CHAPTER 4 MATERIALS AND EXPERIMENTAL PROCEDUREshodhganga.inflibnet.ac.in/bitstream/10603/10080/9/09_chapter 4.pdfCHAPTER 4 MATERIALS AND EXPERIMENTAL PROCEDURE 4.1 GENERAL To accomplish

59

CHAPTER 4

MATERIALS AND EXPERIMENTAL PROCEDURE

4.1 GENERAL

To accomplish the objective of this study, the experimental work

was carried out on nine GFRP reinforced concrete deep beams with and

without web reinforcement. The details of the materials used for casting the

deep beam specimens, fabrication of GFRP reinforcements and testing

procedures are given in the subsequent sections.

4.2 MATERIALS AND ITS PROPERTIES

4.2.1 Concrete

The concrete used for casting was prepared in the testing laboratory

using a portable concrete mixture machine. All the specimens which were

tested were cast by using cement concrete and the cement used was

confirming to the specification of IS 8112 (1989) code. The concrete was

designed to achieve the 28 day compressive strength of 40 N/mm2 (M40

Grade). The concrete mix proportion adopted was 1: 1.02: 1.93 with water

/cement ratio of 0.38. The material proportions per cubic meter of concrete:

1) 1059 kgs of coarse aggregate (maximum size 20mm)

2) 560 kgs of natural river sand (sp.gr =2.53)

3) 548.5 kgs of ordinary port land cement(43 grade)

4) 208 litres of water

60

The coarse aggregate used was crushed granite stones of 20 mm

and less in size with specific gravity of 2.58 and fineness modulus of 6.9. The

locally available river sand which was sieved by passing through 2.36 mm

sieve and retained on 75 m sieve and which had a fineness modulus of 3.1

was used as the fine aggregate. The cement used for preparing the concrete

was tested in the laboratory conforming to IS 8112 (1989) standards with

respect to initial and final setting time, specific gravity, fineness and

compressive strength.

Along with each batch of the concrete prepared for casting the deep

beam specimens, concrete cubes were cast to determine the material

properties and to confirm its strength. Testing of these cubes was done after

7 days and then 28 days after casting. The cubes were de-moulded the next

day of casting and it was cured in water before it is tested. The details of cube

tests results are shown in Table 4.1.

Table 4.1 Concrete cube test results

BEAMS

Compressive

Strength N/mm2

(7Days )

Compressive

Strength

N/mm2 (28 Days )

GFRDB-1 28.7 41.9

GFRDB-2 30.9 43.3

GFRDB-3 28.8 40.8

GFRDB4 34.9 40.1

GFRDB-5 33.4 41.3

GFRDB-6 28.5 42.9

GFRDB-7 29.8 42.5

GFRDB-8 29.5 40.7

GFRDB-9 30.5 41.9

61

4.2.2 About GFRP Constituent Materials

4.2.2.1 Glass Fiber

The glass fiber used for fabricating the GFRP reinforcements is of

E- Glass fiber type of Saint-Gobain Vetrotex with identification number as

RO 99 2400 P 566. The fiber which was available in roving of 2400 normal

liner density (tex) was specifically selected for the purpose of manufacturing

by ‘Manual Fiber-Trusion’. The proprietary sizing system of the selected P

566 type fiber was designed to give a high level of performance with greater

compatibility with Polyester, epoxy, vinyl and phenolic resins. The test

specifications supplied by the fiber manufacturer are as shown in Table 4.2.

Table 4.2 Specification of E-Glass Fiber

PROPERTY VALUE TEST METHOD

Glass content (%) 60-65 BS 3691

Tensile Strength (Mpa) 1700-1800 BS 3691

Tensile Modulus (Gpa) 65-75 BS 3691

Some of the other advantages of using this type of fiber include

fast and complete wet-out of fibers with most resins, good resin wettability by

fibers, excellent spreading on the mandrel which gives a smooth and regular

surface, high mechanical properties, good composite translucency. These

were found to be needed for this type of manufacturing process to be efficient

and hence this type of fiber was adopted.

