Chapter 7

Braced Cuts

Sometimes construction work requires ground excavations with vertical or near-

vertical facesfor example, basements of buildings in developed areas or
underground transportation facilities at shallow depths below the ground sur -
face (cut-and-cover type of construction). The vertical faces of the cuts need
to be protected by temporary bracing systems to avoid failure that may be
accompanied by considerable settlement or by bearing capacity failure of
nearby foundation(s).

Figure 7.1 shows two types of braced cut commonly used in construction
work. One type uses the soldier beam (Figure 7.1a), which is driven into the
ground before excavation. Soldier beams are vertical steel or timber beams.
After the excavation is started, laggings are placed between the soldier
beams as the excavation proceeds. Laggings are horizontal timber planks.
When the excavation reaches the desired depth, wales and struts are
properly installed (wales and struts are horizontal steel beams). The strut s
act like horizontal columns. Figure 7.1b shows another type of braced
excavation. In this case, interlocking sheet piles are driven into the soil
before excavation. As the excavation proceeds, wales and struts are
inserted immediately after reaching the appropriate depth.

To design braced excavations (that is, to select wales, struts, sheet piles,
and soldier beams), one must know the lateral earth pressure to which
the braced cuts will be subjected. This topic is discussed in Section 7.2;
subsequent sections cover the procedures of analysis and design of braced cuts.
7.2
Lateral Earth Pressure in Braced Cuts

In Chapter 5 we learned that a retaining wall rotates about its bottom (Figure 7.2a).
With sufficient yielding of the wall, the lateral earth pressure can be approximated to
be equal to that abtained by Rangkines coulombs theory
In contrast to retaining walls, braced cuts show a different type of wall yielding
(see Figure 7.2b). In this case, the deformation of the wall gradually increases
with the depth of excavation. The variation of the amount of deformation will
depend on several factors, such as the type of soil, the depth of excavation, and
the workmanship. However, one can easily visualize that, with very little wall
yielding at the top of the cut, the lateral earth pressure will be close to the at-
rest pressure. At the bottom of the wall, with a much larger degree of
yielding, the lateral earth pressure will be substantially lower than the Rankine
active earth pressure. As a result, the distribution of lateral earth pressure will
vary substantially in comparison to the linear distribution assumed in the case of
retaining walls.

A theoretical evaluation of the total lateral force, P, imposed on a wall can be

made by using Terzaghi's general wedge theory (1943a) (Figure 7.3a), in which
the failure surface is assumed to be the arc of a logarithmic spiral, defined by the
equation

where = angle of friction of soil

A detailed outline for the evaluation of P is beyond the scope of this text; readers
should check a soil mechanics text for more information (for example, Das,
1979). However, a comparison of the lateral earth pressure for braced cuts in sand
(with angle of wall friction S = 0) with that for a retaining wall (5 = 0) is shown in
Figure 7.3b. If 5 = 0, a retaining wall of height H will be subjected to a Rankine
active earth pressure, and the resultant active force will intersect the wall at a
distance of nH measured from the bottom of the wall. For this case, n = 1/3. In
contrast, the value of n for a braced cut may vary from 0.33 to 0.5 or 0.6. The
general wedge theory can also be used to analyze braced cuts in saturated clay
(for example, see Das and Seeley, 1975).

In any event, when choosing a lateral soil pressure distribution for design of
braced cuts, one should keep in mind that the nature of failure in braced cuts

is much different from that in retaining walls. After observation of sever

braced cuts, Peck (1969) suggested using design pressure envelopes for bra
cuts in sand and clay. Figure 7.4 shows Peck's pressure envelopes, to which t
following guidelines apply:
1. Figure 7.4a is for braced cuts constructed in dry or moist sand. Note is the
Rankine active earth pressure coefficient.

2. For cuts in clay, first calculate the value of (where c = undrained

cohesion of the clay located on the sides of the cuts; ) = 0 concept). If is
less than or equal to 4, the pressure envelope shown in Figure 7.4c should be
used. The value of p a, varies between 0.2 and 0.4 , with an average of
0.3yH. If is greater than 4, the pressure envelope shown in Figure 7.4b
should be used. In this case, may be equal to [1 (4c/ )] or 0.3 ,
whichever is greater. Peck's pressure envelopes are sometimes re ferred to as
apparent pressure envelopes.

