Ercam Book

Lecture Notes on Multidisciplinary Industrial Engineering
Series Editor: J. Paulo Davim
M. Vinyas
Amelia Loja
Krishna R. Reddy Editors
Advances in
Structures,
Systems and
Materials
Select Proceedings of ERCAM 2019
Lecture Notes on Multidisciplinary Industrial
Engineering
Series Editor
J. Paulo Davim , Department of Mechanical Engineering, University of Aveiro,
Aveiro, Portugal
“Lecture Notes on Multidisciplinary Industrial Engineering” publishes special
volumes of conferences, workshops and symposia in interdisciplinary topics of
interest. Disciplines such as materials science, nanosciences, sustainability science,
management sciences, computational sciences, mechanical engineering, industrial
engineering, manufacturing, mechatronics, electrical engineering, environmental
and civil engineering, chemical engineering, systems engineering and biomedical
engineering are covered. Selected and peer-reviewed papers from events in these
fields can be considered for publication in this series.
More information about this series at http://www.springer.com/series/15734

M. Vinyas Amelia Loja
• •
Krishna R. Reddy
Editors
Advances in Structures,
Systems and Materials
Select Proceedings of ERCAM 2019
123
Editors
M. Vinyas Amelia Loja
Department of Mechanical Engineering Department of Mechanical Engineering
Nitte Meenakshi Institute of Technology Instituto Superior de Engenharia de Lisboa
Bengaluru, Karnataka, India Lisbon, Portugal
Krishna R. Reddy
Department of Civil and Environmental
Engineering
University of Illinois at Chicago
Chicago, USA
ISSN 2522-5022 ISSN 2522-5030 (electronic)

Lecture Notes on Multidisciplinary Industrial Engineering
ISBN 978-981-15-3253-5 ISBN 978-981-15-3254-2 (eBook)
https://doi.org/10.1007/978-981-15-3254-2
© Springer Nature Singapore Pte Ltd. 2020
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.
This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface
The ERCAM proceedings includes the abstracts from selected manuscripts from
peer-reviewed papers under various themes, namely Materials, Mechanics,
Structures and Systems of the Second International Conference on “Emerging
Research in Civil, Aeronautical and Mechanical Engineering” (ERCAM 2019),
organized by Nitte Meenakshi Institute of Technology (NMIT), Bangalore, India,
from 25 to 26 July 2019.
NMIT, established in 2001 under well-known NITTE Education Trust, is one
of the renowned technical institutions of the country and known as the premier
institution in Karnataka in particular. This institution is affiliated to the
Visvesvaraya Technological University (VTU) and approved by UGC/AICTE/
Government of Karnataka. Recognizing its performance in both teaching and
research, University Grants Commission has granted autonomous status from the
year 2007, accredited at the “A Grade” by the National Assessment and
Accreditation Council (NAAC), and four UG programmes got accredited by the
National Board of Accreditation under Tier 1 for three years from 2014 and again in
2017. It is the only unaided private engineering college in Karnataka state that was
selected by the Government of India for World Bank funding under TEQIP Phase II
Subcomponent 1.1.
Teaching, learning, research and development—the four major pillars—are
sustaining the edifice of higher education. Multidisciplinary research has become an
indispensable imperative in the development of a wide range of engineering
materials, processes and systems.
Higher education is now in the process of a paradigm shift, with research gaining
increasing attention from policy planners and educationists. Contribution to the
research field is prioritized right from the undergraduate level in the universities,
making teachers enthusiastic and involving student community. NMIT is conscious
and cognizant of this phenomenon and is organizing ERCAM 2019 to create a
platform of interaction to the researchers across the world to share/exchange their
knowledge. ERCAM 2019 is therefore intended to bring together specialists and
practitioners from industries, R&D institutes and academicians, providing a forum
of exchange of knowledge on researches in recent trends in major sectors of
v
vi Preface
materials, processes and systems pertaining to various fields under civil, aeronau-
tical and mechanical engineering.
The conference succeeded in providing a solid platform for researchers around
the globe to share insights, experiences and various aspects of evolving technolo-
gies related to civil, aeronautical and mechanical engineering. Proceedings covers a
wide range of topics which are relevant to the current needs of the industry. All the
papers received for ERCAM 2019 conference on various aspects of focused areas
were rigorously triple-reviewed by a pool of experts in respective fields from all
over the world under Technical Review Committee, first-level review being pla-
giarism check limiting 20% including self-plagiarism. The committee selected the
best 28 papers for oral and poster presentation in the conference to publish in
association with Springer and AIP accordingly. The management of NMIT and
Team-ERCAM 2019 express their sincere gratitude to the reviewers for their
contribution of rigorous review and also authors for revising their papers to
incorporate the comments and suggestions by the reviewers.
We are immensely grateful to all the organizers and members of the Program
Committee for their diligent work, which ensured high-quality publications. We
highly appreciate the guidance and continuous support provided by the manage-
ment of NMIT and Advisory Committee being with us spending their valuable time
for creating ERCAM 2019 a brand image. We recognize the involvement of faculty
members of all three departments throughout.
We thank all our sponsors for their support, especially AICTE, Government of
India; VTU, Belagavi; and The Supreme Industries Ltd. We are grateful to our
fellow professionals from academia and industry who have volunteered to review
papers and accepted our invitation to deliver plenary talk despite being of their
preoccupied schedules. We would like to thank Springer Nature, AIP publications,
JVC Printers and USB Printers for their fruitful collaboration during the preparation
of the proceedings. The proceedings of ERCAM 2019 is organized in a single
volume. We hope that the ERCAM 2019 proceedings will serve as an important
intellectual resource for civil, aeronautical and mechanical engineering researchers,
pushing forward the boundaries of these fields and enabling better collaboration and
exchange of ideas. Our special thanks to all authors for their contributions, and we
anticipate extended contributions in coming years of the flagship event of NMIT—
ERCAM—an international conference.
Bengaluru, India M. Vinyas

Lisbon, Portugal Amelia Loja
Chicago, USA Krishna R. Reddy
Contents
1 Cerium Oxide and Silicon Carbide Reinforced Al6063 Metal

Matrix Composites Comparative Evaluation for Mechanical
Properties and Fractography Studies . . . . . . . . . . . . . . . . . . . . . . . 1
Abdul Nazeer and Mir Safiulla
2 Feasibility Study of Detonation Wave Initiation by Multiple
Structured Blockages in a Plain Tube . . . . . . . . . . . . . . . . . . . . . . . 11
S. Srikrishnan, V. Jayakumar and P. K. Dash
3 Spectral Density at Roof of Setback Tall Building Due to Time
Variant Wind Load . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
Amlan Kumar Bairagi and Sujit Kumar Dalui
4 Distribution of Wind Pressure Around Different Shape
Tall Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
5 Material Characterization of Pin-Core Latex Polymer Foam
Under Static and Dynamic Loads . . . . . . . . . . . . . . . . . . . . . . . . . . 39
K. Venkataramana, R. K. Singh, Anindya Deb, Vivek Bhasin,
K. K. Vaze and H. S. Kushwaha
6 Effect of Delamination on Static Behaviour of Point Supported
Composite Conoidal Shells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Kamalika Das and Suman Pandey
7 Wind Excitation Study of a Corner-Modified Square Tall
Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
Arghyadip Das and Sujit Kumar Dalui
8 Experimental Investigations on SCC with Indigenously
Developed Geopolymer Aggregates . . . . . . . . . . . . . . . . . . . . . . . . . 75
M. P. Naveena, G. Narayana, Vijayalakshmi Akella and C. M. Thejas
vii
viii Contents
9 Experimental Studies and Numerical Validation on Bearing

Capacity of Skirted Footings on c-U Soils . . . . . . . . . . . . . . . . . . . 85
Arekal Vijay, Vijayalakshmi Akella and B. K. Raghu Prasad
10 Effect of Aerodynamic Modifications on a Tall Building
with Horizontal Irregularity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
Shanku Mandal, Sujit Kumar Dalui and Soumya Bhattacharjya
11 Flexural Behaviour of 2D Cellular Lattice Structures
Manufactured by Fused Deposition Modelling . . . . . . . . . . . . . . . . 109
Harshini Gullapalli, Syed H. Masood, Syed Riza
and Panneer Ponnusamy
12 Power Spectral Density on Principal Building Due to Setback
Interfering Building . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
Soumya Mukherjee and Amlan Kumar Bairagi
13 Effect of Solar Optical Properties of Building Envelope on Time
Lag, Decrement Factor and Energy Saving of Buildings . . . . . . . . . 127
Debasish Mahapatra and T. P. Ashok Babu
14 Study and Estimation of Static Stiffness of Machine
Tool Spindle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
Vibish S. Bhardwaj, Ramesh H. Aralaguppi, Ashok N. Badhe,
Bhargav and Arun R. Rao
15 Effect of Surface Corrosion of Steel Reinforcement on Bond
Strength Characteristics of Concrete . . . . . . . . . . . . . . . . . . . . . . . . 153
Vineet Kothari and Hemanth Kamplimath
16 Stress Distribution in an Infinite Plate with Circular Hole
by Modified Body Force Method . . . . . . . . . . . . . . . . . . . . . . . . . . 163
Shrikrishna Badiger and D. S. Ramakrishna
17 Wind Load Analysis of a Tall Structure with Sharp
and Corner Cut Edges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
Abhipsita Rej and Amlan Kumar Bairagi
18 Comprehensive Characterization of Carbon Fiber-Reinforced
Epoxy Composite for Aerospace Application . . . . . . . . . . . . . . . . . 185
D. Bino Prince Raja, B. Niharika,
R. S. Manoj Kumar and C. G. Tejaswini
19 Design and Simulation of Single-Axis MEMS Accelerometer
for Low Acceleration Applications . . . . . . . . . . . . . . . . . . . . . . . . . 199
K. J. Rudresh, Kin Gopalakrishna, K. Bharath Gowda,
R. Harshith Gangatkar and Hemanth Kumar
Contents ix
20 Optimization of Neem Biodiesel Blend and Injection Pressure

of CI Engine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
C. Srinidhi, Sunil Panshetty, Madhusudhan and S. V. Channapattana
21 Design and Manufacturing of Hand Injection Mold for Flip
Badge—Analysis of Alternate Air Venting Patterns . . . . . . . . . . . . 227
Vijaykumar Kalwa and Balaji Thottempudi
22 Design and Optimization of Foot Locus Trajectory
of Theo Jansen Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
N. Prashanth, S. Karthik, G. R. Rahul and T. B. Sandarsh
23 Performance Evaluation of Concentric and Eccentric
Buckling Restrained Braces on the Dynamic Behaviour
of RC Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
H. J. Prajwal Shankar, Sunil Lamsal, Praveen Shrestha,
Bharathi Ganesh and R. Prabhakara
24 Analysis of Rotation Capacity of RC Beams Over Formation
of Plastic Hinges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
R. Prabhakara, B. Nambiyanna, R. Nakul and Bharathi Ganesh
25 Influence of Metallurgical Parameters on Wear and Impact
Characteristics in High Chromium Manganese Irons . . . . . . . . . . . 267
G. J. Naveen, P. Sampathkumaran, S. Seetharamu, T. Pramod,
Ashirwad Pandey and A. Sathyanarayanaswamy
26 Simulation of High-Temperature Solid Particle Erosion
of HVOF-Sprayed NiCrBSiFe and WC-Co/NiCrBSiFe-Coated
Wall Using CFD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
K. Raghavendra Naik, R. K. Kumar, V. Saravanan, S. Seetharamu
and P. Sampathkumaran
27 Optimization of Friction Stir Welding Parameters Using Taguchi
Method for Aerospace Applications . . . . . . . . . . . . . . . . . . . . . . . . 293
K. Ramesha, P. D. Sudersanan, N. Santhosh, G. Ravichandran
and N. Manjunath
28 Assessment of the Shear Strength of Fly Ash-Based
Geopolymer Concrete . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
S. Kumar, S. Rajendra and K. S. Sreekeshava
About the Editors
Dr. M. Vinyas is an Assistant Professor in the Department of Mechanical

Engineering, Nitte Meenakshi Institute of Technology (NMIT), Bangalore, India.
He received his Ph.D in Mechanical Engineering from the National Institute of
Technology Karnataka, Surathkal in 2018. He received the university medal in his
Masters of Technology at Visvesvaraya Technological University. He is also a
visiting research faculty in foreign universities and founder of Composite Research:
Experimentation, Analysis and Design (CREAM) lab, NMIT, Bangalore. He has
published more than 70 articles in reputed international journals and is also serving
as a reviewer for many refereed journals. Recently, considering his research
achievements, Dr. Vinyas received the CV Raman post-doctoral award from the
Indian Institute of Science, Bangalore. His research areas include finite element
analysis, composite structures, coupled structural analysis, and vibration control.
Dr. Amelia Loja is presently Adjunct Professor at the Mechanical Engineering

Department of the Engineering Institute of Lisbon (ISEL/IPL), and Senior
Researcher of the Mechanical Engineering Institute (IDMEC/IST). Her academic
background integrates a BSc with honours in Marine Engineering from the
Portuguese Nautical School and a BSc in Computer Science. Her MSc degree in
Mechanical Engineering was conferred by the Technical University of Lisbon, and
later she obtained her PhD in Mechanical Engineering also from this University.
Her major areas of interest include the scientific areas of Computational Solids
Mechanics, Optimization and Reverse Engineering. Until now she published 41
papers in scientific international journals. Amelia Loja is also Chairperson of the
ECCOMAS thematic conference SYMCOMP (International Conference on
Numerical and Symbolic Computation: Developments and Applications) and since
2017, she has been invited by the European Commission to evaluate project pro-
posals in different subjects related to her competences.
Dr. Krishna R. Reddy is a Professor of Civil and Environmental Engineering, the

Director of Sustainable Engineering Research Laboratory, and also the Director
of the Geotechnical and Geoenvironmental Engineering Laboratory in the
xi
xii About the Editors
Department of Civil and Materials Engineering at the University of Illinois,

Chicago, USA. Dr. Reddy received his Ph.D. in Civil Engineering from the Illinois
Institute of Technology, Chicago, USA. He received gold medals for being first in
his class of B.E. (Civil Engineering) at Osmania University and M.E. (Civil
Engineering) at the Indian Institute of Technology, Roorkee.
Dr. Reddy has over 25 years of teaching, consulting and research experience
within the fields of geotechnical engineering, environmental engineering, and
sustainable engineering. Specifically, his expertise includes: (1) remediation of
polluted soils, sediments, groundwater and stormwater, (2) waste management and
landfill engineering; (3) life cycle assessment and sustainable engineering; and
(4) geotechnical engineering (foundations, earth structures, etc.). His research is
funded by the U.S. National Science Foundation, the United States Environmental
Protection Agency, several prominent state and local government agencies, and
industries. He has published 4 books, 206 journal papers, and 182 full conference
papers (with h-index of 52). Dr. Reddy has given 185 invited presentations in the
USA and 18 other countries (Brazil, Canada, China, Colombia, France, Germany,
Greece, Hong Kong, India, Italy, Japan, Macedonia, Spain, Sri Lanka, South Korea,
Turkey, Thailand, and U.K.). Dr. Reddy is an Associate Editor or Editorial Board
Member of over 10 different journals. He has been an active member of various
professional societies, including the ASCE.
Dr. Reddy has received several awards for excellence in research and teaching,
including the ASTM Hogentogler Award, the University Distinguished Researcher
Award, the University of Illinois Scholar Award, and the University of Illinois
Award for Excellence in Teaching. He is a Fellow of the American Society of Civil
Engineers and a Diplomate of Geotechnical Engineering.
Chapter 1
Cerium Oxide and Silicon Carbide
Reinforced Al6063 Metal Matrix
Composites Comparative Evaluation
for Mechanical Properties
and Fractography Studies
Abdul Nazeer and Mir Safiulla
Abstract Aluminum matrix material is considered as advanced material because

of its very good mechanical, tribological properties which have been used in auto-
mobile, aerospace industries at the larger end, and now aluminum matrix materials
have been used in medical and electronics industries because of its good and high
strength to weight ratio and low coefficient of thermal expansion. The present work
is focused about the improvement in mechanical properties and fracture graphic
studies of aluminum matrix reinforced with ceramic particulate silicon carbide and
cerium oxide of laboratory grade size, respectively. Here, the reinforcement is varied
in terms of weight percent ranging from 0 to 8% in the steps of 2%, and the com-
posites were fabricated using a stir casting route. Prepared specimens were tested as
per ASTM standard and performed tensile, compressive, impact and hardness test
and fracture surfaces studies done under the scanning electron microscope (SEM)
and X-ray diffraction (XRD). The results reveal that with increase in percentage of
reinforcement, the mechanical properties of the composite system have increased,
SEM images clearly reveal the modes of fracture, and XRD shows the presence of
reinforcement.
Keywords Aluminum matrix composite · Stir casting · Mechanical testing ·

SEM · XRD
1.1 Introduction
The term “composites material” deals with mixing two different materials, where one
is called as matrix and other as reinforcement. Composite materials are becoming
increasingly popular because of its good mechanical, tribological properties and
good resistance to wear, good fatigue life, prevention to corrosion and ability to the
A. Nazeer (B) · M. Safiulla

Mechanical Engineering, Ghousia College of Engineering, Ramnagaram, Karnataka, India
e-mail: [email protected]
© Springer Nature Singapore Pte Ltd. 2020 1

M. Vinyas et al. (eds.), Advances in Structures, Systems and Materials,
Lecture Notes on Multidisciplinary Industrial Engineering,
https://doi.org/10.1007/978-981-15-3254-2_1
2 A. Nazeer and M. Safiulla
coefficient of thermal expansion. Composite material is broadly classified on the

basis of matrix and reinforcement. On the matrix, it is classified as polymer matrix
composite (PMC), Metal matrix composite (MMC), and ceramic matrix composite
(CMC), on reinforcement is classified as continuous fibers, discontinuous fibers,
Whiskers and particulates. The function of reinforcement is to carry the load, and the
matrix is used to transfer and distribute the applied load. These composite materials
have been used in many industries mainly in automotive and aerospace industries
due lightweight, high load carrying capacity and good resistance to wear, and metal
matrix composite is gaining more importance because of easy processing and cost
effectiveness. Other advantage of composite metal matrix is that it can be fabricated as
hybrid composite in order to increase the properties for specific requirement, where
in a single matrix two different reinforcements are introduced such as Al6061 as
matrix and SiC and Al2 O3 as reinforcement [1]. In order to increase tensile strength,
hardness and toughness, hybrid composites are prepared by reinforcing fly ash and
SiC [2]. Researchers have done work on aluminum alloy with alumina and boron
carbide and found that tensile strength, flexural test, impact test and Brinell hardness
test values have marginally increased.
Lightweight aluminum alloy matrix composites (AMCs) become increasingly
popular due to great mechanical and tribological properties as compared to conven-
tional aluminum alloys. Cast particle-reinforced composites show good results in
the automobile sector due to inherent simplicity in fabrication combined with an
advantage of mass production [3]. Silicon carbide (SiC)-reinforced aluminum matri-
ces have been widely accepted as potential candidates for weight-critical automobile
and aerospace applications.
1.2 Experimental Details
1.2.1 Matrix and Reinforcement Material
Aluminum 6063 is used as metal matrix, silicon carbide and cerium oxide as
reinforcement in laboratory grade.
Silicon carbide of laboratory grade 10–20 µm size is chosen as reinforcement
owing its high hardness, low coefficient of thermal expansion and highly wear
resistant.
Cerium oxide of 5 µm size was considered as a reinforcement. It is an important
commercial product and an intermediate in the purification of the element from the
ores.
1 Cerium Oxide and Silicon Carbide Reinforced Al6063 … 3
Fig. 1.1 Preparation of composite system with various reinforcements and testing for mechanical
properties. a Preheating of mold, b pouring of molten material into mold, c tensile test specimen,
d compression test specimen, e hardness test specimen, f Charpy impact specimen
1.2.2 Preparation of Composites
Aluminum 6063 alloy was melted utilizing a 6 kW electric heater. Hereliquid met-
allurgy route stir casting technique adopted where mechanical stirrer is turned to
make a vortex, reinforcement SiC and CeO2 powders is at first preheated and grad-
ually mix in liquid metal with consistently stirred. The stirring time was 10 min.
The molten composites maintained at a constant temperature of 740 °C. The rein-
forcement induced 0–8% in steps of 2%, and Fig. 1.1a, b shows the preparation of
composite.
1.2.3 Scanning Electron Microscopy (SEM)
A scanning electron microscope (SEM) is a type of electron microscope that pro-

duces images of a sample by scanning the surface with a focused beam of electrons.
The analysis was conducted on Hitachi SU 3500 make at Centre for Incubation Inno-
vation Research and Consultancy (CIIRC) Jyothy Institute of Technology, Bangalore
(INDIA). The fractured surfaces of tensile test specimen were studied under SEM.
1.2.4 X-Ray Diffraction
When the geometry of the incident X-rays impinging the sample satisfies the Bragg
Equation, constructive interference occurs and a peak in intensity occurs. The powder
Fig. 1.2 Variation of micro hardness with increased content of SiC and CeO2 for cast matrix alloy
and its composite systems
was collected from the samples of Al6063–SiC and CeO2 composites for all weight
percentage of (0, 2, 4, 6, 8), and diffraction patterns were recorded to confirm the
presence of SiC and CeO2 in the composite. The sample is in powder form so that it
can be easily placed in sample holder of dimension 24.6 mm × 1.0 mm.
1.3 Result and Discussion
1.3.1 Micro Hardness Test
Micro hardness tests were performed by applying load of 10 N for a period of 20 s. It

can be observed from Fig. 1.2 that the hardness value of composite goes on increases
with increase in reinforcement for all composites with different reinforcement which
indicate that hardness of the base alloy can be increased with increase in reinforce-
ment which clearly shows the effect of hard reinforcement. It also found that hardness
value is more for reinforcement SiC as compared to CeO2 . Addition of 8 wt% of SiC
increases hardness of matrix alloy 31.6%, while addition of CeO2 is 22.25%.
1.3.2 Compression Strength
The results reveal that as percentage of reinforcement increased, the compressive

strength also increases. This is because SiC and CeO2 are strong and hard particulate
added to aluminum alloy. The strength of composite systems increases which requires
high energy to compress. The variation compressive strength for composite systems
is very much similar to other researchers (Fig. 1.3).
Fig. 1.3 Variation of compressive strength with increased content of SiC and CeO2 for cast matrix
alloy and its composite systems [2, 4]
Fig. 1.4 Variation of tensile strength with increased content of reinforcement for cast matrix alloy
1.3.3 Tensile Test
Figure 1.4 summarizes the tensile strength of cast alloy and its composite systems
and tries to identify which reinforcement will give the maximum ultimate tensile
strength, and from Fig. 1.4 it clearly observed that the maximum tensile strength is
for silicon carbide reinforcement, as silicon carbide is harder than cerium oxide. The
maximum ultimate tensile strength at 8 wt% composites is found for Al6063–SiC in
comparison with Al6063–CeO2. It was found 51.9% of increase in tensile strength
with reinforcing 8% of SiC and 43.6% increase in tensile strength with reinforcing
8% of CeO2 as compared with base metal matrix alloy.
1.3.4 Charpy Impact Test
It is observed from Fig. 1.5 that as the percentage of reinforcement is increased, the
ductility of the material is decreased and brittleness is increased, so due to the impact,
the energy absorbed by the composite is decreased with increase in percentage of
reinforcement, and similar results are obtained by other researchers [4, 5].
Fig. 1.5 Variation of impact strength with increased content of SiC and CeO2 for cast matrix alloy
1.3.5 Fractography Studies
The study of fracture surfaces for different weight percentages of SiC and CeO2
reinforcement under scanning electron microscope (SEM) technique is shown in
Fig. 1.6 as the SEM photographs of the tensile test fractured surfaces of the cast
Al6063 and its composite systems. It is evident that the base matrix alloy has got
larger dimples when compared with different composite system studied for given
content of reinforcement. Al6063 matrix alloy showing very large dimples indicates
0%
Crack &
Delamination
2% SiC 2%Ceo2 4% SiC 4% Ceo2
6% SiC 6% Ceo2 8% SiC 8% Ceo2
Fig. 1.6 SEM of tensile fractured surface of Al6063 0–8 wt% composites system
ductile fracture as shown in Fig. 1.6 for 0% reinforcement, whereas in case of Al6063-
8 wt% composite, particles of reinforcement and medium-sized dimples are visible
as evident from Fig. 1.6, and these observations are similar to other researchers [2].
1.3.6 X-Ray Diffraction Analysis
XRD graphs and results given below clearly indicate the presence of reinforcement
in the matrix alloy with varying percentage, XRD result shows a variation in the
amount of reinforcement embedded in matrix alloy, and this is because XRD samples
in powder form in minute quantity from a particular region are taken.
In X-ray diffraction pattern (Fig. 1.7a–i), many peaks are obtained in the 2 span
ranging from 5 to 90, but the common peaks at 2 of 38.44°, 44.7°, 65.32° and 77.2°
belong to pure Al and the peaks at 2 of 28.5, 38.83°, 50.80°, 78.88° and 82.43° belong
to CeO2 and for SiC 35.1960, 38.370 and 38.31,83.0. With the help of Match software,
graphs are compared with JCPDS card JCPDS file #04-0787 for SiC (Table 1.1).
Fig. 1.7 XRD pattern for Al6063 matrix alloy and its composites
8
Table 1.1 Analysis of XRD data using Match software and identification of the presence of reinforcement with following percentage
Figure 1.7 a b c d e f g h i
Name Al6063 Al6063-2 wt% Al6063-4 wt% Al6063-6 wt% Al6063-8 wt% Al6063-2 wt% Al6063-4 wt% Al6063-6 wt% Al6063-8 wt%
SiC SiC SiC SiC CeO2 CeO2 CeO2 CeO2
Amount 100% 92% Al 93.1% Al 86.1% Al 78.4% Al 99.3% Al 97.8% Al 99.6% Al 80.6% Al
(%) 8% SiC 6.9% SiC 13.9% SiC 21.6% SiC 0.7% CeO2 2.2% CeO2 0.4% CeO2 19.4% CeO2
A. Nazeer and M. Safiulla
1.4 Conclusion
• Successful synthesis and characterization of Al 6063-silicon carbide and cerium

oxide composites are obtained.
• Tensile strength of composites increased significantly with increased content of
SiC and CeO2.
• Energy absorbed during the impact test on composites decreased with increased
content of SiC and CeO2 in matrix alloy under identical test condition.
• Micro hardness of composites increased significantly with increased content of
SiC and CeO2.
• It has been noted that the compressive strength of composite increased with
increased content of reinforcement and is higher than base aluminum alloy, and
Al6063 SiC composites have higher compressive strength than CeO2 for identical
test condition.
• SEM studies clearly show the modes of fractured occurred.
• XRD analysis confirms the presence of SiC and CeO2 in Al6063 composite with
varying percentage.
References
1. Chakraborty, A., Mitra, R., Kamat, S.V., Jain, M.K., Subramanyam, J.: Structure property
relations for MOSi2 and Nb particle reinforced Al2O3 matrix composite. In: Proceeding of the
Third International Conference on Advances in Composites, pp. 547–558 (2013)
2. Ravi kumar, N.V., Pai, B.C., Dwarakadasa, E.S.: Effect of particle content and matrix stength
on the mechanical properties of Al-ZN-MG/SiCp composites. In: ADCOMP, IISc, pp. 295–301
(1996)
3. Schwartz, M.M.: Composite Materials Handbook. McGraw-Hill Book Company, New York
(1984)
4. Rajesh, P., Rana, R.S., Verma, C.S.: Fabrication of Al–SiCp composite through powder
metallurgy process and testing of properties. Int. J. Eng. Res. Appl. 2, 420–437 (2012)
5. Sekunowo, O.I., Adeosun, S.O., Lawal, G.I., Balogun, S.A.: Mechanical characterization of
aluminium bronze-iron granules composite. Int. J. Sci. Technol. Res. 2(4) (2013)
6. Surappa, M.K.: Metal matrix composites. Mater. Res.: Curr. Scenar. Futur. Proj. 301–318
(2003)
7. Hosamani, G.G., Manoji, V.B., Biradar, V.R., Kallappagol, V.N., Haragapure, L.I.: Character-
ization of silicon carbide reinforced aluminum matrix composite by using stir casting method.
Int. J. Mech. Prod. Eng. (Special issue)
8. Ezatpour, H.R., Sajjadi, S.A., Sabzevar, M.H., Huang, Y.: Investigation of microstructure and
mechanical properties of Al6061-nanocomposite fabricated by stir casting. Mater. Des. 921–
928 (2016)
9. Ramnath, B.V., Elanchezhian, C., Jaivignesh, M., Rajesh, S., Parswajinan, C., Siddique Ahmed
Ghias, A.: Evaluation of mechanical properties of aluminium alloy–alumina–boron carbide
metal matrix composites. Mater. Des. 58, 332–338 (2014)
10. Jagadish, B.S.: Synthesis and characterization of aluminium2024 and graphene metal matrix
composites by powder metallurgy. Int. J. Mech. Eng. (SSRG-IJME) 2(7) (2015)
11. Gupta, M., Surappa, M.K.: Effect of weight percentage of SiC particulates on the ageing
behaviour of 6061/SiC metal matrix composites. J. Mater. Sci. Lett. 141283–141285 (1995)
12. Mohal, S.: Microstructural investigation of Aluminium–silicon carbide particulate metal matrix
composite fabricated by stir casting. Int. J. Innov. Res. Sci. Technol. (IJIRST) 3(11) (2017)
13. Ray, S.: Review synthesis of cast metal matrix particulate composites. Mater. Sci. 28, 5397–
5413 (1993)
14. Dewangan, S.K., Arora, A.: Mechanical properties of aluminum 6063 alloy based graphite
particles reinforced metal matrix composite. Int. J. Sci. Res. Develop. (IJSRD) 3(05), 2321-
0613
Chapter 2
Feasibility Study of Detonation Wave
Initiation by Multiple Structured
Blockages in a Plain Tube
S. Srikrishnan, V. Jayakumar and P. K. Dash
Abstract Pulse detonation engines, due to its simplistic design, have gone through
critical acclaim in recent years as a viable solution of high-speed propulsion in
the aerospace industry. Through various experimental and theoretical methods,
researchers are trying for the most explicit method for initiating a controlled detona-
tion in a plain tube. Studies are being conducted on various types of cross sections
and multiple forms of blockages to evolve detonation using several chemical and
mechanical procedures. Previous study shows a rectangular blockage which blocks
about 60% of the flow at 2/3rd distance from the origin produces best detonation
wave initiating a sustained exit velocity and increment in pressure. The current paper
studies the variation effect of placing multiple blockages in different permutations
along the length of the plain tube and their effect on the onset of detonation. It has
been observed that multiple blockages placed throughout the length inhibit the pos-
sibility of formation of the detonation wave. Further, it has been summarized that
an ascending rectangular cascade produces the best onset of detonation wave with a
proximal maximum velocity of 10,858 m/s at the outlet when a mixture of kerosene
and air at a maximum pressure of 100 MPa is given at the inlet of the tube.
Keywords Detonation · Blockage · Pulse detonation engine
S. Srikrishnan (B)
Institute of Aeronautical Engineering, Hyderabad, India
V. Jayakumar
Amrita School of Engineering, Amrita Vishwa Vidyapeetham Chennai, Chennai, Tamil Nadu,
India
P. K. Dash
NITTE Meenakshi Institute of Technology, Bangalore, India

https://doi.org/10.1007/978-981-15-3254-2_2
12 S. Srikrishnan et al.
2.1 Introduction
Detonation is normally signified by short pulses of combustion using hypergolic mix-

tures like hydrogen and oxygen, and however, any practical application of detonation
requires a steady and constant flow of gases for the effective supply of thrust in any
vehicle. Tubular combustion and transition to detonation is limited by the interval of
feasible detonation. The detonation is possible only by completely filling the control
volume with both fuel and oxidizer mixture and firing in supersonic speed, where
each pulse will give a burst of thrust and the cycle can be repeated. But on the longer
run, the downstream of such vehicles will experience both bursts of high-pressure
waves and vacuum intermittently causing the loss of thrust. To reduce such a loss, the
frequency of the detonation occurring in the tube has to be increased. The frequency
of detonation is primarily dependent on the detonation transition time which in turn
is dependent on internal geometry, fuel and oxidizer selection, initial spark energy
and ambient conditions.
Georgina et al. [1] discussed the effects of orifice plates as a type of blockage
in a tubular section and the propagation of detonation waves through experimental
methods and high-speed photography to show the onset in transparent plastic tubes.
A mixture of hydrogen and oxygen at an initial pressure of 60 kPa is fired in a 1.55 m
plastic tube to try and predict the critical inlet pressure required for detonation initia-
tion and transition. The experiment with 50% blockage ratio did have an increment of
velocity and pressure, but the critical pressure required for the initiation of detonation
could not solved in the experimental procedure.
Bo et al. [2] discussed the effect of spiral blockages on the detonation propagation
of hydrogen–oxygen mixtures with an inclusion of a surface roughness factor for the
blockages. It has concluded that higher the roughness fraction lower the chance of
detonation initiation as the higher surface roughness acts as a detonation quencher
and inhibits the flow. The spiral section initiates better detonation transition speeds,
and the flow is more sensitive across the spiral due to which failure of detonation
is mitigated. Sun et al. [3] studied extensively rotating detonation engines burning a
mixture of hydrogen and oxygen under both premixed and non-premixed conditions.
In a density-based solver across multiple inlet mass flow rates, the rotating detonation
engine behaves similar to a single-wave engine at both premixed and non-premixed
conditions. In case of a higher mass flow rate, the single-wave operation is only
possible during premixed condition and in a non-premixed condition dual-wave flow
occurs. It had been concluded that the thrust offered by the engine burning hydrogen–
oxygen mixture is higher in premixed condition than the non-premixed condition.
Lu et al. [4] discussed the effects of back pressure wave propagation on valve-less
two-phase air-breathing pulse detonation engine. A successful model of pulse deto-
nation combustor and subsequent operations were performed under different firing
patterns where all the tubes were fired in sequential and simultaneous order. It was
found that the pressure peaking feasibility is higher during simultaneous firing of all
the tubes in comparison to the sequential firing. Forlov et al. [5] performed experi-
ments on continuous detonation of ternary hydrogen, liquid propane and air mixture
2 Feasibility Study of Detonation Wave Initiation … 13
in a large-scale annular combustor. The introduction of hydrogen in propane liquid

increases the possibility of detonation wave rather than enriching the air with oxygen
and a continuously rotating detonation wave was achieved for a small duration.
Wenhu et al. [6] simulated an integrated process of flame acceleration and DDT
resulting in detonation propagation in micro-scale and macro-scale channels. It is
found that the viscosity and turbulent flame development play a significant role in
determining the detonation propagation factor in both these channels and that the
rate of detonation propagation across both is not similar. In macro-channels, strong
detonation waves decay over the length and are transformed into quasi-detonation
waves. Zhenhua et al. [7] studied the detonation propagation modes of ethylene–
oxygen mixtures and initiated the fabrication of a helical detonation channel to aid
in the onset of detonation. It has been found that for a stoichiometric mixture of
ethylene and oxygen, a curved section increases the possibility of detonation wave
initiation by morphing the detonation waves from steady to unsteady detonation
waves across the helical section. Continuous failure, restart and regeneration occur
in the helical tube by which a steady stream of detonation wave has been achieved.
Joshua and Deanna [8] studied the effects of nanosecond repetitive pulsed plasma
discharge in the enhancement of detonation waves for a hydrogen–oxygen mixture.
While the plasma pulse is not sufficient for ignition of the fuel oxidizer mixture, it
has been observed that the plasma pulse can enhance the probability of detonation
initiation and transition. Sergio et al. [9] studied the effect of a single blockage in a
flow stream of hydrogen–oxygen enriched air mixture when passed through a tube.
The tube constrained the flow through a convergent-divergent section acting as a
blockage, and it has been found that detonation is initiated in short-run space which
is useful for practical applications.
Pandey and Debnath [10] have reviewed the recent advances in pulse detonation
engines and studied the phenomena of combustion in a PDE combustor through
the effects of fuel–air mixture, combustion, blowdown and purging as a factor of
efficiency. It was observed that detonation wave flow path, ejectors and Mach number
are the main parameters responsible for improving the propulsion performance of
PDE. Further scope of work was also discussed. Hu et al. [11] studied the numerical
simulation of DDT of iso-octane vapor in an obstacle filled tube through a single-
step reaction mechanism under Navier–Stokes equation. It has been found that the
detonation initiation in an iso-octane environment is a function of both pressure and
flame parameters. The temporal variation of total energy release rate and flame front
location parameters were studied, and the flame was categorized into four stages.
In all the above literature, DDT is studied extensively by the variation of fac-
tors like choosing the correct fuel oxidizer mixture, selection of premixed or non-
premixed combustion, position of block and blockage ratio, variation of flow param-
eters, etc. Most of the literature concentrates on hydrogen–oxygen enriched air mix-
ture, but the disadvantage of such a mixture is that they tend to be hypergolic. A
more stable fuel oxidizer mixture is required for sustained ignition and propagation
of deflagration to detonation waves. Previous presented research have extensively
studied the geometry of a blockage and blockage ratio, though not studied exten-
sively the effect of placing multiple blockages throughout the flow stream and their
effect on detonation initiation and propagation [12]. This paper concentrates on the
internal geometry of the tubular section and the effect of placing multiple blockages
throughout the tubular section in relation to the detonation transition [13].
2.2 Methodology
2.2.1 Modeling
The analysis of a pulse detonation engine done in this paper is based on a shock tube
design being a plain tube with multiple blockages at various intervals. A plain tube
of 1 m length is considered for the design where for the purpose of analysis, and the
inlet is designated to be a pressure inlet where a mixture of kerosene gas and air is
supplied at various pressures simulating a diaphragm being blown and the pressure
being expanded into the tube. The diameter of the tube is 1 inch or 2.53 cm, and
the blockages that have been provided at various distances across the tube have a
blockage ratio of 60% of the diameter. The blockage ratio is given by:
Diameter of Tube − Height of Blockage

Blockage Ratio =
Diameter of Tube
The blockages are positioned throughout the length of the tube as a function of
fraction of the tube length like 1/4th, 1/3rd, 1/2nd, 2/3rd and 3/4th of the length of
the tube, which simplifies the distribution of blockages during modeling stages and
capture of the properties across blockages during analysis stages. Figure 2.1 shows
the various forms of models that have been considered for the analysis in this paper.
Model ‘a’ shows a rectangular spiral setup where three blockages are given at
1/3rd, 1/2nd and 2/3rd the distance of the plain tube. It is similar to a coil spring
inserted in the mean of the tube and the blockage ratio of the rectangular element is
60% in each case. Model ‘b’ represents a cascade blockage where three rectangular
blocks are set up in a cascade arrangement continuously at the 1/3rd, 1/2nd and
2/3rd position of the tube. Model ‘c’ shows a plain tube with multiple blockages
which set different positions in the tube length where the blockage height is steadily
Fig. 2.1 Models considered for the analysis of DDT

increased from 20 to 60% of the diameter of the tube. Model ‘d’ shows an eccentric
spiral arrangement. This is almost symmetric to the multi-height blockage stream and
inverse of the same. But in this model, the number of blockages has been increased
to study the effect of placing multiple blockages in the flow stream and determine
the critical limit of blockages present across the tube. Model ‘e’ is the minimized
version of model ‘d’ where the number of blockages is reduced so that a comparative
study between both can conclude whether restricting the number of blockages in the
flow field is a legitimate solution.
2.2.2 Numerical Analysis
The two-dimensional plain tube is modeled using ‘GAMBIT’ as the meshing tool
which has a characteristic ease of usage in 2D meshing and import to any viable
commercial software for analyzing fluid flow where an unstructured mesh is con-
structed to analyze the models in a computational fluid dynamic domain. The scaling
factor is considered in meters for the modeling of the tube as shown in Fig. 2.1. The
meshing has been done at an average rate of 1800–2000 nodes and 1500–1700 ele-
ments so that the analysis is simpler in CFD software tool used. The initial boundary
conditions that were used are listed in Table 2.1.
The models are further analyzed in a computational fluid domain having a fluid
mixture of kerosene vapor and air pass through the tube using the k–ε model. ANSYS
16.0 workbench is the selected solver and Fluent 16.0 standalone system is used to
analyze the above boundary conditions and the results are tabulated and detailed
graphical images have been recorded for the easy assimilation of the data. The
standard equations of the model are given in Eqs. 1 and 2.

∂k ∂[u i k] ∂ νt ∂k
+ = + Pk − (1)
∂t ∂ xi ∂ xi σk ∂ xi

∂ ∂[u i ] ∂ νt ∂ 2
+ = + C1 Pk − C2 (2)
∂t ∂ xi ∂ xi σ ∂ xi k k
Table 2.1 Input parameters for analysis

S. no. Description Input
1 Fuel and oxidizer Kerosene vapor and air
2 Volume fraction 0.5 for kerosene vapor
3 Inlet pressure Range from 1 to 100 MPa
4 Inlet velocity 0.15 m/s
5 Back pressure Vacuum condition
6 Wall function Standard wall functions without slip or wall heating
7 Temperature 1500 K at the inlet for the mixture
where ‘k’ represents the energy present in eddies of various sizes which is governed
by the transport equation with terms that are model of exact terms, and ‘ε’ represents
the small actions of all eddies which aid in the dissipation of the kinetic energy.
Turbulent boundary conditions are used near the wall to resolve steep gradients.
The model is then analyzed in a transient boundary condition, and the time step is
defined for 1 s divided into 1000 steps as the movement speed predicted through the
analysis being very high, anything higher than the division may not be feasible for
the capturing of the various properties of the flow at the exit during the process of the
flow. It is made sure that the flow is controlled by controlling the residue frequencies
and the residue is monitored such that divergence due to temperature increment and
large Courant number does not occur.
2.3 Result and Discussions
The models are analyzed in a domain of fluid where the input is a mixture of kerosene
vapor and air where the inlet pressures range from 1 to 100 MPa. Table 2.2 shows
the increment of exit velocity of the burnt mixture of kerosene vapor and air for the
various types of blockages as shown above.
For a minimum inlet velocity of 0.15 m/s inside the tube, the multi-height blockage
and ascending cascade perform well by increasing the velocity at the exit across all
pressure inputs. The maximum velocity at the exit is achieved by using an ascending
cascade blockage ranging at 10,858 m/s when the input pressure is 100 MPa. Further,
it can also be seen that during the comparison of ascending cascade blockage and
eccentric spiral blockage, even though the blockage is similar in structure, the eccen-
tric spiral blockage performance is inherently low at 7856 m/s for the same input
pressure. It can be concluded that increasing the number of blockages throughout the
length of the tube may inhibit the flow, and the exit velocity dissipation rate would
increase causing an imminent loss. The graphical representation of the above result
is shown in Fig. 2.2.
The variation of exit pressure with variable pressure input for different arrange-
ments of blockages is shown in Table 2.3. The onset of detonation is marked by the
Table 2.2 Input pressure versus exit velocity

Input Exit velocity in m/s
pressure Multi-height Eccentric Cascade Ascending Rectangular
(MPa) blockage spiral blockage cascade spiral
blockage blockage blockage
1 996 779 822 1065 677
10 3198 2470 2611 3378 2146
100 10,500 7856 8809 10,858 6780
Fig. 2.2 Graphical representation of input pressure versus exit velocity
Table 2.3 Input pressure versus exit pressure

Input Exit pressure in MPa
1 1.005 1.02 1.02 1.04 1.013
10 10.06 10.27 10.22 10.47 10.14
100 100.07 102.79 102.75 104.96 101.47
increment of both pressure and velocity at the exit due to the various stages of expan-
sion across the structures of blockages in the flow stream. In the previous comparison,
the multi-height blockage and the ascending cascade were analyzed stating that the
performance of the two types of blockages is similar. But in case of exit pressure as
a factor of detonation initiation and propagation, it is observed that the ascending
cascade blockage performs better than that of the multi-height blockage. The other
types of block pale in comparison with ascending cascade blockage and gross lower
pressures at the exit than the ascending cascade blockage. Across the input pressure
range, the multi-height blockage has the lowest possible exit pressures stating that
the type of blockage may not be feasible for the onset of detonation where it only
increases the exit velocity and not the exit pressure which is almost similar to that of
the inlet pressure or close enough to consider that there is variation of pressure across
the tubular flow. The above statement is contradictory to the jet flow function, but the
placement of the blockages at the various positions with an incremental permutation
Fig. 2.3 Graphical representation of exit pressure versus input pressure
of the height of the blockage may happen to increase the pressure due to the jumping
effect of the flow across the cascade of multiple height blockages.
Further comparing the eccentric spiral, cascade and rectangular spiral blockage,
it is clearly stating that the pressure increment across the input of various pressure is
significantly lesser compared to the ascending cascade blockage but fares well when
compared to the multi-height blockage. Slight change in the placement and height of
the blockage in the eccentric spiral and rectangular spiral blockage may mimic the
results of the ascending cascade but said analysis has not been formulated as ascend-
ing cascade blockage have given the optimum result. The graphical representation
of the comparison between exit pressure and inlet pressure of the mixture is shown
in Fig. 2.3.
The variation of turbulence kinetic energy across the tube with various permuta-
tions of blockages is shown in Table 2.3. The turbulence kinetic energy is physically
formulated from the root mean square of the velocities and is the kinetic energy
associated with the eddies in the turbulent flow. The turbulent kinetic energy across
the comparative data is the highest with the ascending cascade blockage. This shows
that the ascending cascade blockage arrangement to the most suitable type of the
blockage arrangement in a tubular pulse detonation engine design. Figure 2.4 shows
the contour representation of turbulence kinetic energy through an ascending cascade
blockage where the input pressure of the mixture is 1 bar (Table 2.4).
Characteristically, the flow remains almost steady till it reaches the imminent
blockage and once it crosses the blockage, the flow kinetic energy rapidly increases
due to the orifice effect. Subsequent blockages increase the turbulent kinetic energy
due to the flow jumping and jet flow effect evolving a maximum of 54,311 m2 /s2
at the exit of the tubular section. This value is comparatively lower than that of the
multi-height blockage and a cascade blockage, but other significant properties like
pressure and velocity at the exit for an ascending cascade blockage are higher in
Fig. 2.4 Turbulent kinetic energy (100 MPa input pressure)
Table 2.4 Turbulence kinetic energy versus input pressure

Input Turbulence kinetic energy in m2 /s2
1 58,678 37,913 66,849 54,311 35,159
10 591,655 389,691 681,500 546,620 358,082
100 4,014,101 2,572,688 4,180,167 4,478,380 3,096,003
case of an ascending cascade blockage. It can also be seen that at 100 MPa pressure
input the maximum value of turbulent kinetic energy occur when using an ascending
cascade blockage having 4,478,380 m2 /s2 .
2.4 Conclusion
With reference to the above results and the detailed discussions, the following
conclusion remarks can be achieved:
1. The ascending cascade blockage is the most viable type of flow restrictor and
produces the ideal exit velocity of 10,858 m/s and exit pressure of 104.96 MPa
for an inlet pressure of 100 MPa in a mixture of kerosene vapor and air jet.
2. The number of blockages across the flow stream must be restricted as 3 blocks
per meter of the tube length. Exceeding this limit starts to inhibit the flow causing
pressure and velocity loss at the exit as shown by the eccentric spiral blockage
when compared to the multi-height or the ascending cascade blockage.
3. Turbulent kinetic energy plays a significant role in the onset of detonation but
can be compromised when other salient properties like exit pressure and velocity
are higher which aid in the initiation and propagation of detonation waves.
References
1. Georgina, R., Deepinder, J.S.A., Gaby, C.: Visualization of detonation propagation in a round
tube equipped with repeating orifice plates. Combust. Flame 198, 205–221 (2018)
2. Bo, Z., Hong, L., Cheng, W.: On the detonation propagation behavior in hydrogen-oxygen
mixture under the effect of spiral obstacles. Int. J. Hydrogen Energy (2017)
3. Sun, J., Zhou, J., Liu, S., Lin, Z.: Numerical investigation of a rotating detonation engine under
premixed/non-premixed conditions. Acta Astranautica (2018)
4. Lu, J., Zheng, L., Wang, Z., Peng, C., Chen, X.: Operating characteristics and propagation
of back-pressure waves in a multi-tube two-phase valveless air-breathing pulse detonation
combustor. Exp. Therm. Fluid Sci. 61, 12–23 (2015)
5. Forlov, S.M., Aksenov, V.S., Ivanov, V.S., Shamshin, I.O.: Continuous detonation combustion
of ternary “hydrogen-liquid propane-air” mixture in annular combustor. Int. J. Hydrogen Energy
1–13 (2017)
6. Wenhu, H., Yang, G., Chung, K.L.: Flame acceleration and deflagration-to-detonation transition
in micro- and macro-channels: an integrated mechanistic study. Combust. Flame 176, 285–298
(2017)
7. Zhenhua, P., Jun, Q., Jianfeng, P., Penggang, Z., Yuejin, Z., Mingyue, G.: Fabrication of a
helical detonation channel: effect of initial pressure on the detonation propagation modes of
ethylene/oxygen mixtures. Combust. Flame 192, 1–9 (2018)
8. Joshua, A.T.G., Deanna, A.L.: Enhancement of the transition to detonation of a turbulent
hydrogen–air flame by nanosecond repetitively pulsed plasma discharges. Combust. Flame.
199, 258–266 (2019)
9. Sergio, B., Joshua, A.T.G., Julius, R., Jonas, P.M., Oliver, C.P., Jorn, S.: Detonation initiation
in pipes with a single obstacle for mixtures of hydrogen and oxygen-enriched air. Combust.
Flame 198, 290–304 (2018)
10. Pandey, K.M., Debnath, P.: Review on recent advances in pulse detonation engines. J. Combust.
1–16 (2016)
11. Hu, M., Zhenjuan, X., Wei, G., Changfei, Z., Dong, W.: Numerical simulation of the deflagration
to- detonation transition of iso-octane vapor in an obstacle-filled tube. Int. J. Spray Combust.
Dyn. 10(3), 244–259 (2018)
12. Srikrishnan, S., Dash, P.K., Jayakumar, V.: Evaluation of critical blockage ratio and pulse length
in a pulse detonation engine using CFD and MATLAB. In: MATEC Web of Conferences, vol.
172, pp. 1–8 (2018)
13. Srikrishnan, S., Dash, P.K.: 2D CFD analysis of deflagration to detonation transition in closed
pipe using different blockage. IJMET 8(6), 447–454 (2017)
Chapter 3
Spectral Density at Roof of Setback Tall
Building Due to Time Variant Wind Load
Abstract Turbulence due to wind around conventional square shape bluff body
creates the pressure difference of different planes of the body. Nevertheless, the
unconventional bluff body creates a large amount of turbulence around its face and
on the roof region. This paper highlights the pressure variation of top roof and setback
roof of square and setback tall building with respect to different time domains. The
frequency of the roof due to wind also affects the pressure fluctuation on neighbor
faces. Most of the pressure fluctuation develops at 0.06 s which forms the initial time
and maximum pressure difference occurred at the setback roof for along wind con-
dition. The present study concentrated on the fluctuation of time-dependent pressure
between top and setback roof to take special care during the design.
Keywords Spectral density · Setback tall building · Roof pressure
3.1 Introduction
Velocity of wind changes continuously with the building height. Similarly, the pres-
sure around the building always fluctuates with respect to time, and therefore, the
frequency and the spectral density change abruptly. The turbulence around the con-
ventional tall building changes its vortex for the unconventional setback tall building.
The turbulence, pressure, frequency and spectral density on the top roof of the build-
ing have some amount of difference in setback roof of the building. A number of past
studies were highlighted the wind effect on different unconventional tall buildings.
Kim et al. [1] studied the tapered tall building aeroelastic models with different taper
ratios. Kim and Kanda [2] highlighted the static and dynamic wind pressure distri-
bution on tapered and setback tall building. Tanaka et al. [3] presented the wind flow
A. K. Bairagi (B) · S. K. Dalui

Department of Civil Engineering, Indian Institute of Engineering Science and Technology,
Shibpur, Howrah, India
S. K. Dalui

https://doi.org/10.1007/978-981-15-3254-2_3
22 A. K. Bairagi and S. K. Dalui
characteristics on thirty-four number of tall buildings with analytical and experimen-

tal tools. Bairagi and Dalui [4, 5] estimated the optimum distance where interference
effects nullify and principal building behave like isolate building. Mukherjee and
Bairagi [6] focused the wind behavior around ‘N’ plan shape tall building by CFD
analysis. Mendis et al. [7] developed the solution of different mistakes of CFD analy-
sis. Xu and Xie [8] evaluated the optimization of aerodynamic effect on tall building
and perfect fitted wind angle. Roy and Bairagi [9] presented the wind-induced pres-
sure and force variation on setback tall building with different geometrical shapes.
Tamura et al. [10] estimated the dynamic response on tall building for different
configurations. Elshaer and Bitsuamlak [11] optimized the opening of tall build-
ing to reduce the wind-induced load. Bairagi and Dalui [12] compared the different
aerodynamic parameters between various setback tall buildings. Namchu et al. [13]
highlighted the pressure coefficients on tall chimney for different wind terrain condi-
tions. Bairagi and Dalui [14] highlighted the aerodynamic effects and power spectral
density on setback roof compared to the top roof of setback model. Mukherjee and
Bairagi [15] studied the interference effect on square plan shape tall building due to
setback model for different orientations. Rajasekarababu and Vinayagamurthy [16]
studied the experimental analysis of sharp edge setback model of aspect ratio 1:5.
The study described the hybrid turbulence models, which used delayed detached
eddy simulation (DDES) and improved delayed detached eddy simulation (IDDES)
and introduced the treatment of wall for roughness parameter combination.
The present study is based on computational fluid dynamics (CFD) simulation and
highlights the variation of pressure on top roof and the setback roof on different time
intervals and the spectral density of that particular region for along and across wind
conditions. The considered models are square plan shape bluff body and a setback
tall model of the setback roof on both sides and at half of the height from base of the
model.
3.2 Model Detail
Two numbers of analytical models have been placed inside the domain and analyze
the spectral density and pressure variations on the top surfaces of the models. The
square (SQ) model has length (L): breath (B) which was 1 and height (H): length (L)
which was 2. A setback (SB) model also considered with the same aspect ratio of SQ
model. The setback distance used 20% of the length of the model and placed on both
sides and half-height of model from base. As the setbacks are equally distributed, the
total area of setback roof and top roof of the SB model compared it to SQ model. The
wind incidence angels are considered from 0° to 90° at 15° intervals (see Fig. 3.1).
3 Spectral Density at Roof of Setback Tall … 23
Fig. 3.1 Model details for a SQ and b SB
3.3 Analytical Domain and Mesh
The computational fluid dynamics (CFD) is a widely accepted method of for the
analytical application of wind engineering. The present study based on this CFD
simulation. The models are placed inside the analytical domain at 5H from the
extreme edge of the model to the inlet and sidewalls of the domain and 15H and 6H
for outlet and top of model, respectively, as stated by Frank et al. [17] (see Fig. 3.2a).
The height of analytical model is presented by H. The boundary conditions are free
slip for the sidewall of domain, i.e., U wall = 0, τ w = 0 and no slip for model wall,
i.e., U wall = 0. Here, U wall is velocity normal to the wall, and τ w is the wall shear
stress. The power law is introduced in this study to estimate the velocity around the
experimental model as explained in SP:64 (S&T) [18] as shown in Eq. 3.1.
U/U H = (Z /Z H )α (3.1)
where U represents the horizontal wind speed at an elevation Z; U H represents the

10 m/s speed at the reference elevation Z H ; α represents the power law index 0.133
for terrain category 2; and Z H is 1.0 m. The kinetic energy of turbulence and its
Fig. 3.2 a Computational domain for CFD simulation; b mesh detail of SQ model; and c mesh
detail of SB model
Fig. 3.3 a Validation of velocity and turbulence intensity of wind between present studies with
Kim Kanda [2]; b pressure tapping points on the roof of SQ; and c for SB
dissipation rate at the inlet section are calculated according to Eq. 3.2.
2
k = 1.5 Uavg × I and ε = Cμ0.75 × k 1.5 /l (3.2)
where U avg represents the mean velocity at the inlet; I represents the turbulence
intensity; and l denotes the turbulence integral length scale. The k − ε turbulence
model has air temperature 25 °C with tetrahedron meshing (see Fig. 3.2a, b).
3.4 Comparison with Analytical Study
This analytical study also validated with the experimental study of setback model
discussed by Kim and Kanda [2]. The experimental model was 40 m × 24 m ×
160 m with 8 m setback in 1:400 scale was a study in 1.8 m × 1.8 m × 12.5 m wind
tunnel at the University of Tokyo. The simulated model has the same aspect ratio
with the same boundary conditions. The velocity profile and turbulence intensity of
present study validated with Kim and Kanda [2] as shown in Fig. 3.3a.
3.5 Results and Discussion
Pressure variation of unconventional rooftop is quite different compared to the con-

ventional rooftop of tall building. In this study, the setback model has three numbers
of roof. One is top and another two are setback at half of the building height. The
attacking wind creates a huge amount of turbulence near the setback roof and sur-
rounding surface of the model. Bairagi and Dalui [19] conferred the variation of
spectral density on the setback roof of tall model. The authors were considered an
unconventional setback model which had setback on one side. This study is discussed
the pressure variation, and the power spectral density calculation is carried out in
this study for along the wind and across the wind conditions. The pressure tapping
points are considered at the edge of the roof of SQ and SB model (see Fig. 3.3b, c).
The points #A and #B are located at 0.04 L from the edge of top roof of SQ model.
Similarly, for the SB model has #1, #2, #3 and #4 are the pressure tapping points.
The points #1 and #4 are situated at 0.15L from the edge and #2 and #3 are 0.24 L
and 0.96 L, respectively, from the edge.
3.5.1 Pressure-Transient Analysis
The change of pressure over time is the pressure-transient analysis. This study high-
lighted the pressure fluctuation at the rooftop of SQ and SB model followed by
Eq. 3.3.