4.2.2.2 Epoxy resin

Epoxy resin is the extensively used matrix material compared to

other category of resins particularly for reinforcement bars due to their high

62

mechanical property to bond with fibers. Epoxy is a thermosetting polymer

formed from after mixing a peroxide resin with a hardener in a suitable

proportion.

Epoxy resins have high strength, good corrosion resistant and can

very well bond with fibers, high glass transition temperature, superior

electrical properties, high dimensional stability, and low shrinkage. Epoxy

resins which have a wide range of mechanical properties can be used in all

FRP manufacturing processes.

The slow curing epoxy resins used in this work takes a long gel

time to become solid, which is considered to be more advantageous for the

‘Manual Fiber-Trusion’ process as it requires more time for moulding the

uncured impregnated fibers pulled from the Fiber-Trusion die.

Epoxies used for this work was found to have an average tensile

strength of 95 MPa. Their density varied between 1.2 -1.4 Kg/m3. The

percentage of cure shrinking was observed to be 3.1%.

4.2.2.3 Proportion of mix for resin

To provide the correct calculated quantity of fiber in the GFRP bars

and stirrups the fibers were cut to a desired length using the technical

specification provided by the glass fiber manufacturer. From the specification

supplied by the fiber manufacturer, the weight of fiber required per unit

length was calculated and the required amount of fiber to achieve 75% of

fiber volume fraction in the GFRP composite reinforcement material was

calculated. The calculated amount of fiber was then counted and removed

from the rovings to the desired length of the reinforcement. The glass fiber

cut-out from the roving was suitably impregnated in a resin tub containing

25% of resin and hardener mix by volume.

63

The amount of fibers used for fabricating the GFRP bars and

stirrups varied with respect to the required diameter. For a greater diameter of

the bar, the required amount of fiber was higher to maintain the 75% fiber

content in the finished product.

A number of trails were made and tested to finally arrive at a

particular combination of fiber and resin which was finally adopted to

produce bars of 8mm, 10mm and 22mm diameters. The details of calculation

showing the amount fiber -resin mix required for this manufacturing process

has been illustrated in Appendix 1.

4.2.2.4 Properties of GFRP used in the experiment

Tension tests on GFRP bars were conducted to know the ultimate

tensile strength and modulus of elasticity for the bars used for this

experiment. The bars were placed inside the specially made pipe fixtures

fitted at its ends and filled with a mixture of epoxy and sand to avoid slippage.

The results of which prove to be greater than values of the FRP bars

manufactured by conventional pultrusion method. The calculated values of

some of the physical properties from the test results are tabulated in Table 4.3

Table 4.3 Properties of Fabricated GFRP material

Material Average Tensile

Stress (f frp) Mpa

Ultimate tensile

Strain ( )

Modulus of Elasticity

(EGFRP) Mpa

GFRP 780 0.017515 44,530

To obtain good bond strength with concrete, the GFRP bars which

were coated with a layer of coarse and fine sand sieved particles on the rebar

surface one over the other. After providing this type of sand coat a very

effective bond between GFRP bars and concrete was obtained. Pull-out tests

64

were carried out on sixteen rebar specimens and the results proved to be better

in terms of bond strength, when compared to other GFRP bars with different

surface finish.

4.3 MANUAL FIBER-TRUSION

4.3.1 ‘Manual Fiber-Trusion’- a new method of Manufacturing

GFRP Reinforcement

FRP Composites are being manufactured by various techniques and

with emerging new technologies, production is easy and fast. Among the

various methods of manufacturing FRP composites, pultrusion by machine is

one way of manufacturing. The Pultrusion method employs machines to

manufacture FRP composite which make the process expensive. Moreover

they require high initial investment and the process of manufacturing is slow.

In addition to all this, there exist some shortcomings with respect to

the strength criteria of FRP bars manufactured through mechanised

pultrusion. This is due to the use of low strength filler material during the

manufacture of pultruded FRP composites. These filler materials are used for

the smooth production of pultruded sections and also to reduce the gel time

during the pultrusion process. The presence of filler material in FRP

composites is an impediment to achieve higher tensile strength. The filler

material in these FRP bars reduces the volume of fiber and matrix materials

and this in turn reduces the strength of the FRP composite material.

