the anti-backlash gear, espe-
3
cially when the transmission angle error is larger than the X JFi + JHi
backlash. Based on Fig. 4b, compared with Eq. (3a), the Jm = J0 + , (8)
i
i=1 Nj2
Q
equivalent stiffness is not zero, but the combination of the
j =1
freewheel gear train and the torsional spring stiffness when
the motion is in the backlash area (|1θ | ≤ b), and if the mo-
tion is outside the backlash, the main gear train, freewheel where J0 is the total moment of inertia of the motor rotor,
gear train and torsional spring will have a comprehensive ef- driving gear and shaft; JFi is the total moment of inertia of
fect on transferring torque, so the transmission torque of the the gear and shaft of each stage main gear train (the last stage
anti-backlash gear train can be written as follows: is not contained ); JHi is the total moment of inertia of the
 gear and shaft of each stage freewheel gear train (the last
Kan (1θ − b) + bKab , 1θ > b
 stage is not contained).
Tl = Kab 1θ, |1θ | ≤ b (6a) Then, the equivalent moment of inertia of load can be cal-
 culated by
Kan (1θ + b) − bKab , 1θ < −b

!
n
Y Jl = Jl0 + JF4 + JH4 , (9)
Tms = Tl / Ni . (6b)
i=1
where Jl0 is the moment of inertia of pure load; JF4 is the
According to the two-inertia modeling method, the differen- total moment of inertia of the gear and shaft of last stage
tial equations of the anti-backlash gear train can be written main gear; JH4 is the total moment of inertia of the gear and
as shaft of last stage freewheel gear.
From Fig. 6, the Keq1 can be calculated with the series of
Jm θ̈m = Tm − Tmf − Tms (7a)
each stage main gear stiffness KFi ; Keq2 can be calculated
Jl θ̈l = Tl . (7b) with the series of each stage freewheel gear stiffness KHi .
The KFi and KHi can be calculated by
Based on above analysis, the differential equations of the
anti-backlash geared servo system can be obtained by com-
bining Eqs. (1) and (4)–(7). Using the Laplace transform, the KXgi KXsi
KXi = , (i = 1, . . ., 4, X = F or H), (10)
block diagram of model of the anti-backlash geared servo KXgi + KXsi
system in s domain is shown in Fig. 5.
The model has been established, but the key parameters of where KFgi is the mesh stiffness of each stage main-driven
the model are unknown. At the mechanism design stage, the gear and KHgi is the mesh stiffness of each stage freewheel-

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

310 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Figure 5. Block diagram of the anti-backlash geared servo system in s domain.

can be written as
 
Keq1 = KF1 KF2 KF3 KF4 (N2 N3 N4 )2 (13)

/ KF1 KF2 KF3 (N2 N3 N4 )2 + KF1 KF2 (N2 N3 )2

2
+ KF1 KF3 (N2 ) + KF2 KF3
 
2
Keq2 = KH1 KH2 KH3 KH4 (N2 N3 N4 ) (14)
Figure 6. Schematic of the anti-backlash gear train. 
/ KH1 KH2 KH3 (N2 N3 N4 )2 + KH1 KH2 (N2 N3 )2

driven gear. They can be calculated by (Zhou et al., 2009) 2
+ KH1 KH3 (N2 ) + KH2 KH3 .

