Chris Oram, Roger Bullivant
This paper has been prepared in response to concerns raised by the author at numerous levels that Annex A within BS 8004:2015 does not do enough to explain how steel helical piles work, and thus how to approach design.
This paper is designed to be read in conjunction with the aforementioned annex, and the recently revised ICE Speciﬁcation for Piling and Embedded Retaining Walls (SPERW), which now includes a section covering the installation of steel helical piles. It is not intended to act as a replacement for either of the documents, although it is hoped that it can be drawn upon for future revisions of the British Standard.
It is also not the intention to recommend the helical piling system over any other for specific loading conditions, as such a decision would be subject to a variety of considerations based on individual project specifics. Likewise, any derivation of actions under BS EN 1990, for design to BS EN 1997-1, would include the correct and relevant partial and combination factors applied to any actions imposed on the foundation, taking into account the magnitude and frequency over the design life. With reference to using the helical piling system for cyclical loading conditions, considerations given in cl.184.108.40.206 of BS 8004:2015 for cyclical loading would still be relevant, and over-arching.
While cl.A.2.4, Note 1, references the reader to Howard A Perko’s publication Helical Piles: A Practical Guide to Design and Installation for details on the design of helical piles, it is a US publication, and will offer the reader very little input in terms of adapting the design for use with the Eurocodes. Where possible, this paper makes recommendations to any adaptations to the design of helical piles in order facilitate a Eurocode design, although this guidance should be used for reference only, and no liability is assumed for its use.
As a final note, to avoid confusion, the majority of references made in this paper are for BS 8004:2015 (unless otherwise stated).
2.0 Compressive resistance of helical piles
With steel helical piles there are two accepted methods of design: the individual bearing plate method and the cylindrical shear method.
The individual bearing plate method applies when the spacing between plates is large enough so that each helix will act independently of the other(s). If the spacing between the plates is small, then the helical plates will act as a group and the bearing capacity of the pile will incorporate the bearing of the bottom plate and the side shear along the cylinder of soil that is formed between each plate, as first recommended by Mooney et al (1985). This soil cylinder is misleadingly referred to as a “plugged shaft” in Annex A of BS 8004:2015; as many helical piles use an open steel tube, a “plugged shaft” could allude to plugging at the end of the tube. This comment suggests incorporating the end bearing resistance of the tube itself which is small in relation to the bearing resistance of the helix plates.
If the pile has one bearing plate then only the individual bearing method can be adopted for design. If piles have more than one plate, it is prudent to use both methods and limit the result to the lowest calculated value. While the exact point at which the transition between individual bearing failure and cylindrical shear is not known, and will vary depending on the soil type, it is prudent to use a helix spacing-to-diameter ratio of three as a rule-of-thumb when applying the note in cl.A.5.1, checking sufficient vertical spacing between the helices as it prevents the overlap of the stress bulbs beneath each plate. The helix spacing to diameter ratio is still very much up for debate: experimental results from Rao et al (1993) put it at a value of around 1.5, whereas Bassett (1978) suggests the transition occurs at a ratio of 2.1 to 3.4.
Generally speaking, under Eurocode piles should be preliminary tested prior to final design to ensure design verification, and improved confidence in the design (via a reduced model factor, or SLS verification factors), regardless of which method of design is used.
The helical pile designer/supplier should be able to demonstrate clearly in calculations which method has adopted and these should contain sufficient enough detail as to how their soil parameters have been derived. Naturally, this would include reference to a detailed soil investigation with a satisfactory number boreholes undertaken to a suitable depth, encapsulating the full length of the proposed pile, with adequate soil testing to BS EN 1997-2. This information would facilitate comparison of the design to the installation records and the subsequent designer/checker tasked with the design review.
3.0 Shaft friction
As per cl.A.5.1.3, shaft friction is generally ignored in the design of helical piles, but the reasons for this are not outlined. Generally speaking, most manufactured helical piles are smooth steel-shafted tubes and have coupling sleeves which are slightly larger in diameter than the shaft which create a void/space around the shaft during installation. Similarly, the bolts which hold these section into place will also carve a path of increased diameter throughout the soil during installation. Square-shafted piles, such as the A B Chance system, can create a round hole of loosened soil immediately adjacent to the shaft during installation. Wobbling during installation may also cause the soil to separate from the pile shaft along the uppermost sections of the pile, especially if piles are installed without a guide mast. As it is difficult to quantify many of these reasons shaft adhesion is often just ignored in the pile design but, in reality, it is present regardless of installation method and it is not unreasonable to assume large diameter piles may develop a large proportion of their capacity in shaft friction.
