135 CHAPTER 7 FINITE ELEMENT ANALYSIS OF SOLID ROLLER BEARING 7.1 MODELING AND MESHING The two dimensional finite element model of the roller bearing of specifications given in the Table 4.2 (Chapter 4) is developed for solid roller bearing in ANSYS finite element tool. Because of the symmetrical shape of the bearing, half of the model is taken for analysis. The roller is modeled with four symmetrical quarter circles of radius 7.5 mm and glued together to get symmetrical meshing of the complete model. The inner race and the shaft are combined in the model and are modeled with two symmetrical quarter circles of radius 41.75 mm and are glued together as shown in Figure 7.1. The outer race is modeled with two partial quarter annulus of inner of 56.75 mm and outer radius of 62.5 mm and glued together. The models of rollers, inner race and outer race are meshed with PLANE42 elements with the roller effective thickness of 20.4 mm as shown in Figure 7.2. All the bearing elements are meshed with the PLANE42 elements of global element edge length of 0.4 mm as shown in Figure 7.3 to get accurate results. Since the analysis is extended for contact analysis, the elements at the contact interface between inner race and roller and the interface between outer race and roller are refined with the element edge length of 0.04mm as shown in Figure 7.4. The element can be used either as a plane element (plane stress or plane strain) or as an axisymmetric element (Kogut and Etsion 2003). The element is defined by four nodes having two degrees of freedom at each node: translations in the
21
Embed
CHAPTER 7 FINITE ELEMENT ANALYSIS OF SOLID ROLLER …shodhganga.inflibnet.ac.in/bitstream/10603/17483/12/12_chapter 7.pdf · CHAPTER 7 FINITE ELEMENT ANALYSIS OF SOLID ... analysis
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
135
CHAPTER 7
FINITE ELEMENT ANALYSIS OF SOLID
ROLLER BEARING
7.1 MODELING AND MESHING
The two dimensional finite element model of the roller bearing of
specifications given in the Table 4.2 (Chapter 4) is developed for solid roller
bearing in ANSYS finite element tool. Because of the symmetrical shape of
the bearing, half of the model is taken for analysis. The roller is modeled with
four symmetrical quarter circles of radius 7.5 mm and glued together to get
symmetrical meshing of the complete model. The inner race and the shaft are
combined in the model and are modeled with two symmetrical quarter circles
of radius 41.75 mm and are glued together as shown in Figure 7.1. The outer
race is modeled with two partial quarter annulus of inner of 56.75 mm and
outer radius of 62.5 mm and glued together. The models of rollers, inner race
and outer race are meshed with PLANE42 elements with the roller effective
thickness of 20.4 mm as shown in Figure 7.2. All the bearing elements are
meshed with the PLANE42 elements of global element edge length of 0.4 mm
as shown in Figure 7.3 to get accurate results. Since the analysis is extended
for contact analysis, the elements at the contact interface between inner race
and roller and the interface between outer race and roller are refined with the
element edge length of 0.04mm as shown in Figure 7.4. The element can be
used either as a plane element (plane stress or plane strain) or as an
axisymmetric element (Kogut and Etsion 2003). The element is defined by
four nodes having two degrees of freedom at each node: translations in the
136
nodal X and Y directions. The element has plasticity, creep, swelling, stress
stiffening, large deflection, and large strain capabilities. The option available
to suppress the extra displacement shapes is used in the meshed model.
Figure 7.1 2D model of the solid cylindrical roller bearing
Figure 7.2 Meshed model of the solid cylindrical roller bearing
137
Figure 7.3 Meshed model of solid cylindrical roller
Figure 7.4 Refined mesh at contact interface
138
The geometry, node locations, and the coordinate system for
PLANE42 element are shown in Figure 7.5. The element input data includes
four nodes, a thickness and the orthotropic material properties. Orthotropic
material directions correspond to the element coordinate directions. Surface
loads are designated by a label and a key. The label indicates the type of
surface load and the key indicates where on the element the load acts. For
PLANE42 element type, the surface load list of "Pressure: face 1 (J-I), face 2
(K-J), face 3 (L-K), face 4 (I-L)" indicates that pressure loads are available on
4 faces: the line from node J to node I defines the element's face 1 (identified
on surface load commands with key = 1), and K-J (key = 2), L-K (key = 3),
and IL (key = 4). The nodal forces, if any, should be input per unit of depth
for a plane analysis (except for KEYOPT(3) = 3) and on a full 360° basis for
an axisymmetric analysis. KEYOPT(2) is used to include or suppress the
extra displacement shapes.
