MODEL STUDIES ON IMPROVEMENT OF LATERAL CAPACITY OF PILES IN SAND

E.Saibaba Reddy, Professor of Civil Engineering, JNTUH, esreddy1101@gmail.com 

Rakesh Reddy.E, UG Student, JNTUH, rakesh15794@gmail.com

ABSTRACT: This paper presents the details of two series of experiments carried out on model piles embedded in sand. The test results are compared with the results obtained from an equivalent plain pile. The additions of attachments have shown considerable increase in the lateral capacity of a pile. For these piles, theoretical estimates are made and compared with the experimental results. The theoretical estimates and the experimental results are found to be in good agreement. The second part of the paper presents the details of the test series carried out on model vertical piles under oblique pull. Experiments were carried out on three model piles of different lengths. The load and the displacement of the pile head were observed till failure of the pile. The uplift capacity is analyzed as a function of obliquity of load. The theoretical estimates of the pile capacity under oblique pull were made.

Key word: Pull-out, oblique, attachments, pile, capacity.

1.0 INTRODUCTIONLateral resistance of pile increases with depth up to a critical depth. Beyond this depth, the lateral capacity of a pile does not increase significantly with depth. The lateral capacity of a pile, in such a situation, can be improved by increasing its lateral dimensions. Broms has suggested a few attachments to pile [1, 2] at shallow depths, to improve its capacity to lateral loads. So far no attempts appear to have been made, either in laboratory or in field, to examine the effect of these attachments on the lateral capacity of a pile. An experimental investigation was carried out on model piles to study the effect of four attachments, suggested by Broms, in improving the lateral capacity of a pile. A number of investigations on vertical piles under inclined downward forces were reported [3,4,5,6] however, only a few references are available on the behavior of piles under oblique pull [7,8,9] the second part of this paper presents the details of an experimental investigation carried out on vertical model piles under oblique pull. The tests were carried out under different oblique angles. The ultimate pullout capacity of pile under oblique load observed during experiments is compared with the theoretical estimates. In the rest of this paper the details of the two series of experiments are presented. The experimental results are compared with the respective theoretical estimates.

2.0 MODEL STUDIESModel laboratory studies were carried out on a number of model piles with and without attachment to improve the lateral capacity of piles. All tests were conducted on piles embedded in sand. The properties of sand used for the investigation, details of model piles used and the test procedure is explained below.

2.1 Soil Properties: The sand used in both the test series was a dry angular silica sand having D10 = 0.3mm and uniformity coefficient Cu = 2.8 was used. Direct shear test indicated a friction angle (Ø) of 35o for the initial porosity of n = 41% used in the tests, corresponding to a unit weight (γ) of 16 KN/m3 and a relative density of Dr = 0.6 (60%).

2.2 Model Studies on Improvement of lateral Capacity of Piles

For this investigation a total of 15 model piles were fabricated with four types (Type-1, 2, 3 and 4) of attachments. Four piles were fabricated with attachment placed at different depths, along the pile (Table 1). Besides these piles, three plain piles (without attachments) were fabricated with different (360, 480 and 600mm) lengths. All piles were fabricated with steel rod of 12mm diameter. The details of each model pile are presented below.

Table 1. Details of model testsPile Model Test Conditions Remarks

Plain pile D = 360, 480 and 600mm --Type – 1 a = 100, 200 and 300mm D = 600mmType – 2 a = 100, 200 and 300mm D = 600mmType – 3 a = 100, 200 and 300mm D = 600mmType – 4 D = 360, 480 and 600mm a = 50mm

Model pile type – 1In this, three 12mm diameter and 600mm long piles were fabricated; each was welded with a pair of two steel rods as shown in Fig. 1. The attachment was positioned at three elevations (a = 100, 200 and 300 mm) as shown in Fig. 1.

Model pile type – 2In this, three 12 mm diameter and 600 mm long piles were used. Each of these piles was welded with a cross beam, of 12 mm diameter and 180mm long, at different elevations as shown in Fig. 2.

Hook for loading

2d a 1

d

100

D=6

00

a= (d1+d2)/2 a=100, 200 and 300 Figure not to scale, All dimensions are in mm

(a) Front view

12 mm dia. steel rods

Weldin

3.5

(b) Top view

Fig.1.Model Pile Type-1

a

180

D=6

00

a=100, 200 and

300 Figure not

to scale All

dimensions are

in mm

(a) Front view12 mm

180

Fig.2. Model Pile Type-2

a 10

0

D=6

00

a=100, 200 and 300 Figure not to scale All dimensions are

in mm

(a) Front view

Vanes

(b) Top view

Fig.3. Model Pile Type-3

Model pile type – 3The third model was developed by welding four vanes as shown in Fig. 3. The vane dimensions and their locations are presented in Fig. 3.

