Old Dominion University

ODU Digital Commons

Civil & Environmental Engineering Theses & Civil & Environmental Engineering
Dissertations

Spring 2018

Flexural Behavior and Strength of Doubly-Reinforced Concrete

Beams with Hollow Plastic Spheres
Rutvik R. Patel
Old Dominion University, [email protected]

Follow this and additional works at: https://digitalcommons.odu.edu/cee_etds

Part of the Civil Engineering Commons

Recommended Citation
Patel, Rutvik R.. "Flexural Behavior and Strength of Doubly-Reinforced Concrete Beams with Hollow Plastic
Spheres" (2018). Master of Science (MS), Thesis, Civil & Environmental Engineering, Old Dominion
University, DOI: 10.25777/09vp-a362
https://digitalcommons.odu.edu/cee_etds/30

This Thesis is brought to you for free and open access by the Civil & Environmental Engineering at ODU Digital
Commons. It has been accepted for inclusion in Civil & Environmental Engineering Theses & Dissertations by an
authorized administrator of ODU Digital Commons. For more information, please contact
[email protected].
FLEXURAL BEHAVIOR AND STRENGTH OF DOUBLY-REINFORCED
CONCRETE BEAMS WITH HOLLOW PLASTIC SPHERES

BY

Rutvik. R. Patel

CEE Graduate student, ODU

This Thesis is Submitted to the Faculty of Old Dominion University in Order to Fulfill the

Requirement for the Degree of

MASTER OF SCIENCE (CIVIL ENGINEERING)

OLD DOMINION UNIVERSITY

May 2018

Approved by:

Zia Razzaq (Director)

Duc T. Nguyen (Member)

Yunbyeong Chae (Member)

 ABSTRACT

FLEXURAL BEHAVIOR AND STRENGTH OF

DOUBLY-REINFORCED CONCRETE BEAMS
WITH HOLLOW PLASTIC SPHERES

Rutvik. R. Patel
Old Dominion University, 2018
Advisor: Dr. Zia Razzaq

This thesis presents the outcome of an investigation into the experimental and

theoretical flexural behavior and strength of doubly-reinforced concrete beams with and

without hollow plastic spheres. Tests are conducted on two types of beams having simply

supported end conditions. To obtain experimental results, a gradually increasing two-point

loading is used up to collapse. The experimental load-deflection and load-strain curves are

recorded. Theoretical analysis is based on developing non-linear moment-curvature

relationships for cross sections with and without hollow spheres. These moment-curvature

relationships are then coupled with three separate numerical methods namely, finite-

difference method, finite integral method and Newmark’s method to predict load-deflection

relationships for both beams. These three approximate analysis methods gave practically the

same results. In addition, a theoretical study is conducted to predict the load-deflection curves,

and the cracking and collapse load indices of full-scale beams with and without hollow

spheres. The predicted cracking and peak load values are in good agreement with those found

in the laboratory experiments. The study shows that the use of hollow plastic spheres in

reinforced concrete beams results in a substantial decrease in self-weight without

compromising the ultimate strength.

Copyright, 2018, by Rutvik. R. Patel, All Rights Reserved.
This thesis is dedicated to my family members.
 ACKNOWLEDGMENTS

First of all, I would like to thank Dr. Zia Razzaq who provided me excellent advice,

support, and comments on this thesis. His guidance and encouragement has provided me

motivation throughout my master’s program. Also I would like to thank my friends and Ph.D.

candidate, Herish Hussain, Jose Carrasquillo, and Ali Parva who provided their time in

assisting with the testing setup. A special thanks to Thomas Galloway for his technical

assistance.

Most importantly, I am grateful to my parents, Rajendrabhai Patel and Sonal Patel

for their support and encouragement. They have always stood with me and believed in me. I

also thank my brother, sister-in-law, and my lovely nephew, Kevin Patel, Krishna Kevin Patel,

and Vihaan Patel, Because of them I never felt home sick in this foreign country. Special

thanks to my lovely wife, Riya Patel, she constantly encouraged me to finish my master’s

degree.

Last but not least, I would like thank all CEE graduate students, and faculty

members for helping me throughout my study.

 NOMENCLATURE

Area of tension steel.

′ Area of compression steel

′ Resultant forces from compression zone.

Cc Concrete compression force.

c Neutral axis.

d Distance from top of beam to tension steel.

d’ Distance from top of beam to compression steel.

Ec Concrete modulus of elasticity.

fc Computed concrete compression stress.

fc’ Specified concrete compression strength.

Concrete modulus of rupture.

T Resultant forces from tension zone.

I Moment of inertia.

Mc Collapse moment.

Mcr Cracking moment.

ɸ Curvature.

∈ Bottom Strain
є Strain.

є˳ Concrete strain at maximum stress.

v Deflection.

W Applied Load.

R12 Conjugated load.

wself Self-weight of beam.

Wcr Cracking load.

Wc Collapse load.

h Segment length.

S.G. Strain gage.

ηcr Ratio of cracking load to self-weight of beam.

ηc Ratio of collapse load to self-weight of beam.

R Radius of hollow sphere.

 i

TABLE OF CONTENTS
Page

LIST OF TABLE………………………………………………………………………………………iv
LIST OF FIGURES……………………………………………………………………………………vi

1. INTRODUCTION ................................................................................................................................ 1

1.1 Background ............................................................................................................................... 1

1.2 Literature Review ...................................................................................................................... 3

1.3 Problem Statement .................................................................................................................... 7

1.4 Objectives and Scope ................................................................................................................ 9

1.5 Assumptions and Conditions................................................................................................... 10

2. THEORITICAL ANALYSIS.............................................................................................................. 11

2.1 Material Properties .................................................................................................................. 11

2.1.1 Stress-Strain Relation for Concrete ......................................................................................... 11

2.1.2 Stress-Strain Relation for Steel ............................................................................................... 12

2.2 Moment-Curvature Relation ................................................................................................... 13

2.2.1 Solution Algorithm for Moment-Curvature Relation for Solid Cross Section........................ 19

2.2.2 Solution Algorithm for Moment-Curvature Relation for Cross Section with Hollow Sphere 21

2.3 Numerical Results ................................................................................................................... 22

2.4 Finite-Difference Analysis for Predicting Load-Deflection Curves ....................................... 24

2.4.1 Boundary Conditions .............................................................................................................. 25

2.4.2 Finite Difference Formulation................................................................................................. 25

2.4.3 Finite-Difference Algorithm. .................................................................................................. 26

2.4.4 Numerical Results ................................................................................................................... 27

2.5 Finite Integral Analysis ........................................................................................................... 29

2.5.1 Finite Integral Algorithm ........................................................................................................ 32

2.5.2 Numerical Results ................................................................................................................... 33

 ii

2.6 Analysis based on Newark’s method ...................................................................................... 34

2.6.1 Computation of Moments in beam .......................................................................................... 34

2.6.2 Computation of Deflection Value ........................................................................................... 35

2.6.3 Algorithm based on Newmark method ................................................................................... 36

2.6.4 Numerical Results ................................................................................................................... 37

2.7 Theoretical Study of Large Beams ........................................................................................... 38

2.8 Numerical Results .................................................................................................................... 41

3. EXPERIMENTAL INVESTIGATION .............................................................................................. 50

3.1 Material Properties ................................................................................................................... 50

3.1.1 Stress-Strain Relation for Steel ................................................................................................ 50

3.1.1.1 Test Specimen and Specimen Cross Section Properties ......................................................... 50

3.1.1.2 Experimental Results .............................................................................................................. 51

3.1.2 Compression Test of Concrete ................................................................................................. 52

3.2 Test Procedure for Beams ........................................................................................................ 53

3.3 Experimental Results ............................................................................................................... 56

3.3.1 Flexural Test to Determine the Peak Load Capacity................................................................ 56

3.3.2 Flexural test on Beam 1............................................................................................................ 56

3.3.3 Flexural Test on Beam 2 .......................................................................................................... 63

4. COMPARISION OF RESULTS ......................................................................................................... 69

4.1 Comparison of Moment Curvature Relation ............................................................................ 69

4.2 Comparison of Load-Deflection Relation ................................................................................ 69

4.3 Experiments versus Theory ...................................................................................................... 70

5. Conclusion and Future Research ......................................................................................................... 82

5.1 Conclusion ............................................................................................................................... 82

5.2 Future Research ........................................................................................................................ 83

Reference .................................................................................................................................................... 84

APPENDIX A ..................................................................................................................................... 86
 iii

(a) Computer programing for Finite Difference Method .................................................................... 86

(b) Computer program for Finite Integral Method .............................................................................. 88

(c) Newmark’s method ........................................................................................................................ 94

APPENDIX B ..................................................................................................................................... 96

Computer Programming for Non-Linear Moment Curvature Relation............................................... 96

APPENDIX C ..................................................................................................................................... 99

Experimental Data............................................................................................................................... 99

VITA ......................................................................................................................................................... 101

 iv

LIST OF TABLES

Tables Page

1. Stiffness test report summary [8] ......................................................................................... 6

2. Cracking and collapse loads indices .................................................................................. 48

3. Cross-Sectional properties of specimen............................................................................. 50

4. Compression Test data (fc’)................................................................................................ 52

5. Cracking load and peak load results .................................................................................. 56

6. Comparison of bending moments and curvature ............................................................... 69

7. Comparison of experimental and theoretical results ......................................................... 70

8. Summary of results ............................................................................................................ 81

9. N matrix for Finite-difference Method .............................................................................. 87

10. N inverse matrix .............................................................................................................. 87

11. Curvature and Deflection values ..................................................................................... 88

12. [N] Matrix for Finite Integral method ............................................................................. 89

13. N2 Matrx .......................................................................................................................... 90

14. Nn2 Matrix ........................................................................................................................ 91

15. Nn2 z matrix ..................................................................................................................... 92

16. Matrix .................................................................................................................... 93

17. Deflection and curvature values ...................................................................................... 94

19. Experimental results of Beam 1 ....................................................................................... 99

20. Experimental results of Beam 2 ..................................................................................... 100

 v

LIST OF FIGURES

Figures Page

1. Semi-precast section for a slab ........................................................................................ 2

2. Cast-in-situ system for a slab ........................................................................................... 2

3(a) Solid reinforced concrete beam…………………………………………………………8

3(b) Solid cross section……………………………………………………………………....8

3(c) Reinforcement detail ....................................................................................................... 8

4(a) Beam with hollow spheres ............................................................................................... 9

4(b) Cross section of beam ...................................................................................................... 9

4(c) Reinforcement details for beam with hollow spheres ...................................................... 9

5. Stress-Strain relation for concrete .................................................................................. 12

6. Stress-strain relation for steel ......................................................................................... 13

7. Schematics of the cross section with geometry of the hollow sphere ............................ 16

8. Cross-sectional strain, stresses, and forces for solid R.C. beam..................................... 16

9. Cross-sectional strain, stresses and forces of hollow sphere beam ................................ 17

12. Strain, stresses, and force diagram for solid beam ......................................................... 18

13. Strain, stresses, and force diagram for beam with hollow sphere .................................. 19

14. Moment-curvature relation for solid cross section ......................................................... 23

15. Moment-curvature relation for cross section with hollow sphere .................................. 23

16. Finite difference discretization ....................................................................................... 24

