Permafrost, Phillips, Springman & Arenson (eds) 2003 Swets & Zeitlinger, Lisse, ISBN 90 5809 582 7

Design of helical pier foundations in frozen ground

ABSTRACT: Helical piers have been used as foundations for boardwalks, utilidors and fences in rural Alaska in cold and warm permafrost. The piers are becoming more and more popular due to their light weight, ease of transportation and installation, and resistance against frost jacking. However, engineers have been hesitant to specify them as foundations for buildings due to the lack of design guidelines for frozen ground. To increase the use of helical piers as a cost effective foundation alternative, Alaska Science and Technology Foundation funded a study to create these guidelines. This paper presents the current experience with helical piers in Alaska and gives a design example using the developed finite element analysis and creep results.

1 INTRODUCTION Helical piers create a lightweight foundation that can be used either in compression or tension. Due to their light weight, ease of installation and minimal ground disturbance they are becoming very popular in remote building sites such as arctic villages in Alaska (Figs 1, 2). Conditions in these places often include permanently frozen soils, which changes the dictating

design criteria from bearing capacity and consolidation settlement to creep settlement. Current design methods are not suitable for helical piers in frozen ground, which may limit their use. Therefore, the Alaska Science and Technology Foundation (ASTF) is funding research to develop guidelines for design of helical piers in frozen ground. The purpose of this paper is to show how the Finite Element Analysis (FEA) is used to create design curves and aid in design of helical piers in frozen silt.

2 HELICAL PIERS The helical piers typically consist of a central shaft made from square or round sections that can be either solid or hollow. To this shaft is connected between one to four spiral plates (Fig. 3) that are designed to mobilize more soil resistance under normal loading conditions than a conventional smooth pile.

Figure 1. Helical pier stockpile in St. Michael, Alaska.

Figure 2. Insignificant ground disturbance. 1307

Figure 3. Helical pier lead sections (A.B. Chance Co. 1996).

In frozen ground, additional vertical pressure is often required. First pier installations into very cold (5C) permafrost were conducted into 50 mm pilot holes to ensure straight piers. Installers found, however, that piers can be installed correctly even into the cold permafrost without pilot holes.

5 DESIGN METHODS The current design method for determining the capacity of helical piers in thawed soils is based on bearing capacity theory. According to Ladanyi & Johnston (1974), it is not appropriate to analyze deep circular pile foundations in frozen soils on the basis of Prandtl-type bearing capacity equations and a separate settlement analysis using Boussinesqs stressdistribution theory and compressibility of soil. In frozen soil, the temperature and undrained creep become predominant in the determination of allowable foundation pressures. Therefore, Ladanyi & Johnson (1974) developed an alternative method for predicting the time and temperature dependent creep settlement and the bearing capacity of frozen soil under deep circular loads. The asymptotic ultimate pullout resistances calculated with the model seem to agree with field results, but it is not clear if the model can be used for long term creep under compressive loads. Therefore, further analysis using FEA is warranted. The purpose of the project is to develop a finite element model that will be used to create design curves with different soil temperatures and soil properties. The analysis developed includes four models that are Large Model, Small Model, Installation Failure Model, and Creep Model (Liu et al. 1999a). The Large Model analyzes the soil stresses and displacements immediately after the pier is subjected to its design load. These data are critical for the development of more detailed analysis using sub-modeling techniques. The Small Model is a sub model of the Large Model. It analyzes the stresses developed within the spiral structure by using results from the Large Model analysis. The Installation Failure Model is a detailed model of the spiral structure subjected to a torsional load during installation. This model provides insight into the failure mechanism of helical piers during construction. The Creep Model analyses the longterm displacement and soil stress in frozen ground. Figures 6 and 7 provide examples of the deformations and stresses in the pier itself and the surrounding soil obtained from the Large Model. The finite element analysis shows that the traditional bearing capacity equations are not adequate design tools for helical piers with multiple helixes (Liu et al. 1999a, 1999b, 2000). The stress distribution in the soil does not agree with the traditional method (bearing capacity
1308

Figure 4. Typical uses for helical piers: boardwalks and utilidors.

Figure 5. Utilidor under construction in St. Michael, Alaska.

3 APPLICATIONS The helical piers are used for a variety of applications. In compression, they are used as building foundations for new structures, underpins for existing structures, and as support for boardwalks, utilidors, fence posts, light posts, etc. In tension, the helical piers, often called helical anchors, are used as soil anchors for retaining walls, and to resist wind loads for tall structures. In Figure 4, the utilidors in Selewik, Alaska, are founded on helical piers, but the boardwalk is founded by conventional methods. The helical piers are wellsuited to resist the frost jacking that is a common problem with pile foundations in frozen ground. Figure 5 shows helical piers installed for a utilidor in St. Michael, Alaska.