Cp (t) = [P(t) − P0 ]/ 0.5ρVz2 (3.3)
where P(t) is the pressure at time t, P0 is the reference static pressure of that particular
point, ρ is the density of air and V z is the mean velocity at the height of the roof.
Here, negative sign denotes suction, and positive sign presents the pressure.
Pressure coefficient at top roof and setback roof with respect to time for along and
across wind conditions has been plotted (see Fig. 3.4). From the figures, it is clear that
the high amount of pressure fluctuation is developed at first 0.06 s. The maximum
pressure coefficient (3.52) observed at the beginning of the flow of the SB model at
the tapping point #3 and maximum suction (−1.8) noticed at 1.43 s at tapping point
Fig. 3.4 Comparison of pressure transient for SQ and SB model at a, b top roof along wind and
across wind; c, d setback roof of SB model along wind and across wind
#A of SQ model for along wind condition as shown in Fig. 3.4a. The point #B and
#3 has maximum pressure 1.38 for 2.85 s and 1.66 for 2.82 s and suction is −1.6
for 1.46 s and −1.25 for 1.2 s. For across wind condition, maximum pressure (1.38)
is developed at #B for SQ model at 2.87 s and maximum suction −2.08 at #3 for
1.42 s (see Fig. 3.4b). After that, it may be said that the pressure coefficient of SB is
greater than the SQ model. Again, the pressure in the setback zone at #1 and #4 has
1.55 and 1.87 for 2.82 s and 0.02 s for along wind condition (see Fig. 3.4c), where
maximum suction is 1.44 in 1.28 s at #4 point. For across wind condition, pressure
has 1.07 and 1.30 in 2.56 s and 2.79 s. The suction is −2.02 is same for #1 and #4
points for 1.34 s (see Fig. 3.4d). According to the above condition, it may be said
that the pressure variation is maximum at setback region due to a large amount of
turbulence due to setback for along wind condition.
3.5.2 Spectral Density on Top and Setback Roof
Power spectral density (PSD) is the process, which defines the strength of energy as
a function of frequency variation of a particular point. In this connection, anyone can
calculate the frequency and amplitude of a variable signal in a time series. Another
non-dimensional part Strouhal number takes place in this oscillating flow mechanism.
The equations of PSD and Strouhal number are described under Eqs. (3.4)–(3.5),
respectively.
[Sp ( f )]/σ 2 (3.4)
f B/U (3.5)
where S p is the power spectral density in energy/frequency, f is the frequency in Hz,

σ is the standard deviation of pressure variation with respect to time, U is the flow
velocity in m/s and B is height of pressure tapping point marked as #.
Figure 3.5 represents the PSD of top roof and the setback roof of SQ and SB
models for across and along wind conditions at the points as stated in the previous
section. For along wind condition, the spectral density of farthest point #B has high
value compared to the #A at the Strouhal number 0.1 (see Fig. 3.5a). Similarly, the
farthest point #2 has high spectral density with respect to #1 and has Strouhal number
0.05. Again for across wind condition, the SB model has less spectral value compared
to SQ model at the Strouhal number 0.07 in Fig. 3.5b. In the case of setback roof,
the spectral value of #1 is 0.12 for the fB/U value 0.025, whereas at point #4 has
spectral value 0.05 for fB/U value 0.038 (see Fig. 3.5c). From this graph, it may be
said that the high amount of pressure fluctuation observed in the setback zone of
SB model for along wind condition. Figure 3.5d presents the spectral density curve
for across wind condition. Here, the spectral density fluctuation is minimum except
the Strouhal number 0.076. In this zone, high amount of spectral density variation is
Fig. 3.5 Comparison of spectral density of SQ and SB model at a, b top roof along wind and across
wind; c, d setback roof of SB model along wind and across wind
compared to #1 and #4. According to those graphs, it is clear that the SB model has
higher spectral density variation in its setback region and affects the neighbor faces
to change its frequency due to the high amount of turbulence. Therefore, special care
should be adopted to design the setback roof and the neighbor face compared to the
conventional roof of a tall building.
3.6 Conclusion
The CFD simulation has been studied in this paper for square and setback tall model
to calculate the pressure transient and power spectral density at the top roof and
setback roof. The number of simulations has been conducted in CFD analysis for
along and across wind conditions. The results convey the message to the designer to
consider special care to design the setback roof and adjacent wall of the setback tall
building. The number of important features is observed in this study and stated as
follows.
• High amount of pressure fluctuation is observed at first 0.06 s for both square and
setback model.
• Maximum pressure (3.52) has been developed on the top roof of the setback model
at the initial time for along wind condition. However, the square model has less
pressure at the same time series.
• The maximum pressure variation has been observed for across wind condition on
the setback model for 0.15 s.
• The farthest setback roof for along wind condition has maximum pressure coef-
ficient 1.87 for 0.02 s and 1.30 for 2.77 s and has the same suction (−2.02) for
1.34 s.
References
1. Kim, Y.M., You, K.P., Ko, N.H.: Across-wind responses of an aeroelastic tapered tall building.
J. Wind Eng. Ind. Aerodyn. 96, 1307–1319 (2008)
2. Kim, Y.C., Kanda, J.: Wind pressures on tapered and set-back tall buildings. J. Fluids Struct.
39, 306–321 (2013)
3. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M., Kim, Y.C., Bandi, E.K.: Aerodynamic and flow
characteristics of all buildings with various unconventional configurations. Int. J. High-Rise
Build. 2, 213–228 (2013)
4. Bairagi, A.K., Dalui, S.K.: Evaluation of interference effects on parallel high-rise buildings for
different orientation using CFD. In: 3rd World Conference on Applied Sciences, Engineering
& Technology, pp. 764–774. Kathmandu, Nepal (2014)
5. Bairagi, A.K., Dalui, S.K.: Optimization of interference effects on high-rise building for
different wind angle using CFD simulation. Electron. J. Struct. Eng. 14, 39–49 (2014)
6. Mukherjee, A., Bairagi, A.K.: Wind pressure and velocity pattern around ‘N’ plan shape tall
building—A case study. Asian J. Civil Eng. (BHRC) 18, 1241–1258 (2017)
7. Mendis, P., Mohotti, D., Ngo, T.: Wind design of tall buildings, problems, mistakes and solu-
tions. In: 1st International Conference on Infrastructure Failures and Consequences, Melbourne,
Australia (2014)
8. Xu, Z., Xie, J.: Assessment of across-wind responses for aerodynamic optimization of tall
buildings. Wind Struct. 21, 505–521 (2015). (Techno-Press Ltd.)
9. Roy, K., Bairagi, A.K.: Wind pressure and velocity around stepped unsymmetrical plan shape
tall building using CFD simulation—A case study. Asian J. Civil Eng. (BHRC) 17, 1055–1075
(2016)
10. Tamura, Y., Xua, X., Tanakac, H., Kima, Y.C., Yoshidaa, A., Yangd, Q.: Aerodynamic and
pedestrian-level wind characteristics of super-tall buildings with various configurations. In: 10th
International Conference on Structural Dynamics, EURODYN, vol. 199, pp. 28–37. Procedia
Engineering (2017)
11. Elshaer, A., Bitsuamlak, G.: Multiobjective aerodynamic optimization of tall building openings
for wind-induced load reduction. American Society of Civil Engineers (2018)
12. Bairagi, A.K., Dalui, S.K.: Comparison of aerodynamic coefficients of setback tall buildings
due to wind load. Asian J. Civil Eng. Build. Housing 19, 205–221 (2018)
13. Namchu, A.D., Bairagi, A.K, Chakroborty, S.: Aerodynamic coefficients of steel stacks under
different terrain category. In: Proceeding of International Conference on Frontier in Engineering
Applied Science and Technology (FEAST’18), pp. 133–138. NIT Tiruchirappalli (2018)
14. Bairagi, A.K., Dalui, S.K.: Aerodynamic effects on setback tall building using CFD simulation.
In: 2nd International Conference on Advances in Dynamics, Vibration and Control, pp. 381–
388. NIT Durgapur (2018)
15. Mukherjee, S., Bairagi, A.K.: Interference effect on principal building due to setback tall
building under wind excitation. In: SEC18, Proceedings of the 11th Structural Engineering
Convention—2018. Jadavpur University, Kolkata, India (2018)
16. Rajasekarababu, K.B., Vinayagamurthy, G.: Experimental and computational simulation of an
open terrain wind flow around a setback building using hybrid turbulence models. J. Appl.
Fluid Mech. 12, 145–154 (2019)
17. Franke, J., Hirsch, C., Jensen, A., Krüs, H., Schatzmann, M., Westbury, P., Miles, S., Wisse, J.,
Wright, N.G.: Recommendations on the use of CFD in wind engineering. COST Action C14.
Impact of Wind and Storm on City Life and Built Environment. Von Karman Institute for Fluid
Dynamics (2004)
18. SP 64 (S&T): Explanatory Hand Book on Indian Standard Code of Practice for the Design
Loads (Other than Earthquake) for Buildings and Structures (Part-3. Wind Loads) [IS:875
(Part-3):1987]. Bureau of Indian Standards, New Delhi, India (2001)
19. Bairagi, A.K., Dalui, S.K.: Comparison of pressure coefficient between square and setback
tall building due to wind load. In: SEC18, Proceedings of the 11th Structural Engineering
Chapter 4
Distribution of Wind Pressure Around
Different Shape Tall Building
Abstract Wind effect is the most interesting and important parameter for different
structural elements like sidewall, roof and around the building also. A number of
researchers were presented their thoughtful inspections on different, unconventional
tall buildings due to wind issue. The present study focuses on the pressure distribution
around the square and setback tall buildings due to wind load. The models have
single and double type setback at different elevations. The pressure calculation was
conducted by analytical study of plane and highlights the pressure fluctuation. Some
amount of pressure bulb was observed on the leeward side of setback model, which
mean the increase of suction on that particular region. The excessive amount of
suction envelop recognized at the top roof of setback model compared to square
model.
Keywords Setback model · Pressure coefficient · Tall building
4.1 Introduction
According to architectural point of view, setback tall building always robbed the
feather of elegance. Environmental effects like wind also claim to reveal around
the building and its surrounding region. Mendis et al. [1] enumerated simple quasi-
static treatment of wind load on tall buildings. Irwin et al. [2] established the energy
in tall building increased with the increase in the height of a tall building. Kim
et al. [3] carried out for three aeroelastic, tapered, tall building models with taper
ratios of 5, 10 and 15%. Kim and Kanda [4] focused the wind pressure on the
setback and tapered shape tall buildings in both static and dynamic for different
flow condition. Tanaka et al. [5] studied the aerodynamic response due to wind and
A. K. Bairagi (B) · S. K. Dalui

S. K. Dalui

https://doi.org/10.1007/978-981-15-3254-2_4
flow characteristics of tall buildings with thirty-four numbers unconventional shapes

in wind tunnel test and CFD simulation. Bairagi and Dalui [6, 7] evaluated the
interference effects, pressure coefficients and optimum distance on parallel high-rise
buildings for different orientation using CFD. Mukherjee and Bairagi [8] studied the
‘N’ plan shape tall building and evaluated the pressure, force and velocity around the
model. Mendis et al. [9] discussed a number of problems, mistakes and solutions for
CFD wind analysis. Baby et al. [10] presented an overview of the optimal external
shape and structural system for tall buildings subject to aerodynamic loads and the
response of a structure through a comprehensive investigation of the building. Xu and
Xie [11] focused the aerodynamic optimization of tall buildings and best compromise
wind issues. Roy and Bairagi [12] discussed wind pressure and force coefficients on
stepped tall building at different geometrical shape placed on above to each other like
rectangular, square and triangular. Tamura et al. [13] conferred pedestrian level and
aerodynamic wind characteristics of super-tall buildings with various configurations
and conducted the dynamic wind response. Mittal H et al. [14] investigated the effect
of building shape (square, tapered and setback) and wind direction on pedestrian level.
Bairagi and Dalui [15] discussed comparison of aerodynamic coefficients between
two setbacks tall buildings due to wind load. Namchu et al. [16] highlighted the
pressure coefficients on tall chimney for different wind terrain condition. Bairagi and
Dalui [17] highlighted the aerodynamic effects and power spectral density on setback
roof compare to the top roof of setback model. Bairagi and Dalui [18] focussed on
pressure coefficient on square and both side setback tall building and concluded
the pressure coefficient of the setback roof was 205.4% more effective than the top
roof. The present study discussed the pressure coefficient around the square and the
setback tall building at different plane for along and across wind condition.
4.2 Description of Model
Different types of unconventional modes were analyzed by the wind tunnel test and
the test is expensive. To overcome this situation, computational fluid dynamics (CFD)
is a widely acceptable simulation process of wind analysis. This study based on this
type of simulation. Three sets of models, namely S1, S2 and S3, are used in this
study. All the models have l/b ratio is 1 and the h/b is 2. The S1 model has same plan
area along height, but S2 and S3 changed their plan area h/2 and h/3 level from the
ground. The S2 model has 20% setback at h/2 level and S3 has 10% setback at each
h/3 and 2h/3 level. The attacking wind angles are from 0° to 180° at 15° interval. The
considered h, b and l are 500 mm, 250 mm and 250 mm, respectively (see Fig. 4.1).
4 Distribution of Wind Pressure Around Different Shape … 33
Fig. 4.1 Elevation and plan of model. a S1, b S2 and c S3
4.3 Analytical Domain and Mesh
The three models were placed inside the domain for CFD simulation. The boundary
of domain has been set 5H from inlet and both sidewalls and 6H from the base of
model as stated by Frank et al. [19], Revuz et al. [20] (see Fig. 4.2a). Here, H is
the height of the analytical model. The boundary conditions for different walls and
different non-dimensional parameters are stated in Table 4.1.
Fig. 4.2 a Computational domain for CFD simulation. b Mesh detail of square model
Table 4.1 Boundary conditions and non-dimensional parameters

Condition Parameters
Flow regime Subsonic
Turbulence intensity Low (1%)
Method of mesh Tetrahedron
Inlet U/U H = (Z/Z H )α
Relative pressure of outlet Zero
Sidewall Free slip
Model wall No slip
Air temperature 25 °C
Model wall roughness Smooth wall
Velocity of wind 10 m/s
Ground roughness (α) 0.133
k 3/2 (U avg × I)2
(3/4)
ε Cμ × k (3/2) /l
Where U is the horizontal wind speed at an elevation Z; U H is the speed at the

reference elevation Z H ; which was 10 m/s; Z H is 1.0 m; k is the kinetic energy of
turbulence and ε is the dissipation rate; U avg is the mean velocity at the inlet; I is
the turbulence intensity; l is the turbulence length scale; C μ is the turbulence model
constant, i.e., 0.09. The mesh detail of the S1 model (see Fig. 4.2b).
4.4 Comparison with Analytical Study
The analytical study validated with the experimental study of square model discussed
by Kim and Kanda [21]. The experimental study was conducted in Eiffel-type wind
tunnel at the University of Tokyo. The experimental model had L = 100 mm, B =
100 mm, H = 400 mm and length scale 1/400 with α = 0.13. The blockage ratio was
1.2% with wind flow at 6.5 m/s. The simulated square model has the same aspect ratio
and has same non-dimensional parameters adopted by the author. The validation of
turbulence intensity and mean wind speed as shown in Fig. 4.3.
Pressure contour around the models S1, S2 and S3 are presented in this study. The
pressure distribution on YZ and XZ plane has been studied for along the wind and
across wind conditions. Figure 4.4a shows the wind pressure contour at YZ plane
and at the center of model for along wind condition for S1 model. For along wind
Fig. 4.3 Validation of turbulence intensity and mean wind speed of square model with Kim and
Kanda [21] and present study
Fig. 4.4 Wind pressure coefficient around S1 model for a along wind condition, b across wind
condition
condition, the windward face encircled by the ranged from 1.2 to 0.1 and the top and
leeward faces ranged from −0.1 to −0.5. In this connection, the excessive amount of
suction shown in purple color at −0.5 on leeward zone. Similarly, the pressure contour
at the XZ plane and the center of the model for across wind conditions showcased in
Fig. 4.4b. For this condition, the pressure envelope ranged from −0.1 to −0.75 and
the values are high suction. A comparative study also carried out between S2 and S3
models on YZ and XZ plane for 0°, 90° and 180° wind incidence angles as shown in
Table 4.2. Mou et al. [22] found the large amount of pressure difference on leeward
surface further established the negative and positive pressure around the building
depends upon the width of the building. No special change has been focused for
along wind condition for 0° wind on YZ and XZ plane except the bulb region on the
leeward face of S2 and S3 models compared to S1 model. However, large amounts of
pressure variant take place on opposite sides of the setback face on the YZ plane for
90° wind angle. S2 model has line −0.85 near the top and bottom of setback zone.
However, S3 model has the same line at the lowest part of setback zone. A beautiful
violet color zone (−0.7) observed between Y ranges from 0.1 to 0.2. Simultaneously,
model S2 has a light green color (−0.45) on that particular Y range. For along wind
Table 4.2 Wind pressure coefficient around models S2 and S3 at YZ and XZ plane for 0°, 90°
and 180° wind angle
Wind Pressure S2 S3
angle plane
0° YZ
XZ
90° YZ
XZ
180° YZ
XZ
condition on the XZ plane has a pressure contour line −0.3, which covered in most
of the area for the S3 model. However, S2 model has a line range from −0.2 to −
0.6. Therefore, it is clear that the large amount of suction developed in the leeward
region due to decrease of number of setback roof for 90° wind angle. For 180° wind
angle, the windward face has line 1.2 and −0.1 at the leeward face of S2 model. On
the other hand, the S3 model has a large area of line −0.6 on the leeward side, but
no lines of 1.2 are visible at windward face. The XZ plane for 180° wind angle has
large amount of suction on the top roof at line −0.7 for S2 model and line −0.6 at
the rooftop of S3 model. It is clear that the pressure on top of the setback roof has
maximum suction (−0.7) for S2 model compared to S1 model (−0.35).
4.6 Conclusion
The present paper focused on the wind pressure variation around the square and
different setback models for along the wind and across the wind conditions. The
following conclusions are made after a lot of analytical simulations.
• The leeward pressure is maximum on S1 model for 0° wind of line −0.5. A bulb
region created on the leeward side of S2 and S3 models of line −0.4.
• The excessive amount of suction matured at leeward side due to decrease of number
of setback roof for 90° wind angle.
• The pressure coefficient is quite large for along wind condition on S2 model for
180° wind angle.
• The pressure on the rooftop of S1 model has maximum compared to other two
models.
References
1. Mendis, P., Ngo, T., Haritos, N., Hira, A., Samali, B., Cheung, J.: Wind loading on tall buildings.
Electron. J. Struct. Eng. Special Issue: Loading on Structures. 41–54 (2007)
2. Irwin, P., Kilpatrick, J., Robinson, J., Frisque, A.: Wind and tall buildings: negatives and
positives. Struct. Des. Tall Spec. Build. 17, 915–928 (2008)
3. Kim, Y.M., You, K.P., Ko, N.H.: Across-wind responses of an aeroelastic tapered tall building.
J. Wind Eng. Ind. Aerodyn. 96, 1307–1319 (2008)
39, 306–321 (2013)
5. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M., Kim, Y.C., Bandi, E.K.: Aerodynamic and flow
characteristics of tall buildings with various unconventional configurations. Int. J. High-Rise
Build. 2, 213–228 (2013)
different orientation using CFD. In: 3rd World Conference on Applied Sciences, Engineering
& Technology, pp. 764–774. Kathmandu, Nepal (2014)
7. Bairagi, A.K., Dalui, S.K.: Optimization of interference effects on high-rise building for
8. Mukherjee, A., Bairagi, A.K.: Wind pressure and velocity pattern around ‘N’ plan shape tall
building—A case study. Asian J. Civil Eng. (BHRC) 18(8), 1241–1258 (2017)
9. Mendis, P., Mohotti, D., Ngo, T.: Wind design of tall buildings, problems, mistakes and solu-
tions. In: 1st International Conference on Infrastructure Failures and Consequences, Melbourne,
Australia (2014)
10. Baby, S., Jithin„ P.N., Thomas, A.M.: A study of wind pressure on tall buildings and its
aerodynamic modifications against wind excitation, Int. J. Eng. Dev. Res. 3 (2015)
11. Xu, Z., Xie, J.: Assessment of across-wind responses for aerodynamic optimization of tall
buildings. Wind Struct. 21, 505–521 (2015)
tall building using CFD simulation—A case study. Asian J. Civil Eng. (BHRC) 17, 1055–1075
(2016)
13. Tamura, Y., Xua, X., Tanakac, H., Kima, Y.C., Yoshidaa, A., Yangd, Q.: Aerodynamic and
pedestrian-level wind characteristics of super-tall buildings with various configurations. In: 10th
International Conference on Structural Dynamics, EURODYN, vol. 199, pp. 28–37. Procedia
Engineering (2017)
14. Mittal, H., Sharma, A., Gairola, A.: Pedestrian level wind characteristics around tall build-
ings: effect of building shape and wind direction. In: 9th Asia-Pacific Conference on Wind
Engineering, Auckland, New Zealand (2017)
due to wind load. Asian J. Civil Eng. Build. Housing 19, 205–221 (2018)
16. Namchu, A.D., Bairagi, A.K, Chakroborty, S.: Aerodynamic coefficients of steel stacks under
different terrain category. In: Proceeding of International Conference on Frontier in Engi-
neering, Applied Science and Technology, (FEAST’18), pp. 133–138. NIT Tiruchirappalli
(2018)
In: 2nd International Conference on Advances in Dynamics, Vibration and Control, pp. 381–
388. NIT Durgapur (2018)
18. Bairagi, A.K., Dalui, S. K.: Comparison of pressure coefficient between square and setback
tall building due to wind load. In: SEC18: Proceedings of the 11th Structural Engineering
19. Franke, J., Hirsch, C., Jensen, A., Krüs, H., Schatzmann, M., Westbury, P., Miles, S., Wisse,
J., Wright, N.G.: Recommendations on the use of CFD in wind engineering. In: COST Action
C14. Impact of Wind and Storm on City Life and Built Environment. Von Karman Institute for
Fluid Dynamics (2004)
20. Revuz, J., Hargreaves, D.M., Owen, J.S.: On the domain size for the steady-state CFD modelling
of a tall building. Wind Struct. 15 (2012)
21. Kim, Y.C., Kanda, J.: Characteristics of aerodynamic forces and pressures on square plan
buildings with height variations. J. Wind Eng. Ind. Aerodyn. 98, 449–465 (2010)
22. Mou, B., Heb, B.J., Zhaoc, D.X., Chaud, K.: Numerical simulation of the effects of building
dimensional variation on wind pressure distribution. Eng. Appl. Comput. Fluid Mech. 11(1),
293–309 (2017)
Chapter 5
Material Characterization of Pin-Core
Latex Polymer Foam Under Static
and Dynamic Loads
K. Venkataramana , R. K. Singh, Anindya Deb, Vivek Bhasin, K. K. Vaze

and H. S. Kushwaha
Abstract Cellular polymer foams find extensive applications as energy absorbers

under static, impact and blast loads due to their capacity to absorb energy under con-
stant stress up to full densification strain (Ashby et al. Metal foams: a design guide.
Butterworth-Heinemann, Oxford, 2000 [1]; Gibson and Ashby, Cellular solids: struc-
ture and properties. Cambridge University Press, Cambridge, 1999 [2]). In the present
paper, the material characterization of natural latex cellular polymer foam used in
the field air blast experiments (Venkataramana et al. Proc Eng 173:547–554, 2017
[3]; Venkataramana et al. Numerical simulation of blast wave mitigation using foam
impregnated with water. DAE BRNS Symposium on Multiscale Modeling of Mate-
rials and Devices (MMMD-2014), 2014 [4]) performed to assess the blast mitigation
effect of fluid-filled open-cell pin-core polymer foam is presented. Static compres-
sion tests were performed to study the quasi-static behavior of the pin-core latex
polymer foam. The Ogden hyper-elastic material parameters are determined from
the analysis of static compression test data using ADINA (automatic dynamic incre-
mental nonlinear analysis) v 9.2, ADINA R&D, Inc., Watertown [5]). Further, results
of the static compression tests on dry foam and water-saturated foam are compared
and discussed. In addition, the dynamic behavior of the pin-core latex cellular foam
is investigated by conducting drop tower impact experiment and numerical simu-
lation of the drop tower experiment using LS-DYNA (LS-DYNA v971, Livermore
Software Technology Corporation, Livermore [6]). The stress–time history from the
drop tower experiment is compared with that of simulation, and good correlation is
found between the two sets of results.
Keywords Blast protection · Fluid-filled foam · Blast waves · Polymer foam ·

High explosive · Near-field blast
K. Venkataramana (B) · V. Bhasin

Reactor Safety Division, BARC, Mumbai 400085, India
R. K. Singh · K. K. Vaze · H. S. Kushwaha
Raja Ramanna Research Fellow, DAE, BARC, Mumbai 400085, India
A. Deb
CPDM, Indian Institute of Science (IISc), Bengaluru 560012, India
https://doi.org/10.1007/978-981-15-3254-2_5
40 K. Venkataramana et al.
5.1 Introduction
Cellular polymer foams consist of a three-dimensional array of interconnected cells

with solid cell walls and cell faces. When the cell faces are open, they are called
open-celled foams, and when the cell faces are closed, they are called closed-cell
foams. The cellular polymer foams are characterized by their relative density defined
as (ρ ∗ /ρs ), where ρ ∗ is the overall density of the foam, and ρs is that of the solid
of which the foam is made. The relative density of foams varies from less than
0.3 to as low as 0.003. The porosity of the foam is defined as 1 − (ρ ∗ /ρs ). Low-
density open-cell polymer foams have porosity up to 0.97. The densification strain
for open-cell elastomeric polymer foams is given by [2] εD = 1 − 1.4(ρ ∗ /ρs ).
Flexible polymer foams such as open-cell latex polymer foam absorb energy by
elastic bending and buckling of cell walls and release the energy after the load is
removed. Though their use as energy absorbers under static and low-velocity impact
loads in automobile and packaging industry is well known, low-density open-cell
polymer foams have also drawn the attention of researchers for military applications
as energy absorbers under blast and impact loads [7–10]. Dawson [7] analyzed the
fluid-filled open-cell polymer foam for blast mitigation, and Venkataramana et al. [3,
4, 11] have experimentally and numerically investigated the application of open-cell
natural latex polymer foam filled with fluid for mitigating the close-range air blast
effects on structures. However, the material properties of the pin-core latex polymer
foam used in the blast experiments are not available in open literature.
Typical compositions and manufacturing methods of the pin-core natural latex
polymer foam are discussed by Rani (Chap. 9) in Ref. [12]. The natural latex pin-
core open-cell polymer foam used in the current experiments is manufactured by
MM foam Ltd. [13]. The foam is categorized as low-density elastomeric cellular
polymer foam with an overall density of 20 kg/m3 , relative density (ρ ∗ /ρs ) of 0.022
and 97.8% porosity. This paper presents results of static compression tests and drop
tower impact experiments performed to characterize the open-cell pin-core latex
polymer foam under static and dynamic loading conditions. In addition, numerical
simulations of the static compression tests and drop tower impact experiments are
performed, and the simulation results are compared with the experimental data.
5.2 Drop Tower Experiment on Foam
The drop tower experiments were performed on a miniature drop tower, which is
a small-scale drop tower designed to perform experiments on soft materials like
polymer foams. The drop tower experimental setup is shown in Fig. 5.1. Drop tower
tests were conducted on 100 mm thick foam block of base size 150 mm × 150 mm.
The drop weight is allowed to fall freely under gravity on the foam block from
a predetermined
√ height. The impact of the drop weight imparts an initial velocity
V0 = 2gh to the top surface of the foam block. The force–time history acting
5 Material Characterization of Pin-Core Latex Polymer … 41
Fig. 5.1 Experimental setup for drop tower experiment on foam block of 150 mm × 150 mm ×
100 mm
on the foam block is acquired by the data acquisition system attached to the drop
tower. The force–time history (Fig. 5.2) is converted into acceleration–time history
and then is integrated twice to get the displacement–time history of the foam block.
The displacement–time history is used to calculate the strain–time history developed
in the foam and finally the dynamic stress–strain curve is derived. The experimental
dynamic stress–strain response of the foam is shown in Fig. 5.3 which shows a
0.26 MPa stress at 19% strain.
Fig. 5.2 Force–time history, with time from the point of contact between the drop weight and the
foam
Fig. 5.3 Experimental dynamic stress–strain curve of foam block for up to 19% strain
5.2.1 Simulation of Drop Tower Experiment
Simulations of the drop tower experiments are performed with LS-DYNA [6] code.
The simulation setup is shown in Fig. 5.4. The foam block and the drop weight are
meshed with brick elements. The bottom of the foam block is fixed in all degrees of
freedom. The drop weight is modeled as an elastic steel plate. In order to simulate the
experimental conditions, the plate is given an
√ initial velocity corresponds to the veloc-
ity acquired by the free-falling mass, V = 2gh. The interaction of the plate with the
foam block is enforced by surface-to-surface contact algorithm in LS-DYNA. The
dynamic behavior of the foam is modeled by MAT_FU_CHANG_FOAM material
model (Mat_083) available in LS-DYNA. This material model is suitable to model
dynamic compressive behavior of low-density foams, and it allows the stress–strain
curves from drop tower experiments to be directly used as input to LS-DYNA material
model. The stress–strain values of the foam derived from the drop tower experiment
are given as input for the FU_CHANG_FOAM material model [14]. The LS-DYNA
program internally calculates the other material parameters using the input test data.
The force–time history from the drop tower experiment is converted into stress–time
history and compared with that of simulation shown in Fig. 5.5. It is seen that there
is a good agreement between the experimental and simulated stress–time history.
Fig. 5.4 Simulation of drop tower test using LS-DYNA

Fig. 5.5 Normal stress–time response of the foam block for 0.13 m height fall of drop weight of
13 kg
5.3 Static Compression Tests on Foam
The foam block used in the actual blast experiments is 100 mm in thickness. The foam
is made from natural latex using pin-core method of manufacturing. The pin-core
method introduces uniformly distributed cylindrical holes of approximately 5 mm
diameter in the foam block. Foam specimens of full thickness (100 mm thick) of foam
block used in the field air blast experiments are used in the static compression tests in
order to determine the properties of the material in conditions as close as possible to
the actual field conditions in which the foam is used. The test specimen is of square
base size 150 mm × 150 mm and 100 mm thickness. The specimen size is greater
than the minimum size (50 mm × 50 mm base and 25 mm thickness) recommended
by the ASTM standard D3574-11 [15] for compression testing of flexible polymer
foams. Figure 5.6 shows the natural latex polymer pin-core foam used in the field
Fig. 5.6 Natural latex

pin-core polymer foam used
in the field air blast
experiments [3, 11]
Fig. 5.7 Dry foam testing a foam at initial condition and b foam at 95% compression
Table 5.1 Summary of compression tests performed on dry foam

Test no. Compression stroke Crosshead speed Peak force (N) Original thickness
length (mm) (mm/s) of foam specimen
(mm)
1 95 10 2173 100
2 95 20 1980 100
3 94.2 40 1688 100
air blast experiments. The foam specimen during the compression test is shown in
Fig. 5.7.
The uniaxial compression tests were performed at various crosshead speeds.
Table 5.1 summarizes the compression tests performed on the dry foam. Figure 5.8
shows the compressive stress–strain response of the dry foam. There was no perma-
nent set observed in the compression test. The foam block completely recovered its
initial shape on removal of the load. The foam thus remained elastic during one static
uniaxial compression loading and unloading cycle.
As it can be seen from Table 5.1, as the crosshead speed is increased from 10 mm/s
to 40 mm/s, the peak load is decreased from 2173 to 1688 N and a 22% reduction in
peak-load-carrying capacity at 95% compression. This reduction of peak load may
be attributed to early crushing of cell walls at high crosshead speeds (high strain
rates). When the foam is loaded slowly, the foam cells get enough time to come
to equilibrium with the applied load till peak load is reached, whereas at higher
crosshead speeds, early crushing of the foam cell walls results in reduction in the
peak-load-carrying capacity of the foam.
Compression tests were also performed on foam fully saturated with water, and the
corresponding stress–strain response is compared with that of dry foam in Fig. 5.9.
It has been observed that water in the pores of the foam contributed to the load-
carrying capacity of the foam. Peak stress in the dry foam is observed to be 96,489,
Fig. 5.8 Experimental stress–strain curves of foam at crosshead speeds of 10, 20 and 40 mm/s
(corresponding strain rates are 0.1 s−1 , 0.2 s−1 , and 0.4 s−1 , respectively). The peak stress at
40 mm/s crosshead speed is less than the peak stress at 10 and 20 mm/s speed
Fig. 5.9 Comparison of experimental stress–strain curves of foam at crosshead speeds of 10 mm/s
(strain rates are 0.1 s−1 ), for dry foam and water-saturated foam
and 118,489 Pa in water-saturated foam, an increase of 22.8% with the presence of

water in the pores. This increase in strength is due to the additional work required to
expel the fluid (water) from the foam during its compression [4]. It is further observed
from Fig. 5.9 that the unloading curve is steeper for the water-saturated foam, with
the stress decreasing rapidly at almost constant strain.
This is due to the loss of elasticity of the foam in the presence of water during
the unloading cycle as the foam remained at maximum strain like a perfectly plastic
material until the load is almost completely removed. The sudden change in the
slope in the unloading curve shown in Fig. 5.9 is due to the sudden elastic recovery
of the foam at the end of unloading cycle, when the load is almost removed. The
compression tests with water-saturated foam were repeated, and they gave the same
stress–strain curve as shown in Fig. 5.9.
5.4 Simulation of Static Compression Tests
The static uniaxial compression tests on the dry foam are simulated using ADINA
[5] finite element software. The 150 mm × 150 mm × 100 mm3 test specimen
is represented by a plane strain model. The bottom support and the top plate are
treated as rigid surfaces. Frictionless contact between the foam and the supports
is assumed. The top rigid plate is given downward displacement corresponding to
the compression stroke during the uniaxial compression experiment. The problem is
solved in small time steps considering the material and geometric nonlinearities.
5.4.1 Material Model
The dry foam is modeled with Ogden hyper-elastic material model [5] with its param-
eters determined from the uniaxial compression experimental data. Ogden material
model is based on the following expression for strain energy density (i.e., strain
energy per unit volume) [16]:
9
μn αn
WD = λ1 + λα2 n + λα3 n − 3 N/m2 (5.1)
n=1
αn
where μn and αn are the Ogden material constants and λi ’s are the square roots of
the principal stretches of the Cauchy–Green deformation tensor. Choosing only μn ,
αn = 0, n = 1, 2, 3, the standard three-term Ogden material description is recovered
from the general model given by Eq. (5.1). The Ogden material model coefficients
are determined by least-squares fit between the test data and the strain energy density
function. The coefficients determined using ADINA for the latex polymer foam are
given in Table 5.2, and the curve fit to compression experimental data is shown in
Fig. 5.10.
Table 5.2 Ogden material

μ parameter Value (N/m2 ) α parameter Value
model parameters for the
natural latex pin-core cellular μ1 681.5047 α1 1.3
polymer foam μ2 11.45471 α2 6.3
μ3 43.09203 α3 −2.5
Stress (Pa) x 103
Strain m/m
Fig. 5.10 Curve fit to the latex polymer foam compression test data in ADINA
Using the Ogden material parameters from Table 5.2, the stress–strain curves of
the foam material from the UTM compression test and the simulation are shown in
Fig. 5.11 which shows good agreement between the two.
Fig. 5.11 Comparison of experimental and simulation stress–strain curves for compression test on
latex foam
5.5 Conclusions
The dynamic response of pin-core natural latex open-cell polymer foam is deter-
mined by conducting drop tower impact experiments on the foam. Simulations of
the drop tower experiments are performed with LS-DYNA, and good agreement is
obtained for the force–time history derived from the experiments and simulation.
Further, static uniaxial compression tests were carried out on the foam using univer-
sal testing machine (UTM). The compression test data were analyzed to determine
the Ogden hyper-elastic material model parameters for the foam. In addition, numer-
ical simulations of the uniaxial static compression tests on the foam were done using
ADINA code, and good agreement was found between the test data and simulation
results for the uniaxial compression stress–strain response. Furthermore, the effect
of the water on the uniaxial compressive behavior of water-saturated open-cell latex
polymer foam is demonstrated experimentally.
References
1. Ashby, M.F., Evans, A.G., Fleck, N.A., Gibson, L.J., Hutchinson, J.W., Wadley, H.N.G.: Metal
Foams: A Design Guide. Butterworth-Heinemann, Oxford (2000)
2. Gibson, L.J., Ashby, M.F.: Cellular Solids: Structure and Properties, 2nd edn. Cambridge
University Press, Cambridge (1999)
3. Venkataramana, K., Singh, R.K., Deb, A., Bhasin, V., Vaze, K.K., Kushwaha, H.S.: Blast protec-
tion of infrastructure with fluid-filled cellular polymer foam. In: 11th International Symposium
on Plasticity and Impact Mechanics (IMPLAST-2016), IIT New Delhi, 11–14 Dec 2016. Proc.
Eng. 173, 547–554 (2017)
4. Venkataramana, K., Singh, R.K., Deb, A., Bhasin, V., Vaze, K.K., Kushwaha, H.S.: Numerical
simulation of blast wave mitigation using foam impregnated with water. In: DAE BRNS Sym-
posium on Multiscale Modeling of Materials and Devices (MMMD-2014), BARC, Mumbai,
30 Oct–02 Nov 2014
5. ADINA (Automatic Dynamic Incremental Nonlinear Analysis) v 9.2, ADINA R&D, Inc.,
Watertown, MA 02472, USA
6. LS-DYNA v971, Livermore Software Technology Corporation, Livermore, CA, USA
7. Dawson, M.A.: Composite plates with a layer of fluid-filled, reticulated foam for blast protection
of infrastructure. Int. J. Impact Eng. 36(10–11), 1288–1295 (2009)
8. Yost, A.L.: Fluid-filled helmet liner concept for protection against blast-induced traumatic
brain injury. , Masters Thesis, Mechanical Engineering, Massachusetts Institute of Technology,
Cambridge, MA (2012)
9. Goel, R.: Study of an advanced helmet liner concept to reduce TBI: experiments and simula-
tion using sandwich structures. Masters thesis, Aeronautics and Astronautics, Massachusetts
Institute of Technology. Cambridge, MA (2011)
10. Christou, G.A.: Development of a helmet liner for protection against blast induced trauma. Mas-
ters thesis, Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge,
MA (2010)
11. Venkataramana, K.: Blast effects on mild steel plates and blast protection using fluid-filled
polymer foam. Ph.D. thesis, Indian Institute of Science (IISc), Bengaluru, May 2017
12. Rani, J.: Latex foam. In: Eves, D. (ed.) Handbook of Polymer Foams. Rapra Technol-
ogy Limited, Shawbury, Shrewsbury, Shropshire, UK (2004). http://www.rapra.net, ISBN:
1-85957-388-6
13. MM foam Ltd., Chennai, Tamil Nadu, India. www.mmfoam.com

14. LS-DYNA Keyword Reference Manual, Vol-2, Material Models, Livermore Software Tech-
nology Corporation (LSTC), Livermore, CA (2013)
15. ASTM-D3574-11: Standard test methods for flexible cellular materials—slab, bonded, and
molded urethane foams. Test C—compression force deflection test. ASTM International (2003)
16. ADINA Theory and Modeling Guide. vol. 1, ADINA Solids and Structures. ADINA R&D Inc.,
MA, USA
Chapter 6
Effect of Delamination on Static
Behaviour of Point Supported Composite
Conoidal Shells
Kamalika Das and Suman Pandey
Abstract Large column-free areas are a need of today’s industrial units. This need
can be fulfilled by thin laminated shell structures made of composite material.
Conoidal shell, being a doubly curved shell structure with a ruled surface, is compar-
atively easy to construct and is ideal for roofing purposes of such large areas. High
specific strength of composite materials and curved form of shells provide greater
stability to these shell roofs. The behaviour of these shell structures under loading
largely depends on support conditions. Now, separation of layers, known as delam-
ination, is considered to be a major critical defect of composite laminates, which
occurs due to repeated cyclic stresses, impact and manufacturing defects. Delamina-
tion causes remarkable loss in mechanical toughness and ultimate failure. The most
notable thing regarding delamination is that, in most cases, it is not visible from
outside. Hence, this type of damage needs a wider attention from researchers. In this
paper, the composite conoidal shells with delamination under static loading have
been studied. A finite element formulation has been developed with Fortran code to
observe the change in deflection in a conoidal shell with varying number of support
conditions. An in-depth analysis is done based on number of supports, laminations,
stacking sequences and percentage of damage; thereafter, some helpful observations
have been highlighted so as to narrate the characteristics of this shell structure as a
roofing unit.
Keywords Composites · Conoidal shell · Deflection · Delamination · Finite

element method · Thin shell
K. Das (B) · S. Pandey

Civil Engineering Department, Techno India University, Kolkata, India
S. Pandey

https://doi.org/10.1007/978-981-15-3254-2_6
52 K. Das and S. Pandey
Notations
μ12 Poisson’s ratio

E 11 , E 22 Elastic moduli
G12 , G13 , G23 Shear moduli of a lamina with respect to 1, 2 and 3 axes of fibre
hh, hl Higher and lower heights of conoid, respectively
6.1 Introduction
Recent developments in manufacturing methods of composite materials have resulted

in easy fabrication of composite shell structures of complex shell shapes: synclastic,
anticlastic or a combination of both. Due to their high specific strength, composite
materials result in low cost of supporting members and foundation.
With the increase in use of composite materials in all advanced engineering struc-
tures, the investigation on characteristics of damaged laminated composite shell
structures is important so that one may judge the performance and stability of the
structure. Delamination is a major type of damage in laminated shells, which, if
unchecked, can result in reduction of strength, leading to sudden buckling and possi-
ble failure of the structure. Singly ruled surfaces, such as conoidal shells, are a popular
choice for roofing purposes in civil engineering structures, due to their simplicity and
elegance. Hence, problems associated with laminated conoidal shells require more
attention from researchers and scientists, and separation of layers is one among the
major problems.
Dey et al. [1] investigated the behaviour of composite shells during bending using
finite element formulation. The shape of the shells used for the study is paraboloid
of revolution. Gim [2] studied the transverse shear deformation on delaminated dou-
ble cantilevered plates by developing a plate finite element formulation based on a
lamination theory. Bolotin [3] discussed the research on delamination and related
crack-like defects of laminated and fibre composites. Parhi et al. [4] analysed lam-
inated composite plates for the first ply failure. The plates used for the study have
random delamination located at multiple places. Kumari et al. [5] studied the static
behaviour of damaged conoidal shells of graphite epoxy composite material by con-
sidering twelve different laminations, which include angle-ply and cross-ply shells
with uniformly distributed loading. Nardi et al. [6] worked on sparse sensing detection
of delaminated carbon-fibre-reinforced-plastic (CFRP) plates. Nikolayevich et al. [7]
studied the recent increase in interest in thin shell structures for civil constructions.
They explicitly stated the opportunities of the application of composite materials
in construction of thin shell structures for civil engineering applications and indus-
trial purposes and exemplified the same in their paper. Ismail et al. [8] worked on
the buckling behaviour of damaged composite plate with woven reinforcement and
evaluated the critical buckling load.
6 Effect of Delamination on Static Behaviour … 53
The above literature review suggests that the use of composite shell structures has
been steadily rising since the last two decades, and even if the technology to detect
delamination inside the laminations of a shell structure exists, a lot more remains to
be known about the behaviour of conoidal shells with delamination damage. Hence,
the purpose of this paper is to seek more information and insight on how damaged
composite conoidal shells deflect with changes in number of supports, lamination
layers, orientation of fibres and delamination area.
6.2 Mathematical Equations
An eight-noded curved quadratic isoparametric finite element with five degrees of

freedom, u, v, w, α, β at each node is employed for solution [5]. Here, the displacement
field can be written as:
u(x, y, z) = u 0 (x, y) − zα(x, y)

v(x, y, z) = v0 (x, y) − zβ(x, y)
w(x, y, z) = w0 (x, y)
where u0 , v0 , w0 are the displacement components along the x-, y-, z-coordinate
directions, respectively, and α, β are rotational displacements about y- and x-axes,
respectively, of a point on the mid-plane, i.e. z = 0 [9].
Hence, the static equilibrium problem can be written as:
[K ]{δ} = {P} (6.1)
In the above equation,

[K] global stiffness matrix
{P} global load vector
{δ} global displacement vectors.
6.2.1 Constitutive Relations for Undelaminated Shells
Since the shell consists of a number of laminae, the total stress resultants of an n-layer
laminate can be represented by Kumari et al. [5]:
{F} = [D]{ε} (6.2)
In the above equation,

[D] laminate stiffness matrix
{F} generalised force vectors

{ε} represents the in-plane strain vectors.
6.2.2 Constitutive Relations for Delaminated Shells
Figure 6.1 represents cross-sectional geometry of a laminated shell having ‘n’ number
of layers and ‘p’ number of delamination.
Figure 6.2 represents cross-sectional details of the junction, marked as ‘o’, of
undelaminated part (segment 1) and delaminated part (segment 2 and 3) of a laminated
shell. In-plane strain and curvature can be related by the following equation for
undamaged segments of a thin delaminated conoidal shell:

{ε} L = ε0 1 + z 0L {k}1 (6.3)
Fig. 6.1 Cross-sectional geometry of laminated shell showing multiple delaminations [5]
Fig. 6.2 Cross-sectional details of undelaminated and delaminated junction [5]

Table 6.1 Values of maximum non-dimensional downward deflection (×10−4 ) for different
laminations and SSSS boundary condition
Lamination Das and Chakraborty [10] Present approach
0°/90° −5.63 −5.629
0°/90°/0° −4.70 −4.701
45°/−45° −3.12 −3.121
45°/−45°/45° −2.41 −2.413
Note E 11 = 25 E 22 , G12 = G13 = 0.5 E 22 , G23 = 0.2 E 22 , μ12 = 0.25, hl = hh = 0.25, a = b,
c = d, h = 0.01
where suffix ‘L’ is for delaminated portion, suffix ‘1’ is for undelaminated portion,
and{ε0 } represents the vectors of in-plane strain present at the mid-surface, {k}
are curvature vectors and z 0L is the distance along transverse direction between the
mid-surface of delaminated and undelaminated portions, respectively.
Consequently, Eq. 6.2 becomes

{F} L = [D] L ε∗ 1 (6.4)
where {ε∗ } are the generalised strain vectors.
6.3 Benchmark Problems
For validating the accuracy of present computer code and formulation, two
benchmark problems have been solved.
The results obtained from the first benchmark problem, Das and Chakraborty [10],
and present formulation are, to a great extent, similar. This shows the correctness of
the conoidal shell formulation.
Again, output of present computer code and that of the second benchmark prob-
lem, Acharyya et al. [11], are in close agreement. For this, Rxy (twist radius of
curvature) has been assigned a very large value and Ry (radius of shell surface along
y-direction) is kept constant by authors. Hence, exactness of delaminated composite
shell is also confirmed (Tables 6.1 and 6.2).
6.4 Results and Discussions
The delaminated area (c × d) is centrally located in the square-shaped (a/b = 1)

composite conoidal shell. The ratio c/d is unity, and the extent of delamination has
been expressed with the ratio c/a (e.g. see Fig. 6.4). Here, ‘a’ and ‘b’ are length
Table 6.2 Maximum transverse non-dimensional deflections (×10−4 ) of corner-supported

delaminated cylindrical shells with different laminations
Lamination c/a Acharyya et al. [11] Present approach
0°/90°/0°/90° 0.25 59.379 59.3789
0.5 71.736 71.7361
45°/−45°/45°/−45° 0.25 26.011 26.0106
0.5 30.371 30.3709
c = d, h = 0.01
and width of shell in plan, respectively, and ‘c’ and ‘d’ are length and width of
delaminated area in plan, respectively (Fig. 6.3).
The purpose of this study is to analyse the deflection of damaged laminated
conoidal shells of composite material subjected to uniformly distributed load, for
which the following parametric variations have been considered:
1. The extent of delamination, i.e. c/a is variegated from 0 to 56.25% with respect
to the total plan area.
Fig. 6.3 A typical discretization of 8 × 8 mesh on plan area with element and node numbers (c/a =
0) [12]
Fig. 6.4 An example of c/a = 0.25, i.e. 6.25% delamination (delamination area is shown by the
shaded portion)
2. Twelve separate laminations are studied which includes angle-ply and cross-ply
of both symmetric and anti-symmetric sequences.
3. Three separate point-support conditions are taken (shown by black dots), where
Case 1 (see Fig. 6.5) has four, Case 2 (see Fig. 6.6) has eight and Case 3 (see
Fig. 6.7) has sixteen supports, respectively.
Fig. 6.5 Case 1

Fig. 6.6 Case 2
Fig. 6.7 Case 3
Case 1: The results of Table 6.3 indicate that:

• In most of the cases, there is an increase in maximum transverse deflection with
increase in delamination area.
• The maximum deflection decreases with increase in number of lamination layers
for each case of damage and in most of the cases for both symmetric and anti-
symmetric laminations. For example, 45°/−45° and 45°/−45°/45°/−45° for c/a
= 0.25, the maximum deflections are 30.4 × 10−4 and 15.5 × 10−4 , respectively.
• For the same number of laminae, the maximum deflections for each case of delami-
nation are lower in symmetric laminations than that for anti-symmetric laminations
for cross-ply shells, whereas this is reverse for angle-plies.
Case 2: The results of Table 6.4 indicate that the:
• Maximum transverse deflection is observed to increase with increase in the
delamination area in most cases.
Table 6.3 Maximum non-dimensional deflections (×10−4 ) and the corresponding nodes for each
lamination (in parentheses below corresponding deflections) of Case 1
Stacking sequence c/a Maximum deflection
Case 1
0 0.25 0.5 0.75
0°/90° 40.61 40.69 40.77 40.57
[70] [70] [70] [78]
45°/−45° 30.07 30.4 31.9 32.4
[78] [78] [78] [78]
0°/90°/0° 25.1 26.1 29.5 36.2
[53] [53] [69] [69]
45°/−45°/45° 31.4 32.6 36.7 25.2
[78] [78] [95] [83]
0°/90°/0°/90° 24.8 26.5 30.6 37.2
[78] [78] [78] [78]
45°/−45°/45°/−45° 14.9 15.5 17.4 22.9
[78] [78] [78] [78]
0°/90°/90°/0° 22.8 23.8 26.9 32.1
[69] [69] [78] [69]
45°/−45°/−45°/45° 20.3 21.2 25.3 32.2
[78] [78] [95] [104]
0°/90°/0°/90°/0°/90° 23 24.1 26.52 29.8
[70] [70] [70] [70]
45°/−45°/45°/−45°/45°/−45° 13.4 13.8 15.1 17.9
[78] [78] [78] [78]
0°/90°/0°/0°/90°/0° 22.3 23.5 26.51 31.2
[78] [78] [78] [78]
45°/−45°/45°/45°/−45°/45° 15.5 16 17.8 22.1
[78] [78] [78] [95]
c = d, h = 0.01
• Maximum transverse deflection decreases as compared to Case 1 for almost all

the cases.
• For cross-ply laminations, the maximum deflections increase from the 0°/90° to
the 0°/90°/0° laminations, then decrease with further increase in number of lam-
inations. For angle-ply laminations, the maximum deflections increase from the
45°/−45° to the 45°/−45°/45° laminations, then decrease with further increase in
number of laminations.
• In general, the maximum deflections for each case of delamination are lower in
anti-symmetric laminations.
• The overall performance of angle-ply shells is better than that of the cross-ply ones
for the same number of laminae.
Case 3: The results of Table 6.5 indicate that:
Case 2
0 0.25 0.5 0.75
0°/90° 3.31 3.47 3.56 3.53
[52] [74] [74] [74]
45°/−45° 3.99 4.03 4.13 4.5
[14] [14] [14] [14]
0°/90°/0° 5.13 5.15 5.27 5.82
[43] [43] [43] [43]
45°/−45°/45° 4.23 4.27 4.37 13.4
[52] [52] [52] [57]
0°/90°/0°/90° 2.47 2.49 3.06 3.88
[41] [40] [74] [87]
45°/−45°/45°/−45° 2.45 2.46 2.51 3.26
[14] [14] [14] [40]
0°/90°/90°/0° 3.87 3.88 4.04 4.22
[44] [44] [44] [27]
45°/−45°/−45°/45° 2.9 2.94 3.16 16.8
[4] [4] [4] [84]
0°/90°/0°/90°/0°/90° 2.41 2.42 2.63 2.96
[41] [40] [74] [74]
45°/−45°/45°/−45°/45°/−45° 2.3 2.31 2.36 2.7
[14] [14] [14] [40]
0°/90°/0°/0°/90°/0° 3.18 3.16 3.14 3.25
[44] [44] [44] [44]
45°/−45°/45°/45°/−45°/45° 2.65 2.66 2.72 3.23
[4] [4] [4] [31]
c = d, h = 0.01
• In most of the cases, there is an increase in maximum transverse deflection with

increase in delamination area.
• Maximum transverse deflection decreases further, from Case 1 and Case 2, as
number of supports increases.
• Maximum deflection decreases with increase in lamination layers for most cases.
For example, with c/a = 0.5, as the number of laminae increases from 0°/90°/0°/90°
to 0°/90°/0°/90°/0°/90°, maximum deflection reduces from 0.953 × 10−4 to 0.902
× 10−4 .
• Maximum deflections are lower in anti-symmetric laminations for most cases.
• There is no noticeable trend in two- and three-layered laminations, but with
increase in lamination layers, the maximum deflections are lower in cross-ply
Case 3
0 0.25 0.5 0.75
0°/90° 1.17 1.29 1.3 1.1
[48] [87] [61] [48]
45°/−45° 1.26 1.268 1.29 1.35
[48] [62] [62] [48]
0°/90°/0° 1.13 1.14 1.12 3.59
[27] [27] [27] [83]
45°/−45°/45° 1.04 1.06 1.21 1.44
[60] [60] [60] [60]
0°/90°/0°/90° 0.879 0.902 0.953 0.113
[61] [74] [74] [48]
45°/−45°/45°/−45° 0.914 0.935 1.05 1.29
[61] [61] [74] [48]
0°/90°/90°/0° 0.908 1.01 11.6 2.07
[48] [87] [139] [35]
45°/−45°/−45°/45° 0.922 0.945 1.18 1.6
[60] [60] [60] [59]
0°/90°/0°/90°/0°/90° 0.846 0.88 0.902 0.914
[61] [74] [74] [48]
45°/−45°/45°/−45°/45°/−45° 0.881 0.895 0.957 1.11
[61] [61] [61] [48]
0°/90°/0°/0°/90°/0° 0.779 0.781 0.785 0.976
[48] [48] [48] [48]
45°/−45°/45°/45°/−45°/45° 0.937 0.953 1.04 1.23
[60] [60] [60] [60]
c = d, h = 0.01
lamination for most cases. For example, in six-layered anti-symmetric lamina-

tions, the maximum deflections for c/a = 0.5 are 0.902 × 10−4 for cross-ply and
0.957 × 10−4 for angle-ply laminations.
6.5 Conclusion
In the present work, an eight-noded isoparametric finite element is used to generate a

software system. This system is used to investigate the static behaviour of composite
conoidal shells with delamination. Following conclusions can be drawn based on the
present study:
1. Results obtained using present system are nearly matching with the two bench-
mark problems. This confirms the usefulness of the software system used
here.
2. The above research on deflection of composite conoidal shells concludes that
maximum deflection decreases as the number of supports and number of
laminates increases.
3. It is noted that maximum deflection increases as the extent of delamination
increases from 0 to 56.25%. However, as the number of supports increases, maxi-
mum deflection in each case of delamination decreases (except for 0°/90°/90°/0°
lamination of Case 3 with delamination area c/a = 0.5).
4. It is also observed that in cases with four supports, the maximum deflection is
lower in symmetric laminations for cross-ply laminates, whereas it is lower in
anti-symmetric laminations for angle-ply laminates. With the increase in num-
ber of supports, the conoidal shell shows lower maximum deflection in the
anti-symmetric sequences for angle-ply and cross-ply shells as well, with the
exception of the six-layered cross-ply laminates for Case 3.
5. The best performing laminate for both Case 1 and Case 2 has been observed to
be 45°/−45°/45°/−45°/45°/−45°, and for Case 3, it is 0°/90°/0°/0°/90°/0°.
References
1. Dey, A., Bandyopadhyay, J.N., Sinha, P.K.: Finite element analysis of laminated composite
conoidal shell structures. Comput. Struct. 43(3), 469–476 (1992)
2. Gim, C.K.: Plate finite element modeling of laminated plates. Comput. Struct. 52(1), 157–168
(1994)
3. Bolotin, V.V.: Delaminations in composite structures: its origin, buckling, growth and stability.
Compos. B Eng. 27(2), 129–145 (1996)
4. Parhi, P.K., Bhattacharyya, S.K., Sinha, P.K.: Failure analysis of multiple delaminated
composite plates due to bending and impact. Bull. Mater. Sci. 24(2), 143–149 (2001)
5. Kumari, S., Chakravorty, D.: Bending of delaminated composite conoidal shells under
uniformly distributed load. J. Eng. Mech. 137(10), 660–668 (2011)
6. Gaudenzi, P., Nardi, D., Chiappetta, I., Atek, S., Lampani, L., Pasquali, M., Sarasini, F., Tir-
illó, J., Valente, T.: Sparse sensing detection of impact-induced delaminations in composite
laminates. Compos. Struct. 133, 1209–1219 (2015)
7. Nikolayevich, K.S., Lvovna, S.S., Bock, H.C.A.: Thin-walled composite and plastic shells for
civil and industrial buildings and erections. Mater. Sci. Forum 895, 45–51 (2017)
8. Ismail, M.R., Ali, Z.A.A.A., Al-Waily, M.: Delamination damage effect on buckling behavior
of woven reinforcement composite materials plate. Int. J. Mechanical & Mech. Eng. 18(5),
83–93 (2018)
9. Reddy, J.N.: Mechanics of Laminated Composite Plates—Theory and Analysis, 2nd edn. CRC
Press, Boca Raton (2004)
10. Das, H.S., Chakraborty, D.: Design aids and selection guidelines for composite conoidal shell
roofs—a finite element application. J. Reinf. Plast. Compos. 26(17), 1793–1819 (2007)
11. Acharyya, A.K., Chakravorty, D., Karmakar, A.: Bending characteristics of delaminated com-
posite cylindrical shells—a finite element approach. J. Reinf. Plast. Compos. 28(8), 965–978
(2009)
12. Kumari, S.: Finite Element Bending and Vibration Characteristics of Delaminated Composite
Conoidal Shell Roofs. PhD Thesis, Jadavpur University, West Bengal (2014)
Chapter 7
Wind Excitation Study
of a Corner-Modified Square Tall
Structure
Arghyadip Das and Sujit Kumar Dalui
Abstract The modern buildings are becoming taller due to lack of land space and
this makes the tall buildings more sensitive to lateral loads such as wind. The outer
shape of the building is one of the main parameters which affects the design wind
loads. Various types of minor corner modifications on tall buildings result a huge
change in force and pressure. The present study is carried out to find the effect of
corner modification on square plan-shaped tall building. The corners are modified as
corner recessed, corner chamfered, and corner rounded, and a series of simulation
is done in ANSYS CFX (k − ε model) to find out the effects on force coefficients,
pressure coefficients, wind flow pattern around the buildings, etc. The numerical
analyses are done considering the model scale as 1:300 and for 0° angle of wind
attack.
Keywords Tall buildings · Pressure coefficients · Force coefficients · Domain ·

Boundary layer · Meshing · Wind attack
7.1 Introduction
The population of the world is increasing drastically, and to accommodate this pop-
ulation, the modern-day technology prefers to build the high-rise buildings due to
lack of land space. So, these tall buildings are very much sensitive in lateral forces
such as wind. The outer shape is one of the main parameters which affects the wind
loads and responses. Various International Standards like IS:875 (part-3):2015 [7],
ASCE-16 [2], and AS-NZS: 1180-2011 [1] are providing guidelines to calculate
the wind-induced loads and responses for regular plan-shaped buildings but there
are no such guidelines for the irregular and unconventional plan-shaped buildings.
A. Das (B) · S. K. Dalui

