Behaviour of interlocking mortarless hollow block walls under in-plane loading.
Mortarless structural wall systems constructed using blocks with interlocking keys are an attractive alternative to conventional block walls. The main feature of the interlocking hollow block mortarless system is the elimination of mortar layers and the blocks are interconnected through interlocking keys (protrusions and grooves). The goal in interlocking system is to assure efficient construction forming with well-aligned masonry structures even without skilled masons. In this kind of construction, dry joints will be formed between the block courses. Therefore, no render or finishing material is applying in this mortarless masonry.
There are several attempts to develop interlocking hollow blocks in different parts of the world. In Putra Block system, the blocks are stacked on one another and three-dimensional interlocking protrusions are provided in the blocks to integrate the blocks into walls (Safiee et al. 2011, Thanoon et al. 2004). The structural behaviour of this system is assessed by performing experimental tests under axial and eccentric loads (Fares 2005 and Najm 2001). Based on the available experimental results, strength correlation between individual blocks, prisms and basic wall panels of interlocking mortarless hollow block masonry are developed (Jaafar et al. 2006). Further analytical and experimental studies are conducted on the interlocking mortarless masonry system (Alwathaf et al. 2005; Alwathaf 2006; Jaafar,Thanoon, et al. 2006; Safiee et al. 2011; Thanoon et al. 2008a, 2008b). Masonry walls are usually subjected to simultaneous gravity load and lateral loading resulting from wind and/or seismic excitation. However, limited researches are carried out on to understand the structural behaviour of interlocking mortarless masonry walls under in-plane lateral loading (Alwathaf 2006; Alwathaf et al. 2005). As such, additional research in this field will improve the existing knowledge and understanding of its complex behaviour. This paper presents test results of full scale interlocking mortarless walls subjected to in-plane vertical and lateral loading. It also explores the possibility of using simple equilibrium models to estimate the in-plane lateral load capacity of the walls tested.
2. Experimental programme
A total of five unreinforced interlocking mortarless wall panels were tested under prescribed vertical pre-compressive loading ranging between 1 and 4 N/[mm.sup.2] to consider the gravity load and load from the upper storeys. This load was kept constant throughout the test. The in-plane lateral loads were gradually applied at one of corner edge of wall until it reached failure in order to evaluate the in-plane structural behaviour. The wall panels were constructed using Putra blocks and were assembled without mortar layers and without reinforcements. The dimensions and pre-compressive loads for the five walls were summarised in Table 1.
2.1. Fabrication of wall panels
All the wall panels were built according to the geometry shown in Figure 1. The wall panels were constructed using Putra block through running bond pattern. The Putra hollow blocks have three different configurations known as stretcher block, corner block and half block. Each block was incorporated with interlocking key at its top. Table 2 shows the details of each block. The average values for the mechanical properties of individual block units based on 18 samples for each unit. The stretcher blocks commonly used as a main unit in the construction of the walls. The corner block unit was used to fit at the end of the walls, while the half block was used to complete the courses of the wall so that vertical joints will be staggered. Figure 2 shows the arrangement of each block to form a prism which utilises the function of interlocking key in aligning the courses and locking them in their position.
2.2. Test Set-up and Instrumentations
The wall panels were constructed in a steel frame equipped with vertical and horizontal hydraulic jacks. The vertical loading was distributed on the top of wall through a distribution I-shape steel beam having an overall depth of 305 mm and flange width of 150 mm. The horizontally mounted hydraulic jack was used to apply lateral loads at the corner top edge of the wall. To avoid local failure of masonry unit at the point of application of lateral load, a 50 mm thick steel plate was used to transfer the load from the jack. The first course (lowest course) of the wall was horizontally restrained as shown in Figure 1, in order to avoid any sliding effects of the wall panel during lateral loading. The in-plane lateral displacements were measured by linear variables displacement transducer (LVDT) at three different height levels which placed at top, middle and base of the walls tested as shown in Figure 3. Another LVDT set were also placed on the wall surface to monitor any out of plane deformations at three different locations on walls: base, middle and top of walls. The strains in walls tested were measured by strain gauges at several locations to monitor its development. Figure 3 shows the experimental set-up and locations of LVDTs and strain gauges.