Sometimes one encounters layers of both sand and clay when constructing a
braced cut. In this case, Peck (1943) proposed that an equivalent value of
cohesion ( = 0 concept) should be determined in the following manner (refer to
Figure 7.5a):

[ ( ) ]

.
where H = total height of the cut
= unit weight of sand
= height of the sand layer
= a lateral earth pressure coefficient for the sand layer
= angle of friction of sand
= unconfined compression strength of clay
n' = a coefficient of progressive failure (ranges from 0.5 to one; average
value 0.75)
The average unit weight, , of the layers can be expressed as

[ ( ) ]

where = saturated unit weight of clay layer

Once the average values of cohesion and unit weight are determined, the .
pressure envelopes in clay (Figure 7.4b and c) can be used to design the cuts.

In a similar manner, when a number of clay layers arc encountered in the cut
(Figure 7.5b), the average undrained cohesion can be expressed by the.
Equation
 [ ]

w h e r e c 1 , c 2 , . . . , C n , = undrained cohesion in Layers 1, 2, . . . , n

H1,H2, Hn = thicknesses of layers 1, 3, . . . , n
The average unit weight, yo, can be given as

[ ]

7.3.
Design of Various Components of a Braced Cut

Struts
In construction work, the struts should have a minimum vertical spacing of about
2.75 m or more. The struts are actually horizontal columns subject to bending.
The load-carrying capacity of columns will depend on the slenderness , ratio, 1/r.
The slenderness ratio can be reduced by providing vertical and horizontal
supports at intermediate points. For cuts with large widths, it may be necessary to
splice the struts. In the case of braced cuts in clayey soils, the depth of the first
strut below the ground surface should be less than the depth of tensile crack,
zc;..From Eq. (5.11)

A simplified conservative procedure can be used to determine the strut

loads. This procedure will vary depending on the engineers involved in the
project. Following is a step-by-step outline of it (refer to Figure 7.6).
1. Draw the pressure envelope for the braced cut (see Figure 7.4). Also show the
proposed strut levels. Figure 7.6a shows a pressure envelope for a

sandy soil; however, it could also be for a clay. Also, in this figure, the strut levels
are marked A, B1, C, and D. The sheet piles (or soldier beams) can be assumed to
be hinged at the strut levels, except for the top and bottom ones. In Figure 7.6a,
the hinges are at the level of struts B and C. (Many designers also assume the
sheet piles, or soldier beams, to be hinged at all strut levels, except for the top.)

2. Determine the reactions for the two simple cantilever beams (top and
bottom) and all the simple beams in between. In Figure 7.6b, these reactions
are A, B1, B2, C1, C2, and D.
3. The strut loads in Figure 7.6 can now be calculated as follows:

PA = (A)(s)
PB = (B1 + B2)(s)
PC = (C1 + C2)(s)
PD = (D)(s)

Where PA, PB, Pc, P D = loads to be taken by the individual struts at levels
A, B, C, and D, respectively
A, B 1 , B 2 , C 1 C 2 ,D = reactions calculated in Step 2 (note unit: force/unit
length of the braced cut)
s = horizontal spacing of the struts (see plan in Figure 7.6a)

4. Knowing the strut loads at each level and the intermediate bracing
conditions, one can now select the proper sections by using the steel con -
struction manual.

Sheet Piles
In order to design the sheet piles, perform the following steps:
1. For each of the sections shown in Figure 7.6b, determine the maximum bending
moment.

2. Determine the maximum value of the maximum bending moment (M max )

obtained in Step 1. Note that the unit of this moment will be, for example, kN-
m/meter length of the wall.
3. Obtain the section modulus of the sheet piles:

where = allowable flexural stress of the sheet pile material

4. The sheet pile section can now be chosen from a table such as Table 6.1.
Wales

1. Wales can be treated as continuous horizontal members if they are

spliced properly. Conservatively, they may also be treated as though they are
pinned at the struts. For the section shown in Figure 7.6a, the maximum
moments for the wales (assuming that they are pinned at the struts) are as
follows:

where A, B 1 , B 2 , C 1 C 2 , and D are the reactions under the struts per unit
length of the wall (Step 2 of strut design).
2. Determine the section modulus of the wales:

The wales are sometimes fastened to the sheet piles at points that satisfy the
lateral support requirements.