Figure 7.5 Geometry of the element PLANE42
The material properties of the PLANE42 elements like Young’s
modulus and Poisson’s ratio are defined for the meshed models as 2.3E5
N/mm2 and 0.3 respectively. Kogut and Etsion (2002) recommended mesh
refinement, and the mesh in the current analysis is orders of magnitude more
refine, as necessitated by mesh convergence.
X
Y
Element coordinate system(shown for KEYOPT(1)=1)
1
2
4
3
I
J
KL
X
Y
139
7.2 PARAMETERS FOR CONTACT ANALYSIS
The multi-body contact phenomena are used to perform the contact
analysis of the roller bearing so as to ensure local iteration process converged
and increase the computational efficiency (Werner Schiehlen et al 2006). Two
contact pairs are developed in the FE model for inner race contact and outer
race contact by using contact manager of the tool. Contact elements are
modeled at the contact lines of inner race, roller and outer race. For the inner
race contact, the curved lines of inner race are taken as target surface and
modeled by TARG169 elements and the curved lines at top of the rollers are
taken as contact surface and modeled by CONTA172 elements. Similarly for
the outer race contact, the curved lines of outer race are taken as target surface
and modeled by TARGE169 elements and the curved lines at bottom of the
rollers are taken as contact surface and modeled by CONTA172 elements as
shown in Figure 7.6. The mash is constructed using 1287 contact elements at
the inner race contact and 1451 contact elements at the outer race contact.
CONTA172 is used to represent contact and sliding between 2D
"target" surfaces and a deformable surface, defined by this element. The
elements are located on the surfaces of 2D solid elements with mid side
nodes. It has the same geometric characteristics as the solid element face with
which it is connected. Contact occurs when the element surface penetrates one
of the target segment elements on a specified target surface. Coulomb friction
coefficient is given as 0.3 for steel roller contact with steel inner and outer
races (Prasanta Sahoo et al 2010). The geometry and node location of the
CONTA172 element is shown in Figure 7.7.
140
Figure 7.6 Contact elements at inner and outer race interfaces
Figure 7.7 Geometry of the element CONTA172
The element is defined by three nodes (the underlying solid element
has mid side nodes). The element X-axis is along the I-J line of the element.
The 2D contact surface elements are associated with the 2D target segment
elements TARGE169 via a shared real constant set. For either rigid-flexible or
X
Y
Surface of Solid element
Contact element
Associated Target surface
Contact normal
J I
141
flexible-flexible contact, one of the deformable surfaces must be represented
by a contact surface. This element supports various 2D stress states, including
plane stress, plane strain, and axisymmetric states. The stress state is
automatically detected according to the stress state of the underlying element.
TARGE169 is used to represent various 2D "target" surfaces for the
associated contact elements CONTA172. The contact elements themselves
overlay the solid elements describing the boundary of a deformable body and
are potentially in contact with the target surface, defined by TARGE169. This
target surface is discretized by a set of target segment elements and is paired
with its associated contact surface via a shared real constant set. It can be
imposed any translational or rotational displacement on the target segment
element. Also it can be imposed forces and moments on target elements. For
flexible targets, these elements will overlay the solid elements describing the
boundary of the deformable target body (Stolarski et al 2006).
Figure 7.8 Geometry of the element TARGE169
The target surface is modeled through a set of target segments;
typically, several target segments comprise one target surface. The target and
associated contact surfaces are identified by a shared real constant set. This
real constant set includes all real constants for both the target and contact
Surface-to-SurfaceContact Element
Node-to-Surface ContactElement
M LN
M
K
I JK
n
n
I Jn
Target Segment
142
elements. Each target surface can be associated with only one contact surface,
and vice-versa. However, several contact elements could make up the contact
surface and thus come in contact with the same target surface. Likewise,
several target elements could make up the target surface and thus come in
contact with the same contact surface. The nodes are ordered so that, for a
2D surface, the associated contact elements CONTA172 must lie to the right
of the target surface when moving from target node I to target node J. The
mesh has extensively been verified for model convergence (Jackson et al
2005 and Quicksall et al 2004).
7.3 LOAD AND BOUNDARY CONDITIONS
The boundary conditions are applied such that, the displacement of
outside surface of the outer race is arrested in all directions and the
displacement of top surface of the inner race is arrested only in X direction.
The displacement of inner race in Y direction is kept free to get bearing
deformation (Biqiang Xu and Yanyao Jiang 2002). The rollers are positioned
between the inner and outer races and the degrees of freedom are kept free.