In the present day mechanised pultrusion process, the presence of

filler material cannot be avoided for the reason that it is this filler material

which ensures the smooth running of pultrusion manufacturing process

without interruption. Moreover, the cost of pultrusion machine and its

maintenance is very high which in turn has a bearing on the production cost of

FRP composite materials.

65

To overcome this problem, it was needed to develop a new way of

manufacturing FRP composites. The outcome was the conception of a simple

manufacturing method called “Manual Fiber-Trusion” developed during the

course of this research work. The “Manual Fiber-Trusion” method of

manufacturing FRP bars and stirrups was found to be cost effective with an

advantage of having greater tensile strength. By using this novel method, FRP

bars can be made to any size and shape which is not possible in the machine

pultrusion. The greatest advantage in this Manual Fiber-Trusion method is

that, FRP bars can be bent during and also after the process to any desired

shape, which cannot be achieved by machine pultrusion. A schematic diagram

showing the various stages of ‘Manual Fiber-Trusion’ process is shown in

Figure 4.1.

Figure 4.1 The ‘Manual Fiber-Trusion’ process

Although there are many FRP composite manufacturing companies

in India, most of them mainly manufacture GFRP bars for electrical

applications only. In such GFRP bars, the insulation property is considered

primarily important than the strength of the bar. When tested the strength of

these bars were found to be less as their fiber content was less than 65% and

RESINTUB

GLASS FIBRE CREELS

DIE

MANUAL

FIBER-TRUSION

RE BARROLLERS

DIRECTION OFMANUAL PULL

66

a substantial amount of filler materials was present in it. Such a combination

of low fiber and high matrix materials used for manufacturing these GFRP

bars leads to low strength and thereby reduces their load carrying capacity.

These GFRP bars which were manufactured for electrical applications were

tested earlier, before adopting this ‘Manual Fiber-Trusion’ method of

manufacturing. Thus the option of using the readily available GFRP bars from

the market was not preferred.

To overcome the entire above problem, it was decided to produce

the GFRP reinforcement bars having high strength. After experimenting by

various methods, many trails were carried out to manufacture high strength

bars economically. Some of the trail specimens are as shown in Figure 4.2.

Figure 4.2 Some trail GFRP reinforcement specimens fabricated

During the trail fabrication, the GFRP rebar specimens were

manufactured by altering the composition of individual constituent materials

and also by adopting different techniques of manufacturing.

67

Fabrication of GFRP rebars and stirrups with open moulds of

different sizes and shapes, fabricating stirrups using continuously looped fiber

impregnated in matrix, casting of rebars using ribbed flexible pipes as mould,

were some of the techniques that were tried during initial trails for fabricating

the GFRP rebars and stirrups.

Some of the other procedures adopted were; altering the percentage

volume of the resin and fiber combinations in the composites. Ultimately,

with a considerable effort, a new method suitable for fabricating the GFRP

reinforcement material was found. This new method of fabricating FRP rebars

through manual Fiber pulling resulted in the naming of this method as

“Manual Fiber-Trusion” method by the author.

4.3.2 Arrangement of Parallel Fibers for Fiber-Trusion

Based on the percentage of fiber to be adopted for making the

GFRP bars of required diameter and also based on the weight per unit length

of the fiber supplied by the fiber manufacturer, the required numbers of

strands were removed from the roving creels and looped. The fibers were

looped between two fixed points of required length and after looping they

were removed from the arrangement and kept ready for sinking in a tub of

resin. Looping is mainly done to facilitate proper gripping of fibers during

manual pulling. The arrangement made for looping the fibers is shown in

Figure 4.3. The arrangements of fibers were carefully maintained until the end

of the manufacturing process to ensure the parallel alignment of fibers.

68

Figure 4.3 Looping of glass fibers

4.3.3 “Manual Fiber-Trusion” Process

The fibers which were looped were tied to a steel rod which has a

provision to hook the fiber loop at one end. This steel rod with hook at one

end and with a handle at the other end was used to pull the impregnated fiber

after the wet fibers pass through the die. The looped fibers are then pulled

through the preset die arrangement which is fixed over the table through

which it is passed. This was done to get the fibers molded to the required size

and shape of the die and also to remove the excess resin from the wet fiber

which is recycled into the resin bath. The “Manual Fiber-Trusion”

arrangement used for fabricating the GFRP main bars of 22mm diameter is

shown in Figure 4.4. Each batch of resin prepared has to be utilised before it

starts curing. It took about 30 minutes to complete one cycle of fabrication of

a GFRP element.