S. K. Dalui

https://doi.org/10.1007/978-981-15-3254-2_7
66 A. Das and S. K. Dalui
Many research works are done on various irregular plan-shaped buildings to find
the wind-induced loads and responses. Tanaka et al. [16] provide some guidelines
on aerodynamic forces and wind pressure on various unconventional configurations
like corner cut, corner chamfered, tilted, tapered, helical, and cross-opening with
basic square-shaped model by a series of wind tunnel investigation. Charkraborty
et al. [3] investigated the ‘+’-shaped tall building for 0°–45° wind incidence angle.
Gomes et al. [6] investigated on experimental outcomes of ‘L’- and ‘U’-shaped mod-
els with 1:100 length scale. Tse et al. [17] carried out his research to find out the
wind loadings and wind-induced responses of square tall buildings with different
sizes of chamfered and recessed corners. Li et al. [11] suggested that among the
horizontal modifications, 10% corner cut model is most greatly reduced along wind
load for specific wind direction. Verma et al. [18] investigated the variation of the
pressure distribution of tall square plan-shaped building for various wind incidence
angle. Kumar and Dalui [9] compare the pressure coefficients and force coefficients
for regular and angular cross-plan-shaped buildings. Pal and Dalui [12] studied the
pressure and force coefficients for ‘Z’ plan-shaped tall buildings. Sanyal and Dalui
[14] studied the variation of pressure on courtyard and opening of a rectangular plan-
shaped building. Elshaer et al. [4] studied the building corner modification effects
on square-shaped tall building and concluded that the corner round model has less
drag coefficients among all models. Kwok et al. [10] studied on finned and slotted
finned corner buildings and concluded that the fins and slotted fins increase the along
wind responses and reduce the across wind responses. Kawai [8] investigated on
square sections with rounded, chamfered, and recessed corners and concluded that
the small chamfers and recessions are effective in preventing aeroelastic instability
but rounded corners increase the aerodynamic damping. Tamura et al. [15] carried out
his research on square sections with rounded and chamfered corners using smooth
uniform flows and concluded about the reliability of CFD in predicting the wind
loads.
7.2 Scope of the Work
In this study, the square model is considered as the basic model with cross
section 250 mm × 250 mm and 750 mm height with length scale 1:300 shown
in Fig. 7.1a. The corner recessing is done in the next model with 25 mm recess that is
10% corner recess and shown in Fig. 7.1b, and the corner rounding and corner cham-
fering are also done successively with 25 mm rounding radius and 25 mm chamfering
distance and shown in Fig. 7.1c, d. The isometric view of square model is shown in
Fig. 7.1e. All the models are prepared using ANSYS CFX software package. The
wind incidence angle is considered as 0° for each case. Pressure distribution is eval-
uated for each model, and pressure coefficients are represented in tabulated form as
well as graphical form.
7 Wind Excitation Study of a Corner-Modified Square Tall Structure 67
Fig. 7.1 Model details of various corner modifications. a Square model, b corner recessed model,
c corner rounded model, d Corner chamfered model, and e isometric view of square model
7.3 Computational Domain, Boundary Condition

and Meshing
The computational domain has some specific sizes as per the guidelines of Frank
et al. [5] and Revuz et al. [13]. As per the guidelines, the upstream, downstream,
sidewall, and top wall clearance is taken as 5H, 15H, 5H, and 6H successively, where
H is the height of the mode. The details of domain size are shown in Fig. 7.2a, b.
The velocity of the wind is considered as 10 m/s, and the turbulence intensity is
taken as 1% (low intensity). The relative pressure at outlet is taken as zero. Sidewalls
of the domain boundary are considered as the free slip walls, and the model walls
are considered as the no slip walls. The ground roughness (α) is taken as 0.133.
Overall temperature of the domain is considered as 25 °C. The numerical analysis
Fig. 7.2 a Plan and b elevation view of computational domain used in CFD
Fig. 7.3 Comparison of a velocity profile and b turbulent profile
data are compared with previously published data of same type of geometric model,
and power law is also used to generate such profile with exponent coefficient as
0.133.
α
U Z
= (7.1)
U0 Z0
where U 0 is the basic wind speed taken as 10 m/s and Z 0 is boundary layer height
of 1 m. A similar type of velocity profile was used by Chakraborty et al [3]. A
comparison of the velocity profile and turbulent profile is shown in Fig. 7.3a, b.
Tetrahedron meshing is used throughout the whole domain, and finer meshing is
provided near the building to measure the accurate responses on the surface of the
building. Comparatively coarser meshing is provided in the outer edges of the domain.
The overall y+ values for all models are kept within the range of 30–300. A typical
meshing of the corner chamfered model is shown in Fig. 7.4.
The average pressure coefficients and force coefficients of corner-modified models

for various faces are calculated for 0° angle of wind attack and shown in Table 7.1.
The average Reynolds number is calculated as 3.72 × 106. The basic square model
has maximum positive pressure in windward face (face A) but side faces (face B
Fig. 7.4 Mesh pattern of corner chamfered model
Table 7.1 Average pressure and force coefficients for all types of model
Sl. No. Corner modification Mean pressure coefficient Force coefficient
A B C D
1 Basic square model 0.83 −0.58 −0.43 −0.58 1.20
2 Corner recessed model 0.82 −0.55 −0.40 −0.55 0.98
3 Corner rounded model 0.82 −0.62 −0.38 −0.62 0.70
4 Corner chamfered 0.80 −0.64 −0.37 −0.64 0.71
model
and face D) and leeward face (face C) are experiencing negative pressure at the same
time, whereas the corner recessed model is experiencing a bit lower positive pressure
compared to the square model. Corner rounded and corner chamfered models are
subjected to lesser pressure in face A, but corner chamfered model is experiencing
the maximum negative pressure in face D (leeward face). Among all four models, the
square model has maximum force coefficient as 1.20 and corner chamfered model
has minimum force coefficient as 0.70 along ‘X’-direction.
7.4.1 Variation of Pressure Coefficients Along the Vertical

Centerline, Flow Pattern and Pressure Contour
of Various Faces of the Building
The pressure coefficients (C p ) along vertical centerline of all faces of the building
models are evaluated and plotted in graph shown in Fig. 7.5. In face A, it is found
that the maximum pressure coefficient is at 600 mm height. The square model is
showing less C p value at 600 mm height compared to corner recessed and corner
chamfered models for face A. In the comparison of face B, the corner recessed model
Fig. 7.5 Comparison of avg. C p values along vertical centerline for various faces of all type of
buildings
is showing maximum suction at 450 mm height, whereas the corner recessed and
corner chamfered models have maximum suction at 600 mm height. Face B and
face D are showing maximum negative pressure along the vertical centerline. The
ratio of height of any point from the base (H) with respect to the overall height of
the building (H o ) is depicted in the Y-axis as the ratio of (H/H o ). Due to symmetry
of the building models and 0° angle of wind attack, the pressure coefficients (C p )
along vertical centerline are similar for sidewalls. The wind flow pattern around the
building models is shown in Fig. 7.6. The vortex generated in the leeward side of the
building models indicates the generation of negative pressure. Due to symmetry of
the building models with respect to both axes and 0° wind angle, the vortex formation
is also symmetric in the wake region of the buildings. Large separation of flow is
observed for corner rounded and corner chamfered building models. The side view
of vortex generation around the corner rounded building is also shown in Fig. 7.6.
The comparison of pressure contour is shown in Fig. 7.7.
(a) Square model (b) Corner recessed model
(c) Corner rounded model (d) Corner chamfered model
(e) Side view of vortex generation around corner rounded model
Fig. 7.6 Flow pattern around all type of building model at 0° angle of wind attack
7.5 Conclusion
The corner modification has a huge impact in reducing the force and pressure coef-
ficients on the building boundaries. Force and pressure coefficients are the most
important parameters of building design, and reduction of those parameters is very
much appreciated in the design point of view. The force coefficients are reduced
around 18% in the case of corner recessed model as compared to the square basic
model. But the best suited model is corner rounded model as it gives around 41%
reduction in force coefficients along ‘X’-direction for 0° angle of wind attack. Corner
chamfered model also gives satisfactory results (reduction of mean drag coefficient
around 40%). Due to some complicacy in construction works for corner rounded
Fig. 7.7 Comparison of pressure contour for face A for a square model, b corner recessed model,
c corner rounded model, and d corner chamfered model
model, the corner chamfered model is the preferred model for reduction of wind-
induced loads and responses. As compared to other minor modifications in building
corners, chamfered model gives maximum utilization of plan area.
References
1. AS/NZS: 1170.2:2011, Structure design actions, Part-2: wind actions. Australian/ New Zealand
Standard; Sydney and Wellington
2. ASCE: 7-16, Minimum design loads for buildings and other structures. Structural Engineering
Institute of the American Society of Civil Engineering, Reston (2016)
3. Chakraborty, S., Dalui, S.K., Ahuja, A.K.: Wind load on irregular plan shaped tall building—A
case study. Wind Struct. 19(1), 59–73 (2014)
4. Elshaer, A., Bitsuamlak, G.T., El Damatty, A.: Wind load reductions due to building corner
modifications. In: 22nd Annual Conference of the CFD Society of Canada Toronto, Canada
(2014)
5. Franke, J., Hirsch, C., Jensen, A.G., Krűs, H.W., Schatzmann, M., Westbury, P.S., Miles,
S.D.,Wisse,J.A., Wright, N.G.: Recommendations on the use of CFD in wind engineering.
In: International Conference on Urban Wind Engineering and Building Aerodynamics:COST
C14: Impact of Wind and Storm on City life and Built Environment (2004)
6. Gomes, M., Rodrigues, A., Mendes, P.: Experimental and numerical study of wind pressures
on irregular-plan shapes. J. Wind Eng. Ind. Aerodyn. 93(10), 741–756 (2005)
7. IS: 875(Part-III):2015, Indian standard code of practice for design loads (other than earthquake)
for buildings and structures, Part 3 (wind loads). Bureau of Indian Standards, New Delhi, India
(2015)
8. Kawai, H.: Effects of corner modifications on aeroelastic instabilities of tall buildings. J. Wind
Eng. Ind. Aerodyn. 74–76, 719–729 (1998)
9. Kumar, D., Dalui, S.K.: Effect of internal angles between limbs of cross plan shaped tall
building under wind load. Wind Struct. 24(2), 95–118 (2017)
10. Kwok, K.C.S., Wilhelm, P.A., Wilkie, B.G.: Effect of edge configuration on wind induced
response of tall buildings. Eng. Struct. 10, 135–140 (2018)
11. Li, Y., Tian, X., Tee, K.F., Li, Q.S., Li, Y.G.: Aerodynamic treatments for reduction of wind
loads on High-rise Building. J. Wind Eng. Ind. Aerodyn. 172, 107–115 (2018)
12. Pal, R., Dalui, S.K.: Wind effects on ‘Z’ plan shaped tall building: a case study. Int. J. Adv.
Struct. Eng. 8(3), 319–335 (2016)
13. Revuz, J., Hargreaves, D.M., and Owen J. S.: On the domain size of steady state CFD modelling
of tall building. Wind Struct. 15(4), 313–329 (2012)
14. Sanyal, P., Dalui, S.K.: Effect of courtyard and opening on a rectangular plan shaped tall
building under wind load. Int. J. Adv. Struct. Eng. 10(2), 169–188 (2018)
15. Tamura, T., Miyagi, T., Kitagishi, T.: Numerical prediction of unsteady pressures on a square
cylinder with various corner shapes. J. Wind Eng. Ind. Aerodyn. 74–76(1), 531–542 (1998)
16. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M., Kim, Y.C.: Experimental investigation of
aerodynamic forces and wind pressures acting on tall buildings with various unconventional
configurations. J. Wind Eng. Ind. Aerodyn. 107–108, 179–191 (2012) (Elsevier)
17. Tse, K.T., Hitchcock, P.A., Kwok, K.C.S., Thepmongkorn, S., Chan, C.M.: Economic perspec-
tives of aerodynamic treatments of square tall buildings. J. Wind Eng. Ind. Aerodyn. 97(9),
455–467 (2009)
18. Verma, S.K., Ahuja, A.K., Pandey, A.D.: Effects of wind incidence angle on wind pressure
distribution on square pan tall buildings. J. Acad. Ind. Res. 1(12), 747–752 (2013)
Chapter 8
Experimental Investigations on SCC
with Indigenously Developed
Geopolymer Aggregates
M. P. Naveena, G. Narayana, Vijayalakshmi Akella and C. M. Thejas
Abstract The excessive use of natural aggregates is having an adverse effect on

the environment and raises major environmental issues; hence, an attempt has been
made to develop artificial aggregates. These artificial aggregates are made from using
GGBS, sodium hydroxide and sodium silicate. This paper focuses on the fresh, hard-
ened and durability properties of self-compacting concrete with artificial geopolymer
aggregates (GPAs) as partial to full replacement of natural aggregates. Mix design
was carried out with varying content of GPA (0–100%) and cements contents (300,
350 and 400 kg/m3 ). Flow, passing ability and segregation resistance were found
to increase when 400 kg of cement and 100% replacement of 10 mm GPA is used
as compared to crushed gravel. However, the compressive strength was found to be
marginally less with the increase in aggregate size. After examining, the properties of
SCC with geopolymer aggregate of size varying from 10 to 20 mm are used in casting
of the sleepers. The sleepers were precast in Malu sleeper yard, Birur, Karnataka, and
subjected to bending, electrical resistivity and durability tests. The results from static
bending test showed that the resistance load of a sleeper with 50% of 6 M aggregate
is 238 and 230 kN for conventional concrete. Electrical resistivity and durability of
a sleeper with 6 M was found better than conventional concrete sleeper. The predic-
tion of self-compacting concrete properties with artificial geopolymer aggregate and
natural aggregate was investigated by using MATLAB.
M. P. Naveena (B) · G. Narayana

Department of Civil Engineering, S.J.C Institute of Technology, Chikkaballapur, Karnataka
562101, India
G. Narayana
M. P. Naveena · V. Akella · C. M. Thejas
Department of Civil Engineering, K.S.School of Engineering and Management, Bangalore,
Karnataka 560109, India
C. M. Thejas

https://doi.org/10.1007/978-981-15-3254-2_8
76 M. P. Naveena et al.
Keywords Geopolymer aggregate · Railway sleeper · Durability · Electrical

resistivity · RCPT · Bond strength
8.1 Introduction
Natural river sand and coarse aggregate have caused rapid depletion of this nat-
ural source due to widely used in the production of cement mortar and concrete
[1]. According to Freedonia, World Construction Aggregates Report, the global
demand for construction aggregates exceeded 48.3 billion tons in the year 2015
and is expected to grow 5.2% annually. In India, the consumption of aggregate was
about 2.2 billion metric tons in 2010, and further, the demand is going to be more
than 5 billion metric tons by 2020 [2].
Since aggregate is the main occupants of concrete (about 65–75% of total concrete
volume), type of aggregates have a significant influence on the fresh and hardened
property of high-strength concrete [3]. The replacement of cement by GGBS not only
increases the compressive strength but also reduces the cement content which even-
tually leads to the decreases in emission of CO2 . Current standards allow only 50%
of total binder content GGBS to be used in the production of concrete. The remain-
ing GGBS goes unused and disposed as landfills [4]. Geopolymer is ceramic-type
inorganic polymers produced at low temperature, generally below 100 °C. The raw
material is mainly minerals of geological origin, and hence, the name is geopolymer.
Creating geopolymer cement requires an alumina silicate material, a user-friendly
alkaline reagent, sodium- or potassium-soluble silicates [5]. Geopolymer possesses
excellent mechanical property, fire resistance and acid resistance [6]. The geopoly-
mer product with fly ash and GGBS shows considerable resistance to chemical attack
compared to the Portland cement product [7]. The property of geopolymer concrete
with a fly ash aggregate mainly depends on the type of method of curing of aggregate
[8]. Aluminosilicate network having Si–O–Al–O bonds in polymeric form, which are
formed by dispersion of a precursor material such as fly ash or metakaolin (which has
silica and alumina) in an alkaline solution which contains reagents such as sodium
hydroxide (NaOH) and sodium silicate (Na2 SiO3 ) or potassium hydroxide (KOH)
and potassium silicate (K2 SiO3 ) [9].
8.2 Mix Design
Designing an appropriate mix proportion of SCC using geopolymer aggregate was

challenging task. EFNARC method of mix design was adopted using different sizes
of graded aggregates making suitable adjustments (Table 8.1).
8 Experimental Investigations on SCC with Indigenously … 77
Table 8.1 Mix design of self-compacted concrete

Sl. Volume Proportion Cement Fly Fine Coarse Water Superplasticizer
No. (m3 ) (kg) ash aggregate aggregate (l) (kg)
(kg) (kg) (kg)
1 1 1:1.47:1.42 300 240 796.5 767 175 5.4
2 1 1:1.42:1.7 350 210 796.5 767 175 5.6
3 1 1:1.37:132 400 178 796.5 767 175 5.78
8.3 Experimental Investigations
Table 8.2 shows various properties of natural and GPA aggregate. From the table, it
is observed that elongation index, flakiness index and angularity index are absent as
coarse aggregate developed is round in nature.
Concrete cubes were cast with aggregate sizes varying from 10 to 20 mm. Fresh
and hardened properties were found for the design mix. Natural aggregates were
replaced in the range of 0–100% by geopolymer aggregates as shown in Table 8.3.
Same investigations have been carried out for both 350 and 400 kg of cement.
Figures 8.1 and 8.2 reflect the data
Table 8.2 Physical characterization of geopolymer aggregate

Sl. no. Tests Geopolymer aggregate Conventional aggregate
1. Fineness modulus 6.28 7
2. Flakiness index (%) 0 19.2
3. Elongation index (%) 0 15
4. Angularity index 1.17 11
5. Specific gravity 2.67 2.72
6. Water absorption (%) 0.66 0.3
7. Bulk density (kg/m3 ) 1463 1542
8. Aggregate crushing value 26 24
(%)
10. Aggregate impact value (%) 21 20
11. Aggregate abrasion value 20 18
(%)
78
Table 8.3 Fresh and hardened properties of SCC

Sl. no. Cement Aggregate Percentage of Slump flow V funnel (s) J ring (mm) L box ratio U box (mm) Compressive
content size (mm) replacement (mm) strength
(kg/m3 ) (N/mm2 )
1 300 10 0 700 8.5 2 1 12 42.04
25 712 8 2 1 11 43.06
50 730 7.9 1 1 10 47
75 742 7.1 0 1 8 46.5
100 760 7.1 0 1 6 43
2 300 12.5 0 678 9.9 4 0.89 16 49.46
25 682 9.2 4 0.9 16 56.6
50 700 8.8 3 0.95 15 58.2
75 710 8.9 2 0.98 14 51.25
100 724 8.1 2 1 12 48.9
3 300 20 0 666 11.5 8 0.8 22 37.5
25 682 11.1 8 0.84 21 41.2
50 695 10.5 7 0.87 18 46.1
75 702 10.2 6 0.88 18 44.2
100 714 10.2 6 0.91 16 39.3
M. P. Naveena et al.
(a)
Slump flow (mm) Slump flow Value for 10mm GPA (b) Slump flow Value for 20mm GPA
Slump flow (mm)

800 750
750
700
700
650 650
0 50 100 150 0 50 100 150
Percentage of replacement of NA (%) Percentage of replacement (%)
300 350 400 300 350 400
kg/m3 of cement kg/m3 kg/m3 of cement
Fig. 8.1 Slump flow value of GPA
(a) (b)
28 days Compressive strength for
Compressive strength (MPa)
28 days Compressive strength for

Compressive strength (MPa)
300 kg cement content 350 kg cement content
100 100
50 50
0 0
0 50 100 150 0 50 100 150
Percentage of replacement (%) Percentage of replacement (%)
10 12.5 20 10 12.5 20
Aggregate size (mm) Aggregate size (mm)
Fig. 8.2 Compressive strength value of GPA
8.4.1 Flow Characteristics
Figure 8.1a, b shows the slump of SCC for 10 mm and 20 mm aggregate, respectively.
It is observed from Fig. 8.1 that higher the cement content higher is the slump. It is
also seen that with increase in aggregate size slump also decreases.
From Table 8.3, it is also observed that V funnel, L box and J ring test show
increase in flowability of SCC. However, the U box values have decreased with
increase in aggregate size. This could be due to the filling ability of geopolymer
aggregate.
8.4.2 Compressive Strength
The compressive strength was observed to be high for SCC mix with 350 kg cement
content and 10 mm aggregate with a value of 56.9 N/mm2 . From Fig. 8.2a, b, it
Table 8.4 Rapid chloride penetration and bond stress of SCC

Sl. No. Cement content Percentage of RCPT (C) Bond stress (MPa)
(kg/m3 ) replacement Aggregate size (mm)
10 12.5 20 10 12.5 20
1 300 0 3500 3500 3560 12.15 12.37 11.71
25 3650 3700 3750 12.37 12.81 12.37
50 3800 3820 3900 12.81 13.04 12.59
75 4000 4020 4010 13.04 13.48 13.14
100 4250 4300 4215 12.59 12.81 12.90
is observed that as the aggregate size increases compressive strength is reducing.

Further studies need to be carried out to study the size effect of the aggregate.
8.4.3 Rapid Chloride Penetration Test (RCPT)
From Table 8.4, it is observed that penetration of chlorine increases with increase
in aggregate size as well as cement content. The aggregates being round in nature
increase the voids.
8.4.4 Bond Strength
Concrete cubes were cast with rods inserted with 200 mm projections. These were
tested in UTM to find the bond strength. It is observed from Table 8.3 that bond
stress is in the range of 12–14 N/mm2 which is satisfactory.
8.4.5 Microstructure Analysis
Figure 8.3a, b is SEM images of natural aggregate and GPA. It is observed the natural
particle size varies from 1.2 to 9.28 µm and geopolymer particle size from 2.98 to
4.70 µm. Geopolymer particles are more densely packed as compared to natural
aggregate.
Fig. 8.3 SEM images of natural and geopolymer aggregate
Fig. 8.4 EDS image of natural aggregate
8.4.6 Energy Dispersive Spectroscopy (EDS)
EDS is an analytical technique used for the elemental analysis or chemical charac-
terization of a sample. It is found that silica content is higher in natural aggregate as
compared to geopolymer aggregate. Calcium content is high in geopolymer aggregate
as shown in Figs. 8.4 and 8.5.
8.5 Regression Analysis
It is of interest to predict strength of compressive strength of concrete for different

percentage replacements of natural aggregates with GPA for varying cement con-
tent. Regression Analysis is carried out in MATLAB. Figure 8.6 shows graphical
representation of predicted values for various percentage replacements of natural
aggregates with geopolymer aggregate. Figure 8.7 represents surface plot of the
Fig. 8.5 EDS image of geopolymer aggregate
Fig. 8.6 Compressive strength for different sizes of aggregates
predicted values.
8.6 Application of Geopolymer Aggregate
Further investigations were carried out to check the suitability of geopolymer aggre-
gate in sleepers. The sleepers were cast in Malu sleepers Pvt. Ltd., who are one of the
suppliers for Indian Railways. The bending and electric resistivity tests of sleepers
were carried out in yard. It was found that the bending strength of sleepers cast by
Fig. 8.7 Surface plot
Table 8.5 Static bending and electrical resistivity test results

Sleeper Center top Center Rail seat Electrical
in (kN) bottom in Resistance Failure load resistivity
(kN) load (kN) (kN) (V)
R.H.S L.H.S R.H.S L.H.S
Sleeper 70 65 238 236 380 378 160
(GPA)
Standard 60 52.50 230 230 370 370 Max. 240
value as per
RDSO
6 M GPA with 50% replacement met with standards prescribed by Indian railways.
It was also found that electrical resistivity was also within the limits from Table 8.5.
8.7 Conclusions
Geopolymer aggregate is indigenously developed in KSSEM Laboratories. It was

observed from the studies that with increase in aggregate size slump decreases. It
was found that V funnel, L box and J ring test show increase in flowability of
SCC. However, the U box values have decreased with increase in aggregate size.
This could be due to the filling ability of geopolymer aggregate. The compressive
strength was observed to be high for SCC mix with 350 kg cement content and
10 mm aggregate with a value of 56.9 N/mm2 . The penetration of chlorine increases
with increase in aggregate size as well as cement content. The aggregates being
round in nature increase the voids. The bond stress is in the range of 12–14 N/mm2
which is satisfactory. SEM and EDS show the morphological properties as well
as chemical properties of the aggregate. It is of interest to predict the compressive
strength for intermediate values. Regression analysis has been carried out using
MATLAB. However, the values should be validated by testing these mixes.
References
1. Sethu Parvathy, S., Sharma, A.K., Anand, K.B.: Comparative study on synthesis and properties
of geopolymer fine aggregate from fly ashes. Constr. Build. Mater. 163, 428–437 (2018)
2. Saint-Gobain and Bosch.: Fredonia, World Construction Aggregates—Industry Study with
Forecasts for 2015 and 2020. The Freedonia Group, USA (2012)
3. Vishalakshi, K.P., Revathi, V., Sivamurthy Reddy, S.: Effect of type of coarse aggregate on the
strength properties and fracture energy of normal and high strength concrete. Eng. Fract. Mech.
(2018)
4. Oner, A., Akyuz, S.: An experimental study on optimum usage of GGBS for the compressive
strength of concrete. Cement Concr. Compos. 29, 505–514 (2007)
5. Davidovits, J.: Geopolymers: ceramic-like inorganic polymers. J. Ceram. Sci. Technol. 8(3)
(2017)
6. Davidovits, J., Davidovits, M.: Geopolymer room temperature ceramic matrix for composites.
Ceram. Eng. Sci. Proc. 9(842), e853 (1988)
7. Sakkas, K., Nomikos, P., Sofianos, A., Panias, D.: Fire resistant geopolymer for passive fire
protection. Fire Mater. (Special issue) D (2014)
8. Venkata Suresh, G., Pavan Kumar Reddy, P., Karthikeyan, J.: Effect of GGBS and fly ash
aggregates on properties of geopolymer concrete. J. Struct. Eng. 43(5), 436–444 (2017)
9. Menon, S.U., Anand, K.B., Sharma, A.K.: Performance evaluation of alkali-activated coal ash
aggregate in concrete. Waste Resour. Manage. 171(1), 4–13 (2018)
Chapter 9
Experimental Studies and Numerical
Validation on Bearing Capacity
of Skirted Footings on c- Soils
Arekal Vijay, Vijayalakshmi Akella and B. K. Raghu Prasad
Abstract This work discusses the suitability of inclined internal skirts along with
outer vertical plates or skirts on bearing capacity of skirted footings on c- soils.
The parameters considered are thickness and depth of vertical plates and the spacing
of inclined internal skirts. The scaled model footings were made up of mild steel
which was considered for this research work. Square and rectangle are the two
shapes adopted. The size was designed in such a way that the lateral confinement
effect does not affect the test results. The foundation soil was in dry condition. The
various parameters considered are thickness, depth and spacing of internal skirts.
In square footing, for a 5-mm-thick skirt, 40-mm-skirt depth and 10-mm-internal
skirt spacing, increase in bearing capacity observed was 2.4 times. If the depth of
skirt is increased to 120 mm keeping all other parameters unchanged, the bearing
capacity was increased to 13 times. For square footing, the highest improvement
in bearing capacity was 14 times for skirt depth of 120 mm with skirt thickness of
5 mm and internal skirts spacing of 10 mm. Increase in spacing of internal skirts
from 10 to 30 mm for square footing with 5-mm-thick skirt with 40-mm-skirt depth
showed a reduction in bearing capacity from 238 to 150%. Numerical validation of
experimental test results was carried out by using Plaxis software.
Keywords Skirted footings · Bearing capacity · Settlement
9.1 Introduction
The present study aims at improving the bearing capacity of foundations soil by con-
finement. The confinement of soil is achieved by providing vertical insertions along
with the periphery of footings. These vertical insertions resist the lateral movement
of foundation soil. These vertical insertions are termed as skirts, and foundations with
these skirts are termed as skirted footings or bucket foundations. These foundations
are employed in offshore structures to replace deep foundations. However, skirted
A. Vijay (B) · V. Akella · B. K. Raghu Prasad

KS School of Engineering and Management, Bengaluru, Karnataka 560109, India

https://doi.org/10.1007/978-981-15-3254-2_9
86 A. Vijay et al.
foundations are yet to be adopted for onshore structures. Presently, researchers are
focusing on this aspect. Most of the previous research focused on footings with verti-
cal insertions. In this work, apart from outer vertical insertions, footings are provided
with additional inner inclined insertions for improved load-carrying capacity.
9.2 Literature Survey
Eid [1] conducted numerical studies on axially loaded skirted shallow foundations.
The results showed that bearing capacity and settlement values of skirted foundations
were close to pier foundations of the same width and depth. Bearing capacity of shal-
low foundation increased with increase in skirt depth and decrease in relative density
of sand. Settlement reduction exceeds 70% for a ratio of skirt depth/foundation width
equal to 2.
Finite element limit analysis conducted by Mana et al. [2] identifies the criti-
cal internal skirt spacing. With the increasing skirt depth, fewer internal skirts are
required. But increasing soil strength heterogeneity requires more internal skirts. The
results also indicate that the reduction in number of skirts reduced bearing capacity.
Laboratory tests were conducted by El Sawwaf and Nazer [3] to study the influ-
ence of soil confinement on the behaviour of a footing resting on granular soil. The
parameters considered are the diameter and depth of cell and the embedded depth of
footing. The results showed soil confinement increase bearing capacity of circular
footing. The cells with small cell diameters exhibit deep foundation behaviour, while
this pattern was not found with large cell diameters.
9.3 Methodology
9.3.1 Model Footings Without Skirt
In the present work, square-shaped footing with and without skirts are used for the
testing. The footing sizes were smaller than (1/5) B (where B = width of container)
to avoid lateral confinement effect. The plan area of footings is 80 mm × 80 mm.
The thickness of footings is 10 mm made of mild steel. Vertical plates of thickness
1, 3 and 5 mm are used as skirts. The skirt depths of 0.5B, 1.0B and 1.5B were used
to understand the effect of bearing capacity due to variation in skirt depth where B
is the least lateral dimension of the model footing.
9 Experimental Studies and Numerical Validation on Bearing … 87
Fig. 9.1 Square footing with

vertical and inclined skirts

vertical and inclined skirts
9.3.2 Model Footings with Skirts
Model footings and skirts are connected by welded joints so as to form a monolithic
footing. The line diagrams of square footing are shown in Figs. 9.1 and 9.2. Totally,
18 model square footings of varying skirt depths (40–120 mm) and skirt thicknesses
(1–5 mm) were used for the study.
9.3.3 c-Φ Soil
An oven-dried soil was used in the experiment program. The soil was sieved as per
Indian Standard specifications. The geotechnical properties of the soil are given in
Table 9.1.
Table 9.1 Geotechnical

Property Value
properties of soil
Specific gravity (G) 2.67
Effective particle size (D10 ), mm 0.022
Particle size (D30 ), mm 0.068
Particle size (D60 ), mm 0.36
Unit weight (γ ), kN/m3 13.73
Angle of internal friction 30°
Cohesion (c), kN/m2 11
88 A. Vijay et al.
Fig. 9.3 Test tank and

loading frame
9.3.4 Test Tank and Loading Frame
The steel tank of size of 0.75 m (length), 0.45 m (width) and 0.75 m (depth) as shown
in Fig. 9.3 was used to hold foundation soil. To avoid buckling of sidewalls of the
tank during loading, stiffeners were provided. Static vertical loads were applied, and
load-settlement readings were taken.
9.3.5 Test Procedure
Raining technique was adopted to prepare the bed of soil. This was done in lift
of 50 mm thick up to full tank height. The soil was poured from a predetermined
height and with controlled intensity of pouring, required relative density of soil was
obtained. To keep the model footing in full contact with the soil, the top of soil layer
was levelled. This will also facilitate the load applied to the footing was vertical
(normal). The unit weight of the soil was maintained at 13.73 kN/m3 in dry state.
After each test cycle, the soil beneath the footing was excavated to a depth of 1.5B.
The footing was placed on the predetermined prepared surface of the soil in the
tank. The vertical load at a constant rate was gradually applied to the footing until
the settlement is seized. Two dial gauges were used to record the settlements of the
footing. For each cycle, the load-settlement readings were recorded.
9.3.6 Numerical Modelling
An extensive 2D finite element modelling and analysis was carried out to evaluate the
behaviour of the footing with and without structural skirts. The axisymmetry model
is used for square footing with 15-node triangular element, and the dimensions were
set for modelling. The dimensions of c-F soil 450 mm × 750 mm and thickness
of 750 mm are modelled in the drawing area, and the standard boundary condition
is assigned. The footing model with and without structural skirts with varying skirt
thickness of 1 and 3 mm, skirt depths of 0.5B where ‘B’ is the least lateral dimension of
the footing is modelled. The interface is given between footing and soil structure and
between the structural skirt and soil structure. The point load is applied concentrically
on the footing. The geometry of the finite element soil model adopted for the analysis
is 450 mm × 750 mm × 750 mm shown. Soft soil model is adopted for soil modelling.
The main feature of soft soil model is that the stiffness of soil is dependent on stress
applied. The failure behaviour of soil is in accordance with the Mohr–Coulomb
criterion. The material properties of c-F soil are defined according to soft soil model.
The stresses initially are more at the bottom of the soil model and gradually
decrease towards the surface of the soil model due to the fixity at the bottom and sides.
The load is applied concentrically as in experimental study for 24 mm settlement.
Stress point nodes are selected to get the load. The deformation of the footing and
soil model is obtained as shown in Fig. 9.4. From the analysis, vertical displacement
is obtained.
Fig. 9.4 Deformed mesh and vertical displacement

90 A. Vijay et al.
Extensive pressure-settlement experiments on square-shaped skirted footings on c-

soil were carried out. The pressure-settlement curves for a square footing of size
80 mm × 80 mm with thickness varying from 1 to 5 mm with constant depth of
40 mm are shown in Fig. 9.5. It is evident from the figure that the bearing capacity
increases with an increase in the skirt thickness. The bearing capacity was improved
by 55.50% (6.90–10.73 kN/m2 ) for a skirt thickness of 1 mm, 125.60% (6.90–15.56
kN/m2 ) for 3 mm skirt thickness and 238.90% (6.90–23.38 kN/m2 ) for 5 mm skirt
thickness for a constant skirt depth of 40 mm. It is due to increased thickness of skirts
increase rigidity of skirt and hence higher resistance to lateral pressure.
The pressure-settlement curves for a model square footing of size 80 mm × 80 mm
are shown in Fig. 9.6. For a concentric load, the variation in bearing capacity without
skirts and with skirts of thickness 1, 3 and 5 mm with internal skirts having 30 mm
spacing and constant skirt depth of 40 mm is plotted. The values of bearing capacity
of footings with skirts are compared with the bearing capacity values of footings
without skirts.
From Fig. 9.6, it is evident that the bearing capacity increases with increase in the
skirt thickness. The bearing capacity was improved by 44.40% (6.90–9.96 kN/m2 )
for a skirt thickness of 1 mm, 73.40% (6.90–11.96 kN/m2 ) for 3 mm skirt thickness
and 150% (6.90–17.24 kN/m2 ) for 5 mm skirt thickness in c-F soil for a constant
skirt depth of 40 mm.
Fig. 9.5 Pressure-settlement curves for a square footing (skirt spacing = 10 mm)
Figures 9.5 and 9.6 indicate that the bearing capacity for the square footing with
10-mm-internal skirt spacing is greater when compared to the model square footing
with 30-mm-internal skirt spacing. The comparison of bearing capacity improve-
ments in model square footings with 10- and 30-mm-internal skirt spacing for a skirt
depth of 40 mm is discussed in Table 9.2.
The pressure-settlement curves for a model square footing of size 80 mm × 80 mm
are shown in Fig. 9.7. For a concentric load, the variation in bearing capacity without
skirts and with internal skirts of thickness 1, 3 and 5 mm having 10 mm spacing and
constant skirt depth of 80 mm is plotted. The values of bearing capacity of model
footings with skirts are compared with the bearing capacity values of model footings
without skirts.
Figure 9.8 shows that the bearing capacity increases with increase in the skirt
thickness. The bearing capacity was improved by 198.80% (6.90–20.62 kN/m2 ) for
a skirt thickness of 1 mm, 269.90% (6.90–25.52 kN/m2 ) for 3 mm skirt thickness
Table 9.2 Bearing capacity improvement in square footings with 10- and 30–mm-internal skirt
spacing for 40-mm-skirt depth
Skirt thickness Bearing capacity improvement in Bearing capacity improvement in
square footing having square footing having
10–mm-internal spacing 30-mm-internal spacing
t = 1 mm 1.6 times 1.4 times
92 A. Vijay et al.
and 375.40% (6.90–32.80 kN/m2 ) for 5 mm skirt thickness in c-F soil for a constant
skirt depth of 80 mm.
In Fig. 9.8, pressure-settlement curve for different thickness of internal skirt with
constant depth and spacing of 80 mm and 30 mm, respectively, is shown. It is evident
that the bearing capacity increases with increase in the skirt thickness. The bearing
capacity was improved by 177.70% (6.90–19.16 kN/m2 ) for a skirt thickness of
1 mm, 211% (6.90–21.46 kN/m2 ) for 3 mm skirt thickness and 255.50% (6.90–
24.53 kN/m2 ) for 5 mm skirt thickness for a constant skirt depth of 80 mm. From
Figs. 9.7 and 9.8, it is clear that the bearing capacity for the same square footing with
10-mm-internal skirt spacing is greater when compared to the square footing with
30-mm-internal skirt spacing. The comparison of bearing capacity improvements
with increasing skirt thickness in model square footings with 10 and 30 mm spacing
for a skirt depth of 80 mm is shown in Table 9.3. Table 9.4 shows similar trend for
120-mm-skirt depth.
• Figures 9.9, 9.10 and 9.11 show numerical analysis results of square footings with
skirts (d = 40 mm, s = 10 mm and t = 1, 3 and 5 mm). From the figures, the
effect of thickness of skirts on the deformation in the foundation soil and stress
concentration in the footing and the skirts can be compared.
Table 9.3 Bearing capacity improvement in square footings with 80-mm-skirt depth and varying
skirt thickness and internal skirt spacing
Skirt thickness Bearing capacity improvement for Bearing capacity improvement for
square footing having square footing having
10-mm-internal spacing 30-mm-internal spacing
t = 1 mm 3 times 2.8 times
Table 9.4 Bearing capacity improvement in square footings with 120-mm-skirt depth and varying
skirt thickness and internal skirt spacing
Skirt thickness Bearing capacity improvement in Bearing capacity improvement in
footing having 10-mm-internal footing having 30-mm-internal
spacing spacing
t = 1 mm 6.6 times 4 times

skirt t = 1 mm
94 A. Vijay et al.
Fig. 9.10 Square footing

with skirt t = 3 mm
Fig. 9.11 Square footing

with skirt t = 5 mm
• The higher intensity of stress is found in the internal inclined skirts as compared
to peripheral vertical skirts as shown in Figs. 9.12, 9.13 and 9.14. The maximum
Fig. 9.12 Max principal

stress (5000 kN/m2 )

stress (766 kN/m2 )

stress (237 kN/m2 )
deformation is found between the skirts and deformation reduces significantly

below the skirt tip.
• Further, the deformation between the inclined skirts is significantly lesser as com-
pared to outside the inclined skirts indicating that inclined skirts resist vertical
loads better than the vertical loads as shown in Figs. 9.15, 9.16 and 9.17.
9.5 Summary
• Results showed that spacing of internal inclined skirts has a significant effect on
the bearing capacity of square footings on c- soil. Increase in spacing of internal
skirts from 10 to 30 mm for square footing with 5-mm-thick skirt with 40-mm-
skirt depth showed reduction in bearing capacity from 238 to 150%. This trend
confirms that smaller cells are more effective in confining foundation soil.
• Pressure-settlement curves showed improvement in bearing capacity with the
increase in skirt depth for both square footings. The maximum effect on bear-
ing capacity value was found to be for a skirt depth of 1.5B. For example, in
square footing, for a 5-mm-thick skirt, 40-mm-skirt depth and 10-mm-internal
96 A. Vijay et al.
Fig. 9.15 Deformed mesh

for skirt thickness 1 mm


skirt spacing, increase in bearing capacity observed was 2.4 times. If the depth of
skirt is increased to 120 mm keeping all other parameters constant, the bearing
capacity was increased by 13 times.
• Increase in skirt thickness for square footing increased the bearing capacity. For
a skirt thickness of 5 mm, the bearing capacity was found to be higher when
compared to 1- and 3-mm-thick skirts. For square footing, the highest improvement
in bearing capacity was 14 times for skirt depth of 120 mm with skirt thickness of
5 mm and internal skirts spacing of 10 mm. This trend is attributed to the increased
thickness of skirts increase rigidity of skirt and hence higher resistance to lateral
pressure.
• The percentage of difference in settlement for square footing (for 40 mm skirt
depth) ranges between 5.5 and 19.83%. Similar trend is observed for 80 mm skirt
depth. The percentage of difference ranges between 6.58 and 29.25%.
• The higher intensity of stress is found in the internal inclined skirts as compared to
peripheral vertical skirts. The maximum deformation is found between the skirts
and deformation reduces significantly below the skirt tip. Further, the deforma-
tion between the inclined skirts is significantly lesser as compared to outside the
inclined skirts indicating that inclined skirts resist vertical loads better than the
vertical loads. It may be concluded that the analytical result closely matches with
experimental results in terms of percentage difference in settlement.
Acknowledgements The authors appreciate the KS School of Engineering and Management for
providing the necessary infrastructure and cooperation to carry out this work.
98 A. Vijay et al.
References
1. Eid, H.: Bearing capacity and settlement of skirted shallow foundations on sand. Int. J. Geomech.,
645–652 (2013). https://doi.org/10.1061/(asce)gm.1943-5622.0000237
2. Mana, D.S.K., Gourvenec, S., Martin, C.M.: Critical skirt spacing for shallow foundations under
general loading. J. Geotech. Geoenviron. Eng. 139(9), 1554–1566 (2013)
3. El Sawwaf, M., Nazer, A.: Behavior of circular footings resting on confined granular soil. J.
Geotech. Geoenviron. Eng. 131(3), 359–366 (2005)
Chapter 10
Effect of Aerodynamic Modifications
on a Tall Building with Horizontal
Irregularity
Shanku Mandal , Sujit Kumar Dalui and Soumya Bhattacharjya
Abstract Impact of wind force is more critical on tall buildings with an increase in
height. Horizontal irregularity on plan shape and randomness of wind makes the struc-
ture more vulnerable. A detail analytical study has been done considering various
wind incidence angle ranging from 0° to 90° at an interval of 30° using computational
fluid dynamics (CFD) on U-plan shaped tall building with corner modifications. The
present study has shown that the introduction of minor aerodynamic modifications
has major a role in minimizing the wind-induced responses. Drag coefficient and
the lift coefficient are reduced due to the application of corner chamfered and corner
roundness on the basic model. Most of the cases both maximum pressure and suc-
tion occur on the different faces of corner chamfered models. It is also noted that the
modified corner exerts more pressure than the basic model.
Keywords Horizontal irregularity · Wind incidence angle · Computational fluid

dynamics · Aerodynamic modification
10.1 Introduction
Corner modifications do not only enhance the aesthetic beauty of the structure but
also play a major role for drastic chance in wind pressure on structural elements
of tall building faces. The unconventional shape of the buildings makes them more
sensitive to wind than those with a regular shape. Kim and You [1] performed several
wind tunnel tests and concluded that the tapering effect has a more significant effect
in across-wind direction than that in along-wind direction. Gomes et al. [2] used a
S. Mandal (B) · S. K. Dalui · S. Bhattacharjya

S. K. Dalui
S. Bhattacharjya
https://doi.org/10.1007/978-981-15-3254-2_10
100 S. Mandal et al.
closed circuit wind tunnel for testing 1:100 scale irregular-plan shaped models with
different wind incidence angles and find out the noticeable difference in pressure
distributions on inner faces due to the influence of the additional wing which trans-
forms the L into U-shaped model. Ilgin and Gunel [3] suggested different design
methods and major aerodynamics modification like tapering, setbacks, sculptured
building tops, varying the shape and openings, etc., should be adopted to ensure the
functional performance of flexible structure and control the wind-induced motion of
tall buildings. Tse et al. [4] investigated the impact of the amount and type of corner
modifications. Recessed corners modifications are shown to be an effective mini-
mizer of the wind-induced loads than chamfered corners. The overall construction
costs can be reduced with the application of corner modifications. Sevalia et al. [5]
numerically studied the force coefficient of different geometric plan configurations
like square, circular, hexagon and octagon tall building having the same plan area.
They conclude that the circular plan shape of building is much better when compare
to the other plan shape of building in terms of both wind pressure coefficient as well
as total drag force on building. Tanaka et al. [6] found out that helical has better
aerodynamics behavior over other configurations like corner cut, setback due to shed
irregularity throughout the height. Xie [7] considered aerodynamic optimization is
the most efficient way to ensure the structure safety in strong winds and control
the wind-induced motion of super-tall buildings. Kim and Kand [8] discussed the
effectiveness of aerodynamic modification in reducing wind loads due to the fact that
change in building shape with height promotes frequent, random and incoherent vor-
tex formation at different levels. Bandi et al. [9] tested different cross-sectional tall
buildings with configurations of straight triangle, corner cut, 60° helical, 180° helical
and 360° helical, and clover to investigate the variations in along-wind and cross-
wind overturning moment coefficients. Sharma et al. [10] experimentally investigate
the advantages of tapered and setback model over regular conventional shapes. Set-
back building model is the effective minimizer of the wind loads compare to taper
modified building. Present study mainly focuses on the effect of rounded corner and
chamfered corner on U-plan shaped tall building at different wind incidence angles.
Comparison of force coefficient and mean pressure coefficient of various models are
shown.
10.2 Scope of Study
The effect of wind force on regular-plan shaped building is available in various codal
provisions but very limited data are available for irregular-plan shaped building.
Detail study should be done to evaluate the wind responses due to the abundant
presence of U-plan shaped buildings. Numerical simulation using computational
fluid dynamics (CFD) is the best-suited alternative where wind tunnel facility is not
available.
Different models as shown in Fig. 10.1 have been analyzed by ANSYS CFX
considering 0° to 90° wind incidence angle at an interval of 30°. The U1 model has a
10 Effect of Aerodynamic Modifications on a Tall Building … 101
Fig. 10.1 Analytical models a plan view of basic model, b elevation view of basic model, c basic
model with face name and wind incident angle (U1), d corner chamfered model with face name
and wind incident angle (U2) and e corner rounded model with face name and wind incident angle
(U3)
uniform cross section throughout the height. U2 model has been modified with corner
chamfer of 0.25b and the same amount of corner roundness has been provided in
case of U3 model. Here, b, b , d and H are considered as 250 mm, 50 mm, 150 mm
and 500 mm, respectively.
10.3 Computational Setup and Meshing
The wind-induced responses of the analytical models have been studied on the
domain recommended by Frank et al. [11] and Revuz et al. [12]. The building model
has been placed at 5H from inlet, sidewalls and roof of the domain, and 15H from
the outlet as shown in Fig. 10.2. The height of the analytical model has been taken as
H. The flow of the wind is not restricted and all the characteristic behaviors of wind
can be observed in this domain. No blockage correction is needed for this domain.
The k-ε turbulence model is considered in our present study. The boundary con-
dition for sidewalls and roof of the domain has been taken as free slip walls, and
the walls of the building and the floor of the domain are taken as no-slip walls. The
terrain category II is adopted as per Indian standard IS: 875-2015 (part 3) [13]. The
analytical model has been modeled at a geometric scale of 1:300. The free stream
velocity of wind at inlet is taken 10 m/s.
The equation of atmospheric boundary layer (ABL) wind profile is taken as
Fig. 10.2 a Plan view and b elevation view of computational domain used for CFD simulation
Fig. 10.3 Domain meshing of U-plan shaped building model
α
Ux x
= (10.1)
U∞ x0
Ux is the velocity at any height ‘x’ from ground.

U∞ is the free steam velocity.
x is the height at any point above ground surface.
x0 is the atmospheric boundary layer height.
α is a power-law exponent which depends on the roughness of the terrain.
The value of the exponent α is taken as 0.133 conforming to terrain category II
as per IS: 875-2015 (part 3) [13]. The boundary layer height (x 0 ) is taken as 1 m.
The domain and the building meshing are done using tetrahedral elements as
recommended by Lo et al. [14]. Finer meshing is employed around building models
to capture the more efficient flow for accurately analyze the higher gradient of the
wind flow as shown in Fig. 10.3.
10.4 Comparison with the Previous Study
Figure 10.4a, b shows the comparison of wind velocity profile and turbulence inten-
sity profile between the data obtained from the present study, wind tunnel study by
Dalui [15] and CFD by Kar et al. [16]. The comparison shows more or less simi-
lar characteristics with the previous study due to generation of similar type of wind
environment on every case. Figure 10.4c displays pressure coefficients along vertical
centerline at face C of U-plan shaped model for 0° wind incidence angle. To validate
the current analytical study, a comparison of pressure coefficient between the current
study and the previous study of Gomes et al. [2] has been done.
Fig. 10.4 Comparison with the previous study a velocity profile, b turbulence intensity and c pres-
sure coefficients along vertical centerline at face C of U-shaped model for 0° wind incidence
angle
The wind-induced effect on different faces of basic and corner modified U-plan
shaped tall buildings are shown considering 0° as well as 30°, 60° and 90° wind
incidence angle. The typical wind flow patterns around basic and corner modified
building models are shown in Fig. 10.5. Symmetrical vortices are formed on the
leeward side of basic model due to normal wind incidence angle, but asymmetric
vortices are formed on the corner modified building models. The formation of vortices
varies significantly with the change in wind angle. The horizontal irregularity of the
U-plan shaped building promotes the production of unsymmetrical and irregular
vortices in the wake region. Corner modification as well as self-interfering limbs of
the building also form irregular vortices with the change in angle of wind attack.
Fig. 10.5 Wind flow pattern around various model a basic, b corner chamfered, c corner rounded
and d 60° wind incidence angle
Comparison of drag coefficient and lift coefficient of basic U-shaped and corner
modified U-shaped models are shown in Fig. 10.6. It is clearly visible that sharp
corner model or basic U-shaped model attracts more wind force in comparison with
corner modified models. Corner modifications are efficient to reduce drag coeffi-
cient as well as lift coefficient. Corner roundness is more effective to reduce wind-
induced responses in along-wind direction. In some cases, corner chamfered shows
less attraction to wind force on across-wind direction with change in attack of wind
angle. At 60° wind incidence angle, the effect of wind-induced responses is critical
in across-wind direction.
Fig. 10.6 Comparison of force coefficient of various models with different wind incidence angle
a drag coefficient and b lift coefficient
Fig. 10.7 Comparison of mean pressure coefficient on all faces of various model at a 0°, b 30°,
c 60° and d 90° wind incidence angle
The mean pressure coefficients on different faces are shown in Fig. 10.7. Mean
pressure coefficient on corner rounded models are less compare to other models in
case of normal wind incidence angle. Corner chamfered models exert more pressure
on side faces in most of the cases. Maximum positive pressure occurred on windward
direction. Side faces and leeward faces exert suction due to side wash and formation
of vortex, respectively. The change in mean pressure coefficient on various faces
changes due to change in wind incidence angle. At 30° wind incidence angle on face
A of basic model, mean wind pressure is negligible due to presence in both positive
and negative pressure, but in case of corner modified models negative pressure is
predominant on this face. At 60° wind angle maximum suction is occur on face A of
corner chamfered model, but at 90° wind angle maximum suction is occur on face
E. It is a clear indication that with the change in wind incidence angle, the variation
in wind-induced responses shows mainly due to the presence of limbs on U-shaped
buildings.
The comparison of pressure coefficient (C p ) along perimeter of building models
is shown in Fig. 10.8 at 250 mm height from the base on horizontal direction. The
important observation is to be noted that on the modified corner the pressure is more
as compare to basic model in most of the cases.
10.6 Conclusion
The role of corner modification on irregular-plan shaped tall building at different

wind incidence angles is presented in the current study. The effects of modification
can be summarized as follows.
Fig. 10.8 Comparison of pressure coefficient along the horizontal center line of different models
at a 0°, b 30°, c 60° and d 90° wind incidence angle
• The drag coefficient is maximum in case of normal wind incidence angle and
minimum in case of 90° wind incidence angle.
• It is noticeable that the drag coefficient is reduced with the introduction of corner
modifications. Though rounded and chamfered modification show almost similar
reduction in drag coefficient, corner roundness is more efficient in reducing the
drag coefficient.
• At 60° wind incidence angle, the lift coefficient is maximum in every case.
• It is observed that on corner rounded model lift coefficient is less than corner
chamfered model at 0° and 30° wind incidence angle, but at 60° and 90° wind
incidence angle the lift coefficient is more on corner rounded models than that of
corner chamfered models.
• For each case, maximum positive pressure is generated on face C at 0° wind inci-
dence angle, and face A of corner Chamfered model is experienced the maximum
negative pressure at 60° wind incidence angle.
• At 90° angle, all the faces except face F exert suction.
• Most of the cases the corner modification exert more positive as well as negative
pressure than that of basic model.
On the basis of obtained results, it is clear that corner modifications play an impor-
tant role in reducing force coefficient on tall building. Providing corner roundness on
basic model is more effective than chamfered corner. Current investigation gives a
serious indication that the designer may provide modifications to minimize the wind
force on structure but should take special care in case of designing the clad struc-
tural elements. The above study shows the necessity of detailed study considering
the various angle of wind attack to find out the worst cases before designing any
structure.
References
1. Kim, Y.M., You, K.P.: Dynamic responses of a tapered tall building to wind loads. J. Wind
Eng. Ind. Aerodyn. 90, 1771–1782 (2002)
2. Gomes, M.G., Moret Rodrigues, A., Mendes, P.: Experimental and numerical study of wind
pressures on irregular-plan shapes. J. Wind Eng. Ind. Aerodyn. (93), 741–756 (2005)
3. Ilgin, H.E., Gunel, M.H.: The role of aerodynamic modifications in the form of tall buildings
against wind excitation. Metu Jfa (2), 2 (2007)
4. Tse, K.T., Hitchcock, P.A., Kwok, K.C.S., Thepmongkorn, S., Chan, C.M.: Economic per-
spectives of aerodynamic treatments of square tall buildings. J. Wind Eng. Ind. Aerodyn. 97,
455–467 (2009)
5. Sevalia, J.K., Desai, A.K., Vasanwala, S.A.: Effect of geometric plan configuration of tall
building on wind force coefficient using CFD. Int. J. Adv. Eng. Res. Stud. (I), 4 (2012)
6. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M., Chul Kim, Y.: Experimental investigation of
configurations. J. Wind Eng. Ind. Aerodyn. (107–108), 179–191 (2012)
7. Xie, J.: Aerodynamic optimization in super-tall building designs. In: Seventh International
Colloquium on Bluff Body Aerodynamics and Its Applications, pp. 104–111 (2012)
8. Kim, Y.C., Kand, J.: Wind pressures on tapered and set-back tall buildings. J. Fluids Struct.
39, 306–321 (2013)
9. Bandi, E.K., Tamura, Y., Yoshida, A., Chul Kim, Y., Yang, Q.: Experimental investigation on
aerodynamic characteristics of various triangular-section high-rise buildings. J. Wind Eng. Ind.
Aerodyn. (122), 60–68 (2013)
10. Sharma, A., Mittal, H., Gairola, A.: Wind-induced forces and flow field of aerodynamically
modified buildings. Environ. Fluid Mech. 1–25 (2019). https://doi.org/10.1007/s10652-019-
09687-9
11. Franke, J., Hirsch, C., Jensen, A.G. Krus, H.W., Schatzmann, M., Miles, S.D., Westbury, P.S.,
Wisse, J.A., Wright, N.G.: Recommendations on the use of CFD in wind engineering. Cost
Action C 1–11 (2004)
12. Revuz, J., Hargreaves, D.M., Owen, J.S.: On the domain size for the steady-state CFD modelling
of a tall building. Wind Struct. Int. J. 15, 313–329 (2012)
13. IS: 875 (Part-3): Code of Practice for Design Loads (other than Earthquake Loads), for Building
and Structures—Wind Loads (2015)
14. Lo, Y.L., Kim, Y.C., Li, Y.C.: Downstream interference effect of high-rise buildings under
turbulent boundary layer flow. J. Wind Eng. Ind. Aerodyn. 159, 19–35 (2016)
15. Dalui, S.K.: Wind Effects on Tall Buildings with Peculiar Shapes. Ph.D. thesis, Indian Institute
of Technology Roorkee (2008)
16. Kar, R., Dalui, S.K., Bhattacharjya, S.: An efficient optimization approach for wind interference
effect on octagonal tall building. Wind Struct. Int. J. 28, 111–128 (2019)
Chapter 11
Flexural Behaviour of 2D Cellular
Lattice Structures Manufactured
by Fused Deposition Modelling
Harshini Gullapalli, Syed H. Masood, Syed Riza and Panneer Ponnusamy
Abstract Cellular lattice structures have applications in crash resistant and

protective equipment. This study involves designing rectangular beams of lattice
structures with different unit cell configurations, fabricating them on fused deposition
modelling (FDM) 3D printing machine and subsequently finding out the structural
design for the best flexural performance in engineering applications. All the test
specimens were fabricated on Stratasys Dimension 1200 ES FDM machine in
acrylonitrile butadiene styrene (ABS) material. Five types of 2D unit cells for building
lattice structures were investigated which included honeycomb, diamond, square,
circle, and triangle shapes. Three-point bending tests were carried out to study flexural
properties like flexural strength and modulus of these cellular lattice structures. After
mechanical testing, the obtained data for the five different lattice structures were
mutually compared for their flexural behaviour and also with the beams built in solid
and sparse shapes. Results of flexural tests indicate that the cellular lattice structures
based on triangular and honeycomb shapes exhibit maximum flexural strength. The
triangular structure also proved to be the highest in the value of flexural modulus
among all the five cellular lattice structures. This study also includes investigating
the effects of varying porosity of a particular cellular structure (honeycomb) on its
flexural behaviour. Results show that for honeycomb structures, the higher porosity
led to higher flexural strength and flexural modulus and shorter build time up to a
porosity of 61%.
Keywords Cellular structures · Flexural behaviour · Fused deposition modelling ·

