See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/332230491

Predicting the Axial Capacity of Screw Piles Installed in Canadian Soils

Conference Paper · October 2007

CITATIONS READS
14 117

2 authors:

Kristen Tappenden David Sego

Government of Alberta University of Alberta
8 PUBLICATIONS   33 CITATIONS    180 PUBLICATIONS   2,254 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Geotechnical Study of Oil Sands Dewatering using Thin Layer Freeze-Thaw View project

All content following this page was uploaded by Kristen Tappenden on 05 April 2019.

The user has requested enhancement of the downloaded file.

 OttawaGeo2007/OttawaGéo2007

PREDICTING THE AXIAL CAPACITY OF SCREW

PILES INSTALLED IN CANADIAN SOILS
Kristen M. Tappenden
AMEC Earth & Environmental, Edmonton, Alberta, Canada
David C. Sego
Civil & Environmental Engineering – University of Alberta, Edmonton, Alberta,
Canada

ABSTRACT
The results of 26 full-scale static axial load tests are presented for screw piles installed in Alberta and British Columbia
since 1998, and the effectiveness of three design methods are evaluated for predicting the axial capacity of screw piles
in cohesive and cohensionless soils. Theoretical formulations for capacity calculations are examined alongside the
LCPC direct pile design method, and an empirical relationship correlating the installation torque to the ultimate screw
pile capacity.

RÉSUMÉ
Les résultats de 26 essais de chargement statiques sont présentés pour des pieux vissés installés en Alberta et en
Colombie Britannique depuis 1998 ainsi que l’évaluation de trois méthodes de conception utilisées pour prédire la
capacité axiale des pieux vissés en sols pulvérulents. Les formules théoriques utilisées pour fins de calcul de la
capacité sont examinées en parallèle avec la méthode du LCPC et une relation empirique établissant la corrélation
avec la capacité ultime d’un pieu vissé.

1 INTRODUCTION Screw piles are frequently installed to depths of less than

10 m, and are favoured for their rapid installation,
This paper presents three design methodologies for typically requiring about 30 minutes per pile using a two
calculating the axial capacity of screw piles loaded in person crew.
uplift or compression in cohesive and cohesionless soils.
An indirect design approach based on theoretical Screw piles are not well-suited for installation into very
formulations of bearing capacity and skin friction is hard, dense, or gravelly soils, where shallow refusal or
presented, followed by the LCPC direct pile design damage to the steel helical plates could occur. However,
method (based on the cone penetration test), and an commercial screw piles have been fabricated with helices
empirical relationship correlating the installation torque to of up to 25 mm thick to offset the likelihood of structural
the axial screw pile capacity. The results of 26 static load damage during installation in adverse conditions.
tests conducted in tension and compression on screw pile
installations in cohesive and cohesionless soils are
summarized. Each of the capacity prediction methods 2 SHALLOW VERSUS DEEP FAILURE IN UPLIFT
are applied to the screw piles tested, and the
effectiveness of the methods are discussed in light of For screw piles loaded in uplift, a distinction must be
discrepancies between the measured and predicted made between shallow and deep embedment. For
capacities. shallow screw piles in uplift, the bearing failure zone
above the uppermost helix will extend to the ground
1.1 General surface, as evidenced by tensile cracking and surface
heave (Narasimha Rao et al., 1993). For deep screw
Screw piles, also known as helical piles or screw piles in uplift, the bearing failure zone above the
anchors, are structural, deep foundation elements used to uppermost helix will be contained below the surface. The
provide stability against compressive, tensile, and lateral demarcation between the shallow and deep failure
loads. Screw piles consist of one or more circular, helical condition is expressed as the critical embedment ratio,
plates affixed to a central shaft of smaller diameter. Hcrit, which is equal to the embedment depth of the
uppermost helix divided by its diameter.
Screw piles are embedded into the soil by the application
of a turning moment to the pile head, causing the helices In cohesionless soils, the critical embedment ratio, Hcrit,
to penetrate the ground in a “screwing” motion without beyond which the deep failure condition prevails, is
creating vibration or spoil. Installation of screw piles is dependent upon the soil friction angle, as given in Table 1
typically accomplished using a torque head affixed to the (after Meyerhof and Adams 1968). For screw piles
arm of a backhoe or a trailer-mounted hydraulic boom. installed in cohesive soils, Narasimha Rao et al. (1993)

