Special Piles: Their Performance Evaluation and Advantages

Dr. Nainan P. Kurian, Ph.D., D.Sc., Retd. Professor of Civil Engineering, Indian Institute of Technology, Chennai.

The aim of this paper is to illustrate the performance of some special type of piles, called upon to transmit loads of varying types. It also highlights their advantages in performing their assigned roles, as confirmed by the author’s own investigations. The piles so chosen are, ‘tapered piles’, underreamed piles’ and ‘screw piles’.

Pile as a Structural Foundation

A pile is a long and slender structural member established in the soil which is called upon to transmit, to the surrounding soil, a vertical load (downward – compressive, or upward – tensile), a horizontal load and a moment transferred to it at the top from the superstructure. A pile may occasionally have also to transmit a twisting moment about its axis. (Kurian 2005)

There are two basic types of piles: (1) ‘displacement’ piles and (2) ‘replacement’ piles. While a displacement pile displaces a volume of soil equal to its own volume, in case of a replacement pile, a volume of soil equal to the volume of pile is removed from the ground which is subsequently replaced by the pile. Precast driven piles and bored cast in situ piles are typical examples of the above two types respectively.

From the point of view of load transfer, piles are basically classified as ‘end bearing’ piles and ‘friction’ piles. In some instances, the load is transferred partially in either mode. Such piles are known as ‘end-bearing friction’ piles, Figure 1. Special piles are piles which are different from normal piles which retain the same cross section over their length. This paper throws light on the behaviour of three special piles.

Tapered Piles

Prismatic piles, irrespective of the shape of cross section provided, normally have their sides parallel, so that the cross section remains identical throughout their length, Figure 1. Sometimes their sides are provided with a taper in which case the cross section decreases uniformly from the butt to the tip (Figure 2). Such piles are known as ‘tapered’ piles.

In case of a friction pile subjected to a downward vertical load, if its sides are parallel, the transfer of load to the surrounding soil is entirely by the shear (friction) at the interface. However, in the case of a tapered pile a portion of the applied load is transferred by direct bearing on the sides over the area in plan shown in Figure 2. This bearing results in an increased normal pressure when compared to the pile without taper, which consequently increases the frictional component of the shearing resistance. Tapered piles are therefore very effective in frictional soils such as sands. On the other hand, in clay, the difference between the capacities of uniform and tapered piles will be marginal, since the adhesion component of shearing resistance is independent of the normal pressure. On the side of tension, however, the advantage in compression is lost as can be realized from Figure 3 (in which the pile width is exaggerated by compressing its length), where heavy resistance is encountered in compression, but virtually no resistance in tension. A wedge (Figure 4) which is difficult to drive in the face of increasing resistance, but easy to pull out, is perhaps the best analogy for the behaviour of a tapered pile.

Investigations

Extensive studies involving both theoretical (by the finite element method) and experimental investigations on tapered piles with different geometrical shapes of cross section, such as square, circular and triangular, by Kurian and Srinivas (1994, 1995) and Kurian and Seetharamayya (1998), have shown that for the same material input, the performance of tapered pile is much superior, in terms of both bearing capacity and settlement, to the corresponding piles of uniform cross section, under compressive axial load, in sand. In each case, piles in the ‘displacement’ mode presented much better performance than ‘replacement’ piles. A significant finding is that triangular piles outperform the other shapes. These studies have been extended by Dawar (1995) for horizontal loads, where tapered displacement piles again revealed their superiority, thanks to the more equitable distribution of pile material at the top where the bending effects are the highest.

Major Conclusions

Even though the major findings from the above studies have been highlighted, the following are some specific conclusions arrived at from these investigations, which include extensive parametric studies, stated in a more quantitative form, as they apply to piles of typical standard dimensions.

