Experimental study on fiber-reinforced concrete panels subjected to shear, Experiments on FRC - Tesi

FRC.

A standard specimen has not been formulated for the uniaxial tension tests: RILEM

Technical Committee 162 (2001) has developed a recommendation for the uniaxial

142

4. EXPERIMENTS ON FRC

tensile test on the steel fiber reinforced concrete (SFRC). Their recommendation

involves testing a notched SFRC cylinder with a nominal cross sectional diameter of

150 mm and a circumferential notch with a dept and a width of 15 mm and 2-5 mm,

respectively. Following the procedures of Jimmy Susetyo, it was decided to use an

unnotched dog-bone shaped specimen instead of the recommended notched cylinder

specimen, in this experimental program.

Reasons are as follows: as previously discusses, the crack in a notched specimen will be

forced to occur at the notch, due to the stress concentration and the reduction of the

cross-sectional area at the notch. In an unnotched specimen with a uniform cross

section, cracks would develop at the weakest point along the length of the specimen.

However, bond failure related problems due to stress concentrations at the interface

between the concrete and the steel loading plate are common. The use of an unnotched

dog-bone shaped specimen allow the development of cracks at the weakest points, along

the length of the specimen, while minimizing the possibility of bond failure between the

steel loading plate and the concrete specimen by lowering the stress at the interface.

Dimensions and details of the dog-bone shaped specimen used in the experimental

program are given in the following Figure 4.12:

Figure 4.12: Uniaxial Tension Test. 143

4. EXPERIMENTS ON FRC

Dimension of midsection of the specimen were 100 x 70 mm, resulting in an effective

area of 7000 mm². The dimension were chosen such that some degree of random fiber

orientation within the specimen could be achieved while maintaining portability of the

specimen and resemblance to the condition in the concrete panel specimens, as done in

the thesis of Jimmy Susetyo.

Chosen of 70 mm dimension was done to reflect the thickness of the panel, following

the experiment done by Jimmy Susetyo, whereas the 100 mm dimension was chosen to

ensure the specimen portability while allowing some degree of random orientation.

The recommended minimum dimension for fiber reinforced concrete was at least three

times the fiber length (ASTM C1018, 1997).

A 50x50 MW9.1/9.1 wire mesh was also embedded in the end regions of the specimens,

as shown in the following Figure 4.13: this was done to ensure that the cracks would

occur in the middle portion of the specimens.

Figure 4.13: Wire mesh in the end region of the uniaxial tension test specimens. 144

4. EXPERIMENTS ON FRC

4.5.3.1 Casting Procedure

A pre-casting procedure was done, to prepare all the moulds that were needed for the

concrete cylinders, and it was made of eight steps.

The most difference following the thesis of Jimmy Susetyo are done by the point 3, 4, 5

and 6:

1. Three dog-bone forms were cleaned and assembled with a particular attention to

the dimension, to be sure that all is correct;

2. A ¾” threaded rod was the screwed into the end blocks such that the rod reached

65 mm into the specimen, with a nut on the outside to hold the rod in place;

3. Using a ruler, the threaded rods were verified to be in line with each other and in

the centre of the concrete specimen (in terms of height and width inside the

form);

4. A nut was placed on the end of the rod inside the form, to help with stress

transfer;

5. The forms were then oiled, and no oil was placed on the threaded rod;

6. Two wire meshes for each end of the dogbones were cut to the correct size,

meant to provide some reinforcement for the end large cross-section of the

specimens.

Once that the pre-casting procedure has been done, the casting was made by the

following steps:

1. Pour in 3 lifts, using a form vibrator to ensure good distribution each time;

2. At the end of vibration for each lift, one of the wire meshes was placed into the

end regions of the dog-bone (as showed in previous Figure 4.13). The wire

mesh was placed under the threaded rod for the first lift, and on top of the

threaded rod for the second lift;

3. Once full, the form was vibrated once more and the top surface was finished

using a trowel. An attempt was made to ensure that any fibers around the edge

of the specimen were removed to prevent sharp edges once the concrete had

cured;

4. The form was then covered with wet burlap and plastic and let cure one day; 145

4.

. EXPERIMENTS ON FRC

5. On the next day, the specimen was demoulded, marked and covered with wet

burlap and plastic

plast and let cure for 6 more days;

6. The burlap and plastic were then removed and the specimen was left to cure in

ambient conditions.

4.5.3.2 Test Procedure

The tests were conducted on an MTS Universal Testing Machine at the University of

Toronto. The machine

ine was a closed loop system with a maximum capacity of 245 kN

(55 kips) where tests were controlled by means of displacements transducers integral

with the MTS machine.

(a) (b)

(c

c) (d

(d)

Figure 4.14: Uniaxial Tension Specimen used in Susetyo experimental tests: (a)

Specimen dimension; (b)LVDTs configuration; (c) LVDTs configuration 2; (d) LVDTs

configuration 3. 146

4.

