Effect of Steel Pins on Interface Shear Behavior of Segmental Concrete Units

Md. Zahidul Islam Bhuiyan* Postgraduate student Department of Civil Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia Email: mdzibhuiyan@gmail.com Faisal Hj Ali Professor Department of Civil Engineering, National Defense University of Malaysia, 57000 Kuala Lumpur, Malaysia Email: fahali@gmail.com Firas A. Salman Senior lecturer Department of Civil Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia Email: firasalman@hotmail.com Abstract - This study mainly focuses on shear strength behavior of newly and locally produced modular block units with and without steel shear pins. A series of direct shear tests was executed to find out the effectiveness of steel shear pins under different normal loading conditions. Test results were outlined in the form of shear forcedisplacement relationship to compare the influence of shear pins on shear strength behavior. Test results revealed that the presence rigid shear pins reduce the shear strength capacity than a purely frictional condition. Key words: interface shear, rigid pin, segmental block, shear strength, shear connector

1. INTRODUCTION
Geosynthetic reinforced segmental retaining walls (GRSRWs) consisting of polymeric reinforcements and precast modular block units have achieved popularity worldwide in the last three decades because of their many fold advantages. They are frequently used in many geotechnical applications. In Malaysia, the use of dry-stacked column of segmental units as a facing column in retaining wall constructions has been extensively practicing for last decades (Lee, 2000). Today, facing stability is an important issue in the current design guidelines (NCMA, 1997; Elias et al., 2001) and it mainly depends on interface shear and connection failures. Past research works (Soong & Koerner, 1997; Bathurst & Simac, 1993; Buttry et al., 1993) reported that facing instability basically occurs due to poor connection strength and inadequate connection systems. To develop interlocking mechanism between successive vertical courses of units, two different types of shear connections are mainly used in retaining wall constructions. One is built-in mechanical interlock in the form of concrete shear keys or leading/trailing lips and another one is the mechanical connector consisting of pins, clips, or wedges. Mechanical connectors are mainly used to help out unit alignment and control the wall facing batter. Bathurst and Simac (1997) reported that shear connectors (mechanical) or shear keys provide additional interface shear capacity of segmental concrete units. As a shear connector, steel pins (rigid) were used in this research to investigate its effect upon interface shear behavior of infilled units. To evaluate interface shear behaviors or performance parameters, a series of full scale laboratory tests was conducted with and without steel pins (NCMA SRWU-2, 1997; ASTM D 6916-03). Shear force-shear displacement graphs were drawn to compare performance of the infilled concrete units with and without shear pins. Shear capacity envelope graphs were also plotted by using Morh-Coulomb failure criteria under peak and service state criteria.

Corresponding Author: Md. Zahidul Islam Bhuiyan Email: mdzibhuiyan@gmail.com

2. MATERIALS 2.1 Segmental Concrete Unit

In this investigation I blocks were used as segmental concrete units. I blocks are machined mold wet cast concrete units (G 30), which have one center web and the tail/rear flange is extended beyond the web (Fig. 1). The rear flange is tapered that allows the blocks to form curve walls. The maximum tapered angle of the I block is 11.3 deg. I blocks are double open-ended units and make an equivalent hole in conjunction with two units, and the equivalent dimensions are around 450 mm in length, 280 mm in width and 300 mm in height. The infill weight is approximately 93 to 94 kg with the aggregate of bulk density of 1527 kg/m3.The physical and mechanical properties of the used block are outlined in Table 1.

gradation requirements. The physical properties of infill are given in Table 2.

2.3 Steel Bar

Galvanized mild steel round bars were used in the study as mechanical connectors that are generally known as shear connectors. According to the hole dimensions of the segmental concrete units, 12 mm dia bars were selected, and the bars were cut into 125 mm in length. The physical and mechanical properties of the used round steel bars are illustrated in Table 3. Table 2: Physical properties of granular infill. Property Bulk density (kg/m3) Specific gravity Void content (%) Coefficient of gradation, Cc Fineness Modulus (FM) Value 1527 2.63 42 1.15 7.16

Table 3: Physical and mechanical properties of steel bar (Courtesy of AMSteel Mills Sdn Bhd, Malaysia). Property Yield strength (MPa) Modulus of elasticity (MPa) Rolling mass (Kg/m) Cross section area (mm2) Value 347 210000 0.859 113.10