2
KXgi = 0.0136EdXgi Wi cos2 λ (11) From Fig. 6, the freewheel gear train is connected to the main
(i = 1, . . ., 4, X = F or H), gear train through a torsional spring. The torque is transferred
by both the freewheel gear train and the torsional spring in
the backlash area, so the stiffness of torsional spring needs to
where E is Young’s modulus of the gears; dXgi is the pitch be considered in the model. The structure of torsional spring
circle diameter of each gear; Wi is the width of each gear; λ is shown in Fig. 7, which is an arc spring, with a rectangular
is the pressure angle of the gears. section and the same material as the gears.
KFsi is the stiffness of each stage main gear shaft, and KHsi Assuming that a force P is applied perpendicular to the
is the stiffness of each stage freewheel gear shaft. They can center line of the arc spring vertically, the deformation is δ,
be calculated by (Zhou et al., 2009) according to the principle of virtual work, the δ can be written
as
4
π GdXsi
KXsi = , (i = 1, . . ., 4, X = F or H), (12) Zπ
32LXsi M(ϕ)M̄(ϕ)R dϕ
δ= , (15)
EI
α
where G is the shear modulus of the gear shafts; dXsi is the
diameter of each gear shaft; LXsi is the length of each gear where α is half of the initial opening angle of arc spring in
shaft. Based on the relations between the input and output free state; R is the radius of the arc spring; I is the mo-
torque of anti-backlash gear train, formulas Keq1 and Keq2 ment of inertia of the spring’s rectangular section. M(ϕ) =

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 311

Table 1. Parameters values of the anti-backlash geared servo system.

Variable name (unit) Value

Motor resistance Ra () 4.9
Motor inductance La (mH) 0.148
Torque sensitivity Kt (N m A−1 ) 0.02
Back-EMF constant Ke (V s rad−1 ) 0.02
Proportional gain Kp 135.4
Integral gain KI 90 890
Motor driver gain Kpwm 2.4
Gear ratio N1 /N2 /N3 /N4 2/2.5/3.6/5
Initial backlash b (◦ ) 0.15
Pressure angle of gears λ (◦ ) 20
Gears’ and torsional spring’s Young’s modulus E (GPa) 205
Shear modulus of gears G (GPa) 80
Pitch circle diameter of each stage-driven gear dXg1 /dXg2 /dXg3 /dXg4 (mm) 20/25/36/60
Width of each stage-driven gear W1 /W2 /W3 /W4 (mm) 4.5/5.9/6.3/8.4
Diameter of each stage gear shaft dXs1 /dXs2 /dXs3 /dXs4 (mm) 2.5/4/6/12
Length of each stage gear shaft LXs1 /LXs2 /LXs3 /LXs4 (mm) 21.5/22/31/75
Width of the rectangular section of arc torsional spring c (mm) 2.5
Height of the rectangular section of arc torsional spring h (mm) 3.0
Radius of the arc torsional spring R (mm) 25
Half of the initial opening angle of arc spring in free state, α (◦ ) 20
The total moment of inertia of the motor rotor, driving gear and shaft J0 (kg mm2 ) 1.3
Moment of inertia of the second stage shaft and gears JF1 + JH1 (kg mm2 ) 0.56
Moment of inertia of the third stage shaft and gears JF2 + JH2 (kg mm2 ) 1.6
Moment of inertia of the fourth stage shaft and gears JF3 + JH3 (kg mm2 ) 7.4
Moment of inertia of the fifth stage shaft and gears JF4 + JH4 (kg mm2 ) 21.7
Moment of inertia of the pure load Jl0 (kg mm2 ) 945

If we let V = (π −α)(0.5+cos2 α)+0.75 sin 2α, then Eq. (16)

can be rewritten as
P R3V
δ= , (17)
EI
where I can be calculated by
ch3
I= , (18)
12
where c is the width of the arc spring’s rectangular section; h
is the height of the arc spring’s rectangular section.
Equation (17) can be rewritten as
EI
P= δ. (19)
V R3
When force P is applied on the spring, the variation value of
the opening angle is 1α. If 1α is small enough, sin 1α =
Figure 7. Structure of the arc torsional spring. 1α = δ/R, then δ = R1α. Substituting this equation into
Eq. (19), the stiffness of the torsional spring can be written
as
P R(cos α − cos ϕ), and M̄(ϕ) = R(cos α − cos ϕ). Substitut- T P R cos α EI cos α
ing these equations into Eq. (15), δ can be rewritten as Kx = = = , (20)
1α 1α VR
P R3 h i where T the torque generated by force P around the center
δ= (π − α)(0.5 + cos2 α) + 0.75 sin 2α . (16) of the arc spring.
EI

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

312 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Based on the above analysis, the values of key parameters

of the system – Jm , Jl , Keq1 , Keq2 and Kx – are obtained
through Eqs. (8), (9), (13), (14) and (20). Now, with these
parameters and the block diagram in Fig. 5, the influences of
key parameters on the frequency response characteristics of
the anti-backlash geared servo system will be simulated and
analyzed.

