9
The 1 st Regional Conference of Eng. Sci. NUCEJ Spatial ISSUE vol.11,No.3, 2008 pp 374-381 NUCEJ vol.11, No.3,2008 Effect of Reinforcement Density 374 The Effect of Reinforcement Density on The Deflection of Reinforced Soil Retaining Wall Dhiaadin Bahaadin Noory Zangana MSc. Civil Engineering / Geotechniqu Lecturer Building and Construction Dept. College of Engineering University of Sulaimani Abstract This research involves the effect of both vertical and horizontal spacing, as well as the length of the reinforcing strips, on the structural behavior of a model reinforced soil retaining wall. The tests were carried out on a small-scale laboratory model, in which uniform graded sand was used as a fill material. It was placed at a high dense state; raining technique was used for this purpose. The sand was reinforced with thin dipped Aluminum strips. The size of these reinforcing strips was chosen sufficient enough to prevent rupture mode of failure. The retaining wall was presented by a plastic transparent sheet. The results showed that the horizontal spacing is more significant parameter on deflection of the wall as compared to the vertical spacing of the reinforcing strips. The critical values of the horizontal and vertical spacing with respect to this research were 20 and 10 cm respectively. A short length of reinforcement proved to cause appreciable deflection of the wall, even under close horizontal and vertical spacing. Keywords Retaining wall, vertical spacing of reinforcing strips, horizontal spacing of reinforcing strips, length of reinforcing strips, deflection. Symboles H Total height of retaining wall S v Vertical spacing of reinforcing Strips. S h Horizontal spacing of reinforcing Strips. L Length of the reinforcing strips Introduction Reinforced soil is defined as, a construction material composed of soil fill limited to cohesionnless free drainage materials, that is strong in compression but weak in tension, and the reinforcing strips, which are relatively high tensile strength materials, placed at stipulated spacing, and they supply the mass with the necessary tensions. Lastly the facing element, which is usually non structure element, acting as an outer membrane to prevent the soil from sloughing [5]. The soil and reinforcing strips will interact by means of friction resistance [8], and results into stable mass, that behaves monolithically and can be used as earth retention and load supporting structures. Reinforced soil is really an attractive and retaining walls, bridge abutments, platform supporting structures, foundation slabs, under water quay and sea walls, dams, sedimentation basins and tunnel linings etc. The applications of reinforced soil to soil retaining and load supporting structures have been studied on a theoretical and analytical basis. The basic assumptions and results of these studies have been checked and found to be realistic by constructing small scale models in the laboratory. Most of the studies that have been traced till now were concerning the modes of failure, seismic effect on the behavior of reinforced soil wall, suitability of materials for reinforcing strips as well as the effect of placement condition of fill materials. On the other hand, the works performed on full scale structures were concentrated on the stress measurement within the soil and stress distribution along the reinforcements. Lee et.al. [7] have performed several studies on small model of reinforced earth Walls, using thin aluminum ties their results indicated that Rankine or Columb theory is suitable for design purposes. Smith and Bransby [10] carried out series of tests on small models of reinforced wall using aluminum foils, they used typical radio graph system to monitor the failure surface , they found that angle of failure plane varies between 20 - 25 degrees with vertical. Brom [3] carried out small scale laboratory tests using fiber reinforcement; his results enabled development of design criteria. Al-Hussaini and Perry [1] tested a full scale experimental wall, the wall was reinforced with galvanized steel and loaded by static surcharge up to failure, they concluded that the lateral pressure at the end of construction were approximately equal to those predicated by Rankine theory for active condition. Richardson et. al. [9] constructed a prototype wall and applied various dynamic loads. They showed that the static design of reinforced soil

The Effect of Reinforcement Density on The Deflection of

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: The Effect of Reinforcement Density on The Deflection of

The 1stRegional Conference of Eng. Sci. NUCEJ Spatial ISSUE vol.11,No.3, 2008 pp 374-381

NUCEJ vol.11, No.3,2008 Effect of Reinforcement Density 374

The Effect of Reinforcement Density on The Deflection of Reinforced Soil Retaining Wall

Dhiaadin Bahaadin Noory Zangana

MSc. Civil Engineering / Geotechniqu Lecturer Building and Construction Dept. College of Engineering University of Sulaimani

Abstract This research involves the effect of both

vertical and horizontal spacing, as well as the length of the reinforcing strips, on the structural behavior of a model reinforced soil retaining wall. The tests were carried out on a small-scale laboratory model, in which uniform graded sand was used as a fill material. It was placed at a high dense state; raining technique was used for this purpose. The sand was reinforced with thin dipped Aluminum strips. The size of these reinforcing strips was chosen sufficient enough to prevent rupture mode of failure. The retaining wall was presented by a plastic transparent sheet. The results showed that the horizontal spacing is more significant parameter on deflection of the wall as compared to the vertical spacing of the reinforcing strips. The critical values of the horizontal and vertical spacing with respect to this research were 20 and 10 cm respectively. A short length of reinforcement proved to cause appreciable deflection of the wall, even under close horizontal and vertical spacing.

Keywords Retaining wall, vertical spacing of reinforcing strips, horizontal spacing of reinforcing strips, length of reinforcing strips, deflection.

Symboles H Total height of retaining wall Sv Vertical spacing of reinforcing Strips.

Sh Horizontal spacing of reinforcing Strips. L Length of the reinforcing strips

Introduction Reinforced soil is defined as, a construction material composed of soil fill limited to cohesionnless free drainage materials, that is strong in compression but weak in tension, and the reinforcing strips, which are relatively high tensile strength materials, placed at stipulated spacing, and they supply the mass with the necessary tensions. Lastly the facing element, which is usually non structure element, acting as an outer membrane to prevent the soil from sloughing [5]. The soil and reinforcing strips

will interact by means of friction resistance [8], and results into stable mass, that behaves monolithically and can be used as earth retention and load supporting structures. Reinforced soil is really an attractive and retaining walls, bridge abutments, platform supporting structures, foundation slabs, under water quay and sea walls, dams, sedimentation basins and tunnel linings etc. The applications of reinforced soil to soil retaining and load supporting structures have been studied on a theoretical and analytical basis. The basic assumptions and results of these studies have been checked and found to be realistic by constructing small scale models in the laboratory. Most of the studies that have been traced till now were concerning the modes of failure, seismic effect on the behavior of reinforced soil wall, suitability of materials for reinforcing strips as well as the effect of placement condition of fill materials. On the other hand, the works performed on full scale structures were concentrated on the stress measurement within the soil and stress distribution along the reinforcements. Lee et.al. [7] have performed several studies on small model of reinforced earth Walls, using thin aluminum ties their results indicated that Rankine or Columb theory is suitable for design purposes. Smith and Bransby [10] carried out series of tests on small models of reinforced wall using aluminum foils, they used typical radio graph system to monitor the failure surface , they found that angle of failure plane varies between 20 - 25 degrees with vertical. Brom [3] carried out small scale laboratory tests using fiber reinforcement; his results enabled development of design criteria. Al-Hussaini and Perry [1] tested a full scale experimental wall, the wall was reinforced with galvanized steel and loaded by static surcharge up to failure, they concluded that the lateral pressure at the end of construction were approximately equal to those predicated by Rankine theory for active condition. Richardson et. al. [9] constructed a prototype wall and applied various dynamic loads. They showed that the static design of reinforced soil