69

Figure 4.4 ‘Manual Fiber-Trusion’ for fabricating GFRP main bars

These fibers pulled out from the die which was in a flexible state

were passed into suitable moulds to form bars of the required shape and size.

4.3.4 Moulding of Bent Bars and GFRP Stirrups

The method of fabricating the stirrups was slightly different from

that adopted for GFRP straight bars. The process was the same till the fibers

were pulled out of the die. Then the flexible wet fibers in the cylindrical form

were placed inside a flexible mould. The type of mould adopted for

fabricating the stirrups is shown in Figure 4.5.

70

Figure 4.5 Fibers impregnated in resin and placed inside a

closed mould to make stirrups

The tension of the fibers was maintained till the resin had cured

considerably. The mould was placed exposed to direct sunlight for about 2 to

3 hours after it was molded. Then it was made to dry under shade for about

12 to 15 hours. Figures 4.6 and 4.7 shown below are photographs of trail

stirrup specimens fabricated during the initial stages before the final

fabrication of bars and stirrups for experimentation. Different types of moulds

such as open mold, glass tube mold and P.V.C. flexible mould were tried

during the trail phase of fabrication.

Figure 4.6 View of trail pieces GFRP stirrup specimens

71

Figure 4.7 GFRP stirrups specimens fabricated and used in this work

4.3.5 Post Curing Process

After allowing the moulded GFRP bars and stirrups to be naturally

cured under atmospheric temperature for about 12 -15 hours under dry

condition, the GFRP material is removed from the mould. To complete the

curing process the GFRP bars and stirrups are further post-cured by placing

these GFRP materials inside an electric oven which is maintained at

120o temperature for about 2 hours. This is done to cure some of the uncured

resins inside the core of the GFRP material. The post curing of GFRP bars

and stirrups in a large size electric oven is as shown in Figure 4.8.

After post-curing the specimens, they were given a finishing

treatment such as by cutting and removing the excess resin which had over-

flown the mould during the moulding process and also cutting and removing

72

the ends of the GFRP fiber material present beyond the mould. This finishing

work for the stirrups is shown in Figure 4.9.

Figure 4.8 Post-curing of GFRP bars and stirrups

Figure 4.9 GFRP stirrups finishing work

73

In order to prevent the glass fibers from penetrating into the skin,

handling of the GFRP finished product was carried out using leather gloves.

The post cured bars and stirrups were found to be extremely hard.

4.4 PREPERATION OF BEAM SPECIMENS

4.4.1 Sand Coating

After the finishing work, the GFRP bars and stirrups were coated

with a layer of epoxy over which coarse sand was spread uniformly over the

entire surface. The coarse sand used for this first layer of sand coating was

obtained after sieving the sand through a 0.5 mm sieve. After curing for about

two days, a second layer of fine sand was applied over the existing layer of

coarse sand. The fine sand used for this purpose was obtained after passing

through a 100 micron sieve. The fine sand fills between the gaps of the coarse

sand in order to develop a complete rough surface over the entire GFRP

material. Figure 4.10 shows sand coated GFRP stirrups kept under the

sunlight for surface curing.

Figure 4.10 Curing of sand coated GFRP stirrups

74

4.4.2 Fixing of Strain Gauges

To understand the internal behaviour of GFRP reinforcement in

deep beams, all the deep beam specimens were affixed with electric foil strain

gauges at critical points where the strains were expected to be more and need

to be measured. In case of non-isotropic materials like the GFRP

reinforcement, the material is very strong when the fibres are unidirectionally

oriented. But in the same material, when the fibers are bent, it becomes weak

leading to failure. To effectively understand the response of GFRP

reinforcements, the strain gauges were placed at vital places over the surface

of the GFRP bars and stirrups, where the cracks in the beam were expected to

cause a considerable amount of strain in the GFRP reinforcement. After fixing

the strain gauges over the GFRP reinforcements, the electrical resistances of

the strain gauges were checked to ensure that all the strain gauges have the

required resistance of 120 ohms. This also ensures the working condition of

the strain gauges before concealing them inside the water proofing latex

cover. Checking the resistance of the strain gauges which were fixed over the

GFRP reinforcements is shown in Figure 4.11.