17. Load-deflection curve for solid reinforced concrete beam ............................................. 28

18. Load-deflection for reinforced concrete beam with hollow spheres .............................. 28

19. Load deflection curve for solid R.C beam ...................................................................... 33

 vi

21. Equation for continuous polygonal curve [16] ............................................................... 36

22. Load-deflection for beam with hollow sphere ................................................................ 37

23. Load-deflection for solid beam ....................................................................................... 38

24. Cross section of Beam 4 ................................................................................................. 39

25. Cross section of Beam 6 ................................................................................................. 40

26. Cross section of Beam 8 ................................................................................................. 40

27. Load-deflection curve for Beam 3 .................................................................................. 42

28. Load-deflection curve Beam 4........................................................................................ 43

29. Load-deflection curve for Beam 5 .................................................................................. 44

30. Load-Deflection curve for Beam 6 ................................................................................. 45

31. Load-Deflection curve for Beam 7 ................................................................................. 46

32. Load-Deflection curve for Beam 8 ................................................................................. 47

33. Cracking load index versus beam length ........................................................................ 48

34. Collapse load index versus beam length ......................................................................... 49

35. Tensile test equipment. ................................................................................................... 51

36. Stress-strain curve for sample specimen......................................................................... 52

37. Compression Test ........................................................................................................... 53

38. Test Setup for Two-point Load system .......................................................................... 54

39. Strain gauge location on (a) top surface (b) bottom surface (c) elevation surface ......... 55

40. Cracking pattern of Beam 1 ............................................................................................ 58

41. Cracking pattern in bottom surface of beam 1 after testing............................................ 58

42. Load-strain curves for Strain Gauge 1, 3, and 4 ............................................................. 59

43. Load-strain curves for Strain Gauge 2, 5, and 6 ............................................................. 60

 vii

44. Load- strain curves for Strain Gauge 7, 8, and 9 ............................................................ 61

45. Combine Strain Gauge curves ........................................................................................ 62

46. Experimental load-deflection curves .............................................................................. 63

47. Testing setup for Beam 2 ................................................................................................ 64

48. Cracking pattern in bottom surface of Beam 2 after test ................................................ 65

49. Load-strain curves for Strain Gauge 1, 2, and 3 ............................................................. 65

50. Load-strain curves for Strain Gauge 4, 5, and 6 ............................................................. 66

51. Load-strain curve for Strain Gauge 7, 8, and 9............................................................... 66

52. Experimental load-deflection curve ................................................................................ 67

53. Combine load-strain curves ............................................................................................ 68

54. Comparison of moment-curvature curves ...................................................................... 71

55. Comparison of theoretical load-deflection curves .......................................................... 71

56. Comparison of experimental load-deflection curves ...................................................... 72

57. Comparison of theoretical and experimental load-deflection curves ............................. 73

58. Comparison of theoretical and experimental load-deflection curves ............................. 74

59. Load-Deflection curves for 15 feet beam ....................................................................... 75

60. Load-Deflection curves for 21 feet beam ....................................................................... 76

61. Load-Deflection curves for 27 feet beam ....................................................................... 77

62. Theoritcal versus experimental load-strain values for Beam 1 ...................................... 78

63. Theoritical versus experimental load-strain curve for Beam 2 ....................................... 78

viii
 1

Chapter 1

INTRODUCTION
1.1 Background

A new reinforced concrete construction technique was invented in Europe around

1990 that combined the use of hollow polyethylene plastic spheres embedded in regions of

low normal stresses of a structure. The use of such hollow sphere results in a reduction of

structural self-weight of about 20 to 25 percent. The present study is focused on assessing the

effectiveness of plastic spheres in a doubly reinforced concrete beam and on determining its

flexural behavior and strength.

A reinforced concrete structure with hollow spheres can be constructed in three

ways: precast, semi-precast, and cast-in-situ. Figure 1 shows an example of a semi-precast

application. This technique involves use of the hollow spheres and most of the main

reinforcement for a given structure. The elements are then stitched on-site through a concrete

placement. Figure 2 shows an example of cast-in-situ setup in which hollow spheres are placed

in modules between the top and bottom steel. The modules are then placed on conventional

formwork followed by placement of concrete. Precast reinforced concrete structural units with

hollow spheres can be delivered to a site as fully precast elements. This option however,

becomes less continent for large span structures. A precast application is generally used and

formed to be economical for slabs.

A review of the existing literature revealed that a number of studies have been

conducted on the performance of slabs with hollow spheres, also known as bubble decks.

However, no studies have been published for reinforced concrete beams with such hollow

spheres. This thesis presents a theoretical and experimental study of simply-supported doubly
 2

reinforced concrete beams of both types with and without hollow spheres under gradually

increasing statics loads up to collapse. Numerical results include the prediction of moment-

curvature and load-deflection relationships. The nonlinear load-deflection curves are arrived

at using finite-difference method, finite-integral, and Newmark’s methods.

Figure 1. Semi-precast section for a slab

Figure 2. Cast-in-situ system for a slab

 3

1.2 Literature Review

As mentioned in Section 1.1 of this thesis, the past studies have been focused on

reinforced concrete slabs and not beams with hollow spheres. Consequently, the literature

review presented below relates to reinforced concrete slabs with hollow spheres.

Churakov [1] presented in his study that in 1990, a new construction technique

which is called bubble deck technology was invented by Berunning to link airspace and steel

using a voided biaxial concrete slab. The bubble deck technology uses spheres made of

recycled industrial plastic to create air voids while providing strength through a section. As a

result, this allows the hollow slab to act as a normal monolithic two-way spanning concrete

slab. Bubble deck slabs can be lighter, stronger, and thinner than conventional reinforced

concrete slabs.

In 2012, a study had been conducted by Ibrahim, Ali and Salman [2] on the flexural

capacities of reinforced two-way bubble deck slabs. A bubble deck slab has a two-dimensional

arrangement of voids within the slabs to reduce self-weight. The behavior of bubble deck slabs

is influenced by the ratio of bubble diameter to slab thickness. To verify the flexural behavior

of bubble deck slabs such as ultimate load, deflection, concrete compressive strain and crack

pattern two-dimensional flexural tests were tested by using special loading frame. Results

have shown that the crack pattern and flexural behavior depend on the void diameter to slab

thickness ratio.

During their research Calin, Gintu, and Dascalu [3] found out that bubble deck slab

can omit a significant volume of concrete in the central core where the slab is principally un-

stressed in flexure. In slabs, the depth of compressed concrete is usually a small proportion of

the slab depth and this means that it almost always involves only the concrete between the
 4

ball and the surface, so there is no sensible difference between the behavior of a solid slab and

bubble deck slab.

In 2012, Teja and Kumar [4] studied the durability of bubble deck slab and

explained it on the basis of creep and shrinkage. A bubble deck element was compared with

a solid concrete block of the same dimension and of the same quality of concrete. Then, the

difference between the shrinkage strains of these two was measured. The results show that

bubble deck element has a negligible larger marginal shrinkage strain than a solid slab with

equivalent dimension, same concrete performances, and under the same exposure to

environmental conditions. The influence of carbonation shrinkage can be neglected in the

design of concrete structures with bubble deck system because only a small part of the

concrete cross section is exposed to this kind of shrinkage.

In 2010, Sharma, Mounika and Purnachandra [5] conducted studies on the fire

resistance of bubble deck slabs. The analysis was first done on a hollow core slab without fire

for two charges, one that leads to elastic dynamic response and the other that causes plastic

behavior and severe concrete cracking. The same blast analysis had been subjected to fire.

There were many difficulties in obtaining a reliable result. A discussion of the Experimental

setup and experimental results are compared with simplified numerical models solved with

the software LS-DYNA. Fire does not change the material and structural properties that fast

as compared to an explosion. The most important conclusion of the analysis is that crack

patterns and blast load dynamic responses are indeed altered by fires with temperature up to

4500C. Yet within the limitations of assumptions concerning boundary conditions, the

examined slabs keep their blast bearing capacity after blast load scenarios up to 1.5kg C4 with

at 1m standoff distance.
 5

In 2009, Lai [6] discussed the acoustic behavior of bubble deck slabs in “Structural

behavior of bubble deck slabs and their applications” and found that bubble deck performs

acoustically in a better way than any other hollow or solid floor surfaces. Because of the three-

dimensional structure and the graduated force flow, the hollow spheres have a positive

influence on sound insulation. The tests reveal that the airborne sound insulation is even higher

than expected. This indicates the bubbles have a positive influence on sound insulation. The

main criteria for reducing noise is the weight of the deck and therefore bubble deck evidently

will not act otherwise than other deck types with equal weight.

Schnellenbach and Pfeffer [7] from the Institute for Concrete Structures and

Materials as the Darnstadt University of Technology conducted another large study on the

punching behavior of bubble deck. Two different depths, 240 mm and 450 mm, were used to

model the shallowest and deepest variety of the slabs. The slab was made of standard B25 and

B35 concrete with a maximum aggregate size of 16 mm and attached to a short column in

order to simulate the response. The slabs were radially supported at eight points and were

monitored by strain gauges, deflection gauges, and extensometers. The tests proved that

although the HDPE spheres did not influence the crack pattern along the slab, the resistance

to punching shear was less than a solid slab. When sawn open, the cross section showed that

the crack angle varied from 30o to 40o. In order to further understand the structural mechanics

of the Bubble deck, the researchers generated a 3D nonlinear finite element model of the slab

with software DIANA. The FEM analysis conformed to the results of the physical

investigations and verified the punching shear behavior of bubble deck. They suggest reducing

the allowable shear area if any bubbles intersect the control perimeter so that those spheres

will not play a role in the punching shear resistance. These findings correspond with other
 6

studies in that they recommend mitigating the punching shear response by excluding HDPE

spheres from the shear perimeter.

Mann [8] at the technical university of Darmstadt in Germany also performed tests

on the stiffness of a bubble deck slab. The results verified with the theoretical analysis and

with the physical tests done in the Netherlands [8]. For the same strength, bubble deck has

87% of the bending stiffness of a similar solid slab but only 66% of the concrete volume due

to the HDPE bubbles. As a result, the typical deflection was marginally higher than that of a

solid slab, as expected. However, the significantly lower deadweight compensated for the

slightly reduced stiffness, and therefore gave bubble deck a higher carrying capacity. Table 1

summarizes the results of their experiments. Analyses have also proven that deflections under

service loads were a little higher than that of an equivalent solid slab. On the other hand, the

reduced permanent load positively affects the long-term response in the serviceability limit

state (SLS) design, which governs crack propagation.

Table 1. Stiffness test report summary [8]

*On the condition of same amount of steel, the concrete itself has 220% greater effect

The Technical University of Denmark and AEC Consulting Engineers Ltd, led by

Professor Nielsen [18], tested both the shear strength and punching shear resistance. They

used a slab depth of 188 mm, which is not a typical bubble deck thickness, and used a

force/Thickness (a/d) ratio of 1.4. They found that shear strength for bubble deck was

approximately 80%. For punching shear they experimented on slabs with depths of 230 mm
 7

and 450 mm. They found that the crack pattern was similar to that of a solid slab and that local

punching failure did not occur within the given load cases. The average experimental value of

the punching shear capacity of this slab was about 90% of a solid slab. The test specimens

actually performed better than the theoretical models, but still not as good as a solid concrete

slab.

