Propulsion and Power Research 2013;2(1):3849

http://ppr.buaa.edu.cn/

Propulsion and Power Research

www.sciencedirect.com

ORGINAL ARTICLE

Stress and deformation of rocket gas turbine disc under

different loads using nite element modelling
Amr Elhefny, Guozhu Liangn
School of Astronautics, Beihang University, Beijing 100191, China
Received 27 September 2012; accepted 23 October 2012
Available online 21 March 2013

KEYWORDS
Rocket turbine;
Turbine disc;
Stress;
Deformation;
Finite element (FE)

Abstract Gas turbine discs have numerous applications in the aerospace industry, such
as in liquid rocket engines. In this study, the stresses and deformations of a turbine disc
were studied. The goal was to highlight the stress and deformation distribution to assist
in the design of a disc as well as to demonstrate the importance of using nite element
(FE) analysis in simulating an actual design case. Then, to present the real model, a twodimensional (2D) axisymmetric model for a non-uniform disc was analysed using FE
analysis. The stresses and deformations developed as a result of the disc operating
conditions at high rotational speeds and thermal gradients were evaluated using two
types of heat transfer modesconduction and convection, taking into consideration the
material behaviour at elevated temperatures. The FE model revealed that the weight of
the disc should be reduced optimally by using a non-uniform thickness because this
results in a huge increase in the applied stresses. The greatest stresses in the disc result
from the thermal load caused by conduction, and they are located at the centre of the
disc. In addition, an analytical method was used to evaluate and predict the stresses
along the disc, and it gave a good estimate of the stress values compared to the FE
model. Based on this estimate, a parametric study was conducted for a range of

Corresponding author.
E-mail address: [email protected] (Guozhu Liang).

Peer review under responsibility of National Laboratory for

Aeronautics and Astronautics, China.

2212-540X & 2013 National Laboratory for Aeronautics and Astronautics. Production and hosting by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jppr.2013.01.002

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

39

rotational velocities under high temperature loads for a series of disc radii. Finally, it was
found that this method can be used for the preliminary design of different turbines.
& 2013 National Laboratory for Aeronautics and Astronautics. Production and hosting by Elsevier B.V.
All rights reserved.

1. Introduction

2. Description of gas turbine disc

Rotating discs are historically of interest to designers in

the aerospace industry because of their vast range of uses.
Gas turbine discs are an important example of such
applications. In liquid rocket engines, rotating discs are
simultaneously subjected to mechanical and thermal loads.
A disc may be under internal pressure because of being
shrink-tted onto its mounting shaft. In addition, the
blade effects installed on its outer periphery cause an
external load to be applied to its outer edge. During
operation, the disc rotates with a signicant angular
velocity as a result of the hot gases passing through the
blades, which create a variable temperature eld on the
disc. In one study, a nite element (FE) analysis was
performed to examine the damage mechanisms of a
turbine disc, along with the critical high stresses caused
by excessive rotational speeds [1]. Another study utilised
an analytical solution to show the magnitudes of the
circumferential stress components with respect to the
radial stress components in a rotating disc [2]. Other
studies used an analysis procedure based on modelling the
disc as a series of connected constant thickness rings. In
this investigation, the criterion of failure was shown to be
the effective stress distribution at the state of yield [3].
Another analytical procedure was performed to evaluate
the elastic stresses and strains on the disc [4,5]. Threedimensional (3D) simulations of rotating metal discs were
also used to evaluate the sensitivity of the disc stability to
material parameters [6]. Furthermore, the axisymmetric
deformations of a rotating disc with variable thickness
were explored [7].
In this work, the FE technique was used to analyse
the stresses and deformations caused by different types
of loads, separate and combined. The stress results were
compared using a simplied analytical routine.

This section provides outlines of the disc geometry,

material, and operating conditions.

