OPTIMIZATION OF RADIAL FAN IMPELLER

USING FINITE ELEMENT ANALYSIS

A PROJECT REPORT

Submitted by

KISHORE KANNA.B 40401114020 MOHAMMED MOHAIDEEN.M 40401114033PANDIARAJ.T 40401114039SATHISH KUMAR.K 40401114049

in partial fulfillment for the award of the degree

of

BACHELOR OF ENGINEERING

in

MECHANICAL ENGINEERING

B.S.ABDUR RAHMAN CRESCENT ENGINEERING COLLEGE,

CHENNAI-48

ANNA UNIVERSITY: CHENNAI 600 025

MAY 2005

BONAFIDE CERTIFICATE

This is to certify that the project work entitle “OPTIMIZATION OF

RADIAL FAN IMPELLER USING FINITE ELEMENT ANALYSIS”

is a Bonafide record of the work done by

KISHORE KANNA.B - 40401114020

MOHAMMED MOHAIDEEN.M - 40401114033

PANDIARAJ.T - 40401114039

SATHISH KUMAR.K - 40401114049

Students of B.E., (Mechanical Engineering) of B.S ABDUR

RAHMAN CRESCENT ENGINEERING COLLEGE, Chennai at Ranipet.

During the period from 31-01-05 to 28-02-05.

We wish them all the success in their future endeavour.

For BHARAT HEAVY ELECTRICALS LIMITED

Mr. R.BABU M.Tech (IIT-Madras) Mr.S.PARAMANANTHAM

Deputy Manager (Fans) H.R.D.Officer

BHEL-BAP BHEL-BAP

Ranipet Ranipet

ANNA UNIVERSITY: CHENNAI 600 025

BONAFIDE CERTIFICATE

This is to certify that the project report ‘OPTIMIZATION OF RADIAL

FAN IMPELLER USING FINITE ELEMENT ANALYSIS’ is the

bonafide work of

KISHORE KANNA.B 40401114020MOHAMMED MOHAIDEEN.M 40401114033PANDIARAJ.T 40401114039SATHISH KUMAR.K 40401114049

Who carried out the project work under my supervision.

Dr.R.GANESAN Mr.P.GANESHHEAD OF THE DEPARTMENT INTERNAL GUIDE

DEPARTMENT OF MECHANICAL ENGINEERINGB.S.ABDUR RAHMAN CRESCENT ENGINEERING COLLEGE

CHENNAI-48

VIVA VOCE EXAMINATION

The viva voce examination of this project work submitted by

B.KISHORE KANNA, REGISTER NO: 40401114020

T.PANDIARAJ, REGISTER NO: 40401114039

K.SATHISH KUMAR, REGISTER NO: 40401114049

M.MOHAMMED MOHAIDEEN, REGISTER NO: 40401114033

is held on

EXTERNAL EXAMINER INTERNAL EXAMINAR

ACKNOWLEDGEMENT

AKNOWLEDGEMENT

It is our great pleasure in presenting the project work undergone at Bharat

heavy electrical limited (BHEL), Ranipet.

At this moment we wish to place on record our sincere gratitude to

prof. S.Peer Mohamed, Correspondent and Dr.K.P.Mohamed, Principal, for

their encouragement.

Our sincere gratitude to our Head of the Department Dr.R.Ganesan

and Project coordinator Mr.J.Bhaskaran for providing us with necessary

infrastructure.

Our sincere gratitude to Mr.P.Ganesh, Lecturer who has been a

resource of encouragement and guidance for our project work. We are

indebted towards him for his valuable suggestions and help without which

our project could not have been completed.

We express our sincere gratitude and thanks to Mr.S.Paramananthan,

HRDC Officer, BHEL- Ranipet.

Our sincere gratitude and thanks to our External guide Mr.R.Babu,

Deputy Manager, BHEL-Ranipet.

Our sincere thanks to all the staff members of our department for

their encouragement and guidance.

Last but not least we thank all of our friends who stood by us and

provided the moral support during the preparation of our project work.

ABSTRACT

Increasing cost of consumable materials has put an enormous pressure

on the pricing as well as the profitability of an organization. Therefore

without any compromise on quality, the variable cost has to be reduced. This

demands novel thinking and creativity for constant improvement in design,

resulting in good profits.

In this project work, an attempt has been made to optimize the design

of aerofoil bladed radial fan impeller using finite element analysis (FEA).

The optimization is done using FEA in “ANSYS – MECHANICAL

UTILITY”

In the FEA section, modeling as well as analysis was done using ANSYS

MECHANICAL UTILITY. In this section, the thickness of the every

component of the impeller was reduced and Stiffeners were provided to

curtail the large deflections in the optimized model. Stresses and deflections

were analyzed for the modified and pre modified model.

The results of the optimization were successful as 18.5% savings in net

weight after FEA optimization. So, the material cost is reduced and also the

space occupied. In FEA the stress and deflection analysis were performed

whose results were well within the limits.