Figure 1: Schematic of used I block. Table 1: Physical and mechanical properties of segmental concrete units. Value 370x300x500 41-42 2166 % 7.1 Water absorption capacity kg/m3 155 Moisture content (%) 3.7 Net compressive strength (MPa) 8.0 * W = Width (Toe to heel), H= Height, L= Length (Parallel to the wall face) Property Dimensions (WxHxL)* in mm Weight (kg) Oven dry density (kg/m3)

3. EXPERIMENTAL DESIGNS 3.1 Test Apparatus

A specially designed and modified large-scale apparatus originally reported by Bathurst and Simac (1993) was used to carry out the performance tests of the I blocks. A photograph of the modified test apparatus is shown in Fig. 2. It is seen that the apparatus was mainly consisting of loading frame, hydraulic actuators, and a fabricated electric hydraulic pump. The vertical actuator was mounted with the loading frame with rollers to allow block movement during the shear test, but in ASTM D 6916-06c test protocol the vertical actuator was kept fixing. The vertical and horizontal actuators were capable of applying around 45 tons of surcharge load and 130 tons of push/pull out force respectively and simultaneously. The electric hydraulic pump was connected to the actuators with pressure hoses, and the pump was capable of delivering flow rate 3 cc per minute. A geosynthetic loading clamp was set with horizontal actuator to apply the tensile load as well as shear load.

2.2 Granular Infill

The hollow cores between the blocks were infilled with 100% crushed limestone aggregate and lightly compacted. The maximum and nominal maximum sizes of the aggregate were 25 and 19 mm, respectively. The particle size distribution of the granular infill meets the lower limit of the NCMA (1997)

Two (2) pressure transducers of 0 to 3,625 psi capacity were mounted over each hydraulic actuator of 150 mm stroke, and the actuators were calibrated by using load cell against the pressure transducers. Two (2) flow regulators (Atos QV06160) were attached with the pump to control the rate of displacement of horizontal (shear) and vertical actuators. The shear displacements were measured using of two 50 mm linear variable displacement transducers (LVDTs) with an accuracy of 0.001mm. Pressure transducers and LVDTs reading were continuously measured and recorded during the test by a data logger. The data were recorded at every 10 second interval.

Pressure transducer Vertical actuator Horizontal actuator

Electric pump

3.2 Interface Shear Tests

Two layers of modular block units were used for interface shear test. The bottom layer/course consisting of two (2) I blocks was installed and braced laterally at the front of loading frame. A single I block was placed centrally over the running joint formed by the two underlying units to simulate the staggered construction procedure used in the field. Two layers of segmental units were connected with shear pins and setback was kept as zero. The hollow sections between the blocks were filled with 19 mm crushed stone aggregate and lightly compacted using a steel rod. To hold the infilled aggregate of top block, two (2) steel plates were used (Fig. 3). Surcharge/normal load was imposed only over the top block through stiff rubber mat and simulated an equivalent height of stacked blocks. The shear load was applied against the top block and immediately above the shear interface to minimize the moment loading at a constant rate of 1 mm/min of the horizontal actuator (ASTM D 6916-03). A steel plate with stiff rubber mat was used with geosynthetic loading clamp to concentrate the shearing load only over the centrally installed top block. A horizontal seating load was applied to the top block to ensure close fitting of the shear pins and after that the load and displacement devices were set to zero. The imposed seating load was 10% of maximum shear strength. Mohr-Coulomb failure criteria were used to find out interface shear capacity at ultimate and service state strength criteria. (1) Where: = Interface shear capacity (kN) = Normal load (kN) = angle of friction (deg.) = interception

Figure 2: Photograph of test apparatus.

Figure 3: Photograph of interface shear test arrangement.

4. TEST RESULTS AND DISCUSSIONS

A series of interface shear tests was performed under a range of normal forces. Shear force - displacement graphs were plotted to evaluate the effects of rigid shear pins on the interface frictional behavior of the tested blocks. Shear capacity envelopes were also drawn to compare the ultimate and service state shear strength of I blocks. Figs. 4(a) and 4(b) compare the magnitude and distribution of shear force with displacement of infilled concrete units. Fig.4 (a) shows the gradually increment of shear force without any pick points that results from the absence of steel pins at joints. In this case, shear force reaches the steady state condition (around 10 kN) after a significant amount of shear displacement. On the other hand, a sudden pick shear force can be seen in Fig. 4(b) prior to serviceability displacement limit and blocks fail quickly at joints (Fig. 6).

