Escolar Documentos
Profissional Documentos
Cultura Documentos
Scholars' Mine
International Conference on Case Histories in (1993) - Third International Conference on Case
Geotechnical Engineering Histories in Geotechnical Engineering
Jun 1st
Recommended Citation
M. T. Davisson, "Negative Skin Friction in Piles and Design Decisions" ( June 1, 1993). International Conference on Case Histories in
Geotechnical Engineering. Paper 7.
http://scholarsmine.mst.edu/icchge/3icchge/3icchge-session15/7
This Article - Conference proceedings is brought to you for free and open access by Scholars' Mine. It has been accepted for inclusion in International
Conference on Case Histories in Geotechnical Engineering by an authorized administrator of Scholars' Mine. This work is protected by U. S. Copyright
Law. Unauthorized use including reproduction for redistribution requires the permission of the copyright holder. For more information, please contact
scholarsmine@mst.edu.
1!1111 Proceedings: Third International Conference on Case Histories in Geotechnical Engineering, St. Louis, Missouri,
~ June 1-4, 1993, SOA No. 1
SYNOPSIS Negative skin friction behavior of pile foundations is described for conditions of pile usage
in the USA. Methods of determining downdrag load and pile resistance to downdrag are explained. Other
Eactors entering into design are discussed such as pile load testing and analysis, pile structural
strength, factors of safety, pile drivability, and reduction of downdrag loads. Observations from seven
~npublished negative skin friction failures are used for illustration.
Negative skin friction can be accommodated in If the LRFO equations presented herein are
LRFD design techniques for both structural and utilized for a site with variable thickness of
geotechnical conditions. The structural equation subsiding stratum, it is possible that the site
is: may be divided into zones in a pattern defined by
the engineer. Each zone may then be analyzed
1.40 + 1.7L +1.4NSF ~ ~pn (Str) (6) according to the particulars of each zone.
Finally, the engineer can make a judgement to use
one value of Rutt or several different Rut-values
Note that NSF has been assigned a load factor of if the choice is made to construct the project
1. 4 when determined. from tension load tests, according to different zones on the job site.
which the writer recognizes as the lowest factor
that can be applied in uniform soil strata(same Designers utilizing the techniques given above
as dead load). ACI 318, however, treats earth should be cognizant of the need for communicating
loads with a factor of 1.7; therefore, others can the resulting requirements in the plans,
argue for a higher factor. It would be specifications, and procedures for pile driving
appropriate to use a higher load factor if NSF is control during construction. An example will
determined by calculation or other indirect make the point. Assume a project where dead load
means, or if the soil stratification is equals live load, NSF is equal to dead plus live
nonuniform. load, a= 0.2, and = 0.7; a required value of
Rut can then be calcu\ated from Equation 7. The
The companion geotechnical equation is: resulting value of Rut will be 2.96(D+L+NSF),
indicating that the required load from a load
1.40 + 1.7L + 1.4NSF ~ test must be approximately 3 times the applied
load. This is higher than the factor of 2
8Rut - (1+a)NSF (Geot) (7) normally encountered, and may not be understood
by field personnel that are not versed in the
The new item in Equation 7 is a, which accounts details of the design. Also, if the factor of
for the variation in NSF, and is used to obtain safety as defined herein is calculated, the
an upper bound on NSF; it is likely to have a result is 2. 46 calculated as (Rut-NSF) I (D+L+NSF) .
value of o. 2 at the lowest.. The expression Even the latter result will not be understood by
requires the factored loads (left side) to be field personnel unless properly communicated.
less than the ultimate pile load reduced by the
NSF (right side) Several other topics related to pile foundation
practice enter .into the designer's actions
concerning NSF, even if the designer may not be
'wo of the case histories in Table 1 (#4, 5) Batter piles in subsiding soil layers are
.nvolve thin-wall pipe piles which have a subjected to bending loads that can easily
elatively small cross-sectional area of steel. destroy the effectiveness of the piles.
~he axial stiffnesses were too low to allow Treatment of this problem typically consists of:
>enetration into the bearing layer with 1) eliminating batter piles, 2) decreasing pile
:ufficient force to develop the required load batter angle until the downdrag is noncritical in
:apacities. Another case history (#7, Table 1) bending, or 3) finding an alternative way to
.nvolved prestressed piles wherein drivabili ty provide lateral stiffness to the supported
ras adequate. However, where piles were furnished structure. If the engineer chooses to use batter
.anger than required, they were overdriven to piles, they must be designed for the interaction
tvoid pile cut-offs, and performed of axial and bending loads, including stability
:atisfactorily. Unfortunately, some piles were considerations.
:horter than actually required and were
mderdriven; they failed because of NSF. This OTHER ANALYTICAL METHODS
.atter case history demonstrates the need to
levelop pile load capacity in the designated Two additions to the analytical methods explored
>earing layer. previously are often used. The first 1.s to
reason that transient live loads do not add load
~he foregoing case histories reveal several to the piles. This is arrived at by reasoning
:actors that must be considered in producing a that a transient load (not to be confused with
>ile foundation that is successful after it is the long-term portion of live load which acts
:onstructed. Discussion and guidance on several similar to dead load) causes a temporary downward
>f these factors is given in the paragraphs that movement of the pile relative to the soil,
:allow. thereby relieving negative skin friction and
replacing it with positive skin friction which
>ILE SHAPE AND DOWNDRAG supports the transient load. Obviously, this can
only work within limits. If a reduction in the
~ load transfer analysis for the conditions on design pile load is made on account of such an
.<igure 2a would show that load in the pile analysis, then some economy will eventually be
lncreases from the superimposed load at the top realized in the cost of piling.
)f the pile to a higher value that includes the
1egative skin friction at (or near ) the bottom Another analytical technique utilizes a knowledge
)f the subsiding layer. Then, the load in the of soil compressibility to determine to what
)ile diminishes with respect to increasing depth depth the pile is moving downward equally with
l.S it is transferred to the bearing layer. the soil. Above this point the soil causes
)bviously, a pile of relatively small perimeter downdrag (negative friction), and below this
vill attract less NSF than a larger perimeter. point the soil is functioning to support the pile
~lso, the pile must be strongest where the load (positive friction) . If this point is
Ls greatest (near the bottom of layer 1, Figure significantly above the bottom of layer 1 in
~a). With these thoughts in mind, it can be said Figure 2a, then a lower NSF load will be
:hat tapered piles such as timber or step-tapered calculated, and will eventually result in some
?iles are the wrong shape for dealing with major economy in the cost of piling. However for
iowndrag loading. (They would serve better if subsiding soil layers where the settlement; will
turned upside down.) be. large. (s~veral inches to feet), the depth at
wh1.ch fr1.ct1.on changes from negative to positive
l"our of the case histories cited in Table 1 will be. near or at the bottom of the subsiding
involved timber piles; two of the supported layer; 1.t may be so close to the bottom that it
structures collapsed to the ground. Timber piles falls within the engineers ability to define the
~nd downdrag are a deadly mixture because of pile depth of the various layers in a soil profile.
shape and timber weakness exacerbated by the 40
percent loss of strength of timber under 10-year
,CKNOWLEDG EMENT
~EFERENCES