 |
 |
Buckling Capacity of Concrete Piles in Sand
Senior Lecturer, Department of Civil Engineering, T. Sivasamy Graduate Student, Department of Civil Engineering, P. Parameswaran Professor, Department of Civil Engineering, |
ABSTRACT
Experimental investigation on the buckling behaviour of concrete piles, 40mm·50mm and 2200mm high, partially embedded in dry sand subjected to axial loads and eccentric loads have been carried out in a sand box of size 600mm·600mm·1500mm. Concrete piles having unembedded length to embedded length ratios of lu/l=0.40, 0.50 and 0.55, were subjected to axial loads. An eccentricity to least section size ratios of e/D=0.5, 1.0 and 1.5, with constant embedment length of 1200mm, were used under eccentric loads. Dry river sand of specific gravity 2.62, uniformity coefficient 1.4, emax= 0.68, emin= 0.47 has been used as foundation medium. Based on the experimental results, the influence of unembedded length and eccentricity have been investigated on the variation of load-deflection along the entire length for each increment of load up to failure, ultimate load and critical load of slender concrete pile. Salient conclusions are reported from the study.
Keywords: Partially Embedded Pile, Concrete Pile, Axial loads, Eccentric loads, Ultimate load, Critical load, Sand.
INTRODUCTION
Pile foundations are often recommended for deep strata of weak soils. Piles are structural members used to transfer the heavy loads to the foundation. Fully embedded pile in a very weak soil is subjected to buckling. Piles used for water front structures such as jetties, trestles etc., which is partially embedded are vulnerable to buckling. In practice, eccentricity may occur during pile installation and hence such pile is critical. Scanty experimental results on buckling of partially embedded concrete pile subjected to axial loads and eccentric loads are available. Field tests though highly desirable, uncertain soil conditions, instrumentation limitations and involvement of huge funds restrict the scope of field tests. In the absence of resources and scope of field tests, laboratory tests conducted on piles in relatively consistent foundation medium prepared under controlled condition with well instrumented pile may serve the purpose to some extent. Therefore, the requirement was felt to undertake laboratory testing to study the buckling behaviour of partially embedded pile under axial loads and eccentric loads in sand.
BRIEF REVIEW OF LITERATURE
The brief review of literature is restricted to the buckling aspects of piles. The purpose is to summarize the available literature in the point of view of partially embedded piles in general and concrete piles in particular.
Several laboratory tests have been conducted by Grandholm (1929), Walter (1951), Bergfelt (1957), Golder and Skipp (1957), Lee (1968), Ramsamooj (1973), apart from few field tests by Glick (1948), Brandtzaeeg and Harboe (1957), and some methods of analysis have been proposed by Forssell (1918), Hetenyi (1946), Davisson (1963), Francis (1964), Toakley (1965) Reddy and Valsangar (1970), Mazindrani and Sastry (1972), Reddy and Panigraphi (1973), Prakash (1987), Fleming et al (1992), Gabr et al (1994), West et al (1997), Heelis et al (1999) Kalaga (2001), Lin and Chang (2002) Petrakis (2005), for the piles subjected to axial loads and that too for the fully embedded condition. Information on partially embedded pile is limited ( Forssell (1926), Sulocki (1955), Davisson and Robinson (1965), Gabr et al (1997), Heelis et al (2001)).
Limited investigations on concrete piles under axial loads are available. (Hromadik (1961), Gouvenot (1975)). However investigations on piles under eccentric loads are scanty. ( Klohn and Hughes (1964)).
Therefore, experimental investigations have been carried out to study buckling behaviour of partially embedded concrete piles under axial loads and eccentric loads in sand.
EXPERIMENTAL SET UP AND TEST PROCEDURE
Experimental Set up
Figure 1 shows the arrangement of experimental setup.
Figure 1. The schematic arrangement and photograph of the experimental setup
Sand Box
The tests were conducted in a specially fabricated wooden box of size 600mmx600mm and 1500mm high, with three detachable arrangements such as base and other two L-shape side attachments. (like form work)
Foundation Medium
Uniform dry river sand was used as a foundation medium. Selection of sand as foundation medium was made because its behaviour is free from time effect and consistent reproducible medium can be obtained reasonably well. The specific gravity and uniformity coefficient of the material were 2.62 and 1.4 respectively. The limiting void ratios were emax= 0.68, emin= 0.47 corresponding minimum and maximum dry densities were 1.599g/cc and 1.782g/cc respectively. The placement density during the test was 1.619g/cc for loose packing. (R.D = 30%)
Test Specimens
The pile specimens had rectangular cross-section of 40mm·50mm, 2200mm height with ends widened and provided with a suitable bearing plate and reinforcement, to ensure uniform distribution of loading and to prevent premature failure at the supports. Rectangular section of specimens was selected to maximize the effects of buckling without any lateral restraint to guide the specimen. The longitudinal reinforcement of the specimen consisted of four mild steel bars of 4mm diameter arranged symmetrically. The transverse reinforcement comprised of 3mm diameter spaced at 40mm center to center.
Local concrete materials such as cement, sand, crushed gravel (maximum size varying 4.75mm to 6mm) have been used to cast the test specimens. The proposed mix was designed to develop characteristic cube strength of 25MPa.
Partially embedded pile specimens in sand with variable unembedded lengths 1200mm, 1100mm, 900mm, having lu/l = 0.55, 0.50, 0.40 respectively, are subjected to axial loads. And partially embedded pile specimens with constant unembedded length of 1200mm, on variable eccentricity of 60mm, 40mm, 20mm, having e/D = 1.5, 1.0, 0.5 respectively, are used.
