2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

Direct CPT method for footing response in sands using a

database approach
P.W. Mayne and F. Illingworth
Georgia Institute of Technology, Atlanta, GA USA

ABSTRACT: A direct CPT method for evaluating footing response on clean sands is
developed based on a special database of 30 shallow foundation load tests situated on
12 different sands. Only large size footings have been considered, having widths from
0.5 m to 6 m. A characteristic stress vs. square root of normalized displacement curve
(s/B) is shown to be applicable for these cases. When the applied stress is normalized
by the cone tip resistance, all footings follow a unified trend in their behavior.

1 INTRODUCTION

The evaluation of shallow foundations in sands using cone penetrometer testing

(CPT) can be made using either indirect (rational) and direct methods. In the rational
approach, a two-step procedure is followed with the first task covering the interpreta-
tion of soil engineering parameters (e.g. t, ', K0, OCR, E', ') from the measured
CPTs and the second task involving the use of theory to assess bearing capacity and
displacements. Commonly, limit plasticity solutions are used to calculate bearing ca-
pacity and elastic continuum theory applied for displacement evaluations. Alterna-
tively, direct methods provide a means for scaling the CPT results immediately to ob-
tain the capacity or settlement estimates in a direct one-step procedure.
The drained bearing capacity of shallow foundations on sands via limit plasticity
theory is usually expressed:

qult = ½ B· * ·N (1)

where B = foundation width, * = operational unit weight (i.e., total or effective unit
weight, depending on water table elevation), and N = bearing factor. One difficulty
that arises is the well-known dependency of Non foundation size. That is, the bear-
ing factor Nhas been shown to decrease with B, as now verified by small- and large-
scale experiments (Kusakabe et al., 1992), centrifuge modeling (Kimura et al., 1985),
and numerical computer simulations (Mase & Hashiguchi, 2009).
The magnitude of foundation displacements or settlements (s) for shallow spread
footings can be calculated using elastic theory, commonly given by:
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

s = q·B ·I·(1-2)/E' (2)

where q = applied footing stress, I = elastic displacement influence factor (e.g., Pou-
los & Davis, 1974), E' = drained elastic soil modulus, and  = soil Poisson's ratio. A
difficulty lies in the highly nonlinear behavior of E' that starts at Emax in the nonde-
structive region and progressively decreases by over one order of magnitude as the
applied stress approaches the strength (Tatsuoka & Shibuya, 1991; Mayne 2005).
In terms of direct CPT methods for shallow foundations, solutions have been pro-
posed based on analytical studies (e.g., Schmertmann, 1978; Eslaamizaad & Robert-
son, 1996; Eslami & Gholami, 2005), experimental chamber tests (e.g., Berardi &
Bovolenta, 2003), and finite element simulations (Lee & Salgado, 2005). Herein, a
new direct CPT is developed from the measured load-displacement behavior of 30
shallow footings on sand.

2 CHARACTERISTIC STRESS-NORMALISED DISPLACEMENT CURVES

The load-displacement behavior of footings has been shown to be unique for a given
sand, when the results are presented in terms of a characteristic stress vs. normalized
displacement curve (Fellenius, 1994; Briaud & Gibbens, 1999; Lutenegger & Adams,
2003; Briaud, 2007). This can be illustrated in Figure 1a with the results of applied
load (Q) vs. measured displacement (s) behavior from 3 footings of varied sizes that
were load tested on loose sand at Fittja, Sweden (data from Bergdahl, et al. 1985).
The individual responses of the footings becomes singular when plotted in normaliza-
tion terms of stress (q = Q/B2) vs. pseudo-strain, s/B, where B = equivalent footing
width, as presented in Figure 1b.
As the observed curve can be represented as a power function (Decourt 1999), a
further post-processing can be adopted using the square root of s/B, as evidenced in
Figure 2. This results in a single parameter (rs) to represent the sand formation, where
rs = slope of applied stress vs. sqrt(s/B):

q  rs  (s / B) (3)
Fittja, Sweden Fittja, Sweden
2.5 1.0
0.9 2.3 x 2.5 m
Applied Stress, q (MPa)

2.0 2.5 m x 2.3m 0.8 1.6 x 1.8 m

Applied Load, Q (MN)

