Skin friction resistance of expansion steel pipe piles in sandy soils

Abstract

This study investigates the use of expansion steel pipe piles for the reinforcement of relatively loose (sandy) soil under small building foundations. The expansion steel pipe piles used in this study are frequently used as rock bolts in tunnel excavation in Europe. The diameter of these pipes expands from 36 to 54 mm through the application of high‐pressure water. In this study, field load tests confirm the reinforcement effect of expansion steel pipe piles. The results can be summarized as follows: (a) The region of the surrounding soil compressed by the expansion of the piles is approximately 8 times (450 mm) the pipe diameter, as measured from the pile center. (b) The maximum friction resistance of the expansion steel pipe piles is approximately 3 times greater than that of conventional steel pipes with the same diameter. (c) The finite cavity expansion theory was used to estimate the maximum friction resistance, and the analytical results correlate strongly with the field test results.

Introduction

In the construction of small buildings, it may be necessary to first reinforce the ground underneath the foundations if soil tests indicate a low bearing capacity or the potential for differential settlement. Common ground reinforcement methods include small-diameter piles (eg, steel pipe piles or precast concrete piles) with diameters of 300 mm or less, deep soil mixing, and surface soil improvement. For small-diameter piles and deep soil mixing, various measures can be taken to obtain a high bearing capacity. For example, in the case of steel pipe piles, tip blades may be included to increase the end bearing capacity of the pile, or helical wings may be attached to increase the shaft resistance of the pile. In the case of deep soil mixing, fixed blades can be attached to prevent the counterrotation of the soil, and an increase in the number of stirring blades attached can lead to more thorough mixing.

Currently, authors are focusing on the outer-diameter expansion of steel pipe rock bolts, which are used to provide a restraining force in loose grounds during tunnel excavation and ground reinforcement piles for small buildings. Previous research has included tensile load tests for the expansion of steel pipe rock bolts for tunnels under various geothermal conditions to assess the anchoring performance and durability. However, research regarding their use as piles could not be found.

This paper summarizes the results of experiments that investigated the effects of outer-diameter expansion on friction. This paper shows (a) a comparison of Swedish weight sounding tests (hereafter referred to as SWS tests ) on the surrounding soil before and after the expansion of the pipes, (b) an investigation of the shaft friction of expansion steel pipe piles via vertical loading tests, and (c) a comparison of the theoretically calculated values with the experimental values of the shaft friction. Certain results have been previously reported.

Description of expansion steel pipe piles

As shown in Figure , an expansion steel pipe pile consists of a main steel pipe with ZAM (zinc-aluminum-magnesium) plating (for corrosion protection) and sealing sleeves at the ends. The ZAM-plated steel pipe used to fabricate the piles is formed using 400 N class structural steel with a Young's modulus of 2.05 × 105 N/mm², a tensile strength of 440 N/mm², and a yield strength of 295 N/mm². The sealing sleeves are constructed of rolled steel for general structures (SS 400). The steel pipe, which has an outer diameter of 54 mm and a wall thickness of 2 mm, is pressed flat and folded into a heart shape, with an outer diameter of 36 mm, and the sealing sleeves are welded to both ends. Approximately 250 000 of these units have been used in tunnel construction.

Schematic of an expansion steel pipe pile

The sleeve on the pile head has a port with Ø = 3 mm that allows water to be injected. Water is injected under high pressure via a special apparatus known as a seal head. When the water pressure inside the pipe exceeds 5 MPa, the folded steel pipe begins to unfold. At 12 MPa, the pipe opens fully (except for a strip approximately 70 mm wide at each end), and the entire tube expands to Ø = 54 mm (−0 to +1 mm) as the pressure approaches 20 MPa. When this condition is reached, a safety device activates, and pressurization stops. When the water injection is completed, and the seal head is removed, the pressure inside the steel pipe is released, but the water remains in the pipe. Figure shows the process of expansion, and Figure shows cross sections of the pipe before and after expansion.