ABSTRACT

CONTENTS

CHAPTER PAGE NO

ACKNOLEDGEMENT i

ABSTRACT ii

CONTENTS iii

LIST OF DIAGRAMS vi

LIST OF TABLES vii

LIST OF NOMENCLATURE viii

CHAPTER 1 INTRODUCTION

1.1 Organization Profile 1

1.2 BHEL – Boiler auxiliary plant ranipet 1

1.3 About the Project. 2

1.4 About the Fan 3

1.5 Classification of fans 3

1.6 Advantages of an Aerofoil Bladed Fan 6

1.7 Constructional features of a radial fan 7

1.8 Methods to drive the Fan 8

1.9 Fan specifications 9

CHAPTER 2 LITERATURE REVIEW

2.1 Introduction 10

2.2 Historical Background 10

2.3 What is FEA? 11

2.4 Need for Finite Element Analysis 11

2.5 The Finite Element Method 12

2.6 Concepts of FEA 13

2.7 General procedure for FEA 14

2.8 Applications of FEA 16

2.9 Setting element attributes 16

2.10 Boundary conditions 17

2.11 Material properties 18

2.12 Element type 18

2.13 Fan laws and Efficiency 22

2.14 Stress Analysis 28

CHAPTER 3 INTRODUCTION ANSYS

3.1 Introduction to ANSYS 31

3.2 ANSYS offers 31

3.3 Examples of ANSYS analysis 31

3.4 Choosing the software 32

3.5 Feature selection 33

3.6 Results required for analysis 33

3.7 Solution speed 33

3.8 Hardware availability 33

3.9 Analysis procedure 34

3.10 Procedure for static analysis 35

3.11 Modal analysis 36

3.12 Procedure for modal analysis 37

3.13 Design optimization 38

3.14 Typical examples of optimized designs 38

CHAPTER 4 METHODOLOGY

4.1. Solid model generation using preprocessor 39

4.2. Meshing contours 39

4.3. Meshing of areas 39

4.4. Defining material 40

4.5. Choosing appropriate element for analysis 40

4.6. Attributing equivalent and actual boundary 40

4.7. Solving the problem using solver 40

4.8. Viewing results 41

4.9. Studying the parameters stress and deflection 41

4.10. Modifying the geometry model by reducing the thickness 41

CHAPTER 5 MODELING

5.1 Segment generation 42

5.2 Real constants for the model 43

5.3 Mapped mesh 45

5.4 Generation of fan impeller 46

5.5 Boundary condition 47

5.6 Solution 48

CHAPTER 6 RESULT AND DISCUSSION

6.1 Static analysis 49

6.2 Optimization 49

6.3 Results for fan impeller original thickness 50

6.4 Results for optimized fan impeller 58

CHAPTER 7 CONCLUSION

7.1. Real constants for the model 66

7.2 Weight reduction 67

REFERENCES

LIST OF FIGURES PAGE NO

FIG.1- LINE PLOT OF FAN IMPELLER SEGMENT 42

FIG.2- AREA PLOT OF FAN IMPELLER SEGMENT 43

FIG.3- REAL CONSTANT NUMBERING FOR THE SEGMENT 44

FIG.4- MAPPED MESH OF THE SEGMENT 45

FIG.5-FAN IMPELLER MODEL WITH REAL CONSTANT NUMBERING 46

FIG.6-MAPPED MESH OF THE IMPELLER MODEL 46

FIG.7- CONSTRAINTS AT THE CENTRE 47

FIG.8- MESH OF IMPELLER WITH CONSTRAINTS 48

FIG.9- ORIGINAL FAN IMPELLER DEFLECTION PLOT 50

FIG.10- ORIGINAL FAN IMPELLER STRESS PLOT 51

FIG.11- ORIGINAL BACKPLATE (BOTTOM) DEFLECTION PLOT 52

FIG.12- ORIGINAL BACK PLATE (BOTTOM) STRESS PLOT 52

FIG.13- ORIGINAL BACK PLATE (TOP) DEFLECTION PLOT 53

FIG.14- ORIGINAL BACK PLATE (TOP) STRESS PLOT 53

FIG.15- ORIGINAL BLADE DEFLECTION PLOT 54

FIG.16- ORIGINAL BLADE STRESS PLOT 54

FIG.17- ORIGINAL COVER PLATE DEFLECTION PLOT 55

FIG.18- ORIGINAL COVER PLATE STRESS PLOT 55

FIG.19- ORIGINAL RING DEFLECTION PLOT 56

FIG.20- ORIGINAL RING STRESS PLOT 56

FIG.21- ORIGINAL FLANGE DEFLECTION PLOT 57

FIG.22- ORIGINAL FLANGE STRESS PLOT 57

FIG.23- OPTIMIZED FAN IMPELLER DEFLECTION PLOT 58

FIG.24- OPTIMIZED FAN IMPELLER STRESS PLOT 59

FIG.25- OPTIMIZED BACKPLATE (BOTTOM) DEFLECTION PLOT 60

FIG.26- OPTIMIZED BACK PLATE (BOTTOM) STRESS PLOT 60

FIG.27- OPTIMIZED BACK PLATE (TOP) DEFLECTION PLOT 61

FIG.28- OPTIMIZED BACK PLATE (TOP) STRESS PLOT 61

FIG.29- OPTIMIZED BLADE DEFLECTION PLOT 62

FIG.30- OPTIMIZED BLADE STRESS PLOT 62

FIG.31- OPTIMIZED COVER PLATE DEFLECTION PLOT 63

FIG.32- OPTIMIZED COVER PLATE STRESS PLOT 63

FIG.33- OPTIMIZED RING DEFLECTION PLOT 64

FIG.34- OPTIMIZED RING STRESS PLOT 64

FIG.35- OPTIMIZED FLANGE DEFLECTION PLOT 65

FIG.36- OPTIMIZED FLANGE STRESS PLOT 65

LIST OF TABLES

TABLE 2.1- ELEMENT TABLE 19

TABLE .5.2- REAL CONSTANTS FOR THE MODEL 43

TABLE.6.1- OPTIMIZATION TABLE 49

TABLE.7.1- DIMENSIONS FOR THE MODEL 66

NOMENCLATURE

N - rpm

D - Fan diameter (mm)

µ - dynamic viscosity

Ns - Specific speed

Q - Volume flow rate (m3/sec)

gH - specific energy (J/kg)

P - Pressure rise (pa)

ρ - fluid density (kg/m3)

CHAPTER 1

INTRODUCTION

CHAPTER1

INTRODUCTION

1.1 Organization Profile

M/S. BHARAT HEAVY ELECTRICAL LTD., popularly

known as BHEL is today, the largest engineering and manufacturing

enterprise of its kind among the public sector undertakings in India. The

company provides products, systems and services in the field of energy and

transportation for domestic and export markets.

The company ranks amongst the worlds top 10 organizations

engaged in the manufacturing of power plant equipment. About 50

countries, extending from USA in the west to Australia and New Zealand in

the far east are BHEL’s customers.

1.2 BHEL - BOILER AUXILIARIES PLANT - RANIPET

BHEL – Trichy launched its phase III expansion for

augmentation of manufacturing capacity to 4,000 MW for boilers and

auxiliaries at Ranipet Tamilnadu in 1982.

The product profile of BAP, Ranipet is

Fans – Radial, Axial, Impulse and Axial reaction

Electrostatic precipitators

Air preheater (Regenerative type)

These auxiliaries play a vital role in the thermal power plants.

There are 43 ancillaries established adjacent to the plant .BHEL gives by

way of technical , raw material and quality control procedure , BAP Ranipet

has technical collaboration with M/S K.K.K, West Germany for fans .

BAP at Ranipet provides direct employment to about 3,000

employees and indirect employment of over 10,000 employees. BHEL is

certified with ISO 9001 and ISO 9002 by BVQI.

1.8 About the Project.

This project has been done to predict and give the results of a

Radial fan impeller under physical operating conditions.

The Radial fan impeller is analyzed before the performance

testing and installation. Stress analysis is performed to find the maximum

stress values. These analysis are done by ANSYS.

In this project static and optimization of fan impeller thickness

have been performed during the analysis using the software.

In static analysis the maximum stress, strain values for the

required boundary conditions are found. In optimization, the impeller

thickness is reduced or optimized without changing or violating the

maximum stress values. So that the weight reduces and hence the cost of the

product also reduces.

HARDWARE REQUIREMENTS

Processor : PENTIUM III

CPU speed : 400Mhz

HDD : 20GB

Main memory capacity: 159MB

SOFTWARE REQUIREMENTS

ANSYS 5.4

1.9 About the Fan

A fan is a turbo machine used for energy transfer. It can be defined as a rotating machine with a bladed impeller, which maintains a continuous flow of air (or) gases.

Fans usually consist of a single rotor with or without a stator element and cause a rise in pressure of the flowing fluid.

PRINCIPLE OF WORKING:

The principle involved is that the mechanical energy owing to the rotation of the fan is converted into the fluid energy (in the form of pressure rise).