0.7 0.55 x 0.65 m

1.5 0.6
1.8 m
0.5
x
1.0 1.6m 0.4
0.3
0.5 0.2
0.65m
0.55m 0.1
0.0 0.0
0 20 40 60 80 100 0.00 0.05 0.10 0.15
Displacement, s (mm) Normalized Displacement, s/B
Figure 1. Measured footing response for 3 large footings at Fittja: (a) load-displacement curves;
(b) characteristic stress vs. normalized displacement behavior (data from Bergdahl, et al. 1985).
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

Fittja, Sweden
1.5
0.55 x 0.65 m Bearing Capacity:
Applied Stress, q (MPa) Stress at s/B = 10%
1.6 x 1.8 m
[sqrt(s/B) = 0.316]
1.0 2.3 x 2.5 m

qcap = 0.64 MPa

0.5 Regression:
n = 21
q = 2.03 sqrt(s/B)
2
r = 0.980
0.0
0.0 0.1 0.2 0.3 0.4 0.5
Square Root (s/B)
Figure 2. Stress vs. square root of (s/B) response for footings on Fittja sand.

For the Fittja sand results presented in Figure 2, the site-specific value of rs = 2.03 is
observed which can be used to regenerate the load-displacement response for all three
footings using equation (3).
As no clear peak is observed during the load testing of foundations on sand, the de-
finition of bearing capacity of sands must be specified. In this case, one well-known
criterion is to define the capacity as the stress required for the settlement to reach a
value equal to 10% of the footing width; i.e., qcap occurs at (s/B) = 0.1. In terms of
the square root plot, Fig. 2 shows that equation (3) can be used to calculate the bear-
ing capacity for the specified sand under study. For Fittja, the bearing capacity can be
interpreted as qcap = 640 kPa for the loose glaciofluvial sands using the s/B = 10% cri-
terion. Otherwise, Lee & Salgado (2005) suggest a bearing capacity defined at s/B =
20%, and thus equation (3) could be used here to obtain comparable results.

3 FOOTING DATABASE

A database of 30 footings on 12 different sands that were subjected to vertical loading

in the field was compiled for this study. The origins of these sands included marine
sediments, dunes, and compacted sand fills. The footings included 18 square, 7 rec-
tangular, and 5 circular shallow foundations, with the equivalent footing width B va-
rying from 0.5 to 6 m. Only large foundations were considered in order to avert
aforementioned difficulties associated with 1-g model tests. Table 1 provides a brief
summary listings of the sand sites, locations, type of sand, and reference sources. Ta-
ble 2 provides the footing geometry and sizes (in terms of an equivalent square foot-
ing with side B), embedment depth (ze), groundwater depth (zw), and footing perfor-
mance results (rs and qcap) from the measured load-displacement data.
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

For each site, a representative average cone tip resistance (qc(av)) has been assigned
to the sand. In actuality, a qc value for each footing was appointed, corresponding to
the mean over 1.5 B deep below the bearing elevation. From theoretical viewpoints,
limit plasticity specifies a value over 1B deep, while elastic theory considers the
range over 2B deep. However, for most all of the cases reported herein, the differ-
ences between arithmetic and geometric means over these depth intervals were small
and thus only a representative qc is given for each site.