Process of expansion

Cross section of an expansion steel pipe

In actual construction, the pipe is buried up to the depth of the foundation, leaving only the sleeve of the pile head exposed. In this study, however, an additional 250 mm of the pile was left exposed above ground to confirm complete expansion of the pipe. The pile was pressed into the soil, with no rotation, to a predetermined depth at a rate of 0.5 m/min.

To confirm the compressive strength of the steel pipe material, steel pipes before processing (original pipes) and pipes folded and then expanded (expansion steel pipes) were each cut into 300 mm segments, on which compression tests were performed. The tests were conducted on 3 specimens of each type. The average proportional maximum load was 96.3 kN (stress equal to 316.7 N/mm²) for the original pipes and 104.6 kN (stress equal to 344.1 N/mm²) for the expansion steel pipes. The strength of the expansion steel pipes increased by approximately 9% via work hardening.

Compaction effect of surrounding soil Test overview

To assess changes in the properties of the soil surrounding the expansion steel pipe piles before and after expansion, test areas of various sizes, including 3200 × 3200 mm, 2900 × 2900 mm, and 2600 × 2600 mm, at a depth of 2.5 m were prepared, as shown in Figure . The soil used was Masado (50.0% gravel, 41.3% sand, and 8.7% fine-grained soil) produced in Shigaraki (Shiga Prefecture, Kouka City). A 60 kg plate rammer was used once for every 200 mm of thickness.

Diagrams of test areas. (A) Pile spacing, 600 mm. (B) Pile spacing, 450 mm. (C) Pile spacing, 300 mm

The buried depth of the expansion steel pipe piles was 2.0 m, and 9 piles were placed at equal intervals of 600 mm, 450 mm, and 300 mm in the order of the arrows shown in Figure A. SWS tests were performed before and after expansion, and the state of compaction of the surrounding soil was determined. The SWS tests were performed in 4 locations: 2 at intermediate points adjacent to the piles and 2 at intermediate points diagonally. It was assumed that the test area consisted of low-capacity ground and, as such, only static penetration would occur. Therefore, the load could be quickly reduced by 1 stage when penetration began, and, when there was no penetration, the load was increased. Measurements were taken at intervals of 50 mm in depth.

SWS test results

Figure shows the average self-weight penetration load (W_sw [kN]) of the 4 points at each depth before and after expansion of the piles, and Table shows the average W_sw measured from the ground surface to the tip of the pile. The table also shows the proportion of the cross-sectional area occupied by the piles before and after expansion in the intermediate area between the center pile and the outer piles (indicated by the dotted lines in Figure ).

Test ground Wsw results. (A) Pile spacing, 600 mm. (B) Pile spacing, 450 mm. (C) Pile spacing, 300 mm

Increase in W_sw for the surrounding ground

Pile spacing	600 mm	450 mm	300 mm
Cross section ratio of piles
Before expansion	0.28%	0.50%	1.13%
After expansion	0.64%	1.13%	2.54%
Average Wsw from 4 points of measurement
Before expansion	0.602 kN	0.580 kN	0.578 kN
After expansion	0.622 kN	0.633 kN	0.697 kN
Increase	3.3%	9.1%	20.6%
Average Wsw from 2 intermediate points adjacent to the piles
Distance from pile	300 mm	225 mm	150 mm
Before expansion	0.606kN	0.575 kN	0.559 kN
After expansion	0.638 kN	0.641kN	0.694 kN
Increase	5.2%	11.5%	24.2%
Average Wsw from 2 intermediate points diagonally
Distance from pile	424 mm	318 mm	212 mm
Before expansion	0.597 kN	0.584 kN	0.597 kN
After expansion	0.606 kN	0.625 kN	0.700 kN
Increase	1.5%	7.0%	17.3%

In Figure , no change in W_sw occurred between the depths of 200 and 250 mm for the 3 pile spacings. This is due to the expansion of the pipes above the ground surface, which caused fissures to form in the surrounding soil. It was noted that when the piles were fully expanded, the fissures were approximately 100-200 mm in length. It was estimated that the fissures propagated to a depth equal to the length, and there were no constraining forces on the piles to the depth of the fissures (the fissures situation is shown in Figure ). Therefore, the effective length (ie, the range where friction resistance is generated) was taken to be the total length minus 200 mm from the ground surface and 150 mm from the tip of the buried end.