The serviceability displacement is identified by vertical dashed line in the graphs, which is around 7 mm according to the block Geometry (2% of the block width). It is also seen that after immediate failure at joints, the shear interface behaves like purely frictional surface and there is no significant rise and fall of shear force with displacement increment rather than near to straight line (Fig. 4(b)). Fig. 4(c) also demonstrates the gradual increment of shear force against shear displacement without any pick points in the curve like Fig. 4(a). Fig. 4(d) reports three (3) repeated tests at almost same surcharge loads, which were controlled using analog pressure controlling valve. Three (3) nominally identical curves also show the accuracy of the performed laboratory tests and the peak shear forces of these tests are less than 10% from the mean of the three tests. The most interesting thing of Fig. 4(d), it is showing two (2) pick points; before and after serviceability line. It happens due to the failure mechanism of concrete blocks at the pin joints. At the time of the experiment, it was observed that one shear pin joint fails first than other because of block irregularity and set up alignments. As a result, shear force increases up to completed failure of both pin joints and drops permanently or becomes a steady state due to aggregate frictions. The test numbers 1 & 2 of Fig. 4(e) illustrate abrupt drops of shear strengths that happen due to sudden relief of frictional contact area of blocks interface like the behavior of tectonic plates. From the Fig. 4(f), it is seen that the steady state shear strength of pins connected infilled blocks reduce to about 15 kN that is less than the purely frictional condition (Fig. 4(e)). It occurs due to the stress concentration at the connection joints, which accelerate the failures of blocks at the flange area (Fig. 6(b)). As a result, the failure interface areas of blocks become loose enough and unable to carry shear force that causes to reduce interface shear strength at the high surcharge load conditions. The same behaviors are also observed in Fig. 4 (h) with respect to Fig. 4(g). The curve of Fig. 4(g) is wavier than others purely frictional graphs that result from a high surcharge load. The increment of normal load makes stress concentrations at different contact points of interface, which leads to rise and fall of shear force with the mobilization of blocks. The data presented in the Figs. 5(a) & 5(b) illustrates the influence of steel shear pins on shear capacity envelopes. It is clear from the Fig. 5(b) that shear pins provide more apparent cohesion (normal force-independent strength) than purely frictional block systems (Fig. 5(a)) at the ultimate condition, although the angle of internal friction is less. Bathurst and Simac (1994), and Bathurst et al. (2008) reported the same behaviors with different types of block geometries and shear connectors. The frictional interface area is a complex surface

consisting of block-block, block-infill, and infill-infill contact areas. The presence of steel shear pins in running bond causes immediately failures (spalling/cracks) in the block-block contact areas, which results in the reduction of concrete contact areas (Fig. 6). As a result, angle of internal friction becomes lower than purely frictional interface systems. Fig. 5(b) reports no serviceability capacity envelope due to the presence of steel pins in the connection system that breaks the blocks before reaching the serviceability limit. So it can be said that this connection system is not effective for service state design of segmental retaining walls with I blocks. As an alternative of steel pins, plastic pins may be used and investigated its effect on interface behavior.

5. CONCLUSIONS
This study investigates the results of interface shear testing executed to find out the effect of steel shear pins on frictional behavior of newly fabricated and modified I blocks. In this research, a series of interface shear tests was performed under several conditions. The following conclusions can be drawn based on the results: 1. Steel shear pins initially increase shear force but after immediate failure at the joints shear force decrease to purely frictional shear force of infilled blocks or less than that. Purely frictional behavior of infilled blocks is very smooth and steadier than infilled blocks with steel pins. Purely frictional systems easily govern service state criterion but the systems with steel pins are unable to follow that criterion. Steel bar is stiffer than concrete, as a result concrete fails easily at the connections prior to any significant shear displacement to mobilize shear strength. So, it is important to find out effective flexible shear connectors that give significant shear displacement as well as shear strength.

2.

3.

4.

30 25 20 15 10 5 Normal force = 8.71kN 0 0 5 10 15 20 25 Shear displacement (mm)

30 25 20 15 10 5 Normal force = 8.13 kN 0 0 5 10 15 20 25 Shear displacement (mm)

Shear force (kN)

Shear force (kN)

(b) (a)
30 25 20 15 10 5 0 0 5 10 15 20 25 Shear Displacement (mm)
30 25 20 15 10 5 Normal force = 12.66 kN Normal force = 12.55 kN Normal force = 12.57 kN 0 5 10 15 20 25

Shear force (kN)

Normal force = 12.58 kN

0

Shear force (kN)

Shear displacement (mm)

(c)
30 25 20 15 10 5 0 0 5 10 15 20 25 Shear Displacement (mm) Normal force = 18.22 kN Normal force = 18.18 kN Normal force = 18.05 kN
30 25 20 15 10 5

(d)

Shear force (kN)

Shear force (kN)

Normal force = 18.03 kN 0 0 5 10 15 20 25 Shear displacement (mm)

(e)
30 25 20 15 10 5 0 0 5 10 15 20 25 Shear Displacement (mm)
30 25 20 15 10 5

(f)

Shear force (kN)

Normal force = 23.55 kN

Shear force (kN)

Normal force = 23.37 kN 0 0 5 10 15 20 25 Shear displacement (mm)

(g)

(h)

Figure 4: Shear force-displacement curves with (b, d, f, and h) and without (a, c, e and g) shear pins.