2.1. Disc geometry

Figure 1 shows an illustration of the turbine disc under
investigation. This turbine disc has a non-uniform thickness that varies from 0.008 m to 0.014 m, with an outer
diameter equals to 0.180 m. The detailed dimensions of
the non-uniform disc are shown in Figure 2. This disc
also has a hub where the turbine shaft is attached.

2.2. Disc material

A chromenickel steel alloy is the current material
used for the gas turbine discs. The material properties
were taken to vary with temperatures from 0 1C up to
800 1C [8], as shown in Table 1.

Figure 2 Detailed dimensions of disc cross section (unit: m).

Data

Temperature/1C
0

100

200

300

400

600

800

K
a/106

22
9.32

22
9.44

22
9.64

22
9.77

24
9.87

24
9.97

26
10.00

Data

Temperature/1C

E
n

Figure 1 Gas turbine disc.

Engineering constants.

Table 1

25

95

205

315

425

595

815

200
0.293

195
0.297

185
0.302

175
0.308

166
0.314

154
0.324

136
0.336

Where E: Youngs modulus (GPa)

n: Poissons ratio
K: Thermal conductivity (W/(m  1C))
r: Density, r 7835 kg/m3
a: Coefcient of thermal expansion, a 10  106 1C
CP: Specic heat, CP 460 J/(kg  1C)

40

Amr Elhefny, Guozhu Liang

2.3. Disc operating conditions

The turbine was operated at a rotational speed of 380
revolutions per second (rps) and a total power equal to
750 kW. The turbine blades, which were attached to the
outer boundary surface of the disc, are not presented in
Figure 1, but their mechanical and thermal effects on
the disc were taken into consideration. The mechanical
effect was a result of the total centrifugal force caused
by the blades, which was simulated by a uniformly
distributed pressure of 45 MPa (outward direction) at
the outer boundary surface of the disc. This simplication ignores the stress concentration at the connection
area between the disc and blades. Although it was
appropriate in evaluating the strength of the disc,
a 3D elasticplastic analysis should be employed for
a fatigue calculation for the blade. On the other hand,
the thermal effect that originated from the attached hot
blades was represented by an abrupt high temperature
of 600 1C at the outer boundary surface of the disc.
Because of thermal conduction, heat was transferred
from the outer boundary surface to the centre of the
disc for the subsequent 100 s, representing one operational life cycle of the turbine. For the torque transmission, the turbine shaft was shrink-tted onto the disc
hub. The amount of interference between the inner
diameter of the hub and the outer diameter of the shaft
was 0.0001 m and extended 0.01 m along the length of
the hub. Analytically, this induced a contact pressure of
26 MPa. Although the shaft was also not shown, to
assess the level of stresses arising from the shrink t, this
value was simulated as a pressure load on the inner
surface of the hub.

3.1.2. Loads and boundary conditions

Four different types of loads were applied to the model:
a rotational load, shrink-tted load, blades outer boundary
load, and thermal load. The rotational load was simulated
by an inertial velocity in the z-direction of 380 rps. The
shrink t load caused by the shaft is represented by a
pressure of 26 MPa, and it was applied to the inner surface
of the hub. The load from the blades was represented by a
uniformly distributed pressure of 45 MPa at the outer
boundary surface of the disc. Finally, the thermal load
was represented by a temperature of 600 1C at the outer
boundary surface of the disc. For symmetrical boundary
conditions for the displacement of the disc, all of the nodes
at the inner vertical surface of the hub were xed in the
z-direction, which represented the actual xation method
for the disc to the turbine shaft.
3.1.3. Results and discussion
The PATRAN FE package was used to perform
linear static analyses to solve ve different load cases.
These load cases were the rotational load (load case 1),
shrink-tted load (load case 2), blades outer boundary
load (load case 3), temperature gradient as a thermal
load (load case 4), and combination of all the loads
(load case 5). The results in terms of the Von Mises
stress distributions are shown below.