To the best of the author’s knowledge, a study of reinforced concrete beams having

hollow spheres has not been published in the past.

1.3 Problem Statement

The main problem addressed in this thesis is to conduct an experimental and

theoretical study of the flexural behavior and strength of simply-supported doubly reinforced

concrete beams. The corresponding beams with no hollow spheres are also studied for

comparison, and are referred to as solid beams. Figure 3a shows a solid beam with a uniform

cross section. The cross section and reinforcement details are shown in Figure 3b and 3c.

Figures 4b and 4c show the cross section and reinforcement details of a beam with hollow

spheres. The theoretical problem is to develop rigorous nonlinear moment-curvature relations

for these cross sections and then couple them with central finite-difference, finite integral and

Newmark’s solution scheme to predict the load-deflection relation of the beam up to collapse.

This theoretical data is then used to compare to the load-deflection curve obtained from the

experimental results. Lastly, the theoretical prediction model is applied to full-scale beams

both with and without hollow spheres.

 8

(a) Solid reinforced concrete beam

(b) (c)
Figure 3. (a) Solid reinforced concrete beam, (b) Cross section, and (c) Reinforcement detail
 9

(a) Beam with hollow sphere

Hollow
sphere

(b) (c)
Figure 4. (a) Beam with hollow spheres, (b) Cross section of beam, and (c) Reinforcement

details for beam with hollow spheres

1.4 Objectives and Scope

With reference to both the solid reinforced concrete beam and the reinforced beam

with hollow spheres, the primary objectives of this research are to:

1. Experimentally study the flexural behavior and strength of doubly reinforced concrete

beams with and without hollow spheres;

 10

2. Generate an algorithm for nonlinear moment-curvature relationships;

3. Predict theoretical load-deflection relationships by coupling a moment-curvature

relationship with three numerical methods, namely, central finite-difference method, finite

integral method and Newmark’s method;

4. Predict both moment-curvature and load-deflection curves for beams with real-life span

and with hollow spheres of three different diameter;

5. Assess the effectiveness of beams with hollow spheres in comparison to solid beam.

In this research, scale down model beams with cross-sectional dimensions of 6 × 6

in., and a span of 36 in. are used. Ultimate strength of concrete used is 4,000 psi. The nominal

yield stress of steel reinforcement is 71,000 psi. The steel reinforcement consists of four no.3

rebar, and stirrups are made from no.2 rebar. Hollow spheres shown in Figure 4c, have a

diameter of 2.5 in.

1.5 Assumptions and Conditions

The main assumptions and conditions adopted in this study are:

1. Beams have simply-supported end conditions;

2. The beams are loaded by two-point loads and gradually increase up to collapse;

3. The bond between the hollow sphere and stirrups is maintained during casting process as

well as during loading;

4. Bernoulli-Navier plane section hypothesis is adopted;

5. The tensile resistance of concrete is neglected in the analysis after the stress at a given

cross section has reached the modulus of rupture;

6. Concrete is assumed to fail when the compressive strain reaches 0.003 in. /in.
 11

CHAPTER TWO

THEORETICAL ANALYSES
This chapter presents detail about the numerical results obtained from the

theoretical study of both doubly reinforced concrete beams with and without hollow spheres.

Numerical results are presented in the form of moment-curvature relationship, and load-

deflection relationships. Once the non-linear moment-curvature relations are developed, the

load-deflection curves are predicted using three different approaches, namely, the finite-

difference, the finite integral, and Newmark’s methods.

2.1 Material Properties

The physical properties of concrete and steel reinforcement used in this study are

summarized in this section.

2.1.1 Stress-Strain Relation for Concrete

In this thesis, concrete with a compressive strength of 4,000 psi is used. The

following nonlinear normal stress-strain relation for concrete given by Lin and Burns [14] is

adopted:

є є
= ′ 2 (1)
є˳ є˳

Where:

= Computed concrete compression stress,

′= Specified concrete compression stress,
є = Strain,
є˳= Concrete strain at maximum stress.
Figure 5 is a graph of Equation 1.
 12

2.1.2 Stress-Strain Relation for Steel

An elastic-plastic stress-strain relationship as shown in Figure 6 is adopted for the

steel reinforcement. Thus, the following relationship are applicable:

= є Es, |є | є (2a)

= fy , | є | є (2b)

=- , є є (2c)

4500
fc'
4000

3500
Stress, fc (psi)

3000

2500

2000

1500

1000

500

0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035

strain, єc (in. /in.)

Fig 5. Stress-Strain relation for concrete

 13

Steel stress, fs, (psi)

0
Steel strain, є , (in. /in.)

Figure 6. Stress-strain relation for steel

2.2 Moment-Curvature Relation

To determine the moment-curvature relationships for the cross sections shown in

Figures 3b and 4b, the location of the neutral axis needs to be determined for each value of

the gradually increasing applied load. With reference to Figure 8, the axial force equilibrium

equation is written as follows:

ΣFx = Cc + C ′ – T = 0. (3)

Where:

T= (4)

′= ′ ′ (5)

In this expression, As and As’ are areas of tensile and compressive steel reinforcements,

respectively, , and ′ are based on either hook’s law or material yield stress, fy, as applicable.

The term Cc represents the total compressive force in concrete and is given by:

Cc = ∫ f b dx (6)
 14

In which dx is the depth of a concrete compressive elemental area shown in Figure 8. For the

solid reinforced concrete cross section, the right hand side integral in Equation 6, can

approximately be determine by using the following expression:

Cc = ∑ (ΔAc)i (7a)

Cc = ∑ (ΔAco)i (7b)

In which is the concrete compressive stress, (ΔAc)i , and (ΔAco)i are the finite elemental area

as shown schematically in Figure 7 given by:

(ΔAc)i = bi (Δx)i (8)

By substituting the value of (ΔAc)i into Equation 7a, Cc is found out for solid cross section. By

using values of T, Cc, and Cs’ in Equation 3, try to satisfy the force equilibrium condition

iteratively.

Similar approach is used for locating the neutral axis location with gradually

increasing load for cross section with hollow sphere. With reference to Figure 9, the axial

force equilibrium equation should be same as Equation 3. Figure 7 shows the location of

hollow sphere inside the cross section. Due to the geometric property of hollow sphere shown

in Figure 7, the width of concrete in each layer varies with the depth of the cross section. To

find the actual width of concrete layers we have to subtract the width, 2Y, which is occupied

by each layer of the hollow sphere and can be expressed as:

Y=√ (9)

As shown in Figure 7, R is a radius of the hollows sphere and X is a vertical length from the

center of sphere to the C.G. of layer. Computed values from Equation 9, are used to compute

(ΔAco)i which is schematically shown in Figure 7 for cross section with hollow sphere given

as:
 15

(ΔAco)i = (b - 2Y)i Δxi (10)

Values computed using given equations are coupled with Equation 7, to find out the Cc for

cross section with hollow sphere. By using the computed values of T, Cc, and Cs’ in Equation

3, try to satisfy the force equilibrium condition iteratively.

The bending moment equilibrium is written as:

Mext = Mint (11)

In this equation, Mext is obtained from the bending moment diagram shown in Figure 10b, and

express as follows:

Mext = W z, 0 < z < Lo (12a)

Mext = W Lo, Lo < z < L/2 (12b)

With reference to Figure 8, the internal resisting bending moment Mint is calculated using the

following expression:

Mint = ∑ 0 (ΔAc)i xi ) - ′ ′ (d-d’) (13)

Where xi is the distance of a finite elemental layer to the tension zone as shown in Figure 7,

and ′ is a compressive steel stress itself taken as a negative quantity.

The computational procedure involves iteratively satisfying the equilibrium

conditions represented by Equation 3 and Equation 13. For given cross sections, the converged

value of c for a specified load, W is used to compute the corresponding curvature value, Φ.

The corresponding moment, M, is obtained from the bending moment diagram which is in

equilibrium with Mint given by Equation 13. Detailed steps for generating moment-curvature

relationships for cross section shown in Figure 3b and 4b are given in the following section

2.2.1 and 2.2.2 respectively.

 16

(ΔAc)i

c (ΔAco)i
d
H
xi

Figure 7. Schematics of the cross section with geometry of the hollow sphere

Strain Stresses and

forces

Figure 8. Cross-sectional strain, stresses, and forces for solid R.C. beam.
 17

Stresses and
Strain forces

Figure 9. Cross-sectional strain, stresses and forces of hollow sphere beam

Δxi Δxi

(a) Solid cross section (b) Cross section with

hollow sphere

Figure 10. Schematics of discretized cross sections

 18

(a) Beam schematics

(b) Bending moment diagram

Figure 11. Beam schematics and bending moment diagram

0.85 ′

Strain Stresses and forces

Figure 12. Strain, stresses, and force diagram for solid beam
 19

0.85 ′

Strain Stresses and forces

Figure 13. Strain, stresses, and force diagram for beam with hollow sphere

2.2.1 Solution Algorithm for Moment-Curvature Relation for Solid Cross Section

This section gives detailed steps for generating moment-curvature relationships for

the cross section shown in Figure 3b. Algorithm steps are as follows:

1. Specify the dimension and material properties of the solid reinforced cross section.

2. Assume a neutral axis depth c and a value for ϵc as shown in Figure 7 and 8.

3. Divide the cross section in-to n number layers as shown in Figure 10a.

4. With strain values from Step 3, compute stress, fci, values using Equation 1.

5. Specify all forces on stress diagram as shown in Figure 8.

6. By using Equation 4, and 5 find out the values for T and Cs’.

7. Using Equation 8, compute the value of (ΔAc)i for each layer.

8. With (ΔAc)i from Step 7, and stress values fci, from Step 4, compute Cc using Equation 7a.
 20

9. With T, Cc, from Step 4, and Cs’ from Step 8, check if Equation 3 is satisfied or not; if not

satisfied then consider different value for c; and go to Step 2 and repeat Steps 2-9 until the

equilibrium condition given by Equation 3 is satisfied.

10. Compute concrete modulus of rupture, f , and elastic modulus, Ec, using the following

ACI equations [11]:

= 7.5 ′ (14)

Ec = 57,000 ′ (15)

11. With fr and Ec from Step 10, compute the bending moment at cracking by using:

Mcr = (16)

12. With fr and Ec from Step 10, compute bottom strain ϵct, using the following equation:

ϵct = (17)

13. With ϵct from Step 12, compute curvature at cracking by using following expression:

∈
Φ= (18)

14. Compute the internal resisting bending moment given in Equation 13, using converged

values of c for each load level.

