Professional Documents
Culture Documents
Modeling Verification of An Advanced Torsional SPR
Modeling Verification of An Advanced Torsional SPR
Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct
A R T I C L E I N F O A B S T R A C T
Keywords: In the article, the authors present an advanced model of hyperbolic torsion spring based on the Finite Element
Torsion spring Method (FEM). This model takes into account the contact forces between the spring arms, which allows you to
Tracked vehicle suspension simulate its operation even in the case of a large twisting angle. On the basis of this FEM model, the spring
Finite element method
stiffness characteristics were constructed. Based on this characteristics, the existing 2S1 track platform model has
ADAMS
Flexible body
been rebuilt. In this model, a much higher mass of the body is allowed than in the original model. On the basis of
the elastic spring model, parameters such as the static deflection of the suspension have also been determined, as
well as its range of mobility compared to the original suspension based on torsion bars. In addition, spring
deflection phases have been presented in terms of internal stresses, which are important for the strength analysis
of the spring. The originality of the described solution lies in the unique design of the so-called a hyperbolic
spring. This design has not been previously described in the literature.
1. Introduction A solution worth considering is the use of torsion springs. At this point, a
question appears. This question is associated with the kind and con
The solution of the original passive suspension system of the 2S1 struction of above mentioned torsion springs. The answer will be put
tracked platform is based on torsion bars. These bars are problematic forward by us in the first section of our paper. But before this, we have
elements. Problems with these elements are associated with their length analyzed constructions of full tracked vehicles and also modeling solu
and location of them in the vehicle. Because of their reactive character, tions, which may be applied in the tracked vehicle model. A complex
which during explosion of an anti-tank mine may destroy the suspension vehicle model was described by Campanelli and Shabana [3]. For
system, they should be replaced with another solution. This destruction analysis of the track link natural frequencies mode they use FEM. Less
could be also harmful for a crew of the vehicle. An alternative solution complex models of tracked vehicles were described by Nabagło et al.
may allow to replace reactive elements from under floor space to the [11,12]. The model developed in [12] was based on proving ground
vehicle sides. One of possible solutions of the problem is application of measurements described by Jurkiewicz et al. [4]. It included many
the helical springs. They were analyzed by Taktak et al. [17], with usage important elements of tracked vehicle model. One of the most important
of the Finite Element Method (FEM). FEM computer applications are a problem in this area is associated with joints between track segments.
good solution also for the analysis of other kinds of springs. This is very Ryu et al. [15] take under consideration single pin track link. He de
popular method of flexible bodies analysis, but there are also others, like scribes the so-called rubber bushing model, which connects two track
method applied by Mellouli et al. [10] for free vibration analysis, the links. Lee [6] in his article describes contact forces between wheels,
mesh-free Radial Point Interpolation Method. The helical springs could track links and ground surface. There is also explained a numerical
be problematic in application to the tracked vehicle, which was previ method of force contacts calculation. Details of this connection forces,
ously based on torsion bars. The suspension of the vehicle would require contact forces between wheels, track links and road surface and also
more interference in its structure than in the case of a torsional system FEM model construction of the flexible body are described in ADAMS
with other parameters. The FEM model application for various kinds of Documentation [2]. Selected parts, with high fidelity of real parts
springs was developed in collective work edited by Shimoseki et al. [16]. shapes, may be designed in other CAD programs. Łukaszewicz [7] in his
* Corresponding author.
E-mail address: nabaglo@agh.edu.pl (T. Nabagło).
https://doi.org/10.1016/j.engstruct.2020.111623
Received 11 April 2020; Received in revised form 27 September 2020; Accepted 17 November 2020
Available online 13 December 2020
0141-0296/© 2020 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
T. Nabagło et al. Engineering Structures 229 (2021) 111623
2
T. Nabagło et al. Engineering Structures 229 (2021) 111623
Fig. 3. Suspension stiffness reduced on the wheel axle (a) stiffness character
istics of the torsionbar (b).
where:
3
T. Nabagło et al. Engineering Structures 229 (2021) 111623
flexible model. Contact forces appear above an angle of 40.42 degrees equal to 10 kNs/m. This damping is reached after exceeding a pene
(see Fig. 8), but without noticeable influence. Radical increase of stiff tration, which is equal to 0.1 mm. Details of this contact model was
ness, because of contact influence, should appear after exceeding the described by Lankarani et al. [5]. These contact parameters are
value of 60 degrees. approximately equal to parameters of the contact between two surfaces
The simulated model is described in the Section 3.2. The spring of steel elements. This contact model neglects the friction phenomena.