Figure 4.11 Checking of strain gauges fixed over the GFRP bars

75

4.4.3 Binding of GFRP Reinforcement

After fixing the strain gauges over the GFRP reinforcements at

selected locations, the bars and stirrups were tied according to the beams’

reinforcement configurations. Self-locking nylon ties were used to bind the

GFRP bars and stirrups. At every junction, two nylon ties were provided in a

criss-cross fashion to ensure that no slippage of reinforcement takes place. A

GFRP cage tied for beam GFRDB-8 by nylon ties is shown in Figure 4.12.

Figure 4.12 Binding of GFRP reinforcement for beam GFRDB-8

The main GFRP bars and horizontal bars were projected beyond the

point of support for effective anchorage. The anchorage length was calculated

as per the guidelines given in ISIS Design Manual 3(2001). To avoid any

rupture in the GFRP bars with the steel stirrups which were provided beyond

the test area, the ribs of the steel stirrups were removed by properly grinding

at the places of contact with GFRP bars. This also helped to prevent the

scraping of the sand coated over the GFRP bars during extreme forces in the

anchor zone usually during the peak loading condition. The cage

reinforcement for all beams is as shown in Figure 4.13.

76

Figure 4.13 GFRP reinforcement cage for deep beams

77

4.4.4 Casting of Deep Beams Specimens

Three steel moulds of the same dimensions were fabricated for

casting the beam specimens to be tested in this study. The moulds were

properly cleaned and greased for easy de-moulding after casting. The concrete

required for casting was prepared using a concrete mixture machine.

Before pouring concrete, the reinforcement cages were placed

inside the mould with suitable small sized cover blocks. The beams were cast

with the concrete which had the required slump as per the IS code

specifications. The concrete was properly compacted using a needle vibrator.

All the beams were cast to the same dimensions of 160 mm depth, 530 mm

height and 1970 mm overall length. The mould used for casting is shown in

Figure 4.14.

Figure 4.14 Mould used for casting deep beams

The concrete beam specimens were de-moulded the day after they

were cast. Subsequently, they were cured in water for 28 days before being

subjected to testing. Along with each batch of concreting, concrete cubes

78

were also cast to ascertain the 7 day and 28 day strength of concrete. The

beams cast is shown in Figure 4.15

Figure 4.15 Deep beams cast and cured in moisture by using gunny bags

and dried hay stalks

4.5 EXPERIMENTAL SETUP AND TESTING

4.5.1 Test Equipment and Instrumentation

To monitor the change in the strain of the internal GFRP web

reinforcement, the strain gauges were fixed at vital locations of the GFRP

reinforcement. The variation of strain during the experiment was observed

using a Multi-Channel digital strain meter. During testing of the deep beam

specimens, the concrete surface strain varies and this surface strain was

measured using a demountable mechanical gauge (De-Mech gauge) along the

line joining the point of loading and the point of support. The deflection of the

beam was measured using a mechanical dial gauge at the mid-span section.

One externally placed GFRP bar which does not form a part of the

beam’s reinforcement, was kept exposed to atmosphere and was connected to

79

the strain meter for auto-correction of strain due to temperature variation. This

was done for temperature compensation of the strain gauge used inside the

concrete deep beam specimens. The size of the strain gauge foil used in this

testing varied from 20 to 50 mm gauge length. The layout of the strain gauges

fixed to the reinforcement of all the deep beam specimens is shown from

Figure 4.16 (a) to 4.16 (e)

Figure 4.16 (a) Location of strain gauge (SM) in main bar of beam

GFRDB-1

The strain gauges are abbreviated as:

1) SS - Strain Gauge at straight portion of stirrup.

