Dynamic Analysis of Gas Turbine Engine
Dynamic Analysis of Gas Turbine Engine
Dynamic Analysis of Gas Turbine Engine
I. INTRODUCTION
Even though the gas turbine has a long history, John
The basic concepts of turbine prime mover refer back to Barber did not make the first important design until
the first known turbine by Hero of Alexandria in 130 1791 in England. The next important development was
BC. However, until 1935 a practical gas turbine has not accomplished in the early part of the nineteenth century
been produced that could compete with any degree of by Sterling and Ericson who developed hot air engines
success with the reciprocating internal combustion that led to the establishment of the sterling constant
engines and the steam plant in the fields of propulsion volume cycle and Ericson constant pressure cycle, both
and stationary power plants. For many years steam is using isothermal compression and expansion. In the
used as the working substance, and steam turbines have middle of the nineteenth century, Joule proposed the
been developed to a high degree of efficiency. But the constant pressure cycle using isentropic compression
production of high pressure and high temperature steam and expansion. This cycle is now used as the
requires the use of a bulky and costly steam generator. fundamental cycle for gas turbine.
The gases produced in the steam generator never reach
Working principle of Gas Turbines: The basic simple
the turbine and they are only used to convert water into
open cycle gas turbine consists of a compressor,
steam. Therefore if the gases can be used in the turbine
combustion chamber and a turbine. The compressor
the costly and wasteful intermediate step of producing
takes in ambient air and raises its pressure. In the
steam could be eliminated. This idea resulted in the
combustion chamber heat is added to the air, raising its
development of a gas turbine.
temperature and hence its heat energy.
Biradar 328
availability of materials for the construction of a turbine
The product from combustion chamber then called
blade and the life required. Therefore the two factors
working fluid is expanded through a turbine to develop
that determine the performance of a gas turbine are
mechanical energy.
component efficiencies and maximum cycle
temperature.
In a gas turbine, fuel must be added not only to increase
the energy of the working media to produce the
necessary work but also to make up for losses which are
radiation, leakage, pressure drops and efficiencies of
component parts. Thus, the problem becomes one of the
largest yields for least fuel consumed.
The Thermal efficiency of an actual simple open cycle
gas turbine is heavily dependent upon the following
operating variables:
1. Pressure Ratio and Turbine Inlet Temperature
2. Turbine Efficiency
Fig. 1. Gas Turbine Open Cycle. 3. Compressor Efficiency
4. Atmospheric or Compressor Inlet Air Temperature.
Gas Turbine Cycle: The analysis of the gas turbine
This thesis is mainly based on turbine blade materials
process is based on Brayton cycle is called ideal cycle
and its life of the components so by consider the turbine
which has dry air with variable specific heats as the
blades and its performance
working fluid.
Turbine blades: These are one of the most important
The Brayton cycle is shown in Fig.1. Atmospheric air is
components in a gas turbine power plant. These are the
compressed isentropic ally to several times its
components across which flow of high pressure gases
atmospheric pressure in the compressor, which enters
takes place to produce work. A blade can be defined as
combustion chamber by injecting and burning the fuel
the medium of transfer of energy from gases to the
in the air stream. Since a normal hydrocarbon fuel
turbine rotor. The shape or profile of a blade in the
burns with a temperature around 1930 °C to 2200 °C, turbo machine cascade is of an airfoil shape. The airfoil
the major part of the air, called secondary air is utilized profile is designed using airfoil theory.
to cool the products of combustion down to the
permissible turbine inlet temperature. The combustion Typical Gas Turbine: Fig. 2 represents the Typical
in the combustion chamber is at constant pressure. Turbofan Gas Turbine engine. It has mainly 4 sections
These hot and high-pressure gases from the turbine are Compressor, Combustor, Turbine and Accessory. The
let out into the atmosphere. Approximately two thirds compressor and Turbine Rotors are identified as critical
of the power developed by the turbine is utilized to parts as they are subjected to heavy Thermo Mechanical
drive the compressor and the auxiliaries and about one loading.
third of the power of the turbine goes for useful work.
A fresh charge of air is now taken into compressor to
begin the next cycle. Therefore in the Brayton cycle,
the cycle is completed through the atmosphere and heat
on supplied by directly injecting the fuel in the air
stream. Brayton cycle is an open cycle where as the
Joule cycle is a closed cycle. In the Joule cycle the
same working substance i.e., air is circulated and
recirculated. Heat exchange is by indirect means in the
heat exchangers. Except for these differences the 1.5 Materials used in Gas Turbines:
Brayton cycle is identical to Joule cycle in all aspects.