1. For uniform and tapered piles having the same volume (i.e., equal material inputs) the ultimate load-carrying capacities of tapered piles have been found to be higher than those of uniform piles by about 10% with the triangular shape registering the maximum difference (12%) followed by the square and circular shapes.
2. Even though the surface areas of the uniform piles are higher than those of tapered piles by about 7% for the same material input in both, the surface resistance of tapered piles is higher by 45% on account of the increased unit friction resulting from bearing on the sides.
3. Tapered piles of circular, square and triangular sections undergo lesser butt settlements (about 25%) under vertical loads than uniform piles of the same shapes at corresponding loads.
4. The load-carrying capacity of the pile increases with the degree of taper, with the triangular pile registering the highest value.
5. The ultimate loads of displacement piles are higher (by 44% for circular to 83% for triangular) than replacement piles of the same shape.
6. In the case of displacement piles, the ultimate load-carrying capacities of tapered piles are higher than those for uniform piles by 4% for circular to 23% for triangular shapes.
7. For the same material input, the ultimate load is highest for the triangular shape, followed by the square and circular shapes, both under the replacement and displacement methods of installation. The same applies to both uniform and tapered piles of the same shape.
8. The displacement piles in all shapes undergo lesser settlements (30% for square to 45% for circular) under vertical load than the replacement piles of the same shape at corresponding loads, as predicted by the analysis and confirmed by tests.

Underreamed Piles

An underreamed pile is a bored cast in-situ pile of short length which is provided with a bulb or underream at the bottom (Figure 5). These piles have been found to be very effective in expansive soils if the bulbs are established in the nonswelling soil below. The top soil, which is susceptible to considerable volume change due to moisture variation, is normally confined to a depth of about 3m. Under a load bearing wall, these piles established at regular intervals along their centre line. These piles are connected by a grade beam at the top on which it constructed the wall. An air gap is required to be ensured between the bottom of the grade beam and the expansive soil below, sufficient to absorb the swelling of the soil preventing it from coming into contact with the grade beam. This prevents the soil exerting its pressure on the grade beam and the corresponding development of tension in the pile stem. Some amount of tension can still develop in the stem on account of the upward drag exerted by the swelling soil surrounding the pile. Under this tension the pile behaves like an anchor pile whose bulb provides the necessary anchorage in the firm soil. Depending upon its length, the stem can be provided with more than one bulb. (Kurian 2005)

The use of the underreamed pile is by no means confined to expansive soils. They can indeed be used in normal soils also as anchor piles when called upon to resist a tensile load. They are also effective under compression on account of the wider bearing area provided by the bulb. These piles can also be used in a group in which case they are connected by a pile cap at the top. Such groups even with two piles (joined by a capping beam) are very suitable, for example, as foundations for the legs of transmission line towers. Kurian (2005) also explains the method of manual construction of underreamed piles using simple underreaming tools.

In a field ruled by empiricism, designers, who want to design underreamed pile foundations have to necessarily resort to codal provisions many of which lack a rational basis. In India we follow the Indian Code on the subject: IS:2911(Part III) – 1980.

Analytical Studies on Underreamed Piles

With a view to gaining a rational understanding of the behaviour of underreamed piles under diverse situations, Srilakshmi(2003) and Kurian and Srilakshmi (2001, 2002, 2003, and 2004) have undertaken an analytical investigation of underreamed piles, with special reference to their geometry, by the finite element method, using the advanced ANSYS package. The nonlinear analysis up to failure have covered underreamed piles of various sizes and shapes, in cohesive and cohesionless soils, under vertical loads – compressive and tensile – and horizontal loads. An interesting finding has been a single cone underream with the top half (Figure 6) is superior to the normal double-cone underream, at considerable saving of material. Kurian (2005) have also given the design of a manual underreaming tool for cutting such half cones. One may note in this connection that this is the shape which is normally produced in the mechanized method of underreaming - as opposed to the manual method.