system shown in Fig. 5b is built with the hyperbolic spring and sets of This situation is close to real conditions when steel torsional springs are
guide links connected between spring and its housing through revolute immersed in oil and friction may be neglected. The FEM algorithm in its
joints. The housing in the form of a rigid drum is rigidly connected to the generalized form is described below. It realizes subsequent steps:
vehicle hull. The torsion axle of the hyperbolic springs is also the axle of
suspension deflection (see Fig. 6). As is shown in Fig. 5a the hyperbolic 1. Discretization and meshing,
shape spring is 2.6 times stiffer than the torsion bar. Then the linear 2. Formulation of stiffness matrices for structural elements,
spring on the wheel axle, for the hyperbolic spring system is reduced to 3. Assembly of the global stiffness matrix,
130 kN/m. In continuation of this way of thinking and based on Eq. (3), 4. Assembly of the global load vector,
the unit sprung mass is equal to 2, 107 kg and the full sprung mass 5. Definition of boundary conditions.
should be equal to 29.5⋅103 kg. These parameters may be applied in the 6. Solving the system of equations,
full vehicle model. Static displacement with such mass and stiffness is 7. Determining internal forces and reactions.
described by Eq. (4) and it is about 0.16 m.
Given the stiffness matrix K, the static equation has the form (6).
Mu ⋅g
zst = (4)
ku K⋅x = F (6)
(z − z )
αst = α0 − arcsin 0 st
(5) where:
d
K – stiffness matrix,
where:
x – displacements vector,
F – force vector.
g gravitational acceleration, 9.81 m/s2 .
Such kind of model may be simulated in the Marc program, but as
This translational displacement is adequate to the angular displace
was said before, based on elements of this model also ADAMS flexible
ment, which is about 30 degrees (see Eq. 5). These parameters are
body model may be performed. Chosen steps of the simulation of the
important for estimation of preload position in the full vehicle model.
above described ADAMS flexible model are shown in Fig. 7. It begins
Contrary to the characteristics of the torsion bar, which were
with the initial position of the torsional spring for angle 0 degrees (see
calculated from the real suspension system, only a computer model of
the hyperbolic spring was performed.
Fig. 6. Torsion spring in suspension system unit (a), torsion spring construction
details (b). Fig. 7. Subsequent steps of simulation of the ADAMS flexible model.
4
T. Nabagło et al. Engineering Structures 229 (2021) 111623
Fig. 7a). In the figures from 7b to d, the results of the classic flexible body
simulation are presented. In Fig. 7d, the material deformation may be
noticed, especially on the external ends of the spring arms.
After Fig. 7e, where angle of 40 degrees was gained, the influence of
the contact forces appears. Vectors of these forces are described with red
arrows and the arrow length is proportional to the values of the forces. It
may be noticed in comparison with Fig. 8, where the stiffness charac
teristics of the torsional spring is presented, the contacts have negligible
influence on the stiffness characteristics shape. This is only true to the
moment when external the ends touch the closed spring. This condition
is fulfilled when the angle exceeds 60 degrees. After exceeding this
angle, stiffness should radically increase, which may be adequate to the
situation when the system is blocked. In Fig. 7f, high deformation are
visible. They are visible on the external ends of the spring arms. It may
suggest the necessity of making some changes in the spring construction.
From the simulation effect of the above model we obtained the stiffness
characteristics of the hyperbolic spring.
The FEM model implemented in the MARC program allowed for the
stresses analysis in the hyperbolic spring [9]. In this model, like in the
previous flexible model, the contact forces between the spring arms are
implemented. Fig. 9 shows the visualization of stresses for a suspension
deflection angle of 50 degrees. We can observe here the most stressed
parts of the spring, which can fail due to material fatigue during long-
term operation.
The co-simulation model with the ADAMS/View software allowed
for the presentation of the forces affecting the suspension mechanism
from the hyperbolic spring [1] (see Fig. 10). The both figures show the
result for the angle of 50 degrees because it was the maximum angle
obtained during the simulation, which allowed to observe both the Fig. 9. Equivalent of stress presented on the FEM model.
greatest forces and stresses in the mechanical system of the spring.