15. Using converged c values for each load level, compute the corresponding curvature values.

In order to determine if the above algorithm provides the similar ultimate moment capacity as

one can obtain from Whitney’s theory [19], the following procedure is also used herein. Using

force equilibrium, the depth of Whitney’s stress block, a, shown in Figure 12 is:

′
a= (19)
.
 21

The ultimate bending moment

Mu = Cc + Cs’ ′ (20)

Where:

Cc = 0.85 × ′ × b × a (21)

The ultimate curvature, Φu, using the strain diagram shown in Figure 13, is given by:

.
Φu = (22)

The above iterative algorithm is programmed to obtain numerical results which are

shown in Figure 14. Complete listing of a computer program to find out moment-curvature

relation based on the provided algorithm is shown in Appendix B.

2.2.2 Solution Algorithm for Moment-Curvature Relation for Cross Section with

Hollow Sphere

This section gives detailed steps for generating moment-curvature relationships

with converge values of c for each load increment of cross section shown in Figure 4b. The

overall solution process is as follows:

1. Specify the dimension and material properties of the cross section with hollow sphere.

2. Assume a neutral axis depth c and a value for ϵc as shown in Figure 7 and 9.

3. Divide the cross section in-to n number layers as shown in Figure 10b.

4. With strain values for each layer from Step 3, compute stress fci values using Equation 1.

5. Specify all forces on stress diagram as shown in Figure 9.

6. By using Equation 4 and 5 find out the values for T and Cs’.

7. Using Equation 9, compute the width 2Y for each layer of hollow sphere.

8. With 2Y from Step 7, compute the value of (ΔAco)i using Equation 10.
 22

9. With (ΔAco)i from Step 8, and stress fci, from Step 4, compute Cc using Equation 7b.

10. With T, Cc, from Step 4, and Cs’ from Step 8, check if Equation 3 is satisfied or not; if not

satisfied then consider a different value for c; and go to Step 2 and repeat Step 2-9 until

equilibrium condition given by Equation 13 is satisfied.

11. Using Equation 14 and 15 find values of fr and Ec.

12. With fr from Step 8, compute bending moment at cracking using Equation 16.

13. With fr and Ec from Step 13, compute bottom strain ϵct using Equation 17.

14. With ϵct from Step 14, compute curvature at cracking by using Equation 18.

15. Compute the internal resisting bending moment given in Equation 13, using converged

values of c for each load level.

16. Using converged c values for each load level, compute the corresponding curvature values.

The above iterative algorithm is programmed to obtain numerical results which is shown in

Figure 15. Complete listing of a program is shown in Appendix B.

2.3 Numerical Results

This section provides graphical representation of moment-curvature relationships

based on algorithms in Section 2.2.1 and 2.2.2. Figures 14 and 15 represent the moment-

curvature curves for both cross sections with and without hollow spheres respectively.
 23

70
65
60
Bending Moment, M (Kip-in)

55
50
45
40
Actual
35
30 Curve fitted
25
20
15
10
5
0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004

Curvature , Φ (radi./in.)

Figure 14. Moment-curvature relation for solid cross section

70
65
Bending Moment, M (Kip-in)

60
55
50
45
40
35
30 For cross section with
25
hollow sphere
20 Curve-fitted
15
10
5
0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035

Curvature , Φ (radi./in.)

Figure 15. Moment-curvature relation for cross section with hollow sphere
 24

2.4 Finite-Difference Analysis for Predicting Load-Deflection Curves

A central finite-difference method [18] to develop load-deflection relations is

described in this section. Briefly, the finite-difference procedure involves expressing

derivatives in terms of the nodal values of dependent variables. As shown in Figure 16, this

approach divides the beam into n number of equal length segments including the points

outside the boundary of the beam called the phantom points.

Lo
Z

Figure 16. Finite difference discretization

The beam slope at any node i is given by:

–
( )i = (23)

The beam curvature, ɸ, at any given node is given by:

–
≈ = -Φ (24)

In Equations 23 and 24, vi represents the deflection value at any node i, h is the segment

length, and i represents the node number. Equation 24 is applied at each node over [0, L/2]

along with boundary conditions provided in Section 2.4.1. Due to symmetry of the beam
 25

geometry and loading conditions, only half of the beam needs to be analyzed, that is i = 1, 2,

3, …, n, where n = 9 is adopted for this study.

2.4.1 Boundary Conditions

Since the deflection and the curvature at i = 1 are zero at the left support,

=0 (25a)

)1 = 0 (25b)

Using Equations 24 and Equation 25b, one gets:

= (25c)

Also, due to zero slope at mid-span:

( )L/2 = 0 (25d)

which upon using Equation 23 for i = 9 leads to:

= (25e)

Equations 25a, 25c, 25e are used in the formulation presented in Section 2.4.2.

2.4.2 Finite Difference Formulation

The numerical procedure is based on a second-order central finite-difference

technique [18] which is, applied to Equation 24 at n equidistant nodes over [0, L/2]. Using

appropriate boundary and symmetric conditions mentioned in the section 2.4.1 together with

the Equation 24 for the nodes i = 0, 1, 2…, 9, the following matrix equation is obtained:
 26

 2 1 0 0 0 0 0 0  v2   2 
 1 2 1 0 0 0 0 0    
  v3   3
0 1 2 1 0 0 0 0  v4   4 
    
0 0 1 2 1 0 0 0  v5   5 
0 0 0 1 2 1 0 0     (-h ) 6 
  2

  v6   
(26a)
0 0 0 0 1 2 1 0  v7   7
0 0 0 0 0 1 2 1    8 
  v8   
0 0 0 0 0 0 2 2  v9   9 
    
  

[N] {v} = - h2 {Φ} (26b)

Appendix B presents the steps to formulate Equation 26b. The vector {Φ}

represents curvature values at various nodes, and {v} represents the deflection vector. The

curvature vector {Φ} in Equation 26b can be computed using the non-linear moment-

curvature relations. Using excel curve-fitting function, curvature equations are formulated in

terms of bending moments, M. Thus for solid cross section,

Φ = 9E-09M3 - 5E-07M2 + 1E-05M - 2E-05, 0 M 62.5 kip-in. (27a)

Φ = 0.0003M2 - 0.0363M + 1.1487, 62.5 M 66 kip-in. (27b)

And for a cross section with hollow sphere,

Φ = 1E-08M3 - 7E-07M2 + 2E-05M - 3E-05, 0 M 62.5 kip-in. (28a)

Φ = 0.0009x - 0.0559, 62.5 M 64.8 kip-in. (28b)

The bending moment, M, at any node i can be computed using Equation 12a and 12b

respectively.

2.4.3 Finite-Difference Algorithm.

A finite-diffrence based algorithm for genrating load-deflection relationships using

Equation 18a is developed and is presented herein. The steps are as follows:

1. Define the total number of segments on the beam.

 27

2. Define the boundary conditions for the beam as shown in Section 2.4.1.

3. For i = 0, 1, 2, …, 9 compute all simultanious equations by using Equation 24.

4. With all differential equations from Step 3, and boundary condition from Step 2, simplify

these equations.

5. With simplified equations from Step 4, generate the matrix Equation 26a.

6. Using Equation 12a and 12b compute the moments for each nodes respectively.

7. With moments from Step 6, compute the curvature vector {ɸ} using Equation 27a, 27b

and 28a, 28b respectively for both beams.

8. With curvature {ɸ} vector from Step 7, and [N] matrix from Step 5, compute Equation

26b in order to find the values of delfection vector {v}.

9. Repeat Steps 6 to 7, until the collapse load condition is reached.

The above iterative algorithm is programmed to compute the load-deflection relationships for

both the beams. A complete listing of computer programming based on the above algorithm

is given in Appendix A.

2.4.4 Numerical Results

This section provides the load-deflection curves based on the algorithm provided

in section 2.4.3 for both reinforced concrete beams with and without hollow spheres. Figure

17 and 18 represents the load-deflection curve obtain theoritically for both beams with and

without hollow spheres respectively.

 28

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Mid-span deflection, v (in.)

Figure 17. Load-deflection curve for solid reinforced concrete beam

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Mid-span deflection, v (in.)

Figure 18. Load-deflection for reinforced concrete beam with hollow spheres
 29

2.5 Finite Integral Analysis

Finite integral method is a promising method of solving differential equations. It is

generally of superior accuracy by comparing with the well-known finite-difference method.

Brown and Trahair [12] used the finite integral approach to obtain numerical solutions of

linear ordinary differential equations. Usami and Galambos [14] used the finite integral

approach in a study of single angle beam-columns. Zhao and Razzaq [13] studied the behavior

of biaxially loaded steel beam-columns subjected to a high temperature using a finite integral

formulation. Mamadou and Razzaq [15] use finite integral method to study inelastic behavior

and strength of steel beam-column with applied torsion.

Briefly, the finite integral procedure involves replacing the continuous differential

equations which must be satisfied everywhere by a series of simultaneous equations which,

represent the differential equations at a series of discrete points. All but the highest differential

coefficients in these equations are eliminated by replacing them by linear combinations of the

highest differential coefficients and of the constants of integration, these combinations being

determined by the method of finite integrals. The resulting simultaneous equations may be

combined with the boundary conditions and solved for the highest differential coefficients.

The discrete values of the dependent variables are then calculated by back-substitution into

the finite integral expressions. If the variation of a function f over an interval zj < z < zj+1 such

that zj+1 - zj = h, is approximated by a parabola:

f = az2 + bz + c (29)

And fitted to three adjacent values of f, it can be shown by the following equation:

5 + 8 fi+1 – fi+2) (30)

4 + 16 fi+1 + 4 fi+2) (31)

 30

For an integral li defined by:

li = (32)

The matrix equation formed by single equation of the Equation 30 and 31 can be written as:

{li}= f (33)

Here;

{l} = {l0 l1 l2 ………. ln}T (34a)

{f} = {f0 f1 f2 ………. fn}T (34b)

N is a square matrix of size (n+1), which is defined as given below:

0 0 0 0 0 0 . .
5 8 1 0 0 0 . .

4 16 4 0 0 0 . .
 
4 16 9 8 1 0 . .
4 16 8 16 4 0 . .
[N] =   (35)
4 16 8 16 9 8 . .
. . . . . . . .
 
 . . . . . . . .

If the function l like f is approximated by a series of parabolas, the second integral m of the

function f given by:

m= (36)

Equation 30 can be approximated by:

{m} = [N]2{f} (37)

From the above equations given by Brown and Trahir [12], we conclude that for a function

F, the integrals are:

 31

′′′ = (38)

′′ = (39)

′= (40)

Fi = (41)

This set of equations can approximately be represented as:

′′′ = (42)

′′ = (43)

′ = (44)

{F} = (45)

Now obtain an equation for deflection and their derivatives of order three or less in terms of

integrals involving fourth order derivatives, and determine the constant of integration by

means of the available boundary condition:

v= " + A1z +A2 (46)

Invoke Boundary condition:

= 0; A2 = 0

=0

A1 = " (47)

v= " " (48)

Now using values of Equation 43, we can rewrite the Equation 48 as:

" "
= (49)
 32

" (50)

Where:

(51a)

[ {Nn}T {Nn}T {Nn}T …… {Nn}T ]T (51b)

Euqation 51b defines the matrix [ ] but matrix also can be defined as last

row of [N2] matrix. In Equation 50, {v”} is the curvature Φ vector which can be found out

using the nonlinear moment curvature relation of two beams. Using excel curve fitting

program Equation 27 and 28 are generated which are curvature, ɸ, equations with respect to

moment, M. Because applied load condition is two-point load system moment, M can be

computed using Equation 12a and 12b respectively for each node. In this present study, n =

16 was selected.