Analysis of Underreamed Pile in Expansive Soil by "Thermal Analogy"

The most significant part of the above work (Srilakshmi 2003) is the extension of the analysis into expansive soils, based on ‘thermal analogy’ the basis of which is explained below. When heat is applied to a body, it undergoes thermal expansion, which if prevented in any manner, will induce stresses in the body. In a like manner expansion takes place in an expansive type of clayey soil if water is input into the soil. Two properties of the material required in a thermal analysis are:

(1) ‘the coefficient of linear thermal expansion’(α) and (2) ‘thermal conductivity’ (K). For the analysis of expansive soil, analogous properties – call them (β) (the coefficient of linear swell) and M (moisture conductivity) – are determined in an analogous manner. These properties are input along with the water content distribution of the soil with depth (i.e., the moisture profile) in a ‘coupled (thermal-structural) field analysis,’ which outputs the desired results. An important result that emerged is that in deep deposits of expansive soils, if the bulb is placed in the expansive zone itself, as suggested by the code, the ultimate load carrying capacity is vastly reduced in tension, but increased in compression, mainly due to the upward swelling pressure acting on the bulb which opposes a downward compressive load, but aids an upward tensile load. The interesting aspect of the above approach is that it is general and applies to any type of foundation in an expansive soil medium. One can determine the peak stresses in the system corresponding to wet and dry seasons just from the respective moisture profiles and the two soil properties. Even though the approach to studying expansive soil using thermal analogy has been suggested earlier, the present analysis appears to be the first instance when the same has been carried out and brought to a useful and practical stage for the analysis of the performance of foundations in expansive soils.

Major Conclusions

The major conclusions from the above analytical studies are presented in the following in a more quantitative form as applying to underreamed piles of normal standard dimensions:

1. The most valuable conclusion from the case of no bulb (prismatic pile of dia. 0.3m) is the increase in the capacity by 260% in compression and 1125% in tension, when the pile is underreamed, for a volume increase of 29%. These results alone establish the case for the underreamed pile.
2. The influence of the location of the bulb is not significant beyond the standard height from the base of the pile for the location of the underream. However, this capacity is 9% less compared to the case of the underream located at the bottom of the pile. This leads to the observation that the standard location used in manual construction, leaving space for the bucket at the bottom, is reasonable.
3. The observation in the case of multiunderreamed piles is that, the capacity increases by 150% in compression and 180% in tension, when the number of bulbs increases from 1 to 3, against a corresponding increase in the volume of the pile by 35%. This leads to the conclusion that multiunderreamed pile is a cost-effective proposition in terms of material input.
4. The shape of the underream has revealed an interesting result. With half bulb (single upright cone) at the top alone, the maximum capacity in compression is found to exceed that of the full underream (double cone) by 20%, indicating thereby a gain in strength with reduced material input (i.e., by 50% for the bulb).
5. Model tests conducted on underreamed piles in expansive soil to validate the analytical approach developed, involving ‘thermal analogy’ and ‘coupledfield analysis’, gave encouraging results.
6. The variation of ‘moisture conductivity’, M, has little influence on the load carrying capacity of the underreamed pile - in compression and tension - and the swell of the soil.
7. For given properties of expansive and non-expansive soils, there is an increase in ultimate load in compression by 25% when the bulb is provided in the expansive zone, when compared with the case of the bulb in the non-expansive soil. As against this, in tension, the ultimate load suffers a decrease of 40%. As already stated, this is due to the fact that swell pressures add to oppose resistance in compression and tension, respectively.
8. Comparison of prismatic and underreamed piles in expansive soil has revealed the enormous contribution of the underream to the load carrying capacity. At an extra material input of 30%, the load carrying capacity registers an increase of 415% in compression. The corresponding figure is far higher in the case of tension since the load carrying capacity of the prismatic pile is found to be very low in tension. This establishes an even stronger case for underreamed piles in expansive soil than in non-expansive soil.

Critique on IS: 2911 (Part III) – 1980

Ever since the above standard was issued, a need was keenly felt for a sound theoretical back-up to corroborate the recommendations made in the code, which have been cited to have an empirical basis. The present work (Srilakshmi 2003) is perhaps the first attempt at a rigorous theoretical treatment of the problem of the underreamed pile in normal and expansive soils.

The main operative part of the code is Table 1, in Appendix B, which gives the permissible loads in compression, tension and lateral thrust, against diameters of pile varying from 200 to 500 mm. It is stated that the given values apply to ‘medium’ sandy (defined by range of N-values) and clayey soils, ‘including expansive soils.’