3 Simulation

The simulation model of the anti-backlash geared servo sys-

tem is built in Fig. 5. The values of basic parameters in the
model are shown in Table 1, and parameter values of com-
puted variables are shown in Table 2. As the input of sys-
tem, the excitation signal is a chirp sinusoidal signal; the fre-
quency ranges from 1 to 300 Hz; the response signal is the
angular velocity of the motor, which is the output of sys-
tem. The sampling time interval is 0.2 ms. The bode diagram
of input/output can be obtained. The influences of backlash,
torsional spring stiffness, load, gear mesh stiffness and gear
shaft stiffness variation on the frequency response character-
istics of system can be obtained.

3.1 The effect of backlash

In the anti-backlash geared servo system, if changing the

magnitude of backlash only and other model parameters
shown in Tables 1 and 2, the backlash ranges from 0.01 to
0.5◦ . The influence of backlash variation on the frequency
response characteristics of system is shown in Fig. 8. The
backlash of the anti-backlash gear train is the total backlash
before the torsional spring installation, which can be adjusted
by the center distance between the driving and driven gear.
When the torsional spring is fixed on the last stage main gear
and freewheel gear, the space backlash of anti-backlash gear
train is very small but not completely eliminated.
The anti-resonance and resonance frequencies decrease
with the increase of backlash. When the backlash ranges
from 0.01 to 0.5◦ , the anti-resonance frequency ranges from
170 to 60 Hz, and the resonance frequency ranges from 190
to 70 Hz. The results show that the anti-resonance and reso-
nance frequencies can be adjusted at a large range by chang-
ing the magnitude of backlash. However, in the practical sys-
tem, if there is no center distance adjusting mechanism, the
total backlash is decided by the assembly technology. It is
difficult to control the magnitude of backlash in the assembly
process. Therefore, the method to improve the frequency re-
sponse characteristics of system by adjusting the total back- Figure 8. Influence of backlash variation on the frequency response
lash is to design the center distance adjusting mechanism. characteristics of system. (a) Bode diagram dependent on backlash
variation. (b) Anti-resonance and resonance frequencies dependent
on backlash variation.
3.2 The effect of torsional spring stiffness

In the anti-backlash geared servo system, only changing the

stiffness of the torsional spring, the influence of torsional

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 313

Table 2. Parameters values of computed variables.

Variable name (Unit) Value

Equivalent moment of inertia on the motor shaft Jm (kg mm2 ) (according to Eq. 8) 1.527
Equivalent moment of inertia of the load Jl (kg mm2 ) (according to Eq. 9) 966.7
Torsional spring stiffness Kx (N m rad−1 ) (according to Eq. 20) 49.91
Equivalent stiffness of each stage main gear KF1 /KF2 /KF3 /KF4 (N m rad−1 ) (according to Eq. 10) 14.24/90.43/322.9/2018.89
Equivalent stiffness of each stage freewheel gear KH1 /KH2 /KH3 /KH4 (N m rad−1 ) (according to Eq. 10) 14.24/90.43/322.9/2018.89