1608
 OttawaGeo2007/OttawaGéo2007

suggest that a critical embedment ratio of about 4.0 be installation disturbance and interference between closely-
used, on the basis of laboratory uplift tests on model spaced helical plates.
screw piles installed in clay. Whether an embedded
screw pile may be considered as shallow or deep will The ultimate screw pile capacity calculated as per the
affect the parameters chosen in the calculation of the cylindrical shear model will be the sum of the bearing
axial capacity. capacity below the bottom helix in compression, or above
the top helix in tension, plus the skin friction acting along
the cylinder of soil circumscribed between the top and
Table 1: Variation of critical embedment ratio with soil bottom helices. Skin friction acting along the section of
friction angle (after Meyerhof and Adams 1968) pile shaft between the uppermost helix and the ground
surface may be considered for compressive loading, or
for uplift loading under the deep failure condition, but
Soil friction angle, Φ Critical Embedment Ratio, Hcrit should be neglected for uplift under the shallow failure
25° 3 condition (Narasimha Rao et al. 1993; Mitsch and
Clemence 1985).
30° 4
35° 5
40° 7 4 CALCULATION OF AXIAL SCREW PILE
45° 9 CAPACITY

Screw pile axial capacity calculations may be performed

3 FAILURE MODELS FOR AXIALLY LOADED using indirect (theoretical) methods, direct methods, or
SCREW PILES empirical methods. Indirect or theoretical methods,
although involving some degree of empiricism, are based
There are two primary failure models established in the on established equations of bearing capacity and skin
literature for describing the behaviour of screw piles friction used to describe the behaviour of piles and buried
under axial load—these are the cylindrical shear model anchors in compression and uplift. The equations
and the individual plate bearing model (Narasimha Rao et depend on the accurate estimation of characteristic soil
al. 1993). The choice of the most appropriate failure parameters. Direct methods eliminate the intermediate
model is considered to be dependent on the screw pile calculation of soil parameters by relating in-situ test
geometry. For multi-helix screw piles, the value of the results, such as from a cone penetration test (CPT), to
inter-helix spacing ratio, which is equal to the spacing (S) the unit bearing resistance and skin friction of a pile.
between adjacent helical plates divided by their average Empirical methods for predicting the axial capacity of a
diameter (D), is the parameter which determines the use screw pile attempt to directly relate the torque measured
of either model. during installation to the ultimate pile capacity; this
concept is analogous to the relationship between pile
3.1 Individual Plate Bearing Model driving effort and capacity. A detailed overview of the
three design methodologies is given in the following
The individual plate bearing model is applicable to single- subsections.
helix screw piles and multi-helix screw piles with an inter-
helix spacing ratio equal to or greater than 3.0 (CFEM 4.1 Theoretical Methods
2006). The model assumes that bearing failure occurs
above or below each individual helix in tension or A summary of the appropriate theoretical equations for
compression, with negligible interference between calculating ultimate unit bearing and frictional resistance
multiple failure zones. of embedded screw piles is given below. The ultimate
screw pile capacity is determined by application of the
The ultimate axial capacity given by the individual plate equations in conjunction with the appropriate failure
bearing model will be the summation of the bearing model, i.e., the cylindrical shear model or the individual
capacity above/below each helix; additionally, skin friction plate bearing model.
along the section of pile shaft between the uppermost
helix and the ground surface may be considered for 4.1.1 Bearing
compressive loading or for uplift loading under the deep
failure condition, but should be neglected for uplift under The ultimate bearing resistance, Qb, mobilized beneath a
the shallow failure condition (Narasimha Rao et al. 1993; screw pile helix is given by Equation [1] for cohesive soils,
Mitsch and Clemence 1985). and Equation [2] for cohesionless soils.