The radial load is applied on the top surface of the inner race as distributed
vertical downward load as shown in Figure 7.9. This boundary condition may
be valid for the modeling of asperity contacts for two reasons. One of the
reason is the asperities are actually connected to a much larger bulk material
at the base and will be significantly restrained there, and another reason is the
high stress region occurs near the contact, the boundary condition at the base
of the hemisphere will not greatly affect the solution because of Saint
Venant’s Principle. Since the problem is nonlinear, small load steps are used
to increment toward a solution in loading. It is important to assign a large
value of stiffness for these contact elements so that negligible penetration
occurs between the surfaces. However, using too high of a stiffness can result
in convergence problems. This work uses a stiffness that is approximately the
143
elastic modulus multiplied by a characteristic length (approximately the
contact radius of the problem).
Figure 7.9 Load and boundary conditions of the FE model
In addition, if the penetration is greater than a defined value
(tolerance), the Lagrangian multiplier method is used. This ensures that the
penetration of the converged solution is less than the assigned tolerance. The
tolerance of the current work is set to 1% of the element width. The contact
elements thus apply forces to the nodes of the elements that are in contact
(Bourdon et al 1999b). There are two ways to simulate the contact problem.
The first applies a force to the rigid body and then computes the resulting
displacement. The second applies a displacement and then computes the
resulting contact force. In both methods, the displacement, stress, and strain in
the elastic body can be determined, as well as the contact pressure. In this
144
work the latter approach is used, where the base nodes of the hemisphere are
displaced a distance approaching the target surface. This method is used
because the resulting solution converges more rapidly than the former. An
iterative scheme is used to solve for the solution, and many load steps are
used to enhance solution convergence (Cui et al 2008). The transient mode of
analysis is selected for various load steps and the results of roller deformation,
bending stress, and contact parameters are obtained for solid roller bearing.
7.4 FEA RESULTS OF SOLID ROLLER BEARING
Finite Element Analysis (FEA) is carried out on the solid roller
bearing model. By considering the experimental results, the critical load is
maximum for 17 numbers of rollers and the cage slip is maximum at 135 N
load and at 2000 rpm shaft speed. Hence the 17 number of roller bearing is
taken for finite element analysis. Initially the rollers are positioned so that one
of the rollers is placed exactly at the load line of the bearing and the other
rollers are placed with equal angular positions of 21.176o at both the sides of
the loaded roller as shown in Figure 7.10. The load and boundary conditions
of the bearing are applied and the results of deformation, contact status and
contact pressure are obtained. Then the rollers are positioned so that the
loaded roller is placed at the angular position with the increment of 1o from
the previous position and the other rollers are placed accordingly. The same
procedure is used until the angular position of roller reaches 10.588o, so that
two rollers are loaded and are equally spaced from the load line. The bearing
deformation for various angular positions of the rollers is obtained by finite
element analysis and the sample results are given in Table 7.1. The contact
status which includes far open, near contact, sliding, and sticking region of
roller positioned at equal angle of 10.588o from load line is shown in
Figure 7.11.
145
The bearing deformation against the applied radial load for various
angular positions of rollers is plotted as shown in Figure 7.12. It shows that
the increased radial load increases the deformation of bearing for any given
angular positions of the roller. It is observed that the deformation of bearing
for the roller position at 0, 1, 2, 3, 4 and 5o are same for any given radial load.
The deformation of bearing for the roller position of 6o is reduced
significantly from the previous value. Hence it is concluded that, for the given
radial clearance of 0.05 mm, only one roller is loaded for the angular position
ranging from 0o to 5o from the load line. And maximum deformation is
obtained at these positions. When the angular position of the roller reached 6o,
two rollers located near the load line carries the bearing load and the
deformation is decreased because the applied load is shared by two rollers.
The bearing deformation is further reduced for the same applied loads with
increasing angular position of the roller from the load line because of the load
distributed to two adjacent rollers. The roller located near to the load line
carries more load than the other roller. It is also observed that the bearing
attains minimum deformation for the given radial loads at 10.588o because the
rollers are equally spaced and equally shares the bearing load. Hence the
specific roller load which is defined as the load acting on individual roller is
decreased and the deformation is correspondingly decreased with increasing
the angular position of the roller from load line. The maximum and minimum
deformation of bearing for the applied load of 1000 N is observed as 0.0162
and 0.0085 mm respectively. By neglecting the inertia forces, the bearing run
out for 1000 N load is expected as 0.0077 mm which is the difference
between maximum and minimum deformations (Harsha 2006). It is also
observed that the run out of the bearing is decreased with decreasing the
radial load.
146
Figure 7.10 Position of rollers in the FE model
Table 7.1 Deformation at various angular position of roller