spring stiffness variation on the frequency response charac- 3.4 The effect of gear and shaft stiffness
teristics of system is analyzed. Due to the size restriction of
the last stage main and freewheel gears, the radius of tor- In the anti-backlash gear train, the size and material of each
sional spring can not be changed. Only the width or height of stage freewheel gear are the same as the main gear of the
rectangular section can be adjusted. In this paper, four kinds same stage, so KHi is equal to KFi . The equivalent stiffness
of torsional spring are chosen. Their stiffnesses are [24.80, of each stage main or freewheel gear is decided by the mesh
49.91, 102.50, 239.57] N m rad−1 , respectively. The result is stiffness of gear pair and the stiffness of gear shaft. However,
shown in Fig. 9. the mesh stiffness is decided by the width and pitch circle
In Fig. 9, the anti-resonance and resonance frequen- diameter of the driven gear, and the stiffness of gear shaft is
cies increase with the increase of stiffness of the tor- decided by the diameter and length of the shaft, so the KHi
sional spring. When the stiffness is [24.80, 49.91, 102.50, and the KFi can be changed by adjusting the size of driven
239.57] N m rad−1 , the anti-resonance frequency is [44.8, gear or gear shaft. In the simulation model, assuming that the
60.4, 69.5, 80.4] Hz, and the resonance frequency is [51.8, variation ranges of KFi and KHi are from 0.5 to 2 times initial
70.5, 81.4, 93.1] Hz. However, the frequency increment is value, namely KFi ∈ [0.5, 2]KFi , KHi ∈ [0.5, 2]KHi , the in-
restricted by the requirements of installation space and elas- fluences of KFi and KHi variation on the frequency response
tic force of the anti-backlash gear train. Compared with the characteristics of system are shown in Fig. 11.
backlash, the method to improve the frequency response From Fig. 11, when the KFi and KHi change, the anti-
characteristics of system by adjusting the size of torsional resonance and resonance frequencies remain the same, so
spring is relatively easy in engineering applications. The re- it is difficult to improve the frequency response character-
sult is validated in Sect. 4. istics of system by adjusting the size and material of main
and freewheel gears. The simulation results also show that
the farther away the gear is from the load, the less the in-
3.3 The effect of load fluence on the frequency response characteristics of system
caused by gear and shaft stiffness variation. The stiffnesses
The anti-backlash gear train is applied in the inertially stabi- of the last stage main and freewheel gear have the greatest in-
lized platform, which should remain the same control preci- fluence. So strengthening the stiffness of the last stage main
sion in different load conditions. While the resonance fre- and freewheel gear should be considered in the design pro-
quency of system is one of the main factors which influ- cess. However, when the mechanism of anti-backlash gear
ence the control precision, the influence of load variation train has been produced and assembled completely, chang-
on the frequency response characteristics of system is ana- ing the size of main and freewheel gears becomes difficult,
lyzed. In the simulation model, only changing the load mo- so the simulation results will not be verified in the next ex-
ment of inertia, the load moment of inertia ranging from 483 periments.
to 2000 kg mm2 . The result is shown in Fig. 10.
From Fig. 10, the resonance and anti-resonance frequen-
cies decrease with the increase of load moment of in- 4 Experiments
ertia. When the moment of inertia ranges from 483 to
2000 kg mm2 , the anti-resonance frequency ranges from 82 The experimental setup of anti-backlash geared servo
to 44 Hz, and the resonance frequency ranges from 92 to system is shown in Fig. 12, which is composed of
52 Hz, so the method to improve the frequency response an anti-backlash gear mechanism, motor driver (Maxon-
characteristics of system is the lightweight design of load. ADS50V/5A), dSPACE1103 hardware-in-loop simulation
However, the lightweight design may decrease the stiffness system and a computer. The tachometer is used to mea-
of the load, which will affect the resonance frequencies too. sure the angular velocity of the motor, and the sensitivity
Thus, there should be a compromise between the lightweight is 0.1 V(deg s−1 )−1 . The motor driver works at the current
design and stiffness decreasing of the load in the mechanical mode; its maximum continuous current is 5 A; the input sig-
design. nal ranges from −10 to +10 V; the current sensitivity is

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

314 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Figure 10. Influence of load variation on the frequency response

Figure 9. Influence of torsional spring stiffness variation on the characteristics of system. (a) Bode diagram dependent on load vari-
frequency response characteristics of system. (a) Bode diagram de- ation. (b) Anti-resonance and resonance frequencies dependent on
pendent on torsional spring stiffness variation. (b) Anti-resonance load variation.
and resonance frequencies dependent on torsional spring stiffness
variation.