Page 2: The Effect of Reinforcement Density on The Deflection of

NUCEJ vol

structure is convenient for moderately intense seismic loading.None of the published works had paid attention to study the effect of reinforstrips spacing on the behavior of reinforced sandy soil retaining wall. It is therefore the task of this research is to highlight the behavior of reinforced sandy soil retaining wall, with various spacing and length of the reinforcing strips.MaterialWorksMaterialsThe soil used in the tests of this research was from Khasawhich is available in large quantities in the local area, and used in most of construction activities. Suitable amount of the soil was washed, and sieved on sieve No. size, for better workabilitSieve analysis was carried out on the soil to determine the grain size distribution of the soil particles, according to the British standard and Bthat the soil was granular uniformly graded sand with a unified soilclassification system destination SW, with a uniformity coefficient Cuconcavity coefficient Cc=l.l and The specific gravity of the sand particles was found to be density of the sand was determined by means of compactminimum density was found by jar test method. The upper and lower values of the density were respectively. The density of the sand was calibrated by sand raining method sand was rained through a mesh 2.87of drop, which gave different values of placing densities. It was decide to use dense state through out the research with density of 40

cm height of raining which yielded a relatiinternal friction between the sand grains was which was determined using direct shear test [

Reinforcing stripsThe reinforcing strips used in the model of this research were dipped aluminum

NUCEJ vol.11,

structure is convenient for moderately intense seismic loading.

None of the published works had paid attention to study the effect of reinforstrips spacing on the behavior of reinforced sandy soil retaining wall. It is therefore the task of this research is to highlight the behavior of reinforced sandy soil retaining wall, with various spacing and length of the reinforcing strips.

Materials And ExperimentalWorks

Materials

Soil

The soil used in the tests of this research was from Khasawhich is available in large quantities in the local area, and used in most of construction activities. Suitable amount of the soil was washed, and sieved on sieve

. 14 and 200, size, for better workabilitSieve analysis was carried out on the soil to determine the grain size distribution of the soil particles, according to the British standard and B.Sthat the soil was granular uniformly graded sand with a unified soilclassification system destination SW, with a uniformity coefficient Cuconcavity coefficient Cc=l.l and The specific gravity of the sand particles was found to be density of the sand was determined by means of compactminimum density was found by jar test method. The upper and lower values of the density were 16.98respectively. The density of the sand was calibrated by sand raining method sand was rained through a mesh 2.87

mm openingof drop, which gave different values of placing densities. It was decide to use dense state through out the research with density of 15

kN

cm height of raining which yielded a relative density of internal friction between the sand grains was 35

degrees at the proposed densitywhich was determined using direct shear

[4]

Reinforcing stripsThe reinforcing strips used in the model of this research were dipped aluminum

, No.3,2008

structure is convenient for moderately intense

None of the published works had paid attention to study the effect of reinforstrips spacing on the behavior of reinforced sandy soil retaining wall. It is therefore the task of this research is to highlight the behavior of reinforced sandy soil retaining wall, with various spacing and length of the

s And Experimental

The soil used in the tests of this research was from Khasa

river bed in Kirkuk, which is available in large quantities in the local area, and used in most of construction activities. Suitable amount of the soil was washed, and sieved on sieve

, to have a suitable particle size, for better workabilitSieve analysis was carried out on the soil to determine the grain size distribution of the soil particles, according to the British

S.I 1377 [4], that the soil was granular uniformly graded sand with a unified soilclassification system destination SW, with a uniformity coefficient Cuconcavity coefficient Cc=l.l and The specific gravity of the sand particles was found to be 2.66 [4]. The maximum density of the sand was determined by means of compaction test minimum density was found by jar test method. The upper and lower values of the

16.98

and respectively. The density of the sand was calibrated by sand raining method sand was rained through a mesh

mm opening, using different heights of drop, which gave different values of placing densities. It was decide to use dense state through out the research with

kN/m3 This was obtained by

cm height of raining which yielded a ve density of 70%. The angle of

internal friction between the sand grains

degrees at the proposed densitywhich was determined using direct shear

Reinforcing strips

The reinforcing strips used in the model of this research were dipped aluminum

structure is convenient for moderately intense

None of the published works had paid attention to study the effect of reinforcing strips spacing on the behavior of reinforced sandy soil retaining wall. It is therefore the task of this research is to highlight the behavior of reinforced sandy soil retaining wall, with various spacing and length of the

s And Experimental

The soil used in the tests of this research river bed in Kirkuk,

which is available in large quantities in the local area, and used in most of construction activities. Suitable amount of the soil was washed, and sieved on sieve

to have a suitable particle size, for better workability conditions. Sieve analysis was carried out on the soil to determine the grain size distribution of the soil particles, according to the British

, it was found that the soil was granular uniformly graded sand with a unified soilclassification system destination SW, with a uniformity coefficient Cu=2.2, concavity coefficient Cc=l.l and D5o=0.4The specific gravity of the sand particles

The maximum density of the sand was determined by

ion test [4], while the minimum density was found by jar test method. The upper and lower values of the

and 14.32 kN/mrespectively. The density of the sand was calibrated by sand raining method [6]. The sand was rained through a mesh 2.87

x using different heights

of drop, which gave different values of placing densities. It was decide to use dense state through out the research with

This was obtained by

cm height of raining which yielded a The angle of

internal friction between the sand grains

degrees at the proposed densitywhich was determined using direct shear

The reinforcing strips used in the model of this research were dipped aluminum 1.0

mm thick

structure is convenient for moderately intense

None of the published works had paid cing

strips spacing on the behavior of reinforced sandy soil retaining wall. It is therefore the task of this research is to highlight the behavior of reinforced sandy soil retaining wall, with various spacing and length of the

The soil used in the tests of this research river bed in Kirkuk,

which is available in large quantities in the local area, and used in most of construction activities. Suitable amount of the soil was washed, and sieved on sieve

to have a suitable particle y conditions.