Practical gas turbines use Brayton cycle. Losses occur
in all components. Therefore the power output for a
given amount of fuel supplied depends upon the
efficiency of the components. Further, power output of
Fig. 2. 3D view of Typical Aeroengine.
the gas turbine plant depends upon the amount of fuel
supplied. The temperature in turn is limited by the
Biradar 329
Since the Aeroengine components are subjected to High and which can be expected to determine the fatigue life
temperature environment, the material chosen should of the part. Thus Fatigue is an important failure mode in
have high strength to weight ratio along with the an Aeroengine. High Cycle Fatigue (HCF) and Low
capability to withstand its properties at high Cycle Fatigue (LCF) are the major drivers for failure in
temperature. The Rotors are made of Titanium alloy. an Aeroengine. In this project we are mainly focusing
the High Cycle Fatigue that occurs due to high
frequency vibrations at relatively large number of
cycles.
Following are the Causes for the HCF in a typical
Aeroengine:
1) Mechanical vibration arising from Rotor imbalance
2) Aerodynamic excitation occurring in upstream
vanes, downstream struts and blades whereby engine
excitation and component response frequencies
corresponding to different modes of vibration may
overlap.
Current design methods to assess the HCF life of
critical components in aircraft gas turbine engines entail
following general steps:
1) Appropriate stress analysis using FE analysis are
performed to determine mean stress level.
2) Structural Dynamics FE simulations are carried out
to determine resonant frequencies and Excitation
modes.
Fig. 3. Typical materials used in Aeroengine parts. 3) The design of component is then carried out in such
a way that it meets: a) criteria for safe life for HCF
Structural Mechanical assessment of Turbines: The
with the appropriate mean stress level (using modified
process of structural Mechanical assessment of a
Goodman diagram).b) No potential resonance related
Turbine rotor component starts from the design mission
which is usually defined in the Engine Specification. problems arise.
The design mission provides the required thrust and Mechanical design mainly includes Stress analysis and
power as a function of time. It is assumed that at this Modal analysis. Thus it is very essential to carry out the
stage the part geometry and the materials are already modal analysis in the preliminary design stage to
defined. In a first step, there are engine performance determine the dynamic characteristics of the Structure.
parameters derived from thrust and power Figure indicates the overview of mechanical design
requirements. The performance parameters consist of process in an Aeroengine.
the temperatures and pressures in the main gas path and
the spool speeds for each point of the design mission. order to meet the HCF criteria of design, frequency
Additionally, the cooling air flows, temperatures and criteria needs to be met. The frequency criteria can be
pressures in the secondary air paths are determined. In a met during design phase by plotting the Campbell
second step, these performance and cooling system diagram to verify the potential resonance. The
parameters are used as boundary conditions for the Campbell diagram is a plot of natural frequency versus
calculation of transient temperature distributions in the speed along with the forcing functions. Modal analysis
engine components. The third step is concerned with needs to be carried out at the operating speed of the
the mechanical analysis. Total stresses are calculated as engine to extract the natural frequencies.
sum of centrifugal stresses (due to part rotation), II. LITERATURE REVIEW
thermal stresses (induced by temperature gradients),
and stresses from pressures and of assembly stresses Rotating shafts are employed in industrial machines
(fits and bolt clamping). Based on the results of the such as steam and gas turbines, turbo generators,
stress analysis, the critical areas of the rotor can be internal combustion engines, reciprocating and
identified. Critical areas are those areas which are centrifugal compressors for power
exposed to the highest stresses and largest stress ranges,
Biradar 330
transmission. On account of the ever increasing demand centrifugal stiffening effect, the temperature effect and
for power and high speed transmission, the rotors of the interactions between the blade and disc. The root
these machines are made extremely flexible, which flexibility effect due to differences in clamping
makes the study of vibratory motion an essential part of condition was then investigated in detail. Aerofoil
design. The shafts of these machines are subjected to untwist effects due to creep were also revealed in this
torsional and bending vibrations and in some cases analysis.
unstable conditions of operation. The effect of blade root flexibility for a turbine blade
According to M.L.J.Verhees [1], the main cause of has been investigated Using advanced FE techniques,
turbine blade failure is high cycle fatigue. Fatigue including a pre-stressed and pre-creped modal analysis
failure is related to repeated cyclic loading on a for the blade. In order to make reliable predictions for
structural member. The fatigue life of a structural various loading conditions, the FE model was validated
member i.e. the number of load cycles it can survive is with experimental test results and close agreement was
in general determined by the magnitude of the stress achieved for all the available test cases. Thermal and
cycles. The exact relation between the magnitude of the centrifugal loads have inherent effects on the vibration
stress and the fatigue life depends on the material characteristics of the blade. For a simple clamped
properties of the structural member. In this research condition, the natural frequency tends to increase with
M.L.J.Verhees considered the turbine blade as a Euiler- the centrifugal stiffening effect and decrease with the
Bernouli beam and he find out the modal analysis for thermal effect, leading to an overall decrease in the
the turbine blade by considering the clamped and natural frequency when the two effects are combined.