4 INSTALLATION Helical piers are installed by applying a torque on the pier shaft. The pier then augers itself into the ground.

equation) that assumes the helixes are not connected with each other. Further, the traditional bearing capacity equations do not consider creep in frozen soil. The analysis and results given in the following sections are based on the creep model that utilizes the Large Model. The creep equation used by Morgenstern et al. (1980) was adopted in the analysis (Equation 1): s & e &c e scuu
n

A full-scale loading test was conducted in the U.S. Army Cold Regions Research and Engineering Laboratory (CRREL) to calibrate the FEA. However, the FEA models could not be calibrated using the CRREL test results due to uneven temperatures in the test cell. 5.1 Development of design guidelines

(1) The design guidelines are developed on the basis of the Creep Model. For frozen ground, the dictating design criterion is the settlement due to the creep of the ice in the ground. The guidelines obtainable in the following sections are results from the Large Model creep analysis. Input parameters for the analysis include material properties for the pier and the soil. The soil parameters include the strength parameters and creep parameters. These are functions of soil type, water content, unfrozen water content, and soil temperature. They vary greatly depending from location and season, and therefore, general guidelines cannot be provided for typical soil types, e.g. for sand, silt or clay. The following sections describe an example of design of helical piers for a silty soil at three different temperatures. To analyze any other soil, the engineer needs to contact the authors for FEA runs to produce the creep curves. 5.2 Materials and model dimensions

where e strain rate; c reference strain rate; n experimental creep parameter, se equivalent stress and cuu temperature dependent total deformation modulus, corresponding to the reference strain.

Figure 6. Vertical displacement distribution in soil volume x 3302 mm, z 3556 mm.

The soil properties for the silty soil are given in Table 1. The pier has a pipe shaft with 90 mm outer diameter and 13 mm thick wall, and a 203 mm diameter, 13 mm thick plate for spirals. For the large-scale model, the spiral is modeled as a flat plate in order to limit the number of elements in the model. The element and material properties for steel appear in Table 2. The soil
Table 1. Soil properties for silty soil. Temperature Unit Unit weight Cohesion Friction angle Creep parameter, n Creep parameter, B Youngs modulus Poissons ratio
1 &c B scui
n

1C 19.07 2413 25 2.04

5C 19.07 6206 25 2.04

10C 19.07 9653 25 2.04

kN/m3 kN/m2

h (h MPan)1 3.81E-7 5.49E-8 1.85E-8 kPa 1800 0.2 7400 0.2 14400 0.2

Figure 7. Vertical stress distribution in soil volume x 3302 mm, z 3556 mm. 1309

Table 2. Pier properties and FEA parameters. Material Steel shaft Steel plate n 0.3 0.3 Element Beam Shell NDF 6 6
Displacement (mm)

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0 10 20 Time (years) 30 28 kN 56 kN 83 kN 111 kN

E MPa 200000 200000

r kN/m3 77.0 77.0

NDF Nodal Degrees of Freedom.

Figure 10. Creep displacement for a pier with one helix at a design temperature of 1C for various axial loads.
1270
-0.8 Displacement (mm)

3556 2286 Soil Volume

-0.7 -0.6 -0.5 -0.4 -0.3 -0.2 -0.1 0.0 28 kN 56 kN 83 kN 111 kN

1651

1651

10 20 Time (years)

30

Figure 8.

Model dimensions for a pier with one helix (mm).

Figure 11. Creep displacement for a pier with one helix at a design temperature of -5C for various axial loads.
-0.5 Displacement (mm) -0.4 -0.3 -0.2 -0.1 0.0 28 kN 56 kN 83 kN 111 kN

203

1270 610 4166 2286 Soil Volume

10 20 Time (years)

30

Figure 12. Creep displacement for a pier with one helix at a design temperature of 10C for various axial loads.
1651 1651
-1.6 Displacement (mm)

Figure 9. Model dimensions for a pier with two helixes (mm).

-1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0 10 20 Time (years) 30 28 kN 56 kN 83 kN 111 kN

surrounding the helical piers is modeled as a cylindrical volume. The dimensions used in the final analysis models are given in Figures 8 and 9. 5.3 Analysis of the results

Creep settlement (displacement) deduced from the FEA as a function of temperature is plotted in Figures 1015. Negative values imply settlement rather than heave. For the given soil, the creep displacement under various design loads is less than 3 mm at temperatures from 1 to 10C, which is less than a typical allowable settlement.
1310

Figure 13. Creep displacement for a pier with two helixes at a design temperature of 1C for various axial loads.

5.4

Design example

In order to design helical piers in the given frozen silt, the following information is needed: design temperature and allowable displacement after 25 years or less.

-0.6 Displacement (mm) -0.5 -0.4 -0.3 -0.2 -0.1 0.0 0 10 20 Time (years) 30 28 kN 56 kN 83 kN 111 kN

-3.0 -2.5 Displacement (mm) -2.0 -1.5 -1.0 -0.5 0.0 0 10 20 Time (years) 30 28 kN 56 kN 83 kN 111 kN

Figure 14. Creep displacement for a pier with two helixes at a design temperature of 5C for various axial loads.
-0.4 Displacement (mm) -0.3 -0.2 -0.1 0.0 28 kN 56 kN 83 kN 111 kN

Figure 16. Design example for a pier with one helix at 1C.
-1.6 Displacement (mm) -1.4 -1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0 10 20 Time (years) 30 28 kN 56 kN 83 kN 111 kN

10 20 Time (years)

30

Figure 15. Creep displacement for a pier with two helixes at a design temperature of 10C for various axial loads.