Fissures in the ground

As shown in Figure and Table , W_sw increased after expansion in all cases, where the difference increased as the pile spacing decreased. It is believed that this was due to the influence of compaction caused by the expansion of the pipes. For each test area, the soil was prepared using the same compaction method. However, the process was more difficult for smaller test areas and, therefore, the values of W_sw before expansion were lower.

The zone of influence of pile expansion was investigated. Figure shows the relation between the ratio of the distance between the piles to the pile diameter and the change (%) in W_sw. From Figure , it can be observed that the effect of expansion became negligible when the relative distance was nearly 8 times the pile diameter. Therefore, in the field experiments, the piles were placed at intervals of at least 1 m.

Relationship between the relative distance to the pile diameter and the increase in Wsw

Investigation of shaft friction resistance via vertical load tests Comparison of expansion steel pipe piles and single pipes

Continuous axial tensile load tests were conducted using expansion steel pipe piles and straight steel pipes (hereafter referred to as single pipes), with the same diameters of Ø = 54 mm (expansion piles were expanded). The maximum friction force per unit area was obtained for both types of pipe. The single pipe was an open-ended type, and it was pressed into the ground without rotation using the same installation method as that for the expansion steel pipe piles.

The tests were performed at Toyonaka and Fukutsu (2 sites). The pipes at Toyonaka were installed at depths of 3.8 m and 4.8 m, and the pipes at Fukutsu were installed at a depth of 4.3 m.

The distances between pipes were 2 m. Axial tensile load tests were performed after expansion of the pipes and 7 days of curing. Soil data for the 2 sites are given in Figure A and B. The distance between the locations of the standard penetration test (SPT) and the test pile was between 8 and 10 m. Figure shows the test apparatus. The reaction force of the load was measured using a reaction plate consisting of a steel H-beam. The axial tensile load tests were conducted using the continuous loading method, with a loading rate of 4 kN/min. The maximum value was taken as the load value either immediately before the displacement of the pile head reached 10% of the outer diameter of the expansion steel pipe or when the displacement of the head increased rapidly without an increase in the load.

Axial tensile load test apparatus

Figure shows the relation between the tensile load and the pile head displacement. It is clear from the figure that the tensile resistance of the expansion steel pipe piles was considerably larger than those of the single pipes. The maximum tensile load is defined as the tensile load when the displacement is within 10% of the pipe diameter (ie, 5.4 mm). The maximum tensile loads are indicated by the arrows (▾) in Figure . Table shows the values from the test. The maximum friction force per unit area τ_s (kN/m²) was obtained by dividing the maximum tensile load by the cross-sectional area of the aforementioned effective length of the pile. For the single pipes, the effective length was taken as the total buried depth. As shown in the table, τ_s was 2.7-3.0 times greater in the expansion steel pipes than that in the single pipes at the same depth.

Relationship between the tensile load and pile head displacement. (A) Toyonaka (buried depth, 3.8 m). (B) Toyonaka (buried depth, 4.8 m). (C) Fukutsu (buried depth, 4.3 m)

Results of axial tensile load tests using expansion steel pipe piles and single pipes

Test area	Pile type	Buried depth (m)	Average N-value (-)	Maximum tensile load (kN)	Pile head displacement at maximum load (mm)	Maximum friction force per unit area τs (kN/m2)
Toyonaka	Expansion steel pipe	3.8	3.3	40.7	2.8	67.8
	Single pipe	3.8	3.4	14.9	1.1	23.3
	Expansion steel pipe	4.8	3.4	52.9	2.7	71.2
	Single pipe	4.8	3.7	21.7	5.4	26.8
Fukutsu	Expansion steel pipe	4.3	8.6	75.0	5.4	105.7
	Single pipe	4.3	8.4	22.7	4.5	35.3

N-value represents the number of blows of SPT.