. EXPERIMENTS ON FRC

The MTS machine was equipped with two sets of universal joints: one set was bolted to

the actuator, and the other was bolted to the loads cell while the free ends of the joints

were threaded to allow a simple connection between the specimen and the machine. The

specimen was instrumented with a set of four Linear Variable Differential Transducers

(LVDTs), one for each side of the specimens. The LVDTs provided continuous

measurement of the displacement over the gauge length throughout the test.

(a) (b)

(c

c) (d

(d)

Figure 4.15: Uniaxial Tension Specimen used in experimental tests: (a) Specimen

dimension; (b)LVDTs configuration used; (c) Susetyo end-blocks;

end blocks; (d) experimental end-

end

blocks.

Jimmy Susetyo used three different types of LVDTs configurations, shown in the

previous Figure 4.14 and due to difficulties encountered along his studies, it was chosen

to adopt just only one of these: difficulties were related to cracks occurring outside the

147

4.

. EXPERIMENTS ON FRC

gauge length. So, working on Jimmy Susetyo’s experience,

experience

, one of three LVDTs set

configuration was used as shown in the previous Figure 4.15.

Jimmy Susetyo also adopted end-blocks

end blocks to transmit the load from the machine to the

t

specimen, gluing the steel end blocks to the specimen shape; in this experimental

programs, steel end-blocks

blocks were not used. As previous view, a ¾” threaded rod was the

screwed into the end blocks ending 65 mm into the specimen,

, with a rod on the outside

outsid

of the form to hold the rod in place: this solution was found to be more appreciable (and

easier) than the use of the steel-end

steel blocks.

Unfortunately, at the beginning of the test, a bad failure was obtained outside from the

LVDTs on the larger part of the specimen, so non-measurable;

measurable; it was then decided to

use experimental end-blocks:

end blocks: steel plates that permit the correct transmission of the load

to the specimen, without primary failure. This experimental procedure was new: steel

plate at the end of the specimen, as showed in the previous Figure 4.14,

4. and detailed in

the previous Figure 4.15,

4. , were connected to round cylinders where the dob-boned

dob

change

ge width, with two rods per size and it was found to be a good way to transmit the

load as shown in Figure 4.16

4. and in Figure 4.17:

Figure 4.16:

4. Detailed view of the experimental end-blocks.

end- 148

4.

. EXPERIMENTS ON FRC

Figure 4.17:

4. Detailed view of the experimental end-blocks.

end-

The ideal configuration of Susetyo’s LVDTs set, was the configuration 1, in which the

only displacement monitored was that in the mid-section

mid section of the specimens, where the

strain distribution was uniform. However, it was found by Susetyo that, in some

specimen,

men, cracks occurred outside the mid-150

mid 150 mm observation region, preventing data

collection of the post-

-cracking behavior.

So LVDTs configuration 2 and 3 were introduced to overcome this difficulty.

The preparation of the specimens was made by following the

th e next steps:

1. Instrumentation locations were marked off using a ruler: lines were drawn for

the instrumentation mounts, to be sure that it could be installed in the direct

centre of each face of the specimen, with a gauge length of 300 mm or 150 mm;

2. The specimens

ecimens were then painted;

3. Instrumentation mounts were then epoxied on to dogbones using a rig designed

to ensure the correct gauge length is maintained.

While the procedures to tests the specimens was as follows:

1. The dog-bone

bone was loaded into the MTS machine; 149

4. EXPERIMENTS ON FRC

2. LVDTs with a stroke of ±5 mm were then attached to the mounts and set to 3

mm (so as to allow 8mm of stroke in the LVDTs). The LVDTs were then

zeroed;

3. Loading was then commenced at a rate of 0.001 mm/s;

4. The test was run without changes until first cracking and then until the

maximum load post-cracking was achieved;

5. At this point, the loading rate was gradually increased to a maximum of

0.01mm/sec;

6. If the LVDTs approached saturation, they were reset;

7. At around 20% of the maximum post-cracking load, instrumentation was

removed;

8. Fibers across the main failure crack were then counted and catalogued. Cross-

sectional dimensions adjacent to the failure crack were measured.

4.5.4 Prisms

Prisms were made to investigate, with the bending tests, the tensile behavior of FRC

when subjected to bending moments and to evaluate the flexural toughness parameters

of the FRC.

As for the dog-bones, it was possible to make notched and unnotched specimen; the use

of an unnotched specimen, as done in the experiment tests, enables the cracks to occur

at the weakest point along the region of the constant moment.

Two different type of tests can be done to observe the behavior of this type of test: the

ASTM C1018 Standard Test, using a non-slotted beam under four-point loading and the

RILEM Technical Committee 162 (RILEM TC162), using a notched beam under three-

point loading.

The presence of a notch in the specimen, however, forces the crack to occur at the

notch: this simplifies the observation of the crack development and enables the

measurement of the crack mouth opening displacement. This weakness in the form of a

150

4. EXPERIMENTS ON FRC

notch introduced in the specimen may not reflect the actual flexural strength of the

concrete.

So, the use of an unnotched specimen was chosen, although the crack mouth