5
T. Nabagło et al. Engineering Structures 229 (2021) 111623
equilibrium position we know, that an angle at preload is about 30 de 3.4. Simulation results of the full vehicle model
grees. In relation to this, as is apparent from the suspension deflection in
the static equilibrium state, this deflection is reached for an angle of 30 As it was said previously, the suspension stiffness increased after
degrees. From Sections 3.1 and 3.2 we know, that a dramatic increase of hyperbolic spring application. Torsion bars are replaced with hyperbolic
the stiffness appears after exceeding an angle of 60 degrees. Between springs which give additional space inside the 2S1 vehicle, and lowers
these two angle positions, the range of 30 degrees exists. Therefore, the overall mass. According to the models calculation, mass of single torsion
deflection of 30 degrees is an angle from the static equilibrium position bar made from steel is about 35 kg, mass of single Archimedes spiral
to the position of the bump stops. Thus, the mentioned increase of spring system is about 21 kg but mass of single hyperbolic spring system
stiffness, perhaps may even support the effects of bumpstops activity. is only about 13.5 kg. Aggregate mass of the hyperbolic are less than
In the full vehicle model a stiffness characteristics taken from the 40% of the aggregate mass of torsion bars. In comparison, 14 torsion
spring flexible model was used. Exact characteristics with its extrapo bars weigh approximately 490 kg, when hyperbolic springs are used, the
lation used in simulation is presented in the Fig. 12 (red/upper curve). weight of springs system decreased to 189 kg. The 14 torsion bars are
The same was done in the case of the basic vehicle model, where the elements of the suspension system of the full vehicle model presented in
stiffness of the torsion bars are based the blue/lower characteristics, Fig. 11a. It should be remembered that the entire suspension stiffness
from the same figure. This figure expressed also, how stiffness is increases about 2.6 times when torsion bars are replaced with the hy
extrapolated in the full vehicle model. Red/upper characteristics de perbolic springs, which is observable in Fig. 12, where torsion stiffness
scribes stiffness of the hyperbolic torsion spring, but blue/lower one characteristics are compared.
describes characteristics of the real torsion bar. Extrapolation in the Due to increased stiffness, stability of the vehicle should be
increased. This may be demonstrated when simulating each vehicle
traversing a 71 mm triangle obstacle under the right track with a vehicle
velocity of 20 km/h (see Fig. 13).
Result of this simulation is shown in Figs. 14 and 15. In the Fig. 14,
roll rate decreased and pitch and yaw rate are comparably the same.
This demonstrates another advantage of hyperbolic spring usage when
stiffness and stability of the vehicle may increase when the entire mass
of the vehicle suspension system decreases.
Fig. 15 demonstrates the vertical forces on road wheel axles for the
vehicle model with the torsion bars and the model with the hyperbolic
springs. It should be remembered that because the stiffness of suspension
with the hyperbolic springs increased 2.6 times, the hull body mass was
also increased about 2.6 times. Therefore, the forces on the wheels of a
vehicle with hyperbolic springs will have much higher values. There
fore, we only analyze the nature and shape of these forces. But even on
the basis of such a simple observation, the stabilizing effect of the sus
pension system based on the hyperbolic springs can be noticed.
We would like to emphasize that our main intention is to focus on the
hyperbolic torsion spring in the context of its application to a tracked
Fig. 12. Torsion stiffness characteristics. vehicle. If there is stabilization effect to the suspension with the
6
T. Nabagło et al. Engineering Structures 229 (2021) 111623
hyperbolic springs, this is not intended. Our intention is to show that the
new solution of hyperbolic springs can successfully replace torsion bars,
bringing with it additional benefits.
4. Conclusion
Summarizing the results of the simulation, the pros and cons of the
presented solution of the hyperbolic torsion spring should be put
forward.
Undoubtedly, the advantages of the solution are the dimensions and
the stiffness characteristic, which reflects the shape of stiffness charac
teristics of the torsion bar. These torsion bars were originally mounted in
the vehicle. Because of the higher stiffness of the torsion spring packages
over the stiffness of the traditional torsion bars, the constructors can
increase the vehicle hull mass. This is another advantage of the hyper
bolic springs. Because of their higher stiffness, which allows to signifi
cantly increase the weight of the vehicle body, the vehicle will be able to
be used as a platform for heavy guns or as a transport vehicle for
transportation of heavy equipment. As is known in the case of military
vehicles, this is a huge advantage. The hyperbolic springs packages al
lows also to resign from axial offsets of the road wheels, so the right and
left wheels may be mounted on the same axles. Usage of the spring
packages gives more space between wheels. This space may be used for
additional equipment or even additional armor. Usage of spiral springs
Fig. 13. Single obstacle being crossed.
also increases the safety level, because the torsion bars solutions are
associated with reactive elements of suspension under the floor of crew
cabin. During explosion of a charge placed under the vehicle, these
reactive elements could be harmful for the crew. The spiral spring
packages are placed on both side of the vehicle and can be covered. As
there was shown in the case of the full vehicle model simulations, the use
of hyperbolic springs can even slightly improve vehicle stability.