3D printing
H. Gullapalli · S. H. Masood (B) · S. Riza · P. Ponnusamy

Department of Mechanical and Product Design Engineering, Swinburne University of
Technology, Melbourne, Australia

https://doi.org/10.1007/978-981-15-3254-2_11
110 H. Gullapalli et al.
11.1 Introduction
The fused deposition modelling (FDM) by Stratasys Inc. is one of the most
widely used additive fabrication technologies that manufacture various complex
geometry parts using layered manufacturing concept [1]. Materials usually used
by FDM process include acrylonitrile butadiene styrene (ABS) plastic, PLA and
polycarbonate materials. In this paper, the build material used is ABS plastic
because it is considered as one of the best materials of the styrene family due to
its characteristics like toughness, hardness, rigidity, good chemical resistance and
dimensional stability. In the FDM process, the ABS raw material filament is heated
and melted inside the nozzle and extruded in the form of thin layers along the two-
dimensional (X-Y ) layer pattern path to produce the parts as per the designed CAD
model. FDM process is one of the additive manufacturing processes that is cost-
effective and provides the user with greater flexibility in manufacturing intricate
complex parts such as cellular structures.
Cellular lattice structures of both stochastic and non-stochastic types are gaining
increasing attention by researchers due to the advent of additive manufacturing (AM)
technologies, which now offer much more convenient methods of manufacturing
such complex structures compared to conventional manufacturing processes. Several
researchers have investigated various AM technologies, in both metals and polymers,
to design and manufacture such lattice structures and to examine their mechanical
properties. There is a wealth of research that has investigated the additive
manufacturing of metallic cellular structures and their mechanical properties like the
extensive work done by Williams et al. [2]. Iyibilgin et al. [3] conducted a thorough
experimental investigation using fused deposition modelling (FDM) process for
evaluating the time taken to build up a given lattice structure and subsequently tested
these structures for their compressive properties. Cansizoglu et al. [4] carried out
experiments to evaluate the non-stochastic lattice structures fabricated by electron
beam melting process to conduct an observational study on the metal foam structures.
Yang et al. [5] investigated cellular core sandwich panels for comparing their bending
properties. Electron beam melting (EBM) and selective laser melting (SLM) are the
two most important technologies for developing cellular metallic structures. Number
of researches can be found that have employed these systems like Yan et al. [6]
studying the behaviour of lattice structures using SLM; Horn et al. [7] carrying
out investigations on Ti6Al4V cellular structures built on EBM and Li et al. [8]
experimenting on the influence of cellular form upon the mechanical properties of
meshes of Ti6Al4V fabricated by EBM. Gunay et al. [9] had performed studies to
calculate the measurement uncertainty for plastic (ABS) material in flexural testing.
A thorough literature survey also indicates that though many types of research can be
found regarding the study of the compressive and material properties of cellular and
lattice structures, very few studies have been carried out for determining the flexural
behaviour of polymeric parts fabricated using the fused deposition modelling AM
process.
11 Flexural Behaviour of 2D Cellular Lattice Structures … 111
In this paper, five different types of cellular lattice structures were manufactured
in ABS material on an FDM machine. The care is taken that all these structures have
more or less the same porosity (~52%). Subsequent testing was done for evaluating
their flexural behaviour and build times. The values obtained were used for mutual
comparison and also for comparison with two other beam structures called sparse
and solid parts based on build styles available on the FDM machine software. Five
different types of 2D geometric unit cells (honeycomb, square, diamond, circle
and triangle) were used to design the cellular beams using Creo computer-aided
design (CAD) software. Honeycomb structures were also studied for mechanical
behaviour under variable porosities. The comparisons eventually lead to determining
which structural design can provide the best flexural properties and performance for
engineering applications.
11.2 Flexural Testing of FDM Cellular Beams
The required lattice beam structures were fabricated using ABS plastic material
on the Stratasys 1200 ES FDM Machine. The dimensions of parts that can be
fabricated on this setup are limited to 254 × 254 × 305 mm. The fabricated
beams were of rectangular structures with 235 mm in length, 35 mm in width
and 15 mm in thickness. Figure 11.1a shows the CAD models of the five beam
structures, namely square, circular, triangular, diamond-shaped and honeycomb
forms. Figure 11.1b shows the fabricated beam structures. Since the flexural strength
of triangular and honeycomb structures was found to be higher than their counterparts,
it was considered appropriate to produce extra samples of the honeycomb type and
investigate these for the effect of varying porosity on their flexural behaviour. Two
other beams (solid and sparse) of the same size were also fabricated using FDM while
employing the solid and sparse build style options of the FDM processing software.
Fig. 11.1 a CAD models of five cellular beam structures, b beams made using FDM in ABS
Fig. 11.2 a Three-point bending set up, b example of honeycomb cellular beam in bending
Three-point flexural tests were performed using on the MTS Criterion Model 43
machine, which combines high performance with a maximum capacity of 50 kN.
The machine is controlled by MTS TestSuite Software. Using the software and data
acquisition unit of the machine, the test parameters can be adjusted in a wide range.
The test specimen was deformed at a speed of 0.5 mm/min, and the resulting recorded
data helps in evaluating the flexural strength and flexural modulus for the fabricated
specimens. Figure 11.2a shows the three-point bending set-up, and Fig. 11.2b shows
the honeycomb cellular beam in bending.
11.3.1 Effects of Cellular Lattice Structures
The flexural behaviour of five different types of cellular lattice structure, as well as
sparse and solid beam structures, is investigated. All five cellular specimens possess
approximately the same porosity (i.e. 52%). The sparse build style of FDM software
allows interior toolpaths with air gaps to create a non-solid internal structure saving
material and build time. Figure 11.3 shows the flexural stress–strain curves obtained
from three-point bending tests conducted on the above-mentioned structures and
beams. From the obtained plots, it is clear that all the stress–strain curves of cellular
structures exhibit similar trends in the elastic region. However, the stress–strain
curves for the sparse and solid build styles indicate slightly different behaviour. The
cellular structures experienced failure within the strain range of 13 and 40%, but for
solid build structures, the strain at failure was 55% and the value was 32% for the
sparse build beams. The parts manufactured with the solid build style seem to be
more elastic in nature. It is observed that the triangular structure has a more brittle
So
HC T
C S
Sp
D
Fig. 11.3 Flexural stress–strain curves for five cellular structures
nature and higher stiffness among all the cellular lattice part structures. Square and
honeycomb structures have high ductile nature and higher elongation. Triangular and
diamond cellular structures are said to be more brittle in nature than other cellular
parts. Triangular and honeycomb structures have high flexural strength, high strain
energy, high toughness and stronger than all other cellular lattice part structures.
The build times for all the tested lattice and solid and sparse structures are plotted
in Fig. 11.4. Minimum build time of 352 min was recorded for the circular structures
while the triangular structures took the maximum time of 550 min. Solid and sparse
beam structures took appreciably fewer times of 195 and 117 min, respectively. The
vast difference between the build times is due to the fact that the investigated cellular
lattice structures needed complex FDM toolpaths for fabrication. The triangular
Fig. 11.4 Build times for various flexural beams in FDM structures of the same porosity
structure took the highest build time of all the five lattice structures because of its
intricate shape and complex toolpaths.
Figures 11.5 and 11.6 each show the comparison of flexural strength and flexural
modulus, respectively, for the five cellular structures with the sparse and solid built
structures. According to the data, triangular and honeycomb lattice structures exhibit
the highest comparative flexural strength, while the triangular structure shows the
maximum flexural modulus among all the five cellular lattice structures. It is clear
the flexural strength of honeycomb and triangular lattice structures is lower than the
solid and sparse structures, but the flexural modulus of honeycomb and triangular
Fig. 11.5 Flexural strength for different cellular structures
Fig. 11.6 Flexural modulus for different cellular structures

structure is comparable to the sparse structure. From the flexural tests, it is clear that
the solid part structure possesses much higher flexural strength and flexural modulus
than all other lattice part structures, but it will use several times more material than
cellular structures.
11.3.2 Effects of Varying Porosity
For honeycomb structures, further investigations were carried out to find the effects
of porosity on flexural properties. Porosity may be defined as the measure of the void
or empty spaces within the mass of material, and it can be calculated as the ratio of
the volume of pores to the total volume of a sample. Figure 11.7 shows the flexural
stress–strain curves obtained for honeycomb lattice structures with three porosities
(50, 56 and 61%). During the initial stages of loading, almost every curve exhibits
a linear elastic behaviour. The range of ‘strain at failure’ is found to be between 25
and 30% for all specimens with different porosities. From the graphs, it is noted that
all the honeycomb structures with varying porosity undergo brittle failure and the
structure with 61% porosity is stronger and possesses higher stiffness than structures
with 50 and 56% porosity.
Figure 11.8 highlights the comparison of the flexural strength and flexural modulus
for the tested honeycomb cellular structures. An increase in the flexural properties is
noticed with rise in porosity. Such behaviour is an expected because an increase in the
porosity within a prescribed limit will not only decrease the amount of material in the
part but also allow more room for flexibility. A sharp increase is noticeable in flexural
strength and modulus for the structures having 61% porosity in comparison with the
other two honeycomb structures with porosities of 50 and 56%. This observation
demands that the porosity effect demands more in-depth investigation over a wider
range of porosity ratios.
While comparing the build times for the varying porosity honeycomb structures,
it can be observed that build time reduces with an increase in porosity. This behaviour
Fig. 11.7 Flexural

stress–strain curves of
HC(61%)
varying porosity of the
honeycomb structure
HC(56%)
HC(50%)
(a) (b)
Fig. 11.8 a Flexural strength versus porosity and b flexural modulus versus porosity for honeycomb
beams
may be due to the reason that the amount of material used in part building decreases
with an increase in part porosity. For a 61% porous honeycomb structure, the build
time was reduced to less than half in comparison with the build time for the structure
with 50% porosity.
11.4 Conclusions
Fused deposition modelling process was used to prepare lattice and solid sparse
structures. Flexural properties and build times were evaluated and compared by
conducting experimental studies. Results show that triangular and honeycomb
cellular lattice structures possess the highest flexural strength, and the triangular
structure possesses the highest flexural modulus of all the five cellular lattice
structures; it is clear the flexural strength of honeycomb and triangular lattice
structures is lower than the solid and sparse structures; and flexural modulus of
the triangular structure is lower than the solid structure. From the flexural tests, it is
clear that a solid part structure possesses much higher flexural strength and flexural
modulus than all other lattice part structures. A significant difference was observed
in build times of the five fabricated FDM structures, which were almost double of the
build times of the solid and sparse structures. Therefore, a contrasting set of properties
can be observed between the cellular structures and the solid and sparse beams in
terms of flexural properties and build time. Several stress–strain curves produced
for the five cellular structures conclusively illustrate that the triangular structure has
more brittle nature but higher stiffness among all the cellular lattice part structures,
while square and honeycomb structures have higher ductility and experience failure
at higher values of elongation. Triangular and honeycomb structures were found to
provide best flexural performance due to their possession of high flexural strength,
high strain energy, high toughness than all other cellular lattice structures created on
FDM.
References
1. Ziemian, C., Sharma, M., Ziemian, S.: Anisotropic mechanical properties of ABS parts fabricated
by fused deposition modeling. In: Mechanical Engineering, InTechOpen (2012)
2. Williams, C.B., Cochran, J.K., Rosen, D.W.: Additive manufacturing of metallic cellular
materials via three-dimensional printing. Int. J. Adv. Manuf. Technol. 53(1–4), 231–239 (2011)
3. Iyibilgin, O., Yigit, C., Leu, M.C.: Experimental investigation of different cellular lattice
structures manufactured by fused deposition modeling. In: Proceedings of the Solid Freeform
Fabrication Symposium, Austin, TX, USA, pp. 895–907 (2013)
4. Cansizoglu, O., Harrysson, O., Cormier, D., West, H., Mahale, T.: Properties of Ti–6Al–4V non-
stochastic lattice structures fabricated via electron beam melting. Mater. Sci. Eng., A 492(1–2),
468–474 (2008)
5. Yang, L., Harrysson, O., West, H., Cormier, D.: A comparison of bending properties for cellular
core sandwich panels. Mater. Sci. Appl. 4(08), 471 (2013)
6. Yan, C., Hao, L., Hussein, A., Raymont, D.: Evaluations of cellular lattice structures
manufactured using selective laser melting. Int. J. Mach. Tools Manuf. 62, 32–38 (2012)
7. Horn, T.J., Harrysson, O.L., Marcellin-Little, D.J., West, H.A., Lascelles, B.D.X., Aman, R.:
Flexural properties of Ti6Al4V rhombic dodecahedron open cellular structures fabricated with
electron beam melting. Addit. Manuf. 1, 2–11 (2014)
8. Li, S., Xu, Q., Wang, Z., Hou, W., Hao, Y., Yang, R., Murr, L.: Influence of cell shape on
mechanical properties of Ti–6Al–4V meshes fabricated by electron beam melting method. Acta
Biomater. 10(10), 4537–4547 (2014)
9. Gunay, A., Fank, S., Gulmez, T., Durakbasa, N.: Calculation of measurement uncertainty for
plastic (ABS) material in flexural testing. Int. J. Metrol. Qual. Eng. 4(1), 29–33 (2013)
Chapter 12
Power Spectral Density on Principal
Building Due to Setback Interfering
Building
Soumya Mukherjee and Amlan Kumar Bairagi
Abstract The problem of getting living and working spaces is increasing with the
rise in human population. To minimize the effects of this problem, modern technology
introduces tall buildings. Wind-based structural analysis of tall buildings helps us
to build tall structures keeping them safe. Interference effect is one such important
wind phenomenon, which affects the structure greatly. That is why the need of the
study on interference effect is so significant these days. In this study, two virtual
building models are used. The principal one is a square-plan tall building, and the
interfering one is a setback tall building, which is placed in front of the principal
building obstructing a time bound wind flow of a gust lasting for 3 s. This particular
study offers power spectral density (PSD) variation with respect to Strouhal number
for top near corner target points at a level of 0.975h on each surface of the principal
model as well as the streamline variation in the altitude level of 0.475h. All the
variations are obtained due to the positional changes of the interfering building, as it
is rotated about its own axis with an interval of 15° from 0° to 90°. The maximum PSD
value is obtained for 15° orientation of the interfering building from the windward
face of the principal building. For the 90° orientation of the interfering building, the
PSD values obtained from leeward and side faces of the principal building are nearly
same.
Keywords Power spectral density · Strouhal number · Interfering model
S. Mukherjee (B)
Techno India University, Salt Lake, Kolkata, India
A. K. Bairagi

https://doi.org/10.1007/978-981-15-3254-2_12
120 S. Mukherjee and A. K. Bairagi
12.1 Introduction
The advancement of human civilization initiates the need of tall structures because
the shrinkage of building spaces is increasing with the growing number of human
population on the earth. Many remarkable tall structures were destroyed by natural
catastrophe; one of these impactful natural forces is wind. That is why wind-loading
analysis of tall structures is very much necessary. One of the important wind-based
phenomenons is interference, which caused by an obstruction by any interfering
object to the flow of wind toward the principal object. Through the years in the
modern era, many research works have been done to study the wind loading and
its nature on tall building structures, some of which played integral parts during
this study. Xie and Gu [1] studied the shielding effect of interference factor using
two and three high-rise buildings having the same height. Lam et al. [2] focused
on the interference effects between two closely spaced high-rise buildings. Blocken
et al. [3–5] analyzed the venturi effects of wind at the pedestrian level between two
tall buildings using CFD method and wind tunnel experimentation. Kim et al. [6]
studied the comparison of wind pressure coefficients between isolated and principal
buildings along with the effects of interference on wind pressure coefficients. Bairagi
and Dalui [7, 8] studied the optimum distance for nullifying the wind interference
effects on a principal building due to the presence of an interfering building for 0°
and 90° wind angles. Kheary and Dalui [9] discussed the wind interference effect
between two buildings for different elevations and distances. Roy and Bairagi [10]
analyzed the wind force and pressure parameters for unconventional tall buildings
having different geometry. Bairagi and Dalui [11–13] analyzed the aerodynamic
behavior of different types of setback tall buildings using power spectral density
(PSD) function. Yu et al. [14] analyzed the envelope interference factors (EIFs) of
wind on principal building for different aspect ratios of interfering building. The
current research deals with application of computational fluid dynamics (CFD) for
studying the wind interference effects on a tall square-plan-shaped principal building
model (PM) due to the obstructive presence of an interfering tall building model (IM),
which is rotating on its own axis. This study focused on the nature of streamlining of
a wind flow generated by a time bound gust, the vibration generated by the same time
bound gust in the principal building model (PM) is analyzed, and the frequencies and
PSDs are accumulated for certain critical faces of the principal building model (PM)
for certain critical positional orientations of the interfering building model (IM).
12.2 Domain, Meshing, and Model Detail
The most important guidelines for domain creation are made based on the concept
by Franke et al. [15], which are the inlet, top, and side walls of the simulative
wind tunnel domain at a distance of 5th multiplier of the overall height [5h] of the
model. The outlet of the domain is at a distance of 15th multiplier of the overall
12 Power Spectral Density on Principal Building Due to Setback … 121
Fig. 12.1 a Domain and b meshing of square bluff body
height [15h] of the model (for the domain details, see Fig. 12.1a). In this study, to
calculate the solution while the simulation is running on, both the domain and the
model meshes are improved at locations depending upon specified adoption criteria.
At these locations, the solution variables are hanging most rapidly while solving
the flow features at these particular locations. The tetrahedron meshing method has
been utilized during this study, and by the usage of this mesh, the regions near the
boundaries can be made so inflated that the flows with unusual characteristics can
be avoided (for the meshing details, see Fig. 12.1b).
In this case, the study of CFD method has been used to analyze the wind loading.
Two numbers of square-plan tall building models are used here, of which the shapes
and the surface areas of the bottom faces are same. One of the model is the principal
model (PM) of which the measurement of wind pressure variation with respect to
time has been taken during a gust, which lasted for 3 s. The other model is the
interfering model (IM), which is obstructing the gust in the way toward the PM. The
measurements have been taken for seven different positional orientations based upon
the rotation of the model IM from 0° to 90° at an interval of 15° about its own axis
along the XY plane with respect to the angle of the flow of the gust. The dimensional
parameters of the PM and the IM are described at length (l) = 100 mm, width (b) =
100 mm, and overall height (h) = 400 mm. Setbacks of (0.2l) = 20 mm are provided
at half the overall height (0.5h) = 200 mm on the IM from two opposite sides. The
spacing between both of the models is (2b) = 200 mm for all seven orientations of
the IM. The four faces of the IM excluding the top are labeled as windward face (A),
leeward face (B), and side faces (C and D) (for the modeling details, see Fig. 12.2a).
12.3 Boundary Condition
The present study has been done using CFD methodology depending upon software-
based simulation. This study is run and analyzed by creating virtual domain and
observing the effects on PM after placing of IM in the path of a gust. The simulated
flow conditions have been generated as per the criteria of terrain category 2 taken from
IS 875: (part 3) [16]. The magnitude of inlet velocity was considered normal to the
Fig. 12.2 a Schematic diagram of interfering and principal models, b validation with Kim Kanda
(2010) and analytical model
boundary. Following the requirements of directional constraints, the flow direction

was considered parallel to the normal of the surface of the boundary. The inlet and
outlet are specified in the virtual domain. The sidewalls of this virtual domain are
specified to be as free slip wall in this study. The walls of both of the virtual models are
specified to be as no slip wall with no surface roughness. In this study, power law has
been used which is shown in Eq. (12.1). The velocity profile of atmospheric boundary
layer is considered as per SP 64 (S&T) [17]. On the other hand, the formulation to
calculate the turbulence intensity is shown in Eq. (12.2).
(u/u h ) = (z/z h )α (12.1)
I = u /u mean (12.2)
where u represents the speed of wind in horizontal direction at a certain elevation

z. uh represents the speed of the wind at a reference elevation zh . The value of uh is
taken as 10 m/s. The value of zh is taken as 1.0 m. α represents the coefficient of
surface roughness 0.133 for terrain category 2. I represents the turbulence intensity,
and u represents the root mean square of turbulence velocity. umean represents the
average velocity of wind.
12.4 Validation of Present Study
This study is based on CFX application, which creates the simulation of a virtual
wind tunnel testing. That is why it is necessary to validate the simulative results with
past wind tunnel test of a same model with the same aspect ratio. The present study
has been validated with Kim and Kanda [18]. The length scale of the model used
in this analytical study is about 1:4, which is same as the experimental research by
Kim and Kanda. The boundary conditions of this study have been created as per the
experimental conditions for open flat area flow of wind. The power law exponent
has been taken as 0.13, and the turbulence intensity has been taken as 15%, which
is same as the experimental work. The virtual domain is created in such a manner
that the blockage ratio of the present analytical study is same as the experimental
study, which is 1.2% as a square-plan model is used to analyze the behavior of wind
excitation for both the studies (for the parametric graphical representation of wind
velocity and turbulence intensity versus the height of the model, see Fig. 12.2b).
12.5 Results and Findings
The study analyzed the effects of wind interference on a PM caused by a gust lasts for
3 s, which is nothing but a strong and intensified flow of wind. This gust is interfered
by another setback tall building IM in its way toward the PM. The measurements
have been done by obtaining the pressure variation with respect to time and plotting
the PSD versus Strouhal number (S t ) graph, which are taken from top corner points
of all the faces of the PM.
12.5.1 Power Spectral Density (PSD) of Principal Model
The concept of using power spectrum to describe the distribution of power into
frequency components initiated by a certain vibration of a particular time series is
done as per Fourier analysis. This vibration is generated due to the force of gust
on the PM, which is interfered by setback IM. When the energy created by the
vibration is concentrated around a certain time interval, the PSD can be calculated.
The expression for PSD is expressed in Eq. (12.3).
PSD = Sp ( f )/σ 2 (12.3)
where PSD represents the power spectral density, S p represents the peak wind spectra
function, f represents the frequency, and σ represents the standard deviation.
12.5.2 Strouhal Number (St ) of Principal Model
The Strouhal number (S t ) is a dimensionless parameter used in dimensional analysis,

where the object is under dynamic loading condition. Strouhal number (S t ) is
generally utilized to describe the oscillating flow mechanisms, and Strouhal number
(S t ) is expressed in Eq. (12.4).
St = f B/U (12.4)
where S t represents the Strouhal number, B represents the width of the model, and
U represents the gradient wind speed.
The marking and directional orientation of all the four faces A, B, C and D on
the model PM as well as the considered impact points on each faces are shown in
Fig. 12.3. The results have been taken by creating certain impact points on the PM.
The impact points have been taken at 0.975h, i.e., near the top corners of each face of
the model PM, which results in two impact points for each faces. The maximum value
of spectral density 0.277 is obtained for Strouhal number 0.031 due to 15° orientation
at face A (see Fig. 12.3a). Again, 90° orientation has maximum spectral densities
0.07, 0.068, and 0.08 for Strouhal numbers 0.048, 0.032, and 0.048, respectively, for
the faces B, C, and D (see Fig. 12.3b–d). An interesting point has been observed for
0° orientation that the minimum spectral densities are 1.90 × 10−4 and 9.3 × 10−5
for same Strouhal number 0.048 at faces C and D. The streamline diagram of the
wind flow of the gust inside the virtual domain has been shown here for 15° and 90°
orientations of the interfering building at the level of 190 mm (0.475h), respectively
(see Fig. 12.4).
Fig. 12.3 Variation of PSD of principal model at topmost edge point (0.975h) for different
orientations of interfering model a face A, b face B, c face C, and d face D
Fig. 12.4 Streamline for velocity at 0.475h for different orientations of interfering model a for 15°
orientation and b for 90° orientation
12.6 Conclusions
The present analytical study is done by a series of simulative iterations, which is a

time-consuming and rigorous method. After the complete simulative analysis, the
following conclusions are made based on the obtained results.
• According to the spectral density calculation, face A of the principal model has
maximum frequency 0.277 for 15° orientation of interfering model, which means
the face A suffering large amount of positive pressure at the topmost corner, and
this pressure is developed due to the side wash of the interfering model.
• Faces B, C, and D have high frequency for 90° orientation, but face D has maximum
spectral density 0.08 at the topmost edge.
• Faces C and D experienced with low frequency 1.90 × 10−4 and 9.3 × 10−5 for
the same Strouhal number 0.048 for 0° orientation.
• The streamlines varies for different orientation of the interfering building model
for the time bound wind flow of the gust.
References
1. Xie, Z.N., Gu, M.: Mean interference effects among tall buildings. Eng. Struct. 26, 1173–1183
(2004)
2. Lam, K.M., Leung, M.Y.H., Zhao, J.G.: Interference effects of wind loading of a row of closely
spaced tall buildings. J. Wind Eng. Ind. Aerodyn. 96, 562–583 (2008)
3. Blocken, B., Stathopoulos, T., Carmeliet, J.: Wind environmental conditions in passages
between two long narrow perpendicular buildings. J. Aerosp. Eng. ASCE 21(4), 280–287
(2008)
4. Blocken, B., Moonen, P., Stathopoulos, T., Carmeliet, J.: A numerical study on the existence
of venturi-effect in passages between perpendicular buildings. J. Eng. Mech. ASCE 314(12)
(2008)
5. Blocken, B., Carmeliet, J.: Pedestrian wind conditions at outdoor platforms in a high-
rise apartment building: generic sub-configuration validation, wind comfort assessment and
uncertainty issues. Wind Struct. 11(1), 51–70 (2008)
6. Kim, W., Tamura, Y., Yoshida, A.: Interference effects of two buildings on peak wind pressure.
In: The Seventh Asia-Pacific Conference on Wind Engineering, Taipei, Taiwan (2009)
different orientation using CFD. In: 3rd World Conference of Applied Sciences, Engineering
& Technology, Kathmandu, Nepal, pp. 764–774 (2014)
8. Bairagi, A.K., Dalui, S.K.: Optimization of interference effects on high-rise buildings for
9. Kheyari, P., Dalui, S.K.: Estimation of wind load on tall building under interference effects.
Jordan J. Civil Eng. 9(1), 84–108 (2015)
tall building using CFD simulation—a case study. Asian J. Civil Eng. (BHRC) 17(8), 1055–
1075 (2016)
Int. J. Mech. Prod. Eng. Res. Dev., 413–420 (2018)
due to wind load. Asian J Civil Eng Build Hous. 19(2), 205–221 (2018). (Springer International
Publishing)
In: 2nd International Conference on Advances in Dynamics, Vibration and Control, NIT
Durgapur, pp. 381–388 (2018)
14. Yu, X., Xie, Z.N., Gu, M.: Interference effects between two tall buildings with different section
sizes on wind induced acceleration. J. Wind Eng. Ind. Aerodyn. 182, 16–26 (2018)
Wright, N.G.: Recommendation on the use of CFD in wind engineering. COST Action C14.
Dynamics (2004)
16. IS:875 (Part-3): 2015. Indian Standard Code of Practice for the Design Loads (other than
Earthquake) for Buildings and Structures (part-3. Wind Loads). Bureau of Indian Standards,
New Delhi, India (2015)
17. SP 64 (S&T): 2001. Explanatory Hand Book on Indian Standard Code of Practice for the Design
Loads (other than Earthquake) for Buildings and Structures (part-3. Wind Loads) [IS:875
(Part-3):1987]. Bureau of Indian Standards, New Delhi, India (2001)
18. Kim, Y.C., Kanda, J.: Wind pressure on tapered and setback tall buildings. J. Fluids Struct. 39,
306–321 (2013)
Chapter 13
Effect of Solar Optical Properties
of Building Envelope on Time Lag,
Decrement Factor and Energy Saving
of Buildings
Debasish Mahapatra and T. P. Ashok Babu
Abstract World has been suffering from energy crisis since 1970. This is because
of the increased population and industrialization. Buildings have a significant
contribution to total energy consumption across the globe. The increased urbanization
leads to boom in construction of buildings. The materials used for the construction
of buildings affect the climate of the city up to a great extent and affect the energy
consumption as well. So, construction materials should be chosen wisely. The surface
temperature of the building envelope is affected by the coatings used. Cool coatings
help in reducing the surface temperature of the building envelope. Time lag and
decrement factor are the indicators of energy saving in building. In this paper, the
effect of cool coating on time lag and decrement factor was studied. Values of time
lag and decrement factor were evaluated in different climatic zones of India. It has
been found that as the absorptivity of the building envelope increases, the decrement
factor increases. By changing the absorptivity from 0.2 to 0.73, a minimum change of
16% (New Delhi) in decrement factor and 11.75% (Jodhpur) in surface temperature
was observed. The maximum change in decrement factor was 56.44% (Jodhpur).
The maximum change in surface temperature was 39.02% (Bangalore). So, it should
always be practised to keep the absorptivity of the building envelope as low as
possible.
Keywords Cool coating · Time lag · Decrement factor · Energy saving ·

Urbanization
D. Mahapatra (B) · T. P. Ashok Babu

Department of Mechanical Engineering, National Institute of Technology, Surathkal,
Karnataka, India
T. P. Ashok Babu

https://doi.org/10.1007/978-981-15-3254-2_13
128 D. Mahapatra and T. P. Ashok Babu
13.1 Introduction
Reasons such as social, political and economic compel individuals to migrate towards
urban areas. They get an easy platform to live because of the comfort lifestyle of
the city. But the drawback of urbanization is that, the heat released from various
human activities such as vehicular emissions, heat released from air conditioners,
industrial heat gets accumulated and creates an island of heat called urban heat
island (UHI). It had been reported by Valson and Bharat that, because of UHI, the
city centre temperature is around 5.6 °C higher than the suburban areas [1]. Increase
in the city centre temperature enforces increased use of air conditioner. Use of air
conditioner should be minimized considering the fact that heating and ventilation
consumes around 31% of total energy in commercial buildings and 7% in residential
buildings as per Centre for Science and Environment (CSE).
In India, creation of UHI is not so far as urbanization is taking place at a great
pace. In 1901, only 11.4% of the people were residing in urban areas. In 2001, the
count was 28.53%, and in 2011, the census rose to 31.1% [2]. Mumbai, Delhi and
Kolkata have witnessed the most of the urban migration in 2011 from rural areas. The
rise in migration is reported to be 4.1%, 3.1% and 2% in Mumbai, Delhi and Kolkata,
respectively, in 2011 as compared to 2001. So, utmost care should be taken while
selecting construction materials in order to avoid the problem of UHI in India as the
effects are adverse in nature. The heat stress caused by UHI increases the risk of
mortality. Heat stress damages thermoregulatory system in the form of heat syncope,
thermal exhaustion and cardio respiratory diseases [3]. The relation between “heat
island” and “Death island” was found from the study of Buechley et al. [4]. The
mortality rate increases exponentially with the maximum temperature. A study in
Italy during 2003 by Conti et al. found that the residents of urban areas are having a
higher risk of death compared to the residents of suburban areas because of UHI [5].
Use of cool coatings is an emerging technique used to mitigate UHI. Cool coatings
are the coatings which are characterized by their high reflectivity. Cool coatings not
only help in mitigating UHI but also help in saving energy. Because of the high
reflectivity of the coating, the surface temperature does not rise much. Because of
the less surface temperature, heat transfer into the building reduces which in turn
helps in energy saving. Synnefa et al. [6] studied the effect of cool coating on energy
saving in various cities worldwide. From their study, it was found that an increase of
solar reflectance from 0.2 to 0.6 reduced up to 75% cooling load. Similarly, increase
of reflectance from 0.2 to 0.85 reduced cooling load up to 93%.
Hernandez-Perez et al. [7] measured the inside and outside surface temperature
for coatings of reflectivity 0.80, 0.84 and 0.33. The outside surface temperature was
measured to be 30.5 °C, 23.7 °C and 38.3 °C to 40 °C for coatings having solar
reflectance of 0.80, 0.84 and 0.33, respectively. Similarly, the inside air temperatures
were 25 °C, 24 °C and 31 °C to 33 °C for coatings having solar reflectance of 0.80,
0.84 and 0.33, respectively. The heat flux was 19 W/m2 , 8 W/m2 and 45.6 W/m2 for
coatings having solar reflectance of 0.80, 0.84 and 0.33, respectively.
13 Effect of Solar Optical Properties of Building … 129
An experimental study was carried out by Shen et al. [8], and the effect of reflective
coatings on surface temperature of buildings and energy consumption was found for
both conditioned and non-conditioned buildings. For non-conditioned building, in
summer, the maximum decrease in exterior building surface temperature was 19.9 °C.
Similarly, in winter, it was 15.6 °C. The maximum surface temperature reduction
was observed in the west direction for both summer and winter seasons.
13.1.1 Time Lag and Decrement Factor
Because of the variation of solar radiation throughout the day, the ambient
temperature also varies. Maximum ambient temperature and maximum temperature
inside the building do not happen at the same time because of the thermal mass of
the wall [9]. The maximum ambient temperature and maximum inside temperature
are not the same as well. So, the time difference between the maximum ambient
temperature and maximum temperature inside the building is called time lag. The
ratio of difference of maximum and minimum temperature inside to outside is called
decrement factor (Fig. 13.1).
Mathematically,
ϕ = tTmax
0
− tTmax
e
(13.1)
And
A0 T max − T0min
f = = 0max (13.2)
Ae Te − Temin
Fig. 13.1 Schematic presentation of time lag and decrement factor. Source Asan [10]
Time lag and decrement factor are significant in places where the variation of
outside temperature is large. For example, places where the outside temperature in
day time reaches 40 °C and the following night it falls to 10–15 °C. If a material of
time lag 10–12 h is used, then the low night temperature will reach inside during day
and vice versa. By doing so, the building can be cooled or heated passively and the
use of air conditioner can be minimized thereby saving energy.
13.2 Methodology
13.2.1 Sole Air Temperature
Equation of heat transfer from the environment to the outside surface of the wall is
given by
q0 = f 0 (t0 − tso ) + aI (13.3)
It has been found convenient to combine the effect of outside temperature and the
solar radiation so an equivalent temperature is introduced called as sol-air temperature
(Fig. 13.2).
q0 = f 0 (te − tso ) (13.4)
where
aI
te = t0 + (13.5)
f0
Fig. 13.2 Heat transfer to

outside surface of a building
wall
From the above equation, it can be noticed that the heat transfer from the
environment to the outside surface of the wall is dependent on the absorptivity of the
surface. The heat transfer can be reduced by reducing the absorptivity of the surface.
As cool coatings have low absorptivity, they can be used to reduce the absorptivity
of the surface thereby reducing the heat transfer from the environment to the wall
surface, which in turn reduce the heat transfer inside the building and will save
cooling load.
13.2.2 Finite Difference Method for Solving

One-Dimensional Heat Transfer Equation
13.2.2.1 Node in the Conduction Region
See Fig. 13.3.

∂t tm + 1 − tm
=
∂x x
B
∂t tm − tm−1
=
∂x A x
The temperature gradient can be expressed as

∂t ∂t
∂ 2t ∂x B
− ∂x A
=
∂x2 m x
Fig. 13.3 Wall representing nodes in the conduction region

tm+1 − 2tm + tm−1

= (13.6)
x 2
The time derivative can be written as
∂t t − tm
= m
∂τ τ
where tm is the temperature at m after 1 Δτ time when the temperature at m was tm .
Substitution of the above values in Eq. 13.1 gives
tm − tm α
= (tm+1 − 2tm + tm−1 )
τ x 2
ατ 2ατ
= tm = (t m+1 + t m−1 ) + 1 − tm (13.7)
x 2 x 2
Suppose
x 2
M=
ατ
Then,

1 2
= tm = (tm+1 + tm−1 ) + 1 − tm (13.8)
M M
From Eq. 13.6, it can be seen that if,
2
>1
M
For higher value of tm the value of tm will be less.

So,
2
<1
M
M >2
x 2
M= >2
ατ
13.2.2.2 Nodes at the Surface and Nodes Next to Surface
See Fig. 13.4.

Energy balance for the node S can be written as
Fig. 13.4 Wall representing nodes on the surface
2k
f (te − ts ) + (tm − ts ) = 0 (13.9)
x
where
“f ” is the convective heat transfer coefficient, and “k” is the thermal conductivity of
the wall.
Equation 13.9 can be simplified to

2k 2k
f + t s = f te + tm
x x

f 2k x
ts = te + tm
f + x
2k
f + x
2k

Bi 2
= te + tm (13.10)
Bi + 2 Bi + 2
where
Bi = fx
k
is a dimensionless number called “Biot number” and is used to represent
convection boundary. For node point “m”, the finite difference is

∂t tm−1 − tm
=
∂x x
A
∂t t m − ts
=
∂x S x 2
2 ∂t
∂ t ∂x A
− ∂∂tx S
=
∂x2 m x
1
= (2ts + tm−1 + 3tm ) (13.11)
(Δx)2
The finite difference approximation of temperature at node m after time τ + Δτ

is

1 M −3
= tm = (2ts + tm−1 ) + tm
M M
For stability, M ≥ 3.
So at M = 3
1
= tm = (2ts + tm−1 )
3
13.2.3 Climatic Zones of India
Indian climatic zone is basically divided into six categories such as hot and dry,
warm and humid, composite, moderate, cold and sunny and cold and cloudy. Time
lag and decrement factor were calculated for Bangalore (moderate), Kolkata (warm
and humid), Jodhpur (hot and dry), New Delhi (composite) and Bhubaneswar (warm
and humid) on peak summer day.
13.2.4 Time Lag and Decrement Factor Calculation
The one-dimensional heat conduction equation was solved by using finite difference
method by considering nodes at the surfaces. The different parameters required for
solving are hourly outside temperature, inside temperature, i.e. room temperature,
incident solar radiation, absorptivity of the surface, outside heat transfer coefficient,
inside heat transfer coefficient, thickness of the wall and thermal conductivity of the
wall.
The outside air temperature was taken from the book ISHRAE Indian weather
data 2017 [11].
Inside room temperature was taken at 25 °C to maintain comfort condition.
The incident solar radiation which is the sum of direct and diffuse radiation was
calculated by using the necessary formula. For calculation of incident solar radiation,
a MATLAB program was developed. The results obtained were validated with the
result of Indian standard value of Delhi [12]. The deviation was within 10%. Hence,
the program was considered as reliable and used for calculating the solar radiation
of other cities. Table 13.1 shows the comparison between the literature value and
values obtained from MATLAB.
Different colours such as red, yellow, green, brown and black were considered
for this study. The reflectance value for these colours was taken from literature. The
Table 13.1 Validation of calculated solar radiation with literature (Indian standard) value
Time E cal E lit Error W cal W lit Error N cal N lit Error S cal S lit Error
6 455.45 458 0.56 65.12 65 −0.19 213.21 216 1.29 65.12 65 −0.19
7 662.13 665 0.43 75.31 76 0.91 229.56 234 1.90 75.31 76 0.91
8 682.88 685 0.31 81.93 81 −1.15 176.99 177 0.01 81.93 81 −1.15
9 606.53 653 7.12 85.90 83 −3.49 112.83 113 0.15 85.90 83 −3.49
10 467.94 463 −1.07 88.76 88 −0.86 88.76 88 −0.86 119.45 121 1.28
11 288.89 295 2.07 90.05 90 −0.05 90.05 90 −0.05 158.35 160 1.03
13 Effect of Solar Optical Properties of Building …
12 90.50 91 0.55 90.50 91 0.55 90.50 91 0.55 172.02 178 3.36

13 90.05 90 −0.05 288.89 295 2.07 90.05 90 −0.05 158.35 160 1.03
14 88.76 88 −0.86 467.94 463 −1.07 88.76 88 −0.86 119.45 121 1.28
15 85.90 83 −3.49 606.53 654 7.26 112.83 113 0.15 85.90 83 −3.49
16 81.93 81 −1.15 682.88 685 0.31 176.99 177 0.01 81.93 81 −1.15
17 75.31 76 0.91 662.13 665 0.43 229.56 234 1.90 75.31 76 0.91
18 65.12 65 −0.19 455.45 458 0.56 213.21 216 1.29 65.12 65 −0.19
135
Table 13.2 Absorptivity of different colour coating taken from literature

Colours Solar reflectivity Absorptivity Source
Yellow 0.59–0.75 0.41–0.25 Sameera et al. [13]
Black 0.27 0.73 Synnefa et al. [14]
Green 0.27 0.73 Synnefa et al. [14]
Brown 0.35 0.65 Thongkanluang et al. [15]
White 0.8 0.2 Hernandez-Perez et al. [7]
absorptivity was calculated by considering the transmittance as zero. The reflectance

value and absorbance value for different colour are given in Table 13.2.
So, based on the literature, the absorptivity values considered were 0.2, 0.25,
0.65 and 0.73. The variation of time lag and decrement factor with absorptivity was
studied.
The thickness of the wall considered was 25 cm.
Thermal conductivity and diffusivity considered were 1.5 W/m °C and 8 × 10−7
2
m /s, respectively [16].
Inside heat transfer coefficient was taken as 7 W/m2 K and outside heat transfer
coefficient was taken as 23 W/m2 K as per CIBSE 2006 [17].

for Composite (New Delhi) Climatic Zone
The time lag and decrement factor for composite climatic zone (New Delhi) were
calculated on peak summer day, i.e. on June 21st. Variation of time lag and
decrement factor with absorptivity of the wall was analysed. Variation of time
lag with absorptivity was random. The decrement factor increased with increase
in absorptivity. The decrement factor was highest in west direction and lowest in
north direction, for all absorptivity value of the wall. Inside surface temperature of
the wall, which majorly affects the cooling load, was also found. It was noticed that,
inside surface temperature varied same as decrement factor. For all the cases, the
inside surface temperature was the highest for the wall facing west and the lowest
for the wall facing south (Fig. 13.5).
Fig. 13.5 a Decrement factor. b Excess inside surface temperature. c Time lag for different
absorptivity in composite (New Delhi) climatic zone

for Warm and Humid (Kolkata) Climatic Zone
Time lag and decrement factor for the warm and humid climatic zone (Kolkata)
were calculated on peak summer day, i.e. on May 15th. Variation of time lag and
decrement factor with absorptivity of the wall was analysed and it was found that time
lag varied randomly. The decrement factor increased with increase in absorptivity.
The decrement factor was highest in west direction and lowest in south direction,
for all absorptivity value of the wall. Inside surface temperature of the wall, which
majorly affects the cooling load, was also found. It was noticed that inside surface
temperature varied same as decrement factor. For all the cases, the inside surface
temperature was the highest for the wall facing west and the lowest for the wall
facing south (Fig. 13.6).
absorptivity in warm and humid (Kolkata) climatic zone

for Moderate (Bangalore) Climatic Zone
The time lag and decrement factor for moderate climatic zone (Bangalore) were
calculated on peak summer day, i.e. on April 21st. Variation of time lag and decrement
factor with absorptivity of the wall was analysed, and it was found that time lag
varied randomly. The decrement factor increased with increase in absorptivity. The
decrement factor was highest in west direction and lowest in south direction, for all
absorptivity value of the wall. Calculation of inside surface temperature showed that
absorptivity in moderate (Bangalore) climatic zone
13.3.4 Time Lag and Decrement Factor Calculation for Hot

and Dry (Jodhpur) Climatic Zone
The time lag and decrement factor for hot and dry climatic zone (Jodhpur) were
calculated on peak summer day, i.e. on June 21st. Variation of time lag and decrement
factor with absorptivity of the wall was analysed and it was found that time lag
varied randomly. The decrement factor increased with increase in absorptivity. The
decrement factor was highest in west direction and lowest in south direction, for all
absorptivity value of the wall. Calculation of inside surface temperature showed that
absorptivity in hot and dry (Jodhpur) climatic zone

for Warm and Humid (Bhubaneswar) Climatic Zone
The time lag and decrement factor for warm and humid climatic (Bhubaneswar)
were calculated on peak summer day, i.e. on May 15th. Variation of time lag and
decrement factor with absorptivity of the wall was analysed and it was found that time
lag varied randomly. The decrement factor increased with increase in absorptivity.
The decrement factor was highest in west direction and lowest in south direction, for
all absorptivity value of the wall. Calculation of inside surface temperature showed
that inside surface temperature varied same as decrement factor. For all the cases, the
absorptivity in warm and humid (Bhubaneswar) climatic zone
13.4 Conclusions
Analysis of variation of time lag, decrement factor and inside temperature of the wall
surface with absorptivity of the wall surface draws the following conclusions
• The time lag shows random variation with absorptivity. So, it can be concluded
that time lag is completely dependent on the thermal mass of the wall.
• Decrement factor increases with an increase in the value of absorptivity of the wall
surface. For all the climatic zones studied, the decrement factor was the highest in
west direction in all cases of absorptivity. The value was the least in south direction
in all cities except New Delhi. For New Delhi, it was lowest in north direction.
• The inside surface temperature showed the same variation pattern as decrement
factor with absorptivity of the wall in all climatic zones. The inside surface
temperature increases with an increase in the value of absorptivity. For all climatic
zones, the inside surface temperature was the least for the wall facing south and
the highest for the wall facing west.
• Decrement factor and inside surface temperature of the wall play an important role
in determining the cooling load of a building. So, it should always be practised
to keep the absorptivity of the wall as low as possible to reduce the cooling load
thereby saving energy. Special care must be taken in case of the wall facing west
as the decrement factor and inside surface temperature are highest for it.
References
1. Valsson, S., Bharat, A.: Urban heat Island: cause for micro climate variation. Arch.-Time Space
& People 20–25 (2009)
2. Urbanization in India. https://en.wikipedia.org/wiki/Urbanisation_in_India. Last accessed 23
Feb 2019
3. Kleerekoper, L., Van Esch, M., Salcedo, T.B.: How to make a city climate-proof, addressing
the urban heat island effect. Resour. Conserv. Recycl. 64, 30–38 (2012)
4. Buechley, R.W., Van Bruggen, J., Truppi, L.E.: Heat island = death island? Environ. Res. 5(1),
85–92 (1972)
5. Conti, S., Meli, P., Minelli, G., Solimini, R., Toccaceli, V., Vichi, M., Beltrano, C., Perini, L.:
Epidemiologic study of mortality during the Summer 2003 heat wave in Italy. Environ. Res.
98(3), 390–399 (2005)
6. Synnefa, A., Santamouris, M., Akbari, H.: Estimating the effect of using cool coatings on energy
loads and thermal comfort in residential buildings in various climatic conditions. Energy Build.
39(11), 1167–1174 (2007)
7. Hernández-Pérez, I., Xamán, J., Macías-Melo, E.V., Aguilar-Castro, K.M., Zavala-Guillén,
I., Hernández-López, I., Simá, E.: Experimental thermal evaluation of building roofs with
conventional and reflective coatings. Energy Build. 158, 569–579 (2018)
8. Shen, H., Tan, H., Tzempelikos, A.: The effect of reflective coatings on building surface
temperatures, indoor environment and energy consumption—an experimental study. Energy
Build. 43(2–3), 573–580 (2011)
9. Arora, C.P.: Refrigeration and Air Conditioning, 3rd edn. Tata McGraw-Hill, New Delhi (2009)
10. Asan, H.: Investigation of wall’s optimum insulation position from maximum time lag and
minimum decrement factor point of view. Energy Build. 32(2), 197–203 (2000)
11. ISHRAE: Indian society of Heating, Refrigerating and Air conditioning Engineers. India (2017)
12. IS 11907-1986: Recommendation for calculation of solar radiation in buildings
13. Sameera, S., Rao, P.P., Divya, S., Raj, A.K., Thara, T.A.: High IR reflecting BiVO4-CaMoO4
based yellow pigments for cool roof applications. Energy Build. 154, 491–498 (2017)
14. Synnefa, A., Santamouris, M., Apostolakis, K.: On the development, optical properties and
thermal performance of cool colored coatings for the urban environment. Sol. Energy 81(4),
488–497 (2007)
15. Thongkanluang, T., Chirakanphaisarn, N., Limsuwan, P.: Preparation of NIR reflective brown
pigment. Procedia Eng. 32, 895–901 (2012)
16. IS: 3792: Indian standard guide for heat insulation of non-industrial buildings. India: Indian
standards institution (1978)
17. CIBSE: CIBSE Environmental Design Guide-A. 7th ed, Chartered Institution of Building
services engineers, London (2006)
Chapter 14
Study and Estimation of Static Stiffness
of Machine Tool Spindle
Vibish S. Bhardwaj, Ramesh H. Aralaguppi, Ashok N. Badhe, Bhargav

and Arun R. Rao
Abstract Spindle stiffness is the ability of spindle to resist deformations under the
influence of cutting forces. The magnitude of deflections at the spindle nose, where
cutting forces from the tool are transmitted, decides the machining accuracy of the
machine tool. In this paper, a method to evaluate the static stiffness of the spindle
with a bearing system subjected to cutting force is presented. The static stiffness of
the spindle is calculated analytically by evaluating the radial stiffness of the bearings
and the deflection at the front end of the spindle. Finite Element Analysis is carried
out on the spindle bearing system, and results are compared with analytical results.
Further, experiments are carried out to validate the analytical and numerical results.
Keywords Bearing force components · Static stiffness of spindle bearing system ·

Finite element analysis · Experimental setup
14.1 Introduction
Machine tools are used to produce high precision components in a short time with
greater accuracy. The global market demands for very low machining tolerances.
The stiffness of the machine tool spindle plays a very important role in achieving
higher machining accuracy. Accuracy of machining operations depends on the cutting
forces and the deflections caused on the spindle. For a spindle to take both radial and
axial forces, it is important to select appropriate bearings and their arrangements.
The optimum bearing arrangement contributes to the stability, working accuracy,
revolving speed and stiffness of the spindle bearing system. Bearing arrangement of
the spindle system affects the surface profile, quality and dimensional accuracy of the
manufactured parts. Static stiffness decides the deflection the spindle undergoes when
V. S. Bhardwaj · Bhargav (B) · A. R. Rao

Department of Mechanical and Manufacturing Engineering, M. S. Ramaiah University
of Applied Science, Bengaluru, India
R. H. Aralaguppi · A. N. Badhe
Dr. Abdul Kalam Center for Innovation, Bharat Fritz Werner Ltd., Bengaluru, India

https://doi.org/10.1007/978-981-15-3254-2_14
144 V. S. Bhardwaj et al.
a static force is acting on the spindle nose. When the operating frequency of these
high-speed spindles falls under the resonant zone, the amplitude of vibration in the
spindle will reduce the stiffness of the spindle bearing system considerably. Hence,
stiffness estimation of spindle with the bearing system in the static and dynamic
conditions is very important.
Shuzi [1] discussed the significance of the radial stiffness during static stiffness
analysis of a spindle bearing system. Influences of various spindle parameters like
bearing span length, overhang and bearing stiffness are studied. The bearing reaction
forces are derived from direct and indirect influence methods, and relations for spindle
stiffness with two and more bearing supports are arrived. To enhance the static
stiffness of the spindle, the stiffness of the front bearing, area moment inertia of
the spindle and reduction in the spindle overhang length are important. However,
there are no experimental validations made for the suggested methods. Sarenac [2]
studied the methods of creating a rigid spindle by optimizing the spindle parameters.
Paper suggests the ratio of bearing span length to the spindle overhang which can
help to decide the spindle stiffness. It also suggests that the higher spindle stiffness
can be achieved by improving the stiffness of the bearing and spindle cross section.
Kutlu [3] worked on the performance requirements of a machine tool spindle by
considering high loads concentrated on the spindle nose and analyzing the static and
dynamic stiffness of the spindle. The static stiffness is arrived by the superposition
method for the spindle bearing system, and the analytical values are verified with
FEA results. Soos [4] estimated the deflection on the spindle nose by considering
the spindle bearing system divided into nodes and solving analytically as the static
parameters. The bending moments on the overhang and the deflection due to bearing
compliances are used to estimate the deflections at the spindle nose. Prakosa and
Wibowo [5] built a mathematical model of the spindle bearing system and studied
the stiffness results by changing the bearing preload, span length, number of bearing
and spindle length. Later, they optimized these parameters to improve the stiffness.
Using Dunkley’s method, the resonating frequencies are obtained and compared with
experimental results.
In this paper, a methodology to estimate the static stiffness of the machine tool
spindle is developed. Radial deflection of the spindle is determined by considering the
bearing compliance when it is subjected to cutting forces analytically. The correctness
of the analytical deflections is verified numerically by using FEA software. Finally,
the deflection results obtained are validated with experimental results.
14.2 Static Stiffness Analysis Using Analytical Approach
In the spindle bearing system, shown in Fig. 14.1, a hollow spindle is supported by
angular contact ball bearings which take both axial and radial forces at high speeds.
Bearings are arranged in a quad back-to-back arrangement to accommodate axial
forces in both directions. The static stiffness of the spindle bearing system under the
14 Study and Estimation of Static Stiffness of Machine Tool Spindle 145
Fig. 14.1 Schematic of spindle bearing system
influence of cutting forces transmitted to the spindle nose during milling operations
is evaluated.
14.2.1 An Evaluation of Axial and Radial Forces on Bearing

Compliance
The axial force is acting parallel to the spindle axis due to the preload applied on
the bearings using a locknut. Tightening torque is applied on the locknut to provide
necessary preload on the bearings. The axial force is derived from the tangential force
generated by the locknut [6]. The relation between the axial force and the tightening
torque is shown in Eq. (14.1).
T
FA = (14.1)
(0.16662d)
where T is the tightening torque, d is the effective diameter of the locknut, 0.16662 is
the locknut pitch angle factor, and F A is the axial force acting on the bearing. Radial
forces are acting perpendicular to the axis of the shaft. When the cutting forces are
transmitted to the spindle nose during milling operation, it creates reaction force at
the bearings as shown in Fig. 14.2. Reaction forces acting in the radial direction are
arrived using moment and equilibrium equations. Reaction forces at the front bearing
FR1 and the rear bearing FR2 are evaluated. From Fig. 14.2, it can be observed that
due to the cutting forces acting on the spindle nose, a reaction force FR1 acts in the
upward direction on the front bearing and reaction force FR2 acts in the downward
direction on the rear bearing.
Fig. 14.2 Bearing reaction forces
14.2.2 Bearing Radial Stiffness
Bearing arrangement and the contact angle of the bearing play a major role in deciding
its radial stiffness. The radial stiffness of the bearings is evaluated by multiplying the
contact angle factor and axial stiffness of the bearing. The axial stiffness value (K A )
and the contact angle (θ Factor ) of the bearing are selected from the bearing catalog
[7]. The front and rear bearings are of the same dimensions. The relation for the
radial stiffness of the bearing is given in Eq. (14.2). Using this equation, the radial
stiffness of the bearing is calculated.
K R = K A ∗ θFactor (14.2)
14.2.3 Static Stiffness of Spindle Bearing System
Various methodologies for evaluating deflections at spindle nose when subjected to

cutting forces are discussed [3]. Deflections are evaluated using the superposition
technique. In the first case, the spindle bearing system is considered to have a rigid
shaft with elastic bearings, and in the second case, the system has an elastic shaft with
rigid bearings. Both the cases are combined to obtain deflection at spindle nose where
the shafts and bearings deform elastically. In another technique [4], the bending
moments acting on the spindle bearing system are combined with the deflections
caused by the bearing arrangements to evaluate the deflection at the spindle nose.
The spindle system shown in Fig. 14.3 is supported by front and rear bearings, and
cutting force P is acting on the spindle overhang section. Deflection at the spindle
nose is evaluated using Eq. (14.3) by considering the spindle and its support as elastic
bodies to arrive at the real-time deflections [1].
Fig. 14.3 Nature of deflection at the spindle nose

a3 La 2 1 a 2 1 a 2
δ=P + + 1+ + (14.3)
3E Ja 3E JL K1 L K2 L
where
A Spindle overhang
E Modulus of elasticity
δ Deflection at spindle nose
J a and J L Area moment of inertia at overhang and bearing span
L Bearing span length
P Cutting force
K1 and K2 = Front and rear bearing stiffness.
Assumptions made to evaluate the deflection at the spindle nose are as follows:
(a) The spindle is a flexible overhang beam body
(b) The beam is simply supported
(c) Flexible supports with stiffness values and
(d) Point load acting at the end of the overhang section.
In Eq. (14.3), the first term signifies the deflections at the overhang section, the
second term signifies the deflection at the span length where the spindle acts as a
simply support beam, and the remaining terms signify the deflection at the spindle
nose due to the deflection of the front and rear bearings. All these deflections act in
the radial direction. The sum of these deflections results in the static deflection at
the spindle nose as shown in Eq. (14.3). The static stiffness of the spindle bearing
system is arrived by using Eq. (14.4).
K = P/δ (14.4)
14.3 Finite Element Analysis
A hollow spindle model is developed, and angular contact ball bearings are mounted
in a quad back-to-back arrangement with respect to the schematic of the spindle
bearing system. COMBIN 14 element type is used to provide radial stiffness to the
bearing. Convergence study for results up to 0.1% difference in results is carried out,
and discretization of 5 mm based on convergence study is performed. The material
properties of steel are specified. Appropriate boundary conditions where the bearings
are constrained in all degrees of freedom with respect to the spindle housing are
specified, and load is applied in X-axis at the spindle nose as cutting forces. Static
structural analysis is carried out using ANSYS Workbench.
14.4 Experimental Setup
Machine tool spindle of a vertical machining center is considered. An experimental

setup is made to measure the deflection on the spindle nose due to static load. Static
load is applied on the tool holder using a beam load cell as shown in Fig. 14.4 to
create a cutting force on the spindle nose. The maximum cutting forces experienced
on the machine tool spindle during machining operations is 2000 N. The load cell
used here has a capacity of load from 0 to 1000 kg.
A Linear Variable Displacement Transducer (LVDT) with up to 0.1 μ resolution
having a magnetic base is used to measure the deflections at the spindle nose and
the tool holder. The magnetic base is placed on the stationary part of the spindle
unit close to the bearing housing. A gradual incremental load is applied, and the
corresponding deflections are measured. One of the LVDT styluses is placed on the
spindle nose and the other on the tool holder as shown in Fig. 14.4; the respective
deflections are identified. Experimental study is carried out to evaluate the spindle
bearing stiffness which is discussed in the next section.
Fig. 14.4 Experimental setup for spindle deflection test

14.5.1 Deflection Result
The deflections on the spindle nose for various external force values applied using
a load cell are evaluated analytically using Eq. (14.3). These results are compared
with experimental results and FEA results, and differences in terms of percentage
are tabulated in Table 14.1.
Table 14.1 shows the percentage difference between the deflections at the spindle
nose arrived from analytical and experimental approach. It can be observed that the
difference tends to reduce as the externally applied force increases. This is due to
the stiffness of the spindle bearing system which resists deflecting the spindle nose
more at lesser applied force. The spindle behavior is similar to that of its operating
conditions, when it performs machining operations along a particular axis. It can
be observed from Sl. No. 5 and onwards in Table 14.1, where the differences are
below 10%. The spindle nose deflections are more realistic after this point. A plot of
external force versus displacement is plotted as shown in Fig. 14.5a, representing the
comparison of displacement values between the analytical and experimental results.
FEA results of the spindle bearing system are used for total deformations.
Table 14.1 shows the results for deflection at the spindle nose using a numerical
approach. These results are compared with experimental values which are tabulated
in Table 14.1. A plot of external force versus displacement is plotted as shown in
Fig. 14.5b, representing the comparison of displacement values between FEA and
experimental results. Study is performed to estimate the deflection of the spindle
Table 14.1 Comparison between analytical, experimental and FEA results

Sl. External Deflection, δ, at spindle nose (μm) Percentage difference
No force F Analytical Experimental FEA Analytical FEA and
(N) result result result and experiment
experiment
1 200 1.23 0.50 1.24 59.34 59.82
2 392 2.42 1.20 2.43 50.41 50.80
3 595 3.68 2.37 3.70 35.59 35.96
4 788 4.87 3.87 4.90 20.53 21.08
5 984 6.08 5.47 6.10 10.03 10.33
6 1194 7.38 7.03 7.41 4.74 5.14
7 1386 8.57 8.63 8.62 0.70 0.05
8 1543 9.54 10.17 9.60 6.60 5.91
9 1778 11.0 11.67 11.06 6.09 5.46
10 1968 12.2 13.37 12.24 9.59 9.16
Fig. 14.5 a Comparison of analytical versus experimental results. b Comparison of FEA versus
experimental results
Table 14.2 Deflection results

Sl. No. External force Deflection δ (μm)
for spindle nose and tool
F (N) At spindle nose At tool holder
holder
1 200 0.50 2.03
2 392 1.20 4.63
3 595 2.37 7.87
4 788 3.87 11.07
5 984 5.47 14.40
6 1194 7.03 17.80
7 1386 8.63 21.13
8 1543 10.17 24.23
9 1778 11.67 27.50
10 1968 13.37 30.77
bearing system when a tool holder is attached to the system. Comparisons between
experimental deflection at spindle nose and tool holder are tabulated in Table 14.2.
It is noticed that there is a considerable difference in deflection results between
both the conditions. Hence, it is also important to evaluate the stiffness of the spindle
bearing system with an attached tool holder.
14.5.2 Stiffness Results
The real-time deflections are considered from the point where the difference in results
is within 10% (Sl. No. 5 in Table 14.1), and the mean static stiffness of the spindle
bearing system for the set of force and deflection values is evaluated using the relation
in Eq. (14.4) and tabulated in Table 14.3. Since the current work concentrates on the
deflections in the spindle nose, the deflection at tool holder is not considered for
Table 14.3 Static stiffness of