Cl.A.5.1.3 and the following note are misleading and it is believed that the shaft friction along the pile could be taken into account if testing gives better than expected results, even when considering designs carried out via the cylindrical shear method. Designers should factor in a reduction of the shear strength of the soil to account for the reduced friction of soil on bare or galvanised steel and may also need to reduce this further for other surface treatments. However, if you are piling in certain soils, ie London Clay, then it would be more prudent to use lower values for an α-value to reflect appropriate soil behaviour during installation. It is also recommended taking shaft friction over an effective length (Heff) rather than the whole pile length to account for any void formation by the plate during installation.
Under Eurocode design (BS EN 1997-1:2004+A1:2013) the two method helical piling design approaches can be applied when using the relevant design approaches. For the resistance factors, due to the fact that the system is not auguring the ground and the plates are seen to displace the soil, a designer can adopt the R4 values for a driven pile as per Table A.NA.6. For calculation shaft friction over Heff in the design, it is recommended to consider adopting the reciprocal of the M2 material set values as per Table A.NA.4 for calculation in the GEO limit state if testing is not adopted, and lower bound M1 values for the STR limit state. Again, the helical pile designer/supplier should be able to demonstrate assumptions clearly in calculations.
While this may seem controversial, what is proposed above when it comes to incorporating any potential shaft friction, in terms of the theory behind how it can be calculated for a helical pile, is detailed in Chapter 4 of Perko (2009), which would also be as per cl.A.2.4, Note 1. If the code does not allow this then it is inconsistent by picking and choosing sections of the source design material to fit its agenda. This paper sets the view that it is not unreasonable to suggest there is scope on a case-by-case basis to consider the contribution of shaft friction, and the decision to do so would come from a contribution of a number of factors: soil type, soil strength, installation performance, test performance and geometry of the pile.
4.0 Pull-out resistance of helical piles
While design of pull-out resistance is briefly mentioned in the annex (cl.A.2.4, Note 2, and cl.A.5.2) it only represents a very basic understanding and should be expanded upon. In theory, the bearing and pull-out capacity of a deeply embedded helical pile can be calculated in a similar way, but as the soil can be disturbed above the helical plates during the pile installation a designer can apply a reduction factor to the tensile capacity. A disturbance factor of 0.87 is recommended by Perko (2009), but it can vary according to soil type and installation performance.
Cl.A.5.2 is also misleading as it is essentially a repetition cl.A.5.1.3, and shaft friction along the effective length of shaft above the top helix (Heff) could be taken into account if testing gives better than expected results and a suitable case for adoption can be reasoned, as per the previous section.
Under Eurocode, adoption of the reciprocal M2 material set values is recommended as per Table A.NA.4 for pull-out design where testing is not adopted, which can be revised either to include the shaft friction for calculation using an appropriate partial factor for tensile resistance over Heff in the GEO limit state, or to include the M1 set should favourable testing results be obtained.
It is also recommended that helixes must achieve a critical depth to ensure a deep mode of behaviour, which is not an active recommendation of Annex A of BS 8004:2015. If a helical anchor is too shallow then the weight of the soil above it will be insufficient for the pile to provide suitable resistance to tension. Shallow failure can occur when the bearing plates are located too close to the ground surface, or for helical pile used as anchors where the plates are located too close to the active soil wedge. A failure would see a shearing of the soil around the helical bearing plates and a lifting up of a cone of soil above the upper-most helix.
Again, the helical pile designer/supplier should be able to demonstrate the approach clearly in calculations.