2.5.1 Finite Integral Algorithm

A finite-integral algorithm for genrating non-linear load-deflection relation using

Equation 49 is developed and presented herein. The steps are as follows:

1. Define total number of segments of the beam.

2. By using Equation 24 and 25 find out the values for each node and compute the [N] matrix.

3. With [N] matrix from Step 2, compute [N2] matrix.

4. Using the [N2] matrix from Step 3, solve for [ ] matrix using Equation 51b.

5. Using [N2] from Step 3, and [ ] from Step 4, compute matrix using Equation 51a.

6. Find the values of moment at each node using Equation 12a and 12b respectively.

7. With moments from Step 6, compute the curvature vector " by using Equation 27a, 27b

and 28a, 28b for both beams respectively.

 33

8. With matrix from Step 5, and " from Step 7, compute Equation 50 in order to find

defelection at each node.

9. Repeat Steps 6 to 9 until the collapse load condition is reached.

The above iterative algorithm is programmed to compute the load-deflection relationships for

both the beams. A complete listing of computer programming based on the above algorithm

is given in Appendix A.

2.5.2 Numerical Results

This section provides the load-deflection curves for both reinforced concrete beams

with and without hollow spheres based on the algorithm provided in section 2.5.1. Figure 19

and 20 represents the load-deflection curve obtain theoritiaclly for solid reinforced concrete

beam and reinforced concrete beam with hollow sphere respectively.

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Mid-span deflection, v (in.)

Figure 19. Load deflection curve for solid R.C beam

 34

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Mid-span deflection, v (in.)

Figure 20. Load deflection for R.C beam with hollow sphere

2.6 Analysis based on Newark’s method

Newmark's [16] numerical procedure is approximate method, but eventually, it

leads to exact moment and deflection values when loading diagrams are made up of segments

that are bounded by a straight line or by an arc of parabolas. By taking more arbitrary segments

towards the beam we can get better and more accurate results for moment and deflection

values.

2.6.1 Computation of Moments in beam

A fundamental part of these procedures depend on the rapid and systematic

calculations of shear and moment in beam subjected to a series of concentrated loads.

Essentially, the process is to compute shears from one end of the beam to other by adding or

subtracting the successive loads, then to compute the moments by adding or subtracting the
 35

successive shears, multiplied by the length of the beam over which the shear acts. A definite

sign convention is adopted in this system, in which moment will be considered as positive

when it produces compression in the top fiber beam. When resultant force to the left of a

section is upwards, shear will consider as positive. Loads will consider as positive when loads

act upward. The procedure is simplified by omitting the length of the segment as a common

factor so the multiplication of the shear by the distance between loads until the end of the

computation. Then the moment will be computed as a numerical quantity, multiply all by a

common factor, which is the factor for the loads multiplied by the distance between loads.

2.6.2 Computation of Deflection Value

For finding deflection value at each node first we need to find out the value of

curvature at each node which can be found out using Equation 26 and 27 solid reinforced

concrete beam and reinforced concrete beam with hollow spheres respectively. This curvature

values are used to find conjugated loads. Triangular loading section which is shown in Figure

21 is used to find conducted loading. Expression for conjugated loading [16] is given as:

R = R1 + R2 (52)

R12 = (ɸi-1 + 4 ɸi + ɸi+1) (53)

In this equation, ɸi-1, 4 ɸi, and ɸi+1 are successive curvature values and h is length

of each segment. Using same iterative process which we use to find out the moments,

deflections are found out from conjugated loading.

 36

ɸi‐1 ɸi ɸi+1

R1 R2

R12 = R1 + R2 = (h/6) (ɸi-1 + 4 ɸi + ɸi+1)

Figure 21. Equation for continuous polygonal curve [16]

2.6.3 Algorithm based on Newmark method

An algorithm for solving non-linear load deflection case using Newmark’s method

is devloped and presented herein. The solution procedure is given below:

1. Define the total number of segments and total nodes on beam.

2. Define the loading condition on particular node point. And mark row 2 as loading case.

3. Now start with assuming any random number for shear at first node on left side of beam

and by adding that shear value from column 1 with sucssesive load value on column 2 you

can find shear value for column 2. Then repate this step untill you find shear value for last

column.

4. Consider moment at first node equal to zero and repeate the same process which is

described in Step 3, until moment value of last node is found out.

5. If needed, then apply linear correction to moment values found out in Step 4; otherwise

consider moment values of Step 4.

 37

6. Using values of moments from Step 5, compute curvature{Φ} using Equation 27a, 27b

and 28a, 28b for respective beam.

7. With curvature values from Step 6, compute conjucated loading using Equation 49.

8. Using conjucated loading from Step 7, repeat procedure which is shown in Step 3 to 5, in

order to find actual deflections at each node.

9. Repeat the Step 2-8, untill the collapse load condition is reached.

The above iterative algorithm is used to compute the load-deflection relationship for both the

beams and entire solution procedure based on this algorithm is provided in Appendix A.

2.6.4 Numerical Results

This section provides the load-deflection curves for both reinforced concrete beams

with and without hollow spheres based on the algorithm provided in section 2.6.3. Figure 22

and 23 represents the load-deflection curve obtained theoritiaclly for solid reinforced concrete

beam and reinforced concrete beam with hollow sphere respectively.

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
Mid-span deflection, v (in.)

Figure 22. Load-deflection for beam with hollow sphere

 38

5
Load, W(kips)

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Mid-span deflection, v (in.)

Figure 23. Load-deflection for solid beam

2.7 Theoretical Study of Large Beams

In this research, theoretical study has been conducted for full scale beams having

different span and having different diameter of spheres. A total of 6 beams are examined in

which 3 beams are solid reinforced concrete beams and the other 3 beams are reinforced

concrete beams having hollow spheres. In this study, beams are designated as Beam 3, 4, 5,

6, 7, and 8. From that Beam 3, 5, and 7 are solid reinforced concrete beam and Beam 4, 6, and

8 are reinforced concrete beam having hollow spheres. Span length of these beams are 15 ft.,

21 ft., and 27ft and hollow spheres which are used for these beams have a diameter of 6 inches,

8 inches and 11 inches respectively. Cross section of the reinforced concrete beams with

hollow sphere are shown in Figure 24, 25 and 26. Solid beams also have the same cross

sections as shown in Figure 24, 25 and 26, just excluding the hollow sphere from it.
 39

To determine the load-deflection relations, first non-linear moment-curvature

relationships are developed for all beams using the same iterative algorithm which is, shown

in section 2.2.1 and 2.2.2. Then by coupling that moment-curvature relation with the central

finite-difference algorithm provided in section 2.4.3 we are able to predict the load-deflection

relation for these beams. In order to compare the load carrying capacity of these beams, ratio

of cracking load, and collapse load to the self-weight of beams are calculated. The ratio of the

beam cracking load, Wcr, to the self-weight wself of beam, can be defined as:

ηcr = (54)

Similarly, the ratio of the beam collapse load Wc to its self-weight wself can be expressed as:

ηc = (55)

Equation 54 and 55 are used to find the ratio ηcr, ηc for each beam.

3 in.

Figure 24. Cross section of Beam

 40

4 in.

Figure 25. Cross section of Beam

Figure 26. Cross section of Beam 8

 41

2.8 Numerical Results

This section provides tabulated numerical results as well as the graphical

representation of load-deflection relationship for all beams. Figure 27 through 32 represent

the load-deflection curve for all beams. Table 2 shows the theoretical output of cracking and

collapse load indices. Figure 33 and 34 represents the comparison of ηcr and ηc with respect

to the length of beams.

 42

45

40

35

30

25
Load, W (kips)

20

15

10

0
0 0.5 1 1.5 2 2.5

Mid-span deflection, v (in.)

Figure 27. Load-deflection curve for Beam 3

 43

45

40

35

30

25
Load, W (kips)

20

15

10

0
0 0.5 1 1.5 2 2.5 3

Mid-span deflection, v (in.)

Figure 28. Load-deflection curve Beam 4

 44

60

50

40
Load, W (kips)

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5

Mid-span deflection, v (in.)

Figure 29. Load-deflection curve for Beam 5

 45

60

50

40
Load, W (kips)

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4

Mid-span deflection, v (in.)

Figure 30. Load-Deflection curve for Beam 6

 46

90

80

70

60
Load, W (kips)

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Mid span deflection, v (in.)

Figure 31. Load-Deflection curve for Beam 7

 47

90

80

70

60
Load, W (kips)

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Mid span deflection, v (in.)

Figure 32. Load-Deflection curve for Beam 8

 48

Table 2. Cracking and collapse loads indices

BeamBeam Wcr Wc wself Span

type No. (kips) (kips) (kips) ηcr ηc (ft.)
3 3.00 40.00 2.35 1.28 17.08 15
Solid 5 4.52 51.20 5.25 0.86 9.76 21
7 6.74 77.00 10.8 0.62 7.13 27
With 4 2.88 38.98 1.92 1.50 20.3 15
hollow 6 4.44 50.00 4.14 1.08 12.1 21
spheres 8 6.48 75.00 7.80 0.83 9.62 27

1.6

For solid R.C

1.4 beam
Cracking load index, ηcr

For R.C beam

1.2
with Hollow
sphere

0.8

0.6

0.4
10 12 14 16 18 20 22 24 26 28
Length (ft.)

Figure 33. Cracking load index versus beam length

 49

23

21

For solid R.C beam

19
Collaspe load index, ηc

17 For R.C beam with

hollow spheres

15

13

11

5
14 16 18 20 22 24 26 28

Length (ft.)

Figure 34. Collapse load index versus beam length

 50

CHAPTER 3

EXPERIMENTAL INVESTIGATION

The outcome of an experimental study is documented in this chapter. Steel stress-

strain relationship, concrete compressive strength, and beam load-strain and load-deflection

curves are presented in this chapter.

3.1 Material Properties

3.1.1 Stress-Strain Relation for Steel

This section provides tabulated as well as graphical results which are obtained by

performing tensile test on steel rebar. Details of test specimens, test apparatus and test result

are given in the following section.

3.1.1.1 Test Specimen and Specimen Cross Section Properties

Total 3 different specimens are used to perform tensile tests. Cross-sectional

properties of these specimens are shown in Table 3. Figure 35 shows the details of test

equipment and location of rebar in the equipment.

Table 3. Cross-Sectional properties of specimen.

Cross-sectional
Specimen No Type of Rebar Length (in.)
Area (in2)

1 No.2 12 0.05

2 No.3 15 0.11

3 No.3 10 0.11
 51

Figure 35. Tensile test equipment.

3.1.1.2 Experimental Results

Experimental results for sample specimen are presented herein. Load is applied at

specific incremental rate and respective strain values are measured. After collecting values till

breaking point, graph for stress-strain relation are generated. Figure 36 shows the graphical

representation of stress-strain relation for specimen.