Methodology Static stiffness of spindle
spindle bearing system
bearing system (N/μm)
Analytical 162.92
Finite Element Analysis 160.80
Experimental (At spindle 166.53
nose)
further calculations. The experimental static stiffness of the spindle bearing system
considering attached tool holder is compared with the static stiffness of the spindle
bearing system without tool holder.
From Table 14.3, it is observed that the difference between the static stiffness
results at the spindle nose is 2.83% between analytical and experimental results and
3.44% between FEA and experimental results. This indicates that various approaches
used, as shown in Table 14.3, for the stiffness analysis yield similar results. Thus,
the analytical and FEA results are validated with the experimental values. The static
stiffness of the spindle nose reduces due to joint stiffness between the spindle nose
and the tool holder when it is attached with tool holder.
14.6 Conclusions
The machining accuracy and working performance of a machine tool directly depend
upon tool deformation under operating condition. This deformation depends upon
the stiffness of the spindle, and the evaluation of this stiffness is very important. In
this paper, a methodology to arrive at the spindle bearing stiffness by identifying the
force components acting on the system is proposed. The deflection of the spindle
nose is evaluated analytically, numerically and validated with experimental data.
The differences in total deformations between the three methods are within 10%.
The stiffness of the spindle bearing system is evaluated using the three approaches.
The difference in results between analytical and experimental approaches is 2.83%
and between numerical and experimental results is 3.44%. Deflections from the tool
holder are also measured experimentally. The stiffness of the spindle unit at tool
holder point reduces to 72 N/μm due to joint stiffness between the tool holder and
spindle. Future studies to identify the joint stiffness between the spindle nose and
the tool holder need to be carried out.
Acknowledgements Authors are thankful to the member organization (Dr. Abdul Kalam Center for
Innovation, Bharat Fritz Werner Ltd., Bengaluru, and M. S. Ramaiah University of Applied Sciences,
Bengaluru) for supporting this work and Advanced Machine Tool Testing Facility (AMTTF) for
supporting with the experimental setup and measurements.
References
1. Shuzi, Y.: A study of static stiffness of machine tool spindles. Int. J. Mach. Tool Des. 21(1),
23–40 (1980)
2. Sarenac, M.: Stiffness of machine tool spindle as a main factor for treatment accuracy. Sci. J.
Facta Univ. 1(6), 665–674 (1999)
3. Asimkutlu.: Design and development of a lathe spindle. Master of science thesis, Stockholm
(2016)
4. Lubomír, S.: Radial ball bearings with angular contact in machine tools. IntechOpen Ltd., London
(2012)
5. Prakosa, T., Wibowo, A., Llhamsyah, R.: Optimizing static and dynamic stiffness of machine
tools spindle shaft for improving machining product quality. J. Kones Powertrain Transp. 20(4),
363–370 (2013)
6. IBC High precision locknuts. http://www.ibc-waelzlager.com. Last accessed 24 May 2018
7. NSK.: Machine tool spindle bearing selection and mounting guide, England (2009)
Chapter 15
Effect of Surface Corrosion of Steel
Reinforcement on Bond Strength
Characteristics of Concrete
Vineet Kothari and Hemanth Kamplimath
Abstract Corrosion is the degradation of a metal by an electrochemical reaction

with its surrounding environment. The corrosion of embedded steel reinforcement in
concrete causes durability issue in reinforced concrete structures and cement concrete
pavements. The impact of corrosion of embedded steel rebars on the performance of
reinforced concrete structures and pavements is a subject of concern for the engineers
and scientists working in the construction industry. One of the effective ways of
studying such influence is to understand the effect of corrosion on the mechanical
and durability properties of concrete. The bond strength between steel rebar and
concrete is also an important parameter of study. This study focuses on probing the
effect of corrosion on bond strength and other mechanical properties of reinforced
concrete. A pull-out test was conducted on 18 concrete cube specimens having steel
bars of 12 and 16 mm diameter Fe500 grade. These bars were corroded with the
help of commercial grade hydrochloric acid solution 10% v/v. Each group of bar
specimens included three levels of corrosion: 0, 3 and 5%. Tension test on steel bars
and compression test for concrete cubes were conducted.
Keywords Reinforced concrete · Rebar corrosion · Bond strength · Pull-out test
15.1 Introduction
Corrosion is the degradation of a metal by an electrochemical reaction with its

surrounding environment. This is an undesirable process which causes premature
failure of structures. The steel bars may get corroded during its stages of service
life such as production, storage, construction and also during the service life of the
structure [1]. Carbonation and increase in the chloride ionic concentration in concrete
are the two main factors which severely corrode the embedded steel in concrete
V. Kothari · H. Kamplimath (B)

Nirma University, Ahmedabad, India
V. Kothari
https://doi.org/10.1007/978-981-15-3254-2_15
154 V. Kothari and H. Kamplimath
beyond acceptable limits [2]. Durability properties of concrete such as porosity, pore
size distribution and permeability affect the rate of corrosion of steel.
Corrosion of steel in concrete is a complex phenomenon and has a higher impact
on the mechanical properties of concrete especially its strength. The bond strength
depends upon bar geometry, surface condition of the bar, presence of confinement
and the properties of concrete [3]. The corrosion of reinforcement causes an increase
in its volume due to the increase in the diameter of the steel bar initially. This in
turn leads to splitting and weakened bond between steel and concrete. This affects
the serviceability as the durability of the structure is decreased [4]. Corrosion of
steel reinforcement causes a reduction in yield strength and bond strength due to
the delamination of rust elements formed on the surface of steel. Corrosion may
lead to reduction in the ductility of rebar. The ratio of yield strength to the ultimate
strength of the rebar at maximum load is reduced. This reduction leads to early
failure of the rebar [5]. Bond strength between rebar and concrete does not have any
standard quantitative definition. Pull-out test is adopted to compare the bond strength
of various types of rebar with concrete [6].
15.2 Research Methodology
This study focuses on some of the factors affecting the serviceability of RC structures
such as
1. Calculating the mechanical properties of corroded rebar.
2. Correlation between rebar corrosion levels and bond strength of reinforced
concrete.
This study focuses on the acceptability of the results of bond strength for various
levels of rebar corrosion, and thus, helping site engineers in deciding the use of
surface corroded steel rebar on the construction site.
15.2.1 Level of Corrosion
The main significance of this study is to assess the effect of rebar corrosion on
the bond characteristics between rebar and concrete. The effect of corrosion on the
mechanical properties of rebar is also assessed. Table 15.1 represents various material
Table 15.1 Material

Material properties Characteristics
properties
Strength of concrete M25
Bar size 12 and 16 mm
Level of corrosion 0, 3 and 5%
15 Effect of Surface Corrosion of Steel Reinforcement … 155
Table 15.2 Level of

Material properties Characteristics
corrosion for trial test
Length of reinforcing bar 400 mm
Initial weight of reinforcing bar 0.179 kg
Weight of reinforcement bar after 6 days 0.148 kg
Reduction in weight 0.031 kg
Percentage reduction in weight 17%
parameters of steel and concrete considered for the pull-out test. Table 15.2 represents
the levels of the corrosion considered for a trial test. In the present study, the pull-out
test was conducted for two levels of rebar corrosion and compared with a corrosion
less specimen.
For the trial test, 400 mm long rebar specimen was completely submerged in
10% v/v solution of commercial grade hydrochloric acid. It was possible to achieve
17% corrosion in just 6 days. Details of the test conducted are given in Table 15.2.
Corroded rebars were prepared in a laboratory by submerging 170 and 400 mm
long rebar in 10% v/v solution of commercial grade hydrochloric acid. It was possible
to achieve 17% corrosion in just 6 days. Details of the test conducted are given below.
The above mentioned method is suitable for corroding rebar up to the desired
limit. To study the effect of corroded rebar on the mechanical properties of concrete
and its bond strength, two sizes of rebars (12 and 16 mm diameter) having a standard
length of 0.8 m have been tested. Figure 15.1 shows 12 mm rebar for the three levels
of corrosion. The 12 and 16 mm diameter bars were corroded up to a length of
170 mm by submerging it in 10% v/v HCl solution. This is due to the fact that the
corroded specimen is inserted into specially made cube molds of size 150 mm ×
150 mm × 150 mm as shown is Fig. 15.2. Table 15.3 provides the details of the
corroded specimens. The rebars are numbered in such a way that B1 represents the
first bar out of twelve bars used, D12 represents the diameter of the bar, i.e., 12 mm,
and L5 indicated the level of corrosion, i.e., 5%. The level of corrosion is calculated
from the equation given below. The length of rebar for all the specimens is 0.8 m,
5% Corrosion
0% Corrosion
3% Corrosion
Fig. 15.1 Twelve mm bars with different corrosion levels

Fig. 15.2 Concrete

specimen prepared for
pull-out test
Table 15.3 Details of corroded rebar specimens

Rebar no. Wt. Initial Embedded Corrosion Final Result (%)
(gm/m) weight length level (%) weight
(gm) (mm) (gm)
B1 D12 L5 891 713 170 5 705.50 4.95
B2 D12 L3 888 712 170 3 707.50 2.98
B3 D12 L5 887 710 170 5 701.67 5.52
B4 D12 L5 891.5 715 170 5 707.65 4.85
B5 D16 L5 1559 1247.6 170 5 1235.10 4.72
B6 D16 L5 1558 1246.7 170 5 1234.24 4.70
B7 D16 L3 1553 1242.3 170 3 1234.80 2.84
B8 D16 L5 1571 1257 170 5 1244.00 4.87
B9 D16 L3 1572 1258 170 3 1249.00 3.37
B10 D16 L3 1568 1267 170 3 1258.00 3.38
B11 D12 L3 883 704.67 170 3 0699.85 3.21
B12 D12 L3 888 711 170 3 0706.4 3.05
and the embedded corroded length is 170 mm.
(Initial Weight of Bar − Final Weight of Bar) × Length of Rebar

% Corrosion = × 100
Embedded Length of Rebar × Inital Weight of Rebar
Table 15.4 Mix

Description of material Quantity per m3 of concrete
proportioning of concrete
Grade of concrete M25
Cement 382 kg
Fine aggregates 712 kg
Coarse aggregates 1140 kg
Water 153 kg
W/C ratio 0.4
Super plasticizer (PCE-based) 0.9% by weight of cement
15.2.2 Mix Proportioning of Concrete
Twelve mm and 16 mm steel rebars were used for the study. M25 grade of concrete
was used. Table 15.4 provides the mix design quantities used for the study.
15.2.3 Fabrication of Specimens for Pull-Out Test
The concrete cubes of size 150 mm × 150 mm × 150 mm and M25 grade of concrete
were prepared and used for the pull-out test as shown in Fig. 15.2. Twelve specimens
were prepared for the pull-out test at different corrosion levels, i.e., 0, 3 and 5%.
15.3 Experimental Program
15.3.1 Compression Strength of Cubes
IS 2770:1967 (PART 1) [7] requires that the average compressive strength of three
cubes representing the concrete used for test specimen made and tested in accordance
with relevant requirements of IS: 516-1959 shall be 200–300 kg/cm2 at the time of
making pull-out tests. The compression testing machine of capacity two thousand kilo
newton was used to evaluate the compressive strength of concrete cube specimens.
Three concrete cubes of size 150 mm were tested in compression as per IS 516 [8].
The equation for calculating the compressive strength of the cube specimen is given
below.

Compressive Strength N/mm2 = P × 103 /A
P = Failure load of cube (KN)
A = Area of cube (150 × 150) mm2
Fig. 15.3 Test setup for

pull-out test of specimen
15.4 Bond Strength of Concrete
Pull-out test is an effective and standard method for comparison of bond

characteristics of corroded rebar with concrete [9]. Concrete cubes of 150 mm ×
150 mm × 150 mm size along with 12 and 16 mm corroded rebar were used to
gauge the bond strength characteristics as per IS: 2770: 1967 (Part 1) [7]. The details
of corroded specimens are provided in Table 15.3. In this test, the load value is noted
at a relative slip of 0.02 mm at the free end of the specimen. The ratio of load value
recorded at failure load versus the surface area of the bar provides the bond strength.
Dial gauges with a least count of 0.01 mm were used for measuring the slip. Dial
gauges are arranged in such a way that it is possible to record the movement of
rebar at both the unloaded and loaded ends. The concrete specimen prepared for the
pull-out test is shown in Fig. 15.2, and the pull-out test setup is shown in Fig. 15.3.
15.5 Tensile Test on Rebar
To study the impact of corrosion on mechanical properties of rebar, tension test has
been performed. The test consists of straining the test specimen by applying tensile
force until fracture. All steel rebar samples were tested to determine the yield and
Fig. 15.4 Test setup for tension test on reinforcing bar
ultimate strength, in accordance with IS: 1608-1995 [10], using a universal testing
machine of 500 KN capacity. The ultimate load and the breaking load are noted.
Stress–strain characteristics are determined to evaluate yield strength of material.
Figure 15.4 presents the test setup for the tension test.
15.6.1 Compressive Strength of Concrete
The 28-day compressive strength of three sets of concrete cubes (a total of nine
cubes) representing the concrete used for pull-out test was evaluated. The average of
the three sets of results is represented in Table 15.5. It has been noted that the average
cube compressive strength is 27 MPa which satisfies the requirements of minimum
compressive strength of pull-out test specimen at the time of test.
Table 15.5 Cube

Cube no. Compressive strength Avg. compressive
compressive strength of
in MPa strength in MPa
concrete
1 25.3 27
2 27.7
3 28.0
Table 15.6 Tensile strength of rebar at various corrosion levels

Corrosion level (%) Yield strength (MPa) Ultimate strength (MPa)
12 mm dia. 16 mm dia. 12 mm dia. 16 mm dia.
0 494.41 571.96 585.49 691.33
3 477.80 539.60 570.60 634.87
5 463.79 524.00 552.70 623.97
15.6.2 Effect of Corrosion on Tensile Strength of Concrete
Tension test has been performed to evaluate the effect of various levels of corrosion
on mechanical properties of steel rebar. The average results of the yield strength and
ultimate strength have been presented in Table 15.6. It was observed that the increase
in corrosion levels decreases the yield and ultimate strength of rebar.
15.6.3 Effect of Corrosion on Bond Strength of Concrete
The bond strength at an age of 28 days has been evaluated for cube specimen prepared
for all three corrosion levels, i.e., 0, 3 and 5%. Average results of three specimens
tested are calculated, and the results are tabulated in Table 15.7.
A graph of average values of bond strength versus the three corrosion levels (0,
3 and 5%) is plotted as shown in Fig. 15.5. The results indicated that the bond
strength of both 12 and 16 mm diameter bars decreased with an increase in corrosion
levels. The average bond strength of 12 mm diameter bar with 3% corrosion level is
10.68 MPa (2.5% decrement), whereas for 5% corrosion level, it is 8.5 MPa (22%
decrement). For 16 mm diameter bars, average bond strength value for 3% corrosion
level is 11.47 MPa (3% decrement) and 9.76 MPa (19% decrement) for 5% corrosion
level. Up to 3% corrosion level, the decrease in bond strength is minor. However, for
5% corrosion level, there was a drastic decrease in bond strength up to 15–20% for
both rebars of size 12 and 16 mm. Figure 15.5 shows the effect of corrosion on bond
strength for both types of reinforcing bars.
The bond strength of the deformed bars may have decreased due to reduction in
mechanical resistance to slip because of decay in twirls provided on TMT bars up
to 5% corrosion level. Due adequate bond capacity, bond failure due to pulling of
Table 15.7 Effect of

Corrosion level (%) Average bond strength (MPa)
corrosion on bond strength of
concrete 12 mm dia. 16 mm dia.
0 10.96 11.84
3 10.68 11.47
5 8.50 9.76
Fig. 15.5 Bond strength versus corrosion level of rebar
Fig. 15.6 Failure pattern of pull-out test specimen for TMT bars
the bar did not occur. The surrounding concrete which was subjected to excessive
circumferential tensile stress failed by splitting for all the pull-out test specimens.
The interacting force between the deformed bar and the surrounding concrete may
have led to the splitting of the concrete into two segments as shown in Fig. 15.6.
15.7 Conclusion
Method of corroding reinforcing bars in laboratory by submerging bars into 10% v/v
solution of commercial grade hydrochloric acid is suitable to stimulate actual field
scenario as well as long-term serviceability scenario quickly. The tensile strength of
the steel bar decreases with the increase of corrosion level. Variation of modulus of
elasticity of the rebar does not change significantly as there is no much effect of the
corrosion level on the initial linearity of the stress–strain curve. With the increase in
the corrosion level of the rebar, the yield strength and ultimate strength of the steel
decrease by approximately 5–10% as compared with the standard rebar. The bond
strength of the corroded rebar decreases slightly for 3% corrosion level of rebar and
marginally for 5% corrosion level of rebar.
References
1. Huang, C.-H.: Effect of rust and scale of reinforcing bars on the bond performance of
reinforcement concrete, ASCE J. Mater. Civ. Eng. 26(4), 576–581 (2014)
2. Ahmad S: Reinforcement corrosion in concrete structures, its monitoring and service life
prediction—a review. Cem Concr Compos 25(4–5), 459–471 (2003)
3. ACI 222-R19: Guide to protection of reinforcing steel in concrete against corrosion. American
Concrete Institute, Farmington Hills, MI, USA (2019)
4. Cabrera JG: Deterioration of concrete due to reinforcement steel corrosion. Cement and
Concrete Composites 18(1), 47–59 (1996)
5. Mansoor, Y.A, Zhang, Z.Q.: The reinforcement bond strength behavior under different
corrosion condition. Res. J. Appl. Sci., Eng. Technol. 5, 2346–2353 (2013)
6. IS 2770 (part-1:1967): Method of testing bond in reinforced concrete part-1 pull out test
7. Shetty, A., Venkataramana, K., Gogoi, I., Praveen, B.B.: Performance enhancement of TMT
rebar in accelerated corrosion. J. Civ. Eng. Res. 2, 12–14 (2012)
8. IS-10262 (2009): Concrete mix proportion guidelines, Bureau of Indian Standards, New Delhi
(2009)
9. Tondolo, F.: Bond behaviour with reinforcement corrosion. J. Constr. Build. Mater. 93, 926–932
(2015)
10. IS: 1608-1995: Mechanical testing of metals, Bureau of Indian Standards, New Delhi (1995)
11. IS: 516-1959: Methods of tests for strength of concrete, Bureau of Indian Standards, New Delhi
(1959)
Chapter 16
Stress Distribution in an Infinite Plate
with Circular Hole by Modified Body
Force Method
Shrikrishna Badiger and D. S. Ramakrishna
Abstract Stress distribution in an infinite plate with circular hole subjected to uni-
form tension is determined by employing a modified body force method. In this
method, the problem of a plate with a hole under uniform tension is considered as a
plate with an imaginary hole. The boundary of the imaginary hole is divided into a
number of divisions. At the mid-point of each division, concentrated forces known
as body forces are applied. The magnitudes of these body forces are computed from
complex potential functions, and stress at an arbitrary point is obtained by the summa-
tion of stresses due to these body forces applied at the mid-point of each division and
stresses due to applied load. Results obtained from the modified body force method
show trends in line with theoretical results. However, more accurate results can be
obtained by using better estimate of body forces which satisfy boundary conditions
at the circular hole. Setting Poisson’s ratio ν = 0 has little effect on the computed
stress distribution.
Keywords Body force method · Boundary force method · Stress concentration
16.1 Introduction
The solution to the problem of stress distribution in an infinite plate with circular
hole subjected to uniaxial loading was first obtained by Kirsch [1]. The details of the
analytical solution are found in Timoshenko [2] and Wang [3]. Complex variables
approach was first introduced into plane elastic problems in 1909 by Kolosov [4, 5],
which was further utilised in solving various problems in elastostatics by Muskhel-
ishvili [6] and others. With improvements in computer performance, numerical tech-
niques like finite element method (FEM) and boundary element method (BEM)
S. Badiger (B)
Ramaiah University of Applied Sciences, Bengaluru, India
D. S. Ramakrishna
Jawaharlal Nehru National College of Engineering, Shivamogga, India

https://doi.org/10.1007/978-981-15-3254-2_16
164 S. Badiger and D. S. Ramakrishna
became popular methods of solving various engineering problems. However, FEM

and BEM have the limitation of increased computing cost and accuracy when discon-
tinuities like holes and cracks are present in the material. Body force method (BFM)
originally proposed by Nisitani [7, 8] is a boundary type technique for stress analysis.
The software program named BFM2D, developed by Nisitani [8] and based on BFM
is reported to provide accurate results even with coarse mesh. More recently, Man-
junath and Ramakrishna have applied BFM and solved various problems including
Flamant and Melan problem [9–11].
16.2 Body Force Method
Body force method is a boundary type technique based on the principle of super-
position. For a plate with circular hole subjected to uniform tension as shown in
Fig. 16.1a, the circular hole is considered as imaginary hole, which is divided into a
number of divisions (imaginary hole is divided into four divisions in Fig. 16.1b), at
the mid-point of each division, body forces ρxi , ρ yi are applied in x- and y-directions,
the magnitudes of which are unknown to begin with. Applying equilibrium condition
to each division in x- and y-directions due to the body forces and external applied
load, we obtain a set of linear equations which can be expressed in the form, Ax = b
where A is influence coefficient matrix obtained from Kelvin’s problem (point load
in an infinite plane) where unit loads are applied in x- and y-directions, x is the
unknown body force vector and b is a vector consisting of forces on each division
in x- and y-directions due to external applied load. The stress at an arbitrary point
P(x, y) (Fig. 16.1a) is obtained by the summation of stresses at point P(x, y) due
to body forces ρxi , ρ yi and stresses due to the applied uniform load as shown in
Fig. 16.1b.
(a) (b)
Fig. 16.1 Plate with a circular hole subjected to uniform tension and its equivalence in BFM
16 Stress Distribution in an Infinite Plate … 165
16.3 Modified Body Force Method
Body forces ρxi , ρ yi can also be computed from complex potential functions which
describe the given loading and boundary conditions. The complex potential functions
for an infinite plate subjected to uniform tensile stress in x-direction are well known
[10]. In this study, body forces are obtained from complex potential function given
by Honein [10]. These body forces are applied at the mid-point of each division. The
stress at an arbitrary point P(x, y) is obtained by the summation of stresses due to
the computed body forces and stresses due to the applied load.
Stresses are related to complex potential functions ϕ(z) and ψ(z) [2, 3, 5, 6] by
the following relations:

σx + σ y = 2 ϕ (z) + ϕ (z) = 4 ϕ (z) (16.1)

σ y − σx + 2iτx y = 2 z̄ϕ (z) + ψ (z) (16.2)
Forces acting on an arc from A to B are related to complex potential functions by

the following relations [12]:
B
Px + Py = −i ϕ(z) + z · ϕ (z) + ψ(z) A (16.3)
Complex potential functions for an infinite plate subjected to uniform tensile stress
in x-direction are as follows [10]:
z
ϕ(z) = · σ∞ (16.4)
4 x
z
ψ(z) = − · σx∞ (16.5)
2
Complex potential functions ϕ(z) and ψ(z) from Eqs. (16.4) and (16.5) when
substituted in Eqs. (16.1) and (16.2), the following stresses are obtained:
σx = σx∞ (16.6)
σy = 0 (16.7)
τx y = 0 (16.8)
Equations (16.6), (16.7) and (16.8) accurately describe the state of stress at
arbitrary point P(x, y) in an infinite plate without any discontinuity.
Complex potential functions from Eqs. (16.4) and (16.5) when substituted in
Eq. (16.3), the following forces are obtained:
Table 16.1 Body force values for four divisions of the imaginary circle
Body forces Px1 Py1 Px2 Py2 Px3 Py3 Px4 Py4
Numerical values 500 0 −500 0 −500 0 500 0
Px = σx∞ · [y B − y A ] (16.9)
Py = 0 (16.10)
Body forces are computed for an infinite plate with unit thickness having a circular
hole of 5 mm radius subjected to uniform tensile stress of 100 MPa. Table 16.1 shows
numeric values of the body forces obtained from Eqs. (16.9) and (16.10), for four
divisions of the imaginary circle.
16.4 Theoretical Results
In an infinite plane with thickness h, the stress in polar coordinates at an arbitrary

point z = x +i · y due to concentrated load Px +i · Py acting at location z = x0 +i · y0
is reproduced from Wang [13].

(3 + ν) Px cos θ + Py sin θ
σr = − (16.11)
4π h r

(1 − ν) Px cos θ + Py sin θ
σθ = (16.12)
4π h r

(1 − ν) Px sin θ − Py cos θ
τr θ = (16.13)
4π h r
Stress at arbitrary point P(x, y) is obtained by summing stresses (Eqs. 16.11–

16.13) due to body forces obtained for Eqs. (16.9) and (16.10) and stresses due to
applied load.
The expression for stresses in polar coordinates for an infinite plate with circular
hole (Fig. 16.1a) is available in Timoshenko [12]. These equations are reproduced
here from [12].

σx∞ a2 σ∞ 3a 4 4a 2
σr = 1 − 2 + x 1 + 4 − 2 cos 2θ (16.14)
2 r 2 r r

σ∞ a2 σ∞ 3a 4
σθ = x 1 + 2 − x 1 + 4 cos 2θ (16.15)
2 r 2 r

σx∞ 3a 4 2a 2
τr θ =− 1− 4 + 2 sin 2θ (16.16)
2 r r
16.5 Numerical Results
A plate with unit thickness having a circular hole of 5 mm radius is considered with
an applied uniform tensile stress of 100 MPa. The radial, hoop and shear stresses
along radial direction at angles 0°, 45° and 90° are computed using body force
Eqs. (16.9) and (16.10) (legend BFM) and are compared with analytical results
(Eqs. 16.14–16.16).
Figures 16.2, 16.3 and 16.4 show stresses along 0° radial line. Radial and hoop
stresses show trends in line with the theoretical results.
Figures 16.5, 16.6 and 16.7 show stresses along 45° radial line. Radial, hoop and
shear stresses show trends in line with the theoretical results. However, it is observed
that the magnitudes of these stresses in the proximity of the circular hole deviate
from theoretical values.
Figures 16.8, 16.9 and 16.10 show stresses along 90° radial line. Radial and hoop
stresses show trends in line with the theoretical results.
Fig. 16.2 Radial stress (σr ) along radial line making 0° with x-axis, and circle is divided into 1024
divisions
Fig. 16.3 Hoop stress (σθ ) along radial line making 0° with x-axis, and circle is divided into 1024
divisions
Fig. 16.4 Shear stress (τr θ ) along radial line making 0° with x-axis, and circle is divided into 1024
divisions
Fig. 16.5 Radial stress (σr ) along radial line making 45° with x-axis, and circle is divided into
1024 divisions
divisions
Fig. 16.7 Shear stress (τr θ ) along radial line making 45° with x-axis, and circle is divided into
1024 divisions
Fig. 16.8 Radial stress (σr ) along radial line making 90° with x-axis, and circle is divided into
1024 divisions
divisions
Fig. 16.10 Shear stress (τr θ ) along radial line making 90° with x-axis, and circle is divided into
1024 divisions
Figures 16.11, 16.12 and 16.13 show radial, hoop and shear stresses along 0°
radial line. Here, circle is divided into 1024 divisions, with Poisson’s ratio (a) ν =
0, (b) ν = 0.32. It is observed that taking Poisson’s ratio as zero has little effect on
Fig. 16.11 Radial stress along radial line making 0° with x-axis, circle is divided into 1024 divisions,
a ν = 0, b ν = 0.32
Fig. 16.12 Hoop stress along radial line making 0° with x-axis, circle is divided into 1024 divisions,
a ν = 0, b ν = 0.32
Fig. 16.13 Shear stress along radial line making 0° with x-axis, circle is divided into 1024 divisions,
a ν = 0, b ν = 0.32
the radial and shear stresses. With nonzero Poisson’s ratio, hoop stress shows little
improvement in its trend towards theoretical results.
16.6 Conclusion
In this study, stress distribution is obtained by summation of stresses due to body

forces derived from complex potential functions for a plate with uniform load in
x-direction and stresses due to applied load (modified body force method). Results
obtained from this method show trends in line with theoretical solution. However,
numerical values obtained deviate from the theoretical solution. One of the possible
reasons for this deviation in numerical values is that the boundary of the imaginary
circle being stress-free may not be completely satisfied. Better results can be obtained
when accurate estimates of the body forces on the boundary of the imaginary circle
are made. Alternately, body forces obtained from a set of linear equations Ax = b
(body force method) can be used to obtain the stress distribution.
Modified body force method may be useful in the instances where theoretical
results are unavailable and this method can give values of stresses approaching actual
stresses.
As suggested by Nisitani [14], it is acceptable to take Poisson’s ratio ν = 0 since
the marginal improvement in computed stresses is observed when nonzero Poisson’s
ratio is used.
Acknowledgements We would like to thank Vivek H Gupta, Arun R Rao and Amit Lal for helpful
discussions and suggestions.
References
1. Kirsch, E.G.: Die Theorie der Elastizität und die Bedürfnisse der Festigkeitslehre. Zeitschrift
des Vereines deutscher Ingenieure 42, 797–807 (1898)
2. Timoshenko, S.P., Goodier, J.N.: Theory of Elasticity, 3rd edn. McGraw-Hill, New York (1970)
3. Wang, C.-T.: Applied Elasticity, pp. 171–208. McGraw-Hill, New York (1953)
4. Kolosov, G.V.: On an application of complex function theory to a plane problem of the
mathematical theory of elasticity. Doctoral thesis, Yuriev (1909)
5. England, A.H.: Complex Variable Methods in Elasticity, Dover edn. Dover Publications,
Mineola, N.Y. (2003)
6. Muskhelishvili, N.I.: Some Basic Problems of the Mathematical Theory of Elasticity, 2nd
edn. Noordhoff International Publishing, Leyden (1977). https://doi.org/10.1007/978-94-017-
3034-1
7. Nisitani, H., Saimoto, A.: Short history of body force method and its application to various
problems of stress analysis. Mater. Sci. Forum 440–441, 161–168 (2003). https://doi.org/10.
4028/www.scientific.net/MSF.440-441.161
8. Nisitani, H., Saimoto, A.: Effectiveness of two-dimensional versatile program based on body
force method and its application to crack problems. Key Eng. Mater. 251–252, 97–102 (2003).
https://doi.org/10.4028/www.scientific.net/KEM.251-252.97
9. Manjunath, B.S., Ramakrishna, D.S.: Body force method for flamant problem using complex
potentials. In: ASME Engineering Systems Design and Analysis, Volume 4: Fatigue and Frac-
ture, Heat Transfer, Internal Combustion Engines, Manufacturing, and Technology and Society,
pp. 99–103 (2006). https://doi.org/10.1115/esda2006-95303
10. Manjunath, B.S., Ramakrishna, D.S.: Body force method for melan problem with hole using
complex potentials. In: ASME International Mechanical Engineering Congress and Exposition,
Volume 10: Mechanics of Solids and Structures, Parts A and B: pp. 861–865 (2007). https://
doi.org/10.1115/imece2007-42885
11. Manjunath, B.S.: Body force method in the field of stress analysis. Ph.D. thesis, Visvesvaraya
Technological University, Belgaum (2009)
12. Nisitani, H., Chen, D.: Body force method and its applications to numerical and theoretical
problems in fracture and damage. Comput. Mech. 19(6), 470–480 (1997). https://doi.org/10.
1007/s004660050195
13. Honein, T, Herrmann, G.: The Involution correspondence in plane elastostatics for regions
bounded by a circle. J. Appl. Mech. 55(3), 566–573 (1988). https://doi.org/10.1115/1.3125831
14. Nisitani, H.: Stress analysis of notch problem by body force method. In: Sih G.C. (ed.)
Mechanics of Fracture 5, Sijthoff & Noordhoff, Chapter 1, pp. 1–68 (1978)
15. Chen, D., Nisitani, H.: Int. J. Fract. 86(1–2), 161–189 (1997). https://doi.org/10.1023/A:
1007337210078
Chapter 17
Wind Load Analysis of a Tall Structure
with Sharp and Corner Cut Edges
Abhipsita Rej and Amlan Kumar Bairagi
Abstract The massive wind load is always harmful to tall building structure, espe-
cially the structure which has sharp edges. The separation of wind near the sharp
edge creates local effects around the neighbor faces of the building and also creates
enormous damage on that region. The present study illuminated the pressure varia-
tion around the sharp edge setback and corner cut setback tall building models. The
models are inducted inside the computational domain and simulated with different
wind angles. One model has a sharp edge corners and other models has a corner cut
at the bottom part of setback zone and rounder corner at top part of setback zone.
The considerable amount of pressure, drag and lift variation has been detected for
across- and along-wind consideration. The tremendous amount of pressure fluctua-
tion observed at corner cut region. The local pressure recognized on the wall due to
sharp edges for the sharp edge model. However, the local pressure is minimized by
the use of rounded edges at the top portion of the setback. The drag and lift coeffi-
cient decrease with the decrease of sharpness of edges. The study tries to catch the
minimum local pressure due to corner cut edges of the tall building model.
Keywords Sharp and corner cut edge · Set back tall building · Pressure variation
17.1 Introduction
Wind effects are the challenges that designers have to deal with in super-tall building
design. In association with high slenderness, low natural frequencies, low inherent
damping level and high wind speed at upper lever, super-tall buildings are more
susceptible to wind. Wind pressures on building surfaces result in both steady and
A. Rej (B)
Department of Civil Engineering, Techno India University, Salt Lake, Kolkata, India
A. K. Bairagi

https://doi.org/10.1007/978-981-15-3254-2_17
176 A. Rej and A. K. Bairagi
unsteady loading, while air motion over and around buildings transports heat, mass
and momentum. In this study, the method of corners cutting and setback condition
is effective for the reduction of wind-induced vibrations in an across-wind, direction
for high-rise buildings with a regular square plane. Wind forces acting on high-rise
buildings with corner cuts in along-wind direction were measured by wind tunnel
tests. During this study, Kawai [1] gave the concept of the corner modifications
promote the instability at low speed. Tanaka et al. [2] determined the aerodynamic
forces and wind pressures acting on the square-plan tall building models with various
configurations like corner cut, setbacks and helical. Elshaer et al. [3] examined the
aerodynamic shape optimization (ASO) example is to reduce the drag force acting
on a tall building by changing the shape of its corners. Roy and Bairagi [4] illustrated
the pressure variation and velocity around the multiple shape setback tall building.
Mooneghi and Kargarmoakhar [5] reviewed the past/recent work on various aero-
dynamic mitigation techniques developed for reducing wind loads on buildings by
modifying their shapes and/or adding simple architectural elements. Bairagi and
Dalui [6, 7] conferred the pressure fluctuation at top roof of setback model and also
highlighted the power spectral density (PSD) at neighbor faces of the setback and
top roof. Alminhana et al. [8] modified corner cuts, reduce significantly the aerody-
namic forces on the building structures and improve flow conditions near the building
locations. The current research is based on the analytical method by computational
fluid dynamics (CFD) where an isolate setback tall building is set inside the domain
to estimate the dynamic behavior of setback corner cut tall building due to wind
excitation. Horizontal pressure coefficient, drag and lift coefficients are observed for
along different wind directions.
17.2 Description of Meshing and Model
In this study, the domain has been created using the concept of Frank et al. [9]
according to which the inlet and the lateral and top boundary of the domain should
be apart for 5 times of the overall height (5H) from the furthest point of the model
on that side. The outlet of the domain should be at a distance of 15 times the overall
height (15H) from the furthest point of the model on that side. The domain details are
shown (see Fig. 17.1a). The meshing has been done to calculate the solution while
the run is ongoing, in such a manner that the mesh can automatically be refined in
locations that depends on specified adoption criteria, where solution variables are
hanging most rapidly so that it can resolve the flow features in these regions. To
avoid unusual flows inside the domain, tetrahedron meshing has been used and it is
inflated near the boundary. By the use of this mesh, the regions near the boundaries
can be made so inflated that the flows with unusual characteristics can be avoided.
The meshing details are shown (see Fig. 17.1b).
The analytical model has been simulated by computational fluid dynamics (CFD)
method. Two setbacks tall building model has same length: breadth ratio (1:1) and
length: height ratio (1:4) has been adopted in this study. The models are square-plan
17 Wind Load Analysis of a Tall Structure with Sharp and Corner … 177
Fig. 17.1 a Model inside the domain and b meshing of corner cut model
shape and setback at half of the model height. The sharp edge model, namely M1 has
setback level H/2 and setback distance 0.2L from both the opposite edges. Similar
pattern model M2 has been considered but has a corner cut at the bottom part and
the curve cut at the top part. The corner cut size has been demarcated by D1 and
curved portion is R1. The attacking wind angle is 0° for along-wind direction and
90° for across-wind condition and an intermediate wind angle 45° also considered
to observe the deviation of flow pattern around the model. The detail description of
model M1 and M2 are shown (see Fig. 17.2a–b).
Fig. 17.2 Dimension of a model M1, b model M2 and c profile of velocity and turbulence inside
the domain
17.3 Boundary Condition of the Domain
This study is conceived and analyzed by creating virtual domain and simulate the
wind flow pattern around the model. The simulated flow conditions have been gen-
erated as per the criteria of terrain category two taken from IS 875 (part 3) [10]. The
magnitude of inlet velocity has been considered normal to the boundary. Following
the requirements of directional constraints, the flow direction was considered parallel
to the normal of the surface of the boundary. The inlet and outlet are specified in the
virtual domain. The sidewalls of this virtual domain are specified to be as free-slip
wall in this study. The walls of both of the models are assumed no-slip walls with
no surface roughness. In this study, power law has been used which is shown in
Eq. (17.1), to calculate the velocity profile of the atmospheric boundary level in this
CFD-based study.
V /Vz = (Z /Z h )α (17.1)
where V represents the speed of wind in horizontal direction at a certain elevation.

V Z represents the speed of the wind at a reference elevation Z h . The value of V z is
taken as 10 m/s. The value of Z h is taken as 1.0 m. α represents the coefficient of
surface roughness 0.133 for terrain category 2.
17.4 Acceptance of Present Study
The numerical simulation used in the computational domain also validated with
different researchers used by experimentally and analytically. The experimental study
on sharp edge setback tall building was presented by Kim and Kanda [11] and
conducted by Eiffel-type wind tunnel (1.8 m × 1.8 m × 12.5 m) at the University
of Tokyo. The adopted factors are 1:400 scale, height of model 160 m, breadth of
model 40 m and the power law exponent was 0.13, and the turbulence intensity was
15%. Wind speed at the tunnel was 6.5 m/s with 1.2% blockage ratio. Same size and
same factors are considered and presented by Bairagi and Dalui [12, 13]. The present
study also considered the same conditions for corner cut model and presented the
profile of velocity and turbulence (see Fig. 17.2c).
17.5 Results
Two unconventional square setback tall building model, namely M1 and M2 have
been simulated in a computational domain for 0°, 45° and 90° attacking wind. The
pressure variation, drag and lift coefficients are observed and presented in this section.
17.5.1 Pressure Variation Around the Models
The pressure contours at different faces of both the models are conducted by using
the formulas as stated in Eq. 17.2 and presented in Table 17.1.

C p = C p(cal) / 0.5 ρVz2 (17.2)
where C p is the pressure coefficient, C p(cal) is the pressure coefficient from the ana-
lytical data, ρ is the density of fluid at 25 °C and V z is the velocity of wind at the
particulate height. For along-wind condition, the windward face has positive pressure
for both the models and suction at the sidewalls and leeward face. An interesting point
comes out from the local pressure zone. The local pressure zone observed in the wall
and to roof of setback part of model M1, but model M2 has minimum amount of wall
and negligible amount of top roof. Therefore, it may be said that the local pressure
depends upon the sharpness of the edge. The top setback part of the M1 model has
a sharp edge, but the model M2 has curved edges. According to this condition, the
flow pattern around the sharp edge of the model M1 has high amount of turbulence
and create tremendous suction. On the other hand, the model M2 has curved edges
at the top setback part, so the wind passed smoothly on that particular edge.
To understand the proper pressure fluctuation, horizontal pressure variation carried
out in this study. A pressure belt considered at height 0.475H from the base of both
the model M1 and M2. The initial pressure tapping point considered at the corner
of face A and assumed a pressure belt around the model. The pressure variation of
different azimuth is presented in (see Fig. 17.3). For 0° wind angle, the initial point
at the corner of face A has 3.82% difference of suction with respect to M1 model.
The high amount of suction detected at the corner cut region of M2 model as shown
in Fig. 17.3a. In this zone, the pressure difference is 163% less than the M1 model.
According to this analysis, it is clear that the large amount of suction developed at
the corner cut region. At the same time, for 45° wind has less pressure fluctuation at
the corner cut zone compare with M1 and M2 model. The similar amount of pressure
difference observed for 90° wind angle.
17.5.2 Drag and Lift Coefficient
The drag and lift coefficient for both the models are calculated by using the equation
as stated in Eqs. 17.3–17.4
Fd = 0.5 ρvz2 Cd A (17.3)
Fl = 0.5 ρvz2 Cl A (17.4)

Table 17.1 Comparison of pressure coefficient between M1 and M2 model for 0°, 45° and 90°
wind angles
Wind Model Legend Face A Face B Face C Face D Roof
angle
M1
0°
M2
M1
45°
M2
M1
90°
M2
Fig. 17.3 Azimuth of pressure coefficient at 0.475H for a 0° wind, b 45° wind and c 90° wind
angle
where F d and F l are drag and lift force, A is the attacking wind area of the model,
C d and C l are the drag and lift coefficient. The comparative study of drag and lift
force coefficient between M1 and M2 model showcased under (see Fig. 17.4). The
maximum drag coefficient for model M1 is 1.06 for 0° wind and gradually decreasing
with the increase of the wind angle. For 45° and 90° wind angle has 0.56 and −0.03,
respectively, for M1 model as shown in (see Fig. 17.4). On the other hand, model M2
has 0.34, 0.23 and zero for the wind angle 0°, 45° and 90°. The lift coefficient values
also vice versa, and the values also increasing with the increase of the wind angle.
For 0°, the wind has minimum value and 90° has maximum value 1.17 for the M1
model as shown in (see Fig. 17.4). However, the M2 has a negative force coefficient
for both the 0° and 45° wind angles. From the graphical representation of the drag
Fig. 17.4 a Drag and b lift coefficient on M1 and M2 model for 0°, 45° and 90° wind angles
and lift coefficient, it may be said that the corner cut model experienced less drag
and lift forces compared with the sharp edge model.
17.6 Conclusion
The numerical simulation conducted in a computational domain to determine the

pressure coefficient of different faces also drag and lift coefficients for different
angles. After a number of iterations, the following conclusions are made.
As both the models are setback, so the top part of setback model has a different
local pressure zone. The sharp edge model has high suction at this local pressure
zone, but the corner cut model has curved zone at the top part of setback section,
therefore, the wind passes smoothly at that corner. Due to this, the local pressure
zone also decreases.
• High amount of pressure difference (163%) observed at the corner cut region at
0.475H distance from the base of the model. The pressure drop is due to the corner
cut of the model.
• The drag coefficients are gradually changed from 1.06 to −0.03 and lift coefficient
from 0.01 to 1.17 for sharp edge corner model. However, for corner cut model does
not satisfy this rule.
• The drag and lift coefficients create a huge amount of variation on corner cut model
compare to the sharp edge model.
References
1. Kawai, H.: Effect of corner modifications on aeroelastic instabilities of tall buildings. J. Wind
Eng. Ind. Aerodyn. 74–76, 719–729 (1998)
2. Tanaka, H., Tamura, Y., Ohtake, K., Nakai, M., Kim, Y.C.: Experimental investigation of
configurations. J. Wind Eng. Ind. Aerodyn. 107–108, 171–191 (2012)
3. Elshaer, A., Bitsuamlak, G., Damatty, A.E.: Aerodynamic shape optimization for corners of tall
buildings using CFD. In: 14th International Conference on Wind Engineering, Porto Alegre,
Brazil (2015)
tall building using CFD simulation—A case study. Asian J. Civ. Eng. (BHRC) 17(8), 1055–1075
(2016)
5. Mooneghi, M.A., Kargarmoakhar, R.: Aerodynamic mitigation and shape optimization of
buildings: Review. J. Building Eng. (2016)
due to wind load. Asian J. Civ. Eng. Building Hous. 19(2), 205–221 (2018)
In: 2nd International Conference on Advances in Dynamics, Vibration and Control (ICADVC-
2018). pp. 381–388. NIT Durgapur (2018)
8. Alminhana, G.W., Braun, A.L., Loredo-Souza, A.M.: A numerical study on the aerodynamic
performance of building cross sections using corner modifications. Lat. Am. J. Solids Struct.
15(7), e88 (2018)
Wright, N. G.: Recommendation on the use of CFD in wind engineering. COST Action C14.
Dynamics (2004)
10. IS: 875 (Part-3), Indian standard code of practice for the design loads (other than earthquake)
for buildings and structures (part-3. wind loads). Bureau of Indian Standards, New Delhi, India
(2015)
39, 306–321 (2013)
Int. J. Mech. Prod. Eng. Res. Dev. 413–420 (2018)
13. Bairagi, A.K., Dalui, S.K.: Comparison of pressure coefficient between square and setback
tall building due to wind load. In: SEC18: Proceedings of the 11th Structural Engineering
Convention—2018, Jadavpur University, Kolkata, India (2018)
Chapter 18
Comprehensive Characterization
of Carbon Fiber-Reinforced Epoxy
Composite for Aerospace Application
D. Bino Prince Raja, B. Niharika, R. S. Manoj Kumar and C. G. Tejaswini
Abstract Carbon fiber-reinforced polymer (CFRP) is used in the aeronautical indus-

try in the manufacture of different aircraft components. This paper is about studying
the mechanical (tensile, flexural, interlaminar shear strength), thermal and moisture
characterizations of the laminate. A carbon-reinforced polymer laminate of G939
material and 913 resin systems are selected to study the effect of moisture and ther-
mal on the properties of the laminate. The composite lamina is made of different
layer orientations like 0°/90°, 0°/45°, 0°/45°/90°, 0°/0° and 90°/90°. The laminate is
fabricated by vacuum bagging and cured using autoclave. Interlaminar shear strength
(ILSS) was carried out for the specimens. Thermal degradation of CFRP is molecular
deterioration as a result of overheating, and as the temperature increases the bonding
between the molecules gets weaker and starts reacting with each other which results
in the change of properties of composites. Laminates of 0–90 orientation are fabri-
cated, and interlaminar shear strength (ILSS) at 50, 100 and 150 °C was carried out
according to Dutch Institute for Norms (DIN) for the specimens. Micro cracks in the
matrix are observed due to moisture diffusion. Five different test liquids are chosen:
water, diesel, petrol, lubricating oil and acid in which specimens are immersed for
2 days, 5 days and 7 days. This work will help composite materials’ designers and
manufacturers in designing high strength composite parts for aerospace.
Keywords Carbon fiber · Epoxy matrix · Mechanical · Thermal characterization ·

Aerospace applications
18.1 Introduction
Composite materials are made up of two or more constituent materials to obtain a

new material with the desired properties. Composite materials are being developed
and made with two kinds of objectives: One is to enhance the material properties
and performance efficiency and another to design materials with combinations of
D. Bino Prince Raja (B) · B. Niharika · R. S. Manoj Kumar · C. G. Tejaswini

Department of Aeronautical Engineering, S.J.C. Institute of Technology, Bangalore, India

https://doi.org/10.1007/978-981-15-3254-2_18
186 D. Bino Prince Raja et al.
desired properties [1]. Carbon fiber is used in industries where high strength and
rigidity are required in relation to weight. We know that material density has a direct
impact on its weight, and carbon fiber composite has a density two times less than
aluminum and more than five times less than steel [2]. We have used carbon prepreg
in which the resin system already includes the proper curing agent. Advantage of
prepreg is it is less mess and less waste. In this paper, we have studied material,
thermal and moisture characterization of CFRP laminate. Material characterization
includes tensile, flexural and interlaminar shear strength. Interlaminar shear strength
is the stress existing between layers of a laminated material [3]. Flexural test is done
to know the bending strength of the laminate; it is the maximum stress acting on the
outermost fiber of the laminate. Tensile test is done to determine the maximum load
that a material can withstand. When moisture diffuses into the material, the strength
of the material is decreased and fiber/matrix is degraded which results in the decrease
of glass transition temperature [4].
A large share of growing market value is accounted for composites which are
being used in both military and commercial applications of aeronautics. Nowa-
days, in the aeronautical industry, carbon fiber-reinforced polymer (CFRP) offers
significant improvements over current conventional materials [5]. Different struc-
tural components are manufactured by composite materials due to their attractive
specific mechanical properties [6]. When compared with metallic materials, poly-
meric composites have high strength to weight ratio, hence CFRP is processed using
thermoset polymers, especially epoxy resins [7]. The bonding material that allows
the fabric to form a composite material is the resin. Resin is a type of matrix which
acts as a medium to transfer load. We have used carbon prepreg in which resin is
pre-impregnated, and it is ready to use in the component. The resin system used is
typically epoxy. Interlaminar shear strength is the stress acting between layers of
a laminated material; usually, it is performed to characterize both fiber and matrix
interfacial bonding [8]. Flexural test is done to know the bonding strength of the
laminate, and it is the maximum stress acting on the outermost fiber of the laminate.
And it is an important tool for optimization of process and evaluation of matrices
and fiber resin interface. In order to verify specifications, quality assurance of project
and also analysis of failure mode, tensile tests are carried out [9]. Moisture content
is absorbed by epoxy resins from their surroundings which results in the decrease
of glass transition temperature, Tg: typically 1% absorbed water reduces the glass
transition temperature by 20 °C [9]. The result at elevated temperature strength and
stiffness of resin is being reduced [9]. Thermal degradation is the process which is
caused due to heat and chemical properties of a substance and changes due to tem-
perature, and as the temperature increases the bonding between the molecules gets
weaker and starts reacting with each other which results in the change of properties
of composites. The result of physical and chemical changes composite degrades with
environment [10, 11].
18 Comprehensive Characterization of Carbon Fiber-Reinforced Epoxy … 187
18.2 Experimental Work
18.2.1 Preparation of Material
The prepreg used in this work is HEXPLY G939, bidirectional (BD) material with
913 epoxy matrix system which is widely used in the fabrication of high strength
composite materials. 913 epoxy matrix systems can be processed using a wide range
of techniques.
18.2.2 Fabrication of Laminate
To fabricate composite laminate, consider material of size 200 × 200 mm which is

prepared with different layers of orientation. A flat plate was considered as a base,
and followed by vacuum bagging procedure the laminate was cured at 135 °C using
autoclave. After completing the curing cycle, laminate was demolded and trimmed
for extra part to required size. Now, the laminate is used for further testing process
(Fig. 18.1).
Fig. 18.1 Vacuum bag of laminates

18.3 Material Characterization of Composites
Five different orientations 0°/90°, 0°/45°, 0°/45°/90°, 0°/0° and 90°/90° were consid-
ered to study tensile, flexural and interlaminar shear strength (ILSS). The orientations
and layers of the laminate are listed in Table 18.1.
18.3.1 Tensile Test
The tensile tests were performed according to ASTM D638 standard using a mini-
mum of ten specimens (250 × 25 × 2 mm) for each laminate family. By bonding end
tabs of carbon fiber/epoxy laminate, the specimens were prepared. The tests were
carried out in an UTM (Fig. 18.2).
Table 18.1 Features of the layers and its orientations

Laminate No. Orientation 0° layers used 90° layer 45° layer Total no. of
used used layers used
1 0/90 4 4 – 8
2 0/45 4 – 4 8
3 0/45/90 3 2 3 8
4 0/0 8 – – 8
5 90/90 – 8 – 8
Fig. 18.2 Tensile test specimens

Fig. 18.3 Flexural test

specimens
18.3.2 Flexural Test
This test was performed according to ASTM D790 standard. This is a bending test
(three point loading) considering five samples (100 × 10 × 2 mm) for every laminate.
The tests were carried out in an UTM at exact speed of 2 mm/min at room temperature
(Fig. 18.3).
18.3.3 Interlaminar Shear Strength
The interlaminar shear tests (ILSS) were performed according to ASTM D2344
standard by considering five samples (short beam: 20 × 10 × 2 mm) for every
laminate. The tests were carried out in an UTM (Figs. 18.4 and 18.5).
Fig. 18.4 ILSS test specimens

Fig. 18.5 ILSS test setup at

UTM
18.3.4 Thermal Degradation
Select the specimen of 0/90 orientation and cut the specimen (20 specimens) as per
ASTM D2344 standard for the ILSS test by increasing the atmospheric temperature.
Now take the five specimens, test in the room temperature using UTM (ILSS) setup
and note down the obtained values. Carry out the same procedure by taking 15 pieces
into three parts and increase the test chamber temperature to 50, 100 and 150 °C,
respectively. Now, calculate the percentage decrease in the strength of the material
as the temperature increases.
18.3.5 Moisture Test
Select the specimen of 90/90 orientation and test the cut down laminate into required
size as per ASTM D2344 standard. Then, weigh the specimen before dipping into
various atmospheric samples like water, petrol, diesel, acid and lubricating oil in the
room temperature and leave the specimen to absorb moisture for 2 days, 5 days and
7 days. Take out the specimens after the specified duration and weigh the specimens.
Now, compare the results of normal room temperature tested specimen with the
moisture absorbed specimens for 2 days, 5 days and 7 days (Fig. 18.6).
Fig. 18.6 Moisture test specimens
18.4.1 Material Characterization
18.4.1.1 Result of Tensile Test
Table 18.2 shows the mean outcome of tensile strength. The outcome obtained is
according to ASTM D638 as represented. It is concluded that the 0°/90° orienta-
tion shows maximum strength, and comparison of different orientation with tensile
strength is represented in Fig. 18.7.
Table 18.2 Tensile test

Sl. No. Orientation Tensile strength (Mpa)
values of the laminates
studied 1 0°/0° 887
2 90°/90° 945
3 0°/45° 857
4 0°/90° 1250
5 0°/45°/90° 998
Fig. 18.7 Tensile values at different orientation
Table 18.3 Flexural values

Sl. No. Orientation Peak load (N) Flexural strength
of the laminates studied
(Mpa)
1 0°/0° 374.36 844.313
2 90°/90° 341.04 835.101
3 0°/45° 390.00 671.26
4 0°/90° 284.20 859.33
5 0°/45°/90° 284.2 686.15
18.4.1.2 Result of Flexural Test
Table 18.3 shows the mean outcome of flexural strength. The outcome obtained is
according to ASTM D790 standard. It is concluded that the 0°/90° orientation shows
maximum strength, and comparison of different orientation with flexural strength is
shown in Fig. 18.8.
18.4.1.3 Result of ILSS Test
Table 18.4 shows the mean results of ILSS. The outcome obtained is accord-
ing to ASTM D2344 standard. It is found that the 0°/0° orientation shows maxi-
mum strength, and comparison of different orientations with ILSS is represented in
Fig. 18.9.
Fig. 18.8 Flexural values at different orientations
Table 18.4 Interlaminar shear values of the laminates studied

Sl. No. Orientation Load (kN) Peak load (N) ILSS (Mpa)
1 0°/0° 2.36 2365.62 81.20
2 90°/90° 2.31 2312.53 80.96
3 0°/45° 2.14 2144.73 74.00
4 0°/90° 2.52 2519.27 75.47
5 0°/45°/90° 2.21 2212.28 75.42
Fig. 18.9 ILSS values at different orientations

Fig. 18.10 ILSS values at different temperature conditions
Table 18.5 Interlaminar shear values of the laminates for thermal degradation
Sl. No. Temperature in Load (kN) Peak load (N) ILSS (Mpa) Mode of failure
°C
1 23.2 (room 2.52 2519.27 75.47 A
temperature)
2 50 2.29 2288.63 71.85 A
3 100 1.84 1840.03 56.67 C
4 150 1.21 1207.77 36.17 C
A = Single shear; C = Plastic deformation
18.4.2 Thermal Degradation
The table below represents the mean outcome of interlaminar shear strength values
for different laminates for thermal degradation. It is observed that as the temperature
increases, the strength of the specimen decreases. Figure 18.10 represents variation
of ILSS values at different temperatures (Table 18.5).
18.4.3 Moisture Test
The table below represents the average results of interlaminar shear strength values
for studied laminates. Considering the ILSS results obtained in accordance with
ASTM D2344 standard, it is observed that there is maximum decrease in strength
as the moisture content increases. And comparison of ILSS at different moisture

conditions is represented in plot 4 (Fig. 18.11; Tables 18.6, 18.7 and 18.8).
Fig. 18.11 ILSS values at different moisture conditions
Table 18.6 Moisture test values for 2 days

Sl. No. Atmospheric Load (kN) Peak load (N) ILSS (Mpa)
environment
1 Water 2.32 2312.99 80.24
2 Petrol 2.30 2299.98 78.88
3 Diesel 2.26 2265.02 79.73
4 Acid 2.33 2327.84 79.37
5 Oil 2.29 2290.09 79.27

Sl. No. Atmospheric environment Load (kN) Peak load (N) ILSS (Mpa)
1 Water 2.29 2292.62 77.81
2 Petrol 2.30 2298.64 77.07
3 Diesel 2.34 2338.65 78.79
4 Acid 2.32 2318.39 77.22
5 Oil 2.36 2358.81 79.25

Sl. No. Atmospheric environment Load (mm) Peak load (mm) ILSS (Mpa)
1 Water 2.16 2158.67 72.76
2 Petrol 2.33 2328.16 76.89
3 Diesel 2.23 2211.99 76.18
4 Acid 2.31 2307.87 76.77
5 Oil 2.34 2344.70 78.57
18.5 Conclusion
• The composite laminate with 0°/90° orientation possesses maximum tensile

strength of 1250 Mpa, which is higher than several other orientations of 0°/0°,
90°/90°, 0°/45° and 0°/45°/90°, respectively.
• The composite laminate with 0/90 orientation possesses maximum flexural
strength of 859.33 Mpa, which is higher than several other orientations of 0°/0°,
90°/90°, 0°/45° and 0°/45°/90°, respectively.
• The composite laminate with 0/0 orientation possesses maximum ILSS of 81.20
Mpa, which is higher than several other orientations of 0°/0°, 90°/90°, 0°/45° and
0°/45°/90°, respectively.
• As the temperature increases, there is decrease in strength, and the laminate studied
can withstand load up to 1207.77 N at 150 °C with ILSS of 36.17 Mpa.
• For the laminate studied, the ILSS decreased with an increase in the amount of
absorbed moisture. The ILSS decreased by 80.24–72.76 Mpa for water, 78.88–
78.81 Mpa for petrol, 79.73–76.18 Mpa for diesel, 77.22–76.77 Mpa for acid and
79.27–79.48 Mpa for oil.
References
1. Kennedy, G.J., Martins, J.R.R.A.: A comparison of metallic and composite aircraft wings using
aero structural design optimization
2. Shivakumar, S., Guggari, G.S.: Literature review of fiber reinforced polymer composites. Int.
J. Adv. Eng. Technol. (IJAET) 1(5), 218–226
3. Bino Prince Raja, D., Stanly J.R., Antony Samuel Prabu, G., Alagu Sundaram, A.: Mechanical,
morphological and thermal characterization of hybrid bamboo/glass fibre reinforced polyster
composites. Rasayan J. Chem. 11(3), 990–998 (2018)
4. Ryan, J.M., Adams, R., Brown, S.G.R.: Moisture Ingress Effect on Properties of CFRP. Material
Research Centre
5. Bino Prince Raja, D., Stanly J.R., Shukla, M.: Analysis of mechanical properties of hybrid
bamboo/jute fiber reinforced vinyl ester composite materials. Int. J. Mech. Eng. Technol. 8(10),
318–324 (2017)
6. Trapani, J.R., Milan, M.T., Spinelli, D., Bose, W.W.: Mechanical performance of carbon-epoxy
laminates. Mat. Res. 9, 115–120 (2006)
7. Shabbier, K.P., Murtaza, M.A.: Optimization of aircraft wing with composite material. 2, ISSN:
2319-8753
8. De Paiva, J.M.F., De Nadaidos Santos, A., Rezenda, M.C.: Mechanical and morphological
characterization of carbon fiber fabric reinforced epoxy composites used in aeronautical field.
Mat. Res. 12, 367–374 (2009)
9. De Paiva, J.M.F., Mayer, S., Rezende, M.C.: Comparison of tensile strength of different carbon
fabric reinforced epoxy composites. Mat. Res. 9, 83–89 (2006)
10. Jhosi, O.M.: The effect of moisture on the shear properties of carbon fiber. Compos. Butterworth
14(3), 196–200 (1983)
11. Go, S.H., Hong, S.W., Kim, H.G., Kwae, L.K.: A study on the degradation of mechanical
properties for the carbon composites at a high temperature and high humidity of environment.
ISOR J. Mech. Civ. Eng. (IOSR-JMCE) 13(15), 17–24 (2016)
Chapter 19
Design and Simulation of Single-Axis
MEMS Accelerometer for Low
Acceleration Applications
K. J. Rudresh, Kin Gopalakrishna, K. Bharath Gowda,

R. Harshith Gangatkar and Hemanth Kumar
Abstract This work involves the design of a MEMS single-axis capacitive-type

accelerometer for low acceleration applications. In this paper, the displacement
amplification complaint mechanism (DaCM) is used to increase sensitivity of the
device by amplifying displacement. The DaCM model was designed using graphical
method, and ANSYS tool is used to compare the displacement results. The DaCM
model with maximum displacement is integrated with capacitive-type accelerome-
ter. This integration was done using Coventerware turbo tool. The results obtained
were compared with the accelerometer without DaCM. The displacement values that
were obtained with and without DaCM are 3.1e−1 and 4.2e−2 µm, respectively.
The results prove that DaCM gives an improved sensitivity.
Keywords MEMS · DaCM · Capacitance
19.1 Introduction
Microelectro mechanical systems (MEMS) had been a fast-growing technology.