2.3. Loading phase
The experimental testing was conducted in two main stages. Initially, a vertical compressive stress was applied by means of the vertical hydraulic jack until the prescribed stress level was imposed on the wall. This prescribed compressive stress maintained at designated level throughout the test. Three different pre-compressive stresses of 1.0, 2.0 and 4.0 N/[mm.sup.2] were applied to the walls, as shown in Table 1. The lateral load was then applied in a small force increment using the horizontal hydraulic jack until complete failure of the wall achieved.
3. Experimental results and discussions
Test results including load-displacement, the interaction between compressive stress and shear strength of walls and failure modes were collected and presented below.
3.1. In-plane lateral displacement
Figure 4 shows the in-plane lateral load vs. in-plane lateral displacement of the five wall panels tested. Lateral displacements were taken at the top of the walls. The in-plane lateral displacements for every two replicates (W1(I) and W1(II); and W2(I) and W2(II)) were similar, indicating high quality control of the construction and testing of walls. Generally, the lateral-load and displacement were characterised by nonlinear behaviour, reflecting the interaction between the block units and the mortarless dry joints during loading.
The variation in displacement behaviour among walls tested depicted in Figure 4 was contributed by the different levels of pre-compressive stress applied to each wall. It also indicates that the mortarless masonry assemblages exhibit a nonlinear behaviour with varying stiffness upon the in-plane loading. As the walls have dry joints in horizontal and vertical directions, the zigzag pattern of displacement depicted in Figure 4 can be attributed to the successive close-up of the vertical dry joints.
Figure 4 shows that the main characteristics of the in-plane lateral displacements were significantly influenced by the value of pre-compressive stress applied to the wall. At early stage, observation on the same applied lateral load, W1 series exhibits highest in-plane lateral displacement compared to W2 and W3 series. In fact, Putra Blocks allowed approximately 2 mm tolerances in horizontal movement before interlocking keys may act. This shows that the interlocking between the blocks played its role after a slip of about 2 mm. Figure 4 also indicates that the walls exhibited higher stiffness for higher pre-compressive stress level except for W4 where for this series the higher pre-compressive load is tend to increase the lateral displacement. After applying more than approximately 50% of the maximum load of each wall, the in-plane lateral displacement rapidly increases. This load vs displacement pattern of this mortarless wall was comparative to the mortared wall (Velmelfoort and Raijmakers 1993). The pre-compressive load is also able to increase the lateral load carrying capacity of walls, caused by large friction forces developed in the mortarless dry joints between block courses.
3.2. Failure modes
Generally, the failures of the wall subjected to in-plane lateral load were shown by diagonal shear cracks and/or moderate toe crushing. For low pre-compressive vertical load level, the wall failure was dominated by diagonal shear cracks. The shear failure was characterised by bed joints sliding and vertical joints opening, contributing to the development of stepped diagonal opening without visible cracking in the block units as shown in Figure 5(a) and (b) for W1 and W2. However, the increasing of pre-compressive stress level contributed to the visibility of cracking of the block unit which observed in W2. These cracks were initiated by tension splitting of masonry in the compression strut formed in the wall as also observed by Voon and Ingham (2007). This type of failure mode produced due to the existence of vertical pre-compressive stress. This failure mode was similar to that exhibited by the dry joint system tested by Lourenco et al. (2005).
For higher pre-compressive vertical stress level, the wall (W3) failed due to toe crushing of blocks assemblage within the wall as shown in Figure 5(c). Since high pre-compressive vertical stress applied to the wall, it delayed any possible sliding along the bed joints but caused crushing at toe. Cracking through the block unit was also noticeable. Toe or corner crushing failure of the wall occurred due to the principal compressive stresses reaching the diagonal compressive strength of block units.