3.2 Cylindrical Shear Model

π (D2 − d 2 )
The cylindrical shear model is applicable to multi-helix
Qb = (NcCu + γ' H) [1]
4
screw piles with inter-helix spacing ratios less than 3.0.
The model assumes that during axial loading, a frictional, π (D2 − d 2 )
Q b = Nq* γ' H [2]
cylindrical failure surface is formed between the top and 4
bottom helices. The failure surface is attributed to

1609
 OttawaGeo2007/OttawaGéo2007

10000
Where:
H = depth of helix below ground surface (m)
D = diameter of helix (m)
d = diameter of pile shaft (m)
Nc = bearing capacity factor for cohesive soil
Cu = undrained shear strength of cohesive soil (kPa)

Bearing capacity factor, Nq*

*
Nq = bearing capacity factor for cohesionless soil
3 1000
γ’ = effective soil unit weight (kN/m )

For calculating the bearing capacity under the lowermost

helix, the variable ‘d’ may be omitted from Equations [1]
and [2], assuming the formation of a soil plug within the
base of the screw pile shaft.

The value of the bearing capacity factor in cohesive soil, 100

Nc, is typically taken as 9.0 (CFEM 2006). In
cohesionless soil, values for the bearing capacity factor
*
Nq , such as those recommended by Vesic (1963) may be
used (Figure 1).

4.1.2 Uplift
10
The ultimate uplift resistance, Qu, mobilized above the 25° 30° 35° 40° 45° 50°
helix of a screw pile in tension is given by Equation [3] for Angle of internal friction Φ, degrees
cohesive soils, and Equation [4] for cohesionless soils. Figure 1: Variation of bearing capacity factor, from
various authors (after Winterkorn and Fang 1975)

π (D2 − d 2 )
Qu = (NcCu + γ' H) [3] 400
4
π (D2 − d 2 ) Φ= 45°
Q u = Fq γ' H [4]
4
100 40°
Shallow Breakout Factor, Fq

Where:
35°
Fq = breakout factor for cohesionless soil

30°
The bearing capacity factor for cohesive soil, Nc, to be
used for uplift should be determined by Equation [5], as 25°
dependant on the embedment depth of the uppermost 10
helix, H1, divided by the helix diameter, D1 (after Meyerhof
1976).

H 
Nc = 1.2  1  ≤ 9 [5]
 D1 
1
0.5 2.5 4.5 6.5 8.5 9.0
H/D
Values for the breakout factor, Fq, to be used in Figure 2: Variation of breakout factor, shallow condition
cohesionless soils have been determined by Das (1990) (after Das, 1990)
based on Mitsch and Clemence’s (1985) theory, and are
shown in Figure 2 for shallow screw piles (H1/D1 ≤ H/Dcrit)
and Figure 3 for deep screw piles (H1/D1 > H/Dcrit).

4.1.3 Shaft Friction

Skin friction may be considered to act along the length of

the screw pile shaft between the uppermost helix and the

1610
 OttawaGeo2007/OttawaGéo2007

400
Table 2: Recommended Ku values for screw piles (after
Mitsch and Clemence 1985)

Lateral earth pressure

Soil friction angle, Φ
coefficient in uplift, Ku
*
Deep Breakout Factor, Fq

25° 0.70
100
30° 0.90
35° 1.50
40° 2.35
45° 3.20

The Ku values suggested by Mitsch and Clemence (1985)

10
are approximately 40 percent less than the Ku values
25 30 35 40 45 proposed by Meyerhof and Adams (1968) for the uplift
Soil friction angle Φ, degrees capacity of buried foundations. The lower coefficients for
Figure 3: Variation of breakout factor, deep condition screw piles are attributed to the disturbance created by
(after Das, 1990) the churning action of the helices during installation;
Meyerhof and Adams’ (1968) values, however, were
based on essentially undisturbed soil (Mitsch and
ground surface for screw piles in compression, or for Clemence 1985).
deep screw piles in uplift. However, a distance equivalent
to one helix diameter should be subtracted from the
available shaft length, to account for the effect of bearing For compressive loading, values for the parameter KS
disturbance above the uppermost helix when in tension, may be calculated by Equation [9].
and for the void-forming or ‘shadowing’ effect above the
uppermost helix when acting in compression (Zhang
1999). The effective shaft length, Heff, over which the β
Ks = [9]
ultimate skin friction may be mobilized, is then as defined tan δ
in Equation [6].

Where:
Heff = H1 − D1 [6] β = skin friction design parameter for displacement
piles in sand (CFEM 2006)
δ = interface friction angle, equal to 0.6Φ for steel
For a screw pile embedded in cohesionless soil, the embedded in sand (Kulhawy 1984)
ultimate unit shaft friction, QS, may be calculated by
Equation [7] for compressive loading, and by Equation [8] In the case of a screw pile embedded in cohesive soil, the
for uplift loading. ultimate shaft friction mobilized under either compressive
or uplift loading may be calculated by Equation [10].