Sieve analysis was carried out on the soil to determine the grain size distribution of the soil particles, according to the British

it was found that the soil was granular uniformly graded sand with a unified soil

classification system destination SW, with , a 0.4.

The specific gravity of the sand particles The maximum

density of the sand was determined by while the

minimum density was found by jar test method. The upper and lower values of the

m3

respectively. The density of the sand was The

x using different heights

of drop, which gave different values of placing densities. It was decide to use dense state through out the research with

This was obtained by

cm height of raining which yielded a The angle of

internal friction between the sand grains

degrees at the proposed density, which was determined using direct shear

The reinforcing strips used in the model of this

mm thick,

20reinforcing strip was friction between the sand particles and the reinforcing strips was by direct shear testusing the box was filled with a wooden block, covered with the reinforcement material, while the upper part of the box was filled with the sand placed at the proposed density. A linrelationship was observed between normal stress and shear stress.

Testing ApparatusThe plan and appropriate section of the testing apparatus of this research is shown in and

The testing apparatus consists 1. 450side thickness and base contains two

20

mm wide.

reinforcing strip was friction between the sand particles and the reinforcing strips was by direct shear testusing 60 x 60

mm shear box

the box was filled with a wooden block, covered with the reinforcement material, while the upper part of the box was filled with the sand placed at the proposed density. A linrelationship was observed between normal stress and shear stress.

Testing ApparatusThe plan and appropriate section of the testing apparatus of this research is shown in and Fig. (2) below respectively

The testing apparatus consists . A well stiffened wooden box of

450

mm internal dimensionsside thickness and base contains two

Zangana

. The breaking strength of the reinforcing strip was 135 N/mmfriction between the sand particles and the reinforcing strips was 25

degree

by direct shear test

mm shear box.

the box was filled with a wooden block, covered with the reinforcement material, while the upper part of the box was filled with the sand placed at the proposed density. A linrelationship was observed between normal stress and shear stress.

Testing Apparatus

The plan and appropriate section of the testing apparatus of this research is shown in

below respectively

The testing apparatus consists A well stiffened wooden box of

mm internal dimensionsside thickness and 30

mm base thicknessbase contains two 20 x 20

mm square holes

Zangana

375

The breaking strength of the mm2. The angle of

friction between the sand particles and the

degrees, determined

. The lower part of the box was filled with a wooden block, covered with the reinforcement material, while the upper part of the box was filled with the sand placed at the proposed density. A linrelationship was observed between normal

The plan and appropriate section of the testing apparatus of this research is shown in Fig

below respectively.

The testing apparatus consists of: A well stiffened wooden box of 800 x 650

mm internal dimensions, with 12.5

mm

mm base thickness.

mm square holes

375

The breaking strength of the

. The angle of friction between the sand particles and the

s, determined

The lower part of the box was filled with a wooden block, covered with the reinforcement material, while the upper part of the box was filled with the sand placed at the proposed density. A linear relationship was observed between normal

The plan and appropriate section of the testing Fig. (1)

650

x

mm . The

mm square holes

Page 3: The Effect of Reinforcement Density on The Deflection of

NUCEJ

provided with control gates for emptying the box.

2. A rigid steel framesupport the wooden box.3. The retaining wall model was selected to be transparent plastic sheet of The thickness of the wall model selected to give a measurable deflection with reasonable accuracy. The wall was simply suppothe base and let free at the top and side edges, to simulate the condition which is mainly faced in actual practice. The simply supported arrangement was made by inserting the plastic wall into width of the base of tcm from the right side of the wall.Figure strips location under different vertical and horizontal spacings used throughout the tests.

Fig

4. Fourty nine number of plates were attached to the plastic wall, at the required spacing, to act as connectors to the reinforcing strips. The reinforcing strips were bolted to the connecting plates through the holes provided for this puconnection arrangements is shown in Figure below

NUCEJ vol.11, No

provided with control gates for emptying the

A rigid steel frame

support the wooden box.The retaining wall model was selected to be

transparent plastic sheet of The thickness of the wall model selected to give a measurable deflection with reasonable accuracy. The wall was simply suppothe base and let free at the top and side edges, to simulate the condition which is mainly faced in actual practice. The simply supported arrangement was made by inserting the plastic wall into 20

mm slot dragged across total width of the base of tcm from the right side of the wall.Figure 3

below shows layout of the reinforcing strips location under different vertical and horizontal spacings used throughout the tests.

Fig (3)Layout of the Reinforcing Strip

Fourty nine number of plates were attached to the plastic wall, at the required spacing, to act as connectors to the reinforcing strips. The reinforcing strips were bolted to the connecting plates through the holes provided for this puconnection arrangements is shown in Figure below

No.3,2008

provided with control gates for emptying the

A rigid steel frame, one support the wooden box.

The retaining wall model was selected to be

transparent plastic sheet of 650The thickness of the wall model selected to give a measurable deflection with reasonable accuracy. The wall was simply suppothe base and let free at the top and side edges, to simulate the condition which is mainly faced in actual practice. The simply supported arrangement was made by inserting the plastic

mm slot dragged across total width of the base of the wooden box at cm from the right side of the wall.

below shows layout of the reinforcing strips location under different vertical and horizontal spacings used throughout the tests.

Layout of the Reinforcing Strip locations

Fourty nine number of 20 x 20plates were attached to the plastic wall, at the required spacing, to act as connectors to the reinforcing strips. The reinforcing strips were bolted to the connecting plates through the holes provided for this purpose. A detail of connection arrangements is shown in Figure

Effect of Reinforcement Density

provided with control gates for emptying the

one meter high to

The retaining wall model was selected to be 650

x 420 x 8

mm

The thickness of the wall model selected to give a measurable deflection with reasonable accuracy. The wall was simply supported at the base and let free at the top and side edges, to simulate the condition which is mainly faced in actual practice. The simply supported arrangement was made by inserting the plastic

mm slot dragged across total he wooden box at 15.22

cm from the right side of the wall.

below shows layout of the reinforcing strips location under different vertical and horizontal spacings used throughout the tests.