3.3. Pre-compressive stress and shear strength relationship
The main outcome from the in-plane wall testing investigation is the development of a relationship between average pre-compressive stress and shear stress at failure, where the average shear stress represents the maximum lateral load divided by the horizontal cross-section of the walls. The pre-compressive vertical stresses and maximum lateral loads at failure for all walls tested were summarised in Table 3. As mentioned previously, the maximum lateral load increases with the increase of the pre-compressive level which best presented using Mohr-Coulomb formulation as simplification approach where the criterion indicates that the shear strength increases as the pre-compressive stress increases. In Mohr-Coulomb formulation, the related parameters represent this behaviour will be the shear stress at failure, [tau], friction coefficient in joint interface, [mu] and the cohesion, c. Figure 6 shows the relationship between the pre-compressive vertical stress and lateral shear stress obtained from the current experimental investigation compared to the studies by LourencoLourenco et al. (2005) and Velmelfoort and Raijmakers (1993) based on linear regression lines to fit test results obtained from each investigation. Figure 6 also indicates that the mortarless wall system was able to resist the shear stress induced by the in-plane lateral load with the assistance of the pre-compressive vertical loads. In Figure 6, shear stresses ([tau]) were obtained by dividing the lateral loads by normal area of bedded face-shell which is calculated as the wall length (1.5 m) times the wall effective thickness. The effective thickness was the net thickness of the face shells only excluding the hollow part of the block. The linear regression computed for the five walls shows a good approximation to the experimental data with a correlation coefficient [R.sup.2] = 0.826. The linear regression relationship between shear and pre-compressive stresses at failure was:
[tau] = 0.2261[sigma] + 0.31 (1)
where the values of 0.2267 represent the friction coefficient, while 0.31 represents the cohesion of the proposed mortarless system. The approximation of linear regression resulted in high cohesion which gives the strength value at zero pre-compressive stress. As a comparison, results from different system of dry joints without interlocking masonry (Lourenco et al. 2005) were also plotted in Figure 6. From the comparison, it can be observed that the friction coefficient produced by proposed mortarless system was less (0.2267) than available dry joint system (0.3235). This obtained friction coefficient also far lesser compared to bonded (mortared) masonry which provide 0.6 and 0.334 for friction coefficients obtained from BSI (1992) and Velmelfoort and Raijmakers (1993), respectively. This was due to irregularity interface due to roughness of block surface which effects the friction resistance of mortarless joints. The friction resistance of bed joint area was not fully utilised, which causes stress concentration at localised areas as observed in Figure 7. Therefore, the unevenness of bed joint significantly decreases the shear strength of the system. However, improvement can be observed in term of cohesive value where the proposed system produced higher compared to available dry joint system of Louren$o et al. (2005).
In addition to this, the effectiveness of mortarless system could be improved by using higher pre-compressive load for both systems with interlocking and without interlocking masonry unit. It was important to note that the imperfection of the block bed is highly affecting the shear strength of the system as it causes a reduction in the contacted areas (Alwathaf 2005).
4. Strut-and-tie model for in-plane loaded masonry walls
There were various failure modes observed in masonry walls under pre-compressive and in-plane lateral loads. The diagonal shear cracking failure mode observed in W1 and W2 series, combined with toe crushing observed in W3 would be best represented by strut and tie models. The strut-and-tie model provides a rational approach by representing a complex structural member with an appropriate simplified truss model. It has not been widely applied to masonry structures in comparisons to reinforced concrete structures. In this section, the strut-and-tie approach will be utilised for the prediction of in-plane lateral load capacity of walls tested.
The strut and tie model was principally based on the failure mechanism observed in the physical wall testing. The equilibrium conditions at nodes were then employed in calculating forces in struts and ties. Figure 8(a) shows an idealised strut-and-tie model for walls failed in diagonal shear, similar to W1 and W2. This model was a combination between parallel and fan struts, similar to the model proposed by Roca (2006).
Considering equilibrium of forces in parallel struts and in-plane loads as shown in Figure 9(b), the following relations between forces were govern:
[H.sub.1] = [S.sub.1] sin [[beta].sub.1] (2)
[V.sub.1] = [S.sub.1] cos [[beta].sub.1] (3)
where [S.sub.1] is the force in parallel struts, [V.sub.1] is the in-plane vertical force acting on the parallel strut zone, [H.sub.1] is the in-plane lateral load required for horizontal force equilibrium and [[beta].sub.1] is the parallel strut angle relative to vertical axis as shown in Figure 8(b).
The in-plane vertical force [V.sub.1] acting on the parallel strut zone can be written in the following form:
[V.sub.1] = V(b - [b.sub.f])/b (4)
where V is the total in-plane vertical load, b and [b.sub.f] are the wall length and fan length, respectively, as shown in Figure 8(a).