unclamped ends and finally he compared the results According to Achuta Rio [3], Rotor dynamics is a
with Euiler-Bernouli beam. In principle there are two collective term for the study of vibration of rotating
ways in which the failure of turbine blades due to parts found in a wide range of equipment including
fatigue problems can be eliminated. Those are a correct turbines, power stations, machine tools, automobiles,
structural design and the prevention of cyclic loading. home appliances, aircraft, marine propulsion systems,
The correct design of a structural member can usually medical equipment and more. In these applications,
eliminate or dramatically reduce fatigue problems. For resonant vibration in which mechanical systems can
a turbine blade this is not always possible. The design oscillate excessively when excited by harmonic loads at
of a blade is usually constrained by aerodynamic their natural frequencies is of particular concern. These
properties, weight, rotor length, etc. which can make large-amplitude vibrations can bend and twist rotating
the elimination of fatigue problems through design shafts, leading to premature fatigue failure in these
modifications. components as well as bearings and support structures.
According to J.Hou and B.Wicks [2], turbine Blades Also, deformation of shafts and other components can
have been traditionally modeled as a cantilever cause rotating systems to impact adjacent parts in which
supported beam in his research blade root flexibility clearances are tight, causing potentially catastrophic
represented by using springs attached to one end of the damage in high-speed equipment. Analysis of rotating
beam. In these assumptions the analytically systems typically involves the study of many different
intermediate interactions between the blade and disks variables related to vibration including the critical
fit-trees are simplified in order to achieve a linear rotational speeds that set up natural-frequency
vibration system. According to his research, vibration is resonances, the response of the entire system to
generally recognized as one of the most significant unbalanced loads and instabilities, deflection of the
causes of high cycle fatigue failure in gas turbine shaft during vibration, torsional vibration in which
engines. The blade root flexibility often cannot be shafts also twist around their axes, and flow-induced
determined easily in the assessment of blade vibration oscillations produced by fluids moving through the
behavior and exclusion of this effect may lead to a false system. Calculation of these and other vibration-related
prediction of vibration characteristics. His report variables can be performed in ANSYS Mechanical
outlines a study of root flexibility and aerofoil untwists software using some of the most advanced rotor
effects on the vibration characteristics of a turbine blade dynamics simulation capabilities available in
using 3-D finite element analysis, together with commercial finite element analysis (FEA) codes. Rotor
laboratory testing. The vibration characteristics of the dynamics usually is best studied in the rotating frame of
blade were analyzed and the predicted results were reference, in which Coriolis terms are used in the
correlated with laboratory test results at ambient equations of motion to describe rotational velocities and
temperature. The vibration characteristics of the blade accelerations.
assembly during engine operation were then predicted
through a pre-stressed modal analysis, including the
Biradar 331
Introducing these Coriolis terms for static, modal, The upstream and downstream excitation can be
harmonic and transient analysis provides a modified explained with the help of an example. Suppose a
equation of motion: [M], [C] and [K] are the structural particular rotor stage is having 7 stator vane count on
mass, damping and stiffness matrices, respectively. the upside and 14 stator vane count on the downside of
[Kc] is the spin softening matrix, and [G] is a rotor, the excitation for the Rotor is 7th Harmonic and
“damping” matrix contribution due to the rotation of the 14th Harmonic. The 1st Harmonic is the fundamental
structure or the Coriolis term. This modified equation frequency and the higher harmonics are the multiples of
of motion is at the foundation of performing the most these frequencies. These forcing harmonics vary
common types of rotor dynamics analyses. depending on the surrounding vane count. It depends on
the Rotor stage considered for the analysis.
III. PROBLEM DEFINITION
Geometric modeling of the NASA Rotor 67: The
Scope for the present work: The scope for the present
work includes analytical assessment of the dynamic NASA Rotor 67 is a transonic fan. The first stage of a
characteristics of the Gas Turbine Rotor. The dynamic NASA Lewis designed two-stage axial-flow fan was
characteristics include the Eigen values (Natural used as the Geometry in this project.
frequencies) and the Eigen Vectors (Mode shapes). The The rotor consists of 22 low aspect ratio blades and is
analytical assessment is done by using the Commercial designed to produce a total pressure ratio of 1.63 for a
FE package ANSYS software. The Dynamic Response mass flow rate of 32.25kg/sec with a tip inlet relative
of the Rotor subjected to centrifugal loading is Mach number of 1.38 at the design point of speed
computed. In the present work the Geometry used for 16,043 rpm which gives design tip speed of 429 m/sec.