In the first place, the Standard Penetration Resistance, viz., the N-value, is an empirical strength parameter of the soil, which while reliably correlated with density index, and therefore (?), in the case of cohesionless soils, is far from being a reliable indicator of consistency in the case of cohesive soils (Kurian 2005). For values of N outside the above range, multiplication factors have been specified on the above codal values for arriving at safe loads. It also mentions that the recommendations are made based on ‘extensive pile load tests’ and criteria have also been stated, based on settlements and deformations, by which safe loads have been arrived at.

As against the above, the present investigations provide the values of the ultimate load (to be used with appropriate values for factor of safety) based on shear strength parameters c and (φ) in the case of the respective soils which does away with the empiricism associated with the codal recommendations based on N-values.

A matter of greater concern is the provision regarding ‘minimum length of the pile’ in deep deposits of expansive soils. The code permits a minimum length of 3.5 m for single underreamed piles, irrespective of the thickness of the expansive soil layer. The present investigations show the difference in load carrying capacities between piles with their bulbs anchored in nonexpansive soil and expansive soils. In the latter case, whereas the ultimate load in compression undergoes a dramatic increase, that in tension experiences a sharp fall. This picture can reverse direction, when the soil undergoes shrinkage, where the load under compression may be subject to reduction and that under tension to increase. In either case the designer should be concerned with minimum values of ultimate loads in compression and tension, irrespective of when it occurs.

This makes the codal recommendations grossly unsound in respect of the minimum length of pile in deep deposits of expansive soil. Under these circumstances it would only be rational to anchor the bulb in the stable zone below irrespective of at what depth the same occurs.

The codal provisions regarding spacing of piles lack rigour. Analysis of a group of two piles indicates that even at the standard spacing of twice the bulb diameter, the group capacity does not reach the sum of the individual capacities, not to speak of the picture at the minimum spacing of 1.5 times the bulb diameter.

Another finding from the investigations, which reveals the scope of saving in cost, is that the half bulb with single cone at the top gives higher load carrying capacity, both in compression and tension, compared to the double cone bulb recommended in the code. It is also shown that such a half bulb can be produced by manually operated rigs with a minor modification in the design of the underreaming tool.

It is fervently hoped that the Bureau of Indian Standards will take due cognizance of the findings of these investigations, when the Code is taken up for its next revision.

Screw Piles

A circular pile in steel with helical blades, installed by applying a torque in the presence of a vertical load, is an old type of foundation which was widely used in many countries before the advent of piles in reinforced concrete. A circular blade, which is the plain counterpart of the helical blade provides increased bearing area. The purpose of making it helical is to facilitate installation and at the same time for deriving the advantage of increased bearing area. The method of design adopted so far, whether the load is compressive, tensile or horizontal, is treating the blade as a circular (annular) plate, ignoring the helical shape.

The original screw pile was typically a hollow steel pipe pile, closed at the bottom with a conical shoe, at the end of which was fixed a helical unit (single turn) (Figure 7). The shaft diameter varied from 160–900 mm and the outer diameter of the helix varied from 2 to 4 times the shaft diameter, subject to a maximum of 2800 mm. The higher bearing area so created at the foot of the pile resulted in higher capacities in compression, tension and horizontal loads.

Modern Screw Piles

Even though the traditional steel screw piles of the type described above went into near oblivion with the arrival of R.C. piles, they have staged a comeback in the field of R.C. piling itself where the screw refers not to the helical blades, but to the manner in which the sides of the boreholes are profiled for forming replacement type cast in-situ R.C.piles. A good number of such piles are available, which are known by the proprietary technique used in creating them. Typical names are the Atlas pile, the Fundex screw pile, the VB-Franki pile, the Omega (Ω) pile, etc. Figure 8 shows the shaft of the Atlas screw pile. The auger head of the Omega pile is shown in Figure 9.

The helical blade (Figure 7), which is spatially curved, is essentially a shell, which qualifies the helical screw pile to be treated as a ‘shell foundation’.