stiffness and gear shaft stiffness. The simulation results of

Sects. 3.2 and 3.3 will be verified in this experimental setup.
0.8 V A−1 . dSPACE1103 provides AD and DA interface; the The initial state of the experimental setup containing the
resolution is 14 bit; the interface modules are embedded in magnitude of backlash is approximately 0.15◦ ; the stiffness
MATLAB/Simulink; the voltage of tachometer can be sam- of torsional spring is 49.91 N m rad−1 ; the moment of iner-
pled and motor driver control signal can be generated; the tia of load is 966.7 kg mm2 ; the size and material of main
bode diagram of output/input can be plotted in MATLAB. and freewheel gears are designed according to the specific
The simulation results of Sects. 3.1 and 3.4 will not be veri- project. First, the frequency response characteristics of ex-
fied, because it is not easy to adjust the backlash, gear mesh perimental setup are tested in the initial state. The resonance

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 315

Figure 11. Bode diagrams of anti-backlash geared servo system with the gears and shafts stiffness variation. (a) KF1 and KH1 . (b) KF2 and
KH2 . (c) KF3 and KH3 . (d) KF4 and KH4 .

and anti-resonance frequencies are obtained from the bode [483, 966.7, 2000] kg mm2 . The frequency response charac-
diagram. Then, three kinds of torsional spring are used to teristics of them are tested, respectively, and the results are
replace the initial one. They have different rectangular sec- shown in Fig. 14.
tions; the calculated values of stiffness are [24.80, 102.50, From Figs. 13 and 14, the resonance and anti-resonance
239.57] N m rad−1 . The frequency response characteristics of frequencies of system increase with the increase of torsional
servo system with these torsional springs are tested, respec- spring stiffness and decrease with the increase of load mo-
tively. The results are shown in Fig. 13. Finally, the stiffness ment of inertia. Compared with the experiment results, the
of torsional spring remains 49.91 N m rad−1 . Only changing maximum error of simulation results is less than 10 Hz,
the moment of inertia of load, three kinds of load are used which is shown in Table 3. These results show that the model
to replace the initial one. The moments of inertia values are of anti-backlash geared servo system is reasonable; the sim-

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

316 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Figure 12. Experimental setup of the anti-backlash geared servo system: (1) dSPACE1103, (2) motor driver, (3) motor, (4) tachometer and
(5) anti-backlash gear train, six-load.

Table 3. Comparative analysis between the simulation and experiment results.

Kx (N m rad−1 ) Simulation Experiment

Torsional spring stiffness
ARF(Hz) RF(Hz) ARF(Hz) RF(Hz)
24.8 44.8 51.8 43.7 55.4
49.91 60.4 70.5 62.3 78.9
102.5 69.5 81.4 70.1 88.8
239.57 80.4 93.1 76.8 93.5
Jl (kg mm2 ) Simulation Experiment
Load moment of inertia
ARF(Hz) RF(Hz) ARF(Hz) RF(Hz)
483 81.4 91.1 79.0 89.8
966.7 59.4 72.3 62.3 78.9
2000 44.2 52.5 42.9 56.9

ulation results of Sects. 3.2 and 3.3 are correct, and the sim- 5 Conclusion
ulation results of Sects. 3.1 and 3.4 are a credible prediction.
The errors of simulation analysis are caused by the model
parameter difference compared with the experimental setup
system. When the mechanism of anti-backlash gear train is The total backlash, torsional spring stiffness and load mo-
assembled, the parameters of model can be obtained through ment of inertia are the main factors that influence the fre-
the experiment identification, and then the experiment results quency response characteristics of the anti-backlash geared
may stay more consistent with simulation results. servo system. The resonance and anti-resonance frequencies
increase with the decrease of the total backlash and the load

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 317

Figure 13. Bode diagram of the experimental setup with different Figure 14. Bode diagram of the experimental setup with different
torsional spring stiffnesses. loads.