Layout of the Reinforcing Strip

20

x 1

mm steel plates were attached to the plastic wall, at the required spacing, to act as connectors to the reinforcing strips. The reinforcing strips were bolted to the connecting plates through the

rpose. A detail of connection arrangements is shown in Figure

Effect of Reinforcement Density

provided with control gates for emptying the

meter high to

The retaining wall model was selected to be

mm.

The thickness of the wall model selected to give a measurable deflection with reasonable

rted at the base and let free at the top and side edges, to simulate the condition which is mainly faced in actual practice. The simply supported arrangement was made by inserting the plastic

mm slot dragged across total 15.22

below shows layout of the reinforcing strips location under different vertical and horizontal spacings used throughout the tests.

mm steel plates were attached to the plastic wall, at the required spacing, to act as connectors to the reinforcing strips. The reinforcing strips were bolted to the connecting plates through the

rpose. A detail of connection arrangements is shown in Figure 4

Fig

5. 15.22plastic wall vertically, They were paced in front of the wall, at the box sides, to easily arrange the dial gauges at the proposed locations.6. vertical center line of the plastic wall, one at spacing of at the top of vertical center line, to measure the deflection of the plastic wall.7. aluminumdiameter and hopper was at the bottom raining hose. The connection wasflexible for easy control of sand raining. This was achieved by The raining hose was of sliding type. It consists of two aluminum pipes, one sliding inside the other, thus raining height2.87end of the hose, to control the rate of flowing sand. The size of the opening was selected to be sufficient for uniform flowing of sand duTesting programThe main parameters concerned in thisresearch were:a. Effect of vertical spacing of the reinforcing strips.b. Effect of horizontal spacing of the reinforcing strips.c. Effect of length of the reinforcing strips.For this purpose the testing program is divided into two groups, as follows:

Effect of Reinforcement Density

Fig(5) Reinforcing Strip Connection with

. Two temporary wooden supports of 15.22 x 40 cm dimension were used to hold the plastic wall vertically, They were paced in front of the wall, at the box sides, to easily arrange the dial gauges at the proposed locations.

. Two dial gauges were placed along the vertical center line of the plastic wall, one at spacing of 10

cm from theat the top of vertical center line, to measure the deflection of the plastic wall.

. Sand hopper which was made of thin aluminum, cylindrical in shapediameter and 100hopper was at the bottom raining hose. The connection wasflexible for easy control of sand raining. This was achieved by The raining hose was of sliding type. It consists of two aluminum pipes, one sliding inside the other, thus raining height, A stainless steel mesh of 2.87

mm square opening was attached to the end of the hose, to control the rate of flowing sand. The size of the opening was selected to be sufficient for uniform flowing of sand during the filling processTesting programThe main parameters concerned in thisresearch were:

a. Effect of vertical spacing of the reinforcing strips.

b. Effect of horizontal spacing of the reinforcing strips.c. Effect of length of the reinforcing strips.For this purpose the testing program is divided into two groups, as follows:

Effect of Reinforcement Density

Reinforcing Strip Connection with the plastic wall

Two temporary wooden supports of cm dimension were used to hold the

plastic wall vertically, before loading process. They were paced in front of the wall, at the box sides, to easily arrange the dial gauges at the proposed locations.

Two dial gauges were placed along the vertical center line of the plastic wall, one at

cm from the

base, and the other at the top of vertical center line, to measure the deflection of the plastic wall.

Sand hopper which was made of thin cylindrical in shape

100

cm height. hopper was at the bottom and connected to the raining hose. The connection wasflexible for easy control of sand raining. This was achieved by 80

mm diaThe raining hose was of sliding type. It consists of two aluminum pipes, one sliding inside the other, thus providing good control of

A stainless steel mesh of

mm square opening was attached to the end of the hose, to control the rate of flowing sand. The size of the opening was selected to be sufficient for uniform flowing of sand

ring the filling process

Testing program

The main parameters concerned in this

a. Effect of vertical spacing of the reinforcing

b. Effect of horizontal spacing of the reinforcing strips.

c. Effect of length of the reinforcing strips.For this purpose the testing program is divided into two groups, as follows:

Reinforcing Strip Connection with the plastic wall

Two temporary wooden supports of cm dimension were used to hold the

before loading process. They were paced in front of the wall, at the box sides, to easily arrange the dial gauges at

Two dial gauges were placed along the vertical center line of the plastic wall, one at

base, and the other at the top of vertical center line, to measure the deflection of the plastic wall.

Sand hopper which was made of thin cylindrical in shape, with 50

. The outlet of the and connected to the

raining hose. The connection was

flexible for easy control of sand raining. This

mm dia. plastic pipeThe raining hose was of sliding type. It consists of two aluminum pipes, one sliding

providing good control of A stainless steel mesh of 2.87

mm square opening was attached to the end of the hose, to control the rate of flowing sand. The size of the opening was selected to be sufficient for uniform flowing of sand

The main parameters concerned in this

a. Effect of vertical spacing of the reinforcing

b. Effect of horizontal spacing of the

c. Effect of length of the reinforcing strips.For this purpose the testing program is divided

376

Reinforcing Strip Connection with

Two temporary wooden supports of 25 x cm dimension were used to hold the

before loading process. They were paced in front of the wall, at the box sides, to easily arrange the dial gauges at

Two dial gauges were placed along the vertical center line of the plastic wall, one at

base, and the other at the top of vertical center line, to measure the

Sand hopper which was made of thin 50

cm The outlet of the

and connected to the

flexible for easy control of sand raining. This plastic pipe. 8.

The raining hose was of sliding type. It consists of two aluminum pipes, one sliding

providing good control of 2.87

x

mm square opening was attached to the end of the hose, to control the rate of flowing sand. The size of the opening was selected to be sufficient for uniform flowing of sand

a. Effect of vertical spacing of the reinforcing

b. Effect of horizontal spacing of the

c. Effect of length of the reinforcing strips.