Fans obviously consume power as they rotate with the help of prime mover and energize the flowing fluid.

1.5 CLASSIFICATION OF FANS:

1.5.1 ACCORDING TO PURPOSE:

1. Primary Air Fan :( PA FAN)

Primary air fans supply the air needed to dry and transport pulverized coal to the furnace of direct-fired boiler.

2. Forced Draught Fan: (FD FAN)

The forced draught fans supply the air-required for the combustion of fuel and normally handle stoichiometric plus excess air required for the satisfactory burning of fuel.

3. Induced Draught Fan: (ID FAN)

The induced draught fans draw the products of combustion from the boiler while creating sufficient draught (negative pressure) in the furnace for balanced draught operation.

1.5.2 ACCORDING TO FLOW OF AIR:

1. Radial Fan:

A radial fan is a one in which the flow enters along the axis and leaves in the radial direction along the blades. It can be used for PA, FD and ID applications.

Based on the configuration of the blade with respect to the direction of rotation of the impeller (AS SHOWN IN THE FIG.) it is called backward curved, forward curved and radial bladed impeller

FIG. 1.1

2. Axial Fans:

An axial fan is a one in which the main flow is along the axis of

rotation both at entry and exit.

2 < 90 2 = 90 2 > 90

Based on the profile these fans are mainly classified into two

types namely,

I Axial Profile Impeller: (AP IMPELLER)

In this type, the impeller has a central hub which is spherical in

nature and has blades with individual shafts located along the periphery.

The hub is a high precision part which is ball turned to get a curved

smooth profile. The individual blades of the impeller are driven with

the help of hydraulic mechanism.

II Axial Non – Profile Impeller: (AN IMPELLER)

In this type, the impeller has a central hub, which is of

hemispherical nature and has blades curved at a fixed angle and welded

to the hub as in case of its radial counterpart.

Both the fans described above have an inlet guide vane (IGV)

and an outlet guide vane (OGV) along with a diffuser at the exit.

AEROFOIL BLADED RADIAL FAN - A GLANCE

An aerofoil bladed radial fan consists of blades, which are

profiled, in an aerofoil shape as shown in the figure below:

FIG .1.2

1.6 Advantages of an Aerofoil Bladed Fan:

a. Since the aerofoil is a profile curved body, it ensures a

smoother flow than a blunt body and hence no flow separation

thereby minimized losses

b. Because of higher efficiency than normal plate bladed impeller

it consumes less power and hence it is economical.

c. An aerofoil bladed fan has the higher half – load efficiency like

an axial fan and the rigidity of that of a radial fan and hence the

combined feature of both.

But the aerofoil bladed is mostly employed as primary air fan.

1.7 CONSTRUCTIONAL FEATURES OF A RADIAL FAN:

The fan as a whole can be divided in to some major sub –

assemblies.

1. SPIRAL CASING:

The spiral casing consists of two parallel sidewalls, spiral wall,

Suction Chamber and inlet cone. It is split horizontally along the shaft

axis plane; if necessary the upper portion will also be vertically split off

at the center so that impeller installation is easy. The inlet cone and the

suction chamber are welded to the sidewalls.

2. IMPELLER:

The impeller is a completely welded structure. It consists of a

center plate (or) back plate, cover plate and blades. The blades are

welded between the back plate and the cover plate. Proper welding

sequence is followed to have minimum distortion.

3. SHAFT:

The shaft is a hollow tube with 2 endpins shrunk-fit at the 2

ends is welded. Torque is transmitted through the fit and the weld is only

for securing purpose. The tube is controlled at the inside diameter. The

shaft ends are machined after welding. A flat split ring is welded on to

the shaft tube for taking up the shaft flange. The complete shaft is

dynamically balanced.

4. BEARINGS:

The impeller is mounted on pillow block bearings. One is a

locating bearing while the other is a non – locating (FREE) bearing. The

bearings are spherical roller type housed in bearing housing. Or the

bearings are of sleeve types that are selected based on the contractual

requirement and or on the basis of the selection requirement.

5. DAMPER ASSEMBLY:

This consists of a single piece casing, damper flaps, damper

bearings and the actuating mechanism. It is welded casing flanged at both

the ends. The bearing pedestals are mounted to the sidewalls by screws.

There are 3 to 5 flaps fixed by screws on to their shafts, which

are supported by pedestals providing dry lubrication. The flat shafts carry

clamping levers and feather keys transmit the adjusting torque and a

linkage connects the individual clamping levers.

6. SEALS:

The sealing for the shaft with the spiral casing consists of a

labyrinth section For axial and asbestos strip for radial sealing. The

asbestos strip ensures that the movement of the spiral casing during hot

conditions relative to the impeller wheel does not attack the fan’s

functioning. The unmachined flanges of the spiral casing are sealed with

asbestos rope.

1.8 Methods to drive the Fan

Various methods are used to the Fan

Prime movers

Electric motors (the most commonly used)

Engines

Turbines (compressed air to steam)

Compressed air jets.

1.8.1Types of motor drives

There are three ways that can be used for an electric motor to drive a Fan:

1. Belt drive

2. Direct drive

3. Gear drive

1.8.2 Types of Electric motors used to drive fans:

1. Three-phase squirrel-cage motors

2. Three-phase wounded-rotor motors

3. Single-phase, single phase induction motors

4. Single-phase, permanent-split-capacitor motors

5. Single-phase, shaded-pole motors

6. Single-phase universal motors

7. Single-phase, inside-out induction motors

1.9 FAN SPECIFICATIONS:

Backward Aerofoil Bladed Fan Application- Primary Air Fan Power- 1500 KW Plant Capacity- 250 MW Fan size- NDZV 20 BAB2 Speed- 1000 rpm Head- 985 mmmw Pressure ~ 9850 Nm/Kg Volume – 50 m3/s Material Used – Naxtra 70

CHAPTER 2

LITERATURE REVIEW

CHAPTER 2

LITERATURE REVIEW

FINITE ELEMENT ANALYSIS (FEA) :

2.1 Introduction:

Finite Element Analysis (FEA) is a computer-based numerical

technique for calculating the strength and behavior of engineering structures.

It can be used to calculate deflection, stress, vibration, buckling behavior

and many other phenomena. It can be used to analyze either small or large-

scale deflection under loading or applied displacement. It can analyze elastic

deformation, or "permanently bent out of shape" i.e., plastic deformation.

Computer is required because of the astronomical number of calculations

needed to analyze a large structure. The power and low cost of modern

computers has made Finite Element Analysis available to many disciplines

and companies.

2.2 Historical Background:

The very basics of the finite element method rose from the

advances in aircraft. It all began with Hrenikoff, in 1941 presenting a

solution to elasticity problems using “the frame work method”. This trend

continued with Courant’s paper based on piecewise polynomial interpolation

in 1943. Turner et al. derived stiffness matrices for truss, beam and other

elements and presented their findings in 1956. But Clough first coined the

term finite element in 1960.