This paper is in agreement with comment cl.A.2.1.9 which states that the design of helical piles should be based on a conventional soil mechanics approach supported by testing when combined by an empirical approach. The paper also agrees with cl.A.2.1.10 that states that helical piles should not be designed solely on empirical rules relating to the applied torque measured during pile installation. What requires further clarification are points cl.A.7.12 – A.7.14, as they refer to installation torque, and design installation torque, as critical values within the installation procedure and yet no mention is given to how these values are determined or their impact on design. As a result, the designer is therefore left in a paradoxical situation where the torque is both of great importance and of little consequence in the design and installation of a helical pile.
While most literature on helical piling will tell you that although it is very difficult to predict, torque can be used as a way of verifying the axial capacity of a pile in both compression and tension. It is widely accepted that the relationship stated by Hoyt and Clemence (1989) is the easiest way to calculate pile capacity from the final installation torque, where a variable capacity-to-torque ratio is used, and is dependent upon a variety of factors: soil conditions, shaft size and shape, and application of the pile (be it tension or compression). The number of helical plates also has an impact on the torque as plates can work against each other depending on installation and soil conditions, often resulting in very high torque.
This paper suggests that rather than enforce capacity-to-torque ratio values into the code for the derivation of torque, helical piling contractors should be able to demonstrate to clients and engineers their methods of calculating the anticipated minimum and design torques in their design calculations, backed up with empirical data through testing. Of course, this would therefore require contractors to both record and maintain adequate installation records, and this is often a commercial/contractual pre-requisite.
Maximum torque values used in design and installation should be determined by the robustness of the structural elements used in the formation of the helical pile. As helical piles are a bespoke product, all contractors should be able to detail the torsional resistance of the steel tube pile shaft to avoid twisting during installation. In a modular helical piling system particular attention must also be made to the bolt connection between sections, as this too can act as the weakest point of the system and to determine the maximum torque values for installation. It is advisable for helical piling contractors to limit the torsional resistance of the structural elements of a pile to the serviceable limits, so as to ensure no weakening of the structure occurs during the installation.
Attention should also be given to the difference between the maximum and design torque of a helical pile during installation, allowing for a safety buffer for the installation crew to have the opportunity to “punch in” should slightly harder bands or a movable obstruction is encountered during installation without over-stressing the piles.
With all this is mind, this paper reiterates that torque alone should not be used as a method of design of helical piles, as per cl.A.2.1.10, and should only be used in conjunction with an approved pile capacity calculation by way of comparison, as per cl.A.2.1.9. However, there are some further issues that need to be considered when trying to link installation torque readings to geotechnical performances. This paper recommends that preliminary or working pile tests would be worthwhile additions to any helical piling scheme. Even with relation to shaft friction during installation, as per chapter 6.4 of Perko (2009), if the soil has been disturbed enough by the helical plates the torque being recorded might be only that of the shaft friction along the pile tube and not indicative to the performance of the plates themselves. The torque-to-capacity correlations detailed in Perko (2009) are somewhat tenuous when compared to the actual spread of the data. There have been numerous studies undertaken to improve on this, such as the idea of developing an energy model as per Perko (2000) and more recently the design approaches taking improved correlations for granular materials and CPT cone testing as per Gavin et al (2013), Spagnoli (2016), Al-Baghdadi et al (2017) and Davidson et al (2018). Any further developments in this field will help improve confidence in relating pile capacity to installation torque.
6.0 Auguring of piles
As per cl.A.7.2, a crowd force is applied to the pile head to ensure that the penetration rate outlined in cl.A.7.1 is achieved. Despite this application of crowd, if the penetration rate falls outside of these limits, the pile can be said to be auguring (or spinning), and the pile capacity should be reassessed (as stated in cl.A.7.3).
The result of this lack of penetration is that a void is formed under the helix, and theoretically only the lead edge of the helix will bear onto the ground. If this occurs at depth, this may invalidate the design. The bearing pressure area in compression is equal to a line load on the lead edge of the helix, and the end of the pile shaft, rather than the complete area of the helix plate. This will also prove an issue in tension as the auguring of the material may also impact on the strength of the soils above the helical plates, particularly in sensitive soils. The end result is that the pile must either be discounted, or the capacity reduced unless testing can be undertaken to check the performance of the pile.