1 
Qs = π ⋅ d ⋅ Heff ⋅  γ ' H1 K s tan φ  [7] Qs = π ⋅ d ⋅ Heff ⋅ Cu ⋅ α [10]
 2 
1 
Qs = π ⋅ d ⋅ Heff ⋅  γ' H1 K u tan φ  [8]
 2  Where:
α = adhesion factor

Where: Values for the adhesion factor, α, such as those given in

Φ = effective soil friction angle the Canadian Foundation Engineering Manual for grouted
anchors installed in cohesive soils, may be used (CFEM
Ks = lateral earth pressure coefficient in compression
2006).
Ku = lateral earth pressure coefficient in uplift
4.1.4 Cylindrical Friction
Mitsch and Clemence (1985) have proposed values for
the parameter Ku on the basis of laboratory scale tests on Under the cylindrical shear failure model, frictional
model screw piles under uplift in sand, as shown in Table resistance is mobilized along the circumscribed cylinder
2. of soil bounded by the top and bottom helices of the
screw pile. Equations [11] and [12] may be applied in
cohesionless soil to calculate the ultimate cylindrical

1611
 OttawaGeo2007/OttawaGéo2007

friction, Qcyl, mobilized in compression and uplift, In an analysis of 91 screw pile load tests in uplift from the
respectively. In cohesive soil, Equation [13] should be published literature and the authors’ private files, Hoyt
used for both uplift and compression. and Clemence (1989) obtained good approximations of
-1
ultimate uplift capacities using Kt equal to 33 m for all
square-shaft screw piles and round-shaft screw piles less
-1
1 
(
Qcyl = π ⋅ D ⋅  γ' K s tan φ  ⋅ Hn2 − H12
2 
) [11] the 89 mm in diameter, 23 m for round shaft piles 89
-1
mm in diameter, and 9.8 m for round, 89-mm-diameter
piles with 219 mm diameter extension shafts (extending
1
Qcyl = π ⋅ D ⋅  γ' K u
2
tan φ  ⋅ ( H


n
2
− H12 ) [12] from the top helix to the ground surface). Hoyt and
Clemence (1989) averaged the installation torque over
Qcyl = π ⋅ d ⋅ Cu ⋅ α ⋅ (Hn − H1) the final distance of penetration equal to three times the
[13]
largest helix diameter, and all piles were multi-helix screw
piles loaded in uplift under the deep failure condition. As
a preliminary estimate, the Canadian Foundation
Where: Engineering Manual (CFEM 2006) suggests that a Kt
-1
Hn = depth of lowermost helix below ground (m) factor of 10 m be used in the absence of any site-
H1 = depth of uppermost helix below ground (m) specific correlations established by pile load testing. As
the torque relationship may only be used to predict the
4.2 Direct Methods (LCPC Method) screw pile capacity after installation has taken place, it is
best-suited for on-site production control rather than for
The LCPC method attempts to calculate the ultimate unit the actual design of piles in the office (Hoyt and
skin friction and end bearing of a pile on the basis of Clemence 1989).
scaling coefficients applied to an in-situ CPT profile of tip
resistance. The method is documented by Bustamante
and Gianeselli (1982), and described in the Canadian 5 FIELD TESTING PROGRAM
Foundation Engineering Manual (CFEM 2006)
As part of the current research program, load-
The scaling coefficients used in the LCPC method are displacement data was compiled for 29 static axial load
determined on the basis of soil type and magnitude of the tests performed on full-scale screw piles installed at nine
measured CPT tip resistance. The method was derived different sites in Western Canada. The load test sites
from the interpretation of 197 static load tests conducted were located in or near Bruderheim, Edmonton,
on piles, mostly of the bored or driven type, installed in Ft. McMurray, Ft. Saskatchewan, and Hythe, Alberta, and
France under a variety of soil conditions. Ft. St. John, British Columbia. All screw pile load tests
were conducted in accordance with the respective ASTM
The LCPC method may be used to calculate components standards for individual piles loaded in compression and
of unit skin friction and bearing to be applied to the screw tension (ASTM Designation: D1143 1981; ASTM
pile failure surface determined by the cylindrical shear Designation: D3689 1990). The incremental application
model or the individual plate bearing model. However, it of load to the test piles was as per the ‘Quick Test’
should be noted that the majority of the conventional pile procedure described in the above standards.
load tests on which the LCPC method was based were
conducted in compression, in soil types described as 5.1 Investigation of Load Test Sites
sand, clay, silt, and chalk.
Geotechnical field investigations of varied scope were
4.3 Empirical Methods conducted at the screw pile load test sites. Cone
penetration tests (CPT’s) were performed at seven of the
Empirical relationships correlating the required nine test sites, supplemented by test borings and
installation torque to the ultimate axial screw pile capacity standard penetration tests (SPT’s) at six of the seven
have been proposed by several authors (Ghaly and sites. The investigations at the remaining two load test
Hanna 1991; Hoyt and Clemence 1989; Narasimha Rao sites were comprised of test borings supplemented by
et al. 1989; Perko 2000). The concept of an empirical SPT’s.
torque to capacity correlation is analogous to the
relationship of pile driving effort to pile capacity (Hoyt and The surficial geologies of the test sites, although varied,
Clemence 1989). predominantly consisted of glacially-derived soils such as
lacustrine clay, clay till, sand, and sand till. Additionally,
The simplest form of the empirical torque relationship near-surface weathered clay shale bedrock was
directly relates the capacity, Q, of a screw pile to the encountered at two of the test sites.
installation torque, T, by means of an empirical scaling
Table 3 summarizes the locations of the nine load test
factor, Kt, as shown in Equation [14].
sites, the field investigation methods and predominant
soil conditions at each site, and details of the
Q = Kt ⋅ T [14] corresponding screw pile load tests conducted in tension
or compression.