Combining Equations (2)-(4), the in-plane horizontal force [H.sub.1] may be written as below:
[H.sub.1] = [V(b - [b.sub.f])/b]tan [[beta].sub.1] (5)
For the case of fan strut, an average angle [[beta].sub.2] of fan strut was chosen to simplify the complexity of fan strut capacity evaluation. Considering the free body diagram for fan strut shown in Figure 8(c) and by following similar steps to the parallel strut capacity calculation, the horizontal component [H.sub.2] of the fan strut capacity was obtained below:
[H.sub.2] = [V.sub.2] tan [[beta].sub.2] (6)
where [S.sub.2] is the fan strut capacity, [V.sub.2] is the vertical component of the fan strut capacity, [[beta].sub.2] is the average fan strut angle relative to the vertical axis. As [V.sub.2] = V[b.sub.f]/b, [H.sub.2] may be written in the following form:
[H.sub.2] = [V[b.sub.f]/b]tan [[beta].sub.2] (7)
Therefore, the total in-plane lateral load capacity of walls was:
[mathematical expression not reproducible] (8)
At ultimate condition and for mortarless/dry joint case, [[beta].sub.1], [[beta].sub.2] and the angle of internal friction [mu] of masonry were related to each other with the following relation (Roca 2006):
[[beta].sub.2] = [[beta].sub.1]/2 = [mu]/2 (9)
The top length [b.sub.f] of the fan strut is calculated from:
[b.sub.f] = a + h tan [[beta].sub.1] (10)
where h is the wall height and a is the node dimension at the toe where crushing occurred. a can be calculated from:
a = [V.sub.2]/t[f.sub.c] = V[b.sub.f]/bt[f.sub.c] = [gamma][b.sub.f] (11)
where [f.sub.c] is the compressive strength of masonry wall, t is the masonry wall thickness and [gamma] = V/(bt[f.sub.c]). Using Equations 9 and 11, the value of [b.sub.f] may be written in the following form:
[b.sub.f] = h tan [mu](1/[1 - [gamma]]) (12)
Therefore, substituting the value of [b.sub.f] in Equation (8), the total in-plane lateral load capacity of walls can be calculated for this strut-and-tie model.
Table 4 presents the prediction of the in-plane lateral load capacity obtained from Equation 8 for W1 and W2. The ratios for the prediction to experimental in-plane lateral load capacities for W1 and W2 were 0.91 and 0.90, respectively. These results indicate that the proposed strut and tie model were able to predict the in-plane lateral load capacity with reasonable accuracy.
An alternative strut-and-tie model to represents the failure occurred in W3 shown in Figure 9 consists of a main diagonal strut with nodes at the top and bottom of the wall where the in-plane lateral loads applied. These nodes represent the crushing failure of the block units. From the wall geometry, the diagonal strut inclination [theta] with the horizontal axis can be calculated from:
tan [theta] = [h - c]/[b - a] (13)
where h and b are the vertical and horizontal dimensions of wall, c and a are the node dimensions as shown in Figure 9. The value of c may be considered the same as the end plate height through which the in-plane lateral load applied. The angle [theta] of the compressive strut for the walls tested in this investigation is [theta] = 47[degrees].
Due to masonry crushing at the toe region, the compressive stresses [[sigma].sub.s] at the wall toe have achieved the effective masonry compressive strength, [f.sub.c] as below:
[[sigma].sub.x] = v[f.sub.c] (14)
where [f.sub.c] is the masonry compressive strength and v is the effectiveness factor of compressive strength of masonry. Several researchers (Foraboschi and Vanin 2013; Lourenco, Alvarenga and Silva 2006; Patrick 2015) were developed strut and tie models for designing masonry walls. The effectiveness factor v proposed by Patrick (2015) will be used in the current investigation as below:
v = 0.8[[beta].sub.s][[beta].sub.a] (15)
where [[beta].sub.s] is the strut efficiency factor, taken as 0.75 and [[beta].sub.[alpha]] is the strut inclination factor, taken as 0.67 for strut angle [greater than or equal to]37.5[degrees] (Patrick (2015)).
Therefore, the in-plane lateral load H for toe crushing failure can be calculated by multiplied [[sigma].sub.x] and the bearing area as below:
H = [[sigma].sub.x] c t = v[f.sub.c] c t (16)
where t is the masonry wall thickness. The prediction of Equation (16) for the in-plane lateral load capacity of W3 was 120.0 kN using the value of obtained from Equation (15), which agrees well with the experimental value of 136.8 kN.