40

40

Vp N tan 43.03 4.47

Peak capacity
30

Vp N tan 32.20 7.89

30

Peak capacity

Shear force (kN)

20

Shear force (kN)

20

10

Capacity @ 7 mm displacement

10

Vs N tan 44.11
0 0 10 20 Normal force (kN) 30 40

0 0 10 20 Normal force (kN) 30 40

(a)

(b)

Figure 5: Shear capacity envelopes of I blocks (a) without and (b) with steel pins. Proceedings of ASTM Symposium on Geosynthetic Soil Reinforcement Testing, San Antonio, Texas, USA. Bathurst, R. J. & Simac, M. R. (1994) Geosynthetic reinforced segmental retainingwall structures in North America. Proceedings of the Fifth International Conference on Geotextiles, Geomembranes and Related Products, Singapore, 1-41. Bathurst, R. J., and Simac, M.R. (1997) Design and performance of the facing column for geosynthetic reinforced segmental retaining walls. In J. W. Balkema, (ed), International symposium on mechanically stabilized backfill. Denver, Colorado. Bathurst, R. J., Althoff, S. and Linnenbaum, P. (2008) Influence of test method on direct shear behavior of segmental retaining wall units, Geotechnical Testing Journal, 31, 1-9. Elias, V., Christopher, B. R., and Berg, R. R. (2001) Mechanically stabilized earth walls and reinforced soil slopes "Design & construstion guidelines", FHWA-NHI-00-043, Washington D.C., National Highway Institute. Lee, C. H. (2000) Design and construction of a 9.6m high segmental wall. Proceedings of Secend Asian geosynthetics conference, Kuala Lumpur, Malaysia. NCMA (1997) Design manual for segmental retaining walls, Herndon, Virginia, National Concrete Masonry Association (NCMA). Soong, T. Y., and Koerner, R.M. (1997) On the required connection strength of geosynthetically reinforced walls, Geotextiles and Geomembranes, 15, 377- 393.

(b) (a)

(c) Figure 6: Photographs of common failure patterns at joints (a) Spalling at bottom blocks, (b) spalling at top block, and (c) Triangular crack at bottom block.

REFERENCES
ASTM D 6916-03 Standard test method for determining the shear strength between segmental concrete units, West Conshohocken, PA, USA, ASTM International. ASTM D 6916-06c Standard test method for determining the shear strength between segmental concrete units, West Conshohocken, PA, USA, ASTM International. Buttry, K. E., Mccullough, E. S., and Wetzel, R. A. (1993) Laboratory evaluation of connection strength of geogrid to segmental concrete units, Washington, DC 20001 USA. Bathurst, R. J., and Simac, M.R. (1993) Laboratory testing of modular concrete block - geogrid facing connections.

AUTHOR BIOGRAPHIES Md. Zahidul Islam Bhuiyan is a postgraduate student at the Department of Civil Engineering, Faculty of Engineering, University of Malaya, and Malaysia. He graduated from the Department of Civil Engineering at Bangladesh University of Engineering and Technology (BUET), Bangladesh in 2009. His research interests include geotechnical engineering. He can be reached at mdzibhuiyan@gmail.com. Faisal Hj Ali is a professor at the Department of Civil Engineering, Faculty of Engineering, National Defense, and Malaysia. He received a doctoral degree in Geotechnical Engineering from University of Sheffield, United Kingdom in 1984. His teaching and research interests include unsaturated residual soils, ground improvement techniques, slope instability, foundation engineering, and reinforced earth. He can be reached at fahali@gmail.com. Firas A. Salman is a senior lecturer at the Department of Civil Engineering, Faculty of Engineering, University of Malaya, and Malaysia. He received a doctoral degree in Geotechnical Engineering from Baghdad University, Iraq. His teaching and research interests include foundation analysis & design, soil investigation, and ground improvement. He can be reached at firasalman@hotmail.com.