Example
7.1

The cross section of a long braced cut is shown in Figure 7.7a.

a. Draw the earth pressure envelope.
b. Determine the strut loads at levels A, B, and C.
c. Determine the section of the struts subjected to the largest load.
 d. Determine the sheet pile section required.
e. Determine a design section for the wales at level B.

Note: The struts are placed at 3 m center-to-center in the plan.

Solution
Part a

So, the pressure envelope will be like the one in Figure 7.4c. This is plotted in
Figure 7.7a with maximum pressure intensity, pa equal to 0.3yH = 0.3(18)(7) = 37.8
kN/m2.

Part b

For determination of the strut loads, refer to Figure 7.7b. Taking the moment
about MB1, = 0
Part c
The struts at level B are subjected to the largest loadthat is, PB = 271.2 kN. For the
struts, effective length (KL: refer to the American Institute of Steel Construction, Manual
of Steel Construction, 1980, pp. 3-29) with respect to x and y axes is 6 m. Accord-
ingly, the section W 250 mm x 49 kg/m (in English units, it is section W 10 x 33) will
be more than sufficient. (Note: Fv = 248.4 MN/m2.)

Part d
Refer to the left side of Figure 7.7b. For the maximum moment, the shear force should
be zero. The nature of variation of the shear force is shown in Figure 7.7c. The location
of point E can be given as

Because the loading on the left and right sections of Figure 7.7b are the same, the
magnitude of moments at F and C (Figure 7.7c) will be the same as E and A,
respectively. Hence, the maximum moment = 27.03 kN-m/meter of wall.

The section modulus of the sheet piles,

7.4
Stability of Braced Cut
Heave of the Bottom of the Cut in Clay

Braced cuts in clay may became unstable as a result of the heaving of the bottom of
the excavation. Terzaghi (19431)) has analyzed the factor of safety of braced
excavations against bottom heave. The failure surface for such a case is shown in
Figure 7.8. The vertical load per unit length of the cut at the level of the bottom of
the cut along the line bd and of is equal to

This load Q can be treated like a load per unit length on a continuous foundation at
the level of bd (and af) having a width of = 0.7B. Based on Terzaghi's bearing
capacity theory, the net ultimate load-carrying capacity per unit length., of this
foundation can be given by the equation [Chapter 3; see Eqs. (3.3) and (3.34)

Hence, from Eq. (7.8), the factor of safety against bottom heave can be given

The preceding factor of safety [Eq. (7.9)] has been derived based on th assumption
that the clay layer is homogeneous, at least up to a depth of O., below the
bottom of the cut. however, if a hard layer of rock or rock -lik material is
located at a depth D < 0.7B, the failure surface will be modified t some extent. In
such a case, the factor of safety can be modified to the form

( )
Bjerrum and Eide (1956) also studied the problem of bottom heave for braced cuts
in clay, and they proposed the following equation for the factor of safety

The bearing capacity factor N, varies with the ratio of H/B and also L/B (where L =
length of the cut). For infinitely long cuts (B/L = 0), N, = 5.14 at H/B = 0 and
increases to a value of /V, = 7.6 at H/B = 4. Beyond thatthat is, for H/B > 4the
value of N, remains constant. For cuts square in plan (B/L = 1),N, = 6.3 at H/B
= 0, and N, = 9 for H/B 4. In general, at any given H/B

Figure 7.9 shows the variation of the value of N,, for L/B = 1, 2, 3, and co. In any
case, a factor of safety of 1.25 to 1.5 is desired.
Stability of the Bottom of the Cut in Sand

The bottom of a cut in sand is generally stable. When the ground water table is
encountered, the bottom of the cut is stable as long as the water level inside the
excavation is higher than the ground water level. If the water level inside the cut is
lowered below the ground water level by pumping, instability may be created as a
result of the upward seepage of water into the cut. Section 7.5 discusses this
problem in more detail.

Lateral Yielding of Sheet Piles

. In braced cuts, some lateral movement of sheet pile walls may be expected (Figure
7.10). Of course, the lateral yield will depend on several factors, the
most important of which is time elapsed after excavation that is required for the
placement of wales and struts. :liana and Clough (1981) analyzed the field
records of several braced cuts in clay from the San Francisco, Oslo (Norway),
Boston, Chicago, and Bowline Point (New York) areas. Under ordinary con -
struction conditions, it was found that the maximum lateral wall yield [ ( )] has a
definite relationship with the factor of safety against heave. This is shown in Figure
7.10. Note that the factor of safety against heave as plotted in Figure 7.10 has
been calculated by using Eqs. (7.9) and (7.10).