Table 1. List of Sand Sites, Footing Geometry, and Reference Information

SITE LOCATION TYPE SAND FOOTINGS REFERENCE SOURCE

Alvin East Texas Alluvial sand 2 Circular: D = 2.2 m Tand, et al. (1994)
Alvin West Texas Alluvial sand 2 Circular: D = 2.35 m Tand, et al. (1994)
Durbin South Africa White fine sand 1 Square: B = 6.09 m Kantley (1965)
Fittja Sweden Glaciofluvial sand 3 Rect: B=0.6m, 1.7, 2.4 m Bergdahl, et al. (1985)
Grabo T1C Sweden Compacted fill (38)* 1 Square: B = 0.46 m Phung Duc Long (1993)
Grabo T2C Sweden Compacted fill (67)* 1 Square: B = 0.63 m Phung Duc Long (1993)
Grabo T3C Sweden Compacted fill (62)* 1 Square: B = 0.80 m Phung Duc Long (1993)
Green Cove Florida Brown silty sand 1 Circular: D = 1.82 m Anderson et al. (2006)
Kolbyttemon Sweden Glaciofluvial sand 4 Rect: B = 0.6, 1.2, 1.7, 2.4 Bergdahl, et al. (1984)
Labenne France Dune sand 5 Square: B = 0.7 and 1.0 m Amar et al. (1998)
Perth Australia Siliceous dune sand 4 Square: B = 0.5 and 1.0 m Lehane (2008)
Texas A&M Texas Pleistocene sand 5 Square: 1, 1.5, 2.5, 3 m Briaud & Gibbens (1999)
*Note: sand relative density, DR (%), shown in parentheses.

Table 2. Additional information on sands, footing performance, and CPT resistances

SAND Embedment GWT Grain Size Slope rs qcap (MPa) CPT Ratio hs =
SITE ze (m) zw (m) D50 (mm) (MPa) at s/B=0.1 qc (MPa) rs/qc

Alvin East 2.2 1.9 0.11 - 0.14 4.06 1.28 6.72 0.60
Alvin West 2.35 1.9 0.11 - 0.14 5.46 1.73 10.46 0.52
Durbin 0 1.2 na 1.96 0.62 3.66 0.54
Fittja 0.4 to 1.1 1.5 0.2 - 0.4 2.03 0.64 3.2 0.63
Grabo T1C 0 >5 0.34 0.46 0.15 0.88 0.53
Grabo T2C 0 >5 0.34 1.85 0.58 3.86 0.48
Grabo T3C 0 >5 0.34 1.66 0.52 2.87 0.58
Green Cove 0.6 3 na 5.41 1.71 9.78 0.55
Kolbyttemon 0.4 to 1.1 8 0.3 - 0.4 5.51 1.74 10.72 0.51
Labenne 0 3 0.32 2.65 0.84 4.01 0.66
Perth 0.5 to 1.0 5.5 0.42 1.51 0.48 3.44 0.44
Texas A&M 0.76 4.9 0.2 4.86 1.54 7.5 0.65
Notes: ze = footing embedment depth, zw = groundwater depth, D50 = mean grain size, slope rs = q vs. (s/B)0.5,
qcap = bearing capacity stress defined at settlement (s) equal to 10% footing width (B), qc = cone tip resistance.
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

4 CPT NORMALIZATION FOR FOOTING RESPONSE

The characteristic qapplied vs. sqrt(s/B) curves for the full set of 12 sandy soil forma-
tions are presented in Figure 3. In each case, least squares regression analyses were
used to obtain a best-fit line (b = 0) and the corresponding slope of stress vs.
sqrt(s/B), designated rs, as listed in Table 2. The coefficient of determination in the
fitting of the regressions gave r2 > 0.94 and the average n = 24, where n = number of
data sets (Note: for the Florida site, however, n = 5). Single lines are used in Figure 3
to represent each of the derived slopes (rs) and the corresponding CPT qc is shown ad-
jacent to each lines for each sand.

2.0
qc = 10.72 MPa Kolbyttemon
Summary:
1.8 qc = 10.46 Alvin West
30 Footings
qc = 9.78 Green Cove
Footing Stress, q applied (MPa)

on 12 Sand Sites
1.6
qc = 7.52 Texas A&M

1.4
qc = 6.72 MPa Alvin
1.2
rs = qapplied/sqrt(s/B)
1.0
qc = 4.01 MPa Labenne
0.8
qc = 3.21 Fittja
qc = 3.66 Durbin
0.6
qc = 3.86 Grabo T2C
qc = 2.87 Grabo T3C
0.4
qc = 3.44 Perth
0.2
qc = 0.88 MPa Grabo T1C
0.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6

Sqrt Normalized Displacement, (s/B)0.5

Figure 3. Summary of stress-square root (s/B) responses for all footing responses at 12 sand sites.