3. Finite element modelling
3.1. Real turbine disc modelling
3.1.1. Meshing
Because of the symmetry of the problem (loads and
geometry), the disc was simulated by a two-dimensional
(2D) axisymmetric cross-section. The disc was modelled
using the commercial multi-purpose FE software package
PATRAN. The geometry of the 2D non-uniform disc was
meshed with six-node solid triangular elements having two
translational degrees of freedom (DOF) per node (UX
and UZ), and each having a global edge length of 0.002 m.
The total number of elements used was 639. The mesh and
element shape are illustrated in Figure 3.

Figure 3 Meshing of 2D non-uniform real disc model.

Load case 1
For the rotational load, as shown in Figure 4, the
maximum stress of 238 MPa occurred at the centre
of the disc, and the maximum displacement of
5.20  105 m occurred at the discs outer radius.
The shape of this deformation was caused by the hub.
Load case 2
For the shrink-tted load, as shown in Figure 5, the
maximum stress of 190 MPa occurred at the hub of
the disc, and the maximum displacement of
7.60  105 m occurred at the discs outer radius.
Load case 3
For the blades outer boundary load, as shown in
Figure 6, the maximum stress of 89 MPa occurred at
the lower surface of the disc, and the maximum
displacement of 3.52  105 m occurred at the discs
outer radius.
Load case 4
A thermal transient analysis was performed to dene
the temperature gradient along the disc at different
time intervals during operation. Figure 7 shows the
variation in the maximum thermal stresses with
the operating time at three selected positions along
the disc and the temperature variation at the centre
of the disc over time under two different types of
thermal loads: conduction only and conduction plus
convection. These three positions were at the centre
(r 0 m), middle (r 0.045 m), and outer radius
(r 0.09 m) of the disc. A transient thermal analysis
was conducted not only for 100 s, which represented

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

Figure 4 Load case 1: rotational load.

Figure 5 Load case 2: shrink-tted load.

Figure 6 Load case 3: blades outer boundary load.

Figure 7 Thermal stress along disc and temperature of discs centre under conduction and convection.

41

42

Amr Elhefny, Guozhu Liang

one life cycle of the turbine, but also extended for

150 s to study the behaviour of the stresses for a
longer operating time. First, from the conduction
heat transfer mode shown in this gure, it is clear
that during the rst 10 s, the value of the thermal
stress at the outer radius of the disc was 694 MPa.
This value was much higher than the value at the
centre because of the concentration of excess heat in
only 0.02 m of the disc from the outer boundary
surface, as shown in Figure 8. Then, as time passed,
this heat was conducted and distributed along the
disc. Therefore, the value of the stress at the outer

radius decreased until reaching the value of 350 MPa

at the end of the operating time. Furthermore, the
stress at the discs centre started with a small value of
162 MPa and increased to a maximum value of
399 MPa at 100 s before decreasing. In addition, at
the discs middle point, the stress started with a value
of 180 MPa, increased to a maximum value of
306 MPa at 60 s, and then decreased. In addition,
the temperature of the discs centre remained at 0 1C
for the rst 40 s and then steadily rose to 45 1C
at 100 s. It then increased until reaching its steady
state. Second, for the convection heat transfer mode

Figure 8 Temperature distributions in disc at different operating times under conduction and convection. (a) Time 10 s (conduction),
(b) time 10 s (conductionconvection), (c) time 50 s (conduction), (d) time 50 s (conductionconvection), (e) time 100 s (conduction) and (f) time 100 s (conductionconvection).

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

43

Figure 9 Load case 4: thermal load at different operating times under conduction only. (a) Time 10 s (conduction), (b) time 50 s
(conduction) and (c) time 100 s (conduction).