2) SB - Strain Gauge at bent portion of stirrup

3) SM - Strain Gauge at Mid-Span of the Main Bar

4) SHB - Strain Gauge at horizontal web reinforcement (bottom)

5) SHT - Strain Gauge at horizontal web reinforcement (top)

The locations of the strain gauges fixed in Series-II beams were the

same as that of their counterpart beams in Series-I.

80

Figure 4.16 (b) Locations of strain gauges in deep beams GFRDB-2 and GFRDB-3

81

Figure 4.16 (c) Locations of strain gauges in deep beams GFRDB-4 and GFRDB-5

82

Figure 4.16 (d) Locations of strain gauges in deep beams GFRDB-6 & 7

83

Figure 4.16 (e) Locations of strain gauges in deep beams GFRDB-8 and GFRDB-9

84

Figure 4.17 Mechanical strain gauge and crack width microscope

The mechanical strain gauge used for measuring the concrete

surface strains had a least count of 1/40 mm. This strain gauge was used for

measuring the diagonal surface elongation by which the surface strains can be

calculated. The surface elongation occurs due to formation of cracks in the

shear span region of the beam. The type of crack width microscope used

during the experimental work is shown in Figure 4.17

Figure 4.18 Multichannel strain meter

85

The strains in GFRP reinforcement were measured using the above

strain meter as shown in Figure 4.18. This digital instrument was connected

simultaneously all strain gauges fixed to a specimen and the strains were

recorded.

4.5.2 Details of Beam Specimens Tested

The GFRP deep beams were tested under two categories, namely

under Series-I and Series-II. In Series-I nine GFRP deep beams were tested

with a ‘shear span to effective depth ratio’ of 0.72 under the four point

bending condition. In Series-II four deep beams with a ‘shear span to

effective depth’ ratio of 1.08 were tested under three point bending. In total

13 deep beams were tested to study their ultimate shear strengths.

Apart from the above mentioned tests conducted on GFRP deep

beams, some tests on GFRP slender beams were also conducted initially to

check and confirm the use of ‘Manual Fiber-Trusion’ bars in concrete beams

before commencing the testing of heavy deep beams. Twelve beams were cast

and tested in which nine beams were cast using only GFRP main bars as

reinforcement and three beams were cast using only steel main bars

reinforcement. The details of slender beam specimens tested are discussed in

section 4.5.5.

Details of the deep beam specimen regarding dimensions, amount

of reinforcement and grade of concrete are as shown in Table 4.4. The details

of the deep beams which were provided with different web reinforcement

configurations are shown in Figures 4.19 (a) to 4.19 (c).

86

Table 4.4 Details of GFRP deep beam specimens in Series-I and Series-II

Sl.

No

Beam

Effective

Length

le

(mm)

Shear

span

a

(mm)

Effective

Depth

d

(mm)

a/d

ratio

Compressive

Strength

of concrete

f 'c MPa

Percentage Main

Bar

Reinforcement

% main

Ultimate

Tensile

Strength

f frp (kN)