2.3 What is FEA?

The finite element analysis is a kind of analysis in which a

complex region defining a continuum is discretized into simple geometric

shapes called finite elements. The material properties and the governing

relations are imposed on these elements and expressed in terms of unknown

values at element corners.

An assembly process duly considering the loading and

constraints, results in a set of equations. Solution to these equations gives us

the approximate behavior of the continuum.

2.4 Need for Finite Element Analysis:

Finite Element Analysis makes it possible to evaluate a detailed

and complex structure, in a computer, during the planning of the structure.

The demonstration in the computer of the adequate strength of the structure

and the possibility of improving the design during planning can justify the

cost of this analysis work. FEA has also been known to increase the rating of

structures that were significantly over designed and built many decades ago.

In the absence of Finite Element Analysis (or other numerical

analysis), development of structures must be based on hand calculations

only. For complex structures, the simplifying assumptions required to make

any calculations possible can lead to a conservative and heavy design. A

considerable factor of ignorance can remain as to whether the structure will

be adequate for all design loads. Significant changes in designs involve risk.

Designs will require prototypes to be built and field-tested. The field tests

may involve expensive strain gauging to evaluate strength and deformation.

With Finite Element Analysis, the weight of a design can be

minimized, and there can be a reduction in the number of prototypes built.

Field-testing will be used to establish loading on structures, which can be

used to do future design improvements via Finite Element Analysis.

2.5 The Finite Element Method:

In general, in the finite element method, a structure is broken

down into many small simple blocks or elements. The behavior of an

individual element can be described with a relatively simple set of equations.

However, there are two general approaches associated with the

finite element method. One approach called the force method uses the

internal forces as the unknown constraints of the problem, while the other,

the displacement method (or) stiffness method uses displacement as the

unknown.

In the finite element method the continuum is discretized into

small inter connected elements called finite elements, and these elements

have a displacement function associated with it. Each inter connected

element is linked, directly (or) indirectly to every other element through

common interfaces including the nodes and boundary line and surfaces. By

using the known stress strain properties of the material making up the

structure, one can determine the behavior of a given node in terms of

properties of every other in the structure. The total set of equations

describing the behavior of each node results in a series of algebraic

equations best expressed in matrix notation.

2.6 Concepts of FEA:

As described earlier the FEA can be used to determine the

stress and deflection of any structure under load.

According to Newton’s II law,

The force on any body due to external load is given by,

F = ma

This under equilibrium conditions the above equation can be represented in

the differential form as,

mä + cå + ka = 0

where,

a = kx

å = dx/dt

ä = d2x/dt2

So,

[m. (d2x/dt2)] + [c. (dx/dt)] + kx = 0

In matrix form is represented as,

[m] * [k] * [δ] = [f]

By solving the above matrix equations with the values given

(or) solved the values for stress and deflection can be determined easily.

2.7 General procedure for FEA:

With the advent of hi-tech computers, the FEA solutions for complex problems are made easy and simple. The general procedure for the FEA is outlined in the form of a flowchart as below:

FLOW CHART. 2.1

Pre – processor

Read the input data and identify the design constraints.Model the continuum.Identify the element type and mesh the model.Define the boundary conditions and load data.

Processor/solution

Compute element stiffness matrices.Assemble element equations.Solve equations for the conditionCompute results.

Pre – processor

Plot the resultsInterpret the results

27.1 Discretization:

It is the process by which a closed form mathematical

expression such as a function (or) a differential (or) integral equation

involving functions, all of which are viewed as having an infinite continuum

of values throughout some domain, is approximated by analogous

expressions that prescribes values at only a finite number of discrete points

(or) volumes in the domain.

2.7.2 Meshing:

A finite element model includes a mesh of nodes and

elements. The best way of creating mesh is to create the part’s geometry,

then generate a mesh on the geometry. Since the finite element model is

associated with the part, any change to the part is automatically reflected in

the nodes and elements of the mesh. Part geometry based meshes are also

used for geometry-based optimization.

There are generally two types of meshes;

1.MAPPED MESH:

It is a kind of mesh in which the points of the mesh are

arranged in a regular way all through the continuum and can be stretched to

fit a given geometry.

2.FREE MESH:

It is a kind of mesh where the points fill the space to be

considered but is not connected with the regular topology. The mesh with an

irregular structure is often referred to as an unstructured (or) free mesh.

FIG. 2.1

DIAGRAMATIC REPRESENTATION OF MAPPED AND FREE

MESH

FREE MESH

MAPPED MESH

2.8 Applications of FEA:

There are several engineering applications of FEA, but some of the notable one’s are mentioned below:

Structural analysis

Structural machines

Aerospace engineering

Solid mechanics and foundation engineering

Rock mechanics and heat conduction

Hydrodynamics and hydraulic engineering

Water resources and nuclear engineering.

2.9 Setting element attributes:

Before generating a mesh of nodes and elements, the element

attributes are to be defined.

Element type

Real constant set

Material Properties set

Element co-ordinate system

2.10 Boundary conditions:

2.10.1 Definition of Boundary condition

Boundary conditions are nothing but the constraints of the model

that is to be analyzed. The constraints may be displacement, inertias, loads

(forces, moments), temperature, fluid velocity, etc., for every model the

boundary conditions are must be specified. Without the impositions of the

boundary conditions, the element and assemblage stiffness matrices, [k] and

[k], are singular; that is, their determinants vanish and their inverse do not

exist.

The physical significance of this is that a loaded body or structure

is free to experience unlimited rigid body motion unless some supports or

kinematic constraints are imposed that will ensure the equilibrium of the

loads. These constraints are the boundary conditions.

2.10.2 Boundary conditions for the model

1.Displacement:

The rotating motor shaft is fixed in the impeller therefore the

displacement on the impeller hole is zero in all degree of freedom.

2.Angular velocity and angular acceleration:

The impeller rotates about the “z” axis at a speed of

1000rpm. Therefore

Angular velocity = (2πN)/60

= 104.7 rad/sec

The angular acceleration is also given as 9810 rad/sec2.

2.11 Material properties:

Once a mesh has been built to describe the domain occupied by

the structure, the rest of the computer model can be built. It is only at this

stage that the description of the physical problem generated in the initial

stage of the analysis process can be related to the computational geometry

described by the mesh of nodes and elements. For each element, its material

properties must be defined together with the boundary conditions on the

faces of the elements, or at the nodes, which form the exterior of the mesh.

It is not necessarily straight forward task to define precisely the

material properties and, frequently, they must be approximated when

compiling the model data for an analysis.

For this model, the constant Isotropic material has been used and

their values are

1.Young’s modulus EX = 21000 kg/mm2

2. Density DENS = 8.002 e-10 mN/mm3

3. Poisson’s ratio NUXY = 0.3

2.12 Element type

Element type used in FEA may be described in terms of their

shape (through the relative positions of its nodes) and degrees of freedom

(possible directions of movements of each node). The element plot and

nodal plot for the model are shown in fig.