The behaviour of interlocking mortarless load bearing hollow block walls under combined in-plane vertical and lateral loadings was investigated. The mortarless walls subjected to in-plane loads exhibited a nonlinear lateral displacement with increasing stiffness upon increasing the pre-compressive vertical load applied on the walls tested. The pre-compressive vertical load was able to increase the lateral load carrying capacity of walls which caused by the higher friction forces that develop in the mortarless dry joints. It was also shown that, when the pre-compressive load increases, shear resistance also increases.
The failure modes of mortarless walls under in-plane loading were controlled by diagonal shear failure or moderate toe crushing depending on the level of the pre-compressive vertical loading. Shear failure characterised by the development of sliding along bed joints and opening of vertical joints contributed to a stepped diagonal opening without visible cracking in the blocks for low and moderate pre-compressive loads. The power relation is the best fit relation that takes into consideration the zero cohesion of dry joints as well as the material failure at the higher pre-compressive loads for the interlocking mortarless wall system. The predictions from the developed strut-and-tie model were very close to the experimental in-plane lateral load capacity of masonry walls tested.
Received 21 July 2017
Accepted 23 January 2018
No potential conflict of interest was reported by the authors.
Alwathaf, A. H., W. A. M. Thanoon, J. Noorzaei, M. S. Jaafar, and M. R. Abdulkadir. 2005. "Shear Characteristic of Interlocking Mortarless Block Masonry Joints." Masonry International, Journal of the British Masonry Society 18 (3): 139-146.
Alwathaf, A. H. 2006. "Development of Finite Element Code for Non-linear Analysis of Interlocking Mortarless Masonry System. PhD thesis, Civil Engineering Department, University Putra Malaysia.
British Standard Institution. 1992. Code of Practice for Use of Masonry Part 1: Structural Use of Un-reinforced Masonry. BS 5628: Part 1. London: BSI.
Fares, A. Shehab. 2005. "Structural Behaviour of Interlocking Hollow Block Panel with Stiffener Subjected to Axial and Eccentric Load." Master thesis, Universiti Putra Malaysia.
Foraboschi, P., and A. Vanin. 2013. "Nonlinear Static Analysis of Masonry Buildings Based on a Strut and Tie Modelling." Soil Dynamics and Earthquake Engineering 55: 44-58.
Jaafar, M. S., W. A. Thanoon, A. M. S. Najm, M. R. A. Kadir, and A. A. A. Abang. 2006. "Strength Correlation Between Individual Block, Prism and Basic Wall Panel for Load Bearing Interlocking Mortarless Hollow Block Masonry." Construction and Building Materials 20: 492-498.
Lourenco, P. B., D. V. Oliveira, P. Roca, and A. Orduna. 2005. "Dry Joint Stone Masonry Walls Subjected to In-Plane Combined Loading." Journal of Structural Engineering 131 (11): 1665-1673.
Lourenco, P. B., R. C. Alvarenga, and R. M. Silva. 2006. "Validation of a Simplified Model for the Design of masonry Infilled Frames." Masonry International. 19 (2006): 15-26.
Najm, A. M. S. 2001. "Structural Behaviour of Load-bearing Interlocking Hollow Block Masonry." Master thesis, Universiti Putra Malaysia.
Patrick, D. 2015. "Shear strength prediction methods for grouted masonry shear walls." PhD thesis, Brigham Young University.
Roca P., 2006. Assessment of Masonry Shear-walls by simple equilibrium model, Construction & Building Materials, 20:229-238
Safiee, N. A., M. S. Jaafar, A. H. Alwathaf, J. Noorzaei, and M. R. Abdulkadir. 2011. "Structural Behavior of Mortarless Interlocking Load Bearing Hollow Block Wall Panel under Out-of-plane Loading." Advances in Structural Engineering 14 (6): 1185-1196.
Thanoon, W. A., M. S. Jaafar, M. R. A. Kadir, A. A. A. Abang, D. N. Trikha, and M. S. N. Amad. 2004. "Development of an Innovative Interlocking Load Bearing Hollow Block System in Malaysia." Construction and Building Materials 18 (6): 445-454.
Thanoon, W. A., A. H. Alwathaf, J. Noorzaei, M. S. Jaafar, and M. R. Abdulkadir. 2008a. "Finite element Analysis of interlocking mortarless hollow block masonry prism." Computers & Structures 86 (6) March: 520-528.