Figure 17. Design example for pier with two helixes at 1C.

When this information is obtained, the following steps will be performed: 1. Select the proper chart matching the design temperature. 2. Select design life and maximum allowable displacement. 3. Select a design load that yields a predicted displacement that is smaller than the allowable displacement. The following example illustrates the design method. A single helical pier foundation should be designed for a 100 m long wall with a design load of 20 kN/m. A pier with an allowable capacity of 111 kN should be used. The design temperature is 1C and the allowable displacement after 20 years is a) 25 mm, b) 1.5 mm (the small displacement is selected for the illustration). 5.4.1 Solution: (a) Select the chart matching -1C (Figures 10, 13). Note that all loads yield smaller displacement than the allowable value of 25 mm. Then, for the most economical design, select a design load of 111 kN per pier. The required spacing between the piers becomes 111 kN/(20 kN/m) 5.55 m. If the structural considerations allow this, the total number of the piers is 100 m/(5.55 m/pier) 1 pier 19 piers. 5.4.2 Solution: (b) Select the chart matching 1C for a single pier (Figure 10). If the piers were loaded to their capacity
1311

of 111 kN, the allowable displacement of 1.5 mm would be exceeded in 11 years. Consequently the piers need to be loaded by a smaller load. After 20 years a load of 83 kN per pier would yield 1.3 mm displacement and this is smaller than the allowable maximum displacement and is therefore acceptable. To obtain a load of 83 kN per pier, the spacing between the piers becomes 83 kN/(20 kN/m) 4.15 m and the total number of the piers is 100 m/(4.15 m/ pier) 1 pier 25 piers. Again, check if the 4.15 m spacing meets the structural requirements. See Figure 16 for an illustration of this design example. Check if piers with a double helix would yield a more economical design. Select the chart matching 1C for a double pier (Fig. 13). Again, all loads yield smaller displacement than the allowable value of 25 mm. Then, for the most economical design, select a design load of 111 kN per pier. The required spacing between the piers becomes 111 kN/(20 kN/m) 5.55 m. If the structural considerations allow this, the total number of the piers is 100 m/(5.55 m/pier) 1 pier 19 piers. Check if 19 double helix piers are less expensive than 25 single helix piers. See Figure 17 for an illustration of this design example. 5.5 Conclusions and recommendations for design Even if specific design guidelines for frozen ground have not been in existence before, the piers are used

successfully as a foundation for boardwalks, utilidors and other structures. However, engineers have been hesitating to specify helical piers as building foundations, because of the lack of design guidelines. The design method suggested here is very simple and helpful in designing piers for the given frozen soil. The design curves do not apply to any other soil, and even if the given creep displacements are insignificant, other soil may produce larger settlements. To properly design any foundations in frozen ground, the engineer still needs the creep parameters for the foundation soil. These parameters are not readily available for the engineering community. Therefore, actual soil testing needs to be conducted for various soils to create a library or database of soils encountered in cold regions. These tests would include strength parameters in thawed conditions, and creep and strength parameters at several frozen temperatures under various loads. Currently, to analyze any other soil, the engineer needs to contact the authors for FEA runs to produce the creep curves.

Ladanyi, B. & Johnson, G.H. 1974. Behavior of Circular Footings and Plate Anchors Embedded in Permafrost. Canadian Geotechnical Journal 11: 531553. Liu, H., Zubeck, H., Schubert, D., Baginski, S. & Hsieh, Y. 1999a. Behavior of Helical Piers in Frozen Ground. NAFEMS World Conference 99. Proc. April 1999, Rhode Island, USA: 12711280. East Kilbride, UK: The International Association for the Engineering Analysis Community. Liu, H., Zubeck, H. & Baginsky, S. 1999b. Evaluation of Helical Piers in Frozen Ground. Proc. Tenth International Conference on Cold Regions Engineering, Lincoln, NH, 16-19 August 1999: 232242. Reston, VA, USA: American Society of Civil Engineers. Liu, H., Schubert, D., Zubeck, H., & Baginski, S. 2000. Application and Analysis of Helical Piers in Frozen Ground. The 5th Annual Alaska Water and Wastewater Management Association International Research and Development Conference on Rural Sanitation, Proc. April 2526, 2000, Fairbanks, Alaska: 139151. Anchorage: AWWWMA. Morgenstern, N., Roggensack, W. & Weaver, J. 1980. The Behavior of Friction Piles in Ice and Ice-Rich Soils. Canadian Geotechnical Journal 17: 405415.

REFERENCES
A.B. Chance Co. 1996. Helical Pier Foundation System Technical Manual, Bulletin 0196.

1312