In several instances, the sheet piles (or the soldier piles as the case may be) are
driven to a certain depth below the bottom of the excavation. This is done to reduce
the lateral yielding of the walls during the last stages of excavation. The lateral
yielding of the walls will cause settlement of the ground surface surrounding the cut.
The degree of lateral yielding, however, depends mostly on the soil type below the
bottom of the cut. If clay below the cut extends to a great depth and yH/c is less
than about 6, extension of the sheet piles or soldier piles below the bottom of the
cut will help considerably in reducing the lateral yield of the walls. However, under
similar circumstances, if yHIc is about 8, the extension of sheet piles into the clay
below the cut does not help to a great extent. In such circumstances, one may
expect a great degree of wall yielding that may result in the total collapse of the
bracing systems. If a hard soil I layer is located below a clay layer at the bottom of
the cut, the piles should be embedded in the stiffer layer. This will have a great
effect in reducing the lateral yield.

Ground Settlement
The Lateral yielding of walls will generally induce ground settlement ( )
around a braced cut. This is generally referred to as ground loss. Based
on several field observations, Peck (1969) has provided curves for prediction of
ground settlement in various types of soil (see Figure 7.11). The magnitude of
ground loss varies extensively; however, Figure 7.11 can be used as a general
guide.

Based on the field data obtained from various cuts in the areas of San Francisco,
Oslo, and Chicago, Mana and Clough (1981) have provided a correlation
between the maximum lateral yield of sheet piles [ ( ) ] and the maximum
ground settlement [ ( ) ]. This is shown in Figure 7.12. It can be seen that

( ) ( )
Figure 7.12 Variation of maximum lateral yield with maximum ground settlement
(after mana and Clough, 1981)

Solution
In Example Problem 7.1, y =18 kN/m3, c = 35 kN/m2, and H= 7 m
Factor of Safety from Eq (7.9)

Factor of Safety from Eq (7.11)

According to Figure 7.9, for H/B = 7/6 = 1.16 and B/L 0, the value of Nc is equal to
6.46. thus
Failure of Single wall Cofferdams by Piping

Sheet piles are sometimes driven for excavations that need dewatering (figure
7.13). in such cases, the factor of safety against piping should be

checked. [Piping is another term for failure by heave, as defined in Section 1.12;
see Eq. (1.48).] Piping may occur when high hydraulic gradient is set up as a
result of the flow of water into the excavation. One can check the factor of safety
by drawing flow nets and determining the maximum exit gradient ( )

that will occur at points A and B. Figure 7.14 shows a flow net to illustrate the
problem. The maximum exit gradient for this flow net can be calculated as

( )
where = length of the flow element at A (or B)
= number of drops (Note: In Figure 7.14, = 8.)

The factor of safety against piping can be expressed as

( )

where = critical hydraulic gradient

The relationship for i n has been given in Chapter 1 [Eq. (1.48)] as

The value of varies between 0.9 and 1.1 in most soils with an average of

about one. A factor of safety of about 1.5 is desirable

Marsland (1958) has suggested the following values of L2 (minimum Penetration) for
a factor of 1.5 against piping for excavations in sand

The maximum exit gradient for sheeted excavations in sands with L 2 = can
also be theoretically evaluated (Ilarr, 1962). (Only the results of these
mathematical derivations will be presented here. For further details, refer to
the original work.) To calculate the maximum exit gradient, refer to Figures
7.15 and 7.16 and perform the following steps:

1. Determine the modulus, m, from Figure 7.15 by obtaining 2L 2/B (or B/2L2)
and 2L1,/B.
2. With the known modulus and 2L 1 /B, refer to Figure 7.16 and determine

L2 ( ) Because L 2 and h will be known, ( ) can be

calculated.
3. The factor of safety against piping can be evaluated by using Eq. (7.15).

Another method to prevent piping and increase the factor of safety is to

lower the ground water level, thereby reducing the head h shown in Figure
7.13. This can be done by pumping from well points or deep wells placed below
the level of the bottom of the sheet piles.