MEMS devices can act both as sensors and actuators. Some of the widely used MEMS
devices are microphone, gyroscope, accelerometers, etc. MEMS accelerometers are
sensing devices used for the detection of acceleration. Capacitive-type accelerome-
ters are the most commonly used. It consists of proof mass and interdigitated fingers
for capacitive sensing. When acceleration is applied to the proof mass, it displaces.
There will be a change in distance (d) or area (A) between any sensing fingers.
When one of the parameter changes, the capacitance also changes. Capacitive-type
K. J. Rudresh (B)
Department
of Electronics and Communication Engineering, Nitte Meenakshi Institute of Technology,
Bangalore, Karnataka, India
K. Gopalakrishna · K. Bharath Gowda · R. Harshith Gangatkar · H. Kumar
Department of Mechanical Engineering, Nitte Meenakshi Institute of Technology, Bangalore,
Karnataka 560064, India

https://doi.org/10.1007/978-981-15-3254-2_19
200 K. J. Rudresh et al.
accelerometers have advantages [1], like high sensitivity, good noise performance,
low-temperature sensitivity and low power dissipation. All the fingers in capaci-
tive type are connected parallel, and if the number of capacitors is increased, then
this increases the sensitivity [2]. Capacitive-type accelerometers can be used for
low g sensing ranging from 2 µg to several g’s. Some of the examples for low g
sensor applications are structural health monitoring, border infiltration, earthquake
detection, etc.
For low g applications since the acceleration that is applied is very low, displace-
ment is also low. In order to improve the device sensitivity, amplification is required.
There are two types of amplification: mechanical amplification and electronic ampli-
fication. Researchers had found how mechanical amplification is preferred over elec-
tronic amplification [3]. Mechanical noise is less compared to electronic noise [3].
There are many types of mechanical amplification like force amplification and dis-
placement amplification. Force amplification has been explained in [4] and with an
amplification of 11. Further researchers have taken the idea of displacement amplifi-
cation to design a displacement amplifier. Displacement amplification was done with
compliant mechanisms. Compliant mechanisms are those that transmit motion, force
or energy by elastic deflection of flexural members. Displacement amplification com-
pliant mechanisms (DaCMs) are single-input-single-output-type mechanism where
an input displacement is applied at the input point which generates an amplified
output at output point.
19.2 Design Methodology
This section explains about modelling of basic MEMS accelerometer and the design
of DaCM using graphical method as well as modelling of integration of basic
accelerometer with DaCM.
19.2.1 Modelling of Basic MEMS Accelerometer
When an accelerometer is subjected to acceleration, there is some force that is induced

and this makes the accelerometer to vibrate in certain direction depending on the
spring and damper. Due to this vibration, there is some displacement of proof mass.
The mechanical sensitivity is defined as the displacement of the proof mass per unit
gravitational acceleration, and sensitivity is given by capacitance per unit gravita-
tional acceleration. The model of a spring-mass-damper system is shown in Fig. 19.1.
The basic equation of the spring-mass-damper system is given by
m ẍ + b ẋ + kx = F = ma (19.1)
19 Design and Simulation of Single-Axis MEMS Accelerometer for Low … 201
Fig. 19.1 Spring-mass-damper model of accelerometer
where m = proof mass, b = damping coefficient, k = spring stiffness, F = force, a

= acceleration, x = displacement of proof mass.
19.3 Design of DaCM Using Graphical Method
Instantaneous centre method
Consider a rigid body in a plane having two points, and each point on the rigid body
has different velocities. There will be point on the plane where there will be zero
velocity, which makes the body to rotate about that point known as instantaneous
centre of rotation. In order to compute this point, one has to draw a perpendicular
line to the velocity and the point where both of them intersect; this is called instant
centre point as shown in Fig. 19.2. The DaCM drawn using this method is shown in
Fig. 19.3. By applying the instantaneous centre method, the DaCM was designed.
The theoretical amplification of 16 was obtained. A solid model was drawn using
solid edge and is shown in Fig. 19.4.
DaCM is a single member of flexure joints, whose required points are fixed. The
designed model can be used for planar amplification that is for anyone of the single
planar amplification. The input displacement and the output displacement are in
the same direction. This DaCM design is integrated with a basic accelerometer for
improving the sensitivity.
19.4 Modelling of Integrated Accelerometer with DaCM
The device is modelled using MEMS CAD tool called Coventerware turbo. The
tool is used in 3D building of model and analysis of the accelerometer device. Here
CoSolveEM which consists of mechanical domain and electrical domain analysis
is used. All the electrodes and suspension ends are fixed, and acceleration of 1 g
Fig. 19.2 Instantaneous centre method
Fig. 19.3 Graphical design of DaCM
is applied in y-direction for the movable proof mass for mechanical domain. For
electrical domain, 5 V is applied to the fixed electrodes and proof mass is connected
to the ground.
Fig. 19.4 DaCM model using solid edge
The basic accelerometer model integration with DaCM is shown in Fig. 19.5.
Here single crystalline silicon is used due to its properties like high melting point
and low thermal expansion. The material has a density (r) of 2330 kg/m3 , Young’s
modulus (E) of 137 GPa, relative permittivity (v) of 11.7 and Poisson’s ratio (m)
of 0.278. Dimensions of the model are shown in Table 19.1. The meshed model of
Fig. 19.5 Integrated accelerometer with DaCM

Table 19.1 Dimensions of integrated accelerometer model

Parameters Length (µm) Width (µm) Thickness (µm)
Proof mass 1000 500 30
Fingers 10 300 30
DaCM spring suspension 50 500 30
DaCM proof mass 1000 1000 30
Folded beam suspension (one side) 1070 10 30
Fig. 19.6 Meshed integrated acceleration model
integrated accelerometer model is shown in Fig. 19.6.
ANSYS simulation of DaCM: The DaCM model that was designed graphically
using instantaneous centre method was simulated using ANSYS tool by applying
required boundary condition and suitable mesh size. The results obtained as shown
in Fig. 19.7. There is a difference between theoretical graphical method and ANSYS
as shown in Table 19.2. This difference is because of elastic losses in DaCM, mesh
approximations done in ANSYS and some constraint that these tool uses to solve
partial differential equations.
Displacement can be calculated using the basic formula
kx = ma ==→ x = ma/k (19.2)
where x = displacement, m = mass, a = acceleration, k = stiffness.

Fig. 19.7 ANSYS simulation
Table 19.2 Displacement

Theoretical ANSYS
amplifying factor comparison
theoretical and ANSYS Displacement amplifying factor 16 9.5
Capacitance is given by
C = (ε A/d)N (general capacitance) (19.3)
where 1 = permittivity, A = area, d = distance between the plates, N = number of

fingers on both the sides.
Figure 19.8 shows the when 1 g of acceleration is applied on an accelerom-
eter without DaCM, a maximum displacement of 4.8 × 10−2 mm is observed.
The displacement in y-direction is 4.170842 × 10−2 mm. The variation of capaci-
tance, displacement and sensitivity with respect to applied acceleration is shown in
Table 19.3.
When the applied acceleration increases, the displacement increases and the
capacitance decreases. But the differential capacitance increases. The capacitance
observed for 1 g acceleration is 1.895073 pF. Table 19.4 shows the results for input
values. The simulation result of MEMS-based capacitive accelerometer integrated
with DaCM is shown in this section. The analysis of displacement and capacitance
is obtained in Coventerware. When 1 g acceleration is applied, a displacement of
Fig. 19.8 Accelerometer without DaCM for 1 g
Table 19.3 Variation of displacement, capacitance and sensitivity with applied accelerations
without DaCM
Acceleration (g) Displacement (µm) Capacitance (pF) Sensitivity (pF/g)
0 0.03696418 1.884471 –
0.1 0.03743861 1.884468 18.84468
0.2 0.03791303 1.884465 9.422325
0.4 0.03886188 1.884460 4.711150
0.6 0.03981073 1.884455 3.140758
0.8 0.04075958 1.884449 2.355561
1 0.04170842 1.884444 1.884444
Table 19.4 Variation of displacement, capacitance and sensitivity with applied accelerations with
DaCM
Acceleration (g) Displacement (µm) Capacitance (pF) Sensitivity (pF/g)
0 0.08767009 1.895127 –
0.1 0.08767850 1.895121 18.95121
0.2 0.08768151 1.895116 9.47558
0.4 0.08769294 1.895105 4.7377625
0.6 0.08770437 1.895094 3.15849
0.8 0.08771580 1.859084 2.368855
1 0.08772723 1.895073 1.895073
Fig. 19.9 Integrated accelerometer with DaCM for 1 g acceleration
8.772723 × 10−2 mm was observed in y-direction, and maximum displacement of

3.1 × 10−1 mm can be observed in Fig. 19.9.
On plotting for various accelerations, the values obtained for displacement vary
linearly and this increases with g. The plot can be observed in Fig. 19.10. The plot
is generated by acceleration versus capacitance.
One can observe from Fig. 19.11 that capacitance decreases with the increase in
acceleration. On comparing the values from Tables 19.3 to 19.4, we can observe that
displacement, capacitance and sensitivity are higher in case of accelerometer with
DaCM (Table 19.5).
19.6 Conclusion
In this paper, single-axis MEMS accelerometer is designed, simulated and analysed

for measuring low acceleration applications. The amplified displacement factor that is
obtained theoretically and in ANSYS is 16 and 9.5, respectively. The results that were
obtained from simulation of accelerometer with DaCM show that for 1 g acceleration;
sensitivity of 1.895073 pF/g was achieved and without DaCM for 1 g acceleration;
sensitivity of 1.88444 pF/g was achieved. The accelerometer model with DaCM
gives higher displacement and sensitivity.
0.08774
0.08773
0.08772
Displacement(μm)
0.08771
0.0877
0.08769
0.08768
0.08767
0.08766
0 0.5 1 1.5
Accelera on(g)
Fig. 19.10 Displacement versus acceleration
1.89513
1.89512
Capacitance(pF)
1.89511
1.8951
1.89509
1.89508
1.89507
0 0.5 1 1.5
Accelera on(g)
Fig. 19.11 Capacitance versus acceleration
Table 19.5 Comparison of displacement with DaCM and without DaCM

Without DaCM With DaCM
Maximum displacement for 1 g (mm) 4.8 × 10−2 3.1 × 10−1
Displacement in y-direction for 1 g (mm) 4.170842 × 10−2 8.772723 × 10−2
Acknowledgements The author would like to thank Centre for Nanomaterials and MEMS, Nitte
Meenakshi Institute of Technology for providing facilities to make this work possible.
References
1. Pandey, K.P., Kumar, A.: Design and analysis of dual axis MEMS capacitive accelerometer. Int.
J. Electron. Eng. Res. ISSN 0975-6450
2. Khan, S., Ananthasuresh, G.K.: A Micromachined Wide-Band In-plane Single-Axis Capaci-
tive Accelerometer with a Displacement-Amplifying Compliant Mechanism. Taylor & Francis
(2014)
3. Liu, C.: Foundations of MEMS. Pearson International Edition (2006). ISBN 0-13-199204-X
4. Mahalik, N.P.: MEMS, 3rd edn. Tata McGraw-Hill Publishers (2007). ISBN-13:978-0-07-
063445-9
Chapter 20
Optimization of Neem Biodiesel Blend
and Injection Pressure of CI Engine
C. Srinidhi, Sunil Panshetty, Madhusudhan and S. V. Channapattana
Abstract Current chapter focuses on the variation of injection of opening pressure

at various engine loading conditions. The biodiesel chosen for experimentation was
neem oil methyl ester, and it was blended with diesel at volumetric concentrations of
25, 50 and 75% to form B25, B50 and B75, respectively. The experimental results
revealed that higher (280 bars) IOP leads to higher carbon dioxide emission and
BTHE. The observations recorded were subjected to grey relation grade analysis for
blend and injection pressure optimization. The GRG technique revealed that IOP of
280 bars and blend concentration of 25% is optimal.
Keywords Grey relation grade · NOx · Blend of biodiesel · Injection opening

pressure
20.1 Introduction
Compression ignition engines dictate the field of commercial transportation and

agricultural machinery as they are known for their ease of operation and higher
power competence to fuel consumption. Petroleum products or crude oil demand
control the global economy as these are derived from fossil source. To this problem,
many investigators have postulated fatty acid methyl esters (FAME) as possible
alternate fuel for CI engines. This FAME commonly called as biodiesel bear similar
thermophysical properties to that of biodiesel. Transesterification is the commonly
inherited method for preparation of biodiesel from edible or non-edible oils [1–4].
C. Srinidhi (B) · S. Panshetty

Department of Mechanical Engineering, Pune Vidyarthi Griha College of Engg and Technology,
Pune, India
Madhusudhan
Department of Mechanical Engineering, Nitte Meenakshi Institute of Technology, Yelahanka,
Bangalore, India
S. V. Channapattana
Department of Mechanical Engineering, D.Y. Patil College of Engineering, Ambi, Pune, India

https://doi.org/10.1007/978-981-15-3254-2_20
212 C. Srinidhi et al.
Biodiesel is distinguished to three categories based on source of feedstock. If non-

edible feedstock is used for preparation of biodiesel, then they are categorized as
second-generation type and edible sources as first generation. Second-generation
biodiesels are usually preferred as best possible feed stock as they do not raise fuel
to food crises [5].
20.1.1 Experimental Analysis of Using Biodiesel
Researchers extensively worked on increasing usage of Biodiesel by altering var-

ious engine parameters like compression ratio (CR), fuel injection timing (FIP)
and injection opening pressure (IOP) [6–8]. Looking at the current engine designs,
diesel engine has fuel delivering systems which pressurize fuel to high pressures
and later induce them in the combustion chamber. For such phenomenon, the current
chapter focuses on the introduction of fuel injection pressure. For last few decades,
research has been extensively carried out. But it is very much unclear that which
blend of biodiesel and fuel injection pressure level suits best, which is yet to discov-
ered. Studies indicate that incrementation of FIP has resulted in higher peak pressure
and heat release rate [9–11]. Collaterally, the brake thermal efficiency also rises up
making the NOx emission to cross the threshold. [9–11] On the other hand, reducing
the IOP leads to fall in BTHE and leads to higher CO and HC emissions [12, 13].
So, it is quite unclear that what should be the optimum injection opening pressure
that has to set up for diesel engines to get better performance and lesser emissions
[2].
20.1.2 Novelty
As there is no literature of actual optimal blend and IOP, current chapter provides
an insight of optimization of parameters like blend percentage and injection opening
pressure.
Experimental Methodology and Set-up: The current work is phased in two stages.
In the first step, neem biodiesel is prepared by base catalytic transesterification pro-
cess. KOH was used as a catalyst to reduce the operating temperature of the reaction
[5]. In the next step, neem oil methyl ester (NOME) obtained is washed with dem-
ineralized water to remove extracts of excess KOH and methanol. The finally derived
NOME is blended in volumetric proportions of 25, 50 and 75% with diesel. Finally,
three test fuels are prepared which are ready to experimental test on CI engine.
The experimentation was performed on a single-cylinder diesel engine. The engine
standards are given in Table 20.1. The exhaust emissions were measured using a
Naman-make analyser. The range and specifications are mentioned in Table 20.2
20 Optimization of Neem Biodiesel Blend and Injection … 213
Table 20.1 Specifications of

Manufacture Rocket Engg. Corpn. Ltd.
engine
SFC 251 g/kWh
Rated power 4.8 kW @ 1500 RPM
Standard CR 17.5:1
Bore 87.5 mm
Stroke 110 mm
Injection timing 230 before TDC
Inlet valve open bTDC 4.5 bTDC
Exhaust valve open 35.5 bBDC
Inlet valve close 35.5 aBDC
Exhaust valve close 4.5 aTDC
Table 20.2 Device

Device Range Accuracy Uncertainties
specifications and
specification
terminology
Carbon 0–10.00% ±0.01% ±0.1
monoxides
(CO)
Carbon 0–20.00% ±0.01% ±0.15
dioxides (CO2 )
Oxides of 0–5000 ppm ±1 ppm ±0.2
nitrogen (NOx )
Oxygen (O2 ) 0–25.00% ±0.01% ±0.1
Hydrocarbons 1–1500 ppm ±1 ppm ±0.2
(HC)
Exhaust gas 0–500 °C ±1 °C ±0.1
temp. (EGT)
Tachometer 0–10,000 rpm ±10 rpm ±0.2
Fuel flow metre 1–30 cc ±0.1 cc ±0.5
Pressure 0–500 bar ±1 bar ±0.1
transducer
20.1.3 Design Factors and Response Variable
In design of experiment (DOE) technique, response fluctuations are unfamiliar func-

tion of the process variables, called as design factors. The three parameters considered
for performance and emission study are braking torque, fuel blend percentage and
injector pressure. The ranges implied are given in Table 20.3 (Table 20.4).
Table 20.3 Design factors

Design factor Levels
and their levels
1 2 3
Injector opening pressure (A) 200 240 280
Blend % (B) 25 50 75
Braking torque (C) 15 20 25
20.1.4 Results and Discussion
The test fuels were tested for their thermophysical properties and are given in
Table 20.5. The procedure for testing these was as per ASTM standards.
20.1.5 Performance Characteristics
Brake Thermal Efficiency (BTHE): BTHE signifies the output power obtained to
heat generated during combustion. Figure 20.1 plots the observations of BTHE at
various IOPs and varying braking torques. It was observed that, as the IOP was incre-
mented, the BTHE values also increased. The maximum BTHE value was recorded
for B75 blend at engine parameter of 280 bar [4, 9–11, 14]. The calculation of S/N
ratio was done on the basis of “higher the better”. The regression equation obtained
is (Table 20.6).
BTHE = 16.1 + 0.0174 IOP + 0.398 LOAD + 0.0257 BLEND
Brake-Specific Fuel Consumption (BSFC): BSFC signifies the fuel consumed for
a unit power derived. Also, for any alternative fuel, it would be desirous that the
fuel consumption should be least. From figure, it was noted that the minimum and
maximum BSFC values were recorded for B25 blend at 240 bars and B75 blend at
200 bar IOP, respectively [6, 7]. “Lesser the better” way was chosen for calculation
S/N ratio of BSFC. The regression equation for the BSFC in regard to independent
variables is (Fig. 20.2 and Table 20.7).
BSFC = 0.437 − 0.000183 IOP + 0.000240 BLEND − 0.00447 LOAD
20.1.6 Emission Characteristics
Carbon Dioxide Emission (CO2 ): Figure 20.3 illustrates the variation of CO2 emis-
sion at various blends and IOPs. It was observed that as the blend concentration
Table 20.4 L9 orthogonal array containing experimental values as per Taguchi design
Expt. no IOP Blend Load O2 CO2 CO HC NOx BTHE BSFC
1 200 25 15 14.8 8.27 0.1 102 66 26.032 0.343
2 200 50 20 11.8 10 0.3 135 89 29.46 0.312
3 200 75 25 7.06 8.93 0.2 141 101 31.594 0.306
4 240 25 20 13 10 0.1 110 95 30.328 0.295
5 240 50 25 12.6 14.3 0.1 115 121 30.283 0.304
6 240 75 15 17.4 9.56 0.1 119 81 26.79 0.361
20 Optimization of Neem Biodiesel Blend and Injection …
7 280 25 25 12.4 12.6 0.1 103 132 30.778 0.29

8 280 50 15 17.4 8.64 0.1 86 79 27.884 0.33
9 280 75 20 21.5 7.73 0.1 102 114 32.607 0.297
215
Table 20.5 Properties of test fuels

Sr. no. Properties B100 B25 B50 B75
1 Density (kg/m3 ) 912 831 856 891
2 Calorific value (MJ/kg) 36.72 40.3 39.16 37.21
3 Viscosity (cSt) 4.9 2.9 3.1 4.5
4 Flashpoint (°C) 182 91 118 149.25
5 Carbon residue 0.15 – – –
6 Cloud point (°C) −1 – – –
7 Pour point (°C) −8 – – –
Fig. 20.1 Variation of BTHE at various IOPs and torques
Table 20.6 Grey relation grade for BTHE

Sr. no. Expt. readings Normalized Deviation Grey relational S/N ratio
of BTHE S/N ratio of sequences for coefficient
BTHE BTHE BTHE
1 26.0319 0 1 0.333333333 28.31011735
2 29.4596 0.549295373 0.450704627 0.525925704 29.38453691
3 31.5941 0.859921184 0.140078816 0.781153801 29.99211977
4 30.3283 0.678346981 0.321653019 0.608529377 29.63696135
5 30.2833 0.67175322 0.32824678 0.60368481 29.62406398
6 26.7901 0.127489701 0.872510299 0.364295991 28.55948669
7 30.7777 0.743668092 0.256331908 0.661085424 29.76472918
8 27.8843 0.305259653 0.694740347 0.418500975 28.90720402
9 32.6066 1 0 1 30.26611338
Fig. 20.2 Variation of BSFC at various IOPs and torques
Table 20.7 Grey relation grade for BSFC

Sr. no. Expt. readings Normalized S/N Deviation Grey relational S/N ratio
of BSFC ratio of BSFC sequences for coefficient
BSFC BSFC
1 0.3431 0.7663 0.2336 0.6814 9.2901
2 0.312 0.3307 0.6692 0.4276 10.1169
3 0.3062 0.2455 0.7544 0.3985 10.2787
4 0.2945 0.0666 0.9333 0.3488 10.6182
5 0.3035 0.2054 0.7945 0.3862 10.3548
6 0.3611 1 0 1 8.8466
7 0.2902 0 1 0.3333 10.7447
8 0.3296 0.583 0.4169 0.5453 9.638
9 0.2967 0.1009 0.899 0.3573 10.5532
increased, the CO2 percentage also increased. Also, the presence of incremental IOP
evaluated the fact that IOP promotes better combustion of neem biodiesel-diesel
blends. It is eminent that as biodiesel-diesel blends bear higher density and viscos-
ity, it is better to inject these fuels at [15] higher IOP [11, 16]. The maximum CO2
emissions were recorded for B75 blend at 280 bars. Also, “Higher the better” prin-
ciple was used to calculate the S/N ratio of CO2 emissions. The regression equation
obtained is (Table 20.8)
CO2 = 3.54 + 0.0074 IOP − 0.0310 BLEND + 0.312 LOAD
Carbon Monoxide Emission (CO): CO emission is regarded for any alternate fuel
investigator as it highlights incomplete combustion of fuel. From Fig. 20.4, it was
Fig. 20.3 Variation of carbon dioxide emission at various IOPs and torques
Table 20.8 Grey relation grade for carbon dioxide emission

Sr. no. Expt. readings Normalized Deviation Grey S/N ratio of
of CO2 S/N ratio of sequences for relational CO2
CO2 CO2 coefficient
CO2
1 8.27 0.890353697 0.109646303 0.820147678 18.35011019
2 8.96 0.580294157 0.419705843 0.543652086 20.00868155
3 7.87 0.765677364 0.234322636 0.680899615 19.01702918
4 10.01 0.580294157 0.419705843 0.543652086 20.00868155
5 8.77 0 1 0.333333333 23.11279268
6 11.7 0.654982614 0.345017386 0.591703802 19.60915785
7 11.13 0.201497538 0.798502462 0.385058954 22.03494148
8 10.45 0.819284292 0.180715708 0.734521026 18.73027485
9 13.52 1 0 1 17.76358988
observed that as the concentration of neem methyl ester increased, the percentage
of CO emission also increased. This phenomenon leads to an understanding that
neem biodiesel-diesel blends being denser and more viscous need better time for
combustion [1, 12, 13]. Maximum CO emission was recorded for B75 blend at
240 bars of IOP and the least CO percentage at 240 bars for B25 blend [4]. “Smaller
the better” principle was taken for calculation of S/N ratio. The regression equation
for CO emission w.r.t. independent variables is (Table 20.9)
CO = 0.333 − 0.00125 IOP + 0.00067 BLEND + 0.00333 LOAD

Fig. 20.4 Variation of CO emission at various IOPs and torques
Table 20.9 Grey relation grade table for CO emission

Sr. no. Expt. readings Normalized Deviation Grey S/N ratio for
of CO S/N ratio of sequences for relational CO
CO CO coefficient CO
1 0.05 0 1 0.333333333 26.02059991
2 0.05 0.630929754 0.369070246 0.575327486 20
3 0.05 1 0 1 16.47817482
4 0.05 0 1 0.333333333 26.02059991
5 0.05 0.630929754 0.369070246 0.575327486 20
6 0.32 0 1 0.333333333 26.02059991
7 0.08 0.717684818 0.282315182 0.639128591 19.1721463
8 0.46 0 1 0.333333333 26.02059991
9 0.14 0 1 0.333333333 26.02059991
Unburnt Hydrocarbon Emission (HC): Emittents like HC coming out of engine

exhaust signify the presence of incomplete burning of fuel [8, 17]. Figure 20.5 shows
the variation of HC emission at various blend ratios and IOPs. It was noted that as
the volumetric concentration of neem biodiesel in the blend increased, the amount
of HC rose up. It shows that biodiesel combustion needs more time. Also, biodiesel-
diesel blends have higher ignition lag which promotes the HC emission [1, 4, 12,
13]. Maximum HC emission was recorded for B75 blend at 240 bar and least HC
for B25 blend at 200 bar. The presence of incrementation of IOP did not help in
the reduction of HC. The regression equation for HC emission w.r.t. independent
variables is (Table 20.10)
HC = 149 − 0.363 IOP + 0.313 BLEND + 1.73 LOAD

Fig. 20.5 Variation of HC emission at various IOPs and torques
Table 20.10 Grey relation grade for HC emission

Sr. no. Expt. readings Normalized Deviation Grey S/N ratio for
of HC S/N ratio of sequences for relational HC
HC HC coefficient HC
1 102 0.654892454 0.345107546 0.591640676 40.17200344
2 96 0.087953083 0.912046917 0.354095883 42.60667537
3 91 0 1 0.333333333 42.98438225
4 110 0.502170712 0.497829288 0.501087717 40.8278537
5 86 0.41226248 0.58773752 0.459669719 41.21395681
6 137 0.34310695 0.65689305 0.432192068 41.51093923
7 100 0.635159593 0.364840407 0.57814135 40.25674449
8 140 1 0 1 38.68996902
9 119 0.654892454 0.345107546 0.591640676 40.17200344
Exhaust Oxygen (EO): Figure 20.6 demonstrates graphically the obtained exhaust
oxygen at various loads and injection opening pressures. For biodiesel as fuel for
combustion, the chemical structure itself contains embedded oxygen. The recording
reveals that as the braking load increased, the amount of oxygen content decreased.
Also, the amount of least oxygen was found for B50 blend at IOP of 200 bars. The
presence of IOP helped in deriving higher EO. The presence of higher EO in exhaust
signifies that the combustion occurring is complete in nature. This might be due
to ignition delay. The amount of EO for B50 and B75 blends at IOPs 240 bars ad
280 bars almost showed the same trend. The regression equation obtained for EO
with reference to independent variables is (Table 20.11)
EO = 6.54 + 0.0728 IOP + 0.0380 Blend − 0.585 Load

Fig. 20.6 Variation of exhaust oxygen at various IOPs and torques
Table 20.11 Grey relation grade for exhaust oxygen

Sr. no Expt. readings Normalized Deviation Grey relational S/N ratio of
of O2 S/N ratio of sequences for coefficient O2 O2
O2 O2
1 14.84 0.331788559 0.668211441 0.428004711 23.42867802
2 17.52 0.537212885 0.462787115 0.519325604 21.44499795
3 17.93 1 0 1 16.97609402
4 13 0.450859748 0.549140252 0.476580704 22.27886705
5 14.23 0.478257388 0.521742612 0.489360034 22.01430173
6 11.48 0.189675229 0.810324771 0.381584788 24.80099544
7 13.22 0.496269329 0.503730671 0.498141597 21.84036942
8 10.01 0.190710907 0.809289093 0.38188663 24.79099442
9 11.91 0 1 0.333333333 26.63259435
Oxides of Nitrogen (NOx ): Biodiesel-diesel blends possess higher molecular weight

density due to when combusted leads to higher exhaust gas temperature result-
ing in higher peak pressures. Most investigators reported that higher NOx emis-
sion results for higher volumetric concentration of biodiesel in the blend. Similar
results were obtained for neem biodiesel-diesel blends at various braking torques and
IOPs. Figure 20.7 recordings observed for B75 blend at 280 bar injection pressure
showed the highest NOx . The regression equation of NOx is (Table 20.12)
NOx = 94.2 − 0.158 IOP + 0.100 BLEND + 0.200 LOAD

Fig. 20.7 Variation of NOx emission at various IOPs and torques
Table 20.12 Grey relation grade for NOx

Sr. no. Readings of Normalized Deviation Grey S/N ratio of
NOx S/N ratio of sequences for relational NOx
NOx NOx coefficient
NOx
1 66 0.371758779 0.628241221 0.443167641 36.39087871
2 89 0.070273983 0.929726017 0.349717354 37.38463439
3 101 0.216707107 0.783292893 0.389622668 36.9019608
4 95 0.852197627 0.147802373 0.771840334 34.80725379
5 121 0 1 0.333333333 37.61627185
6 81 0.070273983 0.929726017 0.349717354 37.38463439
7 132 0.622912366 0.377087634 0.570068464 35.56302501
8 79 0.622912366 0.377087634 0.570068464 35.56302501
9 114 1 0 1 34.32006687
20.2 Optimization Using Grey Relational Grade (GRG)

Generation
In order to find the optimum combination of load, blend of fuel and injection pres-
sure, its dependence on the output responses should be known. In the current work,
dependent parameters like BTHE and CO2 were given high weighting factor, and
other parameters like CO, HC, NOx and BSFC were highlighted as less weighting
factor. By doing so, the grey relation grade (GRG) is achieved by which it is possible
to compute the optimum combination of the engine design/operating parameters,
i.e., load, blend of fuel and IOP. After getting the grey relation grade and with the
weighting factor with the help of MINITAB® software, it is possible to analyse the
Taguchi design of L9 orthogonal array. Higher GRG signifies the experimental result
which is closer to the ideally normalized value. In other words, optimization of the
complicated multiple characteristics can be converted into optimization of a single
grey relational grade. The grey relational grade is given in Table 20.13. Since the
experimental design is orthogonal, it is then possible to separate out the effect of
each parameter on thereby relational grade at different levels. Basically, the larger
the value of relational grade, the better is the multiple performance characteristics
(Fig. 20.8 and Table 20.14). Numbers in bold in Table 20.14 indicates best suitable
levels in reference to better performance and lower emission.
Confirmation test: The confirmation test for the optimal parameters with its levels
was conducted to evaluate quality characteristics. Table 20.15 shows the highest grey
relational grade, indicating the process parameter set of A3B1C2 for the best multiple
performance characteristics among the nine experiments. By using the A3B1C2
set-up, it confirms an experimental result as given in Table 20.15.
Table 20.13 Grey relation grades with corresponding S/N ratio and its order
Expt no. IP Blend Load Grey relational grades Ranking
1 1 1 1 0.518732283 5
2 1 2 2 0.470811535 8
3 1 3 3 0.654796977 2
4 2 1 2 0.511979074 6
5 2 2 3 0.454418923 9
6 2 3 1 0.493261048 7
7 3 1 3 0.523565388 4
8 3 2 1 0.569088016 3
9 3 3 2 0.659382871 1
Fig. 20.8 Graph of grey

relational grades versus
experimental numbers
Table 20.14 Grey relation grades with corresponding S/N ratio and its order
Parameter Level 1 Level 2 Level 3 Max–Min Ranking
IOP (A) 0.5481136 0.486553 0.584012 0.097459 2
Blend (B) 0.60248 0.49810613 0.51809227 0.104374 1
Load (C) 0.5270271 0.5473912 0.54426 0.020364 3
Table 20.15 Results of

Initial parameter Optimal process parameters
confirmation test
Predicted Experimental
Levels A1B1C1 A3B1C2
O2 14.84 20.468889 21.46
CO2 8.27 7.211111 7.73
CO 0.05 0.045556 0.05
HC 102 94.88889 102
NOx 66 58.88889 66
BTHE 26.0639 28.59839 32.60661148
BSFC 0.343156 0.34107 0.296713401
Predicted response formula is used to find the optimal level of combination in

Table 20.15, and it can be calculated by using Minitab software
Predicted Response = Average A3 + Average B1 + Average C1 − 2 × mean
response
20.3 Inferences
Based on the study conducted for multiple response parameters (engine performance
and emission study) and its optimization on independent variables using grey rela-
tional analysis, the obtained optimized combination is found to be A3B1C2 which
means that the blend percentage of 25% and IOP of 28 bars could be used as best suit-
able higher performance and lesser emission standards. The methodology adopted
in this study can also be used to optimize other engine operating parameters like
compression ratio and injection timing.
References
1. Sayin, C., Gumus, M.: Impact of compression ratio and injection parameters on the performance
and emissions of a DI diesel engine fueled with biodiesel-blended diesel fuel. Appl. Therm.
Eng. 31, 3182–3188 (2011). http://dx.doi.org/10.1016/j.applthermaleng.2011.05.044
2. Channapattana, S.V., Pawar, A.A., Kamble, P.G.: Investigation of DI-CI four stroke VCR engine
at different static injection timing using Bio-Fuel derived from non-edible oil source as a fuel.
Biofuels (2016, Taylor & Francis). https://doi.org/10.1080/17597269.2016.1187540
3. Gumus, M., Sayin, C., Canakci, M.: The impact of fuel injection pressure on the exhaust
emissions of a direct injection diesel engine fueled with biodiesel-diesel fuel blends. Fuel 95,
486–494 (2012)
4. Channapattana, S.V., Pawar, A.A., Kamble, P.G.: Effect of injection pressure on the perfor-
mance and emission characteristics of VCR engine using Honne biodiesel as a fuel. Mater.
Today Proc. 2(4–5), 1316–1325 (2015, Elsevier). Available online at www.sciencedirect.com
5. Srinidhi, C., Channapattana, S.V., Hole, J.A., Pawar, A.A., Kamble, P.G.: Investigation on
performance and emission characteristics of C.I. engine fuelled with Honne oil methyl ester.
Int. J. Eng. Sci. Invention 3(5), 59–66 (2014)
6. Banapurmath, N.R., Tewari, P.G., Hosmath, R.S.: Performance and emission characteristics of
a D.I.C.I. Engine operated on Honge, Jatropha & sesame oil methyl esters. Renew. Energy 33,
1982–1988, 2208 (2019)
7. Kannan, G.R., Anand, R.: Effect of injection pressure and injection timing on DI diesel engine
fuelled with biodiesel from waste cooking oil. Biomass Bioenergy (2012)
8. Srinidhi, C., Madhusudhan, A., Channapattana, S.V.: Effect of NiO nanoparticles on perfor-
mance and emission characteristics at various injection timings using biodiesel-diesel blends.
Fuel 235, 185–193 (2019). https://doi.org/10.1016/j.fuel.2018.07.067
9. Birch, S.: Fuel Injection: The Inside Story. Tech Briefs SAE (2000)
10. Kannan, G.R., Anand, R.: Experimental evaluation of DI diesel engine operating with diestrol at
varying injection pressure and injection timing. Fuel Process. Technol. 92, 2252–2263 (2011).
https://doi.org/10.1016/j.fuproc.2011.07.015
11. Nanthagopal, K., Ashok, B., Thundil Karuppa Raj, R.: Influence of fuel injection pressures on
Calophyllum inophyllum methyl ester fuel led direct injection diesel engine. Energy Convers.
Manag. 116, 165–173 (2016). http://dx.doi.org/10.1016/j.enconman.2016.03.002
12. Sukumar, P., Jegan, R., Bala Subbramanian, K., Nagarajan, G.: Effect of injection pressure
on performance, emission and combustion characteristics of high linoleic linseed oil methyl
ester in a DI diesel engine. Renew Energy 34, 1227–1233 (2009). http://dx.doi.org/10.1016/j.
renene.2008.10.001
13. Purushothaman, K., Nagarajan, G.: Effect of injection pressure on heat release rate and emis-
sions in CI engine using orange skin powder diesel solution. Energy Convers. Manag. 50,
962–969. http://dx.doi.org/10.1016/j.enconman.2008.12.030
14. Kannan, K., Udayakumar, M.: Experimental study of the effect of fuel injection pressure on
diesel engine performance and emission. ARPN J. Eng. Appl. Sci. 5(5), 42–45 (2010)
15. Imtenan, S., Ashrafur Rahman, S.M., Masjuki, H.H., Varman, M., Kalam, M.A.: Effect of
dynamic injection pressure on performance, emission and combustion characteristics of a
compression ignition engine. Renew. Sustain. Energy. Rev. 52, 1205–1211 (2015). https://
doi.org/10.1016/j.rser.2015.07.166
16. Venkanna, B.K., Venkataramana Reddy, C.: Influence of injector opening pressures on the per-
formance, emission and combustion characteristics of DI diesel engine running on calophyllum
inophyllum linn oil (Honne oil). Int. J. Renew. Energy 6(1), January–June 2011
17. Srinidhi, C., Madhusudhan, A.: A diesel engine performance investigation fuelled with nickel
oxide nano fuel-methyl ester. Int. J. Renew. Energy Res. 7(2), 676–681 (2017)
Chapter 21
Design and Manufacturing of Hand
Injection Mold for Flip Badge—Analysis
of Alternate Air Venting Patterns
Vijaykumar Kalwa and Balaji Thottempudi
Abstract The increased demand for plastic components and greater requirement in
the market has created necessity for wider study and research in the area of plas-
tic flow behavior for intricate profiles of core and cavity in mold. Polymers rapidly
replaced other materials because of its chemical and physical characteristics. Best
part of the polymer is being lighter in weight and functionally suitable to large amount
of parts/components produced day to day. The influence of venting and the injection
time on the mold impression filling were examined. It was observed that improper
injection time and incorrect venting parameters lead to either short filling of the
impression or flash on the component. Also, component profile has vital role in plac-
ing the venting region. The controlled injection time and optimum venting parameters
found to have zero flash and 100% impression filling. The functionality/application
of the produced part is also found to be satisfactory.
Keywords Flip badge · Air trap analysis · Shrinkage · Venting
21.1 Introduction
21.1.1 Background Theory
A hand mold is a simple mold used for low-quantity work. It is used in the injec-
tion molding for replicating components in large quantity. In injection molding, hand
molds refer to simple molds that have no provision for heat, cooling, or ejection. They
are usually only of a two- or three-plate design because of the simplicity of the parts.
This research for hand mold comprises planning, detailed design, manufacturing,
V. Kalwa (B)
Ramaiah University of Applied Sciences, Bangalore 560058, India
B. Thottempudi
CreditAccess Grameen Limited, Bangalore 560058, India

https://doi.org/10.1007/978-981-15-3254-2_21
228 V. Kalwa and B. Thottempudi
machining non-identical vent profiles, costing, inspection, and trials of the compo-
nent. The purpose of the examining over the distinct venting pattern is to make aware
of true venting parameters, alternate vent profile pattern for unique component, flow
analysis, and analysis of air venting method and manufacturing procedures theoreti-
cally as well as practically. This paper reveals the analysis of alternate venting pattern
by considering component profile and important factors like design, manufacturing,
and troubleshooting the online and offline occurred difficulties. Component selected
for making hand injection mold is flip badge. A small object (such as tag) that is held
up by a person so that it can be easily seen, that has writing and often a picture on it,
represents the occasion or identity of person.
21.1.2 Objectives
1. To design a prototype hand mold for the given product (flip badge) and prove the
molding.
2. To manufacture the mold by following all the procedures included in a
production/multicavity molds.
3. To study in detail the total process involved like venting in mold making by
practically doing it.
4. To never compromise on quality of the product and produce each and every
element of the mold as per the given drawings and within the tolerance limits.
5. To understand the difficulties involved for filling the component caused by the
entrapped air in the impression and troubleshoot the venting issues.
21.2 Experimental Work
Experimental work is designed to successfully reach the objectives set for the research
work. It broadly consists of tool description, tool design, validation, tool fabrication,
and testing of tool and analysis.
21.2.1 Tool Description
This section highlighted the parameters adopted for tool design. It consists of descrip-
tion of the component, geometrical details, material properties, and tool specifications
[2, 3] (Table 21.1).
21 Design and Manufacturing of Hand Injection Mold for Flip Badge … 229
Table 21.1 Component

Particulars Specifications
specification
Component Flip Badge
Area of the component 2431.64 mm2
Area of the feed system 29.04 mm2
Total projected area 2460.63 mm2
Thickness of the component 2.5 mm
Mass of the component 5.6 grams
Material Polypropylene
Shrinkage 1.50%
Melting temperature 160 °C
Density of material 946 kg/m3
21.2.2 Shrinkage Calculation
The product quality of injection-molded plastic parts is the result of a complex com-
bination of the material used, the part and mold designs, and the process conditions
used to manufacture them. If plastics are processed by injection molding, deviation
of the molding from the dimensions of the cavity cannot be avoided. These devi-
ations from the nominal size are summarized under the term shrinkage [1]. In the
injection-molding technique, shrinkage S, as explained by Eq. (21.1),

S
D = A+ A∗ (21.1)
100
21.2.3 Detailed Design
In this paper, for analyzing the objectives, tool design was carried out in 2D drafting
application (AutoCAD) which included plan view for core half and sectional front
view. Also, 2D plan view is validated by modeling with same dimension as 2D in
3D environment with the help of Creo 2.0. It confines that dimensions applied were
proper and no interference found between the parts of the mold (Fig. 21.1; Tables 21.2
and 21.3).
21.2.4 Tool Validation
The validation of the designed tool was carried out in SOLIDWORKS mold flow
analysis package, especially for fill time and air trap analysis. In the obtained results,
Fig. 21.1 Prototype hand

mold for flip badge
Table 21.2 Shrinkage calculation for core and cavity

S. No. Actual dimension (mm) Shrinkage calculation (achieved dimension, mm)
1 58 58.87
2 50 50.75
3 46 46.69
4 42 42.63
5 11 11.16
6 2.5 2.53
Shrinkage calculation for cavity
S. No. Actual dimension (mm) Shrinkage calculation (achieved dimension, mm)
1 42 42.63
2 0.5 0.50
expected errors were taken into consideration for troubleshooting by employing

suitable machining parameters and methods. The below-mentioned result images
prove that required fill time for selected component is 1.6 seconds and highlighted
on image with dot marks for machining air vents (Figs. 21.2 and 21.3).
Table 21.3 Tool

Specifications Description
specification
No. of impressions 01
Type of mold Two-plate mold
Type of runner Round runner
Type of gate Edge gate
Type of ejection Pin ejection
Mold size 110 * 20 * 125 mm3
Fig. 21.2 Fill time analysis
Fig. 21.3 Air trap analysis

Fig. 21.4 Photographic view of milling machine with rotary table attachment for investigation
Table 21.4 Parts’ machining details

S. No. Particulars Material Machines/tools
1 Core/cavity D2 steel CNC milling, spark EDM,
conventional milling, co-ordinate
drilling, grinding, lapping
2 Ejector plates and bottom plate OHNS Milling, grinding, co-ordinate
drilling
3 Ejector pins/pushback pins/rest OHNS Facing, turning, cylindrical
buttons grinding
21.2.5 Fabrication of the Tool
The parts for the hand mold were precisely machined by utilizing conventional
and non-conventional machines. The critical parts of the mold like core/cavity are
machined by CNC milling and spark EDM. Once after pre-machining, secondary
operations were carried out for all the parts of the mold. As like design consider-
ations, required tolerance on parts was maintained to maintain accurate assembly
of the parts [4, 5]. In the below-mentioned table, fabrication details of the tool are
briefly documented (Fig. 21.4; Table 21.4).
21.2.6 Surface Validation for Core and Cavity
Once after machining, core and cavity with required tolerance need to be checked for
surface matching to ensure no gap between the plates. To ensure the proper mating of
core and cavity, blue applied over both surfaces and assembled. On a assembled core
and cavity plates, pressure applied to ensure even spreading of blue on both plate
surfaces. After performing blue matching on core and cavity, both surfaces found to
be flat (Fig. 21.5).
Fig. 21.5 Blue matching for

core and cavity
21.2.7 Air Vents’ Machining
Standard dimensions preferred for venting slots are 0.05 and 3 mm deep with alternate
shapes. To escape the entrapped air in the impression, air vents should be facilitated
by machining. To ensure proper escaping of entrapped air in cavity, air vents to
be machined on to the core plate of the mold in variety shapes and with different
machining methods. In this experimentation, machining for air vents is carried out
which is shown in below given photographs (Fig. 21.6; Tables 21.5 and 21.6).
Pattern2
Pattern 4
Pattern 1
Pattern
Fig. 21.6 Representation of alternate vent pattern machined
Table 21.5 Machining details for air vents

Vent pattern Machining method Parameters Remarks
Pattern 1 Surface grinding Linear cut (0.05 mm deep Component short filled
and 3 mm wide)
Pattern 2 Milling machine Circular slot (dia. 8 and Component short filled
0.2 mm deep)
Pattern 3 Milling machine Diagonal cut to the Component short filled
impression (8 mm dia and
0.2 mm deep)
Pattern 4 Milling machine Milling cut which connects Filled
from impression to the
circular slot (0.2 mm deep)
Table 21.6 Characteristics of machines adopted for tool fabrication

S. No. Characteristics Specification Machines/feed rate
1 Tool material Copper, 1 mm diameter Spark EDM
2 Spark gap 10–125 µm
3 Spark frequency 210–450 kHz
4 Peak voltage 40–250 V
5 Dielectric fluid Kerosene and water with Glycol
9 Tool material High carbon steel (face mill 5 Vertical milling feed rate
teeth) (0.02 mm/rev)
10 Cutter diameter 76 and 120 mm
12 Tool material High carbon steel Turning/facing feed rate
13 Cutting tool Single-point cutting tool (right (0.1–0.2 mm/rev)
hand)
15 Tool material Abrasive wheel (aluminum Surface grinding feed rate
oxide) (0.6 mm/rev)
21.3 Discussion and Conclusions
21.3.1 Testing and Inspection
In this last stage, testing the fabricated tool by injecting molten plastic (polypropy-
lene) was done by the plunger-type machine. Details of the trails are taken, and
troubleshooting of occurred problems has done and briefed below (Fig. 21.7;
Table 21.7).
Fig. 21.7 Component trails

for tool validation
Table 21.7 Trails’ observation

Trails Status Remark/issue
Trail 1 Short filled Air vents/injection time
Trail 2 Short filled Venting/Injection time
Trail 3 Short filled Venting
Trail 4 Short filled Venting
Trail 5 Filled Break at weld line
Trail 6 Filled Optimum injection time and proper venting provided
21.3.2 Observation
(I) Problems occurred during component trial

• Impression short filling.
• Flash on the component.
(II) Troubleshooting by
• Providing venting.
• Reducing injection time.
21.4 Conclusion
In this experimental process, the obtained component was inspected for its quality
and application and found to be satisfactory. Due care was taken to manufacture
the tooling following all the procedures, simulation results, and guidelines of a pro-
duction mold. Summary of air vent patterns and troubleshooting of occurred errors
noted in the paper understandably. The component could be practically used for its
application. The objectives of research were successfully achieved, and the processes
involved were clearly analyzed and summarized for further research work.
References
1. Ali, E.E., Balal, A.A., Abbas, R.A.: Shrinkage of plastic raw materials (a comparative study of
mold). JECS 18(3), 54–59 (2017)
2. DME Company (India) Ltd.: Mold Base Plastic Materials and Basic Assemblies Manual (2008)
3. Edutech NTTF IndiaPvt Ltd.: Mold Design Standards Manual. Training Centre Mumbai (2001)
4. Gerling, H., Heller, K.H.: All About Machine tools. New Age International Publishers, New
Delhi (2009)
5. HajraChowdhary, S.K.: Workshop Technology. Media Promoters and Publishers (1971)
Chapter 22
Design and Optimization of Foot Locus
Trajectory of Theo Jansen Mechanism
N. Prashanth, S. Karthik, G. R. Rahul and T. B. Sandarsh
Abstract The single degree of freedom legged robots are functional in terms of
its simplicity and performance on both even and uneven terrains. The performance
stability of these legged robots is determined by the trajectory of their foot locus. In
the present work, Theo Jansen mechanism is used as locomotive drive mechanism
for legged robot. The mechanism is designed and optimized using the Synthesis
and Analysis of Mechanism (SAM) software during the conceptual design phase.
The embodiment design was completed using CATIA. Simplex and evolutionary
algorithms were utilized for optimization of foot locus to improve its step height
and stride length. The influence of the link length and the input crank angle on the
foot locus trajectory has been illustrated and compared for three different trials. The
obtained results are compared with standard foot locus. Further, the results derived
in this work can be utilized in selecting the link length for Theo Jansen mechanisms
for various applications.
Keywords Planar mechanisms · Theo Jansen · Optimization · Synthesis · SAM ·

CATIA · Legged robot
22.1 Introduction
The legged locomotive drive mechanism for robots are advantageous compared to
that of wheeled drive mechanisms because of their all-terrain capability. Legged
robots employ a mechanism which can be broadly characterized as multi degrees
of freedom or single degree of freedom. Multi degrees of freedom legged robot
have high manoeuvrability and they require complex control schemes with multiple
gaits. Multi degrees of freedom legged robots are complex in mechanical design,
expensive and often results in poor reliability [1]. Single degree of freedom legged
mechanisms have several advantages to their credit because of its simplicity and
N. Prashanth (B) · S. Karthik · G. R. Rahul · T. B. Sandarsh

Department of Mechanical Engineering, Nitte Meenakshi Institute of Technology affiliated to
Visvesvaraya Technological University, Belagavi, India

https://doi.org/10.1007/978-981-15-3254-2_22
238 N. Prashanth et al.
low cost. Single degree of freedom legged mechanisms adopt planar mechanisms
which are predictable and require simple control schemes. The inherent nature of
planar mechanisms ensures that they do not suffer from redundant mobility. Due to
these advantages, they are widely used in several applications. The performance of
legged robots can be analysed using various metrics such as stability margin, energy
efficiency and foot locus trajectory. For single degree of freedom legged robot, foot
locus trajectory influences the robot’s ability to overcome obstacles and also the
energy efficiency of the mechanism [2]. Foot locus is the path traced by the leg
in one cycle of crank. The trajectory with which the foot of a walking mechanism
comes in direct contact with the ground is very important. As the crank rotates, the
foot traces out a cyclical path. The foot locus of the leg is defined by mainly four
phases; stride/drive, lift, return and lower phases.
Theo Jansen mechanism is a single degree of freedom mechanism which is widely
used as locomotive drive mechanisms for legged robots. The Theo Jansen’s linkage
is a planar leg mechanism designed by the kinetic sculptor Theo Jansen to simulate a
smooth walking motion. Jansen’s linkage bears artistic as well as mechanical merit
for its simulation of organic walking motion using a simple rotary input [3]. The
foot locus of single degree of freedom legged robot should be optimized and the
trajectory of the foot locus should be controlled for stable locomotion of the robot.
The step height and stride length of legged locomotion determine the terrain in
which the robot can move and efficiency of the robot locomotion [4]. In the present
work, the optimization and the performance analysis of Theo Jansen mechanisms
are studied using Synthesis and Analysis of Mechanisms (SAM) software. SAM
was developed by Artas Engineering software. SAM is an interactive PC-software
package for design, analysis (motion and force) and optimization of arbitrary planar
mechanisms. Mechanisms can either be generated via the design wizards or they
can be assembled from basic components including beams, sliders, gears, belts,
springs, dampers and friction elements [5]. CATIA was employed to design the basic
structures of the robot and planar mechanism. Digital mock-up (DMU) kinematic
analysis on the designed mechanism was performed to check the interference of
linkages and structures of the robot.
22.2 Optimization of Mechanism
Figure 22.1 shows the method adopted for analysis and design of Theo Jansen mech-
anism. The standard link lengths of the Theo Jansen mechanism are obtained from
the 11 holy numbers given by Jansen [6]. It is scaled down to the link lengths to
our requirement by taking the ratios of the 11 holy numbers. The scaled link lengths
were then utilized to construct the Theo Jansen mechanism in SAM and the corre-
sponding foot locus was obtained. Table 22.1 shows the link lengths and the foot
locus traced by the standard Theo Jansen mechanism. From the standard foot locus,
optimization was performed based on the foot locus requirement for uneven terrains.
The parameter of the foot locus is shown in Fig. 22.2.
22 Design and Optimization of Foot Locus Trajectory of Theo Jansen … 239
Analysis Phase Design Phase
Single DoF Legged Need

Mechanism Embodiment Design of the
Robot
Construction of Planar
Mechanism using SAM Design
DMU Kinematics for
Interference check
CATIA
Simulation of Foot Locus in Optimization of Link Cross
Simulation
SAM Section Thickness
Iteration
Analysis of Performance Detailed

Metrics such as Angular Engineering
Velocity, Angular Acceleration Analysis
Drawing
for different Crank Angle Input
at different foot locus nodes
Fabrication of the
Mechanism & the Robot
Synthesis of Theo Jansen
Mechanism to Optimize the Synthesis
Performance
Fig. 22.1 Methodology
Table 22.1 Standard link length and foot locus of Theo Jansen mechanism
Links Standard link length (mm) Standard foot locus result (mm)
1 60.108 Step height: 23.074 mm
2 57.812 Stride length: 114.254 mm
3 12
4 23.840
5 78.698
6 88.691
7 63.281
8 56.723
9 56.143
10 60.455
11 101.825
12 83.304
13 93.595
Fig. 22.2 Foot locus A

parameter [4]
Step
Height Lower
LiŌ
Drive
B C
STRIDE
Fig. 22.3 Hodograph of

Theo Jansen mechanism
obtained from SAM
The hodograph obtained of Theo Jansen mechanism for one complete cycle is
shown in Fig. 22.3.
The optimization of standard foot locus trajectory was performed by the following
steps.
• Modelling of Linkage Mechanism: Standard Theo Jansen mechanism was
constructed in SAM using the beam elements.
• Configuring the Inputs (Actuators) and Constraints: The input motion is given
to the crank for the rotation of the crank.
• Tracing Standard Foot Locus: The standard foot loci for the standard link lengths
were obtained. The coordinates of the standard foot locus were obtained.
• Creating the Optimized Foot Locus Coordinates: Based on the standard foot
locus coordinates, a set of new target coordinates were derived to optimize the step
height and stride length.
• Input Target Coordinate for Synthesis of Links: The coordinates for target path
were selected in the software.
• Define the Reference Frames: The fixed supports of the mechanism are selected
in the optimization process.
Optimized target
foot locus Standard
foot locus
Modified Step
Height
31.163
Std Step
Height
23.074 Standard Stride
114.254
Length
Modified Stride
116.574
Length
Fig. 22.4 Standard and optimized foot locus
• Select the Optimization Algorithm: In the optimization, one can choose between
user-controlled modus and automatic modus. In the user-controlled mode, a global
exploration of the parameter space is performed based on evolutionary algorithms
leading to a set of solutions. The user can then select any of these solutions and fur-
ther refine it via a local optimization, which can be either based on an evolutionary
approach or on a simplex method. In the automatic mode, the best solution of the
global search is automatically refined in a local optimization (again, either based
on an evolutionary or a simplex approach). In the present work, the automatic
modus is chosen as it gives the best solution.
• Optimize: After defining all the inputs, the optimization process is commenced;
the software performs multiple iterations to obtain the target foot locus path
obtained as shown in Fig. 22.4. During the process of optimization, the software
will automatically synthesize the link length to obtain the desired foot locus.
• Tabulation of New Link Lengths: Once the desired foot locus was obtained
through optimization, the updated link lengths were tabulated for CAD modelling.
As shown in Fig. 22.4, the target coordinates were entered as the reference file for
the optimization process. The thick lines represent the standard foot locus and the
dotted line represents the optimized foot locus. Using the above foot locus, several
other foot loci were obtained by changing the link lengths to achieve longer stride
length and higher step height.
Figure 22.5 shows the comparison of three different foot loci with that of the
standard. First, the stride length and step height were compared with the standard
foot locus and all the linkages. Percentage of increase or decrease of the link length
was computed to identify the links which influenced the foot locus the most. These
results were further tabulated as shown in Table 22.2. In the first trial, the stride
length and step height are increased by 2.03% and 35.05%, respectively. This foot
locus has higher step height and longer stride length when compared to all other
trials; however, the main problem associated with this foot locus is the mechanism
suffers from interference between the links during motion, the reason for interference
is the increase in link length leading to higher step height, in order to overcome this
problem, the cross-section thickness of link should be very small, however, reduction
in cross-section thickness will further lead to poor leg strength, therefore, trial-1 was
Trial-1 Trial-2 Trial-3

5 5 5
6 6 7 6 7
7 4 4 4
2 1 2 3 1 2 3
1 3
10 10 13
10 13 13
8 8
8
9
9 9
11 12 11 12
11 12
Fig. 22.5 Various foot loci in comparison with standard foot locus
Table 22.2 Nodes identified for the comparison of Theo Jansen foot locus
Link No. Standard Trial-1 Trial-2 Trial-3
link Link % of Link % of Link % of
length length change length change length change
(mm) (mm) in link (mm) in link (mm) in link
length length length
(%) (%) (%)
1 60.108 60.108 0.00 60.32 0.35 61.832 2.87
2 57.812 57.812 0.00 57.812 0.00 57.812 0.00
3 12 12 0.00 12 0.00 12.12 1.00
4 23.84 25.153 5.51 20.96 −12.08 23.408 −1.81
5 78.698 72.952 −7.30 80.273 2.00 78.698 0.00
6 88.691 88.691 0.00 87.96 −0.82 91.419 3.08
7 63.281 63.281 0.00 63.281 0.00 63.281 0.00
8 56.723 58.317 2.81 69.57 22.65 64.089 12.99
9 56.143 56.679 0.95 58.8 4.73 58.601 4.38
10 60.455 63.01 4.23 71.57 18.39 68.92 14.00
11 101.825 102.1 0.27 91.813 −9.83 93.677 −8.00
12 83.304 76.1 −8.65 68.965 −17.21 72.902 −12.49
13 93.595 90.623 −3.18 105.627 12.86 103.7 10.80
Stride 114.254 116.574 2.03 95.5 −16.41 102.6 −10.20
length
Step 23.074 31.163 35.06 20.492 −11.19 27 17.01
height
not considered for further analysis. In the second trial, the stride length and step
height were decreased to 16.41% and 11.19%, respectively. This type of foot locus
will be more suitable for even terrains, since its step height is less compared to all
other foot loci, and hence, this locus was not further considered for analysis. In the
third trial, there is an increase in the step height by 17.01% which is suitable to
overcome the obstacles of up to 27 mm and there is a decrease in the stride length
by 10.2%. Due to the increased step height, it is inferred that foot locus obtained in
trial-3 is suitable for both even and uneven terrains. From the above results, the foot
locus obtained during trial-3 was chosen as the optimal and it was further considered
for embodiment design.
22.3 Design of Legged Robot
Theo Jansen mechanism was designed using 3D CAD software CATIA V5. The
legs of the robot were modelled using the link lengths obtained from SAM software.
These links were then assembled to form the leg mechanism. Total of eight Theo
Jansen mechanisms was modelled. The chassis for the robot was designed using
sheet metal workbench. Pair of each leg was actuated by one geared DC servo motor,
and gear drives were used to transmit power to the legs of the robot. Figure 22.6
shows the assembly of the legged robot.
To ascertain the interference between links and chassis in the robot, digital mock-
up (DMU) kinematics analysis was performed in CATIA. Any interference in the
Fig. 22.6 Assembly of the legged robot