It is evident that the site-specific slopes (rs) for each sand are proportional with their
corresponding representative qc values. Briaud (2007) has suggested the normaliza-
tion of the stress axis with in-situ test measurements; in his preferential arguments,
using the limit pressure from pressuremeter tests. Herein, the same concept is utilized
with the footing stresses normalized by their respective qc values of the sands, as pre-
sented in Figure 4.
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

Alvin East, Texas

0.30
y = 0.597x Alvin West, Texas
Regression (Best-Fit Line): n = 337
Normalized Stress, q applied /qc
0.25 r2 = 0.933 Durbin, South Africa

Fittja, Sweden

Grabo T1C
0.20 q cap = 0.18 q c
Grabo T2C
Clean to slightly silty
Grabo T3C
0.15 fine-medium quartz
to siliceous Sands Green Cove, Florida

Kolbyttemon, Sweden
0.10
Labenne, France
q (stress)  3q  (s / B)
0.05 5 c Perth, Australia

Texas A&M
s/B = 10% criterion All 30 Footings
0.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6 Linear (All 30 Footings)

Sqrt Normalized Displacement, (s/B)0.5

Figure 4. Characteristic relationship for shallow foundation response on clean sands in terms of CPT-
normalized stress (qapplied/qc) vs. square root (s/B)

The normalization provides a standalone approach to evaluating the load-

displacement behavior of spread footings on sands, given by:

q applied 3 s
 5
 (4)
qc B

This statistical expression, derived from full scale experimental footing test results,
compares favorably with the finite element results given by Lee & Salgado (2005).
For the adopted s/B = 10% criterion, the above equation (4) gives a simple
straightforward evaluation of bearing capacity for shallow foundations on sand as:

qcap = 0.18 qc (5)

which agrees similarly with the value of 0.16 recommended by Eslaamizaad and Ro-
bertson (1996).
A validation on the approach can be further checked by comparing the measured
footing applied stresses vs. the calculated stresses given by equation (4). Figure 5
presents the statistical fitting and shows that a reliable evaluation is obtained, having
an associated coef. of determination r2 = 0.976 and standard estimate of the indepen-
dent value of SEY = 0.083 MPa
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

Footing Response on Sands

2.5

q(stress)  53 q c  (s / B)
Applied stress, q (MPa) 2.0
All 30 footings on
12 sands
1.5 n = 339

1.0
Regression Results:
0.5 q (Meas) = 1.004 q (Pred)
r2 = 0.976
S.E.Y. = 0.083
0.0
0.0 0.5 1.0 1.5 2.0 2.5
Predicted stress, q (MPa)
Figure 5. Comparison of applied vs. predicted stress using the direct CPT approach on sands.

5 CONCLUSIONS

The characteristic curves for applied foundation stress vs. pseudo-strain (s/B) for
footings on sands can be linearized using sqrt(s/B) plots. A database of 30 large shal-
low spread footings (0.5 ≤ B ≤ 6 m) resting on 12 different sands has been assembled
to develop a direct CPT method for evaluating footing response. Footing stresses are
normalized by the mean cone tip resistance of the sand (over a depth of 1.5B) to ob-
tain a dimensionless expression that is useful for site-specific evaluations of clean
quartz to silica type sands.

ACKNOWLEDGMENTS

The authors appreciate the support of the US Dept. of Energy (DoE), Savannah River
Site, ConeTec, and Fugro in their sponsorship of the In-Situ Research Group at GT.

REFERENCES

Amar, S., Baguelin, F., Canepa, Y. and Frank, R. (1998). New design rules for the bearing capacity of
shallow foundations based on Menard PMT. Geotechnical Site Characterization, Vol. 2 (ISC-1,
Atlanta), Balkema, Rotterdam: 727-733.
Anderson B.J., Townsend F.C., and Rahelison L. (2007). Load testing and settlement prediction of
shallow foundation, Journal of Geotechnical and Geoenvironmental Engrg, 133 (12): 1494-1502.
 2nd International Symposium on Cone Penetration Testing, Huntington Beach, CA, USA, May 2010