Analysis of Screw Piles

As already stated, all the works on various aspects of the screw pile problem disregarded the actual geometrical shape of the helical blades. It is only the versatility of the finite element method that can account for the actual geometry of a spatial structure such as the helical blade at the micro level. The work by Shah (2005) which is presented here, is perhaps the first attempt where helical pile, of the type shown in Figure 7, has been subjected to the finite element analysis, to study the response within the elastic and elasto-plastic ranges by continuum analysis. The analysis has been validated using the results of a prototype test available in literature.

The results have revealed that the screw pile is superior to its plain counterpart in all modes of load transfer.

 

Major Conclusions

The major conclusions from the above work are stated below:

1. The helical shape of the screw pile is found to contribute to higher ultimate strength over its plain circular counterpart. The increase is found to be of the order of 140%. One may note in this connection that more material, however, goes into forming the helix over its plain counterpart.
2. Interface friction is found to contribute to higher ultimate strength n tension than in compression by about 30%.
3. The ultimate horizontal load increases by 325% when the blade diamteter increases from 300 to 900 mm.
4. When the pitch of the pile was decreased by 30%, the ultimate load increased by 110%.
5. When an extra helical turn was added, the ultimate load increased by 150%. This is, however, at the expense of 100% extra material input for the helix.

Conclusion

The investigations in respect of the three special piles reported above should enhance the level of confidence with regard to their adoption. As such one would expect to see them enjoy wider use in foundation engineering practice in the years to come.

References

• Dawar, M. (1995) ‘Studies on the behaviour of displacement piles under lateral loads,’ M.Tech. Thesis, Department of Civil Engineering, Indian Institute of Technology, Chennai.
• Kurian, N.P. (2005) ‘Design of Foundation Systems: Principles and Practices,’ 3rd Edn. Narosa Publishing House, New Delhi.
• Kurian, N.P. and Seetharamayya, G. (1998) ‘Comparative behaviour of displacement and replacement piles under axial load,’ Proc. 6th East Asia- Pacific Conference on Structural Engineering and Construction, Taipei, Vol.3, pp. 1835–1840.
• Kurian, N.P. and Srilakshmi, G. (2001) ‘Studies of underreamed piles by the finite element method,’ Proc. Indian Geotechnical Conference, Indore, pp. 135–138.
• Kurian, N.P. and Srilakshmi, G. (2002) ‘Studies of underreamed piles in sand and c-φ soils by the finite element method,’ Proc. Indian Geotechnical Conference, Allahabad, Vol. 1, pp. 569– 572.
• Kurian, N.P. and Srilakshmi, G. (2003) ‘Studies on the behaviour of underreamed piles In clay and sand under lateral load by the finite element method,’ Proc. 12th Asian Regional Conference on Soil Mechanics and Geotechnical Engineering, Singapore.
• Kurian, N.P. and Srilakshmi, G. (2004) ‘Studies on the geometrical features of underreamed piles by the finite element method,’ Proc. International e-Conference on Modern Trends in Foundation Engineering: Geotechnical Challenges and Solutions, Theme 1: Special Foundations, Foundation Techniques, paper No.5, 14pp.
• Kurian, N.P. and Srinivas, M.S. (1994) ‘Comparative behaviour of uniform and tapered piles,’ Proc. 3rd International Conference on Deep Foundation Practice, Singapore, pp. 155–159.
• Kurian, N.P. and Srinivas, M.S. (1995) ‘Studies on the behaviour of axially loaded tapered piles by the finite element method,’ International Journal for Numerical and Analytical Methods in Geomechanics, Vol. 19, Issue No. 12, pp. 869–888.
• Shah, S.J. (2005) ‘A Continuous Spring Model for Soil-Structure Interaction Analysis of Beams, Plates and Shells on Elastic Foundations,’ Ph.D.Thesis, Department of Civil Engineering, Indian Institute of Technology, Chennai.
• Srilakshmi, G.(2003) ‘Studies on the Behaviour of Underreamed Piles in Normal and Expansive Soils by the Finite Element Method,’ Ph.D. Thesis, Department of Civil Engineering, Indian Institute of Technology, Chennai.