Percentage of

Vertical Web

Reinforcement

% V

Percentage of

Horizontal Web

Reinforcement

% h

SERIES -I

1 GFRDB-1 1050 350 485 0.72 41.9 0.979 780 - -

2 GFRDB-2 1050 350 485 0.72 43.3 0.979 780 0.4 -

3 GFRDB-3 1050 350 485 0.72 40.8 0.979 780 0.8 -

4 GFRDB-4 1050 350 485 0.72 40.1 0.979 780 - 0.4

5 GFRDB-5 1050 350 485 0.72 41.3 0.979 780 - 0.6

6 GFRDB-6 1050 350 485 0.72 42.9 0.979 780 0.4 0.4

7 GFRDB-7 1050 350 485 0.72 42.5 0.979 780 0.4 0.6

8 GFRDB-8 1050 350 485 0.72 40.7 0.979 780 0.8 0.4

9 GFRDB-9 1050 350 485 0.72 43.4 0.979 780 0.8 0.6

SERIES -II

10 GFRDB- 1(a) 1050 525 485 1.08 42.3 0.979 780 - -

11 GFRDB- 3(a) 1050 525 485 1.08 41.9 0.979 780 0.8 -

12 GFRDB- 5(a) 1050 525 485 1.08 42.1 0.979 780 - 0.6

13 GFRDB -9(a) 1050 525 485 1.08 40.9 0.979 780 0.8 0.6

87

Figure 4.19 (a) Cross section of GFRP deep beams tested

88

Figure 4.19 (b) Cross section of GFRP deep beams tested

89

Figure 4.19 (c) Cross section of GFRP deep beams tested

4.5.3 Experimental Set-up for Deep Beams Testing

4.5.3.1 Experimental setup for deep beams in Series-I

The testing of deep beams in this work was carried out using a

1000 kN Universal Testing Machine (UTM) that was available at the

institution for conducting the experiments. The beams which weigh nearly

360 Kgs were lifted and placed over the supports in the base frame of UTM

using a mobile crane.

All the deep beam specimens were designed to be of the maximum

possible dimensions that the UTM can support during testing. Due to capacity

constraints and also due to constraints in increasing the overall depth of the

specimen, the loading and support points of the beam were decided on the

90

basis of the maximum available support span. The details of all the beams

tested are shown in Table 4.4

The deep beams were tested under four point bending condition in

Series-II as shown in the schematic diagram in Figure 4.20.

Figure 4.20 Experimental Set Up for testing deep beams with a

‘Shear span to effective depth’ ratio of 0.72 (Series-I)

The top loading point of the UTM is placed over a spreader beam

and below the spreader beam two steel rollers were placed to transfer the

loads at two points both placed equidistant from the centre line of the beam.

The position of the point load was decided after finalising the ‘shear span to

effective depth ratio’. A shear span of 350 mm was adopted for this series of

testing. Marking of all loading and support positions by marker pen was done

prior to testing. To identify the cracks and to record its propagation, grids

were formed and marked over on the front face of the specimen between the

supports. The effective span of 1050 mm was divided into 14 divisions while

the 530 mm depth was divided into 7 divisions of equal dimension. The

columns were numbered 1 to 14 while the rows were named A to G.

a =350 mmSPREADER BEAM

DEEP BEAM

Le =1050 mm

L =1970 mm

H=530 mm

91

A typical test setup of a GFRP deep beam with a/d ratio of 0.72 is shown in

Figure 4.21

Figure 4.21 Experimental setup for testing deep beams with a /d ratio 0.72

4.5.3.2 Experimental setup for deep beams in Series-II

After conducting experiments with GFRP reinforced concrete deep

beams with a ‘shear span to effective depth’ ratio of 0.72, some more deep

beams were cast and tested under a single point loading, i.e. under three point

bending condition. This was done to investigate the influence of ‘shear span

to effective depth’ ratio in some of the GFRP reinforced deep beams already

tested. Adopting the same dimensions and reinforcement configuration of

some of the beams already tested, some more beams were freshly cast but

tested under a higher ‘shear span to effective depth’ ratio.

92

Figure 4.22 Testing of GFRP reinforced deep beams with a/d =1.08

The beam GFRDB-9 of Series-I performed well under the four

point bending condition. The ultimate capacity of this beam GFRDB-9 was

decided to be investigated under an increased ‘shear span to depth’ ratio. For

this, the beam GFRDB-9(a) which is identical to beam GFRDB-9 was tested

under three point loading condition with a ‘shear span to effective depth’ ratio

of 1.08. To evaluate the individual contributions of GFRP web reinforcements

of the beam GFRDB-9(a) under three point loading, three more deep beams

were tested. A schematic diagram showing the loading conditions of beams

tested in Series-I has been shown in Figure 4.22.

The four deep beams in Series-II which were identical to the beams

GFRDB-1, GFRDB-3, GFRDB-5 and GFRDB-9 were cast and tested in a

way similar to that of Series-I, except for the change of the a/d ratio from 0.72

to 1.08.

4.5.4 Testing of GFRP Reinforced Deep Beams

During testing, all the beams were initially loaded to 5% of the

predicted ultimate load and subsequently the loads were released. This was

a =525 mm

DEEP BEAM

L e =1050 mm

L =1970 mm

H=530 mm

93

done to make sure that the beam was properly seated over the supports before

commencing the experiment.