ELEMENT PICTORIAL VIEW TYPE

2-NODED BEAM ELEMENT

1 - D

3 – NODED BEAM ELEMENT

1 – D

3 – NODED TRIANGULAR ELEMENT

2 – D

6 – NODED TRIANGULAR

ELEMENT

2 – D

4 – NODED AREA ELEMENT

2 – D

8 - NODED AREA ELEMENT 2 – D

8 – NODED BRICK ELEMENT 3 – D

4 –NODED PYRAMID ELEMENT

3 – D

TABLE 2.1

ELEMENT TABLE

2.12.1 Element type for the model

For this analysis 4-noded area element (SHELL 63) is used.

SHELL 63 element is well suited for mapped meshing for this model.

Usually for any area of a model can be meshed using 4-noded area element

(SHELL 63) in a uniform manner (mapped meshing). It is a kind of mesh in

which the points of the mesh are arranged in a regular way all through the

continuum and can be stretched to fit a given geometry so that the results

will be more accurate when compared to free mesh results.

2.12.2 Choosing the element type

1.The range of elements and testing the elements:

It is not possible to present a set of universal guidelines to

develop any finite element model as such structural problem and element

type have their own particular features. It is not even possible to give rules

for what appears in packages to be identical element types since their

formulation can be different.

Any test for element behavior should be more complicated than

the situation of a simple rectangular geometry with a constant load, since

simple situations can give a false impression of the convergence

characteristics for realistic problems.

Quadratic elements, be they membrane or solid elements, give

the best compromise between accuracy and efficiency for general use.

When modeling a structural problem that can be classified, as

having bending deformation and the geometry is either flat or curved, then

the preferred choice of elements is always the general shell element.

Curved surfaces should not be modeled using flat elements as the

discontinuity at element boundaries introduces significant error.

2. Using a Hierarchy of elements:

Analysts should develop a model using a step by step approach.

This means that they should start with a simple approximation, say a beam

model, and make it more precise as the finite element modeling progresses.

Never tackle a real problem directly as this is likely to be time consuming

and wasteful of resources. Remember, that more results that are generated

the more effort that will be necessary to check that they are reliable and

relevant.

3. Restricting the dimensions of a problem:

Avoid the use of solid elements to model a problem where the

length in one of the spatial dimensions, for example the material thickness,

is much less than the lengths in the other two dimension.

4. Plate and shell elements:

Plate and shell elements have historically been the most difficult

to use in terms of achieving reliable and cost effective solutions. In

particular these elements in a static analysis do not give an acceptable

solution if the displacement of the nodes normal to the surface of the

material is greater than the thickness of the material.

5. The role of compatibility:

Elements must have the same order, all though one can mix three

sided and four sided elements.

There must be connection between the corner nodes of

neighboring elements and, if present, continuity between the edge nodes of

adjacent elements.

6. Elements of model contact:

Before developing a three dimensioning model for a problem

with contact between different parts, check that the package has three

dimensional contact algorithms.

2.13 Fan laws and Efficiency

2.13.1 Fan laws:

There are certain fan laws that are used to convert the

performance of a fan from one set of variables to another.

1.Conversation of fan performance

Suppose a fan of a certain size and speed has been tested and its

performance has been plotted for the standard air density. We then can

compute the performance of a fan of geometric similarity by converting the

performance data in accordance with these fan laws without running a test

on the other fan. It called as general fan laws.

2.Variation in fan speed

In order to convert the performance of a fan at one speed to

another speed, We take a number of points on the performance graph and

convert the corresponding data for air volume, static pressure, bhp,

efficiency and noise level fro the speed of the graph to the desired speed

using the following rules.

The air volume (cmf) varies directly with the speed

(cfm2/cfm1) = (rpm2/rpm1)

The pressure vary as the square of the speed

(Sp2/Sp1) = (rpm2/rpm1)

The brake horse power varies as the cube of the speed

(bhp2/bhp1) = (rpm2/rpm1)

The efficiency remains constant but, of course, shifts to the new air

volume values.

Variation in fan size

This law is used to convert the performance of one fan to

another fan when they are geometrically similar .

The fan laws for size , however , can be used only if the two

fans are in geometric proportion .

Both fans have the same number of blades .

Both fans have the same blade angles and any other angles on

the fan wheel and fan housing.

If the diameters of the two wheels are D1 and D2 for a size

ratio D2/D1, all other corresponding dimensions of wheel and housing have

the same ratio.

The air volume (cfm) varies has cube of the size. (cfm2/cfm1)

= (D2/D1)3

The pressure vary as the square of the size .

(sp2/sp1) = (D2/D1)2

The bhp varies as the fifth power of the size .

(bhp2/bhp1) = (D2/D1)5

Variation both fan size and fan speed

If both the fan size D and the fan speed (rpm) are varied , the

two sets of rules discussed above can be applied consecutively , in either

sequence .

(cfm2/cfm1) = (D2/D1)3 *(rpm2/rpm1)

(sp2/sp1) = (D2/D1)2 *(rpm2/rpm1)2

(bhp2/bhp1) = (D2/D1)5 *(rpm2/rpm1)3

(ME2/ME1)=1

Variation in Density

This fan law is used when the fan operates at high altitude

where the air density is less , where the fan handles hot or cold air (the air

density is inversely proportional to the absolute temperature) , or where the

fan handles a gas other than air , while the size and speed of the fan remains

constant.

The air volume remains constant

(cfm2/cfm1) = 1

The pressure vary directly as the density ρ

(sp2/sp1) = (ρ2/ρ 1)

The bhp varies directly as the density ρ

(bhp2/bhp1) = (ρ2/ρ 1)

The efficiency remains constant

2.13.2 Fan Efficiency

Fan work can be equated to the system resistance. Fan pressure has

the dimension of work per unit volume. Thus the system resistance may also

be regarded as the work required per unit volume of gas.

Power kw = Q X Ps

Work = force X Distance

Power = force X Velocity

= pressure X Area X Velocity

= pressure X Volume

The ratio of this air power to the power required to drive the fan is

the fan efficiency. The pressure may be total (including the velocity

pressure) or static and resulting efficiencies may also be “total” or “static”.

Selecting a fan of higher efficiency normally results in higher first

cost, but in lower operating cost.

1. Size and type limitations to good efficiency

High operational efficiencies are only achievable with certain types

and sizes of fan; indeed, the definition of “good” and “high” efficiency

depends on the class and quality of the fan being considered.

For a particular type of fan the best efficiencie swill be achieved by

higher specific speed fans of backward curved, backward inclined or aerofoil

bladed design with the fans being medium or large diameter and operating at

Reynolds numbers in excess of 20x105.

Where Reynolds number = (ρuD) / μ

D – Fan dameter

ρ – Fluid density

μ – Dynamic viscosity

2. Specification of fan requirements

Aerodynamic duty:

An important factor in ensuring a successful fan system is the correct

specification of the fan. The starting point in the specification of the fan is

knowledge of flow rate, which the fan (or fans) is required to handle. Often

there will be a range of flow rates over which the fan will be required to

operate and if this range is large it may be necessary to consider employing a

number of fans in parallel.