1612
 OttawaGeo2007/OttawaGéo2007

Table 3: Load Test Sites, Screw Pile Geometries and Ultimate Axial Capacities
Test Vicinity of Test Site Predominant Soil Soil Investigation Load Test Pile No. of Helix Diameter(s) Shaft Inter-Helix Installatio Ultimate
Site No. Conditions Technique(s) Designation Length Helices Diameter Spacing Torque Axial
D1 D2 D3 d Ratio Capacity
(Tension, T, or
Compression, C)
(m) (cm) (cm) (cm) (cm) S/D (kN·m) (kN)
C1 5.0 3 35.6 35.6 35.6 21.9 1.5 20.3 180
C2 3.0 3 35.6 35.6 35.6 21.9 1.5 15.6 160
C3 5.0 2 35.6 35.6 -- 21.9 3.0 19.5 210
1 Edmonton, AB Stiff silty clay Test Holes, SPT, CPT
T1 5.0 3 35.6 35.6 35.6 21.9 1.5 22.1 210
T2 3.0 3 35.6 35.6 35.6 21.9 1.5 20.3 140
T3 5.0 2 35.6 35.6 -- 21.9 3.0 22.9 210
C4 5.0 3 35.6 35.6 35.6 21.9 1.5 44.7 470
C5 3.0 3 35.6 35.6 35.6 21.9 1.5 40.7 420
Loose to compact silty C6 5.0 2 35.6 35.6 -- 21.9 3.0 44.7 380
2 Bruderheim, AB Test Holes, SPT, CPT
sand T4 5.0 3 35.6 35.6 35.6 21.9 1.5 50.8 360
T5 3.0 3 35.6 35.6 35.6 21.9 1.5 42.7 190
T6 5.0 2 35.6 35.6 -- 21.9 3.0 47.9 360
C7 4.6 1 45.7 -- -- 17.8 -- 25.6 212
3 Ft. Saskatchewan, AB Stiff silty clay Test Holes, CPT C8 4.6 1 45.7 -- -- 21.9 -- 34.8 268
C9 5.5 2 50.8 45.7 -- 17.8 3.1 31.5 372
Hard clay till over
4 Lamont, AB Test Holes, CPT C10 9.3 2 50.8 45.7 -- 24.4 3.2 118.6 1177
weathered clay shale
C11 5.9 1 76.2 -- -- 27.3 -- 85.4 1094
C12 6.0 2 76.2 76.2 -- 27.3 3.0 97.6 1375
5 Ft. McMurray, AB Hard clay till Test Holes, SPT, CPT
T7 5.9 1 76.2 -- -- 27.3 -- 81.3 800
T8 6.0 2 76.2 76.2 -- 27.3 3.0 122.0 1325