analysis purposes corresponds to NASA Rotor 67. The The aspect ratio for these blades is around 1.56. The
Geometry is created using the NASA design code rotor inlet and exit tip diameters are 51.1 and 49.5 cm
generated co-ordinates. The geometry created is respectively, and the rotor inlet and exit hub/tip radius
validated with the Actual Geometry of the NASA Rotor ratios are 0.375 and 0.475 respectively. The running
with the experimental data. The validation is done by rotor tip clearance is 0.5 mm. Fig.3.2 represents the
comparing analytical frequencies obtained with the Rotor Blade manufacturing co-ordinates system used by
experimental data of the NASA Rotor. The Dynamic the NASA design code. Fig.3.3 represents the Rotor
Analysis is carried out considering the prestressed blade sections at three span wise locations. Table.AII.1
effect due to centrifugal loading. Analysis is also represents the Rotor blade manufacturing co-ordinates
carried out without the prestressed effect. The effect of generated by the NASA design code.
response with and without pre-stress effect is being
studied to verify the impact on Campbell diagram. The The Geometry is created in Ansys V11.0. The local co-
analysis is carried out at various speeds to see the effect ordinate system is created at each of the Rotor sections.
on natural frequencies and on the resonance condition. The Blade manufacturing co-ordinate system is used as
The Dynamic analysis is carried out at following a reference system at each of the cross sections. The co-
RPM’s: ordinates are read into Ansys V11.0 and then rotated
with the proper given stacking angle. Using these co-
1) 0 RPM, Room temperature ordinates splines are generated in Ansys at each of the
2) 5000 RPM cross sections. Using these splines volumes are built to
3) 10000 RPM create the required Rotor blade profile. Table.AII.2
4) 16043 – Operating speed of the Rotor. represents the Ansys generated blade co-ordinates used
Finally a Campbell diagram is plotted to see the impact for the creation of blade.
on the resonance due to various speeds with and
without the prestressed effect. The Frequencies and
mode shapes are provided for all the cases of analysis.
The Campbell diagram represents the plot of speed
versus frequency along with the forcing frequencies.
The forcing frequencies are the excitation frequencies
due to rotor unbalance, upstream and downstream
vanes. The upstream and downstream excitation
depends on the number of the vane count on the
upstream and down stream of the Rotor. The resonance
is identified at the crossing point of the forcing Fig. 4.
frequencies and the natural frequencies.
Biradar 332
FE modeling of the Rotor: Since all the Rotors are IV. SOFTWARE DESCRIPTION AND
assumed to behave in a similar manner one rotor is ANALYSIS PROCEDURE
considered for the Dynamic analysis. One rotor
Overview of ANSYS: ANSYS is a general-purpose
corresponds to 360/22 Sector. Since the blade Hub is
finite element-modeling package for numerically
rigid portion the hub is not modeled in the present
solving a wide variety of mechanical problems. These
analysis. Also we are focusing the frequencies of the
problems include: Static/Dynamic Structural Analysis
blade. The Geometry created in Ansys is meshed with
(both linear and non-linear), Heat Transfer and Fluid
the Linear Hexahedron elements. The element types
Problems, as well as Acoustic and Electro-Magnetic
used for this purpose are Solid 45. The mesh size used
problems.
for the model is around 0.038”. SOLID45 is used for
the 3-D modeling of solid structures. The element is In general, a finite element solution may be broken into
defined by eight nodes having three degrees of freedom the following three stages. This is a general guideline
at each node: translations in the nodal x, y, and z that can be used for setting up any finite element
directions. The element has plasticity, creep, swelling, analysis.
stress stiffening, large deflection, and large strain
capabilities. While meshing the Geometry a 2D regular • Preprocessing: defining the problem
hexa mesh is prepared using which the 3D elements are The major steps in preprocessing are given below:
being generated. The 2D elements used for this purpose
are MESH 200 elements that have no effect during • Define key points/lines/areas/volumes
finite element computation. • Define element type and material/geometric
properties
• Mesh lines/areas/volumes as required
The amount of detail required will depend on the
dimensionality of the analysis (i.e. 1D, 2D, axis-
symmetric, 3D).
• Solution: assigning loads, constraints and solving
Here we specify the loads (point or pressure),
constraints (transnational and rotational) and finally
solve the resulting set of equations.
• Post processing: further processing and viewing of
the results: In this stage one may wish to see
• Lists of nodal displacements
• Element forces and moments
• Deflection plots
Fig. 5. Sectional volumes of the blade geometry before • Stress contour diagrams
meshing.
WHAT IS ANSYS? ANSYS IS FINITE ELEMENT
ANALYSIS SOFTWARE THAT ENABLES ENGINEERS TO
PERFORM THE FOLLOWING TASKS.