A disadvantage of the proposed torsion springs system is the limited
range of spring torsion, which can reach a maximum angle of 60 degrees.
Nevertheless, this range is obtained before obtaining a bumpstops po
sition. For this reason, the aforementioned disadvantage may be negli
gible under conditions assumed in the full vehicle models. A
phenomenon, which should be noted, is the material tension, which
appears on the ends of the spring arms. But it can be significantly
reduced by using the appropriate design solutions.
References
[1] ACSI Adams ver. 2016, Help Documentation assoc. with Adams Co-Simulation
Interface. MSC.Software, 2016.
[2] ADAMS ver. 2016, Help Documentation assoc. with ADAMS/View and Solver.
MSC.Software, 2016.
[3] Campanelli M, Shabana A, Choi J. Chain vibration and dynamic stress in three-
dimensional multibody tracked vehicles. Multibody Syst Dyn 1998;2:277–316.
https://doi.org/10.1023/A:1009758701296.
[4] Jurkiewicz A, Nabagło T, Kowal J, Apostoł M. A new suspension system of an
autonomous caterpillar platform. J Theoret Appl Mech 2014;52:857–67.
[5] Lankarani HM, Nikravesh PE. Continuous contact force models for impact analysis
in multibody systems. Nonlinear Dyn 1994;5:193–207.
[6] Lee K. A numerical method for dynamic analysis of tracked vehicles of high
mobility. KSME Int J 2000;14:1028–40.
[7] Łukaszewicz A. Modelling of solid part using multibody techniques in parametric
Fig. 15. Comparison of vertical forces on wheel axles. cad systems. In: Solid State Phenomena. Trans Tech Publ.; 2009. p. 924–9.
7
T. Nabagło et al. Engineering Structures 229 (2021) 111623
[8] Mallek H, Jrad H, Wali M, Kessentini A, Gamaoun F, Dammak F. Dynamic analysis [12] Nabagło T, Kowal J, Jurkiewicz A. High fidelity model construction and its
of functionally graded carbon nanotube–reinforced shell structures with verification based on 2s1 tracked vehicle. In: Innovative control systems for
piezoelectric layers under dynamic loads. J Vib Control 2020;26:1157–72. tracked vehicle platforms. Springer; 2014. p. 205–15.
[9] Marc Mentat ver. 2016, Help Documentation assoc. with Marc and Mentat. MSC. [13] Prawoto Y, Ikeda M, Manville S, Nishikawa A. Design and failure modes of
Software, 2016. automotive suspension springs. Eng Failure Anal 2008;15:1155–74.
[10] Mellouli H, Jrad H, Wali M, Dammak F. Free vibration analysis of fg-cntrc shell [14] Reimpell J, Stoll H, Betzler JW. The automotive chassis: Engineering principles,
structures using the meshfree radial point interpolation method. Comput Mathe 1995.
Appl. 2020;79:3160–78. http://www.sciencedirect.com/science/article/pii/ [15] Ryu H, Bae D, Choi J, Shabana A. A compliant track link model for high-speed,
S0898122120300316, doi: 10.1016/j.camwa.2020.01.015. high-mobility tracked vehicles. Int J Num Methods Eng 2000;48:1481–502.
[11] Nabaglo T, Kowal J, Jurkiewicz A. Construction of a parametrized tracked vehicle [16] Shimoseki M, Kuwabara T, Hamano T, Imaizumi T. FEM for Springs. Springer
model and its simulation in msc. adams program. J Low Freq Noise Vib Active Science & Business Media; 2003.
Control 2013;32:167–73. [17] Taktak M, Dammak F, Abid S, Haddar M. A finite element for dynamic analysis of a
cylindrical isotropic helical spring. J Mech Mater Struct 2008;3:641–58.
[18] Wong J. Dynamics of tracked vehicles. Vehicle Syst Dyn 1997;28:197–219.