7.0 Horizontal loading
No guidance is given in the annex as to the design of the lateral resistance of helical piles. However, the lateral resistance of the pile is due to the performance of the steel tube that forms the pile shaft, and the strength of the surrounding soils. Therefore, any number of industry-accepted methods can be adopted in line with cl.6.4.5 of BS 8004:2015 to calculate lateral resistance and displacement, including elasticity theory, p-y curves, subgrade reaction models, or any other approved numerical models.
Due to the modular nature of the system, there are many different products and solutions offered by a number of helical piling contractors that can help improve the lateral performance of the system. These range from adding an oversized, or cruciform collar to the top of the pile to add lateral resistance by increasing surface area, welding steel plates to the top of the pile to increase the surface area, or by simply increasing the thickness or diameter of the upper sections of tube to improve the moment capacity of the pile. Not all of these solutions may be appropriate for use depending on the various site and project constraints, but the designer/supplier must appreciate any implications of each one adopted in the design, for example when using an oversized connection the effect of the creation of a void or space around the shaft during installation. It is therefore the responsibility of the contractor/supplier to demonstrate the lateral capacity of a bespoke system, and where practicable, a lateral load test should be carried out in order to verify the suitability of the method adopted.
8.0 Pile spacing and grouping
Cl.A.2.3.2 infers that helical piles are not to be spaced closer than four time helix diameters apart (centre-to-centre on plan), and this is in line with guidance from Report AC358, ICC-Evaluation Services (2007) and is a standard across the helical piling industry.
In terms of group effects, the ultimate capacity of a pile group is determined using a similar method to the cylindrical shear method, and must be considered in the design.
9.0 Pile settlement
As part of a Eurocode pile design it is now down to the pile designer to predict the settlement of the pile at working load as a serviceability check. Reference should be made to cl.6.4.4 of BS 8004:2015 for approved methods of calculating settlement, although these are no substitute for a static pile load test. It could be argued that due to the lack of exposure a number of clients and engineers have with helical piles, testing would help improve confidence in their adoption as a mainstream foundation solution.
This paper suggests two key points should be considered. Firstly, if the shaft friction was ignored in the design then it should also be discounted in the settlement prediction. If, as previously discussed, the pile behaves better than expected, then re-introducing it can be considered in both the pile design and the settlement calculation. Secondly, thought should also be given to the settlement prediction of a helical pile with multiple plates, especially in ground of variable strata. Elastic shortening of the steel under working load should also be considered.
10.0 Structural design
Elaborating on the points outlined previously the pile shaft section requires a check for buckling resistance, as well as a moment and axial force check.
The pile is unlikely to fail in buckling, although a buckling check should be carried out as a standard when a pile is installed through very soft strata. A helical pile is most likely to fail under bending and so the check for MEd ≤ MN,Rd is critical. An estimated pile fixity point is used in these design checks, and this can be determined either by software/modelling, or by the calculation methods outlined in cl.6.4.5 of BS 8004:2015. Where a modular system is used this point of fixity must not fall below, or clash with, the connection between the top two sections of the pile.
All steel piles are at risk from attack from electrochemical corrosion, rather than sulphate chemical attack as per concrete piles. Corrosion rates of soil is dependent on a variety of different factors, such as low pH values, chloride salt content, moisture content, oxygen availability and the presence of certain bacteria. Stray currents and the electrical connection of the structure to another metal are also factors that can affect the corrosion rate of the pile. The general method to deal with corrosion of a helical pile is a combination of utilising a coating of galvanising, and by including a sacrificial thickness of steel in the pile wall. Sacrificial anodes can also installed on some piles where soil corrosivity is classified as severe. Cathodic protection can also be utilised to deal with stray currents and electrical connection, usually in the form of a wire, or strip of metal running away from the pile and into the ground.
Individual helical piling contractors should be able to advise further as to their methods for countering corrosion, and provide some level of empirical data to satisfy any potential concern with the design life of their piles.
It is prudent that any structural check for a helical pile be carried out with the reduced thickness of steel, so as to ensure consistent performance throughout its design life. Failure to do so may result in the requirement of remediation work further down the line.
Finally, although this is more of a fabrication issue rather than a design issue, it is important to note that the welds on helical piles between the plate and the steel tube are a particular vulnerability. Welding is covered briefly in section B7.6 of the third edition of the ICE SPERW, where the relevant ISO standards covering quality control are listed. It is imperative that all welds are checked thoroughly for quality before installation to ensure the system is fit for purpose.