44

Amr Elhefny, Guozhu Liang

combined with the conduction to analyse the effect of

using another type of heat transfer mode on the
stresses, it was assumed to be a forced convection
mode with a convection coefcient h 200 W/
(m2  1C), and the temperature of the hot gases
surrounding the disc during its operation was 200 1C.
Actually, under real operating conditions, these values
will be larger because of the high rotation speed of the
turbine in a hot medium. Therefore, these assumed
values were acceptable for the analysis.
An inspection of the curves in Figure 7 shows that the
behaviour of the change in the thermal stresses is the
same in both modes, but a lower value is seen in the
convection mode. This reduction in the stress level for
all of the points along the disc was caused by the excess
quantity of heat that had been added from the
surrounding hot gases to the disc by convection, which
increased the rate of temperature change for all of the
points. This decreased the temperature gradient and
therefore reduced the thermal stresses. Based on this,
the temperature of the discs centre remained at 0 1C
for the rst 5 s, and then, it steadily rose to 225 1C at
100 s. The stress at the discs centre reached its
maximum value at 60 s, which is shorter than the time

Figure 10

in the case of conduction only. This was also repeated

at the discs middle point, where the maximum value
was reached at 40 s. It was noticed that the conduction
mode played the predominant role in affecting the
stresses all over the disc. From this point of view,
Figure 9 shows the thermal load of the disc at three
different times10 s, 50 s, and 100 sto represent the
state of stress at the beginning, midway through, and
end of the operation, respectively when utilising the
conduction heat transfer mode only.
Load case 5
The combination of four different loads (load case 5)
was applied to the model at the start and end of the
operation (10 s and 100 s), using the maximum conduction thermal load. The results for the stress and
displacement are shown in Figures 1012. Based on
the FE results, the maximum stresses for the different
load cases were tabulated. Table 2 summarises the
values and positions of the maximum Von Mises
stresses applied to the disc and the maximum displacements caused by different types of loads at the
end of 100 s. It is observed from this table that
the thermal load had the greatest effect on the disc.
The maximum stress caused by the thermal load

Load case 5: stress caused by rotation, shrink tting, blades, and conduction thermal loads at 10 s.

Figure 11 Load case 5: stress caused by rotation, shrink tting, blades, and conduction thermal loads at 100 s.

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

Figure 12

Table 2

45

Load case 5: displacement caused by rotation, shrink tting, blades, and conduction thermal loads at 100 s.

Maximum stresses and displacements caused by different loads at 100 s.

Type of loads

Max stress
/MPa

Position of max stress

Max displacement
/m

Position of max displacement

Rotation (load case 1)

Shrink tted (load case 2)
Blades outer boundary (load case 3)
Thermal (load case 4)
Combining all the loads (load case 5)

238
190
89
432
800

Disc centre
Disc hub
At radius r 0.02 m
Disc centre
Disc centre

5.20  105
7.60  105
3.52  105
3.32  104
3.94  104

Disc
Disc
Disc
Disc
Disc

accounted for about 50% of the maximum stress

from all the loads (Figure 11).
In addition, the maximum stress caused by the rotation load was about 25% of the total maximum
stress. The maximum stresses for the rotational and
thermal loads occurred at the centre. Moreover,
for the shrink-tted load, the maximum stress values
were totally localised in the hub. On the other hand,
as shown in Figure 10, for the rst 10 s of operation, a
maximum stress of 600 MPa was observed at the
outer radius of the disc. The real operating conditions
represented in load case 5 proved that the maximum
stresses also occurred at the discs centre. These
detailed stress analyses for a disc subjected to different
types of loads show the importance of using FE
analysis in simulating a real design case because it
takes into consideration the non-uniform complex
shape and real operating conditions.

radius
radius
radius
radius
radius

as those in the non-uniform real disc model. To study the

effect of mass removal on the stresses when using a nonuniform disc thickness, the following four models were
analysed.