6 Ft. McMurray, AB Very dense sand till Test Holes, SPT, CPT T9 4.9 1 76.2 -- -- 40.6 -- 257.6 2025

Firm to stiff clay till over

7 Hythe, AB Test Holes, SPT, CPT C13 7.5 1 40.0 -- -- 21.9 -- 120.9 1075
weathered clay shale
C14 10.4 2 91.4 91.4 -- 32.4 1.8 79.0 634
8 Ft. St. John, BC Firm to stiff silty clay Test Holes, SPT
C15 6.1 3 50.8 50.8 50.8 14.0 3.0 19.7 270
C16 5.0 2 45.7 45.7 -- 11.4 3.3 13.5 245
9 Ft. St. John, BC Stiff silty clay CPT
C17 4.0 1 45.7 -- -- 11.4 -- 8.0 169

5.2 Axial Load Test Results 6.1 Theoretical Predictions

The ultimate axial capacity of each of the test piles was Consistently accurate capacity predictions were
interpreted from the load-deflection curve using the accomplished using the theoretical methodology, as
Brinch-Hansen 80% Failure Criterion (Fellenius 1990). shown in Figure 4. The ultimate screw pile capacities
The interpreted ultimate axial capacities of the screw predicted by the theoretical method were plotted as a
piles are summarized in Table 3. Additionally, the ratio of the measured capacities, and exhibited a
specific geometry of each screw pile and the required maximum over-prediction of 11 percent. Of the 26
installation torques at the finished pile depths are also capacity predictions made, 21 fell within 20 percent of the
shown in Table 3. measured capacities, and the remaining five capacity
predictions outside this boundary erred on the side of
conservatism, with a maximum underestimation of 42
6 PILE CAPACITY PREDICTIONS percent.

Class 3 ultimate axial capacity predictions were made for 6.2 LCPC Predictions
each of the 26 test piles using the indirect theoretical
methodology described above, the LCPC direct pile Using the LCPC method, capacity predictions for the test
design method, and the empirical torque correlation. The piles showed significant variation from the measured
failure surfaces were approximated by the cylindrical capacities, as shown in Figure 4. While good predictions
shear model for multi-helix screw piles having an inter- within 20 percent of the measured capacities were
helix spacing ratio (S/D) less than 3.0, and by the obtained for the screw piles in firm to stiff clay deposits,
individual plate bearing model for single-helix screw piles significant over predictions of 2.25 to 4.75 times the
and multi-helix piles with S/D ≥ 3.0. measured capacity occurred in the hard and very dense
glacial till materials.
Soil strength parameters used in the theoretical
methodology were estimated on the basis of published The use of the LCPC method for screw piles installed in
correlations to SPT and CPT data (Lunne et al. 1997, glacial till soils is hindered by the limited empirical scaling
Peck et al. 1974; Stroud and Butler 1975). Capacity factors defined within the method. The scaling factors
predictions by the LCPC method were accomplished for are grouped into categories based on the soil type,
23 of the 26 test piles, as CPT’s were not included in the whether clay, silt, sand, or chalk, and the soil stiffness
geotechnical investigations at test site numbers 7 and 8. indicated by the cone penetration tip resistance at the
-1 appropriate depth (CFEM 2006). For clay soils, as
A Kt factor of 9.2 m was derived for use in the empirical
torque correlation by linear regression of the measured concerns piles T7, T8, C11, and C12, installed in clay till,
screw pile capacities versus the required installation the available scaling factors are for ‘compact to stiff clay
2
torques, with an R value equal to 0.96. and compact silt’ (Bustamante and Gianeselli 1982)
having tip resistance readings greater than 5000 kPa.
The clay till soils into which piles T7, T8, C11, and C12