moment of inertia and increase with the increase of the tor- manufacture and assembly technology. From the mechanical
sional spring stiffness. design aspect, designing the center distance adjusting mech-
The simulation results show that when the total backlash anism can decrease the initial total backlash, and the load
ranges from 0.01 to 0.5◦ , the anti-resonance frequency ranges lightweight design can also improve the frequency response
from 170 to 60 Hz, and the resonance frequency ranges characteristics of system, but the stiffness of the load should
from 190 to 70 Hz; when the stiffness of torsional spring is be considered simultaneously. These measures will be ap-
[24.80, 49.91, 102.50, 239.57] N m rad−1 , the anti-resonance plied in our future work.
frequency is [44.8, 60.4, 69.5, 80.4] Hz, and the resonance
frequency is [51.8, 70.5, 81.4, 93.1] Hz; when the load mo-
ment of inertia ranges from 483 to 2000 kg mm2 , the anti- Data availability. All data generated or analyzed during this study
resonance frequency ranges from 82 to 44 Hz, and the res- are included in this published article.
onance frequency ranges from 92 to 52 Hz; when the varia-
tion ranges of gear mesh stiffness and gear shaft stiffness are
from 0.5 to 2 times initial value, the anti-resonance and res- Author contributions. LZ and HL designed the methodology,
onance frequencies remain the same. The experiment results created the model, simulated the model and designed the experi-
show that the maximum error of simulation results is less mental setup of servo system. DF provided the ideas, financial sup-
port and supervised the research activity planning and execution. SF
than 10 Hz. The model of anti-backlash geared servo system
designed the experimental software and collected the experimental
is reasonable. The simulation results of Sects. 3.2 and 3.3 are data. JZ designed the mechanical structure of anti-backlash geared
correct, and the simulation results of Sects. 3.1 and 3.4 are a setup and the arc springs. LZ designed the experiments and carried
credible prediction. them out. LZ and HL prepared the manuscript with contributions
From the mechanical parameter point of view, increasing from all co-authors.
the stiffness of torsional spring can improve the frequency
response characteristics of system. This target is easy to real-
ize by changing the width or height parameters of rectangular Competing interests. The authors declare that they have no con-
section of the torsional spring. Reducing the total backlash flict of interest.
of anti-backlash gear train is another method of system im-
provement. The magnitude of backlash is influenced by the