3.2. Effect of mass removal from disc on stresses

The main reason in using a non-uniform disc thickness
is to reduce the weight of the turbine. Weight is a major
parameter for optimisation when designing and manufacturing turbine discs. Because removing material will
mainly affect the rotational load, the rotational load was
simulated using an inertial velocity in the z-direction of
380 rps, and it was the only load applied for all of these
models. In addition, the same element type, shape,
material properties, and boundary conditions were used

outer
outer
outer
outer
outer

Disc with uniform thickness

Figure 13 shows a uniform disc. Because of the
symmetry of the problem (loads and geometry), this
uniform disc was simulated using a 2D rectangle, in
which the length of the rectangle represents the
radius (0.09 m) of the disc, while the height of the
rectangle represents its thickness (0.014 m). The
maximum stress at the discs centre was 194 MPa,
the deformation was in the radial direction, and the
stresses were distributed uniformly along the discs
radius.
Disc with two thicknesses
Figure 14 shows a disc with a thickness step from
0.014 m to 0.008 m. The maximum stress at the
discs centre was 189 MPa, and the deformation
was in the radial direction.
Disc with two thicknesses and a hub
Figure 15 shows a disc with a thickness step from
0.014 m to 0.008 m and a hub. The maximum stress
at the discs centre was 196 MPa, and the deformation was in the radial direction but bent towards
the hub.
Disc with two thicknesses and a central mass
Figure 16 shows a disc with a thickness step from
0.014 m to 0.008 m and a central mass. The maximum

46

Amr Elhefny, Guozhu Liang

Figure 13

Radial stresses with deformed shape caused by rotational loads for uniform disc.

Figure 14

Figure 15

Radial stresses with deformed shape for disc with two thicknesses.

Radial stresses with deformed shape for disc with two thicknesses and hub.

stress at the discs centre was 244 MPa, and the

deformation was in the radial direction.
Table 3 summarises the masses of the different disc
geometries and the corresponding maximum stresses and
displacements caused by the rotational load. From this
table, it is clear that the difference in mass between the
discs with the uniform and non-uniform thicknesses was

about 0.5 kg, and this mass removal resulted in an increase

in the stress of about 25%. Moreover, the central mass
had the main inuence on the high stress values at the
discs centre. In addition, the hub caused the direction of
deformation of the disc to be bent towards the hub, with a
maximum displacement of 4.67  105 m. Finally, the FE
model revealed that reducing the weight of the disc by
using a non-uniform thickness should be optimised.

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

Figure 16

Radial stresses with deformed shape for disc with two thicknesses and central mass.

Effect of mass removal from disc on stress and displacement caused by rotational load.

Table 3
Model
Disc
Disc
Disc
Disc
Disc

with
with
with
with
with

non-uniform thickness (refer to Figure 4)

uniform thickness
two thicknesses
two thicknesses and a hub
two thicknesses and a central mass

Table 4
Radius r/m

0
0.09

47

Mass/kg

Max stress at centre/MPa

Max displacement/m

2.24
2.79
2.10
2.23
2.11

238
194
189
196
244

5.20  105
4.10  105
3.71  105
4.67  105
3.74  105

Analytical results for solid uniform disc.

Rotational load

Thermal load

Rotational and thermal loads

sr/MPa

sy/MPa

sr/MPa

sy/MPa

sr/MPa

sy/MPa

197
45

197
109

412
45

412
322

564
45

564
256

4. Simple method to predict stresses

of turbine disc
4.1. Analytical solution for stress in thin rotating
solid uniform disc
A simple method is proposed to predict an approximate value for the stresses in a thin uniform turbine disc.
This method would be a good tool and a simple way to
assist designers in the preliminary design step to determine the level of stress in a disc, which would allow them
to choose the most suitable material for it. From the
detailed study of the FE stress analysis of the nonuniform real disc and according to Table 2, it was clear
that the main loads affecting the disc are the rotational
and thermal loads. Based on these results, the use of just
these loads is acceptable for a preliminary disc design.
The applied load on the disc is a combination of
rotational and thermal loads. For a thin disc rotating at
a constant speed, the stresses and deformations induced by
the centrifugal forces are symmetrical about the z-axis.
Because the disc is thin, the radial and hoop stresses are
constant throughout the thickness. Thus, there are no

