Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

ANCHORED BULKHEADS WITH SLOPING

DREDGE LINES
By W. L. Schroeder, 1 M. ASCE and Philippe Roumillac 2

ABSTRACT: Tests were conducted to determine the effect of a sloping dredge

line on passive earth resistance in front of a bulkhead and bending moments
in sheet piling. Results are interpreted relative to moment reduction procedures
in current use.

INTRODUCTION

Embedment depth and section m o d u l u s for anchored bulkheads are

typically determined by analysis using a free earth s u p p o r t or fixed earth
support method (5). Considerable economy m a y b e achieved by using
moment reduction procedures in conjunction with the former (4). Both
basic procedures were developed u n d e r the p r e s u m p t i o n that the fore-
slope or dredge line in front of the bulkhead was horizontal. M a n y in-
stallations with sloping dredge lines h a v e b e e n designed a n d built, h o w -
ever. It is presumed, in the absence of an accepted design p r o c e d u r e ,
that considerable judgment was involved in the selection of a n embed-
ment depth and a sheet pile section m o d u l u s for these projects. The
studies described herein were undertaken to evaluate the effect of dredge
line slope on passive resistance in the foreslope a n d sheet pile b e n d i n g
moments. The results indicate that these effects can be considerable a n d
provide guidance for selection of a suitable sheet pile section. Further
studies are planned to extend the findings over a w i d e r range of con-
ditions, and to establish a well d o c u m e n t e d design procedure.

EXPERIMENTAL PROCEDURES AND RESULTS

Experiments were conducted in the installation s h o w n schematically

in Fig. 1. Physical arrangements were included which provided for fore-
slope adjustment and anchorage of the bulkhead at its top. The dimen-
sions of the container were selected so as to minimize the effects of side
wall friction on the results (1). A fine, uniform dry sand was chosen for
the tests and placed at a uniformly loose unit weight. The sand h a d
99.6% of its particles passing the N o . 50 U.S. Standard sieve a n d less
'Prof., Civ. Engrg., Oregon State Univ., Corvallis, Oreg. 97331.
2
Grad. Asst., Civ. Engrg., Oregon State Univ., Corvallis, Oreg. 97331.
Note.—Discussion open until November 1, 1983. To extend the closing date
one month, a written request must be filed with the ASCE Manager of Technical
and Professional Publications. The manuscript for this paper was submitted for
review and possible publication on August 23, 1982. This paper is part of the
Journal of Geotechnical Engineering, Vol. 109, No. 6, June, 1983. ©ASCE, ISSN
0733-9410/83/0006-0845/$01.00. Paper No. 18043.

845

J. Geotech. Engrg. 1983.109:845-851.

Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

FIG. 1.—Experimental Apparatus (1 cm = 0.394 In.)

than 0.1% passing the No. 200. Coefficients of uniformity, C„, and cur-
vature, C c , were 1.4 and 0.97, respectively. The minimum dry unit weight
was 13.7 kN/m 3 (87.5 pcf), and the maximum dry unit weight was 16.5
kN/m 3 (105.4 pcf). The average unit weight for the tests was 13.9 kN/
m3 (88.4 pcf). Direct shear tests at the test unit weight gave an internal
friction angle of 31°. The static angle of repose was 31.5°. Cornforth's
procedure (3) indicated an internal friction angle of 32° at the test unit
weight. This value was used in the analysis of test results. For conven-
ience, sheet pile embedment depth was chosen using Tschebotarioff's
hinge at dredge line procedure (7).
Passive Resistance Tests.—An initial set of experiments was run to
determine the effects of dredge line slope on passive resistance in the
foreslope. A bulkhead 12.7-mm (0.5-in.) thick was chosen to provide the
low flexibility necessary for a free earth support condition. Failure of the
foreslope was induced by applying a horizontal force to the sheet pile
at the dredge line level. No backfill was included above dredge line level
on the active side of the bulkhead. Horizontal deflections were mea-
sured at the dredge line and vertical deflections were measured at the
top of the bulkhead using dial gages. Forces in the anchor and the ap-
plied force at the dredge line were measured using load cells. Results
of these tests for several dredge line slopes are shown in Fig. 2, for both
the applied force and anchor force. Results from Fig. 2 were used to
back calculate the passive resistance coefficient, KP, from