Fig. 22.7 Interference check

in CATIA
individual link lengths was overcome by optimizing the thickness of the links and
also by creating the required stand-off distances between the links (Fig. 22.7).
22.4 Performance Analysis
In this section, the standard foot locus is compared with that of the three different loci
obtained during optimization. The foot locus trajectory of Theo Jansen mechanisms
is unique for each trial, and hence, they cannot be compared in terms of its path alone;
therefore, critical nodes were identified along the foot locus which is common. These
nodes are shown in Fig. 22.8.
Table 22.3 shows the influence of identified nodes on the foot locus generated by
the Theo Jansen mechanism.
Table 22.4 shows the angular velocity and angular acceleration of the standard
Theo Jansen robot leg (link which makes contact with the ground) as the crank
rotates. Tables 22.5, 22.6 and 22.7 show the crank angle versus angular velocity and
angular acceleration of the robot leg (link which makes contact with the ground)
for the three trials, respectively. The analysis of angular acceleration of leg plays an
important role, since the acceleration and frictional forces have interaction during
Fig. 22.8 Major points A

affecting the foot locus
Step
Height
E
B C D
Stride Length
Table 22.3 Influence of nodes identified for the comparison of Theo Jansen foot locus
Nodes Influence on mechanism
A Step height determines the ability to overcome obstacles
B Touchdown determines the point of contact with ground
C and D Determines the stability during the stride length
E Post lift-off retardation and approach towards maximum step height
Table 22.4 Performance parameters of Theo Jansen mechanism for standard link lengths
Nodes Crank angle (rad) Angular velocity (rad/s) Angular acceleration (rad/s2 )
A 3.362 −4.696 −151
B 5.037 1.003 15.41
C 5.875 0.6952 11.6
D 6.713 2.194 9.959
E 4.2 −4.132 −35.16
Table 22.5 Performance parameters of Theo Jansen mechanism for trial-1

A 3.382 −2.09 −69.662
B 5.128 0.347 2.776
C 5.996 0.348 2.927
D 7.252 1.56 1.055
E 2.693 0.08 −10.114

A 3.492 −3.771 −15.32
B 5.06 0.769 13.617
C 5.972 0.865 9.613
D 6.951 2.116 5.919
E 3.07 −1.856 −37.78

A 3.58 −2.621 −64.04
B 4.837 1.363 15.63
C 5.675 0.380 10.96
D 6.512 1.773 10.6
E 3.161 −4.937 −98.36
the locomotion of the robot, and hence, it is imperative to understand the magnitude
and directions of various parameters such as velocity and acceleration [7–9].
Figures 22.9 and 22.10 represent the crank angle versus angular velocity and
crank angle versus angular acceleration, respectively, for the optimized foot locus,
i.e. for trial-3. From the above graphs, one can observe that there is a rapid change in
velocity from node A to B and Node D to E resulting in change of acceleration. The
velocity from point B to D is almost uniform and therefore it results in minimum
acceleration. The rapid change in velocity and acceleration happens during the lift
and lower phase.
Figure 22.11 shows the comparison of angular acceleration between the standard
Theo Jansen mechanism and the optimized foot loci. It can be observed that between
nodes B, C and D, the variation of angular acceleration across the standard foot
locus and optimized foot loci is negligible. However, at nodes A and E, the angular
acceleration varies across all the foot loci. In the standard foot locus, at node A, the
leg experiences maximum acceleration and its acceleration reduces as its approach
Crank Angle Velocity

7 3
6.512
1.363
Angular Velocity (rad/sec)

5.675 2
6 1.773
1
Crank Angle (rad)
5
0
4.837 0.308
4 3.58 -1
3.161
3 -2
-3
2 -2.621
-4
1 -4.937 -5
0 -6
A B C D E
Nodes
Fig. 22.9 Trial-3 crank angle versus angular velocity
Crank Angle Acceleration

7 40
6.512
Acceleration (rad/sec2)
15.63 10.96
6 20
Crank angle (rad)
10.6
5 5.675 0
-20
Angular
4 3.58 4.837
3.161
-40
3
-60
2 -64.04 -80
1 -100
-98.36
0 -120
A B C D E
Nodes
Fig. 22.10 Trial-3 crank angle versus angular acceleration

Standard Trial1 Trial2 Trial3

40
20
Angular AcceleraƟon (rad/sec2)
0
-20
-40
-60
-80
-100
-120
-140
-160
A B C D E
Nodes
Fig. 22.11 Comparison of angular acceleration between standard and optimized foot loci
towards node B. In trial-3, it can be observed that the difference between angular
acceleration between nodes A and E is considerably less than the standard foot locus.
22.5 Conclusions
While comparing the standard foot locus with that of the optimized foot loci, one
can observe that in the standard link the maximum step height is achieved at the
centre of the stride length, and hence, when the leg approaches the lower phase, it
would not overcome the obstacle. In the optimized foot locus, the maximum step
height is achieved during the lower phase and such foot loci are suitable for uneven
terrains or overcome obstacle (obstacle size based on step height). Hence, it can be
used in several applications. The results obtained in the present work can be further
utilized to incorporate Theo Jansen mechanism for various applications. Further,
the optimized mechanism will be fabricated and integrated with the robot to test its
performance under real-world conditions such as different terrain, gaits and phase
angles.
References
1. Tenreiro Machado, J.A., Silva, M.: An overview of legged robots. In: International Symposium
on Mathematical Methods in Engineering, pp. 1–40 (2006)
2. Nansai, S., Rajesh Elara, M., Iwase, M.: Dynamic analysis and modeling of Jansen mecha-
nism. Procedia Engineering 64, 1562–1571 (2013)
3. Moldovan, F., Dolga, V.: Analysis of Jansen walking mechanism using CAD. In: Solid State
Phenomena, vol. 166, pp. 297–302. Trans Tech Publications Ltd (2010)
4. Shigley, J.E.: The mechanics of walking vehicles. No. ATAC-LL71. ARMY TANK-
AUTOMOTIVE CENTER WARREN MI (1960)
5. SAM-The Ultimate Mechanism Design Software. [Online]. https://www.artas.nl/en/sam/
features/general
6. Jansen, T.: Theo Jansen’s strandbeest (2011). Retrieved February 26, 2011
7. Ghassaei, A., Choi, P.P., Whitaker, D.: The Design and optimization of a crank-based leg
mechanism. Pomona, USA (2011)
8. Mohsenizadeh, M., Zhou, J.: Kinematic analysis and simulation of Theo Jansen mechanism.
Diss. Lamar University (2015)
9. Ingram, A.J.: Numerical kinematic and kinetic analysis of a new class of twelve bar linkage for
walking machines. Diss. Rand Afrikaans University (2004)
Chapter 23
Performance Evaluation of Concentric
and Eccentric Buckling Restrained
Braces on the Dynamic Behaviour of RC
Structures
H. J. Prajwal Shankar, Sunil Lamsal, Praveen Shrestha, Bharathi Ganesh
and R. Prabhakara
Abstract Energy absorption and damage control are the two fundamental principles
adopted in the earthquake resistant design of structures. This is possible through the
usage of appropriate building materials, components, construction methodology and
utilizing the advanced technology. Several lateral resisting systems were developed
in order to enhance structural performance against earthquakes. Buckling restrained
braces (BRBs) have been evolved as one of the promising lateral load resisting sys-
tems. The present study deals with the comparison of seismic performance of an RCC
frame with BRBs over bare frame (BF) and shear wall (SW) lateral load resisting
systems in a RC structure of 19 stories located in seismic Zone V, analyzed using
an integrated structural analysis and design software ETABS version 16. 2. 0 using
response spectrum method. It was observed with BRBs as lateral load resisting sys-
tem performed better over BF and SW systems compared over several performance
parameters.
Keywords Concentric buckling · Eccentric buckling · Buckling restrained braces ·

Energy absorption · Damage control
H. J. P. Shankar · S. Lamsal (B) · P. Shrestha · B. Ganesh

Department of Civil Engineering, Nitte Meenaski Institute of Technology, BSF Campus
Yalahanka, Bangalore, Karnataka, India
H. J. P. Shankar
P. Shrestha
B. Ganesh
R. Prabhakara
Structural Engineering Division, PGS VTU, Belagavi, Karnataka 590018, India

https://doi.org/10.1007/978-981-15-3254-2_23
250 H. J. P. Shankar et al.
23.1 Introduction
The objective of seismic design of a structure is to limit the deformation within the
permissible limits so as to ensure that the structures resist the earthquake of assumed
intensity falling under the designated zone. The specified limit of serviceability is
achieved by selecting an appropriate lateral load resisting systems such as moment-
resisting frame (MRF), diagonal bracing (DB) and centre braced frame (CBR) [1].
However, these structures typically generate large inelastic deformation which can
cause severe damage to the structure and subsequently increase the complexity of
handling the damaged structure and cost of structural repair post-earthquake. It has
been observed in the recent years, buckling restrained braces (BRBs) have become an
attractive and feasible alternative to the conventional steel brace system, provided for
dissipating earthquake energy [2]. BRB is an unbounded bracing member consisting
of a steel core plate or a section encased in a concrete-filled steel tube over its length
(Figs. 23.1 and 23.2) [3–5]. The former component resists axial force while the later
component restrains the brace from buckling in compression.
The better seismic performances of BRBs have widely replaced the conventional
steel braces in recent years. BRBs benefit from symmetric and stable hysteretic
behaviour in compression with normal braces, and as a result, BRBs have higher
energy absorption capacity. The energy dissipation in bracing is through yielding of
brace in tension. In compression, the buckling before yielding results in less energy
dissipation and causes instability of a structure [5]. This weakness of lateral bracings
Fig. 23.1 Buckling restrained bracing [6]
Fig. 23.2 Components of a BRB [7, 8]

23 Performance Evaluation of Concentric and Eccentric Buckling … 251
is tried to overcome with a typical brace that can provide restrained against buckling
in compression and named as buckling restrained brace which exhibits symmetric
and stable cyclic behaviour and also high energy absorption [9, 10].
23.2 Present Study
The present study focuses on the analysis of a 19 storey RCC framed structure with
concentric and eccentric BRBs to record the seismic performance in comparison with
that of shear walled RCC framed structure. The geometric and material details of the
structure are provided in Table 23.1. The plan of a live project building is selected for
the study with minor modifications suitable to the present research study (Fig. 23.3).
There are 16 bays in major direction and nine bays in the other direction. Different
models considered for the analysis are the structure with bare frame, structure with
Table 23.1 RCC geometric

Type of building Residential apartment
and material details of framed
structure No of stories 19
Floor to floor height 3m
Total height of building 57 m
Plan area 25.025 m × 52.75 m
Beam size 200 × 550 mm
Column size 450 × 750 mm
Slab thickness 125 mm, 150 mm
Concrete grades M30, M40
Steel grades Reinforcing: Fe 500, Fe 415
Structural: Fe 345
Fig. 23.3 Typical floor plan of the structure

shear wall and also the structure with concentric and eccentric BRBs as lateral load
resisting systems [11–13].
23.3 Methodology
Following methodology is adopted while carrying out the analysis of the structure.
ETABS ver 16.2.0 integrated software for analysis and design of structures is used
to carry out linear dynamic analysis/response spectrum analysis. The geometric and
material details of the structure are as follows (Table 23.1).
23.4 Various Models Considered for Analysis
RCC bare frame without lateral load resisting system, located in seismic Zone V
is considered as a reference model. As per IS1893-2016, a Response Reduction
Factor (RRF) of 3 corresponding to an ordinary moment resisting frame was used
for the design. An RC wall 220 mm thick, made of M30 grade concrete, provided
with shear wall type of lateral load resisting system is taken up for the analysis. A
structure having RRF of 4.5 as per IS1893 (Part 1)-2016 provided with concentric
and eccentric buckling restrained bracings (BRBs)—a combination of shear wall and
steel bracings is considered for the analysis [14].
23.5 Procedure Followed in the Analysis
The procedure adopted for analyzing the structure is explained as follows.

a. Preparation of geometrical model for bare RCC frame.
b. Analyzing with and without shear wall for dynamic behaviour. Various types of
live loads are considered as per IS 875: 2008 [15] as provided in Table 23.2.
Defining the parameters of BRBs and applying calculated loads as per various
load combinations as given in Table 23.3. Calculated loads are assigned on the
respective floor areas in the model (Fig. 23.4).
c. Assigning the BRBs in predefined positions of bare RCC frame, in different
patterns, both concentric and eccentric separately.
d. Carrying out seismic analysis of RC frame with concentric and eccentric BRB
frames and records the results.
e. Comparing the seismic response of various types of model of RCC frame and
draw a rational conclusion.
f. Comparing the results of software analysis with manual calculations.
Table 23.2 Area loading

Sl. No. Description Loads in kN/m2
details
1. Corridors 3.0
2. Staircase 3.0
3. Rooms 2.0
4. Toilet area (75 mm sunken) 1.5
5. Terrace area (accessible) 2.0
6. Utility (150 mm sunken) 3.0
7. Balcony area (100 mm sunken 2.0
slab)
Table 23.3 Load

Sl. No Factored load combinations
combinations (response
spectrum analysis) 1 1.5TDL + 1.5TLL
2 1.5TDL + 1.5RS-X
3 1.5TDL + 1.5RS-Y
4 1.2TDL + 1.2TLL + 1.2RS-X
5 1.2TDL + 1.2TLL + 1.2RS-Y
6 0.9TDL + 1.5RS-X
7 0.9TDL + 1.5RS-Y
8 1.5TDL + 1.5WX
9 1.5TDL − 1.5WX
10 1.5TDL + 1.5WY
11 1.5TDL − 1.5WY
12 1.2TDL + 1.2TLL + 1.2WX
13 1.2TDL + 1.2TLL − 1.2WX
14 1.2TDL + 1.2TLL + 1.2WY
15 1.2TDL + 1.2TLL − 1.2WY
16 0.9TDL + 1.5WX
17 0.9TDL − 1.5WX
18 0.9TDL + 1.5WY
19 0.9TDL − 1.5WY
23.6 Design Using Conventional Method as Per Is456:2000
The design of BRBs was carried as per ASCE-7-05-Design Rule for Relative Strength
in a tall BRBF [16]. The steel of Fe 345 grade was used for core and plane cement
concrete of unit weight of 24 kN/m3 for casing members. A design excel sheet
was created to carry out the design with varying parameters and illustrated over the
following steps.
Fig. 23.4 Plan of a floor with load assignment
1. Calculation of the length of BRB (L) required in a bay by using trigonometric

functions.
2. Assuming the length of non-yielding core by keeping the stiffness of BRB as an
important parameter, (Reference publications) to achieve higher stiffness.
3. Defining the dimensions of non-yielding core such as width of core, thickness of
core, outer core concrete thickness in both directions and the thickness of outer
steel casing to achieve high strength and high stiffness as the main parameter.
4. Calculate the weight of non-yielding core.
5. Defining the dimensions of yielding core such as width of core, thickness of
core, outer core concrete thickness in both x and y directions and the thickness of
outer steel along with casing to achieve high strength and high stiffness as main
parameters.
6. Calculation of the weight of yielding core (W ).
7. Calculation of the total weight of BRB such as weight of yielding core and weight
of non-yielding core.
8. Calculation of the total stiffness of BRB using the Eq. 23.1.
K = (AE/L) cos 2θ (23.1)
where
A = area of BRB core, E = young’s modulus of steel used, 2.1 × 105 N/mm2
L = length of BRB, θ = angle of inclination of bracing.
23.7 Results
The present investigation was on the analysis of behaviour of 19 storey RC structures

located in Zone V. Four types of models, namely bare frame, shear wall, concentric
inverted V and eccentric inverted V BRBs, subjected to an earthquake force with Z
= 0.36 are analyzed. The performances such as storey displacement, storey drifts,
storey stiffness and storey shear of each model are tabulated Tables 23.4, 23.5, 23.6
and 23.7 and studied in comparison with the performances of bare framed model.
Table 23.4 Comparison of seismic responses for different cases considered

Sl No. Model Max. lateral Max. drift
displacement
(mm)
x y x y
1 Bare frame 140.8 80.7 0.003 0.002033
2 Shear walled 92.9 42.5 0.002025 0.000933
3 Concentric inverted V BRBF 84.3 48.1 0.001992 0.001221
4 Eccentric inverted V BRBF 84.6 49.6 0.001996 0.001267
Table 23.5 Comparison of seismic responses for different cases considered

Sl No. Model Max. storey stiffness Max. storey shears
(kN/m) (kN)
x y x y
1 Bare frame 4780808.8 6956439.3 15739.7 19172.2
2 Shear walled 13066206 42877887 24989.4 46095.4
3 Concentric inverted V BRBF 5235352 8381545 11429.3 14612.2
4 Eccentric inverted V BRBF 5201813 8003786 11337.9 14212.5
Table 23.6 Normalized values of performance of the structure over the in terms of performance
of bare frame model
Sl No. Model Max. lateral Max. drift
displacement (mm)
x y x y
1 Bare frame 1 1 1 1
2 Shear walled 0.65 0.52 0.68 0.46
Table 23.7 Normalized values of performance of the structure in terms of performance of bare
frame model
Sl No Model Max. storey stiffness Max. storey shears
(kN/m) (kN)
x y x y
1 Bare frame 1 1 1 1
2 Shear walled 2.73 6.16 2.73 6.16
23.8 Conclusions
The conclusions from the above analysis are as follows.

1. The maximum storey displacement for bare frame model in X directions has
exceeded the limit of displacement. This was directly proportional to the storey
stiffness and shear wall exhibited higher stiffness compared over BF and BRBs
models. However, storey shear was higher in the shear wall compared over other
two.
2. Modal participating mass ratio for a bare frame model shows a value of more
than 90% mass participation at the end of the twelfth mode.
3. Maximum storey displacement of shear walled model satisfies the upper limit
of 114 mm (H/500 = 57*1000/500) for a nineteenth storeyed building. The
maximum storey drift recorded is between storey 6 and storey 8 in X direction
and is within the specified limit of drift as per IS1893:2016.
4. Variation in terms of storey displacements of BRBs was found to be 9.2% less
than shear wall indicating that both concentric and eccentric BRBs patterns are
equally better in resisting the earthquake forces.
5. Storey displacement of shear wall was 17.3% better and storey drift of shear
walled model was better than that of BRBs along Y direction. The variation in
terms of storey drift is minimal between concentric X BRBs eccentric inverted V
models indicating that these patterns are equally good in resisting the earthquake
force.
6. The storey stiffness of BRBF model is lesser by 81.33% than that of shear walled
model. Since BRB system is more effective in resisting earthquake in Zone V
than shear walled system, reduced stiffness may lead to cost optimization also.
7. The variation in terms of storey shear is minimal indicating both concentric and
eccentric BRBF systems are equally good in resisting the earthquake force. The
storey shear at different storeys of RC structure with BRBF model is 68.75%
lesser than that of shear walled model signifying the use of BRBF in place of
shear wall in resisting the earthquake in Zone V. Use of BRBF system may
contribute to the optimization of mass of structure and hence cost also.
References
1. Medhekar, M.S., Kennedy, D.J.L.: Displacement-based seismic design of buildings—theory.

Eng. Struct. 22(3), 201–209 (2000)
2. Genna, F., Gelfi, P.: Analysis of the lateral thrust in bolted steel buckling-restrained braces. II:
Engineering analytical estimates. J. Struct. Eng. 138(10), 1244–1254 (2012)
3. Piedrafita, D., Cahis, X., Simon, E., Comas, J.: A new modular buckling restrained brace for
seismic resistant buildings. Eng. Struct. 56, 1967–1975 (2013)
4. Hosseini, A., Hassanipour, A.: Numerical modeling of BRB frame systems with and without
concrete. J. Multidisc. Eng. Sci. Technol. 2(8)
5. Sun, H., Jia, M., Zhang, S., Wang, Y.: Study of buckling-restrained braces with concrete infilled
GFRP tubes. Thin Walled Struct. 136, 16–33 (2019)
6. Piedrafita, D., Cahis, X., Simon, E., Comas, J.: A new perforated core buckling restrained
brace. Eng. Struct. 85, 118–126 (2015)
7. https://ascelibrary.org/doi/abs/10.1061/
8. https://www.sciencedirect.com/science/article/pii/S0141029613003805
9. Xie, Q., Zhou, Z., Huang, J.H., Meng, S.P.: Influence of tube length tolerance on seismic
responses of multi-storey buildings with dual-tube self-centering buckling-restrained braces.
Eng. Struct. 116, 26–39 (2016)
10. Fahnestock, L.A., Sause, R., Ricles, J.M.: Seismic response and performance of buckling-
restrained braced frames. J. Struct. Eng. 133(9), 1195–1204 (2007)
11. Prinz, G.S., Richards, P.W.: Seismic performance of buckling-restrained braced frames with
eccentric configurations. J. Struct. Eng. 138(3), 345–353 (2011)
12. Della Corte, G., D’Aniello, M., Landolfo, R.: Field testing of all-steel buckling-restrained
braces applied to a damaged reinforced concrete building. J. Struct. Eng. 141(1), D4014004
(2014)
13. Skalomenos, K.A., Inamasu, H., Shimada, H., Nakashima, M.: Development of a steel brace
with intentional eccentricity and experimental validation. J. Struct. Eng. 143(8), 04017072
(2017)
14. IS 1893 (Part 1): 2016.: Criteria For earthquake resistant design of structures—general
provisions and buildings
15. IS 875 (Part 1 & 2) 2008.: Code of practice for design loads (other than earthquake) for buildings
and structures dead loads and imposed loads
16. Aukeman, L.J., Laursen, P.: Evaluation of the ASCE 7-05 standard for dual systems: Response
history analysis of a tall buckling-restrained braced frame dual system. In: Structures Congress
2011, pp. 2707–2717 (2011)
Chapter 24
Analysis of Rotation Capacity of RC
Beams Over Formation of Plastic Hinges
R. Prabhakara, B. Nambiyanna, R. Nakul and Bharathi Ganesh
Abstract Rotation capacity of a joint in a structural member under constant moment

depends on the characteristics of the formation of plastic hinge. The rotation capacity
and plastic hinge formation in RC member depends on many parameters and there-
fore makes it different over the formation of plastic hinge in steel structural members.
It has been observed from the literature that the rotation capacity in RC structural
member is influenced by several parameters such as mechanical properties of con-
crete and steel, geometry of the member, external load and boundary conditions, thus
makes it to be an interesting structural behaviour which has to be explored very well.
Parametric analysis on the rotation capacity and formation plastic hinge was car-
ried out by collecting more than hundred data from the available literature. Relevant
graphs were plotted to demonstrate the variation of plastic hinge length (lp) over
the influencing parameters such as cross-sectional dimensions of the beams, grade
of the concrete, tension reinforcement and span-to-depth ratio. The length of plastic
hinge was calculated based on existing analytical models given by the researchers.
Owing to the advancement in the analysis of nonlinear behaviour of structures using
sophisticated software, the characterization of plastic hinges has to incorporated in
the analysis. These analytical models are very useful and provided more knowledge
R. Prabhakara (B)
Structural Engineering Division, PGS VTU, Belagavi, Karnataka, India
B. Nambiyanna
Civil Engineering Department, Ramaiah Institute of Technology, Bangalore, Karnataka, India
R. Nakul
Civil Engineering Department, Presidency University, Ittgalpura, Rajankunte, Bangalore,
Karnataka, India
B. Ganesh
Department of Civil Engineering, Nitte Meenaski Institute of Technology, BSF Campus
Yalahanka, Bangalore, Karnataka, India

https://doi.org/10.1007/978-981-15-3254-2_24
260 R. Prabhakara et al.
on the behaviour of RC structures. In this study, existing analytical models are eval-
uated based on the different parameters. The plastic hinge length was determined
by nonlinear regression analysis. It was found from the analysis that, large variation
was found over the length of plastic hinge for the existing plastic hinge models and
the proposed nonlinear regression analysis model yielded better results.
Notations
d Effective depth of beam, mm

l Effective length of the beam, mm
lp Plastic hinge length, mm
l p.reg Plastic hinge length by regression mm
z Distance from critical section to point of contraflexure, mm
f c Designed cylinder compressive strength in MPa
fy Specified yield strength of the reinforcement, MPa
ρ Tension zone reinforcement in %
ρ Compression zone reinforcement in %
θp Plastic rotation (rad)
24.1 Introduction
Engineers are facing the challenging situation in the analysis of RC structures

while providing safe and economical buildings to the society. Nonlinear analysis
has become one of the important means of achieving the above goal. RC structures
are made out of structural elements such as column, beam and slab; also their elas-
tic and inelastic behaviour determines the load response behaviour of the structure.
Rotation capacity and formation of plastic hinge in the structural members is one
of the main issues in understanding the inelastic behaviour of the structure. More
specifically in the earthquake design of structures, ductility plays a key role in the
design criteria set by many countries’ code provisions.
When the load applied is amplified, the hinge begins to form at locations where
the moment is equal to the plastic moment, if further increase in the applied load
then hinges transforming into plastic hinges and converting into mechanism followed
with failure. Framed structures are designed to achieve the above by proportioning
the beams and columns so that the majority of the plastic hinges are formed in
the beams not in the column. For this reason, an important parameter is the plastic
hinge, which controls the rotation capacity of the structural members. There are two
different types of plastic hinges formed in RC structure, such as reversing plastic
hinges and unidirectional plastic hinges. The reversing plastic hinge sustains both
positive and negative inelastic rotations in the same region, and unidirectional plastic
24 Analysis of Rotation Capacity of RC Beams over Formation … 261
hinge sustains both in different regions of the beam element during earthquake [1].
It was observed that plastic hinge length lp in RC beams influenced by a number of
parameters such as cross-sectional dimensions, yield and ultimate curvature, material
characteristics, tension and compression reinforcement ratios, support type, intensity
and nature of loading [2, 3].
The main objective of the present study is to propose a soft computing method
that can estimate the rotation capacity and length of plastic hinge by using the avail-
able data from experimental studies conducted on RC beams [4–9]. The regression
analysis was carried out to find out the influencing parameters on which the rotation
capacity and plastic hinge length and also to demonstrate the relationship between
the plastic hinge length and rotation capacity.
24.2 Rotation Capacity and Plastic Hinge Length
In the present investigation, the rotation capacity is defined as the rotation of critical
section of a beam from yield stage to the ultimate stage, and the plastic hinge length is
defined as the length of inelastic zone over which bending moment will be larger than
or equal to yield moment over the length of the beam. It is generally agreed that the
inelastic rotations are concentrated over a length called “plastic hinge length” where
M ≥ M y as shown in Fig. 24.1. The length of plastic hinge is dependent on many
parameters, such as the shape of the ultimate bending moment diagram, and is also
affected by the length between zero moment points and distance between support,
reinforcement ratio, the characteristic strength of concrete and depth of section [2].
The corresponding plastic hinge length is determined with respect to integration of
the curvature distribution for distinctive members.
The correlation between the cross section, ductility and the length of plastic hinge
needs to be accurately determined. It has been observed there is a large variation in
the plastic hinge length formulae proposed by the researchers as shown in Table 24.1.
Sawyer, Corley and Mattock, for example, considered only the member’s length and
depth. Priestley and Park consider the length and diameter of reinforcement of the
beam.
Fig. 24.1 Length of plastic

hinge
Table 24.1 Existing length

Reference Length plastic hinge
of plastic hinge (l p ) models
expression (l p )
[8]
Baker (1956) k(z/d)1/4 d
Sawyer (1964) 0.25d + 0.075z
√
Corley (1966) 0.5d + 0.2 d (z/d)
Mattock (1967) 0.5d + 0.05z
Priestley and Park (1987) 0.08z + 6d b
Paulay and Priestley(1992) 0.08z + 0.022d b
Sheikh and Khoury (1993) 1.0 h
Coleman and Spacone (2001) Gcf /[0.6f c (ε20 − εc + 0.8
f c /E c )]
Panagiotakos and Fardis 0.18z + 0.021d b
(2001)
Fig. 24.2 Variation of plastic hinge length (l p )
To be able to demonstrate the variations in the length of plastic hinge, the geometric
and material properties were taken as an example from [6] l = 6000 mm, b = 300 mm,
d = 540 mm, f’c = 30.9 MPa, z = 3000 mm and l/d = 11.1. For the above data,
plastic hinge lengths were obtained using the above equations and they are shown
in Fig. 24.2. It was observed a large variation in the plastic hinge length which
encouraged us to study on the parameters effect and regression analysis.
24.3 Parametric Study
In the present study, the central point loading configuration and different equivalent
length of beams were selected from the literature [4–9]. The range of the parameters
such as characteristics strength of concrete and steel, width and depth of beams,
Table 24.2 Range of

Parameters Parameter range
parameters used
f c (MPa) 20–129.1
f y (MPa) 250–678
l (mm) 1000–12,000
b (mm) 50–500
d (mm) 90–950
z (mm) 500–6000
ρ (%) 0.13–6.45
ρ (%) 0–4.84
l/d 4.17–21.82
tension and compression steel percentages and all other essential parameters that
affect the beam behaviour under central point loading is shown in Table 24.2.
24.4 Determination of Length of Plastic Hinge
The length of the plastic hinge lp was determined based on different parameters
obtained from the previous studies. The beam plastic rotation (θ p ) was calculated
from Eq. (1) [10].
0.004
θp = (1)
(xu /d)
The plastic hinge length lp was calculated based on Eq. (2) [2], and curvature at
yielding moment (ϕ y ) and the curvature at ultimate moment (ϕ u ) and plastic hinge
length values were analytically calculated.
θp θp
lp = = (2)
(ϕu − ϕy ) ϕp
Illustrative graphs have been plotted to observe the variation of lp against the
parameters which are shown in Fig. 24.3a–f.
It has been observed from Fig. 24.3a–f that the length of plastic hinge has not
shown much variation with grade of concrete, and however, reduction in lp was
observed with increase in reinforcement ratio and further it was observed that lp
increased when the depth of the beam increased, l/d ratio, z and z/d ratio.
Fig. 24.3 Plastic hinge length (lp) versus various parameters
24.5 Nonlinear Regression Analysis
Nonlinear regression analysis is a method of mathematical modelling. In this study,

cross-sectional dimensions, material characteristics, yielding and ultimate deflec-
tions, tension and compression reinforcement ratios were observed and parameters
that affect on length of plastic hinge are analysed by using the nonlinear regression
analysis. Formulated our regression equation for plastic hinge length (lp ) as follows.
0.059266
l
l p = 0.37 × b0.00896 × d 0.667 × f c−0.0016 × z 0.3264 × ρ −0.00207 × (3)
d
It has been observed from Fig. 24.4 that the comparison between regression plastic
hinge length l p reg versus existing models l p trend lines shows that the proposed model
is predicted well with Mattock, Sawyer, Corely and Baker’s models and R2 values
are almost equal to unity.
It has been observed from Fig. 24.5 that the comparison between regression plastic
hinge length l p reg versus θ p trend lines shows that there is no significant relationship
between l p and θ p .
24.6 Conclusions
Based on the parametric study and proposed nonlinear regression analysis, the
following conclusions are drawn.
Fig. 24.4 Comparison between regression l p reg and existing models l p
Fig. 24.5 Comparison

between regression l p and θ p
• The plastic hinge length results are obtained by using plastic rotation and curvatures
at ultimate and yield load (vide Eqs. 1 and 2).
• Demonstrative graphs have been plotted to show the variation of influencing
parameters. It was observed from Fig. 24.3 that the length of plastic hinge has
not shown much variation with grade of concrete, and however, reduction in lp
was observed with increase in reinforcement ratio.
• Further, it was observed that l p amplified with increase in depth, l/d ratio, z and
z/d ratio.
• Analytical expressions given in the literature indicated that only the cross-sectional
dimensions are under the bending effect of beam and other influencing parameters
were not considered.
• The plastic hinge length (l p ) obtained by expressions given by the researchers gave
different values and are not comparable (Fig. 24.2).
• It has been observed that analytical plastic hinge expression and proposed
nonlinear regression expression yielded results that are comparable.
• Based on the comparison between l p calculated by existing model and l p.reg calcu-
lated by nonlinear regression model, the results show that the nonlinear regression
plastic hinge model (l p.reg ) is versatile and can be used all types RC beams.
• Based on comparison between rotation capacity and length of plastic hinge, there
was no significant relation between them and hence, further investigation has to
be done over large number of data.
Acknowledgements We would like to thank the Management of Gokula Education Foundation,

Principal and HOD, Civil Engineering Department of RIT affiliated to VTU, Belagavi, for their
support extended to carry out this work.
References
1. Mendis, P.: Plastic hinge lengths of normal and high-strength concrete in flexure. Adv. Struct.
Eng. 4(4), 189–195 (2002)
2. Kheyroddin, A., Naderpour, H.: Plastic hinge rotation capacity of reinforced concrete beams.
Int. J. Civil Eng. 5(1), 30–47 (2007)
3. Gopinath, A., Nambiyanna, B., Nakul, R., Prabhakara, R.: Parametric study on rotation and
plastic hinge formation in RC beams. J. Civil Eng. Tech. Res. 2, 393–401 (2014)
4. Bigaj, A., Walraven, J.C.: Size effect on rotational capacity of plastic hinges in reinforced
concrete beams. CEB Bull. Inf. 218, 7–23 (1993)
5. Bosco, C., Debernardi, P.G.: Influence of some basic parameters on the plastic rotation of
reinforced concrete elements. CEB Bull. Inf. 218, 25–44 (1993)
6. Lopes, S.M.R., Bernardo, L.F.A.: Plastic rotation capacity of high-strength concrete beams.
Mater. Struct. 36(1), 22–31 (2003)
7. Fransson, H.: Rotation capacity of reinforced high strength concrete beams. Licentiate thesis.
TRITA-BKN. Bulletin 32 (1997) ISSN 1103-4270 ISRN KTH/BKN/B--32—SE
8. Zhao, X., Wu, Y.F., Leung, A.Y., Lam, H.F.: Plastic hinge length in reinforced concrete flexural
members. Procedia Eng. 14, 1266–1274 (2011)
9. Manyoung Ko, M.Y., Kim, S.W., Kim, J.K.: Experimental study on the plastic rotation capacity
of reinforced high strength concrete beams. Mater. Struct. 34(5), 302–311 (2001)
10. Carpinteri, A., Corrado, M., Mancini, G., & Paggi, M.: Size-scale effects on plastic rotational
capacity of reinforced concrete beams. ACI Struct. J. 106(6)
Chapter 25
Influence of Metallurgical Parameters
on Wear and Impact Characteristics
in High Chromium Manganese Irons
G. J. Naveen, P. Sampathkumaran, S. Seetharamu, T. Pramod,

Ashirwad Pandey and A. Sathyanarayanaswamy
Abstract Nickel chromium irons such as Ni-hard of different grades (high

chromium iron) are being employed as wear-resistant materials in thermal power
generation applications in view of their excellent inherent material characteristics.
They are being deployed in bull ring segments, orifice, rolls, MPO, liners, etc., in ther-
mal power plants and other engineering industries. But, these materials being brittle
fail to withstand transient load. To overcome this problem, manganese is added to
chromium iron to improve its impact characteristics apart from being wear resistant.
In this context, the erosion and abrasion properties affected by metallurgical features
are considered important from the point of enhancing its impact energy, without much
sacrificing the wear resistance. Considering these aspects, high chromium manganese
irons have been prepared in grey cast iron metal mould of section size of 24 mm,
followed by heat treatment. The erosion, abrasion and impact energy have been stud-
ied using jet erosion, rubber wheel abrasion and drop weight impact test setup for
varying manganese content in the range of 5, 10 and 15%. It is very much evident
from the metallurgical investigations involving light microscopy that the least Mn
G. J. Naveen · P. Sampathkumaran · A. Pandey · A. Sathyanarayanaswamy

Department of Mechanical Engineering, Sambhram Institute of Technology,
Bengaluru 560097, India
P. Sampathkumaran
A. Pandey
A. Sathyanarayanaswamy
S. Seetharamu
Department of Mechanical Engineering, Nitte Meenakshi Institute of Technology,
T. Pramod (B)
The Energy and Resources Institute, Bengaluru 560071, India

https://doi.org/10.1007/978-981-15-3254-2_25
268 G. J. Naveen et al.
content sample (5% manganese content) shows the highest wear resistance, whereas
on the other hand, the highest manganese addition of 15% exhibits highest impact
energy. The hardness and the metallurgical features support these findings as one
could visualize the phases and carbide morphological features getting transformed
during the heat treatment process, thus favouring the abrasion, erosion and impact
properties. These results have been compared and analysed with the results obtained
from high chromium iron samples.
Keywords High chromium manganese iron · Erosion · Abrasion · Impact energy ·

Carbide size · Retained austenite
25.1 Introduction
The family of white irons including cast irons and nickel chromium irons have
been investigated and reported [1, 2] extensively in the literature especially the high
chromium irons, in view of its possessing good wear resistance property due to the
presence of hard carbides in a hard and tempered martensitic matrix [3, 4]. But,
there is one drawback with regard to its usage in power, mining, cement and other
industries, and it is the brittle nature of it for not taking sudden load. Many researchers
[5, 6] have made attempts on the development of wear-resistant high chromium cast
iron material to withstand load under impact situations, through the addition of
gamma forming elements such as manganese (Mn), copper (Cu) and nickel (Ni).
Among them, Mn being cost-effective and productive has been tried up to about
5% by weight, beyond which the literature information is scanty. The influence of
microstructure on the wear properties has been studied and reported in [7, 8]. Basak
et al. [9] in their work have demonstrated the role of addition of Mn (4% by weight) in
high chromium iron for enhanced impact energy but at the same time not sacrificing
much on the wear resistance property. Thus, a trade-off between wear and impact
characteristics is very much required at this stage. Keeping the literature points in
view, the alloying element manganese is added in the range of 5–15% by weight to
high chromium iron to improve the load-bearing capacity alongside wear property,
as manganese is known to be a strong austenite-stabilizing element. In this work, the
manganese concentration has been varied at 5, 10 and 15 weight percent to chromium
(16–18 weight percent), and the castings are produced in a metal mould of section
size 24 mm in an induction melting furnace, followed by heat treatment. The erosion
and abrasion characteristics have been evaluated in a jet erosion and rubber wheel
abrasion test rigs, respectively. The impact energy and hardness measurements are
carried out using drop weight impact tester and Rockwell hardness tester. The results
have been compared with high chromium iron without manganese addition in it. The
light microscopy has been taken to examine and interpret the microstructures.
25 Influence of Metallurgical Parameters on Wear … 269
25.2 Materials and Methods
The materials chose are high chromium (16–18%)-alloyed iron and high chromium
(16–18%) manganese-alloyed iron with manganese levels of 5, 10 and 15 weight
percent. The castings are made by induction melting route by pouring liquid metal
into the grey cast iron metal mould of section size 24 mm having the dimensions of
150 × 100 × 25 mm3 . The test coupons are cut to a size of 75 × 25 × 6 mm3 using
abrasive cutting machine from the casting as per the schematic diagram shown in
Fig. 25.1 for erosion and abrasion tests and sample size of 55 × 10 × 10 mm3 for
impact energy test.
The test samples are given heat treatment in a muffle furnace with an austeniti-
zation soak for 6 h at 950 °C followed by stress relieving treatment for 30 min at
200 °C. The erosion and abrasion tests were carried out in a jet erosion test rig and
rubber wheel abrasion, respectively, as per the guidelines given in ASTM standard
[10, 11]. The impact energy test is done in accordance with ASTM standard [12]
guidelines using drop weight impact tester using an un-notched specimen. The hard-
ness has been measured using Rockwell C hardness tester (conical indenter; with
vertex angle 120°; hemispherical tip of radius 200 µm), at a test load of 150 kg. The
metallographic procedure has been adopted to arrive at the microstructure and car-
bide volume using optical microscopy coupled with an image analyser. The retained
austenite measurement has been carried out using X-ray stress analyser equipment
with chromium as the target material. The details of characterization procedures are
reported in [13, 14]. The chemical composition has been determined using optical
emission spectroscopy.
Fig. 25.1 Schematic diagram of casting made, showing the location of test samples for various
tests
The sample designation along with its chemical composition is given in Table 25.1,
and the hardness, retained austenite and carbide volume are given in Table 25.2,
respectively. All the test data provided in this study are pertaining to the heat-treated
conditions only.
25.3.1 Erosion and Abrasion
The erosion test conducted at 45° and 90° impact angles are shown in Fig. 25.2.
It is very well seen from the erosion data that the maximum loss takes place at
45° impact angle and the least at 90° impact angle for HiCr5Mn, HiCr10Mn and
HiCr15Mn samples. For HiCr iron sample, the highest erosion is noticed at 90° and
the lowest at 45° impact angle. As per the theory [15], the highest erosion loss is
reported at 45° for ductile materials and least erosion loss at 90° for brittle materials.
In the present case also, the erosion loss of HiCr closely resembles with the data
pertaining to the brittle material, and hence, it is exhibiting highest hardness and in
turn low erosion resistance. This is on the expected lines as chromium carbide in HiCr
which is having highest hardness as well as carbide volume (Table 25.2) and supports
the erosion test data. In the case of HiCrMn iron samples, HiCr15Mn is showing the
highest erosion loss compared to HiCr5Mn. Accordingly, the erosion data varies
Table 25.1 Chemical composition of HiCrMn iron samples

Sample identification Composition (wt%)
C Mn Si Cr Ni Mo S P
16–18% Cr (HiCr) 2.87 0.60 0.78 16.50 0.98 1.96 0.042 0.040
5% Mn 2.39 4.60 2.15 18.65 0.92 1.82 0.044 0.090
17–19% Cr (HiCr5Mn)
10% Mn 2.35 9.70 1.85 17.60 0.82 1.80 0.040 0.085
15% Mn 2.43 14.80 1.98 18.15 0.87 1.75 0.042 0.080
Table 25.2 Hardness,

Sample Rockwell C Retained Carbide
retained austenite and carbide
designation hardness austenite (%) volume (%)
volume of HiCrMn iron
samples HiCr 64.7 10 29.0
HiCr5Mn 61.0 50 27.5
HiCr10Mn 53.0 60 25.5
HiCr15Mn 50.9 64 24.0
Fig. 25.2 Erosion loss at 45° and 90° impact angles
Fig. 25.3 Abrasion loss
with respect to hardness and carbide volume. Thus, it is seen that the addition of
manganese content from 5 to 15% lowers the hardness and carbide volume and in
turn increases the erosion loss.
The trend in abrasion loss (Fig. 25.3) remains the same like in the case of erosion
loss with respect to the manganese addition. Higher the manganese level, higher the
abrasion loss and vice versa. The highest abrasion loss is obtained for HiCr15Mn and
least for HiCr5Mn as the hardness and carbide volume values support these trends
(Table 25.2). The HiCr iron samples exhibit the least, as it is hard and brittle.
25.3.2 Impact Energy
It is very well known that higher is the austenite content, lower is the carbide volume
and better is its toughness characteristics. In the present case also, higher manganese
Fig. 25.4 Impact energy
content sample (HiCr15Mn) has yielded higher impact energy and vice versa as can
be seen from Fig. 25.4. HiCr iron sample being brittle is showing very low impact
energy, and hence, it cannot take any sudden shock. The retained austenite values
also give credence to the data, i.e., retained austenite content of HiCr iron is about
10%, whereas HiCr15Mn sample is possessing retained austenite content in the range
50–64%. The results obtained are in line with the published reports [16–18], wherein
with increase in manganese content, the wear resistance decreases. But, the impact
energy increases due to the higher concentration of manganese present in it which
acts as austenite-stabilizing agent. Hence, the reported data and the present results
pertaining to the impact energy are in good agreement with each other. Thus, the
main objective of obtaining higher impact energy in HiCr iron samples has been
achieved by the addition of manganese to chromium iron at three levels. Further,
it is envisaged that increase in manganese concentration beyond 15% would yield
enhanced impact energy levels, which is not attempted in this work.
The impact energy value obtained in the present work is about 50% better than
the value reported [9], whereas the wear property decreases by 20%. Thus, the use
of HiCr15Mn seems to be beneficial to withstand the sudden load and at the same
time not forgoing wear resistance.
25.3.3 Light Microscopic Features
The light micrographs of HiCr, HiCr5Mn and HiCr10Mn at a magnification of 500×

and 1000× are shown in Figs. 25.5 and 25.6, respectively. The microstructure of
HiCr iron reveals primarily eutectic carbides of different sizes in a matrix of predom-
inantly martensite. Further, the carbides are distributed non-uniformly in the matrix.
The retained austenite content obtained in this case is about 10%. The HiCr5Mn
(Fig. 25.5b) shows fine carbides and is more or less uniformly distributed in the
matrix of austenite. The occurrence of hexagonal carbides along with fine carbide
Fig. 25.5 Light microstructure of a HiCr, b HiCr5Mn and c HiCr10Mn samples at a magnification
of 500×
Fig. 25.6 Light microstructure of a HiCr, b HiCr5Mn and c HiCr10Mn samples at a magnification
of 1000×
network is observed in the matrix. The addition of 10% manganese (HiCr10Mn)

reveals long carbides with hyper-eutectic carbides in the matrix of austenite. Thus, all
these micrographs help to discourse the erosion, abrasion and impact energy results
in a meaningful way. It is quite logical about the fact that by considering the micro-
graphs of HiCr iron, HiCr5Mn and HiCr10Mn, the carbide morphological features,
i.e., size, its distribution and matrices decide its properties. In fact, microstructure
dictates the characteristics of HiCr family. The literature reports [19–21] give more
information on high chromium manganese irons regarding the impact energy aspects
influenced by varied manganese levels, mould type, size as well as the heat treatment
effect.
In summary, the results pertaining to the wear and impact properties of HiCr5Mn
have yielded best wear resistance compared to HiCr10Mn and HiCr15Mn, whereas
HiCr15Mn exhibited highest impact energy level compared to HiCr5Mn and
HiCr10Mn. HiCr although has shown very poor impact energy, but its wear resis-
tance is much superior compared to high chromium manganese iron samples. HiCr
sample has exhibited brittle behaviour, i.e., the erosion loss is higher at 90° impact
angle and lower at 45° impact angle which are in line with the reported predictions in
the literature [15]. Thus, in the HiCrMn samples, wear resistance is in between brittle
and ductile behaviour in view of the fact that for ductile samples, the wear is the low-
est at normal impact angle (90°) and highest at shallow impact angles (30° and 45°)
as reported in [15]. Further, supporting experimental values, viz., hardness, retained

austenite content, carbide volume and microstructures compiled, lend credence to
the above findings.
Thus, the work may summarize that the dual objectives of obtaining best wear
resistance and highest impact energy in high chromium manganese iron samples
have been achieved based on different manganese concentrations in high chromium
iron.
25.4 Conclusions
Based on the work carried out on HiCrMn iron samples, the following conclusions
are drawn:
• HiCr sample shows the best erosion and abrasion resistance, but poor impact
resistance in view of its material and metallurgical characteristics.
• HiCr5Mn iron samples exhibits the highest wear resistance compared to
HiCr10Mn and Hicr15Mn samples.
• Impact energy of HiCr15Mn shows the highest among the high chromium
manganese iron and high chromium iron samples investigated.
• A trade-off between wear and impact properties is desirable by choosing 10% Mn
addition to HiCr in industries requiring good wear resistance property coupled
with higher impact energy level.
All these test results have been substantiated with the supporting evidences such
as hardness, retained austenite and carbide morphological features. It is inferred from
this study that the addition of 15% of Mn to chromium iron is preferred for application
involving impact load, whereas HiCr is best suited for erosion and abrasion resistance
properties.
Acknowledgements The authors wish to acknowledge with thanks the managements of Sambhram
Institute of Technology, Nitte Meenakshi Institute of Technology and The Energy and Resources
Institute to present and publish this paper.
References
1. ELsawy, E.E.T., EL-Hebeary, M.R., EL-Mahallawi, I.S.E.: Effect of manganese, silicon and
chromium additions on microstructure and wear characteristics of grey cast iron for sugar
industries applications. Wear 390–391, 113–124 (2017)
2. Jumaev, A.A.: Comparative study of the structure of casting from white wear resistant cast
iron. Int. J. Adv. Res. Sci. Eng. Technol. 5(12), 7575–7577 (2018)
3. Gundlach, R.B., Parks, J.L.: Influence of abrasive hardness on the wear resistance of high
chromium irons. Wear 46(1), 97–108 (1978)
4. Qui, B., Xing, S.M., Dong, Q., Liu, H.: Effect of impact energy on wear behaviour of high
chromium white iron produced by liquid die forging. In: International Conference on Metal
Material Processes and Manufacturing, MATEC Web of Conferences, vol. 207, 03011, pp. 1–5
(2018)
5. Tong, J.M., Zohu, Y.Z., Shen, T.Y., Deng, H.J.: The influence of retained austenite in high
chromium cast iron on impact abrasive wear. Wear 135(2), 217–226 (1990)
6. Patwardhan, A.K., Mehta, M.L., Sharma, C.P.: Manganese as an additive to chromium white
iron for producing wear resistant microstructure. AFS Int. Cast Metal J. 6, 3–9 (1981)
7. Laird, G.: Some comments on white cast iron microstructure and wear properties. AFS Trans.
128, 497–504 (1993)
8. Kadhim, M.J., Abood, A.N., Yaseen, R.S.: The role of manganese on microstucture of high
chromium white cast iron. Modern Appl. Sci. 5(1), 179–1185 (2011) (Canadian Centre of
Science and Education)
9. Basak, A., Penning, J., Dilewijns, J.: Effect of manganese on wear and impact strength of 12%
chromium cast iron. AFS Int. Cast Metal J., vol. 6 (1981)
10. ASTM G76-18: Standard test method for conducting erosion tests by solid particle impingement
using gas jets. ASTM International, West Conshohocken, PA (2018)
11. ASTM G65-04: Standard test method for measuring abrasion using the dry sand/rubber wheel
apparatus. ASTM International, West Conshohocken, PA (2004)
12. ASTM E23-12: Standard test methods for notched bar impact testing of metallic materials.
ASTM International, West Conshohocken, PA (2012)
13. Sampathkumaran, P., Ranganathaiah, C., Seetharamu, S., Kishore, : Effect of increased man-
ganese addition and mould type on the slurry erosion characteristics of Cr-Mn iron systems.
Bull. Mater. Sci. 31(7), 1001–1006 (2008)
14. Sampathkumaran, P., Ranganathaiah, C., Seetharamu, S., Kishore, : Effect of manganese con-
tent and mould size on abrasion and slurry erosion behavior of chromium-manganese iron
systems investigated by positron lifetime spectroscopy. Wear 267(9 & 10), 1558–1565 (2009)
15. Raask, E: Erosion Wear in Coal Utilization. Book, Taylor & Francis Inc. Publication, pp. 1–621
(1988)
16. Guo, Z.H., Xiao, F., Lu, S., Li, H., Liao, B.: Effects of heat treatment on the microstructure and
wear resistance of a high-chromium cast iron for rolls. Adv. Mater. Sci. Eng., pp. 1–7 (2016)
17. Hadji, A., Bouhamla, K., Mouche, H.: Improving wear properties of high chromium cast iron
by manganese alloying. Int. J. Metal Cast. 10(1), 43–55 (2015) (Springer Publications)
18. Kvon, S.V.S., Kulikov, V.Y., Filippova, T.S., Omarova, A.E.: Using high chromium iron as
material for production of the equipping components of mine shafts. Metalurgija 55(2), 206–
208 (2016)
19. Krishnamoorthy, P.R., Seetharamu, S., Sampathkumaran, P.: Erosion wear in thermal power
plants. 55th R & D Session, CBI & P (1989)
20. Agunsoye, J.O., Ochulor, E.F., Talabi, S.I., Olatunji, S.: Effect of manganese additions and wear
parameter on the tribological behaviour of NFGrey (8) cast iron. Tribol. Ind. 34(4), 239–246
(2012)
21. Sampathkumaran, P., Ranganathaiah, C., Seetharamu, S., Kishore.: Processes induced struc-
tural transformations and the impact properties of Cr-Mn iron systems. Mater. Manuf. Process.
23, 320–325 (2008) (Tayler & Francis Publications)
Chapter 26
Simulation of High-Temperature
Solid Particle Erosion of
HVOF-Sprayed NiCrBSiFe
and WC-Co/NiCrBSiFe-Coated Wall
Using CFD
K. Raghavendra Naik, R. K. Kumar, V. Saravanan, S. Seetharamu

and P. Sampathkumaran
Abstract The erosion characteristics of HVOF-sprayed NiCrBSiFe and WC-

Co/NiCrBSiFe coatings made on SS347 grade steel have been investigated for high-
temperature behaviour, especially in the boiler tubes used in thermal power genera-
tion. The computational fluid dynamic model has been adopted to study the influence
of the impingement angle on the erosion properties of the coatings. The self-fluxing
NiCrBSiFe alloys have shown the best erosion resistance of about two times at room
temperature and about four times at 700 °C when compared with a composite pow-
ered coating of WC-Co/NiCrBSiFe. At 700 °C, The WC-Co/NiCrBSiFe undergoes
thermal cracking between nickel rich and WC metallic phases due to low oxidation
and thermal expansion mismatch between them leading to degradation of the coat-
ing; however, at this temperature, the self-fluxing alloy of NiCrBSiFe performed
better when compared to WC-Co/NiCrBSiFe coating. The simulation is done with
turbulence model, namely shear stress transport (SST) model with a base on user-
defined function predicted from the experimental results. The computational fluid
dynamic simulated data obtained from SST model and the experimental results were
compared, and they are in good agreement with each other.
Keywords Erosion rate · Solid particle impact · Coatings · Computational fluid

dynamics (CFD) · SST turbulence model
K. Raghavendra Naik (B) · R. K. Kumar · V. Saravanan

Material Technology Division, Central Power Research Institute, Bengaluru 560080, India
S. Seetharamu
Department of Mechanical Engineering, Nitte Meenakshi Institute of Technology,
P. Sampathkumaran
Department of Mechanical Engineering, Sambhram Institute of Technology,

https://doi.org/10.1007/978-981-15-3254-2_26
278 K. Raghavendra Naik et al.
26.1 Introduction
Solid particle erosion is a serious problem in power plants, which reduces the service
life of the mechanical components operating at high temperature due to impingement
of coal and ash particles [1–3]. For instance, boiler tubes undergo severe degradation
due to impaction of ash particles, containing quartz leading to the forced shutdown
of the plant [4, 5]. The coating has been proven to be one of the advanced methods
employed to set the best erosion resistance due to solid particle impact. There are
several methods of deposition process known, and it is dependent on the type of
application. For example, the coating processes such as weld overlay, thermal spray,
diffusion, laser cladding, etc., adopted in thermal power plants face certain drawbacks
such as very thin coatings are possible in the diffusion process, embrittlement taking
place at the interface due to repeated welding and leading to crack formation in weld
overly and very difficult to repair or refurbish at site. In view of the problem faced in
this process, thermal spray coating gains momentum [6, 7] as they overcome them.
The factors like hardness, velocity, impingement angle, size and shape of the
erodent need to be considered to study the erosion behaviour in detail both at room
and elevated temperature among them; velocity and impact angle play a major role
in selecting the coatings [8]. As per the theoretical predictions, ductile materials
erode more rapidly at shallow impact angle (10°–30°); however, brittle material
erodes more at normal impact angle (70°–90°) [9, 10]. The erosion rate at elevated
temperatures is much higher at shallow as well as at normal impact angles compared
to the room temperature erosion [11].
The high-velocity oxy-fuel (HVOF) process comes under the thermal spray, and
it is being deployed in industries effectively from the point of erosion protection. The
advantages of this spraying process are that it yields lower porosity levels, higher
density, lower oxide content and superior bond strength compared to the conventional
coatings [12].
Self-fluxing thermal spray coating like Ni-base will provide high erosive wear
resistance at low and moderate temperature, and hence, it has received widespread
applications [13]. The distributed hard WC particles in the Ni matrix will increase
the hardness of the coating, thereby improving the erosion resistance at a higher
temperature. The published information on 35% WC with NiCrBSiFe has shown
superior wear resistance up to 500 °C in view of higher hardness and minimal porosity
level achieved [14–16]. In the present study, an attempt has been made on the wear
characteristics evaluation beyond 500 °C.
In recent trends, computational fluid dynamic (CFD) approach has become a
useful tool to assess the erosion rate both at room and high temperature, and many
works have been reported on this to account for the elevated temperature erosion
behaviour of bulk material and coatings. The experimental data, therefore, serves
as a validation tool to predict erosion behavior computationally. Thus, CFD is one
of the advanced methods employed to predict the erosion phenomena for different
flow conditions both at room temperature and high-temperature regimes. The CFD
approach in conjunction with experimental data has been effectively used [17–21]
26 Simulation of High-Temperature Solid Particle Erosion … 279
to study the wall thinning of pipes in the pulverised fuel system, slurry and silt
erosion, ash erosion in pressure parts, etc., evolving amicable solutions to the critical
problems faced in thermal power plants.
The published information on the topic (up to 800 °C) for plasma-sprayed coating
with lower velocity of about 4 m/s indicates lower erosion of NiCrBSi and highest
erosion of WC-CO/NiCrBSi. However, in the present work, the attempt has been
made with HVOF self-fluxing alloy of NiCrBSiFe and the distributed hard 35% WC
particles in the nickel matrix with high velocity up to 36 m/s, which simulating boiler
environment and compared the erosion rate intensity through CFD modelling.
26.2 Experimentation
26.2.1 Materials and Methods
The SS-347 steel is the substrate material used as this finds extensive applications in
a pulverised coal-fired thermal power plant. The coatings were made with a velocity
of 680 m/s on the SS-347 base material of size 150 × 150 × 7 mm3 . The test
samples of size 15 × 15 × 7 mm3 were prepared using electro-discharge machining
cutting process for testing and microstructural examination. The regular metallurgical
procedure has been adopted for polishing and etching. The input conditions employed
in the coating process and powder coating details are shown in Tables 26.1 and 26.2.
26.2.2 Erosion Test at Elevated Temperature
An erosion test set-up was designed and built to achieve the real service condition
experienced in boiler involving high temperature as shown in Fig. 26.1. The mor-
phology of erodent, namely silica sand, is shown in Fig. 26.2. The erosion test was
Table 26.1 HVOF spray parameter

Powder: Praxair 1350 VM size: −45 to +15 µm, JP5000HVOF
1 Type of fuel: kerosene @170 PSIG, SCFH 5.8
2 The flow rate of oxygen @210 PSIG, SCFH 1850
3 Distance between nozzle to target, metre 0.381
4 Nitrogen as carrier gas @ 50 PSIG, SCFH 23
5 The velocity of particle pertaining to spray, m/s 680a
6 Per pass thickness of deposit formed 15
7 The feed rate of powder, kg/min 0.09
a Provided by the supplier of the coating
Table 26.2 Powder coating details

Code of sample Chemical composition (weight percentage) of Range of powder size (µm)
coating
D1 NiCrBSiFe −45 to +15
(14.7 Cr–3.21 B–4.35 Si–3.75 Fe–Bal–Ni)
N1 35 WC-Co/65 NiCrBSiFe −45 to +15
(30.6 W–4.2 Co–10.4 Cr–47.6 Ni–1.8 B–2.6
Si–2.8 Fe)
1. Air Compressor 11. Sample holder