Figure 17 Radial and hoop stresses caused by rotational and

thermal loads for solid uniform disc.

stresses in the z-direction. On the other hand, because of

the non-uniform temperature distribution within the disc,
thermal stress components would develop along the radial
and circumferential directions. The stress distributions
caused by combining the rotational and thermal loads are
represented by using the following formulas available in
textbooks [9,10]:


Z
B
3v
2 2 aE
sr A 2 
Tr dr
1
ro r  2
r
8
r

48
Table 5

Amr Elhefny, Guozhu Liang

Comparison between FE for non-uniform disc and analytical solution for uniform disc.

FE for non-uniform disc

Analytical solution for uniform disc

Rotational load/MPa

Thermal load/MPa

Rotational and thermal loads/MPa

238
197

432
402

800
551



Z
B
1 3v
aE
2 2
sy A 2 
ro r 2
Tr draET
r
8
r

where A and B are constants found from the boundary

conditions, r is the radius, o is the rotational speed, sr is the
stress in the radial direction, sy is the stress in the hoop
direction and T is the temperature.
If the disc is continuous from the centre to the outer
radius, then it is apparent that, unless B 0, the stresses
would become innite at r 0, so that constant B must
be equal to 0. To nd constant A, it is necessary to use
the boundary condition at the outer radius where sr 0
in the case of a solid disc with an unloaded outer
boundary (no blades are attached to the disc). However,
for loaded boundary conditions, if a number of blades
are attached to the outer radius of the disc, each of these
has a centrifugal force component, which will be present
at the outer periphery of the disc. Given the mass of
each blade, its effective centre of mass and the number
of blades, we can now compute the total centrifugal
force as a uniformly distributed load. Dividing this total
centrifugal force by the thickness of the outer boundary
area provides the value of stress in the radial direction
at the outer radius of the disc, in this case sr a0. In our
case, for a disc with a loaded boundary at outer radius
r 0.09 m, the radial stress is sr 45 MPa, and the
rotational speed o 380 rps.
For the thermal gradient, it is assumed that the
temperature variation along the disc is linear, having
the form T rab, where a and b are constants found
from the temperature boundary conditions. In our case,
it is assumed that the temperature is equal to 50 1C at
r 0 and is equal to 600 1C at r 0.09 m. The temperature constants can be calculated to be a 6111 1C and
b 50 1C. Finally, the integration for the linear temperature distribution along the disc is computed from
r 0 m to r 0.09 m to be [ar30.5br2]. Thus, the stress
distributions caused by the rotational and thermal loads
(separately and combined) can be calculated using the
same formula. Table 4 shows the values of the radial
and hoop stresses caused by these loads, where T 0 in
the case of applying the rotational load only and o 0
in the case of applying the thermal load only.
The results are demonstrated in Figure 17, which shows
that the thermal loads are higher than the rotational
loads, the maximum stress caused by the rotational load
is at the discs centre, the maximum stress caused by the
thermal load is at the discs centre and the maximum

Figure 18 Stresses in steel solid uniform disc caused by rotational and thermal loads at 550 1C temperature gradient between
disc centre and outer radius.

stress caused by combining the two loads is also located

at the discs centre, with a value of 564 MPa.
Table 5 compares the FE results for the non-uniform
disc and the analytical solution for the uniform disc
under rotational and thermal loads, where all of the
stress values were obtained at the discs centre.
An inspection of these values reveals that the simple
analytical method compares favourably with the FE
results for the separate rotational and thermal loads. On
the other hand, when combining the rotational and
thermal loads, it is shown that the stresses derived with
the analytical method are about 30% smaller than the
FE results, which proves that the simple analytical
method gives a good prediction of the stress level for
the disc in the earlier design step.