KP = '— + KA (1)
-7D2cos8
2
in which 7 = unit weight of sand; D = height of wall imbedded in sand;
KA = active earth pressure coefficient; and 8 = wall friction angle equal
to 2cf>/3. A theoretical value of KA was calculated as 0.29 for test con-
ditions. Passive resistance coefficients determined are shown on Fig. 3
and compared with theoretical values from Caquot and Kerizel (2). The
passive resistance tests and subsequent calculations produced KF coef-
ficients exceeding the theoretical values, but varying uniformly with
846

J. Geotech. Engrg. 1983.109:845-851.

Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

nl l l I I I I I 1 I I o*-—' I I I I I I I I
0 2 4 6 8 10 0 2 4 6 8 10
Displacement of sheetpile Displacement of sheetpile
at the dredge line (cm) at the dredge line (cm)

FIG. 2.—Forces and Displacements During Passive Resistance Tests, Rigid Sheet
Pile, Free Earth Support (1N = 0.225 lb, 1 cm = 0.394 in.); (a) Force Applied (F*
x W) versus Displacement of Sheet Pile at Dredge Line; (b) Force in Anchor (A*
x W) versus Displacement of Sheet Pile at Dredge Line

dredge line slope. It is believed that side friction in the test setup was
partly responsible for the differences between measured and theoretical
values, despite attempts to minimize it by selection of apparatus
dimensions.
Effects of Slope on Bulkhead Bending Moments.—The principal ob-
jective of these studies was to measure slope effects on bulkhead bend-
ing moments. Rowe's experiments (4) were set up to insure model-field
similitude by scaling the test bulkheads to have flexibility similar to that
of field installations. The parameter p = H 4 /EI was chosen to represent
flexibility, in which H = total bulkhead height, in feet; £ = Young's
modulus for bulkhead, in pounds per square inch; and I = bulkhead
moment of inertia, in in.4 Rowe's moment reduction procedure employs
this parameter and his test results, and is widely used for bulkhead de-
sign. For the bulkhead height in the present experiments, a thickness of
1.59 mm (1/16 in.) provides a flexibility of 6.12 x 1CT3 ft4/Tb-in.2 per foot
of wall (English units shown to correspond to Rowe's published works).
This flexibility is the same as that of PS 28, PMA 22, and PDA 27 USS
steel sheet piles 8.2 m, 12.2 m, and 15.8 m (27 ft, 40 ft, and 52 ft) long.
The model bulkhead was instrumented with resistance strain gages along
its center line on both faces and at several elevations so that bending
effects could be measured.
Sand fill was placed with a horizontal surface up to the dredge line
level on both sides of the bulkhead. The remainder of the fill to the top
of the bulkhead was then placed on the active side, and bending strains
for that condition were noted. The tests were then completed by steep-
847

J. Geotech. Engrg. 1983.109:845-851.

Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

Dredge line slope angle, /3 (degrees)

FIG. 3.—Comparison of Experimental and Theoretical, KP, Coefficients

ening the dredge line angle in increments and measuring corresponding

bending strains. Bending moments, M, were calculated from
Ele
M =— (2)
c
in which previous definitions apply, and e = bending strain at sheet pile
face; and c = one-half sheet pile thickness. Results of these tests are
presented in Fig. 4.
Figure 4 shows that as the dredge line slope angle increases, the de-
gree of fixity provided by passive restraint in the foreslope is reduced,
from a fixed condition (horizontal dredge line) to a free condition when
the dredge line is at the angle of repose. Positive bending moments in-
creased nearly 100% as the support conditions changed from fixed to
free earth support. Therefore, unequivocally, it can be said that a sloping
dredge line has a significant effect on bending moments in a bulkhead
installation.