Example 7.3

Refer to Figure 7.13. Given: h = 4.5 m, L 1 = 5 m, L 2 = 4 m, B = 5 m, and

L3 = Determine the factor of safety against piping.
Problems

7.1 Refer to the braced cut shown in Figure P7.1. Given: y = 112 lb/P, =
32, and c = 0. The struts arc located at 12 ft center-to-center in the
plan. Draw the earth pressure envelope and determine the strut loads at
levels A, B, and C.
7.2 For the braced cut described in Problem 7.1, determine:
a. the sheet pile section;
b. the section of the strut subjected to the largest load;
c. the section of the wales at level A.
7.3 Redo Problem 7.1 with y = 116 lb/ft 3 , = 35, c = 0, and center-to-center
strut spacing in plan = 10 ft.

7.4 Determine the sheet pile section required for the braced cut described in
Problem 7.3.
7.5 Refer to Figure 7.5a. For the braced cut, given: Hs = 6 m; Hs , = 2 m; ys, =
16.2 kN/m 2; angle of friction of sand , , = 34 0; H. = 4 in, y, = 17.5 kN/m 2;
and unconfined compression strength of clay layer, qu = 6S kN/m2.
a. Estimate the average cohesion (cav) and average unit weight (yav) for the
construction of the earth pressure envelope.
b. Plot the earth pressure envelope.
7.6 Refer to Figure 7.5b on p. 311, which shows a braced cut in clay. Given:
H = 22 ft, = 6 ft,H1 = 2125 lb/ft2, y1, = 111 lb/ft', H2 = 8 ft, c2 = 1565 lb/ft2, y2 =
107 lb/ft3, H 3 = 8 ft, c3, = 1670 lb/fr2, y3 = 109 lb/ft 3.
a. Determine the average cohesion (c av.) and average unit weight (yav) for
calculation of the earth pressure envelope.
 b. Plot the earth pressure envelope.
7.7 Refer to Figure P7.7. Given: y = 17.5 kN/m 2 , c = 30 kN/m 2 , and center-
to-center spacing of struts = 5 m. Draw the earth pressure envelope and
determine the strut loads at levels A, B, and C.

7.8 For the braced cut described in Problem 7.7, determine:

a. the sheet pile section;

b. the section of the strut subjected to the largest load.
7.9 Redo Problem 7.7 assuming c = 60 kN/m2.
7.10 Determine the factor of safety against bottom heave for the braced cut
described in Problem 7.7. Use Eqs. (7.9) and (7.11). For Eq. (7.11),
assume the length of the cut, L = 18 m.

7.11 Determine the factor of safety against bottom heave for the braced cut
described in Problem 7.9. Use Eq. (7.11). Length of the cut is 12.5 m.
References
American Institute of Steel Construction (1980). Manual of Steel Construction,
Eighth Edition, Chicago.
Bjerrum, L., and Eide, 0. (1956). "Stability of Strutted Excavation in Clay,"
Ceotechnique, Vol. 6., No. 1, pp. 32-47.
Das, B. M. (1979). Introduction to Soil Mechanics, Iowa State University Press,
Ames, Iowa.
Das, B. M., and Seeley, C. R. (1975). "Active Thrust on Braced Cut in
Clay,"Journal of the Construction Dicision, American Society of Civil
Engineers, Vol. 101, No. C04, pp. 945-949.
Harr, M. E. (1962). Ground Water and Seepage, McCraw-Hill, New York.
Mana, A. I., and Clough, C. W. (1981). "Prediction of Movements for Braced Cuts in
Clay," Journal of the Geotechnical Engineering Division, American Society of
Civil Engineers, Vol. 107, No. CT8, pp. 759-777.
Marsland, A. (1958). "Model Experiments to Study the Influence of Seepage on the
Stability of a Sheeted Excavation in Sand," Ceotechnique, Vol. 3, p. 223.
Peck, R. B. (1943). "Earth Pressure Measurements in Open Cuts, Chicago (Ill.) Sub-
way," Transactions, American Society of Civil Engineers, Vol. 108, pp. 1008-
1058. Peck, R. B. (1969). "Deep Excavation and Tunneling in Soft Ground,"
Proceedings, Seventh International Conference on Soil Mechanics and
Foundation Engineering, Mexico City, State-of-the-Art Volume, pp. 225-290.
Terzaghi, K. (1943a). "General Wedge Theory of Earth Pressure," Transactions,
American Society of Civil Engineers, Vol. 106, pp. 68-97.
Terzaghi, K. (1943b). Theoretical Soil Mechanics, Wiley, New York.