For this purpose the testing program is divided

Page 4: The Effect of Reinforcement Density on The Deflection of

NUCEJ vol.11, No.3,2008 Zangana 377

1. Group I One test was done on non-reinforced sandy soil retaining wall. 2. Group II Twenty seven tests were done on reinforced sandy soil retaining wall, as follows: a. Nine tests to show the effect of vertical spacing Sv, using L=25, 37.5 and 50 cm for different horizontal spacing Sh equal to 10,20 and 30 cm. b. Nine tests to show the effect of horizontal spacing Sh, using Sv=5,10 and 15 cm for different length L equal to 25,37.5 and 50 cm. c. Nine tests to show the effect of length L, using Sh=10,20 and 30 cm for different vertical spacing Sv equal to 5,10 and 15 cm. While other parameters such as properties of the reinforcement, properties of the fill and placement density of the sand were kept constant. Testing procedure The testing procedure was done according to the following steps: 1. Placing the plastic wall model in its specified location by inserting it into 20 mm deep slot which was dragged in the wooden base, to act as simply supported of the bottom edge of the plastic wall. 2. Placing the two temporary wooden supports at their specified location in front of the plastic wall to hold it vertically. 3. Attaching the dial gauges at their proposed locations along the vertical center line of the plastic wall, and setting their reading to zero. 4. Placing the back fill soil by opening the lock of the raining hose. The height of falling was controlled by sliding the hose up and down, keeping a constant height of 40 cm above the sand surface. Th.ic II'QC o'Ji'ilv Hone with the aid of guide markers along the outer part of the hose. This height was chosen through the preliminary tests in order to obtain a high dense state of backfill, the raining hose was continuously moved forward and backward and in transverse directions. The sand was placed in layers of equal thickness, using marking lines on the interior sides of the box as a guide. When the level of the reinforcing strips was reached according to the testing program, leveling the sand surface was performed using a 2 mm thick steel strip across the full width of the box. Almost perfect leveling was needed to give a good contact between the reinforcing strips and the sand bed. This was achieved by gentle movement of the leveling strip forward and backward several times

5. Laying the reinforcing strips and connecting them to the supporting plates attached to the plastic wall. 6. Raising the temporary wooden supports, after completion of the backfill process, to release the plastic wall's free edges. 7. Recording reading of the dial gauges. 8. Opening the locks of the base holes control gates, at the end of the experiment, to discharge the sand from the box into a collection container. The collected sand was then transferred to the hopper to be reused in the next tests Results And Discussion Nature of the problem The wall is assumed as an elastic plate under pure bending, subjected to transverse active pressure of the fill material, and resisting forces from the reinforcing strips. It is also assumed to be simply supported at the bottom edge, while free at the top edge and vertical sides. It is important to be mentioned that the amount of measured deflections is relatively small in comparison with the height of the wall. The maximum deflection is limited to half of the wall thickness. More than this value leads to an inaccurate results, due to development of the membrane stresses (i.e.normal stresses on the middle plan of the plate) [2] Discussion of the results Table 1 below shows the results of deflections, along the vertical center line of the wall, using different spacing and length of the reinforcements

Page 5: The Effect of Reinforcement Density on The Deflection of

NUCEJ

L (cm)

50

37.5

25

None

NUCEJ vol.11, No

Table (1) Results of Deflections

(cm)

Sv

(cm)

5

10

15

37.5

5

10

15

5

10

15

None- reinforced

No.3,2008

Results of DeflectionsSh

(cm) at

top of the wall (mm)

10 0.3020 0.7330 0.8410 o.7120 1.1730 1.3710 1.1220 1.5730 1.8510 0.4920 0.7730 1.2310 0.9720 1.5530 2.1710 1.1720 2.030 2.5910 0.820 1.5430 2.010 1.4220 2.7130 3.4210 2.020 3.6130 4.34

reinforced

4.5

Effect of Reinforcement Density

Results of Deflections

at top of the wall (mm)

at 10

cm

from the base of the wall (mm)

0.30

0 0.73

-0.0070.84

-0.00971

-0.01

1.17

-0.02

1.37

-0.05

1.12

-0.07

1.57

-0.09

1.85

-0.12

0.49

-0.01

0.77

-0.03

1.23

-0.06

0.97

-0.08

1.55

-0.10

2.17

-0.15

1.17

-0.19

2.0

-0.22

2.59

-0.25

0.8

-0.03

1.54

-0.05

2.0

-0.08

1.42

-0.10

2.71

-0.15

3.42

-0.20

2.0

-0.25

3.61

-0.30

4.34

-0.35

4.5

2

Effect of Reinforcement Density

at

cm

of the

0.007

0.009

Figure deflection of the wall

It has been observed that the deflections follow the same trend in all the cases. The wall deflects in the same manner of a propped cantilever beam, subjected to triangular distribution load. Thus the maximum deflection occurred at the top part of the wallstrips in the lower half of the wall are more efficient in preventing the deflection than the upper strips, which means development of higher frictional resistance at the lower half of the wall. Also it has been observed that a radeflection occurs with depth. A point of zero deflection occurs at a depth slightly below the midthe wall consists of a unit plate, and due to low efficiency of upper strips, it deflects in a manner that it moved upper part, while it showed an inward movement at the lower part. This revised deflection can be referred to different efficiency of the strips and rigidity of the wall. Effect of vertical spacing SIt is found that for Lvertical spacing has slight effect on the deflection of the wall when Sh whilst when Shobtained. To clarify this, the variation of the maximum deflection with vertical spacing, at

Effect of Reinforcement Density

Figure 5

below shows typical distribution of

deflection of the wall

It has been observed that the deflections follow the same trend in all the cases. The wall deflects in the same manner of a propped cantilever beam, subjected to triangular distribution load. Thus the maximum deflection occurred at the top part of the wallstrips in the lower half of the wall are more efficient in preventing the deflection than the upper strips, which means development of higher frictional resistance at the lower half of the wall. Also it has been observed that a radeflection occurs with depth. A point of zero deflection occurs at a depth slightly below the midthe wall consists of a unit plate, and due to low efficiency of upper strips, it deflects in a manner that it moved upper part, while it showed an inward movement at the lower part. This revised deflection can be referred to different efficiency of the strips and rigidity of the wall.