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021

318 L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis

Acknowledgements. The authors would like to express their tion control of flexible structures, Mechatronics, 24, 632–639,
thanks to the supporting agencies. https://doi.org/10.1016/j.mechatronics.2014.01.011, 2014.
Jesper, B.: Transmission error in anti-backlash conical involute gear
transmissions, a global–local FE approach, Finite Elem. Anal.
Financial support. This research has been supported by the Na- Des., 41, 431–457, https://doi.org/10.1016/j.finel.2004.04.007,
tional Key R&D Program of China (grant no. 2019YFB2004700) 2004.
and the Preliminary Research Project of National University of De- Jia, S. X., Howard, I., and Wang, J. D.: The Dynamic Model-
fense Technology (grant no. ZN2019-7). ing of Multiple Pairs of Spur Gears in Mesh, Including Fric-
tion and Geometrical Errors, Int. J. Rotating Mach., 9, 437–442,
https://doi.org/10.1080/10236210390241592, 2003.
Review statement. This paper was edited by Dario Richiedei and Kolnik, I. and Agranovich, G.: Backlash compensation for motion
reviewed by two anonymous referees. system with elastic transmission, 2012 IEEE 27th Conv. Elect.
and Elect. Eng., Eilat, Israel, 14–17 November 2012, 1–5, IEEE,
https://doi.org/10.1109/EEEI.2012.6377140, 2012.
Kwon, Y. S., Hwang, H. Y., Lee, H. R., and Kim, S. H.: Rate loop
control based on torque compensation in Anti-backlash geared
References servo system. Proc. 2004 Am. Control Conf., Boston, USA, 30
June–2 July 2004, ThP03.3, Piscataway, NJ, USA, IEEE, 3327–
Allan, P. M. and Levy, N. M.: The determination of min- 3332, https://doi.org/10.23919/ACC.2004.1384422, 2004.
imum pre-load torque for antibacklash gears in a posi- Lee, D. H., Lee, J. H., and Ahn, J. W.: Mechanical vibration reduc-
tional servomechanism, IEEE Trans. Ind. Electron., 27, 26–29, tion control of two-mass permanent magnet synchronous motor
https://doi.org/10.1109/TIECI.1980.351657, 1980. using adaptive notch filter with fast Fourier transform analysis,
Asada, H., Youcef-Toumi, K., and Koren, Y.: Direct-Drive robots, IET Electr. Power Appl., 6, 455–461, https://doi.org/10.1049/iet-
theory and practice, J. Dyn. Syst. Meas. Control, 111, 119–120, epa.2011.0322, 2012.
https://doi.org/10.1115/1.3153012, 1989. Li, Z. F., Zhu, C. C., Liu, H. J, and Gu, Z. L.: Mesh stiffness
Baek, J. H., Kwak, Y. K., and Kim, S. H.: Backlash estima- and nonlinear dynamic response of a spur gear pair consider-
tion of a seeker gimbal with two-stage gear reducers, J. Adv. ing tribo-dynamic effect, Mech. Mach. Theory, 153, 103989,
Manuf. Technol., 21, 604–611, https://doi.org/10.1007/s00170- https://doi.org/10.1016/j.mechmachtheory.2020.103989, 2020.
002-1378-z, 2003a. Masoumi, M. and Alimohammadi, H.: An investigation into the vi-
Baek, J. H., Kwak, Y. K., and Kim, S. H.: Analysis on the influ- bration of harmonic drive systems, Front. Mech. Eng., 8, 409–
ence of backlash and motor input voltage in geared servo sys- 419, https://doi.org/10.1007/s11465-013-0275-5, 2013.
tem, 11th IEEE Mediterr. Conf. Control and Automat., Rhodes, Nordin, M., and Gutman, P. O.: Controlling mechanical sys-
Greece, 18–20 June 2003, T1-010, 2003b. tems with backlash-a survey, Automatica, 38, 1633–1649,
Bahn, W., Kima, T. I., Li, S. H., and Cho, D. I.: Res- https://doi.org/10.1016/S0005-1098(02)00047-X, 2002.
onant frequency estimation for adaptive notch filters Shi, H. and Yang, X. H.: Spring’s elastic force com-
in industrial servo systems, Mechatronics, 41, 45–57, putation for double gear, Radar ECM, 3, 45–50,
https://doi.org/10.1016/j.mechatronics.2016.11.004, 2016. https://doi.org/10.19341/j.cnki.issn.1009-0401.2000.03.009,
Chung, J. W., Park, I. W., and Oh, J. H.: On the design and develop- 2000 (in Chinese).
ment of a quadruped robot platform, Adv. Robot., 24, 277–298, Shi, J. F, Gou, X. F., and Zhu, L. Y.: Calcula-
https://doi.org/10.1163/016918609X12586214966992, 2010. tion of time-varying backlash for an involute spur
Dwivedula, R. V. and Pagilla, P. R.: Effect of Compliance gear pair, Mech. Mach. Theory, 152, 103956,
and Backlash on the Output Speed of a Mechanical Trans- https://doi.org/10.1016/j.mechmachtheory.2020.103956, 2020.
mission System, J. Dyn. Syst. Meas. Control, 134, 031010, Shim, S. B., Park, Y. J., and Kim, K. U.: Reduction
https://doi.org/10.1115/1.4005493, 2012. of PTO rattle noise of an agricultural tractor using
George, E. and Gao, Z. Q.: Cures for low-frequency mechanical an anti-backlash gear, Biosyst. Eng., 100, 346–354,
resonance in industrial servo systems, IEEE 36th IAS Annu. https://doi.org/10.1016/j.biosystemseng.2008.04.002, 2008.
Meet.Chicago, USA, 30 September–4 October 2001, 252–258, Slamani, M. and Bonev, I. A.: Characterization and experimen-
https://doi.org/10.1109/IAS.2001.955419, 2001. tal evaluation of gear transmission errors in an industrial robot,
George, W. Y.: Compensating structural dynamics for Servo Driven Ind. Rob., 40, 441–449, https://doi.org/10.1108/IR-07-2012-387,
Industrial Machines with acceleration feedback, IEEE 39th IAS 2013.
Annu. Meet., Seattle, USA, 3–7 October 2004, 50p4, 1881–1890, Su, Y. X., Duan, B. Y., and Zheng, C. H.: Mechatronics de-
https://doi.org/10.1109/IAS.2004.1348726, 2004. sign of stiffness enhancement for direct-drive motor sys-
Hale, L. C. and Slocum, A. H.: Design of anti-backlash transmis- tem using ER variable damper, J. Robot. Syst., 19,419–425,
sions for precision position control systems, Precis. Eng., 16, https://doi.org/10.1002/rob.10050, 2002.
244–258, https://doi.org/10.1016/0141-6359(94)90001-9, 1994. Szabat, K. and Orlowska-Kowalska, T.: Vibration suppression in
Hilkert, J. M.: Inertially stabilized platform tech- a two-mass drive system using PT speed controller and addi-
nology, IEEE Control Syst. Mag., 28, 26–46, tional feedbacks comparative study, IEEE Trans. Ind. Electron.,
https://doi.org/10.1109/MCS.2007.910256, 2008. 54, 1193–1206, https://doi.org/10.1109/TIE.2007.892608, 2007.
Hoogendijk, R., Heertjes, M. F., van de Molengraft, M. J.
G., and Steinbuch, M.: Directional notch filters for mo-