1613
 OttawaGeo2007/OttawaGéo2007

were installed exhibited tip resistance readings in the capacities, as shown in Figure 4.
order of 10,000 to 20,000 kPa at the depth of the piles.
The ultimate capacities of test piles C16 and C17 were
Similarly, test pile T9 was installed into very dense sand underestimated by 50 and 60 percent, respectively, using
till, and the scaling factors available for use within the the torque method. These discrepancies could be due to
LCPC method were for ‘compact to very compact sand the fact that the two piles were fabricated with the
and gravel’ (Bustamante and Gianeselli 1982), exhibiting smallest shaft diameter of any of the test piles, 11.4 cm.
CPT tip resistance values of 12,000 kPa or greater. Tip There may be justification for the use of a larger Kt factor
resistance values of 15,000 to 60,000 kPa were recorded for screw piles of relatively small shaft diameter, as
in the sand till at the T9 test site. suggested by Hoyt and Clemence (1985). Based on the
two load test results for piles C16 and C17, a Kt factor of
On the basis of the capacity predictions shown in Figure -1
4 for piles C11, C12, T7, T8, and T9, it is concluded that 18 m would provide reasonable predictions of the
the LCPC method is not directly applicable to glacial till ultimate pile capacities for screw piles with shaft
soils, as the scaling factors defined within the method can diameters of 11.4 cm. However, use of a Kt factor in
-1
only be accurately applied to those soils for which they excess of 10 m , as is recommended by the Canadian
were explicitly intended, i.e. clay, silt, sand, or chalk. Foundation Engineering Manual (CFEM 2006), should be
cautioned, on the basis of the limited number of small-
Additionally, the uplift capacity of pile T5 was shaft screw piles examined under the current
overestimated by approximately 150 percent using the investigation.
LCPC method. This test pile was installed in loose to
compact silty sand, at a depth corresponding the shallow One significant over prediction of 110 percent occurred
failure condition. Based on the preceding theoretical using the torque method for test pile T5, installed in sand
discussions, the capacity of a helix installed in sand will and loaded in uplift under the shallow failure condition. It
be significantly lower in uplift than in compression, should be noted that he use of a single Kt factor within
especially in the case of a shallow failure condition. As
the empirical torque relationship does not capture the
the LCPC method was developed primarily on the basis
difference between the screw pile capacity in uplift as
of compression tests, it may not be directly applied to
opposed to compression. This discrepancy is especially
piles loaded in tension in cohesionless soils.
significant for screw piles loaded in uplift under the
shallow failure condition, as the bearing capacity above a
However for ‘deep’ screw piles in cohesive soils, the
shallow buried plate in uplift will be significantly less than
bearing capacity and skin friction in uplift will be
above a deep plate in uplift or below a buried plate in
essentially equivalent to that in compression, and
compression.
therefore the LCPC method may be applied with some
degree of confidence.

6.3 Torque Predictions

Of the 26 ultimate capacity predictions made using the

-1
empirical torque relationship with a Kt factor of 9.2 m , 23
fell within 30 percent of the measured screw pile