Berardi R., and Bovolenta R. (2003). Stiffness values and deformation behavior of soil for the settle-
ment analysis of foundations, Deformation Characteristics of Geomaterials, Vol. 1 (Proc. IS-
Lyon), Swets & Zeitlinger, Lisse: 849-855
Bergdahl, U., Hult G., and Ottosson E. (1985). Calculation of settlements of footings in sands, Proc.
11th Intl. Conf. on Soil Mechanics and Foundation Engineering, Vol. 4, San Francisco: 2167-2170.
Briaud J.L. (2007). Spread footings in sand: load settlement curve approach. Journal of Geotechnical
and Geoenvironmental Engineering 133 (8): 905-920.
Briaud, J.L. and Gibbens, R.M. (1999). Behavior of five large spread footings in sand. Journal of Geo-
technical & Geoenvironmental Engineering 125 (9): 787-797.
Decourt, L. (1999). Behavior of foundations under working load conditions. Proc. XI Panamerican
Conf. on Soil Mechanics and Geotechnical Engrg, Foz do Iguassu, Brazil, Vol. 4: 453-487
Eslami, A., and Gholami, M. (2005). Bearing capacity analysis of shallow foundations from CPT data.
Proc. 16th International Conference Soil Mechanics and Geotechnical Engineering, Vol. 3
(ICSMGE, Osaka), Millpress, Rotterdam: 1463-1466.
Eslaamizaad S. and Robertson P.K. (1996). Cone penetration test to evaluate bearing capacity of foun-
dations in sands. Proc. 49th Canadian Geotechnical Conference: Frontiers of Geotechnology, Vol.
1, St. John’s, Newfoundland: 429-438.
Kimura, T., Kusakabe, O. and Saitoh, K. (1985). Geotechnical model tests of bearing capacity prob-
lems in a centrifuge. Geotechnique 35 (1): 33-45.
Kusakabe, O., Maeda, Y. and Ohuchi, M. (1992). Large-scale loading tests of shallow footings in
pneumatic caisson. Journal of Geotechnical Engineering 118 (11): 1681-1695.
Lee J., and Salgado R. (2005). Estimation of bearing capacity of circular footings on sand based on
Cone Penetration Test, Journal of Geotechnical and Geoenvironmental Engrg 131 (4): 442-452.
Lehane B.M., Doherty J.P., and Schneider J.A. (2008). Settlement prediction for footings on sand. De-
formational Characteristics of Geomaterials, Vol. 1, IOS-Millpress, Rotterdam: 133-150.
Lutenegger A.J. and Adams M.T. (2003). Characteristic load-settlement behavior of shallow founda-
tions, Proceedings, International Symposium on Shallow Foundations (FONDSUP), Vol. 2, Labo-
ratoires des Ponts et Chaussées, Paris: 381-392.
Mayne, P.W. (2005). Integrated ground behavior: in-situ and lab tests. Deformation Characteristics of
Geomaterials, Vol. 2 (Proc. IS-Lyon), Taylor & Francis Group, London: 155-177.
Mase, T. and Hashiguchi, K. (2009). Numerical analysis of footing settlement problem by subloading
surface model. Soils & Foundations 49 (2): 207-220.
Phung Duc Long, P.D., 1993. Footings with settlement-reducing piles in non-cohesive soil. Report 43,
Swedish Geotechnical Institute, Linkoping: 45-93.
Poulos, H.G. and Davis, E.H. (1974). Elastic Solutions for Soil & Rock Mechanics, Wiley & Sons,
New York: 418 p. Centre for Geomechanics, Univ. Sydney, Australia. Downloadable PDF version
from: www.usucger.org
Robertson, P.K., and Cabal, K.L., 2007. Guide to cone penetration testing. Second Edition, Published
by Gregg In-Situ, Signal Hill, California, 126 p
Schmertmann, J. H., 1978. Guidelines for cone penetration test, performance and design, Report
FHWA-TS-78209, Federal Highway Administration, Washington DC: 146 p.
Tand, K., Funegard E., and Warden P.E., 1994. Footing load tests on sand. Vertical and Horizontal
Deformations of Foundations & Embankments, Vol. 1 (GSP 40), ASCE, Reston/VA: 164-178.
Tatsuoka, F. and Shibuya, S. (1992). Deformation characteristics of soils and rocks from field and la-
boratory tests. Report of the Institute of Industrial Science, Vol. 37, No. 1, Serial 235, University of
Tokyo: 137 pages.