 52

120.0

100.0

80.0
Stress (ksi)

60.0

40.0

20.0

0.0
0.00045
0.00088
0.00121
0.00149
0.00174
0.00195
0.00217
0.00238
0.00255
0.00260
0.00265
0.00286
0.00291
0.00215
0.00311
0.00351
0.00384
0.00427
0.00486
0.00577
0.00671
0.00777
0.00964
0.01073
0.01201
Strain (in. /in.)

Figure 36. Stress-strain curve for sample specimen

3.1.2 Compression Test of Concrete

This section gives details about the results which are obtained from performing

compression test on the concrete cylinders. A total of 3 cylinders are casted using same

concrete mix which is used to cast concrete beams. Each cylinder is having a height of 8 in.

and diameter of 4 in. Equipment which is used to perform compression test is shown in Figure

37. Table 4 provides the experimental results which are obtained from the test.

Table 4. Compression Test data (fc’)

Specimen Number Compression strength (fc’)

Specimen 1 3993 Psi

Specimen 2 3983 Psi

Specimen 3 3997 Psi

 53

Figure 37. Compression Test

3.2 Test Procedure for Beams

For testing concrete beams, two point load system which is shown in Figure 38, is

used. For setting up the loading system, concrete beams are first placed on simply supported

ends. To apply load on the beam pneumatic pump and hydraulic jack are used. To transfer the

point load into two point load system, steel I-beam and two metal blocks are used. To measure

the increments of applied load, a load cell unit is attached with the hydraulic jack. Dial gages

are setup to measure the deflection of beams at different locations. Dial gage D.G. 1, D.G. 3

measure the deflection values for nodes which are 8 in. away from the mid-span. Dial gage

D.G. 2 measures the deflection value at the mid-span. As load values increase, strain starts to

develop in concrete. So, to measure the amount of strain in concrete, uniaxial strain gages are
 54

used. In this test 9 strain gages are mounted on three different surfaces of the beam. Location

of each strain gage is shown the Figure 39.

After assembling the entire set up, mounted strain gages are connected in strain

gage box which records the strain for each increment of load. Once the dial gages are set and

strain gages are connected, calibration has been made to set the reading of the dial, and strain

gage to zero. The test is now ready to commence.

At first, certain load is applied so that the top of the metal block came in contact

with the hydraulic jack. It is now required to ensure that the reading of dial gages do not

change. Change in the reading indicates that setup is not done properly. Now load increments

are applied gradually using the hydraulic jack. The dial and strain gauges readings are

recorded. Again the load increments are applied and readings are measured. To obtain better

graphical representation load increments are limited to 0.2 kips. This procedure is repeated

till the collapse condition occur. This testing procedure is used for testing all concrete beams.

Figure 38. Test Setup for Two-point Load system

 55

(a) Top view

(b) Bottom view

(c) Elevation view

Figure 39. Strain gauge location on (a) top surface (b) bottom surface (c) elevation surface
 56

3.3 Experimental Results

This section gives a brief idea about the experimental results which are obtained

by performing the flexural test on the beams.

3.3.1 Flexural Test to Determine the Peak Load Capacity

The main purpose of this test was to determine the cracking load, and peak load of

both beams under such kind of loading. A test is conducted on a total of 4 experimental beams.

These beams are designated as Beam A, Beam B, Beam C, and Beam D. From that, Beam A

and Beam C are solid doubly reinforced concrete beam, Beam B and Beam D, are doubly

reinforced concrete beam with the hollow spheres. For this test, strain, and dial gages were

not used. After conducting this test we are able to predict the cracking and collapse load

pattern of these beams. Accumulated experimental results are shown in Table 5.

Table 5. Cracking load and peak load results

Number of beam Cracking load (kip) Collapse load (kip)

Beam A 1.55 5.90

Beam B 1.45 5.45

Beam C 1.50 5.85

Beam D 1.45 5.50

3.3.2 Flexural test on Beam 1

After determining the peak and collapse load capacity of these beams, another test

is conducted using the solid doubly reinforced concrete beam. The solid beam is designated

as Beam 1. Beam 1 is tested with simply supported end condition. Dial gages D.G. 1 and D.G.

3 were setup to measure side point’s deflections and D.G. 2 was setup to measure mid-span
 57

deflection as shown in Figure 38. Total 9 strain gages were mounted on Beam 1 at the

locations shown in Figure 39 to measure the strain amount. Now using the same test

procedure, which is described in section 3.2, is conducted. Data is accumulated by applying

specific load increments to beam up to the collapse condition.

Table 19 given in Appendix C presents the accumulated experimental results of

Beam 1 and these results are used to compute the graphical representation of load-strain and

load-deflection curves. Figure 45 shows the combined results from all 9 strain gages. Figure

42 presents the load-strain curves for strain gages S.G. 1, 3, and 4. Figure 40 presents the load-

strain curves for strain gages S.G. 2, 5, and 6. Figure 44 shows the load-strain curves for strain

gages S.G. 7, 8, and 9. Load-deflection relationships are shown in Figure 46.

While conducting the test, small cracks start to develop from the bottom part of the

beam and with the increase in load amount, cracks starts to stretch up which is shown in Figure

40 and 41.
 58

Figure 40. Cracking pattern of Beam 1

Figure 41. Cracking pattern in bottom surface of beam 1 after testing.

 59

Load, W (kips)
4
Strain gauge 3

Strain gauge 1 3

Strain gauge 4
2

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0

Strain, Є × 10-6 (in./in.)

Figure 42. Load-strain curves for Strain Gauge 1, 3, and 4

 60

Load, W (kips)
4

Strain gauge 2
3

Strain gauge 5

2
Strain gauge 6

0
‐400 ‐350 ‐300 ‐250 ‐200 ‐150 ‐100 ‐50 0

Strain, Є × 10-6 (in./in.)

Figure 43. Load-strain curves for Strain Gauge 2, 5, and 6

 61

5
Load, W (Kips)

Strain gauge 7
3
Strain gauge 8

Strain gauge 9
2

0
0 50 100 150 200 250 300 350 400

Strain, Є × 10-6 (in./in.)

Figure 44. Load- strain curves for Strain Gauge 7, 8, and 9

 62

Load, W (Kips)
4

Strain gauge 3
3
Strain gauge 2
Strain gauge 1
Strain gauge 4
Srain gauge 5
2
Strain gauge 6
Strain gauge 7
Strain gauge 8
Strain gauge 9
1

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0 500 1000

Strain, Є × 10-6 (in./in.)

Figure 45. Combine Strain Gauge curves

 63

6.5

5.5

4.5
Load, W (Kip)

3.5
At mid-span
3

2.5
At 8 in. from center

1.5

0.5

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35

Deflection, v (in.)

Figure 46. Experimental load-deflection curves

3.3.3 Flexural Test on Beam 2

Beam 2 was a doubly reinforced concrete beam with hollow sphere. Beam 2 is also

tested with simply supported boundary conditions. To conduct test on Beam 2, the same setup,

which is described in section 3.3.2, is used and test procedure is also the same as given in

section 3.2. In order to compare the results of both, test load increments are kept the same

which is 0.2 kips. Experimental data were accumulated by applying specific load increments
 64

to beam up to the collapse condition. Tabulated experimental results which are accumulated

from dial and strain gages are shown in Table 20 in Appendix C.

Experimental results are shown in the form of load-deflection and load-strain

relationships. Figure 49 shows the graphical representation of load-strain curve for strain

gages S.G. 1, 2, and 3. Figure 50 and 51 shows the graphical representation of load-strain

curve for S.G. 4, 5, 6 and S.G. 7, 8, and 9 respectively. Load-deflection relationships are

shown in Figure 52. Combined load-strain data for all 9 strain gages are shown in Figure 53.

With increasing the load values small cracks starts to develop in the bottom of the beam and

as load values increases the cracks are getting thicker and thicker. Location of cracks are

clearly shown in Figure 48 with red lines.

Figure 47. Testing setup for Beam 2

 65

Figure 48. Cracking pattern in bottom surface of Beam 2 after test

Load, W (kips)
4

Strian gauge 1
3

Strain gauge 3
2

Strain gauge 4
1

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0

Strain, Є × 10-6 (in./in.)

Figure 49. Load-strain curves for Strain Gauge 1, 2, and 3

 66

Load, W (kips)
4

3
Strain gauge 2
2
Strain gauge 5
1
Strain gauge 6

0
‐200 ‐180 ‐160 ‐140 ‐120 ‐100 ‐80 ‐60 ‐40 ‐20 0

Strain, Є × 10-6 (in./in.)

Figure 50. Load-strain curves for Strain Gauge 4, 5, and 6

5
Load, W (Kips)

Strain gauge 7
3
Strain gauge 8
2
Strain gauge 9

0
0 50 100 150 200 250 300 350 400 450

Strain, Є × 10-6 (in./in.)

Figure 51. Load-strain curve for Strain Gauge 7, 8, and 9

 67

4
Load, W (kips)

At mid-span
3

At 8 in. from
center
2

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Deflection, v (in.)

Figure 52. Experimental load-deflection curve

 68

Strain gauge 1

Strain gauge 3

Load, W (Kips)
3
Strain gauge 2

Strain gauge 4

Strain gauge 5

Strain gauge 6 2

Strain gauge 7

Strain gauge 8

Strain gauge 9 1

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0 500 1000
Strain, Є × ‐6
10 (in./in.)

Figure 53. Combine load-strain curves

 69

Chapter 4

COMPARISION OF RESULTS

4.1 Comparison of Moment Curvature Relation

The comparison of moment-curvature relationships of both cross sections with and

without hollow sphere are discussed in this section. Table 6 compares the theoretical values

of moment and curvature of both cross sections used in this research. Figure 54 presents the

graphical comparison of the moment-curvature values of both cross sections.

Table 6. Comparison of bending moments and curvature

Bending
Cross Curvature, ɸ
Condition moment, M
Sections (rad. /in.) × 10-5
(kip-in.)
Cracking 17.07 4.30
Solid
Ultimate 66.00 337

Cracking 16.78 4.34

With hollow

sphere Ultimate 64.80 315.7

4.2 Comparison of Load-Deflection Relation

Experimental and theoretical comparison of load-deflection relationships are

discussed in this section. Figure 55 and 56 compares the theoretical and experimental load-

deflection relationship for both the beams. Figure 57 and 58 compares the theoretical versus

experimental results for solid reinforced concrete beam and reinforced concrete beam with

hollow spheres respectively. From the figures we can see that load-deflection values of
 70

theoretical and experimental investigations are in good agreement to each other. Figure 59,

60, and 61 shows the difference in theoretical results of 15 feet, 21 feet, and 27 beams

respectively.

4.3 Experiments versus Theory

The comparison of load-deflection and load-strain relationships from experiment

and theory are discussed in this section. Table 7 compares the peak load, deflection, and strain

data obtained from experimental and theoretical study for both beams used in this research.

Figure 57 and 58 presents the comparison of the load-deflection relationships of Beam 1 and

Beam 2 respectively. As seen from the figures, the peak deflections from theory and

experiment are in good agreement. Figure 62 and 63 present the comparison of load-strain

relationships of Beam 1 and Beam 2. Based on the results in Table 7, it can be seen that there

is a marginal amount of difference in experimental and theoretical results of both beams.