Relation between shaft friction resistance per unit area and soil parameters

Tensile load tests were conducted on a total of 10 expansion steel pipe piles at 3 test sites, Toyonaka, Fukutsu, and Karatsu. After pipe expansion and 7-10 days of curing, the relationships between the maximum shaft frictional resistance and soil parameters were investigated through an axial tensile load test and static axial compressive load test.

The axial tensile load tests were performed using a single-cycle stage loading method, with at least 8 loading stages and hold times of 30 minutes. The same apparatus used for the previous tests (Figure ) was used. The maximum tensile strength was taken as the tensile load when the displacement of the pile head reached 10% of the diameter of the expansion steel pipe or the maximum load that could be held if the load could not be held for 30 minutes at 10% or less.

As shown in Figure , the static axial compressive load test was performed by first pressing the pile to a predetermined depth, raising it by 20-30 mm before expansion, and then loading it with no tip resistance. Heavy machinery was used to provide the force. As with the axial tensile load tests, the loading was conducted using a single-cycle stage loading method, and the maximum compressive strength was determined in the same manner as for the axial tensile strength.

Procedure for axial compressive load tests

The soil boring logs for each test site, the SPT results, and the SWS test results at the positions closest to the SPT positions are shown in Figure . The SPTs were conducted within 10 m of the test piles, and the SWS tests were conducted 0.5 m from the test piles prior to the installation of the piles.

Soil boring logs and converted N-values of each test site. (A) Toyonaka (Toyonaka City Osaka). (B) Fukutsu (Fukutsu City Fukuoka). (C) Karatsu (Karatsu City Saga)

Figure shows the relation between the tensile load and the pile head displacement obtained from the tensile load tests, and Figure shows the relation between the compressive load and the pile head displacement obtained from the compressive load tests. Table summarizes the test results, where the letter in the test number is the first letter of the name of the test site. F1 had a particularly large value because the tip of the pile reached a sandy layer with a large N-value (>GL −4 m). Although K4 and K5 have almost the same pile length, it is thought that the difference in the maximum friction force per unit area is due to the larger average converted N-value at K5.

Relationship between the load and pile head displacement (axial tensile load test)

Relationship between the load and pile head displacement (axial compressive load test)

Vertical load test results

Test no.	Effective buried depth (m)	Maximum strength (kN)	Maximum friction force per unit area τs (kN/m2)	Average N-value $\overline{N}$ (-)	Average N′-value ${\overline{N}}^{\prime }$ (-)
Axial tensile load test
T1	4.25	55.6	78.4	3.4	3.5
T2	3.4	40.1	70.9	3.3	2.9
F1	4.25	80.1	109.0	8.4	7.5
F2	3.4	42.0	73.2	9.0	3.8
F3	2.4	24.3	60.1	7.9	3.8
Axial compressive load test
K1	4.4	60.0	81.6	8.4	4.8
K2	3.9	44.0	67.7	8.0	4.7
K3	3.2	45.0	84.7	6.8	5.4
K4	2.65	26.0	53.2	6.5	3.4
K5	2.4	32.0	80.9	6.3	4.3

Figure shows the relation between the N-value at the effective buried depth of the expansion steel pipe piles obtained from the SPTs and the maximum friction force per unit area τ_s. Figure shows the relation between τ_s and the converted N-values obtained from the SWS tests and Equations 1 and 2:[Image Omitted. See PDF][Image Omitted. See PDF]where N′ represents the converted N-value (−), W_sw represents the load (in kN), and N_sw represents the number of half revolutions per meter.

Relationship between the maximum friction force per unit area (τs) and N¯-value

Relationship between the maximum friction force per unit area (τs) and N¯′-value

The representative soil of each site was classified as sandy ground, as the SPT results showed this to be the most prevalent soil type. For this reason, the slopes in the initial stages of loading ware different in tests T1 and T2 because of enclosed cohesive soil layers.