4.2. Parametric study to obtain stresses for thin

rotating solid uniform disc
Eqs. (1) and (2) were solved under rotational and
thermal loads with variations in the rotational velocity
(180530 rps) and disc radius (0.030.15 m). The resulting curves shown in Figure 18 serve as a guide for
preliminary design to estimate the maximum stresses at
the discs centre caused by rotational and thermal loads
for a uniform solid disc.

5. Conclusions
This paper presented an FE analysis of the stresses
and deformations developed in a gas turbine disc used
in a liquid rocket engine. A 2D axisymmetric disc was
created using general-purpose FE software for analysis.
The results were compared with a simplied analytical

Stress and deformation of rocket gas turbine disc under different loads using nite element modelling

solution that gave approximate values for the stresses at

the disc.
The 2D disc modelling was quite adequate as a
demonstration of a real disc taking into consideration
all of the applied loads. Several load cases were
investigated to assess the relative contribution of the
various loads imposed on the turbine during operation,
namely, the stresses caused by the rotational load,
blades outer boundary load, shrink-tted load, and
temperature gradient as a thermal load. A linear static
analysis and transient thermal analysis were conducted,
taking into consideration the material behaviour at
elevated temperatures. In addition, the deformations
of the disc caused by different types of loads and the
shape of the disc were studied. The effect on the stresses
of removing mass from the disc to reduce its weight was
presented and showed an increase in stress of about 25%.
Finally, conduction was found to be the most severe
heat transfer mode for the disc, but convection
increased the rate of temperature change and could
potentially lead to a less severe temperature gradient
across the disc. In addition, an inspection of the results
revealed that the deformation occurred in the radial
direction except when modelling the hub. In that case,
the direction of deformation was bent towards the hub.
It was also observed that the greatest stresses in the disc
resulted from the thermal loads. The thermal stresses on
the disc were quite high because of the sharp temperature gradient that existed in the turbine disc. For the
present design, the stresses at the discs centre were
shown to be higher than those at any other area of
the disc.

49

References
[1] L. Witek, Failure analysis of turbine disc of an aero
engine, Engineering Failure Analysis 13 (1) (2006) 917.
[2] H. Callioglu, M. Topcu, A. Tarakcilar, Elasticplastic
stress analysis of an orthotropic rotating disc, International Journal of Mechanical Sciences 48 (9) (2006) 985
990.
[3] H. Jahed, B. Farshi, J. Bidabadi, Minimum weight design
of inhomogeneous rotating discs, International Journal of
Pressure Vessels and Piping 82 (1) (2005) 3541.
[4] V. Vullo, F. Vivio, Elastic stress analysis of non-linear
variable thickness rotating discs subjected to thermal
load and having variable density along the radius,
International Journal of Solids and Structures 45 (20)
(2008) 53375355.
[5] M.H. Hojjati, A. Hassani, Theoretical and numerical
analyses of rotating discs of non-uniform thickness and
density, International Journal of Pressure Vessels and
Piping 85 (10) (2008) 694700.
[6] M. Maziere, J. Besson, S. Forest, Overspeed burst of
elastoviscoplastic rotating discs, part I: analytical and
numerical stability analyses, European Journal of
Mechanics A/Solids 28 (1) (2009) 3644.
[7] G.J. Nie, R.C. Batra, Stress analysis and material tailoring in isotropic linear thermoelastic incompressible functionally graded rotating discs of variable thickness,
Composite Structures 92 (3) (2010) 720729.
[8] C.L. Clark, High Temperature Alloys, Pitman, New York,
1993.
[9] A.R. Ragab, S.A. Bayoumi, Engineering Solid Mechanics,
CRC Press, Boca Raton, FL, 1999.
[10] P.P. Benham, R.J. Crawford, Mechanics of Engineering
Materials, John Wiley and Sons, New York, 1987.