Figure 64 and 65 present the comparison of experimental results versus the theoretical

predicted load-deflection relationships of both beams with and without hollow spheres

respectively.

Table 7. Comparison of experimental and theoretical results

Beam type Experimental Theoretical

Peak load (kips) 5.8 5.5

Solid beam
Deflection, VL/2 (in.) 0.30 0.26
(Beam 1)
Peak Strain, Єc 0.00305 0.003

Peak load 5.4 5.5

Beam with hollow
Deflection, VL/2 (in.) 0.35 0.41
sphere (Beam 2)
Peak Strain, Єc 0.0029 0.003
 71

70
65
60
Bending Moment, M (Kip-in)

55
50
45
40
For solid cross
35 section
30
25 For cross section
20 with hollow sphere
15
10
5
0
0 0.0005 0.001 0.0015 0.002 0.0025 0.003 0.0035 0.004

Curvature , Φ (radi. /in.)

Figure 54. Comparison of moment-curvature curves

6.00

5.00

4.00
Load, W (Kips)

3.00 For beam with

hollow spheres
2.00 For solid beam

1.00

0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Mid‐span deflection, v (In.)

Figure 55. Comparison of theoretical load-deflection curves

 72

5
Load, W (kips)

For Solid
beam

3
For beam with
hollow spheres

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

Deflection, v (in.)

Figure 56. Comparison of experimental load-deflection curves

 73

6.00

5.50

5.00

4.50

4.00
Load, W (kips)

3.50

Theortical values
3.00

2.50 Experimental
values

2.00

1.50

1.00

0.50

0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Mid-span deflection, v (in.)

Figure 57. Comparison of theoretical and experimental load-deflection curves

 74

6.00

5.50

5.00

4.50

4.00
Load, W (kips)

3.50

3.00 Theoritical
values
2.50
Experimental
values
2.00

1.50

1.00

0.50

0.00
0.00 0.03 0.05 0.08 0.10 0.13 0.15 0.18 0.20 0.23 0.25 0.28 0.30 0.33 0.35 0.38

Mid‐span deflection, v (in.)

Figure 58. Comparison of theoretical and experimental load-deflection curves

 75

45

40

35

30

25
Load, W (kips)

20

For hollow sphere beam

15

For solid beam

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4

Mid-span deflection, v (in.)

Figure 59. Load-Deflection curves for 15 feet beam

 76

60

50

40
Load, W (kips)

30

For beam with hollow Spheres

20
For solid beam

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4

Mid-span deflection, v (in.)

Figure 60. Load-Deflection curves for 21 feet beam

 77

90

80

70

60
Load, W (kips)

50

40 For beam with

hollow spheres

30
For solid beam

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Mid span deflection, v (in.)

Figure 61. Load-Deflection curves for 27 feet beam

 78

Load, W (kips)
5

Experimental values 3

2
Theoritical values

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0

Strain, Є × 10-6 (in./in.)

Figure 62. Theoritcal versus experimental load-strain values for Beam 1

Load, W (kips)
4

Theoritical values 3

Experimental values
2

0
‐3500 ‐3000 ‐2500 ‐2000 ‐1500 ‐1000 ‐500 0

Strain, Є × 10-6 (in./in.)

Figure 63. Theoritical versus experimental load-strain curve for Beam 2

 79

6.00
5.75
5.50
5.25
5.00
4.75
4.50
4.25
4.00
3.75 FDM for solid beam
3.50
Load, W (Kips)

3.25 Newmark's for solid

3.00
beam
2.75
2.50
FIM for solid beam
2.25
2.00 Experimental values
1.75
1.50
1.25
1.00
0.75
0.50
0.25
0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Mid‐span deflection, v (In.)

Figure 64. Comparison of experimental results to various predicted

load-deflection relations for solid beam

 80

6.00
5.75
5.50
5.25
5.00
4.75
4.50
4.25
4.00
3.75 FDM for beam with
hollow spheres
Load, W (Kips)

3.50
3.25 Newmark's for beam
3.00 with hollow spheres
2.75
FIM for beam with
2.50
hollow spheres
2.25
2.00 Experimental values
1.75
1.50
1.25
1.00
0.75
0.50
0.25
0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40

Mid‐span deflection, v (In.)

Figure 65. Comparison of experimental results to various predicted

load-deflection relations for beam with hollow spheres

4.4 Discussion

Based on the experimental and theoretical results it can be seen that by using

hollow spheres in the reinforced concrete beam, reduced self-weight is achieved without

affecting the strength of the beams. Results obtained from theoretical study of large span

beams shows that cracking and collapse load index of beam with hollow spheres are way
 81

higher than the solid beams. Also, experimental results in Table 8 present the same thing for

and . The study shows that the beam with hollow spheres has a higher amount of

deflection values than the solid beam. Thus, it is clear that hollow spheres plays an important

role in reducing the self-weight of beams without affecting their strengths while slightly

increasing the deflections by a small amount.

Table 8. Summary of results

Solid Beam with

beam hollow spheres

Weight (lbs.) 112 100

Experimental collapse load index, 51.78 55.00

Experimental cracking load index, 12.70 14.00

Ultimate moments (kip-in.) 66.00 64.30

Ultimate curvature (rad. /in.) 0.00337 0.003157

Elastic stiffness, k for theoretical load- 83.89 59.17

deflection curve

Elastic stiffness, k for practical load-deflection 32.69 27.83

curve
 82

Chapter 5

Conclusion and Future Research

5.1 Conclusion

Based on the experimental and theoretical results, the following conclusions are

drawn:

1. Theoretical load-deflection relationships based on the finite-difference, finite integral, and

Newmark’s methods are in excellent agreement with each other as well as with the

experimental results.

2. For the scale down beam tests, 12 pounds of concrete is replaced by 0.30 pounds of plastic

spheres.

3. The use of hollow spheres in the reinforced concrete beam tested resulted in reduction of

beam self-weight by 12 percent.

4. The theoretical study of full-scale beams showed that the use of hollow spheres gave a

nearly 22 percent of reduction in self-weight.

5. The peak load values from the experimental study are in good agreement with those from

the analysis.

6. The load-strain relations from the experimental study are in good agreement with those

computed theoretically.

7. The test beam with hollow spheres has a 10 and 6 percent chance of higher values of

cracking and collapse load indices, respectively, as compared to the solid beam.

8. The theoretical study of full-scale beams shows that the beams with hollow spheres has a

33 percent chance of higher values of cracking and collapse load indices, as compared to

solid beam.
 83

5.2 Future Research

Research needs to be conducted on steel reinforced concrete indeterminate beams with

hollow plastic spheres. Additional research also needs to be conducted on the effective use of

hollow plastic spheres for various types of concrete structures.

 84

References

1. Churakov, A., “Biaxial Hollow Slab with Innovative Types of Voids,” Construction of
Unique Building and structure, Vol. 6(21), 2014, pp. 70-88.
2. Ibrahim, A., Ali, K. N., and Salman, W. D., "Flexural Capacities of Reinforced Concrete
Two-way Bubble Deck Slabs of Plastic Spherical Voids,” Diyala Journal of Engineering
Sciences, Vol. 06, June, 2013, pp. 9-20.
3. Calin, S., Gintu, R., and Dascalu, G., “Summary of Tests and Studies done on the Bubble
Deck System,” The Buletinul Institutului Politehnic din Ia i, LV
4. Teja, P. P., and Kumar, V. P., “Structural Behavior of Bubble Deck Slab,” JISBN: IEEE,
Vol. 81, ISBN: 978-81-909042-2-3, March, 2012, pp. 383- 388.
5. Anusha, S., Mounika, C. H., and Purnachandra, “Study on Fire Resistance of Bubble Deck
Slab,” JISBN:IEEE, , Vol. 81, ISBN: 978-81-909042-2-3, 2010.
6. Lai, T., “Structural Behavior of Bubble Deck Slab and Their Application to Lightweight
Bridge Decks,” Master’s Thesis, Massachusetts Institute of Technology, 2009.
7. Pfeffer, K., and Schnellenbach, M., "Punching behavior of biaxial hollow slabs," Cement
& Concrete Composites, Vol. 24, Issue 6, December, 2011, pp. 551-556.
8. Mann, K., “Bubble Deck Voided Flat Slab Solutions,” Bubble Deck UK Technical Manual,
Vol. 6, June, 2006.
9. Chung, J. H., Choi, H. K., and Lee, S. C, “Shear Capacity of Biaxial Hollow Slab with
Donut Type Hollow Sphere,” Procedia Engineering, Vol. 14, 2011, Pp. 2219 -2222.
10. Lin, T. Y., and Burns, N. H., Design of Prestressed Concrete Structures, John Wiley &
Sons, Incorporated, New York, 2010.
11. ACI Committee 318. Building Code Requirements for Structural Concrete: (ACI 318-95);
and Commentary (ACI 318R-95). Farmington Hills, MI, American Concrete Institute,
1995.
12. Zhao, Y., “Thermo-Elasto-Plastic Behavior of Biaxially Loaded Steel Beam-
Columns Including those from World Trade Center”, Ph.D. Dissertation, Old Dominion
University, 2013.
13. Usami, T., and Galambos, T. V., “Eccentrically Loaded Single Angle Columns”,
Publications, International Association for Bridge and Structural Engineering, Vol. 31-II,
Zurich, 1971.
 85

14. Mamadou, K., “Inelastic Behavior and Strength of Steel Beam-Columns with Applied
Torsion”, Ph.D. Dissertation, Old Dominion University, 2015.
15. Ajay, J., “Structural Behavior of Bubble Deck Slab”, Master’s report, Department of Civil
Engineering, St. Joseph’s College of Engineering & Technology, Palai, Register No.
202181.
16. Newmark, N. M., “Numerical Procedure for Computing Deflection, Moments and
Buckling Loads,” Journal of Structural Engineering, ASCE, paper No. 2202, May, 1942.
17. Nielsen, M. P., “Punching Shear of Bubble Deck Beam,” Technical report AEC consulting
Engineers Ltd., Technical University Denmark, 2004.
18. Burden, R. L., Faires, J. D., and Reynolds, A. C., Numerical Analysis, Second Edition,
Prindle, Weber and Schmidt, Massachusetts, 1981.
19. Nilson, A. H., Darwin, D., and Dolan, C. W., Design of Concrete Structure, Thirteenth
Edition, McGraw Hill, New York, 2004.
 86

APPENDIX A

Computer Programming for Load-Deflection Relation

This appendix presents a listing of computer programs which are generated using

Microsoft excel to predict the load deflection curve. Programs are based on:

(a) Central finite difference method.

(b) Finite Integral method.

(c) Newmark’s Approach.