Effect of vertical spacing SIt is found that for Lvertical spacing has slight effect on the deflection of the wall when Sh whilst when Shobtained. To clarify this, the variation of the maximum deflection with vertical spacing, at

Effect of Reinforcement Density

below shows typical distribution of

deflection of the wall

It has been observed that the deflections follow the same trend in all the cases. The wall deflects in the same manner of a propped cantilever beam, subjected to triangular distribution load. Thus the maximum deflection occurred at the top part of the wall. This indicates that the strips in the lower half of the wall are more efficient in preventing the deflection than the upper strips, which means development of higher frictional resistance at the lower half of the wall. Also it has been observed that a radeflection occurs with depth. A point of zero deflection occurs at a depth slightly below the mid-height of the wall. Since the wall consists of a unit plate, and due to low efficiency of upper strips, it deflects in a manner that it moved upper part, while it showed an inward movement at the lower part. This revised deflection can be referred to different efficiency of the strips and rigidity of the

Effect of vertical spacing SIt is found that for L=50

and vertical spacing has slight effect on the deflection of the wall when Sh whilst when Sh=30

cm a significant effect is obtained. To clarify this, the variation of the maximum deflection with vertical spacing, at

below shows typical distribution of

It has been observed that the deflections follow the same trend in all the cases. The wall deflects in the same manner of a propped cantilever beam, subjected to triangular distribution load. Thus the maximum deflection occurred at the top

. This indicates that the strips in the lower half of the wall are more efficient in preventing the deflection than the upper strips, which means development of higher frictional resistance at the lower half of the wall. Also it has been observed that a rapid decrease in deflection occurs with depth. A point of zero deflection occurs at a depth slightly

height of the wall. Since the wall consists of a unit plate, and due to low efficiency of upper strips, it deflects in a manner that it moved outward in the upper part, while it showed an inward movement at the lower part. This revised deflection can be referred to different efficiency of the strips and rigidity of the

Effect of vertical spacing Sv

and 37.5

cm vertical spacing has slight effect on the deflection of the wall when Sh =10 and 20

cm a significant effect is obtained. To clarify this, the variation of the maximum deflection with vertical spacing, at

378

below shows typical distribution of

It has been observed that the deflections follow the same trend in all the cases. The wall deflects in the same manner of a propped cantilever beam, subjected to triangular distribution load. Thus the maximum deflection occurred at the top

. This indicates that the strips in the lower half of the wall are more efficient in preventing the deflection than the upper strips, which means development of higher frictional resistance at the lower half of the wall. Also it has

pid decrease in deflection occurs with depth. A point of zero deflection occurs at a depth slightly

height of the wall. Since the wall consists of a unit plate, and due to low efficiency of upper strips, it deflects in

outward in the upper part, while it showed an inward movement at the lower part. This revised deflection can be referred to different efficiency of the strips and rigidity of the

cm , the vertical spacing has slight effect on the

20

cm,

cm a significant effect is obtained. To clarify this, the variation of the maximum deflection with vertical spacing, at

Page 6: The Effect of Reinforcement Density on The Deflection of

NUCEJ vol

the top vertical center liin Fig

Fig

The maximum deflection increased by increasing the vertical spacing. The rate of increase was found to be constant and gentle for all values of length of the strips. The deflection increased at a higher rate at the beginning, then decreased when S10

cmthe slopes of the curve. Effect of horizontal spacing ShFigure maximum deflection of the wall, with the horizontal spacing Sh at the top vertical center line of the wall.

Fig

NUCEJ vol.11,

the top vertical center liFig. 6 below

Fig6 variation of vertical spacin

The maximum deflection increased by increasing the vertical spacing. The rate of increase was found to be constant and gentle for all values of length of the strips. The deflection increased at a higher rate at the beginning, then decreased when S

cm. This is indicated by the difference in the slopes of the curve. Effect of horizontal spacing ShFigure 7

below shows the variation of maximum deflection of the wall, with the horizontal spacing Sh at the top vertical center line of the wall.

Fig 7 variation of the Horizontal spacing

, No.3,2008

the top vertical center line of the wall is plotted

variation of Maximum deflection with vertical spacing

The maximum deflection increased by increasing the vertical spacing. The rate of increase was found to be constant and gentle for all values of length of the strips. The deflection increased at a higher rate at the beginning, then decreased when S

This is indicated by the difference in the slopes of the curve.

Effect of horizontal spacing Sh

below shows the variation of maximum deflection of the wall, with the horizontal spacing Sh at the top vertical center

variation of Maximum deflection, with orizontal spacing

ne of the wall is plotted

aximum deflection with g

The maximum deflection increased by increasing the vertical spacing. The rate of increase was found to be constant and gentle for all values of length of the strips. The deflection increased at a higher rate at the beginning, then decreased when Sv exceede

This is indicated by the difference in

Effect of horizontal spacing Sh

below shows the variation of maximum deflection of the wall, with the horizontal spacing Sh at the top vertical center

aximum deflection, with orizontal spacing

ne of the wall is plotted

aximum deflection with

The maximum deflection increased by increasing the vertical spacing. The rate of increase was found to be constant and gentle for all values of length of the strips. The deflection increased at a higher rate at the

exceeded This is indicated by the difference in

below shows the variation of maximum deflection of the wall, with the horizontal spacing Sh at the top vertical center

aximum deflection, with

It is found that the rate of increase of maximum deflection when Sfrom the other range ,as illustrated by the slope of this portion ofEffect of length of the strips L It is noticed that the length of the reinforcing strips has a profound effect on the deflection of the wall, for any spacing of the strips. Howaffected by the the deflection increased with decreasing length of the stripsvariation of maximum deflection of the wall, with the length of the strips, at the vertical center line of the wall.

Fig

It is observed that the maximum deflection decreases with incThe relation is nondecreases at different rates. The greatest rate was f25slope of the curve. The effect of length on maximum deflection is significantly by the change of spacing. This is specially appearing when SCombined effect of spacing and length of the strips: deflection at the vertical center line of the wall with spacing and length of the reinforcing strips.

It is found that the rate of increase of maximum deflection when Sfrom 10 to 20

cm

the other range ,as illustrated by the slope of this portion of

the curve.

Effect of length of the strips LIt is noticed that the length of the reinforcing

strips has a profound effect on the deflection of the wall, for any spacing of the strips. However the lower part of the wall is slightly affected by the the deflection increased with decreasing length of the strips. variation of maximum deflection of the wall, with the length of the strips, at the vertical center line of the wall.

Fig 8 variation of maximum deflection with the length of the strips

It is observed that the maximum deflection decreases with incThe relation is nondecreases at different rates. The greatest rate was found when the length was changed from 25

to 37.5 cm. slope of the curve. The effect of length on maximum deflection is significantly by the change of spacing. This is specially appearing when SCombined effect of spacing and length of the strips: Figure 9

shows the variation of maximum deflection at the vertical center line of the wall with spacing and length of the reinforcing strips.

Zangana

It is found that the rate of increase of maximum deflection when S

cm, is more than the change in

the other range ,as illustrated by the slope of the curve.