Bending moment ( N - m ) / m

FIG. 4.—Moment Distribution (1 N-m/m = 0.225 Ib-ft/ft; 1 cm = 0.394 in.)

848

J. Geotech. Engrg. 1983.109:845-851.

 ANALYSIS OF DESIGN APPROACHES

Depth of Embedment.—The test results described and presented in

Fig. 3 show that the passive resistance coefficients proposed by Caquot
and Kerizel (2) properly account for slope effects and are at least con-
servative. They are, therefore, suitable for use in determining required
bulkhead embedment, with an appropriate factor of safety chosen cor-
Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

responding to the design procedure to be utilized. If the generally ac-

cepted free earth support method is used to determine embedment depth,
a factor of safety of two is appropriate (6), or, alternately, the full value
of KP may be chosen and the embedment depth increased arbitrarily (7).
The former procedure is preferred for sloping dredge lines as no prec-
edent exists for a suitable arbitrary length increase.
Rowe's moment reduction procedure specifies the use of a KP value
equal to 0.67 times the Coulomb theoretical value for 8 = 0 to determine
minimum embedment depth. Had sufficient tests been run in the pres-
ent study to define moment reduction curves similar to Rowe's, and also
to account for slope effects, use of the same KP value might prove pref-
erable to avoid confusion on similar subjects in the literature.
Table 1 provides a comparison of passive resistance coefficients sug-
gested for possible use in design. Those suggested by Rowe are gen-
erally lower than the Caquot and Kerizel values; however, the latter val-
ues would usually be reduced to provide a factor of safety. The former
values would be used directly. The ratio KPCK/KPR generally lies between
1.74 and 3.35. Therefore, reduction of KPCK to provide the usual factor
of safety of two will result in a passive resistance coefficient for embed-
ment depth analysis close to that suggested by Rowe. Both methods,
therefore, will produce comparable embedment depths.
Bending Moment Ratio.—Figure 5 shows the test results summarized
in the form of ratios of actual bending moment to bending moment for
the horizontal dredge line case. It is known, of course, that actual bend-
ing moment in a bulkhead is dependent on flexibility of the sheet piling
and relative density of the soil in which it is embedded, in addition to
bulkhead height and anchor location. Therefore, the results in Fig. 5 are
specifically representative of the test conditions only. For corresponding

TABLE 1.—Comparison of Passive Resistance Coefficients

Angle of Slope Angle (3, in degrees
internal Wall friction
0 10 20
friction 4>, angle 8,
in degrees in degrees KpcK KFR KpcK KPR KpcK
(1) (2) (3) (4) (5) (6) (7) (8)
30 -15 2.00 4.85 1.47 3.28 1.03 1.79
30 -20 2.00 5.56 1.47 3.77 1.03 2.05
35 -17.5 2.46 6.87 1.76 4.38 1.23 2.56
35 -23.4 2.46 8.24 1.76 5.25 1.23 3.07

Note: Slope angle downward from bulkhead is positive. KPR = Rowe's passive
coefficient, 2/3 Coulomb Kp for 8 = 0. KPCK = Caquot and Kerizel's passive
coefficient.