11.0 Installation and testing
The installation process of a steel helical pile is covered in great depth in both BS 8004:2015 Annex A and sections B7 and C7 of ICE SPERW. This paper does not address or alter these sections. Helical piling contractors should, however, be able to provide site specific method statements and risk assessments outlining their processes when dealing with the issues raised in the above documents, in particular to their reporting of installation torque, monitoring penetration, and their re-design and justification processes for those piles which are deemed to be auguring, or do not achieve minimum or design torque.
Static load testing of steel helical piles is also covered in depth in sections B7.8 and C7.8 of ICE SPERW.
This paper aims to offer better explain some of the clauses in Annex A of BS 8004:201, and how to approach the design of a steel helical pile, especially to the Eurocodes. If read in conjunction with the aforementioned annex and ICE SPERW, a designer or checker should be able to cover most, if not all, of the design quirks of the system. Simultaneously, helical piling pile designers/suppliers should be able to demonstrate to clients and engineers within calculations a variety of considerations, both geotechnical and structural, and these should contain sufficient detail as to how parameters have been derived. It should also be possible to demonstrate methods of calculating the anticipated minimum and design torques in design calculations, backed up with empirical data through testing, and from adequate site installation records.
Dialogue is underway so that the details of this paper can be drawn on for future revisions of the British Standard but, with an upcoming revision of the Eurocodes in 2020, it is anticipated that a further revision of this paper is likely.
Al-Baghdadi, T, Davidson, C, Brown, M, Knappett, J, Brennan, A, Augarde, C, Coombs, W, Wang, L, Richards, D and Blake, A (2017). CPT-based design procedure for installation torque prediction for screw piles installed in sand. 8th International Conference on Site Investigation and Geotechnics. London, UK, Society for Underwater Technology (SUT OSIG).
BS 8004:2015, BSI (2015)
BS EN 1993-5:2007 (E), BSI (2007)
BS EN 1997-1:2004+A1:2013 + UK National Annex, BSI (2013)
Bassett, RH (1978). Underreamed Ground Anchors. Bulletin of Engineering Geology and the Environment, Vol 18, No 1, December Springer, Berlin/Heidelberg, pp. 11–17.
Davidson, C, Al-Baghdadi, T, Brown, M, Brennan, A, Knappett, J, Augarde, C, Coombs, W, Wang, L, Richards, D, Blake, A, and Ball, J (2018). A modified CPT based installation torque prediction for large screw piles in sand. Proceeding of 4th International Symposium on Cone Penetration Testing (CPT’18). 21-22 June 20138. Delft, The Netherlands.
Gavin, KG, Doherty, P, and Spagnoli, G (2013). Prediction of the installation torque resistance of large diameter helical piles in dense sand. Proceeding of 1st International Geotechnical Symposium on Helical Foundations. Amherst, Massachusetts.
ICE Specification for Piling and Embedded Retaining Walls, Third Edition, ICE/Thomas Telford (2016)
Mitsch, MP and Clemence, SP (1985). The Uplift Capacity of Helix Anchors and Sand. Uplift Behavior of Anchor Foundations in Soil, ASCE, pp. 26–47.
Mooney, JS, Adamczak Jr, S, and Clemence SP. (1985). Uplift Capacity of Helix Anchors in Clay and Silt. Uplift Behaviour of Anchor Foundations in Soil, ASCE pp. 48–72.
Perko, H (2000). Energy method for predicting installation torque of helical foundations and anchors. In Geotechnical Special Publication 100, New technological and design developments in deep foundations, ASCE, 342-352.
Perko, H (2009). Helical Piles: A Practical Guide to Design and Installation.
Perko, H (2007). Installation Torque as a Predictor of Helical Pile Axial Capacity
Rao, SN and Prasad, YVSN (1993). Estimation of Uplift Capacity of Helical Anchors in Clays. Journal of Geotechnical Engineering, Vol. 119, No. 2, pp.352–357.
Report AC358, ICC-Evaluation Services, Inc. (2007)
Spagnoli, G (2016). A CPT-based model to predict the installation torque of helical piles in sand. Marine Georesources & Geo