The next parameter to be defined is the pressure rise required of the

fan to move the gas through the system. This requires a knowledge of the

layout the system including pipe lengths, pipe diameters and elevations, a

knowledge of all the pipe work components in the system and information

on the properties of the gas being moved, particularly its density and

viscosity. From this it is possible to calculate the pressure losses in the

system.

If the fan duty varies with time then the use of variable geometry or

variable speed is almost certainly economically beneficial in terms of whole

life costs, if the duties of the fan are anticipated to increase with the time e.g.

: - if the output of the process system in which the fan operates is expected

to grow, then it may be possible to commence operation with a reduced

diameter (or reduced width) impeller fitted in to a standard casing and, as the

demand increases, fit the standard impeller. this will generally be preferable

to initially running the fan at a flow rate way below its design condition and

offer advantages of reduced power consumption and reduced bearing loads.

3. Estimation of fan type, size and speed:

Once the flow rate and pressure are known it is possible to derive

some idea of feasible options for the type of fan required. Simple formula

will allow initial estimates to be made of the probable type(s) and size(s),

which are optimum for a particular installation.

The specific speed, Ns, of a fan is a measure of the fan shape or type.

Ns is defined as Ns = [w (Q) 0.5] / (gH) 0.75

Where w is the rotation speed of the fan (rad / sec)

Q is the volume flow rate (m3 / sec)

gH is the specific energy ( J / Kg )

gH = (p / ρ)

Where p is the fan pressure raise – pa

ρ is the fluid density – Kg / m3

Knowing Q and gH, a range of rotational speeds can be assumed.

Typically these will correspond to 2, 4, or 6 pole motor speeds with a wide

choice available for belt driven fans. The value of Ns defines the optimum

fan type for the duty. If Ns is less than about 1.5 the fan will be a centrifugal

machine; if Ns is greater than about 2.5 the optimum fan will be an axial,

between 1.5 and 2.5 the optimum unit would be mixed flow type.

The next stage is to determine the approximate impeller diameter. For

a centrifugal type fan the dia can be estimated from the relationship, Farrant

V2 tip = gH / (0.8 - 0.23 Ns)

And for an axial or mixed flow machine from

D = 2[ gHQ / w3kL ]0.2

Where the loading coefficient, kL, has a value typically in the range 0.01 to

0.08.

It is thus possible to get an idea of the potential speed and dia of the

fan best suited to the duty. If either the speed or dia appears impractical this

may well point to the need to consider multistage fans or a series of fans in

parallel.

For multistage fans the head per stage reduces, thus raising the

specific speed per stage. For fans in parallel the flow per fan decreases thus

increasing the specific speed per unit.

2.14 STRESS ANALYSIS

Stress is defined as "a force tending to produce strain or

tension and to change the form or dimension of a solid", by dictionary. A

common man's understanding of stress is very much different from that of an

engineer. For an engineer

Limit F

Stress = A 0

A

Where, F = Force vector acting on the small area

M

Ever since the invention of Hooks

Law by the famous English Scientist Robert Hook (1935 -1703) analysis of

stress and strain has attracted many brilliant scientific and engineering

minds. Today, the theory is well developed and is widely used. However, a

general analytical calculation for the state of stress and strain in a general

solid is not yet available and is considered impossible to obtain. ' Stress

analysis problems can be solved using two sets of methods, i.e. experimental

and theoretical Hence, many numerical techniques have been developed and

are widely used in the industries for stress analysis. Finite Element Method

(FEM), one of the numerical techniques, was developed in fifties. Today

Finite Element Method is very popular and widely used in industry. Though

the underlying concept was originally introduced by Argyris in 1954-55, it

was supplemented by Turner, Clough, Martin and Toop in 1956. The

method is widely used since the development of high-speed electronic

digital computers and development of numerical methods to handle difficult

mathematical problems. Though the method was originally developed as a

tool for structural analysis, the theory and formulation have been

progressively refined and generalized and the method has been successfully

applied to many other fields like thermal, fluid, vibration, electrostatics,

Electro-magnetism, etc.

2.14.1 NEED FOR STRESS ANALYSIS

After the Industrial Revolution of nineteenth century,

large and complex machines and structures were built to mankind. As the

time passed, new types of machines and structures were built in critical and

demanding applications, requiring high reliability and economy. These

factors in design, under new environment of competition resulted in

application of analytical methods in the solution of engineering problems.

Design is no longer based upon empirical formulae. The importance of

analytical methods combined with laboratory experiments in the solution of

engineering problems has been recognized and accepted by the engineering

community. The conflicting requirements of increased reliability, reduced

cost and improved performance make the task of designer extremely

complicated.Reduced cost means reduced weight. Increased reliability with

reduced weight can be" achieved only on the basis of careful analysis of

stress distribution in the structure and experimental investigation of the

mechanical properties of the materials. Experimental techniques have

become very refined over the years. Similarly, analytical techniques have

become complex and advanced, leading to better understanding of stress

distribution in complicated solids.

CHAPTER 3

INTRODUCTION TO ANSYS

CHAPTER 3

3.1 Introduction to ANSYS

ANSYS is a general purpose finite element computer program for

the solution of structural, heat transfer engineering analysis. ANSYS

solution to capabilities includes: static analysis, elastic, plastic, thermal,

stress, stress stiffened, large deflections, bilinear elements, dynamic

analysis, model, harmonic response, linear time history, non-linear time

history, heat transfer analysis: conduction, convection, radiation, coupled to

fluid flow, coupled to electric flow, structures, magnetics, etc. Analysis can

be made in one, two, or three dimensions, including axisymmetric and

harmonic element options. ANSYS also contains a complete graphics

package and extensive pre and post processing capabilities.

3.2 ANSYS offers

1.tensive capabilities

2.ailability

3.owth and development

4.pport

3.3 Examples of ANSYS analysis

Examples Special options used

1.Laying ocean cable Dynamic, stress stiffening, large

deflection, hydrodynamic forces.

2.Automatic crash studies Dynamic, large deflection, plasticity,

gaps.

3.Evaluation of golf club swing Large rotations, stress stiffening.

4.Railroad tank car Dynamic, Fluid elements, pressure

vessel fatigue evaluation.

5.Piping system evaluation Static, seismic, gaps, large

deflection.

6.Electric furnaces smelting Heat transfer, thermal- electric

elements.

7.Electronic circuit boards & Heat transfer, radiation, static,

microchips thermal stresses.

8.Offshore power plant Multi-level sub structuring statics,

modal, over 1.5 million dofs.

9.Artificial hip prosthesis Statics, orthotropic materials.

10.Turbine Blade analysis Stress-stiffened, modal analysis.

3.4 Choosing the software

The first thing to consider is how knowledge of structural

mechanics might help you and your organization. To explore this, functional

area are related to structural mechanics must be considered. Structural

analysis may be used to determine the linear static stress and displacement in

structures such as vehicle body shell and the engine under operational loads.