During the testing, the load applied to the beams was gradually

increased in steps of 10 kN up to the beams’ failure. The applied load (P),

mid-span deflection, strains in GFRP and concrete were observed and

recorded along with the crack width. In addition, the crack propagation was

observed and marked over the beam surface.

4.5.5 Experiments on GFRP Slender Beams

Initially, to demonstrate the performance of the GFRP bars

manufactured by ‘Manual Fiber-Trusion’ the bars were tested as

reinforcement in slender concrete beams under three point loading. The

experimental setup for testing the slender beam specimen is shown in Figure

4.23. The details of beam specimen are shown in Table 4.5. The testing of

GFRP reinforced slender beams is not considered for any comparative study

with the analysis of main experiments on deep beams and is to be treated as a

supplementary work.

Nine slender beams reinforced only with GFRP main

reinforcements without any stirrups were cast using different grades of

concrete. The grades of concrete adopted were M20, M30 and M40. Also, the

percentage of main tension reinforcement was varied for each beam to study

the performance of the beams for under, balanced and over reinforced

conditions. Figure 4.24 shows the failure of a GFRP slender beam tested

under three point loading.

94

Figure 4.23 Experimental setup for testing of slender beams

Figure 4.24 Testing of GFRP reinforced slender beams

95

Table 4.5 Details of GFRP and steel reinforced slender beams

Beam

configuration

Concrete

grade

Effective

lengthWidth Height % frp % steel

GFRB – 1 (U)

M20

700 100 150 0.38 -

GFRB – 2 (B) 700 100 125 0.47 -

GFRB – 3 (O) 700 100 150 0.76 -

GFRB – 4 (U)

M30

700 100 150 0.38 -

GFRB – 5 (B) 700 100 90 0.71 -

GFRB – 6 (O) 700 100 150 0.76 -

GFRB – 7 (U)

M40

700 100 150 0.38 -

GFRB – 8 (B) 700 100 125 0.47 -

GFRB – 9 (O) 700 100 150 0.76 -

SRB – 1 (U)

M40

700 100 150 - 0.38

SRB – 2 (B) 700 100 125 - 0.47

SRB – 3 (O) 700 100 150 - 0.76

The beams were cast according to the grade of concrete as mentioned

in the above Table 4.5. The testing of beams was done using a 100 kN UTM.

Along with the beams, concrete cubes were also cast and tested. A view of all

the tested beams is shown in Figure 4.25.

96

Figure 4.25 GFRP and steel reinforced slender beams after testing

4.5.6 Tensile Test on GFRP Bars

The GFRP bars which were fabricated were subjected to tensile

testing to evaluate their tensile strength. Conducting a tensile test on GFRP

bars cannot be done as easily as testing a steel bar, due to the fact that the

lateral strength of GFRP bars is very low compared to its longitudinal strength

which is a typical property of continuous fiber composite bars.

Moreover, since the surface of the fabricated GFRP bars are smooth

and do not have ribs which are present in steel bars, the tension test of such

bars cannot be done directly. So, the GFRP bars to be tested were first coated

with a layer of epoxy over which sand particles were sprinkled uniformly.

This was done in two stages with different sizes of sand particles to achieve

the desired bond strength.

97

For the purpose of providing better grip to firmly hold the GFRP

bar during tensile test, the ends of the GFRP bar were enclosed inside a

gripping attachment filled with epoxy and sand as shown in Figure 4.26

Figure 4.26 Failure of GFRP bar specimen in tension with end gripper

The end grippers were fabricated by using high strength steel pipes

which was machined to form standard threads on the inside and outside

surfaces of the pipe. This arrangement was found to be effective in protecting

the GFRP bars from lateral crushing. The threads formed on the inside of the

pipe played a crucial role in holding the sand-epoxy filler material inside the

pipe. This arrangement gave better results after many trails with other end

gripper setups.

A typical tensile test conducted on a GFRP bar is shown in

Figure 4.27. A graph plotted with the test results of a GFRP bar is shown in

Figure. 4.28. The trend of the stress-strain line plotted in the graph shows a

linear variation which is commonly noticed in GFRP composites.

98

Figure 4.27 Tensile tests on GFRP Bar

Figure 4.28 Stress-strain graph of GFRP Bar tested in tension