Model pile type – 4The fourth model pile was fabricated by welding a large (40 mm) diameter rod of 100mm long over the 12 mm diameter pile as shown in Fig. 4. In this model, unlike other three models, different trials were made by changing the length of the pile as shown in Fig. 4. This was achieved by testing initially with a total length of 600mm then for the next trials the pile length was cut from the bottom, to obtain a total length of 480mm and 360mm as shown in Fig. 4.

Method of sand filling and pile installationThe sand was placed in a circular tank of internal diameter 1120 mm and 1200 mm high (Fig.5). The tank was filled with sand in layers of 100 mm thick. Each layer was compacted, with a rammer under a constant height of fall, to ensure a constant unit weight of 16 kN/m3. The density of sand was measured by placing a number of containers along the depth of the tank (Reddy, 1986).The pile was installed by jacking gradually into the sand up to the required embedded depth.

Dial GaugeHook for loading

1200

Pile

sand

1120

Pulley

Test

Load

BrickMasonry

D=6

00

100 40

D=360, 480 and 600 Figure not to scale All dimensions are in mm

(a) Front view

40 mm Dia

Figure not to scale All dimensions are in mm

Fig.5. Details of test tank

(b) Top view

Fig.4.Model Pile Type-4

Pile Loading After installation of the pile, a dial gauge (least count 0.002 mm) was positioned to measure the horizontal deflection of the pile head. The horizontal load was applied at the pile head by means of a wire rope passing over a pulley mounted on to the side of the tank as shown

in Fig. 5. The load on the pile head was increased in steps of about 10 N. The horizontal deflection of the pile head, under each load, was recorded with time. When the deflection under an applied load was ceased (< 0.01mm/minute), the load on the pile head was increased to the next value. The test was continued till the pile failed by showing continuous increase in the deflection under no increment or a small increment of load. After each test, the tank was emptied and refilled for the next test. The details of tests performed are presented in Table 1. 3.0 ANALYSIS OF TEST RESULTS

Lateral capacity of plain pileThe ultimate lateral capacity of pile is obtained from load-displacement curve, as the load corresponding to the start of the final linear portion of the curve [10].The theoretical pile capacity for a pile under horizontal load (Qh) is obtained from Eq. (1)

Qh = Fbγ D2eu Kb B (1)

Where γ is the unit weight of sand, Deu is the ultimate effective embedded depth of an equivalent rigid pile, Kb is the net lateral soil pressure coefficient, Fb is the lateral resistance factor (0.12) and B is the diameter of the pile. The effective depth ratio (D eu/D) for a given fixity condition at the pile head depends mainly on the relative pile stiffness Kr given by Eq. (2)

Kr = (EpIp) / (Es D4) (2)

Where EpIp is the flexural rigidity of pile and Es is the horizontal soil modulus at pile toe. In the present case, the value of Kr and the corresponding Deu/D for different values of pile depth (D) are presented in Table 2. For the values of Deu/D the Kb is estimated [11] . On substituting the values in Eq. (1), Q h for the three plain (360, 480 and 600 mm long) piles is computed (Table 2).

Table 2. Lateral capacity of plain piles (computed)Pile length, D (mm) Kr Deu/D Deu (mm) Qh (N)

360 63.6 x 10-4 0.90 324 41.8480 20.1 x 10-4 0.78 376 45.6600 8.3 x 10-4 0.70 422 57.4

Details of Modified pile types-1,2 and 3Table 3 and Fig. 6 shows the comparison of the lateral capacity of modified piles (Type-1, 2 & 3) with a plain pile having the same length. It can be observed from Fig. 6 and Table 3 that the attachments have significantly (20 to 124%) increased the lateral capacity of the pile. Among the models tried. Type-3 gave a maximum increase in pile capacity. This is expected because; the laterally projected area in Model Type-3 is large compared to the Model Type-1 and 2.The lateral capacity of a modified pile is computed by considering the increase in lateral resistance of the pile due to attachment, using Eq. (3).

Qhm = (Qh + Qh1 – Qh2) (3)

Where Qhm is the lateral capacity of the modified pile, Qh is the lateral capacity of the plain pile. Qhl is the lateral capacity of, extra projection than pile diameter, considering the attachment length equal to d1 (Fig.1) and Qh2 is the lateral capacity of the attachment

considering its length equal to d2. Where d1 is the depth below the sand surface up to the bottom of the attachment and d2 is the depth below the sand surface up to the top of the attachment (Fig.1). The ultimate lateral capacity of model piles, obtained theoretically and experimentally, with attachments Types-1,2,3 are compared in Fig.7. From Fig.7. it can be observed that the theoretical estimates are close to the experimental results. However, the estimated values are consistently lower than the experimental values. This deviation could be due to the fact that, in computing the lateral capacity of a modified pile, the increase in its stiffness due to the attachment was not accounted. Hence, the computed effective depth of pile will be less than the depth that is due to increased stiffness.