2.5
Qpredicted / Qmeasured

2.0

1.5

1.0

0.5

0.0
C1 C2 C3 T1 T2 T3 C4 C5 C6 T4 T5 T6 C7 C8 C9 C10 C11 C12 T7 T8 T9 C13 C14 C15 C16 C17
Test Pile Designation

Theoretical Method LCPC Method Torque Method

Figure 4: Ratio of predicted to measured ultimate axial screw pile capacities

1614
 OttawaGeo2007/OttawaGéo2007

7 CONCLUSION Das, B.M. 1990. Earth anchors, Elsevier Science

Publishers, New York, NY, USA.
The axial capacities of 26 full-scale screw piles installed Fellenius, B.H. 1990. Guidelines for the interpretation and
in Alberta and British Columbia since 1998 were analysis of the static loading test, 1st ed., Deep
interpreted on the basis of static load test results. The Foundations Institute, Sparta, NJ, USA.
effectiveness of three design methods were evaluated for Ghaly, A. and Hanna, A. 1991. Experimental and
predicting the ultimate axial capacity of the test piles, theoretical studies on installation torque of screw
installed in cohesive and cohensionless soils. The failure anchors. Canadian Geotechnical Journal, 28(3):353-
surfaces were approximated by the cylindrical shear 364.
model for multi-helix screw piles having an inter-helix Hoyt, R.M., and Clemence, S.P. 1989. Uplift capacity of
spacing ratio (S/D) less than 3.0, and by the individual helical anchors in soil. 12th International Conference
plate bearing model for single-helix screw piles and multi- on Soil Mechanics and Foundation Engineering. Rio
helix piles with S/D ≥ 3.0. de Janeiro, Brazil, 2: 1019-1022.
Kulhawy, F.H. 1984. Limiting tip and side resistance, fact
Good predictions of the ultimate screw pile capacities in of fallacy. Symposium on Analysis and Design of Pile
uplift and compression were obtained using theoretical Foundations, ASCE, San Francisco, pp. 80-98.
formulations for appropriate components of bearing Lunne, T., Robertson, P.K., and Powell, J.J.M. 1997
capacity and friction. Most of the capacity predictions Cone penetration testing in geotechnical practice, 1st
made by the theoretical method fell within 20 percent of ed., Routledge, New York, NY, USA.
the measured capacities, with a small number of Mitsch, M.P. and Clemence, S.P. 1985. The uplift
predictions outside the 20 percent range erring on the capacity of helix anchors in sand. Uplift capacity of
side of conservatism. Capacity predictions based on the anchor foundations in soil, ASCE, pp. 26-47.
LCPC direct pile design method, using the results of Meyerhof, G.G. 1976. Bearing capacity and settlement of
static cone penetration tests, were good for screw piles pile foundations. Journal of Geotechnical Engineering,
under uplift and compression in clay, but significant ASCE, 102(GT3): 197-224.
deviations from the measured capacities occurred for Meyerhof, G.G., and Adams, J.I. 1968. The ultimate uplift
piles installed in sand and glacial till materials. A direct capacity of foundations. Canadian Geotechnical
empirical relationship between the installation torque and Journal, 5(4): 224-244.
the ultimate axial screw pile capacity was applied using a Narasimha Rao, S., Prasad, Y.V.S.N., and Veeresh, C.
-1
Kt factor of 9.2 m derived by linear regression of the 1993. Behaviour of embedded model screw anchors
data set. Capacity predictions made using the torque in soft clays. Geotechnique, 43:605-614.
method typically fell within 30 percent of the measured Narasimha Rao, S., Prasad, Y.V.S.N., Shetty, M.D., and
capacities for deep screw piles in uplift and compression. Joshi, V.V. 1989. Uplift capacity of screw pile anchors.
The torque relationship significantly over-predicted the Geotechnical Engineering, 20(2): 139-159.
uplift capacity of a shallow screw pile installed in sand. Peck, C.B., Hanson, W.E., and Thornburn, T.H. 1974.
Foundation engineering, 2nd ed., John Wiley, New
York, New York, USA.
ACKNOWLEDGEMENTS Perko, H.A. 2000. Energy method for predicting
installation torque of helical foundations and anchors.
The authors would like to acknowledge the contribution of New Technological and Design Developments in
Gerry Cyre, R.E.T. This research was made possible in Deep Foundations, ASCE, Special Publication No.
part by scholarships from the Natural Sciences and 100, Denver, Colorado, p. 490.
Research Council of Canada (NSERC) and the Alberta Stroud, M.A. and Butler, F.G. 1975. The standard
Ingenuity Fund. penetration test and the engineering properties of
glacial materials. Proceedings of the Symposium on
the Engineering Behavior of Glacial Materials,
REFERENCES University of Birmingham, England.
Winterkorn, H. F., and Fang, H. Y, Ed. 1975. Foundation
ASTM Designation: D3689 1990. Standard test method engineering handbook, Van Nostrand Reinhold
for piles under static axial tensile load. American Company, New York, 751 pp.
Society for Testing and Materials (ASTM). Zhang, D. 1999. Predicting capacity of helical screw piles
ASTM Designation: D1143 1981. Standard test method in Alberta soils. M.Sc. thesis, Department of Civil and
for piles under static axial compressive load. Environmental Engineering, University of Alberta,
American Society for Testing and Materials (ASTM). Edmonton, Alberta.
Bustamante, M., and Gianeselli, L. 1982. Pile bearing
capacity prediction by means of static penetrometer
CPT. Proceedings of the 2nd European Symposium
on Penetration Testing, ESOPT-II, Amsterdam, Vol. 2,
pp. 493-500.
CFEM. 2006. Canadian foundation engineering manual,
4th ed., Canadian Geotechnical Society, Richmond,
British Columbia, Canada.

1615

View publication stats