In Figures and , it can be observed that the maximum load was reached before the displacement reached 10% of the pile diameter (except for F1). Although there are differences between Figures and , no significant difference was observed in τ_s in both tests; therefore, they can be examined side by side.

As shown in Figure , when the cohesive soil layers are included, different behaviors can be observed. Therefore, the results for T1 and T2 were excluded, and Equations 3 and 4 were obtained through the average values of τ_s/$\overline{N}$ and τ_s/${\overline{N}}^{\prime }$.

Here, σ represents the standard deviation (in kN/m²).

[Image Omitted. See PDF] [Image Omitted. See PDF]

In Figures and , these average lines and the lines for the average values ±σ are shown as solid lines. From a comparison of the figures, the τ_s obtained from N-value is smaller than the τ_s obtained from N′-value at the same values. This difference occurred because, as shown in Table , the N-values were greater than the N′-values as a whole. In addition, there was more scatter in the results in Figure . This difference seems to have occurred because the SWS test measures every 0.25 m, and the SPT measures every 1 m. Although the SPT provides conservative results, the SWS tests are the most common type for assessing soil properties in the design of small structures, and are also more practical from the viewpoint of accuracy.

In Figure , the average line for driven piles, as given by the Recommendations for Design of Building Foundations (hereafter referred to as the J.A.I formula), is indicated by a dashed line. The maximum friction force per unit area of the expansion steel pipe piles is 2.5 times larger compared to this average line value. In addition, Figure shows the average values (σ = 1.4N′) of the maximum shaft friction resistance per unit area in the case of straight rotary penetration piles without tip blades, given by Hirose et al. Because the number of tests for the expansion steel pipe piles was relatively small, the standard deviation was large, approximately 2.9 times that of the straight rotary penetration piles.

Comparison with the calculated results using the elasto-plastic analysis method

This section compares the maximum shaft friction resistance, which increases through the expansion of the steel pipe, with results calculated using an elasto-plastic analysis method based on the finite cavity expansion theory, which has been previously presented in the literature.

Corresponding to the behavior analysis model of the tapered piles subjected to a compressive load presented in the above literature, Figure shows an analysis model adopted for the expansion steel pipes. The maximum shaft friction resistance per unit area τ_s acting on the expansion steel pipe pile in sandy ground in this model is expressed by Equation 5 where σ_h represents the horizontal stress due to steel pipe expansion (kN/m²) from radius r_z (m) to r_z + dU_z (m)[Image Omitted. See PDF]μ represents the friction angle between the soil and steel pipe (radian). The friction angle μ in Equation 4, which is proportional to the internal friction angle φ (radian) of the soil via the coefficient λ (-), is given by Equation 6[Image Omitted. See PDF]

Friction force analysis thin-layer model

The horizontal stress σ_h was evaluated using Equation 7, which assumes an elasto-plastic state of the soil surrounding the piles. This is because, as also shown in the literature, when expanding the cavity from radius r_z to r_z + dU_z, the elasto-plastic limit is reached when dU_z/r_z = 0.005 for the surrounding soil and 0.5 for the expansion steel pipes used in this research.

[Image Omitted. See PDF]

Here, K₀: active soil pressure at rest (-); γ: unit weight of soil (kN/m³); Z: depth (m); G: shear modulus of the soil (= E/[2 (1 + ν)]) (MN/m²); E: modulus of elasticity of the soil (MN/m²); ν: Poisson's ratio of the soil (-).

The experimental results of (B) Fukutsu and (C) Karatsu, with little cohesive soil in Figure , were compared with the results calculated using the above analytical method. The analytical method assumed uniform soil properties along the pile. The values of the internal friction angle φ (degree) and the modulus of elasticity of the soil E were φ = 20N + 20 and E = 1.4N (MN/m²), respectively, as given in the J.A.I. formula using the N-values obtained from the SPTs. The values of the unit weight of the soil, the soil pressure at rest, and Poisson's ratio were γ = 18 (kN/m³), K₀ = 0.5 (-), and ν = 0.3 (-), respectively, which are typical for alluvial sandy soils. According to Ref. 16, the value of the coefficient λ in Equation 5 is in the range of tan(1/2) to tan(3/4) for steel piles; therefore, we used the intermediate value of that range (ie, λ = tan(0.625) = 0.7 [-]).