(a) Computer programming for Finite Difference Method

This section shows the excel programming for finite difference procedure which is

based on solution algorithm provided in section 2.4.3. The program consists of several parts,

part one is formulation of equations and applying boundary conditions, these equations can

be written as:

–
Node i =1, ɸ = (56)

–
Node i =2, ɸ = (57)

–
Node i =3, ɸ = (58)

–
Node i =4, ɸ = (59)

–
Node i =5, ɸ = (60)

–
Node i =6, ɸ = (61)

–
Node i =7, ɸ = (62)

–
Node i =8, ɸ = (63)

–
Node i =9, ɸ = (63)
 87

Part two consists of formulation of N matrix by which is shown in Table 9, part three is to

find out N inverse matrix by using inverse function in excel.

Table 9. N matrix for Finite-difference Method

Node 2 3 4 5 6 7 8 9
2 ‐2 1 0 0 0 0 0 0
3 1 ‐2 1 0 0 0 0 0
4 0 1 ‐2 1 0 0 0 0
5 0 0 1 ‐2 1 0 0 0
6 0 0 0 1 ‐2 1 0 0
7 0 0 0 0 1 ‐2 1 0
8 0 0 0 0 0 1 ‐2 1
9 0 0 0 0 0 0 2 ‐2

Table 10. N inverse matrix

Node 2 3 4 5 6 7 8 9
2 ‐1 ‐1 ‐1 ‐1 ‐1 ‐1 ‐1 ‐0.5
3 ‐1 ‐2 ‐2 ‐2 ‐2 ‐2 ‐2 ‐1
4 ‐1 ‐2 ‐3 ‐3 ‐3 ‐3 ‐3 ‐1.5
5 ‐1 ‐2 ‐3 ‐4 ‐4 ‐4 ‐4 ‐2
6 ‐1 ‐2 ‐3 ‐4 ‐5 ‐5 ‐5 ‐2.5
7 ‐1 ‐2 ‐3 ‐4 ‐5 ‐6 ‐6 ‐3
8 ‐1 ‐2 ‐3 ‐4 ‐5 ‐6 ‐7 ‐3.5
9 ‐1 ‐2 ‐3 ‐4 ‐5 ‐6 ‐7 ‐4

Part four consists of finding deflection values using curve fitting formulas. Deflection values

are shown in column 1 of Table 11. Now using Equation 18 matrix equation is solved out.

Last column of Table 11 gives the actual values of deflections at each node.
 88

Table 11. Curvature and Deflection values

Curvature Deflection
2.4046E-06 -0.034372875
4.4738E-05 -0.068736132
0.00024738 -0.102920437
0.00083235 -0.136115223
0.00213324 -0.165980584
0.00213324 -0.187312985
0.00213324 -0.200112426
0.00213324 -0.204378906

(b) Computer program for Finite Integral Method

This section gives the detail of excel programming to find out the load deflection

curve which is based on finite integral method and the solution algorithm which is shown in

section 2.5.1. By following Step 3 from algorithm N matrix is generated which is shown in

Table 12. Using matrix multiplication function in excel N2 matrix is find out. Step 5 is

computation of Nn2 matrix using Equation 35. Table 14 shows the final computed values of

Nn2 matrix. Now using Step 9 N bar matrix is found out which is shown in Table 16.
Table 12. [N] Matrix for Finite Integral method
89
Table 13. N2 Matrx
90
Table 14. [Nn2] Matrix
91
Table 15. Nn2 × z matrix
92
Table 16. Matrix
93
 94

Table 17. Deflection and curvature values

Curvature Deflection
0 0
1.95487E‐05 ‐0.022083641
5.45932E‐05 ‐0.044067692
0.000152461 ‐0.065830558
0.000425775 ‐0.086775765
0.00118905 ‐0.105217078
0.00118905 ‐0.118880121
0.00118905 ‐0.12690621
0.00118905 ‐0.129581573
0.00118905 ‐0.12690621
0.00118905 ‐0.118880121
0.00118905 ‐0.105217078
0.000425775 ‐0.086775765
0.000152461 ‐0.065830558
5.45932E‐05 ‐0.044067692
1.95487E‐05 ‐0.022086266
0 0

(c) Newmark’s method

During this research for computing load deflection on a theoretical basis, a

computer program has been develop using Newmark’s method. By following a step by

step procedure which is described in the algorithm shown in section 2.6.3 we can formulate

the excel program using Newmark’s method. Details of the excel sheet is shown in Table

18.
Table 18. Software formulation of Newmark’s procedure
95
 96

APPENDIX B

Computer Programming for Non-Linear Moment Curvature Relation

This appendix presents a listing of computer programs, which can be used to find

out the moment curvature relation. Detail script of the programming is shown below:

(a) Moment curvature for solid R.C beam

% Material properties of R.C beam

fc=4000; e_cu=0.003;
fy=71000; Ey=28842000; ey=fy/Ey;
Ec=57000*sqrt(fc);

n=Ey/Ec;
b=6; d_gross=6; d=4.8125; dc=1;
As=0.22; r=As/(b*d);
Asc=0.22; rc=Asc/(b*d);

%Computation of b
if fc<=4000
b1=0.85;
elseif 4000<fc<=8000
b1=0.85-0.05*(fc-4000)/1000;
else
b1=0.65;
end
k=sqrt(((r+rc)*n)^2+2*(r+rc*dc/d)*n)-(r+rc)*n;
fsc=((k*d-dc)/(d-k*d))*fy;
T=As*fy; Cs=Asc*fsc; eo = 0.002;
Aci = 6*0.195*i; i = [0,31];
fcs = fc(2(eci/e0)-(eci/e0)^2);
Cc= symsum(fcs,eci,0,31)*Aci;
while
T-Cc-Cs = 0;
end
%Formulation for Moment curvature
M=[]; phi=[];
%Moment and Curvature at cracking
Ig=(b^4)/12; c=d_gross/2; fr=7.5*sqrt(fc);
e_bottom = fr/Ec;
Mcr=fr*Ig/c; phi_cr = e_bottom/c;
M=[M;Mcr]; phi=[phi; phi_cr];
k=sqrt(((r+rc)*n)^2+2*(r+rc*dc/d)*n)-(r+rc)*n;
fsc=((k*d-dc)/(d-k*d))*fy;
My=As*fy*d*(1-k/3)+Asc*fsc*(k*d/3-dc); phi_y=ey/(d-k*d);
M=[M;My]; phi=[phi;phi_y];
T=As*fy; Cs=Asc*fsc;
a = (T-Cs)/(0.85*fc*b);
 97

Cc=0.85*fc*a*b;
%Moment Curvature at ultimate condition.
ct=0.5*d; c=0;
while abs(c/ct-1)>0.0002;
e_sc=((ct-dc)/ct)*e_cu;
fsc=Ey*e_sc;
Cs=Asc*fsc; Cc=0.85*fc*b*b1*a; T=As*fy;
end
Mu=0.85*fc*b1*a*b*(d-b1*a/2)+Asc*fsc*(d-dc); phi_u=e_cu/(0.85*a);
M=[M;My;Mu]
phi=[phi;phi_y;phi_u]
%Moment curvature relation
plot(phi,M*0.001,'--s');

(b) Moment curvature for R.C beam having hollow sphere

% Material properties of R.C beam

fc=4000; e_cu=0.003;
fy=71000; Ey=28842000; ey=fy/Ey;
Ec=57000*sqrt(fc);
n=Ey/Ec;
b=6; d_gross=6; d=4.8125; dc=1;
As=0.22; r=As/(b*d);
Asc=0.22; rc=Asc/(b*d);
%Sphere_Dia = 2.5; ds = 2.5;

%Computation of b
if fc<=4000
b1=0.85;
elseif 4000<fc<=8000
b1=0.85-0.05*(fc-4000)/1000;
else
b1=0.65;
end
k=sqrt(((r+rc)*n)^2+2*(r+rc*dc/d)*n)-(r+rc)*n;
fsc=((k*d-dc)/(d-k*d))*fy;
T=As*fy; Cs=Asc*fsc; e_o = 0.002;
Acoi = 6*0.195*i; i= [0,31];
fcs = fc(2(eci/e_o)-(eci/e_o)^2);
Cc= symsum(fcs,eci,0,31)*Acoi;
while
T-Cc-Cs = 0;
end

T=As*fy; Cs=Asc*fsc; eo = 0.002;

X = symsum(1.25-0.105*i, i,0,31);
Yi = sqrt((1.25*1.25)-(X*X));
Aci = (6-2*Yi)*0.105;
fcs = fc(2(eci/e0)-(eci/e0)^2);
Cc= symsum(fcs,eci,0,31)*Aci;
while
 98

T-Cc-Cs = 0;
end
%Formulation for Moment curvature
M=[]; phi=[];
Ig=(b^4)/12; c=d_gross/2; fr=7.5*sqrt(fc);
e_bottom = fr/Ec;
Mcr=fr*Ig/c; phi_cr= e_bottom/c;
M=[M;Mcr]; phi=[phi; phi_cr];
k=sqrt(((r+rc)*n)^2+2*(r+rc*dc/d)*n)-(r+rc)*n;
fsc=((k*d-dc)/(d-k*d))*fy;
My=As*fy*d*(1-k/3)+Asc*fsc*(k*d/3-dc); phi_y=ey/(d-k*d);
M=[M;My]; phi=[phi;phi_y];
T=As*fy; Cs=Asc*fsc;
Cc=0.85*fc*a*b;
while
T-Cc-Cs = 0;
a = (T-Cs)/(0.85*fc*b);
end
ct=0.5*d; c=0;
while abs(c/ct-1)>0.0002;
e_sc=((ct-dc)/ct)*e_cu;
fsc=Ey*e_sc;
Cs=Asc*fsc; Cc=0.85*fc*b*b1*a; T=As*fy;
end
Mu=0.85*fc*b1*a*b*(d-b1*a/2)+Asc*fsc*(d-dc); phi_u=e_cu/(0.85*a);
M=[M;My;Mu]
phi=[phi;phi_y;phi_u]
%Moment curvature relation
plot(phi,M*0.001,'--s');
 99

APPENDIX C

Experimental Data

This section gives the detailed tabulated results of the experimental test which is

performed on the beams. Table 19 and 20 show the details of solid R.C beam and R.C beam

with hollow spheres. Tables includes data which are accumulated from strain and dial gages.

Table 19. Experimental results of Beam 1

 100

Table 20. Experimental results of Beam 2

 101

VITA

Rutvik Rajendrabhai Patel

Department of Civil and Environmental Engineering

Old Dominion University

Norfolk, Virginia 23529

Rutvik R. Patel was born in Ahmedabad, India on March 15, 1994. He received his

Bachelor of Engineering (B.E.) in Civil Engineering from Gujarat Technical University,

Gujarat in 2015. He was then employed as an intern structural engineer for Bhagyoday

Builders Pvt. Ltd., where he worked for 6 months. Then he came to the United States of

America in 2016 to pursue his Master’s degree in Civil Engineering at Old Dominion

University. This thesis titled, “FLEXURAL BEHAVIOUR AND STRENGTH OF

DOUBLY-REINFORCED CONCRETE BEAMS WITH HOLLOW PLASTIC SPHERES”

was completed in April 2018 with Dr. Zia Razzaq as the advisor.

The author’s current address is given below.

Current address: 905, E Piney Branch Dr,

Apt #202, Virginia beach, VA, 23451

Phone: 757-754-8812; Email: [email protected]