849

J. Geotech. Engrg. 1983.109:845-851.

 Loose sand (a> = 32°)
/> = 6.I2XI0"° " -/ft
Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

"0 10 20 30
Dredge line slope angle, /3 (degrees)

FIG. 5.—Variation in Bending Moments with Dredge Line Slope Angle

Rowe's curve for

/ 3 * = 0 , q = 0, a=0.7

s Test data points

-3 -2.5
L
°9|0r°

FIG. 6.—Rowe's Moment Reduction Curve and Extrapolated Test Results for Sloping
Dredge Lines (Note: Reader Is Cautioned that Extrapolated Curves Are Tentative
and Should Refer to Text for Basis of Extrapolation)

field conditions, the effect of dredge line slope could be accounted for
by computing bending moments for the level dredge line case propor-
tioning the result using Fig. 5.
Moment Reduction.—Rowe's moment reduction curve (4) for the hor-
izontal dredge line, zero surcharge case is shown in Fig. 6. In practice,
for an embedment depth calculated using Rowe's recommendations, the
theoretical free earth support moment, Mc, is reduced by multiplying it
850

J. Geotech. Engrg. 1983.109:845-851.

 by the ratio M/Mc corresponding to actual bulkhead flexibility. Results
of the present tests are s h o w n to indicate comparability of results a n d
to tentatively extend the original work for the sloping dredge line case.
A discrepancy for the (3 = 0° case should be noted.
It is reasoned that for a sloping dredge line there should be n o m o -
ment reduction (M/Mc = 1) for a given foreslope angle until flexibility
becomes sufficiently large to allow i m p e n d i n g development of an in-
Downloaded from ascelibrary.org by Arnaldo Carrillo-Gil on 03/15/13. Copyright ASCE. For personal use only; all rights reserved.

flection point in the elastic curve of the bulkhead. M o m e n t reduction

curves for the case of a sloping dredge line would therefore originate
somewhere to the right of point 0. The tentative curves s h o w n are,
therefore, drawn through the test data points, parallel to the Rowe curve,
to intersect the abscissa of the plot to the right of 0. In practice they
would be used in the same m a n n e r as the Rowe curves, with the result
that, for a given flexibility, there w o u l d be significantly less m o m e n t
reduction than for the level dredge line case.

CONCLUSIONS

The embedment depth for an anchored bulkhead with a sloping dredge

line may be satisfactorily determined using the Caquot and Kerizel pas-
sive resistance coefficients or Rowe's coefficients, a n d the free earth sup-
port method of analysis. For site conditions and bulkhead flexibility cor-
responding to the test conditions described herein, the calculated moments
in the bulkhead may be reduced using Fig. 5 or Fig. 6. For less flexible
bulkheads, the free earth support m o m e n t should be used in design u n -
til further studies are conducted to define m o m e n t reduction effects for
a greater range of conditions.

APPENDIX.—REFERENCES

1. Bransby, P., and Smith, I., "Side Friction in Model Retaining-Wall Experi-
ments," Journal of the Geotechnical Engineering Division, ASCE, Vol. 101, No.
GT7, July, 1975, pp. 615-632.
2. Caquot, A., and Kerizel, J., "Tables for the Calculation of Passive Pressure,
Active Pressure and Bearing Capacity of Foundations," (rev. trans, by Min-
istry of Works, Chief Scientific Adviser's Division, London), Gauthier-Villars,
Paris, 1948.
3. Cornforth, D. H., "Prediction of Drained Strength of Sands from Relative
Density Measurements. Evaluation of Relative Density and Its Role in Geo-
technical Projects Involving Cohesionless Soils," ASTM STP 523, American
Society for Testing and Materials, 1973, pp. 281-303.
4. Rowe, P., "Anchored Sheet Pile Walls," Proceedings, Institution of Civil En-
gineers, London, Vol. 1, Pt. 1, 1952, pp. 27-70.
5. Terzaghi, K., Theoretical Soil Mechanics, John Wiley and Sons, Inc., New York,
N.Y., 1943.
6. Terzaghi, K., "Anchored Bulkheads," Transactions, ASCE, Vol. 119, 1954, pp.
243-1280 (with discussion, pp. 1281-1323).
7. Tschebotarioff, G., Foundations, Retaining and Earth Structures, McGraw-Hill Book,
Co., New York, N.Y., 1973.

851

J. Geotech. Engrg. 1983.109:845-851.