Table 3. Increase in the lateral capacity of pile due to attachments (observed) Pile model a = 100mm a = 200mm a = 300mm

Qhm(N)

%increase*

Qhm(N)

%increase*

Qhm(N)

%increase*

Type – 1 78.0 30.0 91.0 51.7 104.0 73.3Type – 2 71.5 19.2 84.5 40.8 97.5 62.5Type – 3 91.0 51.7 104.0 73.3 124.0 106.7 Note: * increase is with respect to a plain pile of 600mm long

Fig.6.Comparison of lateral Fig.7.Comparison of theoreticalcapacity of modified Pile and experimental results for

with an equivalent plain pile model piles Type-1, 2 and 3

Model pile Type-4The lateral capacity of the model pile Type-4 for different pile lengths are compared with that of an equivalent plain pile in Fig. 8. From Fig. 8 it can be observed that, the ultimate capacity of the modified pile maintains almost a constant amount of increase in the lateral capacity, over the plain pile, for all lengths. This result is also expected because in each of the trials there is a constant amount of increase in area and it is located at a constant position, from sand surface, for all trials. The lateral capacity of model pile Type-4 is estimated using Eq. (3) by taking Qh2 = 0. The theoretical estimates of the lateral capacity of model pile Type-4 are compared with the experimental results in Fig. 9. It can be seen that the theoretical estimates are agreeing well with the experimental results.

Fig.8.Comparison of lateral capacity of model pile Type-4 with an equivalent plain

pile.

Fig.9.Comparison of theoretical values with experimental results (Model pile Type-4).

4.0 PILES UNDER OBLIQUE PULL

Test detailsThe sand bed was formed in a circular tank of internal diameter 1120 mm and 1200 mm in depth (Fig.10). The sand was placed in layers as explained earlier. For this investigation three model piles 360 mm, 480 mm and 600 mm were fabricated. All piles were of 12 mm diameter so that the D/B ratios of 30, 40 and 50 were achieved. The model pile was jacked into the sand bed to the required depth as explained earlier. After installation of the pile, two dial gauges one to measure horizontal deflection and the other to measure vertical deflection of the pile head, were positioned. The pull out load, on the pile head, was applied by placing dead weights on the hanger connected to a wire rope passing over the pulley mounted on to the side of the tank as shown in Fig. 10. The oblique load angles 0o, 30o and 45o were achieved by positioning the pulley at different elevations (Fig.10). The load on the pile head was applied in steps of 1 kg (10 N). The horizontal and vertical deflection of the pile head under each load was recorded. The vertical deflection of the pile was insignificant when compared to its horizontal deflection [12]. Therefore the horizontal deflection was considered as the criteria for determining the pile capacity. When the horizontal deflection was practically constant under the applied load, the load on the pile head was increased to the next value. The test was continued till the soil failed by showing continuous increase in the horizontal deflection under a constant load.

Analysis of resultsThe load versus horizontal displacement curves for the 600 mm long pile (D/B = 50), under pull-out loads with loading angles θ= 0o, 30o and 45o are presented in Fig. 11. The ultimate pull out capacity Qu of the pile is obtained from the load displacement curve as the load corresponding to the start of the final linear portion of the curve [10] Similar load displacement relationships were observed for the other two embedment depths (D/B = 30 and 40) of pile. The variation of ultimate capacity of pile with the inclination of the load is shown in Fig. 12.From Fig. 12 it can be observed that, the uplift capacity of the pile is increasing with the loading angle (θ). This indicates that, the capacity of pile is small under the horizontal load when compared to vertical pull-out capacity.

Fig.10. Details of test setup

The pile capacity under oblique pull is estimated from Eq.(4) [8,13](Qu Sin Ɵ/ Qv)+ (Qu CosƟ/ Qh)2= 1 (4)

Where Qu is the ultimate pull out capacity of the pile under a load inclined at θ degrees with the horizontal, Qv is the ultimate vertical capacity of the pile (θ = 900) and Qh is the ultimate lateral resistance of the pile (θ = 0). The values of Qv are estimated using Eq. (5).[13,14]

Fig.12. Effect of load Angle on pile capacity Fig.11. Load Displacement curvesfor 600mm long pile

Qv = 0.5 γ D2 B Ku (5)

Where γ is the unit weight of the soil, Ku is the uplift coefficient [5,15], and the other symbols as defined before. For the present case Ku is 9. The computed values of Qv, for the three piles using (5), are presented in Table 4. The theoretical pile capacity of a pile under horizontal load (Qh) is obtained from Eq. (1)[5,14] . The effective depth ratio Deu/D for the given fixity condition of the pile head depends mainly on the relative pile stiffness given by Eq. (2) [2,16]. Using these Kr value, the ultimate effective depth ratio (Deu/D) is computed from Eq. (6) [15].