Mech. Sci., 12, 305–319, 2021 https://doi.org/10.5194/ms-12-305-2021

L. Zhang et al.: Optimization of an anti-backlash geared system using resonance analysis 319

Vassileva, D., Kiyosawa, Y., and Suzuki, M.: Sensorless Yang, M., Wang, C., Xu, D. G., Zheng, W. L., and Lang,
Torque control for a robot with harmonic drive re- X. Y.: Shaft Torque Limiting Control Using Shaft Torque
ducers, Mech. Based Des. Struct. Mach., 39, 253–267, Compensator for Two-Inertia Elastic System With Back-
https://doi.org/10.1080/15397734.2011.550859, 2011. lash, IEEE ASME Trans. Mechatron., 21, 2902–2911,
Walha, L., Fakhfakh, T., and Haddar, M.: Nonlinear dynamics of https://doi.org/10.1109/TMECH.2016.2571304, 2016.
a two-stage gear system with mesh stiffness fluctuation, bearing Yang, Z., Shang, J. Z., and Luo, Z. R.: Effect analysis of friction
flexibility and backlash, Mech. Mach. Theory, 44, 1058–1069, and damping on anti-backlash gear based on dynamics model
https://doi.org/10.1016/j.mechmachtheory.2008.05.008, 2009. with time-varying mesh stiffness, J. Cent. South Univ., 20, 3461–
Wang, J., Qin, D., and Lim, T. C.: Influence of combined assem- 3470, https://doi.org/10.1007/s11771-013-1870-7, 2013.
bly error and bearing elasticity on spur gear tooth contact load Zhou, J. Z., Duan, B. Y., and Huang, J.: Modeling and effects
distribution, Proc. Inst. Mech. Eng. C J. Mech. Eng. Sci., 225, on open-loop frequency for servo system with backlash, Chin.
1507–1521, https://doi.org/10.1177/0954406211399212, 2011. Mech. Eng., 20, 1722–1725, 2009 (in Chinese).
Wang, J. H., Wang, J., and Lim, T. C.: Influence of assembly er-
ror and bearing elasticity on the dynamics of spur gear pair,
Proc. Inst. Mech. Eng. C J. Mech. Eng. Sci., 230, 1805–1818,
https://doi.org/10.1177/0954406215584632, 2016.

https://doi.org/10.5194/ms-12-305-2021 Mech. Sci., 12, 305–319, 2021