The analytic and experimental values (F1 to F3 and K1 to K5 in Table ) of the maximum friction resistance force are shown in Table and Figure . From Figure , it can be observed that although the analytic values were slightly lower, the averages of these ratios were 1.04 times greater, and the results were strongly correlated. From these results, it can be concluded that the mechanism by which the maximum friction resistance increases in expanded pipes and single pipes of equal diameter can be explained using the finite cavity expansion theory. Further investigations planned.

Comparison of calculated values and experimental values of the maximum friction resistance

	F1	F2	F3	K1	K2	K3	K4	K5
Calculated value (A)	64.4	51.2	26.6	72.5	64.2	42.1	31.1	24.6
Experimental value (B)	80.1	42.0	24.3	60.0	44.0	45.0	26.0	32.0
Ratio: (A)/(B)	0.80	1.22	1.09	1.21	1.46	0.94	1.20	0.77

Calculated value (A) and experimental value (B) are in kN.

Comparison of the experimental values and calculated values for the maximum friction resistance

Summary

In this paper, the results of the experimental and theoretical analyses have been presented to investigate the vertical bearing capacity performance of expansion steel pipe piles in sandy soil for use in the ground reinforcement of small building foundations.

The findings obtained in this study can be summarized as follows:

1. SWS tests were performed before and after expansion of the pipe piles in relatively loose artificial soil. The increase in the degree of compaction of the surrounding soil due to expansion decreased with increasing the distance from the piles and generally approached zero at a distance of approximately 8 times the diameter of the expansion pile (450 mm).
2. Three sets of axial tensile load tests on expansion steel pipe piles and single pipes with equal diameters (straight pipes with open ends). The maximum friction force per unit area of the expansion steel pipe piles was 2.7-3.0 times that of the single pipes.
3. The relation between the average N-values obtained through SPTs, the average converted N-values obtained through SWS tests, and the maximum friction resistance per unit area τ_s along the effective buried length (the section where frictional resistance is generated) of the expansion steel pipe piles were as follows:[Image Omitted. See PDF][Image Omitted. See PDF]Because of the small number of data points and the large variation in results, further accumulation of data is necessary.
4. Based on the assumption that an increase in the horizontal stress of the soil acting on the side surfaces of the piles caused by expansion increased the shaft friction resistance, an analysis was performed using the finite cavity expansion theory, and these calculated results were found to correlate strongly with the results gained from the experiments.

Acknowledgments

The authors express their deepest appreciation to those who helped with the experiments in this paper, with a great deal of assistance from Mr. Shinichi Egami of Hokoku Engineering Co., Ltd. and Mr. Ken Katsuno of Space Engineering Co., Ltd.

Disclosure

The authors have no conflicts of interest to declare.

Notes:

¹⁰⁰³Swedish weight sounding. “Swedish weight sounding” test is often called “SWS test” in Japan; it is one type of static penetration test. This method is standardized by the Japanese Industrial Standards (JIS), but uses a different maximum rotational diameter and rod diameter from those specified by overseas standards (European Standards). The SWS equipment consists of a screw point, sounding rod, rotary handle, and 5 weights with a total weight of 1 kN. The screw point attached to the rod is step loaded until it penetrates into the soil stratum, where the load needed to penetrate is described as W_sw. When the screw point cannot penetrate under the maximum load of 1 kN, it is rotated by using the handle, and the number of half turns is converted into a value required for 1 m of penetration, referred to as N_sw.

¹⁰⁰⁴ZAM is a registered trademark of Nisshin-Steel Co., Ltd.