Effect of length of the strips L

It is noticed that the length of the reinforcing strips has a profound effect on the deflection of the wall, for any spacing of the strips.

ever the lower part of the wall is slightly affected by the length, while at the upper part, the deflection increased with decreasing length

. Figure 8

below shows the variation of maximum deflection of the wall, with the length of the strips, at the vertical center line of the wall.

variation of maximum deflection with the length of the strips

It is observed that the maximum deflection decreases with increasing length of the strips. The relation is non-linear, where the deflection decreases at different rates. The greatest rate

ound when the length was changed from . This is clearly indicated by the

slope of the curve. The effect of length on maximum deflection is significantly by the change of spacing. This is specially appearing when Sh more than Combined effect of spacing and length of the strips:

shows the variation of maximum deflection at the vertical center line of the wall with spacing and length of the reinforcing

Zangana

379

It is found that the rate of increase of maximum deflection when Sh was changed

is more than the change in the other range ,as illustrated by the slope of

Effect of length of the strips L: It is noticed that the length of the reinforcing

strips has a profound effect on the deflection of the wall, for any spacing of the strips.

ever the lower part of the wall is slightly length, while at the upper part,

the deflection increased with decreasing length

below shows the variation of maximum deflection of the wall, with the length of the strips, at the vertical

variation of maximum deflection with the length of the strips

It is observed that the maximum deflection length of the strips.

linear, where the deflection decreases at different rates. The greatest rate

ound when the length was changed from This is clearly indicated by the

slope of the curve. The effect of length on maximum deflection is significantly influenced by the change of spacing. This is specially

more than 20 cm. Combined effect of spacing and

shows the variation of maximum deflection at the vertical center line of the wall with spacing and length of the reinforcing

379

It is found that the rate of increase of

was changed is more than the change in

the other range ,as illustrated by the slope of

It is noticed that the length of the reinforcing strips has a profound effect on the deflection of the wall, for any spacing of the strips.

ever the lower part of the wall is slightly length, while at the upper part,

the deflection increased with decreasing length

below shows the variation of maximum deflection of the wall, with the length of the strips, at the vertical

variation of maximum deflection with

It is observed that the maximum deflection length of the strips.

linear, where the deflection decreases at different rates. The greatest rate

ound when the length was changed from This is clearly indicated by the

slope of the curve. The effect of length on influenced

by the change of spacing. This is specially

Combined effect of spacing and

shows the variation of maximum deflection at the vertical center line of the wall with spacing and length of the reinforcing

Page 7: The Effect of Reinforcement Density on The Deflection of

NUCEJ

Figwith

It is found that the horizontal spacing Sh is the most affecting factor, specially when it changes between vertical spacing Sand its effect is rather uniform. The effect of strip length also shows an important effect, which is nonbetween Lconsidered as guidespacing, if the length and permissible deflection are known and vice versa.

ConclusionsBased on the results discussed in the present research, the following conclusions have been drawn:1. The wall deflects in a manner simof a propped cantilever beam. The maximum deflection in all the cases occurred at the top of vertical center line of the wall.2. It is found that the horizontal spacing is the most dominating factor on the deflection, and its effect is clearlbetween 3. Although the vertical spacing proved to have a significant effect on deflection of the wall, it's influence is rather uniform and much less than that of horizontal spacing.4. It is concluded that the uslength Linto excessive deflection, even under small

NUCEJ vol.11, No

Fig. 9 Variation of Maximum Deflection with Spacing and Length of the Strips

It is found that the horizontal spacing Sh is the most affecting factor, specially when it changes between vertical spacing Sand its effect is rather uniform. The effect of strip length also shows an important effect, which is non-linear and most significant between L=25

and considered as guidespacing, if the length and permissible deflection are known and vice versa.

Conclusions

Based on the results discussed in the present research, the following conclusions have been drawn:

The wall deflects in a manner simof a propped cantilever beam. The maximum deflection in all the cases occurred at the top of vertical center line of the wall.

It is found that the horizontal spacing is the most dominating factor on the deflection, and its effect is clearly pronounced in range of Sh between 10 and 20

Although the vertical spacing proved to have a significant effect on deflection of the wall, it's influence is rather uniform and much less than that of horizontal spacing.

It is concluded that the uslength L=25

cm is inconvenientinto excessive deflection, even under small

No.3,2008

Variation of Maximum Deflection Spacing and Length of the Strips

It is found that the horizontal spacing Sh is the most affecting factor, specially when it changes between 10 and 20

cmvertical spacing Sv appear to be less effective and its effect is rather uniform. The effect of strip length also shows an important effect,

linear and most significant

and 37.5 cm. Figure considered as guide

chart to find the adequate spacing, if the length and permissible deflection are known and vice versa.

Based on the results discussed in the present research, the following conclusions have been

The wall deflects in a manner simof a propped cantilever beam. The maximum deflection in all the cases occurred at the top of vertical center line of the wall.

It is found that the horizontal spacing is the most dominating factor on the deflection, and

y pronounced in range of Sh 20 cm.

Although the vertical spacing proved to have a significant effect on deflection of the wall, it's influence is rather uniform and much less than that of horizontal spacing.

It is concluded that the use of short strip

cm is inconvenientinto excessive deflection, even under small

Effect of Reinforcement Density

Variation of Maximum Deflection Spacing and Length of the Strips

It is found that the horizontal spacing Sh is the most affecting factor, specially when it

cm, while the appear to be less effective

and its effect is rather uniform. The effect of strip length also shows an important effect,

linear and most significant Figure 9

can be chart to find the adequate

spacing, if the length and permissible deflection are known and vice versa.

Based on the results discussed in the present research, the following conclusions have been

The wall deflects in a manner similar to that of a propped cantilever beam. The maximum deflection in all the cases occurred at the top of

It is found that the horizontal spacing is the most dominating factor on the deflection, and

y pronounced in range of Sh

Although the vertical spacing proved to have a significant effect on deflection of the wall, it's influence is rather uniform and much less than that of horizontal spacing.

e of short strip

cm is inconvenient, as it yields into excessive deflection, even under small

Effect of Reinforcement Density

Variation of Maximum Deflection

It is found that the horizontal spacing Sh is the most affecting factor, specially when it

while the appear to be less effective

and its effect is rather uniform. The effect of strip length also shows an important effect,

linear and most significant

can be chart to find the adequate

spacing, if the length and permissible

Based on the results discussed in the present research, the following conclusions have been

ilar to that of a propped cantilever beam. The maximum deflection in all the cases occurred at the top of

It is found that the horizontal spacing is the most dominating factor on the deflection, and

y pronounced in range of Sh

Although the vertical spacing proved to have a significant effect on deflection of the wall, it's influence is rather uniform and much

e of short strip as it yields

into excessive deflection, even under small

spacingdeflection appreciably and allows the use of wider spacing in both vertical and horizontal directions.