2. Air Filter 12. Thermocouple
3. Pressure Gauge
4. Moisture controller
5. Digital Mass flow controller
6. Fluid Heater
7. Nozzle system
8. Particle Feeder
9. Specimen chamber
10. Test sample
Fig. 26.1 High-temperature erosion test rig diagram
Fig. 26.2 Structure of erodent (silica sand)

conducted with the target sample fixed in a holder, which can be moved to set the
desired impact angle of 15°, 30°, 45°, 75° and 90° which has been on the varying
intensity levels of damage from the leading edge to the trailing edge. This is an
accelerated test with the oil- and moisture-free dry air obtained from a compressor.
Dry air is mixed with silica sand and then made to impinge on the sample to study
the erosion behaviour both at room temperature and high temperature (700 °C). The
particle velocity has been measured using double-disc device [22]. Table 26.3 gives
the input parameters employed in this work.
26.2.3 Microstructure Observations
Figures 26.3 and 26.4 show the microstructure of the typical cross section of the
Table 26.3 Erosion test conditions

1 Abrasive type Silica sand
2 Particle size 70 AFS
3 Particle velocity 36 m/s
4 Feed rate 142 g/min
5 Impingement angle 90° and 30°
6 Test temperature Ambient temperature and high temperature (700 °C)
7 Test duration 7 min
8 Test specimen size 25 × 25 × 7 mm
9 Diameter of nozzle 5 mm
Fig. 26.3 Cross section of as-sprayed NiCrBSiFe

Fig. 26.4 Cross section of

as-sprayed 35
WC-Co/NiCrBSiFe
D1- and N1-sprayed HVOF sample. The purpose of adding WC to the matrix is
to enhance the hardness level. The microhardness and the porosity were measured
for both coatings. D1 coating shows the hardness and porosity level of about 750
HV0.3 and 2.1%, for N1 coating, hardness of about 1050 HV0.3 and 3.0% porosity,
respectively.
The surface morphology of the D1 and N1 coatings is as shown in Figs. 26.5 and
26.6. From the surface morphology of the coating, it is observed that there is a higher
porosity level in N1 coating. The dispersion of WC in the nickel-rich matrix in N1
coating increases the porosity level when compared to D1 coating, which may result
in an increase in stress level around the pores. As the particle interact the surface,
the cracks nucleate and propagate easily and cause the material to come off easily,
thus resulting in higher erosion loss of the coating and also at 700 °C; the hardness
Fig. 26.5 SEM morphology

of surface as-sprayed
NiCrBSiFe
Fig. 26.6 SEM morphology of surface as-sprayed 35 WC-Co/NiCrBSiFe
decreases and hence results in higher erosion loss; similar observations have been
made by other researchers reported [23, 24].
26.3 CFD Erosion Simulation
The CFD approach has been adopted to identify the critical erosion-prone regions
obtaining low, mild and high erosion regions based on the varied flow conditions,
flow trajectories, varied impact angles keeping the solid burden constant (gram loss
per kg of abrasive used).
The individual model has been developed for five different impact angles of 15°,
30°, 45°, 75° and 90° to get particle trajectories and averaged particle impact velocity,
but in this work, only the impact angles of 30° and 90° are considered for this study;
following this, the simulation work has been done, and the data have been generated.
26.3.1 Development of Erosion Model for Prediction

of Erosion Rate
In the erosion model, material removal takes place mainly by the impact of erodent
particles on the target material due to the high kinetic energy possessed by the particle
transforming it to potential energy. The erosion phenomena being complex in nature
follow a particular sequence; the erosion loss governs the following equation.
E = K · Vpn · (α) M · ṁ (26.1)

Table 26.4 Multiple nonlinear regression analysis data

Temperature (°C) Materials K n M
RT/Ambient D1 3.01e−11 1.69 0.334
RT/Ambient N1 8.76e−12 2.52 0.647
700 D1 1.911e−9 0.928 0.283
700 N1 2.586e−9 0.987 0.5547
where
VP particle impact velocity
K constant reference velocity factor
α impact angle in radian
n, M velocity and angle exponent
ṁ mass flow rate of erodent impacting
Vo reference velocity
Vo (1/kˆ1/n)
Erosion rate/ṁ (V p /V o )n (α)M
The impact angle is expressed in radians, and it is the angle between the approach-
ing particle track and the wall. The experimental values with velocity, impact angle
and temperature ranges are taken to MATLAB to solve multiple nonlinear regression
analysis and the values of K, n and M are obtained by curve fit as shown in Table 26.4.
The above erosion Eq. (26.1) is implemented in Fortran code, with the help of
Ansys CFX. Fortran subroutine is created in the form of user routine and finally
source code Pt_erosion. F is called for solutions to the erosion equation for each
of the material D1 and N1. A user-defined function is derived to solve the erosion
equation which consists of velocity, impact angle, impact location and temperature
of the target surface.
The sieve analysis is carried out to determine the particle size and its distribution,
and for CFD analysis, the following assumptions have been made
1. The shape factor assumed to be 0.75
2. Particle impact is inelastic in nature
3. Perpendicular and parallel coefficient is considered as one
Also, three different velocities (20, 26 and 36 m/s) are considered in this, and the
gap maintained between the jet and the test sample is 10 mm.
26.3.2 CFD Turbulence Model
Turbulence models are generally considered to predict the flow patterns. The shear
stress transport model (SST) is nothing but two equations model which is more
advanced than zero equation model [25]. The SST model is robust, and accuracy is
good when compared with the other models like k-1, k-ω and SSG Reynolds stress
model. In the present CFD work, SST model has been used in view of possessing a
higher accuracy level.
26.3.2.1 SST (Shear Stress Transport) Model
The SST model is developed by Menter [26] and designed by Wilcox [27]. This
model is a combination of the conventional k-epsilon and k-omega model.
k
ω=
Vt
In the SST model, turbulence viscosity V t is defined as
a1 k
Vt =
max (a1 ω, F2 )
where a1 is the 0.30 coefficient; the absolute value of the vorticity is

√ 2
F2 = tan h max 2 k/0.09 ωy, 500 v/ωy 2
where y is the distance from the nearest wall, and v is the kinematic viscosity.
26.3.3 Model Description
There is a total mesh of 2,62,317 nodes and 2,80,666 elements. Ansys CFX15 is
used to perform CFD analysis. Ansys-Pre was used to mesh 3D structures, as shown
in Fig. 26.7a. The target surface and near-wall are used to have finer meshes with the
dominant mesh of hexahedron. The hexahedron-based fluid mesh gives good results
Fig. 26.7 Generation of mesh a domain of control volume b specimen to target surface
Fig. 26.8 Geometrical model set-up used for CFD study
Table 26.5 Boundary

Inflow condition The mass flow rate of erodent and air
conditions adopted for CFD
velocity
simulation study
Target surface No-slip conditions
Roughness of wall As-polished condition
Turbulence model SST model standard
Outlet condition Atmospheric pressure conditions with
opening boundary
in analysing fluid flow near the target wall, as shown in Fig. 26.7b. The SST model
involving steady state is used to simulate the flow conditions and codes developed
using subroutine in the calculation part of erosion rate where the particle flow is in
the opposite of the Z direction.
Figure 26.8 a and b shows the model with an impact angle of 90° and 30°. The
boundary conditions adopted during CFD simulation are shown in Table 26.5.
26.4.1 CFD Simulation Results
The velocity of the particle to predict the erosion behaviour is 36 m/s and it is arrived
at based on the experiments. The solid particle vectors contours is shown in Fig 26.9.
Figure 26.10 shows the distribution of erosion rate density of the sample of both
D1 and N1 coatings at 30° and 90° impact angle using the SST turbulence model.
The erosion scars from the CFD prediction replicate similar to the experimental test
scar as shown in Fig. 26.11.
Fig. 26.9 Erosion rate density of contour a 90° b 30° impact angle
Fig. 26.10 Erosion rate density distribution on the target surface of the specimen a D1 b N1 at 30°
impact angle and c D1 d N1 at 90° impact angle at a temperature of 700 °C
26.4.2 Discussion
Erosion Mechanism: The erosion rate of the self-fluxing alloy (single metallic coat-
ing) of D1 and N1 (two-phase materials) coatings with the impact angle of 30° and
90° for room and elevated temperature (700 °C) is shown in Fig. 26.12.
Figure 26.12 gives the erosion rate versus the impact angle. At room temperature,
both coatings D1 and N1 show superior erosion resistance; however, at 90° impact,
(a) (b)
(c) (d)
Fig. 26.11 Erosion rates on the target surface of specimen a D1 b N1 at 30° impact angle and c D1
d N1 at 90° impact angle at a temperature of 700 °C
Fig. 26.12 Histogram illustrating the room and high-temperature erosion rate of D1 and N1 coatings
at two different impact angles of 30° and 90°
the erosion is higher for both coatings; this may be due to brittle erosion nature at
normal impact angle.
In the case of D1 coating at a higher temperature of 700 °C, the erosion rate is
decreased at both the shallow and normal impact angle. At shallow impact angle
coating will undergo plastic deformation as the material removal is by cutting and
ploughing mechanisms, whereas at a higher angle of 90°, the erosion damage takes
place due to the repeated impact of the erodent. In this case, the erosion mechanism
may be a combination of grooving, cratering and microcracking. The microstructure
of the D1 coating has dense and layered structure and the microhardness of the
coating measured is 780 HV. When the coating is exposed to higher temperatures,
the hardness of the coating is reduced to half (430 HV) the value measured at room
temperature. Even though the coating can withstand erosion at high temperatures, it
is likely that there is an improvement in toughness due to the drop in hardness. It is
well known that when the hardness decreases, the toughness increases [24].
In the case of N1 coating at a higher temperature of 700 °C, the rate of erosion is
less at 30° impact and more at normal impact angle which is a characteristic behaviour
of brittle materials. When the WC content in the coatings varies from 0 to 18%, the
coating hardness increases gradually; however, with the increase of WC content, the
porosity level also increases which leads to larger pores. When the WC content is
more (35%), the brittleness as well as the pore size increases predominantly. These
bigger sized pores may provide a path for the fracture and lead to the propagation of
the crack. If the WC content is less than 25%, the pore sizes are smaller resulting in
crack not getting formed as it absorbs a larger amount of impact energy. Further, a
higher amount of WC (35%) with large pores undergoes the higher amount of plastic
deformation with the formation of network cracks and gives rise to ease pathway by
the crack propagation to occur. In the process, surface layer comes off following the
delamination theory. The hardness of the coating by the addition of 35% WC was
1056 HV by exposing to a higher temperature of 700 °C; the hardness of the coating
was reduced to 520 HV (almost half) which causes the highest erosion rate; lower
hardness results in poor erosion resistance.
The erosion rate at an elevated temperature at normal impact angle was nearly 5–6
times higher for both the coating tested. Compared to stainless steel and other NiCr
alloys, NiCrBSiFe coating has been proven to be very promising. WC agreements the
greater wear resistance at room temperature; however, its poor resistance to oxidation
and thermal expansion mismatch restricts its application to a temperature above 450–
550 °C; similar kind of observation has been reported elsewhere [28–30], and the
erosion loss governs Eq. (26.1) which is as per the literature [31].
26.4.3 Assessment of Simulation and Experimental Results
The erosion rate of D1 and N1 coatings for room and high temperature is shown in
Table 26.6. The average erosion rate on the target surfaces of the specimen through
experiments and computer simulation using the SST model is also shown. The SST
model belongs to the family of eddy dissipation model and has been chosen because
of its high accuracy and accurate prediction of flow separation from a smooth surface
at the time of particle impact; similar kind of prediction is reported [25].
In this context, it is worth mentioning here that the work carried out by Kim et al.
[21] has adopted the SST model for solid particle erosion prediction and compared it
with the k-epsilon model and reported that the SST model is much better compared
Table 26.6 Comparison of experimental and simulation results on erosion

Sample Impact angle (°) Experimental Numerical data % error
data (kg/m2 s) (kg/m2 s)
D1-25 30 3.50e−5 3.89e−5 11.13
N1-25 30 2.78e−5 3.16e−5 13.67
D1-700 30 1.12e−4 1.25e−4 11.60
N1-700 30 4.15e−4 4.54e−4 9.56
D1-25 90 6.09e−5 6.83e−5 12.11
N1-25 90 1.11e−4 1.21e−4 9.00
D1-700 90 1.32e−4 1.46e−4 10.98
N1-700 90 5.65e−4 6.28e−4 11.23
to the other models. Hence, in the present investigation, also SST model is used to the
advantage of predicting the erosion rate in WC-nickel target material with minimum
acceptable error. Further, it is seen that the % error (Table 26.6) obtained between
the experiments and SST simulation lies in the range 9–14% which seems to be in
the acceptable range. Thus, it is summarized both the experimental and simulated
results are in close agreement with each other.
26.5 Conclusions
The experimental results pertaining to erosion characteristics at elevated temperate of

self-fluxing NiCrBSiFe alloy and 35% WC-Co/NiCrBSiFe coatings made by HVOF
technique have been investigated and compared with the CFD numerical simulation
model. The CFD data closely match the experimental values obtained. The following
are the important conclusions drawn.
• The HVOF-sprayed NiCrBSiFe-coated samples have shown higher erosion resis-
tance both at room temperature and elevated temperature (700 °C) compared to
35% WC-Co/NiCrBSiFe HVOF-coated sample irrespective of the impact angle.
• It is desirable to adopt NiCrBSiFe coating for high-temperature erosion application
as it does not seem to undergo oxidation.
• It is preferable to use WC-Co/NiCrBSiFe coating at a shallow angle (20°–30°)
for erosion resistance application at room temperature, but at 90° impact angle,
NiCrBSiFe is preferable.
• The percentage error obtained between the experimental results and CFD simulated
data lies in the range 9% to 14%, which seems to be in the acceptable limits.
• The CFD application with the SST model seems to predict the erosion behaviour
better using Ansys CFX module and the same study may be extended to higher
temperature (greater than 700 °C) regimes.
Acknowledgements The authors wish to thank profoundly the management of CPRI for according
permission to publish this paper. This paper is a part of a bigger project which was under the research
contingency (RC) funding.
References
1. Ramesh, C.S., Seshadri, S.K., Iyer, K.J.: A survey of aspects of wear of metals. Indian J.
Technol. 29(4), 179–185 (1991)
2. Archer, N.J., Yee, K.K.: Chemical vapour deposited tungsten carbide wear-resistant coatings
formed at low temperatures. Wear 48(2), 237–250 (1978)
3. Pawlowski, L.: The Science and Engineering of Thermal Spray Coatings. Wiley & Sons, 30
Apr 2008
4. Uusitalo, M.A., Vuoristo, P.M., Mäntylä, T.A.: High-temperature corrosion of coatings and
boiler steels in the oxidizing chlorine-containing atmosphere. Mater. Sci. Eng. 346(1–2), 168–
177 (2003)
5. Stultz, S.C., Kitto, J.B. (eds.): Steam: Its Generation and Use. Babcock and Wilcox, Barberton,
OH (1992)
6. Uusitalo, M.A., Vuoristo, P.M., Mäntylä, T.A.: High-temperature corrosion of coatings and
boiler steels below chlorine-containing salt deposits. Corrosion Sci. 46(6), 1311–1331 (2004)
7. Miguel, J.M., Guilemany, J.M., Vizcaino, S.: Tribological study of NiCrBSi coating obtained
by different processes. Tribol. Int. 36(3), 181–187 (2003)
8. Hutchings, I.M.: Wear by particulates. Chem. Eng. Sci. 42(4), 869–878 (1987)
9. Sundararajan, G.: A comprehensive model for the solid particle erosion of ductile materials.
Wear 149(1–2), 111–127 (1991)
10. Finnie, I.: In: Nateson, K. (ed.) Corrosion-Erosion Behavior of Materials, p. 437. AIME, New
York (1980)
11. Kulu, P., Hussainova, I., Veinthal, R.: Solid particle erosion of thermal sprayed coatings. Wear
258(1–4), 488–496 (2005)
12. Mohanty, M., Smith, R.W., De Bonte, M., Celis, J.P., Lugscheider, E.: Sliding wear behaviour of
thermally sprayed 75/25 Cr3C2/NiCr wear resistant coatings. Wear 198(1–2), 251–266 (1996)
13. Kim, H.J., Hwang, S.Y., Lee, C.H., Juvanon, P.: Assessment of wear performance of flame
sprayed and fused Ni-based coatings. Surf. Coat. Technol. 172(2–3), 262–269 (2003)
14. Bell, G.R.: Eighth International Thermal Spraying Conference, Mimi, Florida, USA, pp. 396–
406 (1976)
15. Ding, Z., Knight, R., Smith, R.W.: Thermal Spray: A United Forum for Scientific and
Technological Advances, Materials Park, Ohio, USA, pp. 91–95 (1997)
16. Heath, G.R., Kamer, P.A.: Surface Modification Technology IX, Cleveland, Ohio, USA,
pp. 109–113 (1995)
17. Parsi, M., Agrawal, M., Srinivasan, V., Vieira, R.E., Torres, C.F., McLaury, B.S., Shirazi, S.A.:
CFD simulation of sand particle erosion in gas-dominant multiphase flow. J. Nat. Gas Sci. Eng.
27, 706–718 (2015)
18. Duarte, C.A., de Souza, F.J., de Vasconcelos Salvo, R., dos Santos, V.F.: The role of inter-particle
collisions on elbow erosion. Int. J. Multiph. Flow 89, 1–22 (2017)
19. Zahedi, P., Zhang, J., Arabnejad, H., McLaury, B.S., Shirazi, S.A.: CFD simulation of mul-
tiphase flows and erosion predictions under annular flow and low liquid loading conditions.
Wear 376, 1260–1270 (2017)
20. Gandhi, M.B., Vuthaluru, R., Vuthaluru, H., French, D., Shah, K.: CFD based prediction of
erosion rate in large scale wall-fired boiler. Appl. Therm. Eng. 42, 90–100 (2012)
21. Kim, J.H., Joo, H.G., Lee, K.Y.: Simulation of solid particle erosion in WC-Ni coated wall
using CFD. J. Mater. Process. Technol. 224, 240–245 (2015)
22. Ruff, A.W., Ives, L.K.: Measurement of solid particle velocity in erosive wear. Wear 35, 195–
199 (1975)
23. Hidalgo, V.H., Varela, F.B., Menéndez, A.C., Martınez, S.P.: A comparative study of high-
temperature erosion wear of plasma-sprayed NiCrBSiFe and WC–NiCrBSiFe coatings under
simulated coal-fired boiler conditions. Tribol. Int. 34(3), 161–169 (2001)
24. Tu, J.P., Liu, M.S., Mao, Z.Y.: Erosion resistance of Ni-WC self-fluxing alloy coating at high
temperature. Wear 209(1–2), 43–48 (1997)
25. Sklavounos, S., Rigas, F.: Validation of turbulence models in heavy gas dispersion over
obstacles. J. Hazard. Mater. 108(1–2), 9–20 (2004)
26. Menter, F.R.: Two-equation eddy-viscosity turbulence models for engineering applications.
AIAA J. 40(2), 254–266 (2002)
27. Wilcox, D.C.: Reassessment of the scale-determining equation for advanced turbulence models.
AIAA J. 26(11), 1299–1310 (1988)
28. Russo, L., Dorfman, M.: A structural evaluation of HVOF sprayed NiCr-Cr sub 3 C sub 2
coatings. Therm. Spray. Curr. Status Futur. Trends 2, 681–686 (1995)
29. Berger, L.M., Hermel, W., Vuoristo, P., Mantyla, T., Lengauer, W., Ettmayer, P.: Structure,
properties and potentials of WC–Co, Cr3C2–NiCr and TiC–Ni-based hard metal like coatings.
In: Berndt, C.C. (ed.) Thermal Spray: Practical Solutions for Engineering Problems, pp. 89–98
(1996)
30. Berger, Z., Livshitz, M.: The structure and properties of hypervelocity oxy-fuel (HVOF) sprayed
coatings. High-Temp. Mater. Process. 15, 179–186 (1996)
31. Raask, E.: Mineral Impurities in Coal Combustion: Behaviour, Problems, and Remedial
Measures. Taylor & Francis (1985)
Chapter 27
Optimization of Friction Stir Welding
Parameters Using Taguchi Method
for Aerospace Applications
K. Ramesha, P. D. Sudersanan, N. Santhosh, G. Ravichandran

and N. Manjunath
Abstract The current research work investigated the optimization of the input
parameters for the friction stir welding of AA3103 and AA7075 aluminum alloys
for its applications in aerospace components. Friction stir welding is rapidly grow-
ing welding process which is being widely used in aerospace industries due to the
added advantage of strong strengths without any residual stresses and minimal weld
defects, in addition to its flexibility with respect to the position and direction of weld-
ing. Thus, the demand for this type of welding is very high; however, the welding of
aluminum alloys is a key aspect for its use in aircraft components, particularly with
respect to bracket mounting frames, braces and wing components. Henceforth in the
current work, research is focused on optimization of welding of aluminum alloys,
viz. AA 3103 and AA 7075; AA 3103 is a non-heat treatable alloy which is having
good weldability, while AA 7075 is having higher strength. Therefore, the welding
of these aluminum alloys will produce superior mechanical properties. The opti-
mization of input parameters was accomplished in this work based on L9 orthogonal
array designed in accordance with Taguchi method—using which the friction stir
welding experiment was conducted. There were nine experimental runs in total after
formulating the L9 orthogonal array table in Minitab software. The input parameters
which were selected for optimization were—tool rotation speed, feed rate, tool pin
profile. The output parameters which were optimized were hardness, tensile strength
and impact strength. In addition, the microstructure of the fractured surfaces of the
friction stir welded joint was analyzed. It was found from the optimization of the pro-
cess parameters that strong friction stir welded joints for aerospace applications can
K. Ramesha · P. D. Sudersanan
Department of Mechanical Engineering, Dr. T. Thimmaiah Institute of Technology, Kolar Gold
Fields, Karnataka, India
N. Santhosh (B) · G. Ravichandran
Department of Mechanical and Automobile Engineering, CHRIST (Deemed to be University),
Bengaluru, Karnataka, India
e-mail: [email protected]; [email protected]
N. Manjunath
Department of Sciences and Humanities, CHRIST (Deemed to be University), Bengaluru,
Karnataka, India

https://doi.org/10.1007/978-981-15-3254-2_27
294 K. Ramesha et al.
be produced at an optimized set of parameters of tool rotational speed of 1100 rpm,

traverse speed of 15 mm/min with a FSW tool of triangular pin profile of H13 tool
steel material.
Keywords AA3103 · AA7075 · Friction stir welding · H13 tool steel · Taguchi
orthogonal array · Minitab · Aerospace · Components
27.1 Introduction
Friction stir welding is type of welding in which a non-consumable tool is being used
in order to weld two workpieces. The welding of the workpiece occurs heating the
working piece at its line of joint. The heat is generated due to the friction between
the rotating tool and the workpiece, and as the tool is rotating, it intermixes both
the workpiece materials. The mechanical intermixing of the workpiece material will
be accomplished as the tool moves along the line of the joint, and subsequently the
forging of hot and soft material is made by the means of mechanical pressure. The
FSW tool serves two primary functions:
• Heating of the workpiece material.
• Mixing of two materials in a mushy state to produce the joint.
Friction stir welded aluminum alloys are widely used in aeronautical and
aerospace industries for manufacturing airplanes both commercial and defense type,
spacecraft, helicopters, etc. Among the different types of aluminum alloys, AA3103
is a non-heat treatable alloy and having superior strength, whereas the AA7075 is
having higher strength. Thus, the weld obtained from these two alloys will be of
superior quality. But AA7075 aluminum weld is quite expensive. In order to reduce
the expense, the AA7075 aluminum alloy is welded with other grades of aluminum
alloys. The AA3103 aluminum alloy has good weldability and is less expensive.
Therefore, a weld produced from these two alloys will be showing the properties
of both these alloys. For the weldment between AA2024-T365 and AA5083-H111,
heat-treated H13 steel is used as the tool material; the tool is having a concave
shoulder of 20 mm diameter and a pin diameter of 5 mm, and the height of pin is
4.7 mm.
The tool pin profiles used for welding are square, triangular and stepped. The
welding parameters used for the experimentation are 900, 1120 and 1400 rpm for
tool rotational speed and 16, 40 and 80 mm/min for feed rate [1]. The friction stir
welding is conducted between two grades of aluminum alloys 5086 H116 and 5083
H321. Each workpiece plate is having a thickness of 6 mm, the AA 5086 is having
dimensions of 100 * 95 * 6 mm, and the AA 5083 is of 100 * 70 * 6 mm dimensions.
During the welding process, AA 5086 is taken as the advancing side (AS), while AA
5083 is the retracting side (RS), and this is because the advanced side material should
be always a hard material. The tool used for welding is a chromium–molybdenum
27 Optimization of Friction Stir Welding Parameters Using Taguchi … 295
tool which is made of H13 steel. The five different pin profiles used for the weld-
ing experiments are threaded, square, cylindrical, tapered and triangular [2]. The
metallurgical characterization is carried out by SEM and EDS, and the mechanical
properties are developed by tensile and hardness test. From this paper, they come
out with the most efficient result that is of 710 rpm and 28 mm/min, and the tool
design is cylindrical pin [3]. Fatigue behavior of friction stir welded Al–Mg joints
exhibited wide scatter. The Al–Mg joints exhibited two distinct failure modes: kink
crack and interfacial. Intermetallic compounds outside the weld likely led to fret-
ting initiated fatigue. Insufficient material mixing resulted in underperforming joints
[4]. The tool profiles are as follows: The shoulder shape is concave, pin shape is
triangular, shoulder diameter is 12 mm, and pin diameter is 5.4 mm [5].
The tool used for friction stir welding is made up of carbon steel of diameter
20 mm with shoulder diameter 12 mm and with a cylindrical pin of diameter 5 mm.
Eight joints were fabricated as per the design plan from the Taguchi method [6].
Microstructural suggested that there was no rigorous mixing, and there was a small
change in the microhardness across the surface. The microelectron analysis test shows
that there is a bonding at the atomic scale due to interfusion of the alloys in the nugget.
Thus, the interdiffusion of alloying elements and attaining of similar orientation in
the nugget have contributed good mechanical properties [7]. It is concluded that
for AA5083–AA5083 similar material welding, the welding efficiency is 77% from
base metal AA5083, and for the AA6061–AA5083 dissimilar material welding, the
welding efficiency was 34 and 93% compared to AA5083 and AA6061 [8]. The
tool used in this experiment is EN-31 with a shoulder diameter of 20 mm and pin
diameter of 5 mm at the root and 5.75 mm pin length. The weld joints are cut as
per ASTM dimensions for impact test [9, 10]. L8 orthogonal array is used by taking
three process parameters and two levels for optimizing the process parameters for
obtaining higher hardness and tensile strength of the welded region [11, 12].
From the literature review, it is seen that the friction stir welding of aluminum
components has great scope, and studies can be carried out on the process since
friction stir welding depends on the parameters selected, which facilitates the need
to optimize these input parameters so that we could find out how it affects the resulting
weld and the output parameters like strength and quality of weld. To do so, there
are many input parameters, so by going through the findings, it is finalized that
three parameters which are speed, feed rate and tool profile are considered, and each
parameter has to be selected. These values will be then optimized in order to obtain
the perfect combination of the input parameter value so that high quality weld will
be obtained.
27.2 Materials and Methods
The current section gives an overview of the materials used for fabrication of the
plates that are friction stir welded by optimizing the parameters.
27.2.1 Selection of Materials
AA 7075 and AA 3103 are the aluminum alloys which are chosen as the two work-
piece materials having dimensions of 103 * 50 * 6 mm, for friction stir welding in this
experiment. The chemical composition of the workpiece is presented in Tables 27.1
and 27.2.
27.2.2 Selection of Tool
The welding tool is made up of H13 tool steel. The weld tool is manufactured by
casting process and then heat treated. Right after the heat treatment, the tool is kept
in oil bath (for three days). This is done in order to improve the strength of the weld
tool.
27.2.3 Selection of Welding Parameters and Levels
The three main parameters selected for this welding operation are tool rotational
speed, feed rate of tool and tool pin profile. The selection of the welding parameter
values was based on the knowledge obtained from the literature survey. Table 27.3
gives the values selected for the welding parameters and their levels.
27.2.4 Experimental Setup
The entire experiment is carried out in a vertical CNC machine. Both the workpieces
are placed on a rig with AA3103 specimen placed on the advancing side. The rig
for holding the workpiece rigidly is installed on the CNC worktable. The friction
stir welding is carried out based on the combination of input parameter values,
obtained using Minitab software. Computer numerical control (CNC) machining is
a machining process in which the machining tools and traverse path of the cutting tool
are controlled by means of a computer. In this machining process, pre-programmed
computer software will control all the movements of the tool. When a CNC machine
is activated, the desired operations required to be done will be programmed into
the software in the form of G codes and M codes, and these programs shall then
dictate the tool for the corresponding motions to perform the operations. The current
experiment is carried out in a high quality CNC milling machine without using
coolant and with a maximum speed of 1200 rpm. The maximum feed rate used in our
project is 25 mm/min, and the electric drive used is a 25-kw drive motor. The fixture
is used to hold the workpiece tight and in perfect position without any vibration while
Table 27.1 Chemical composition of AA 3103 (weight %)
Elements Mn Fe Cu Mg Si Zn Cr Ti + Zr Al
Percentage composition 1.5 0.7 0.1 0.3 0.5 0.2 0.1 0.1 Balance
27 Optimization of Friction Stir Welding Parameters Using Taguchi …
297
Table 27.2 Chemical composition of AA 7075 (weight %)

Elements Al Zn Mg Cu Cr
Percentage composition 90 5.6 2.5 1.6 0.23
Table 27.3 Welding parameters and their levels

Parameters Unit Symbol Level 1 Level 2 Level 3
Tool rotational speed rpm A 1000 1100 1200
Feed rate mm/min B 15 20 25
Tool pin profile Shape C Square Circle Triangle
Table 27.4 Design of

Experimental Speed of Feed rate Tool pin
experiments matrix with input
trial No. rotation (mm/min) profile
values
(rpm)
1 1000 15 Circular
2 1000 20 Square
3 1000 25 Triangle
4 1100 15 Square
5 1100 20 Triangle
6 1100 25 Circular
7 1200 15 Triangle
8 1200 20 Circular
9 1200 25 Square
welding. The fixture consists of a main plate and above that four supporting sides each
containing a M12 hole, where the tightening screws are inserted to adjust the position
of the workpiece. The experimental trails were carried out based on Taguchi’s L9
orthogonal array. The values of the input parameters were selected based on the
literature survey. A nine run, three factors, three-level orthogonal array was designed
by following Taguchi’s design of experiments, which is given in Table 27.4.
27.2.5 CNC Programming
Since the friction stir welding of the workpieces was carried out in a CNC machine,
a program was required to run the welding experimentation. The CNC program used
for the friction stir welding considered the various aspects of tool rotational speed,
feed rate and depth with preparatory functions (G codes) and miscellaneous functions
(M codes).
A nine trial run experiment was conducted in the vertical CNC milling machine by
giving the input parameters varied at three levels of input parameters. The output
parameters selected for the experiment are tensile strength, hardness and impact
strength, and each of them determined using UTM, Brinell hardness testing machine
and Charpy testing machine, respectively. Further, the microstructure of the fractured
surfaces of the friction stir welded joints is analyzed using a Hitachi make SU 3500
scanning electron microscope. Initially, the specimens of 10 mm * 10 mm * 5 mm
thickness from the fractured surfaces of the friction stir welded joints are cut and pre-
pared suitably for microscopic observations in SEM at various resolutions and mag-
nifications to analyze the fractography and evaluate the fracture mechanics behind
the weld joints, which is presented in this section. Table 27.5 gives the values of
the different properties characterized for the given set of parameters selected for
the experimental trials carried out as per the design of experiments formulated in
accordance with Taguchi’s L9 orthogonal array.
The obtained results are the entered into the L9 orthogonal array with the input
parameter values and analyzed in the Minitab software. As a result of the statistical
approach, the software will optimize and compare both input and output parameters
and provide combinations of input parameters which will produce the best and high
quality weld.
Table 27.5 Resultant values from the experimental trials

Sl. Speed Feed rate Type of Brinell hardness Tensile Impact
No. (rpm) (mm/min) pin profile (HBW * 10/1000) strength energy
(N/mm2 ) (J)
1 1000 15 Circular 90.7 89.9 4
2 1000 20 Square 92.3 96.1 2
3 1000 25 Triangle 96.7 81.2 4
4 1100 15 Square 90.2 97.1 10
5 1100 20 Triangle 98.3 95.4 6
6 1100 25 Circular 96.6 84.4 4
7 1200 15 Triangle 89.6 102.8 12
8 1200 20 Circular 92.3 98.7 4
9 1200 25 Square 92.8 87.7 4
27.3.1 Analysis of Tensile Strength
The tensile strength of the weldment characterized using an Intron make UTM of
1 k N capacity is tabulated and analyzed for the optimization of the values. Fig-
ures 27.1 and 27.2 show main effects plot for signal-to-noise (SN) ratios and for
means. For the main effects plot for the SN ratios, the x-axis represents the input
parameters, whereas the y-axis shows the SN ratios. For the main effects plot for
means, the x-axis represents the input parameters, whereas the y-axis represents
means. The Taguchi analysis has given validation for the SN ratio with the critical
aspect considering the “larger is better” formulations that eventually gives the opti-
mized values for the parameters; it is seen from the graph for SN ratio and main
effects plot for means of tensile test results that the tensile strength of the weld joint
will be maximum for a tool rotational speed of 1200 rpm, feed rate of 20 mm/min
with a tool of square profile.
S/N = −10 ∗ log((1/Y 2 )/n) (27.1)
The empirical formula for “larger is better” is given in Eq. 27.1. The analysis
of variance (ANOVA) was carried out in the Qualitek-4 software to find out the
percentage influence of input parameters on the output parameters. The percentage
influence of input parameters on tensile strength is given in Table 27.6. The feed rate
of the tool is having more influence on tensile strength of the weld.
Fig. 27.1 Main effects plot for signal-to-noise ratios of tensile test results
Fig. 27.2 Main effect plot for means of tensile test results
Table 27.6 Significant percentage of input parameters on tensile strength

Factor Percentage P (%)
Speed 15.591
Feed 69.436
Profile 0
Other error 14.973
Total 100.00
27.3.2 Analysis of Hardness
The hardness of the weld was measured in Brinell hardness testing machine, and a
load of 1000 kg was applied on the weld specimen while testing. Figs. 27.3 and 27.4
show main effects plot for SN ratios and for means. For the main effects plot for the
SN ratios, the x-axis represents the input parameters, whereas the y-axis gives the SN
ratios. For the main effects plot for means, the x-axis represents the input parameters,
whereas the y-axis represents means.
The SN ratios are validated for “larger is better” condition, and it is herewith seen
from the graph that the Brinell hardness of the weld joint is maximum for tool rotation
speed of 1100 rpm, feed rate of 25 mm/min, friction stir welded with triangle tool
insert. The percentage influence of input parameters on Brinell hardness is shown
in Table 27.7. The table shows the feed rate of the tool is having more influence on
Fig. 27.3 Mean effect plot for signal-to-noise ratios of Brinell hardness test results
Fig. 27.4 Mean effect plot for means of Brinell hardness test results
Table 27.7 Significance

Factors Percentage P (%)
percentage of input
parameters on hardness Speed 20.093
Feed 54.225
Profile 15.572
Other error 10.11
Total 100.00
Brinell hardness of the weld.
27.3.3 Analysis of Impact Strength
The impact strength of the weld specimen was determined by a typical Charpy test.
Figs. 27.5 and 27.6 show main effects plot for SN ratios and means, respectively. For
the main effects plot for the SN ratios, the x-axis represents the input parameters,
whereas the y-axis gives the SN ratios. For the main effects plot for means, the x-
axis represents the input parameters, whereas the y-axis represents means. The main
effects plot is analyzed for SN ratio considered to be better for larger values.
The percentage influence of input parameters on impact strength is given in
Table 27.8. The table shows that the feed rate of the tool is having more influence on
Fig. 27.5 Mean effect plot for signal-to-noise ratios of impact test results
Fig. 27.6 Mean effect plot for means of impact test results
Table 27.8 Significance

Factor Percentage P (%)
percentage of input
parameters on impact strength Speed 21.649
Feed 46.391
Profile 15.463
Other error 16.497
Total 100.00
impact strength of the weld.
27.3.4 Fractography of the Weld Joint
The fractography of the weld joint of the aluminum AA 3103 and AA 7075 specimens
is given in Fig. 27.7, and it is clearly evident that the fractured surfaces show a distinct
coarse structure with twinned interfaces and striated surfaces rather than slip bands,
thereby validating the fact that the residual stresses are minimal in friction stir weld
joints; however, striations and luder bands can be observed, thus providing a proof
for plastic deformation that has occurred under tensile loads before fracture. Further,
the SEM image captured under a voltage of 20 k V at a magnification of 500× gives
an account of the striated bands all around the periphery of the atoms inter-dwindled
Fig. 27.7 Mean effect plot for means of impact test results
in the FSW weld structure due to the interatomic forces and coherent bonding that
result in a strong joint.
27.4 Conclusions
After the analysis of the output parameters, it has been found out that if the friction
stir welding is conducted at 1100 rpm tool rotational speed, 15 mm/min feed rate and
using tool of triangular pin profile, an effective weld of AA3103 and AA7075 could be
produced. Further, it has been observed that welds produced at higher tool rotational
speed and lower feed rate give higher output values. Out of all the three types of tool
profile, the weld produced by using triangular pin profile tool gave the best output.
Also, the fractography of the friction stir welded specimens validates the fact that
the friction stir welding of the aluminum alloys reduces the stress accumulation,
distortions and other thermal effects as is evident in other welding processes. Hence,
friction stir welding process can be employed for joints of aerospace components,
in particular bracket mounting frames, braces and wing components, which require
high degree of structural integrity.
References
1. Santhosh, N., Mahamad, M.: Thermomechanical modeling and experimental evaluation of

friction stir welds of aluminium AA6061 alloy. Int. J. Eng. Res. Technol. (IJERT) 2(8), 1494–
1499 (2013)
2. El-Hafez, H.A., El-Megharbel, A.: Friction stir welding of dissimilar aluminum alloys. World
J. Eng. Technol. 6, 408–419 (2018)
3. Kundu, Jitender, Singh, Hari: Friction stir welding of dissimilar AL alloys: effect of process
parameters on mechanical properties. Eng. Solid Mech. 4, 125–132 (2016)
4. Sadesh, P., Kannan, M.V., Rajkumar, V., Avinash, P.: Studies on friction stir welding of AA2024
and AA6061 dissimilar metals. Procedia Eng. 75, 145–149 (2014)
5. Rao, H.M.: Effect of process parameters on mechanical properties of friction stir spot welded
magnesium to aluminum alloys. Mater. Des. 15, 1–34 (2015)
6. Rao, H.M.: Effect of process parameters on mechanical properties of friction stir spot welded
magnesium to aluminum alloy. Mater. Des. 235–245 (2015)
7. Varma, R.R., Ibrahim, A.B.: Mechanical properties of the friction stir welded dissimilar
aluminium alloy joints. Int. J. Mech. Prod. Eng. 2, 1–5 (2014)
8. Kumbhar, N.T., Bhanumrthy, K.: Friction stir welding of Al 5052 with Al 6061 alloys. J. Metall.
2012, 1–7 (2012)
9. Selamat, N.F.M., Baghadi, A.H., Sajuri, Z.: Friction stir welding of similar and dissimilar
aluminum alloys for automotive application. Int. J. Automot. Mech. Eng. 13, 3401–3412 (2016)
10. Shahabuddin, Dwivedi, V.K.: Effect of tool geometry of friction stir welding on mechanical
properties of AA-7075 aluminum alloy. Int. J. Mech. Eng. Technol. 9, 625–633 (2018)
11. Ramakrishna, M.V.A., Mahender, T.: Investigation of friction stir welding parameters of 5083
aluminum alloy by Taguchi method. Int. J. Eng. Technol. Sci. Res. 4, 1078–1082 (2017)
12. Santhosh, N., Ramesha, K.: Mechanical and thermal characterization of friction stir weld joints
of Al–Mg alloy. Int. J. Res. Aeronaut. Mech. Eng. (2017, December 22–23)
Chapter 28
Assessment of the Shear Strength of Fly
Ash-Based Geopolymer Concrete
S. Kumar, S. Rajendra and K. S. Sreekeshava
Abstract Analytical approaches determine the shear strength of geopolymer con-

crete on experimentally evaluated parameters of strength such as compressive
strength, tensile strength and flexural strength. The analytical approach has been
introduced by using the equations of the correlated stress functions to evaluate the
shear strength. The new approach has been introduced over traditional Mohr’s fail-
ure envelope to assess the shear strength. The shear stress evaluated under these
approaches is compared with the solution by the equation suggested by ACI-318-95.
The variation of results from these approaches varied a maximum of 40% with results
by the equation (ACI-318-95). The validation of each approach would be subjected
to further discussion with previous research significance under the same context.
Keywords Geopolymer concrete · Shear strength · Fly ash-based concrete · Shear

stress
28.1 Introduction
There are several research works carried on geopolymers as binder or inorganic

polymer but some intensive research works carried on Geopolymer concrete under
the context of structural applications. Numerous studies by experimentations were
recorded and still continuing on material properties. There are some effective works
which incorporate the classical theories and experimental investigations to elucidate
the material properties under structural applications. Some of the analytical and
experimental approaches are explicated by traditional methods which are being used
for conventional concrete.
S. Kumar (B) · K. S. Sreekeshava

Jyothy Institute of Technology, Bengaluru, Affiliated to Visvesvaraya Technological University,
Belagavi, India
S. Rajendra
Vijaya Vittala Institute of Technology, Bengaluru, Affiliated to Visvesvaraya Technological
University, Belagavi, India

https://doi.org/10.1007/978-981-15-3254-2_28
308 S. Kumar et al.
Several researchers have considered concrete as an isotropic material which has

the same material properties along with all directions. In reality, it is a mixture of
sands and gravels and it is not an isotropic material. Its tensile stress is near 1/10 of the
compressive stress in the perpendicular direction measured on an experimental spec-
imen. It is important to notice that mixtures have the same properties like orthotropic
or anisotropic materials but they are considered as homogeneous materials. So they
do not have the same E1 and E2 and others.
Several experimental investigations have been conducted to study the shear
strength of concrete and predicted the results as a function of the compressive
strength. The shear strength of concrete generally varies with the several associ-
ated factors such as internal frictions due to aggregate interlocking, the strength of
cement paste and aggregates. In this study, the shear strength of concrete is discussed
and predicted by the general strength of concrete as a whole without assessing the
properties of individual materials by which is made up of.
28.1.1 Research Significance
Most of the experimental investigations on the mechanical properties of geopolymer

concrete were compared with the code recommendations. Some of the expressions
on the correlation of strength parameters of the conventional concrete are highlighted
to compare with the present investigation. The correlation of split tensile strength to
compressive strength is expressed as f t = 0.59 f ck0.5 , f t = 0.30 f ck0.5 , f t = 0.23 f ck0.5
by ACI-318-99, CIB FIB and Neville, respectively [1]. The correlation of flexural
strength to compressive strength of conventional concrete is expressed as fr =
0.7 f ck0.5 , fr = 0.62 f ck0.5 , fr = 0.60 f ck0.5 by IS 456-2000, ACI-318 and BS 8110,
respectively [2].
28.2 Basic Materials and Mix Proportion
The geopolymer is prepared by using fly ash, aggregates and alkaline solutions
(NaOH + Na2 SiO3 + H2 O). The fly ash and aggregates used for mix are conform-
ing to IS: 3812 [3] IS: 383 [4], respectively. The geopolymer concrete is prepared
with different mix proportions. The basic material testing and mix proportions are
discussed elsewhere [5]. Same molarity (14 Molar) of alkaline solutions is used for
all the mix. The detail of mix used for this work is shown in Table 28.1.
Table 28.1 Selected mix details of GPC

Mixture Fly Fly ash Coarse Fine NaOH Na2 SiO3 Plasticizer
ash (Kg/m3 ) aggregate aggregate (Kg/m3 ) (Kg/m3 ) (Kg/m3 )
(%) (Kg/m3 ) (Kg/m3 )
FGC-M6 25 521.44 876.02 688.30 89.76 224.4 4.8
28 Assessment of the Shear Strength of Fly Ash-Based Geopolymer … 309
Table 28.2 Test results of the strength parameters of concrete

Mixtures Compressive Split tensile strength Fexural strength ( fr )
strength ( f ck ) MPa ( f t ) MPa MPa
FGC-M1 13.35 1.88 2.52
FGC-M2 15.69 2.13 2.87
FGC-M3 18.56 2.44 3.13
FGC-M4 22.75 3.08 3.68
FGC-M5 26.35 3.52 4.28
FGC-M6 28.56 3.67 4.53
FGC-M7 31.98 4.12 4.98
FGC-M8 33.69 4.33 5.1
FGC-M9 35.58 4.53 5.32
FGC-M10 37.64 4.65 5.77
FGC-M11 40.23 5.14 6.05
FGC-M12 41.54 5.33 6.34
28.2.1 Testing of Concrete
Compressive strength split tensile strength and flexural strength tests of concrete are
conducted as per IS: 516-1959 [6] and presented in Table 28.2.
28.2.2 Correlation of Strength Parameters
Graphically representing correlations of split tensile strength to compressive strength

and flexural strength to compressive strength as shown in Figs. 28.1 and 28.2
respectively.
The correlation of corresponding strength parameters ( f s ) by regression analysis
are presented in the power function in the form f s = f ( f ck ) = k( f ck )n and are given
in Eqs. (28.1) and (28.2).
Fig. 28.1 Correlation of compressive strength to split tensile strength

310 S. Kumar et al.
Fig. 28.2 Correlation of compressive strength to flexural strength
f t = f ( f ck ) = 0.1725( f ck )0.9161 (28.1)
fr = f ( f ck ) = 0.3063( f ck )0.805 (28.2)
28.3 Assessment of Shear Strength
28.3.1 Assessment of Shear Strength by Split Tensile Strength
The shear stress is analysed under classical theories of mechanics by considering

cylindrical specimen subjected to split tensile test as shown in the Fig. 28.3a, b and
c. The shear mechanism is explicated in Eqs. 28.3–28.5. The direct tensile strength
of conventional concrete is expected to be 90% of split tensile strength [7]. Due to
non-availability of the proper experimental evidences on direct tensile strength of
geopolymer concrete, the direct tensile strength of concrete is considered as equal to
split tensile strength to compute principal tensile strength. The relevant multiplication
factor on direct to split tensile strength can be incorporated based on experimental
investigation.
(a) (b) (c)
Fig. 28.3 Shear mechanisms by split tensile strength test

AO = OB = d/2 (where d is the diameter)
ft 4P
= τx y = τc = (28.3)
2 π dl

Principal stress σ1, σ2
2
σx + σ y σx − σ y
σ1 , σ2 = ± + τx2y (28.4)
2 2
ft
σ1 = σx = , σ2 = 0
2
ft
τc = f ( f t ) = (28.5)
2
28.3.2 Assessment of Shear Strength by Flexural Strength
The Shear strength of concrete is analysed by considering the shear stress of prism
without reinforcement subjected to flexure test under two points load placed at a
shear span of “a” as shown in Fig. 28.4. The flexural strength of the beam fr is
assessed from simplified bending equation of two points loading. The shear strength
of the concrete is arrived through Eqs. 28.6–28.13.
3Pa
f cr = (28.6)
bD 2
3P a
f cr = (28.7)
bD D
Fig. 28.4 Flexural strength by two points loading

312 S. Kumar et al.
P
Shear force V = (28.8)
2

fr D
V = bD (28.9)
6 a

fr D
V = bD (28.10)
6 a

V fr D
= τv = (28.11)
bD 6 a

D
τv = 0.167 × × fr (28.12)
a
D Experimental
investigations shows that maximum shear arrived at section when
a
= 1 (when θ = 45°) hence the Eq. (28.4) become
τv = 0.167 fr (28.13)
Shear strength of prism τv is equated to the shear strength of concrete τc , and from
Eq. 28.2, the shear strength of concrete is assessed and explicated in Eqs. 28.14 and
28.15.
fr = f ( f ck ) = 0.3063( f ck )0.805
τc = τv = 0.0511 × f ck0.8063 (28.14)
τc = τv = f ( fr ( f ck )) = 0.0511 × f ck0.805 (28.15)
From Eq. 28.1, the compressive strength of concrete is explicated through

Eq. (28.16)
f t = f ( f ck ) = 0.1725( f ck )0.9161
f ck = 6.8( f t )1.09 (28.16)
From Eqs. (28.15) and (28.16), the shear strength of concrete as a function of
tensile strength is given in Eq. (28.17)
τc = τv = f ( fr ( f ck ( f t ))) = 0.0511 × (6.8( f t )1.09 )0.805
τv = f ( fr ( f ck ( f t ))) = 0.0511 × 4.679( f t )0.87
τv = f ( fr ( f ck ( f t ))) = 0.23 × ( f t )0.87 (28.17)

28.4 Failure Envelopes
Generally in triaxial shear test, the specimen is subjected to like stresses (compressive
stresses) in mutually perpendicular axis and the shear stress is assessed as the function
of compressive stress but the tensile strength of the materials is not considered.
Mohr’s theory of failure provides acceptable strength prediction of concrete. The
parabolic and straight line envelope has been suggested for the family of Mohr’s
circle representing the failure condition. The research works on mechanism of shear
transfer were referred in ASCE-ACI-426 [8], and this shows the failure envelope
drawn on the corresponding circles of stresses to assess the shear strength.
The principal stresses are graphically represented by drawing the Mohr’s circles.
The principal compressive stress is drawn in a positive quadrant, and principal tensile
stress is in negative quadrant of the axis. The Mohr’s failure envelope is drawn as
the function of principal compressive and tensile stresses. The value of shear stress
is the function of principal stresses measured between the points of intersection of
the tangent of the circles on y-axis to the origin.

When τc = f σ1, σ2
Graphically τc = E D
σ1
τc = ; When σ2 = 0
2
τc = 0; Even if σ2 = 0
The value of τc is existed even when the principal tensile stress is zero. The con-
crete is generally weak in tension and can be considered as critical strength parameter
rather than compressive stress parameter. The concrete structure will fail even though
it has good compressive strength when the principal tensile stress is reached to its
maximum limit. Under this condition, the shear strength of the concrete could be
considered on the tensile strength of the concrete. A new approach is introduced
here over modified failure envelope by drawing the triangle AO1O2 with angle as
shown in Fig. 28.5.
In this approach, the shear strength of the material is zero when the tensile strength
of material tends to zero. The shear strength from the modified approach can be
assessed as explained below.
314 S. Kumar et al.
Fig. 28.5 Modified failure envelope
τc is the modified function ( f m ) of σ1, σ2 and φ
σ1 = f ck ; σ2 = f t

τc = f m σ1, σ2 , φ
Graphically, τc = C D
When σ2 = 0 then φ = 0 and τc = 0; the shear strength CD is given in Eq. (28.23),
derived through Eqs. (28.18)–(28.22)
R2
tan θ1 = √ (28.18)
2 (R1 R2 )
R2 − R1
tan θ2 = √ (28.19)
2 (R1 R2 )
σ1
R2 = (28.20)
2
σ2
R1 = (28.21)
2
= θ1 − θ2 (28.22)
σ1
τc = C D = tan (28.23)
2
The shear strength of geopolymer concrete is compared with the empirical
equation suggested by ACI-318-08 [9] as shown below in Eq. (28.24)
1
τc = f c (28.24)
6
Codes are suggesting certain material properties in the absence of experimental
results. Here, ACI-318 suggesting the shear strength is based on the compressive
strength of the concrete. The results of series of shear strength from the respective
equation are shown in Table 28.4 and graphically represented in Fig. 28.6.
Table 28.4 Assessment of shear strength

Mixtures τc = f ( f t ) τc = τc = τc = τc (ACI
MPa f ( fr ( f ck )) f ( fr ( f ck ( f t ))) f m (σ1 , σ2 , φ) 318-95) MPa
(Eq. 28.5) MPa MPa MPa (Eq. 28.24)
(Series1) (Eq. 28.15) (Eq. 28.17) (Eq. 28.23) (Series 5)
(Series 2) (Series 3) (Series 4)
FGC-M1 0.94 0.41 0.34 0.49 0.54
FGC-M2 1.06 0.47 0.45 0.55 0.59
FGC-M3 1.22 0.53 0.50 0.63 0.64
FGC-M4 1.54 0.63 0.62 0.80 0.71
FGC-M5 1.76 0.71 0.69 0.90 0.76
FGC-M6 1.83 0.76 0.72 0.94 0.79
FGC-M7 2.06 0.83 0.79 1.05 0.84
FGC-M8 2.16 0.87 0.83 1.11 0.86
FGC-M9 2.26 0.90 0.86 1.15 0.88
FGC-M10 2.32 0.95 0.88 1.18 0.91
FGC-M11 2.57 1.00 0.96 1.32 0.94
FGC-M12 2.66 1.02 1.00 1.37 0.96
3
Shear strength τc (MPa)
2.5
2 Series 1
1.5 Series 2
1 Series 3
0.5 Series 4
0 Series 5
0 10 20 30 40 50
Compressive strength fck (MPa)
Fig. 28.6 Comparison of results

316 S. Kumar et al.
28.5 Conclusions
The experimental analysis of shear strength is concluded by comparing the results

with the solution by the equation suggested by ACI-318-08 for conventional concrete
which is taken as reference.
1. The shear strength of the concrete evaluated as a function of tensile stress only
(Series 1) shows a higher value. The concrete fails when the maximum tensile
strength is equal to principal tensile strength. This solution can be considered as
the upper bound solution for the maximum shear strength of plain concrete.
2. The shear strength computed on the flexural strength as a function of compressive
strength f ( fr ( f ck )) varied with range maximum of 24% and the minimum of
6% with the result of reference equation (ACI-318-08).
3. The shear strength based on the tensile stress which is the function of compressive
and flexure strength f ( fr ( f ck ( f t ))) shows the maximum of 38% and minimum
of −4% of variations.
4. The shear strength on modified function of principal stresses f m (σ1 , σ2 , φ) shows
the maximum of +40% and minimum of 10% variation.
5. The new approach on modified failure envelope provides consistent solution of
average 13% with the reference equation up to the concrete strength equal to
M-25 grade.
6. The shear strength of the concrete mix grade higher than M-25 showed com-
paratively higher shear strength by modified failure envelope in comparison of
other series of solutions. The shear strength of concrete will be significantly
more for higher grade concrete. The general equation on shear strength could not
be justifiable for high-grade concrete unless modifying the equation or experi-
mental formulation on the range of strength of concrete. The validation of these
approaches is limited with one reference and could be considered for further
discussion with other previous research significance.
7. The geopolymer concrete shows comparatively equal shear strength with con-
ventional concrete and validates with the equations suggested for conventional
concrete. The relationship between the strength parameters can be justifiable with
conventional concrete.
References
1. Lavanya, G., Jagan, J.: Evaluation of the relationship between split tensile strength and com-
pressive strength for geopolymer concrete of varying grades and molarity. Int. J. Appl. Eng. Res.
10(15), 35523–35527 (2015). ISSN 0973-4562
2. Yusuf, I.T., Jimoh, Y.A., Salami, W.A.: An appropriate relationship between flexural strength and
compressive strength of palm kernel shell concrete. Alexandria Eng. J. 55, 1553–1562 (2016)
3. IS: 3812.: Specification for pulverised fuel ash” Part-2, Bureau Of Indian Standard, New Delhi
(2003)
4. IS: 383.: Specification for coarse and fine aggregates” Bureau Of Indian Standard, New Delhi
(1970)
5. Kumar, S., Pradeepa, J., Ravindra, P.M.: Experimental investigations on optimal strength param-
eters of fly ash based geopolymer concrete. Int. J. Struct. Civil Eng. Res. 2(2), 143–152
(2013)
6. IS: 516-1956.: Methods of tests for strength of concrete” Bureau Of Indian Standard (1956)
7. Al-Sahawneh, E.I. A new approach for the determination of tensile and shear strengths of normal
weight concrete. IOSR J. Eng. 05(08), 38–48 (2015). ISSN (e): 2250-3021, ISSN (p): 2278-8719
8. ASCE-ACI Committee 426.: Shear strength of reinforced concrete members. ASCE Proc. 99(6),
1091–1188 (1973)
9. AC1: 318-2008.: “Building Code Requirements for Structural Concrete”, and Commentary
Report (AC1 318-08), ACI Committee (2008)

Ercam Book

Uploaded by

Copyright:

Available Formats

Ercam Book

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ercam Book

Uploaded by

Copyright:

Available Formats

Lecture Notes on Multidisciplinary Industrial Engineering

Series Editor: J. Paulo Davim

More information about this series at http://www.springer.com/series/15734

ISSN 2522-5022 ISSN 2522-5030 (electronic)

Bengaluru, India M. Vinyas

1 Cerium Oxide and Silicon Carbide Reinforced Al6063 Metal

9 Experimental Studies and Numerical Validation on Bearing

20 Optimization of Neem Biodiesel Blend and Injection Pressure

Dr. M. Vinyas is an Assistant Professor in the Department of Mechanical

Dr. Amelia Loja is presently Adjunct Professor at the Mechanical Engineering

Dr. Krishna R. Reddy is a Professor of Civil and Environmental Engineering, the

Department of Civil and Materials Engineering at the University of Illinois,

Abdul Nazeer and Mir Safiulla

Abstract Aluminum matrix material is considered as advanced material because

Keywords Aluminum matrix composite · Stir casting · Mechanical testing ·

A. Nazeer (B) · M. Safiulla

© Springer Nature Singapore Pte Ltd. 2020 1

coefficient of thermal expansion. Composite material is broadly classified on the

1.2 Experimental Details

1.2.1 Matrix and Reinforcement Material

1.2.2 Preparation of Composites

1.2.3 Scanning Electron Microscopy (SEM)

A scanning electron microscope (SEM) is a type of electron microscope that pro-

1.2.4 X-Ray Diffraction

1.3 Result and Discussion

1.3.1 Micro Hardness Test

Micro hardness tests were performed by applying load of 10 N for a period of 20 s. It

1.3.2 Compression Strength

The results reveal that as percentage of reinforcement increased, the compressive

1.3.3 Tensile Test

1.3.4 Charpy Impact Test

1.3.5 Fractography Studies

2% SiC 2%Ceo2 4% SiC 4% Ceo2

6% SiC 6% Ceo2 8% SiC 8% Ceo2

1.3.6 X-Ray Diffraction Analysis

• Successful synthesis and characterization of Al 6063-silicon carbide and cerium

S. Srikrishnan, V. Jayakumar and P. K. Dash

Keywords Detonation · Blockage · Pulse detonation engine

© Springer Nature Singapore Pte Ltd. 2020 11

Detonation is normally signified by short pulses of combustion using hypergolic mix-

in a large-scale annular combustor. The introduction of hydrogen in propane liquid

Diameter of Tube − Height of Blockage

Fig. 2.1 Models considered for the analysis of DDT

2.2.2 Numerical Analysis

Table 2.1 Input parameters for analysis

2.3 Result and Discussions

Table 2.2 Input pressure versus exit velocity

Fig. 2.2 Graphical representation of input pressure versus exit velocity

Table 2.3 Input pressure versus exit pressure

Fig. 2.3 Graphical representation of exit pressure versus input pressure

Fig. 2.4 Turbulent kinetic energy (100 MPa input pressure)

Table 2.4 Turbulence kinetic energy versus input pressure

Amlan Kumar Bairagi and Sujit Kumar Dalui

Keywords Spectral density · Setback tall building · Roof pressure

A. K. Bairagi (B) · S. K. Dalui

© Springer Nature Singapore Pte Ltd. 2020 21

characteristics on thirty-four number of tall buildings with analytical and experimen-

3.2 Model Detail

Fig. 3.1 Model details for a SQ and b SB

3.3 Analytical Domain and Mesh