Thanoon, W. A., A. H. Alwathaf, J. Noorzaei, M. S. Jaafar, and M. R. Abdulkadir. 2008b. "Non-linear Finite Element Analysis of Grouted and Un-Grouted Hollow Interlocking Mortarless Block Masonry System." Engineering Structures 30: 1560-1572.
Velmelfoort, A. Th. and T. M. J. Raijmakers. 1993. Deformation Controlled Tests in Masonry Shear Walls, Part 2. Report TUE/BK0/93.08. Eindhoven: Eindhoven University of Technology.
Voon, K. C., and J. M. Ingham. 2007. "Design Expression for the In-plane Shear Strength of Reinforced Concrete Masonry." Journal of Structural Engineering 133: 706-713.
Nor Azizi Safiee (a), Noor Azline Mohd Nasir (a), Ashraf Fawzy Ashour (b) and Nabilah Abu Bakar (a)
(a) Faculty of Engineering, Department of civil Engineering, university Putra Malaysia, Malaysia; (b) School of Engineering, university of Bradford, Bradford, UK
CONTACT Nor Azizi Safiee ([mail]) firstname.lastname@example.org
Caption: Figure 1. Geometry and boundary condition of tested specimen.
Caption: Figure 2. Assemble of interlocking blocks to form a prism.
Caption: Figure 3. Test set-up and instrumentation.
Caption: Figure 4. in-plane lateral load vs lateral displacement.
Caption: Figure 5. Failure pattern of the wall panels.
Caption: Figure 6. Relationship between pre-compressive stress and shear stress.
Caption: Figure 7. Surface roughness of face shell bed joint of block unit.
Caption: Figure 8. Proposed strut and tie model for W1 and W2.
Caption: Figure 9. Proposed strut and tie model for W3.
Table 1. Detail of test wall panels. Specimen Panel size (H x L) Pre-compressive stress (N/[mm.sup.2]) W1 (I) 1600 x 1500 1.0 W1 (II) 1600 x 1500 1.0 W2 (I) 1600 x 1500 2.0 W2 (II) 1600 x 1500 2.0 W3 (I) 1600 x 1500 4.0 Table 2. Details of putra hollow block. Block type Dimensions (L x Comp. strength H x t) mm [f'.sub.x](N/[mm.sup.2]) Stretcher 300 x 200 x 150 22.85 (COV: 16.5) interlocking key Half 150 x 200 x 150 22.02 (COV: 16.4) Corner 300 x 200 x 150 23.67 (COV: 13.7) Block type Tensile strength, [f'.sub.t](N/[mm.sup.2]) Stretcher 2.06 (COV: 5.31) interlocking key Half 2.16 (COV: 9.72) Corner 2.79 (COV: 15.36) Table 3. Vertical and maximum lateral loads measured. Specimen Pre-compressive Max lateral stress (N/[mm.sup.2]) Load (kN) W1 (I) 1.0 58.30 W1 (II) 1.0 52.05 W2 (I) 2.0 93.72 W2 (II) 2.0 117.20 W3 (I) 4.0 136.8 Specimen Max shear stress (N/[mm.sup.2]) W1 (I) 0.486 W1 (II) 0.434 W2 (I) 0.781 W2 (II) 0.977 W3 (I) 1.14 Table 4. Prediction of strut-and-tie model for W1 and W2. Wall V (kN) b (mm) h (mm) t (mm) [f.sub.c] (N/ [mm.sup.2]) W1 120 1500 1600 150 10.0 W2 240 1500 1600 150 10.0 Wall [[beta].sub.1] [H.sub.exp] [H.sub.predict] (kN) (kN) W1 37 55.2 50.01 W2 43 105.5 95.23 Wall [H.sub.predict]/ [H.sub.exp] W1 0.91 W2 0.90
|Printer friendly Cite/link Email Feedback|
|Author:||Safiee, Nor Azizi; Nasir, Noor Azline Mohd; Ashour, Ashraf Fawzy; Bakar, Nabilah Abu|
|Publication:||Australian Journal of Structural Engineering|
|Date:||Apr 1, 2018|
|Previous Article:||RC walls in Australia: seismic design and detailing to AS 1170.4 and AS 3600.|
|Next Article:||Field studies on sleeper deflection and ballast pressure in heavy haul track.|