References1. "Field Experiment on Reinforced Earth wall". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint No2. Thin Flat Plates". JoDiv1073. Reinforced Retaining Structures". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania4. Soils for Civil Engineering Purposes". Inst, of Civil Engg. London.5. "Design and Field Behavior of Reinforced Earth Wall". Journal of the Geot. Engg. Div., Proc6. Experimental Study of Maximum and Minimum Porosities of Sands". Proc, of the 2nd IntRotterdam7. J. (Journal of soil Mech. Found. Div., Proc. ASCE8. "Seismic Design of Reinforced Earth Walls". Journal of the Geot. Engg. Div., Proc. ASCE, Vol9. LeeReinforced Earth Walls". Journal of Geot. EnggGTl10"The Failure of ReinfoOverturning". Journal of the Geot. London, Vol

Effect of Reinforcement Density

spacing, while the length of deflection appreciably and allows the use of wider spacing in both vertical and horizontal directions.

References

. Al-Hussaini,

"Field Experiment on Reinforced Earth wall". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint No. 3131.

. Bradley, W.Thin Flat Plates". JoDiv., Proc. ASCE107.

. Broms,B.B. (Reinforced Retaining Structures". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint No

. B.S.I. - 1377Soils for Civil Engineering Purposes". Inst, of Civil Engg. London.

. Change, J.C"Design and Field Behavior of Reinforced Earth Wall". Journal of the Geot. Engg. Div., Proc. ASCE, Vol

. KolbuszewskiExperimental Study of Maximum and Minimum Porosities of Sands". Proc, of the

nd Int. ConfRotterdam, Vol

. Lee, K.L., Adams. (1973). "Reinf

Journal of soil Mech. Found. Div., Proc. ASCE, Vol. 9, No

. Richardson, "Seismic Design of Reinforced Earth Walls". Journal of the Geot. Engg. Div., Proc. ASCE, Vol. 103, No. GT

. Richardson, Lee, K.L. (1977Reinforced Earth Walls". Journal of Geot. Engg. Div., ProcGTl.pp. 1-17. 10. Smith, A.K"The Failure of ReinfoOverturning". Journal of the Geot. London, Vol. 26, pp. 376

Effect of Reinforcement Density

while the length of 37.5deflection appreciably and allows the use of wider spacing in both vertical and horizontal

, M. and Perry"Field Experiment on Reinforced Earth wall". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint

.A. (1959). "Thin Flat Plates". Journal of the Engg. Mech.

ASCE, Vol. 85,

. (1978). "Design of Fabric Reinforced Retaining Structures". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh,

Preprint No. 31281377: (1975). "Method of Test for

Soils for Civil Engineering Purposes". Inst, of Civil Engg. London.v

C. and Forsyth"Design and Field Behavior of Reinforced Earth Wall". Journal of the Geot. Engg. Div.,

Vol.103, No. GTKolbuszewski, J.J. (

Experimental Study of Maximum and Minimum Porosities of Sands". Proc, of the

Conf. Soil MechVol. l,pp. 158-165

Adams, B.D. and VagernonReinforced Earth Retaining Walls".

Journal of soil Mech. Found. Div., Proc. No. SM10, pp, G.N. and Lee

"Seismic Design of Reinforced Earth Walls". Journal of the Geot. Engg. Div., Proc. ASCE,

GT1, pp. 1-17., G.N., Ferger, 1977). "Seismic Testing of

Reinforced Earth Walls". Journal of Geot. Proc. ASCE,

K.C. and Bransby"The Failure of Reinforced Earth Walls by Overturning". Journal of the Geot. London,

376-381.

37.5 cm reduce the deflection appreciably and allows the use of wider spacing in both vertical and horizontal

and Perry, E.B. (1978"Field Experiment on Reinforced Earth wall". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint

). "Laterally Loaded urnal of the Engg. Mech.

, No. EM4, pp

Design of Fabric Reinforced Retaining Structures". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh,

3128. Method of Test for

Soils for Civil Engineering Purposes". Inst, of

and Forsyth, R.A. (1977"Design and Field Behavior of Reinforced Earth Wall". Journal of the Geot. Engg. Div.,

GT7, pp. 677-692. (1948). "

Experimental Study of Maximum and Minimum Porosities of Sands". Proc, of the

Soil Mech. Found. Engg165. and Vagernon,

orced Earth Retaining Walls". Journal of soil Mech. Found. Div., Proc.

pp. 745-763. and Lee, K.L. (1975

"Seismic Design of Reinforced Earth Walls". Journal of the Geot. Engg. Div., Proc. ASCE,

.

, D., Fong, A. Seismic Testing of

Reinforced Earth Walls". Journal of Geot. , Vol. 103,

and Bransby, P.L. (1976rced Earth Walls by

Overturning". Journal of the Geot. London,

380

cm reduce the deflection appreciably and allows the use of wider spacing in both vertical and horizontal

1978). "Field Experiment on Reinforced Earth wall". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh, Pennsylvania, Preprint

Laterally Loaded urnal of the Engg. Mech.

pp. 76-

Design of Fabric Reinforced Retaining Structures". Proc. ASCE, Spring Convention and Exhibition, Pittsburgh,

Method of Test for Soils for Civil Engineering Purposes". Inst, of

1977). "Design and Field Behavior of Reinforced Earth Wall". Journal of the Geot. Engg. Div.,

692. ). "An

Experimental Study of Maximum and Minimum Porosities of Sands". Proc, of the

Engg.

, J-M. orced Earth Retaining Walls".

Journal of soil Mech. Found. Div., Proc.

1975). "Seismic Design of Reinforced Earth Walls". Journal of the Geot. Engg. Div., Proc. ASCE,

. and Seismic Testing of

Reinforced Earth Walls". Journal of Geot. , No.

1976). rced Earth Walls by

Overturning". Journal of the Geot. London,

Page 8: The Effect of Reinforcement Density on The Deflection of

NUCEJ vol.11, No.3,2008 Zangana 381

Page 9: The Effect of Reinforcement Density on The Deflection of

This document was created with Win2PDF available at http://www.daneprairie.com.The unregistered version of Win2PDF is for evaluation or non-commercial use only.