Also, optimization may be required to produce body shells with a given

displacement for the minimum material thickness.

To find out which of the available packages may be used, a list of

requirements that the software should meet must be produced. More often

than not, no single package will meet all the requirements, but several

packages will meet some of the requirements.

3.5 Feature selection

The geometry of the structures that may need to be analyzed. This

will show whether a package is needed that can solve problems in two or

three dimensions.

3.5.1 Coupling requirements to other software

In some cases there may be a need to link structural results to heat

transfer simulations or even to fluid flow software. There may be a

requirement to send the results to a proprietary post-processor or to some

other display software, so that software must have interfaces.

3.5.2 The size of the simulation problem

Here something about the number of nodes and elements that a

typical mesh contains needs to be known, together with the number of

degree of freedom that is to be calculated. This information helps to

determine the storage requirements of the programs in terms of both primary

and secondary storage.

3.6 Results required for analysis

Stresses, strains and displacements, possibly as a function of

time.

3.7 Solution speed

Many things affect the time that it takes to produce the

solution. Clearly, this depending on the processing speed of the hardware

used, but it also depends on the structural solver itself.

3.8 Hardware availability

If there is a restriction on the make or type of computer or

graphics terminal that the software can be run on, this should be noted.

3.8.1 The following are some of the requirements that are related to the

software

Quality assurance (QA)

User friendliness

User support

Current users

3.9 Analysis procedure

3.9.1 Static analysis

A static analysis calculates the effects of steady loading conditions

on a structure, while ignoring inertia and damping effects such as those

caused by time varying loads. A static analysis can, however, include steady

inertia loads such as gravity and rotational velocity, and time – varying loads

that can be approximated as static equivalent loads (such as the static

equivalent wind and seismic loads commonly defined in many building

codes ).

3.9.2 Loads in a static analysis

Static analysis is used to determine the displacements,

stresses ,strains, and forces in structures or components caused by loads that

do not induce significant inertia and damping effects.

Steady loading and response conditions are assumed that is ,the

loads and the structure’s response are assumed to vary slowly with respect to

time .The kinds of loading that can be applied in a static analysis include:

Externally applied forces and pressures

Steady- state inertial forces (such as gravity and rotational

velocity)

Imposed (non-zero) displacements

Temperatures (for thermal strain)

Fluences (for nuclear swelling)

3.10 Procedure for a static analysis

3.10.1 The procedure for a static analysis consists of three main steps:

1.Build the model

2.Apply loads and obtain the solution

3.Review the results

The overall equilibrium equations for linear structural static analysis are:

[K] {u} = {F}

OR [k] {u} = {Fq} + {Fq}

N

Where: [K] = total stiffness matrix = ∑ [Ke] M=1

{u} = nodal displacement vector

N = number of elements

[Ke] = element stiffness matrix

{Fq} = total applied load vector

{Fr} = reaction load vector

1.Build the model:

To build the model, define the element types, element real

constants, material properties, and the model geometry.

2.Apply loads and obtain the solution:

In this step, the loads (boundary conditions) are defined and the

solution is obtained.

3.Review the results:

After the solution is completed, the post-processing step gives

the results of the static analysis.

Primary data available

Nodal displacements (UX, UY, UZ, ROTX, ROTY,

ROTZ)

Derived data available

Nodal and element stresses

Nodal and element strains

Element forces

Nodal reaction forces

Etc.,

3.11 Modal analysis

Modal analysis is used to determine the vibration characteristics

(natural frequencies and mode shapes) of a structure or a machine

component while it is being designed. it also can be a starting point for

another, more detailed, dynamic analysis, such as a transient dynamic

analysis, a harmonic response analysis, or a spectrum analysis.

3.11.1 Uses of modal analysis

Modal analysis is used to determine the natural frequencies and

mode shapes of a structure. The natural frequencies and mode shapes are

important parameters in the design of a structure for dynamic loading

conditions.

Modal analysis can be made on a pre-stressed structure, such as

spinning turbine blade. Another useful feature is modal cyclic symmetry,

which allows you to review the mode shapes of cyclically symmetric by

modeling just a sector of it.

3.12 Procedure for modal analysis

3.12.1 The procedure for a modal analysis consists of four main steps:

1. Build the model

2. Apply the loads and obtain the solutions

3. Expand the modes

4. Review the results

3.12.2 Assumptions and restrictions

Valid for structural and fluid degrees of freedom(DOFs)

The structure has constant stiffness and mass effects.

There is no damping.

The structure has no time varying forces, displacements, pressures, or

temperature applied (free vibration).

The equation of motion for an undamped system, expressed in matrix

notation using the above assumptions is:

[M] {u} + [K] {u} = {0}

3.13 Design optimization

Design optimization is a technique that seeks to determine an

optimum design. By “optimum design” all the specified requirements are

met with a minimum expense of certain factors such as weight, surface area,

volume, stress, cost, etc. in other words, the optimum design is usually one

that is as effective as possible.

Any aspect of the design can be optimized: dimensions (such as

thickness), shape (such as fillet radii), placement of supports, and cost of

fabrication, natural frequency, material property and so on.

An optimum design can be defied as the best possible design

satisfying a specific objective and a set of constraints imposed by the

specifications or by the design problem itself.

3.14 Typical examples of optimized designs are:

Design of aircraft, aerospace and automotive structures for

minimum weight.

Design of machines, components, frames, and mechanisms,

handling devices etc., for minimum cost.

Design of pumps, turbines, compressors, engines, and etc.,

for maximum efficiency.

CHAPTER 4

METHODOLOGY

CHAPTER 4

METHODOLOGY

4.1. Solid model generation using preprocessor:

In ANSYS there are three stages. They are pre-

processor, solution and post processor. So before doing analysis, the

geometry of the model should be created. Modeling is done in ANSYS

through pre-processor. There we have lot of option through which the

geometry of the model is created.

4.2. Meshing contours:

After generation of the solid model using pre-

processor, the model should be meshed properly. That is the model should

be disecritised into number of small elements. For meshing of the model, we

should generate meshing contours. That is the lines of the geometry should

be properly divided. So that we can easily mesh the model otherwise without

the contours the mesh won’t be proper and we can’t solve it.

4.3. Meshing of areas:

After generation of contours for proper meshing we

should go for meshing of the model. There are two types of meshing. They

are free mesh and mapped mesh. So in free mesh we can solve it and get the

results but it won’t be accurate.

So for accurate results, we have to go for mapped mesh.

So if we did line element sizing (lesize) properly we get mapped mesh.

4.4. Defining material:

So after completing modeling and meshing we have to

define the material of the model. So there are different properties which will

define a material that is density, young’s modulus, Poisson’s ratio. For

different materials this values are different. By using these properties the

material can be define in pre-processor.

4.5. Choosing appropriate element for analysis:

The basic concept of FEA is to discritise the model into

finite number of smaller elements. There are different element types

available in ANSYS pre-processor. Based on the model the element types

vary. We have to choose a appropriate element for analysis.