Test Procedure
The specimens were tested up to failure using AMSLER Universal Testing Machine (UTM) of 1000kN capacity. The testing machine was suitably modified to allow a maximum specimen height of 2200mm. Selection of UTM for loading the specimen was made because it keeps the specimen as well as assembly set up intact at the time of large lateral deflections of slender specimen, especially under eccentric loads and relatively better control over the test. Each specimen was placed between the machine heads with ball-socket arrangement at top and hinge at bottom, plumbed and secured in the plumb position for axial loads. The required load eccentricity was achieved using widened end with steel bearing plate.
After securing the position of the specimen as well as sand box in place, weighed mass of sand, obtained based on the placement density, was poured manually for every layer of 150mm thick, uniformly rodded to achieve the required density of 1.619g/cc, ensured by checking the sand level with the available graduated level marks in the box.
Five Linear Variable Displacement Transducers (LVDT) were placed to measure the deflection, to effectively represent the deflection along the entire pile length, including three LVDT’s in the foundation medium, attached with deflection rods available in the specimen (fixed at the time of casting).
Specimens were loaded using the load control system of the machine. At every increment of loading, the deflections were recorded besides observing and marking cracks, if any.
EXPERIMENTAL RESULTS AND DISCUSSION
Load – Deflection Diagrams
The basic observations obtained from the experimental tests are applied load and lateral deflections apart from ultimate load of concrete pile, for axial loads and eccentric loads. From the tests, it was observed that the failure readily occurred above the foundation medium in all the tested piles, as expected. Furthermore, it was noticed that the pattern of failure was due to unbound deformation. Therefore, it is clear to establish from the experiments conducted that unembedded portion of the concrete piles is critical for buckling apart from deflection considerations. Hence, the load-deflection curves on the unembedded portion of the pile were presented as shown in Fig. 2 to Fig. 5.
Figure 2. Load versus Deflection at 250mm from top under axial load
Figure 3. Load versus Deflection at 550mm from top under axial load
Figure 4. Load versus Deflection at 250mm from top under eccentric load
Figure 5. Load versus Deflection at 550mm from top under eccentric load
Also, the deflection curves of the pile at various stages of loading were drawn, as shown in fig.6 to fig.11, so as to understand the trend in the deflected shape of the pile. It was observed that the amount of deflection reduces relatively as the pile approaches foundation medium at a given load and it was more pronounced as the eccentricity increases. It was also observed that the increasing load aggravates this trend especially on larger eccentricities and at the verge of failure, deflections of the pile below the foundation medium was considerably small comparing other portions of the pile and it decreases further as the embedment depth increases. It was also observed that the amount of deflection between piles as well as the difference in amount of deflection between unembedded and embedded portion of the pile increases significantly as the eccentricity increases. Therefore, it is noticed that the eccentricity controls the behaviour of such piles and it is critical for consideration and provides scope for further study taking into the account of material and geometric non-linearity of the concrete pile, since the term eccentricity is more applicable for piles other than columns.
Figure 6. Pile Lateral Deformation Curve (lu/l =0.55, axial load condition)
Figure 7. Pile Lateral Deformation Curve (lu/l =0.5, axial load condition)
Figure 8. Pile Lateral Deformation Curve (lu/l =0.40, axial load condition)
Figure 9. Pile Lateral Deformation Curve (e/D = 1.5, eccentric load condition)
Figure 10. Pile Lateral Deformation Curve (e/D = 1.0, eccentric load condition)
Figure 11. Pile Lateral Deformation Curve (e/D = 0.5, eccentric load condition)
Ultimate Load and Critical Load
Results obtained from the tests on the ultimate capacity of the concrete pile are given in Table 1 and Table 2; and for the sake of completeness, the experimental critical load for concrete pile was obtained, using the load-deflection observations on the unembedded portion of the pile, like column, based on the well known method suggested by Kwon and Hancock(1992) are also presented in the table.
Table 1. Test Results of Concrete Piles under Axial Load
| |||
| Embedment ratio |
Ultimate Load (kN) |
Maximum Deflection (mm) |
Experimental Critical Load (kN) |
| |||
| 0.55 | 55.0 | 9.4 | 43.2 |
| 0.50 | 92.0 | 6.9 | 62.8 |
| 0.40 | 122.0 | 2.4 | 72.4 |
| |||
Table 2. Test Results of Concrete Piles under Eccentric Load
| |||
| Eccentricity ratio |
Ultimate Load (kN) |
Maximum Deflection (mm) |
Experimental Critical Load (kN) |
| |||
| 1.5 | 21.0 | 24.1 | 19.7 |
| 1.0 | 25.0 | 12.2 | 23.9 |
| 0.5 | 57.0 | 4.8 | 56.0 |
| |||
Based on the available results, it is noted that the increase in unembedded length of the pile reduces the buckling capacity. It is noteworthy that eccentricity plays a significant role on the buckling capacity of the concrete pile, more importantly in partially embedded condition.
CONCLUSIONS
From the present investigation, the following main conclusions are drawn:
In general, buckling capacity of the partially embedded concrete pile largely depends on unembedded length of the pile.
Buckling capacity of the concrete pile is very critical for the combination of increasing unembedded length with larger eccentricities.
Deflection of the partially embedded pile decreases significantly in the embedded portion as the unembedded length of the pile increases.
Deflection of the partially embedded pile reduces drastically in the embedded length under large eccentricities.
REFERENCES
 | |
| © 2006 ejge | |