Deu/D = 1.65 Kr 0.12 (6)

Values of Qh for the three embedment depth tested (D/B = 30, 40 and 50) are presented in Table 4. The values of Qh and Q v are substituted in (4) to obtain the corresponding Qu

values under different θ values. The computed ultimate pull out capacities of piles under different load angles are in close agreement with the observed values as seen in Fig. 13 and Table 4.

Table 4: Uplift Capacity of Piles under Inclined Loads

Pile LengthUplift Capacity, Qu

(N)

(mm)θ = 00 (Qh) θ = 300 θ = 450 θ = 900 (Qv)

Est. Obs. Est. Obs.Est. Obs. Est.

360 33.6 31.2 33.2 32.536.4 36.0 62.2

480 45.6 44.8 46.8 47.052.5 52.5 110.6

600 57.4 57.5 59.9 64.068.4 74.0 172.8

Note:Est = Estimated value; Obs. = Observed value

5.0 CONCLUSIONSBased on the experimental and theoretical investigation carried out, the following conclusions are drawn.Four attachments were tried for a model pile, to improve its lateral capacity when embedded in sand. The percentage increase in the lateral capacity, due to these attachments, is ranging approximately from 20% to 124% of an equivalent plain pile. The percentage improvement depends on the type of attachment and its depth below ground level. Among the models tried a pile with four vanes shown a maximum improvement.

Fig.13. Comparison between estimated and observed values of pull out capacity

The theoretical estimates of lateral capacity of plain and modified piles are in good agreement with the experimental results. Three model piles were tested each under three different values of oblique pullout loads. From the investigation it was observed that, as the inclination of the load with the horizontal was increasing, the pull-out capacity of the pile was found increasing. The estimated pull-out capacities of piles under oblique loads closely agreed with the observed values.

REFERENCES:1. Broms, B.B., (1972). Stability of flexible structures (piles and pile groups). Proceedings of 5th

European Conference on Soil Mechanics and Foundation Engineering, Madrid. Vol. 2: 239-269. 2. Polous, H.G. and Davis, E.H. (1980), Pile foundation analysis and design. John Willey and Sons. 3. Meyerhof, G.G. and Ranjan,(1972), the bearing capacity of rigid piles under inclined loads in

sand I: Vertical piles. Canadian Geotechnical Journal, G., 9: 430-446. 4. Chari, T.R. and Meyerhof, G.G. (1983), Ultimate capacity of rigid single piles under inclined

loads in sand. Canadian Geotechnical Journal, 20: 849-854. 5. Meyerhof, G.G. and Sastry, V.V.R.N. (1985), Bearing capacity of rigid piles under eccentric and

inclined loads. Canadian Geotechnical Journal, 22: 267-276. 6. Koumoto, T., Meyerhof, G.G. and Sastry, V.V.R.N. (1986), Analysis of bearing capacity of rigid

piles under eccentric and inclined loads. Canadian Geotechnical Journal., 23: 127-131. 7. Broms, B.B. (1965) Design of laterally loaded piles. Journal of Soil Mechanics and Foundation

Engineering, ASCE, 91 (3): 79-97. 8. Meyerhof. G.G. (1973), The Uplift capacity of foundation under oblique loads. Canadian

Geotechnical Journal, 10(1): 64-70. 9. Das, B.M., Seely, G.R. and Raghu, D. (1977), Uplift capacity of model piles under oblique loads.

Journal of Geotechnical Engineering Division, ASCE, 1202 (9): 1009-1013. 10. Terzaghi, K. and Peck, R.P. (1967), Soil Mechanics in Engineering Practice. John Willey and

Sons, Inc., New York. 11. Meyerhof. G.G. (1994), Behavior of pile foundations under special loading conditions: R.M.

Hardy key note address. Canadian Geotechnical Journal, 32: 204-222. 12. Murthy, B.S. (1986), Studies on group of piles subjected to oblique loading. M.Tech. thesis,

Jawaharlal Nehru Technological University, Hyderabad, India. 13. Das, B.M. (1990), Earth anchors. Elsevier Science publishers, B.V. Amsterdam. The Netherlands. 14. Meyerhof, G.G., Sastry, V.V.R.N. and Yalcin, A.S. (1988), Lateral resistance and deflection of

flexible piles. Canadian Geotechnical Journal, 25: 511-522. 15. Sastry, V.V.R.N. and Meyerhof, G.G. (1990), Behaviour of flexible piles under inclined loads.

Canadian Geotechnical Journal, 27: 19-28. 16. Bowles, J.E (1982), Foundation analysis and design. McGraw-Hill Book Company New York. 17. Reddy, K.K. (1986), Lateral capacity of single pile-method of improvement. M.Tech. thesis.

Jawaharlal Nehru Technological University, Hyderabad, India.