consider the base rigid, and the preliminary analysis

suggestions for use of battered piles are defined in

can be performed using a rigid base analysis tool,

EM 1110-2-2906. In using battered piles, consider-

such as the computer program CPGA (Hartman et al.

ation must be given to geometric constraints from

1989). This program performs the pile analysis for

adjacent pile-founded monoliths and sheetpile cutoffs.

two- and three-dimensional loading conditions. For

These constraints can be assessed using the computer

subsequent foundation design and any structural con-

program CPGI (Pile Group Interference Check, CASE

crete design, the pile cap should be treated as a

computer program X0086). Generally, piles from one

flexible base. Therefore, the pile foundation and

monolith should not extend into the area beneath an

structural analysis must be performed using a pro-

adjacent monolith because of the possibility of

gram that will consider the internal stiffness relation-

ship of the structure. The same flexible base analysis

can be used to analyze the piles and the concrete

(2) Pile head fixity. If it is not practical to use

structure. This is possible using the computer pro-

battered piles to resist the lateral loads because of

gram CWFRAME (Jordan and Dawkins 1990) or

geometric constraints, all vertical piles may still be a

other frame analysis or finite element applications

possible solution, but lateral deflections may become

which include pile elements. A complete SSI finite

critical. If lateral deflections are too high using verti-

element model can be useful for this analysis, though

cal piles with a pinned condition at the pile head, the

it is usually much more complicated.

pile may be embedded deeper and analyzed as fixed

at the pile cap. Pile head fixity is discussed in

EM 1110-2-2906. Refer to Castella (1984) for more

information on pile head fixity.

any detailed pile analysis, the pile stiffness coeffi-

cients should be determined by performing single pile

(3) Lateral subgrade moduli. When an accept-

analyses based on available soil data or previous pile

able initial layout is achieved based on pile forces

tests with similar soils and conditions. These coeffi-

and stresses, a comparison of calculated pile head

cients are essentially linear springs that approximate

deflections to those seen in test results or assumed in

the nonlinear behavior of the soil-pile foundation.

the pile stiffness coefficient analysis must be made.

Normally, it is desirable to perform a parametric

Since the pile/soil stiffness degrades with deflection,

analysis where the bounds in variability of the coeffi-

the calculated deflections seen in analyses should

cients can be captured. Once determined, these

compare with the deflections assumed or generated in

coefficients are input to the various programs chosen

the selection of the pile stiffness coefficients.

for the pile group analysis. See EM 1110-2-2906 for

a further discussion of stiffness coefficients.

includes checking for resultant location, sliding, uplift

tance in pile foundations is dependent upon the pile

(flotation), differential settlement, and bearing failure.

type, strong axis orientation and batter angle, and on

One of the single most important elements in the

the assumed or experimental lateral subgrade moduli

design of a soil-founded U-frame lock is the assump-

used in design.

tion regarding the distribution of the effective base

pressure.

(1) Pile orientation. Lateral loads are most effi-

ciently resisted by battered piles. However, battered

piles are more difficult to drive and result in a more

approaches used in determining the distribution of

complicated layout to design and construct. Addi-

base pressures. One is a soil spring approach and the

tionally, battered piles tend to dramatically change the

other is an assumed pressure distribution. The spring

pile force distribution. If the lateral loads are not

method is discussed in paragraph 7-3*e*. The assumed

significant, the designer should consider using all

pressure distribution approaches are a uniform distri-

vertical piles. If the lateral load is significant, piles

bution and a stepped distribution with appropriate

with unequal stiffnesses about the orthogonal axes

corrections for eccentric loading. The methods pre-

(H-piles for example) can be turned to increase stiff-

sented have been derived from analysis of instrumen-

ness in the direction of the load or they can be bat-

tation data where base pressures were measured and

tered. The preliminary batter slope and number of

compared with conventional calculations. A tool

battered piles can be determined by using force vec-

available to compute base pressures is the CASE

tors or similar methods. Capacities, limitations, and

computer program 3DSAD (Tracy and Kling 1982).

A-11

Integrated Publishing, Inc. |