4.6. Attributing equivalent and actual boundary:

After discritising and defining material of a model, we

have to apply the boundary conditions and the loads wherever we required

for the analysis. First, constraints should be applied. So wherever required,

the degrees of freedom should be arrested. After applying constraints, the

loads are applied on nodes or element for the analysis.

4.7. Solving the problem using solver:

Solution is the second stage in ANSYS where the solution

of the given problem is done. So here we won’t do anything the solution

module generate the element matrices and find the stress and deflections

according to the parameters we applied.

4.8. Viewing results:

The results are viewed in post-processor. Where the stress

and deflection can be plotted on the screen. So different colours are plotted

for different stress value. We can view both the maximum and minimum

stress.

4.9. Studying the parameters stress and deflection for existing design:

The above steps are done for the original design and stress

value and deflection for original design can be studied.

4.10. Modifying the geometry model by reducing the thickness:

After studying the stress and deflection for original design.

The stress and deflection for modified design should be studied. The design

is modified by reducing the thickness. Then all the above steps followed for

finding the stress and deflection value.

Finally the results of the original and modified design

should be should be compared in order to obtain a optimized design.

CHAPTER 5

MODELING

CHAPTER 5

MODELING

5.1 SEGMENT GENERATION:

First the segment of the radial fan impeller is created

using ANSYS preprocessor .The created segment is shown below in fig 1.

Area plot for the created segment using ANSYS preprocessor is shown

in the fig 2

COMPONENT ORIGINAL REALCONSTANT VALUES (mm)

MODIFIED REALCONSTANTVALUES (mm)

BACKPLATE(BOTTOM)

25 18

BACKPLATE(TOP)

15 10

BLADE 5 3.15

COVERPLATE 12 8

RING 30 15

FLANGE 80 45

5.2 REAL CONSTANTS FOR THE MODEL:

The table 5.1 shows the real constants for various

components of the original and optimized fan impeller model.

Thicknesses of the various component of impeller are ploted in different

colours which is shown in fig 3.

A1- Back Plate (Bottom)

A2- Back Plate (Top)

A3- Bade

A4-Cover Plate

A5- Ring

A6- Flange

5.3 MAPPED MESH:

Mapped mesh is generated for the model which is shown in fig 4.

Element type used for meshing the model is SHELL

ELEMENT (ET, 1, 63).

5.4 GENERATION OF FAN IMPELLER:

Fan Segment created is copied to 360 degrees along y-

axis with 12 segments (including original) so the fan impeller model is

generated which is shown in fig 6

5.5 BOUNDARY CONDITION:

DISPLACEMENT:

The rotating motor shaft is fixed in the impeller

therefore displacement on the impeller hole is zero in all degree of freedom.

ANGULAR VELOCITY AND ANGULAR

ACCELERATION:

The impeller rotates about the z-axis at a speed of 1000

rpm.Therefore angular velocity=2∏n/60=104.7 rad/sec(105)

Angular acceleration also 9810 rad/sec2.

5.6 SOLUTION:

After completing modeling and giving boundary

conditions the problem has to be solved using the ANSYS-SOLUTION

utility.In the solution utility all the element matrices are formed and it is

solved to find the stress and deflection for the applied boundary condition.

CHAPTER 6

RESULTS AND DISCUSION

CHAPTER 6

RESULTS AND DISCUSSIONS

6.1 STATIC ANALYSIS:

The stress distribution and the deflection of the

impeller are found. The stress distribution and the deflection plot for the

various components of the impeller are plotted in the figures.

6.2 OPTIMIZATION:

For the original fan impeller the stress value is

4.274kgf/mm2 and deflection is .0678mm.The blade in the impeller has the

maximum stress of 4.274kgf/mm2.

COMPONENTS

ORIGINAL FAN OPTIMIZED FAN

STRESS

Kgf/mm2

DEFLECTION

mm

STRESS

Kgf/mm2

DEFLECTION

mm

FAN IMPELLER 4.274 0.0678 5.822 0.09945

BACKPLATE(BOTTOM)

1.284 0.0126 1.291 0.016396

BACKPLATE(TOP)

2.318 0.03426 2.593 0.041143

BLADE 4.274 0.0678 5.822 0.099455

COVER PLATE 2.852 0.0526 3.091 0.064076

RING 2.927 .04845 3.159 0.057363

FLANGE .58453 .00198 .681387 .002394

TABLE .6.1

6.3 RESULTS FOR FAN IMPELLER WITH ORIGINAL

THICKNESS:

6.4 RESULTS FOR OPTIMIZED FAN IMPELLER:

CHAPTER 7

CONCLUSION

CHAPTER 7

CONCLUSION

In this work, an attempt has been made to increase the Fan

efficiency by optimizing the thickness of the various components in the fan

impeller, and analyzing the stress distributions in them. Optimization of the

thickness of the parts of impeller leads to decrease in weight of the Fan

Impeller, and in turn the power required for driving the fan decreases. Pre -

stress conditions are applied to this model, therefore the strengthening and

weakening of the impeller is predicted.

COMPONENT ORIGINAL IMPELLER THICKNESS (mm)

OPTIMIZED IMPELLER THICKNESS (mm)

Back plate(Bottom)

25 18

Back plate(Top)

15 10

Blade 5 3.15

Cover plate 12 8

Ring 30 15

Flange 80 45

7.1. DIMENSIONS FOR THE MODEL:

TABLE.7.1

7.2 WEIGHT REDUCTION:

After optimizing the thickness of the Fan impeller the weight

of the Fan Impeller is reduced.

Existing weight of the Fan Impeller=10 tones

=10,000 kg

After optimizing

% of weight reduced in the =18.5%

Fan Impeller

Amount of weight saved = (18.5/100)*10,000

= 1850 kg.

= 1.85 tones

Cost of steel/kg = Rs.300

In the BHEL-RANIPET there are about 12 Fans made per

year.

Total cumulative weight saved = 12*1850

= 22200 kg

= 22.2 tones

Total cumulative cost saved per year = 22200*300

= Rs. 66,60,000

REFERENCES

REFERENCES

1. John F.Abel, Chandra Kant S Desai (1987), ‘Introduction to the Finite

Element Method’- CBS Publishers and distributors, New Delhi.

2. Kalyanmoy Deb (2000), ‘Optimization for engineering Design’-

Prentice hall of India (p) Ltd, New Delhi.

3. Krishnamoorthy C.S. (1994), ‘Finite Element Analysis’ -Tata

McGraw-Hill publishing company, New Delhi

4. Robert D.Cook, David S. Malkus, Michael E.Plesha (1989), ‘concepts

and applications of finite element analysis’- John Wiley& Sons,

Singapore.

5. Thirupathi R.Chandrapatla, Ashok D.Belegundu (1997), ‘Introduction

to finite elements in Engineering’- Prentice hall of India (p) Ltd,

New Delhi.