1.1 Purposeandscope: 33 of A of Is Fill in
1.1 Purposeandscope: 33 of A of Is Fill in
1.1 Purposeandscope: 33 of A of Is Fill in
l
ria
Introduction
ate
dM
1.1 PURPOSEANDSCOPE
l
aspects of all types of compacted fills, and includes a section on potential
ria
problems with earth structures,with an emphasis on earthen dams. The intent
is to provide information helpful to those involved with the National Dam
Inspection Program, thus serving the dual purpose of improving capabilities
in the inspection and remediationofcompleted earth structures and in the full
range of activities needed for design, inspection, and construction of new
ate
earthworks.Chapter 7 complements Chapter 6with a description of all details
relating to compaction specifications.
Chapter 8 (plus some aspects of Chapter 7, notably Specification Evalua-
tion, Section 7.4) is intended to serve as a manual for fill control procedures: It
is written expressly for the typical inexperiencedyoung geotechnical engineer
dM
or engineering technologist.
As a departure from typical format, chapter glossaries are provided at the
end ofeach chapter to explain certain terms used in the chapter that could not
be explained fully within the text without adversely affecting readability.I sug-
gest that the reader review each chapter’s glossary before reading the chapter
of interest.
hte
The importance of soil compaction and fill control can be emphasized by one
simple declaration: Almost no significant engineered construction occurs
without the movement of soil from one place to another. Furthermore, it
rig
often left out of the sequence. All too often, they are called in (late) to correct a
bad situation or to act as expert witnesses when it i s already too late for
correction.
In the ideal sequence,the geotechnical engineer explores and evaluatesthe
Co
l
The structural engineer is one of the primary recipients of the geotechnical
ria
engineer’s report. In addition to the obvious responsibility of structural de-
sign, it is his typical responsibility to prepare working drawings, including
foundation drawings. The preparation of construction specifications be-
comes a natural extension of this work, typically as notes on the drawings and
(for large projects) separate, additional documents. In the usual situation, the
ate
structural engineer will need to consult architectural drawings and geotechni-
cal reports and communicate with the architect and geotechnical engineer
when questions develop requiring their attention and expertise.
Construction engineers and contractors are, of course, the “doers.” They
bring to fruition the studies, concepts, designs, drawings, and specificationsof
dM
the architect,the geotechnicalengineer, and the structural engineer, thus com-
pleting the sequence of operations of engineered construction.
lifts of fill by the direct application of load, impact, and/or vibration, usually
with a suitable type of compaction equipment. Lift thicknesses vary from
several inches (for clays and silts) to perhaps 2 ft for free-draining fills (sands
and gravels). (Note: One should not confuse soil compaction with consolida-
tion, the long-term reduction of void ratio of a natural soil, usually saturated,
thick deposits of soft clays or silts beneath the water table. This is usually
rig
tive soil sample, obtained from the field, in a steel mold of standard size. The
soil is compacted usually in layers, commonly by dropping a hammer of
specified weight through a specified distance a specified number oftimes. The
energy of such compaction is chosen to simulate that of field compaction,
Co
usuallywith a rollerjudged suitable for the conditions and soil type. The mois-
ture content is varied for a series of filled molds, thus generating a compaction
curve for the soil. A typical curve is shown in Figure 1.1.
I I
Standard Proctor Compaction Test
I (ASTMiD698) I
130 I I I
Silty sand (SM)
I
l
Acceptable
ria
range of field
ri;' 12c /densities
esa
= 114 -
.-b
5 11c OMC
ate
n
\(
90
-
100% Standard Proctor = 120.0Ib/ft3
95% Standard Proctor 114.0Ib/ft3
dM
0 10 20 30
Compaction moisture (%)
Rgwe 1.1. A iypicul compaction curve.
Drainage QwiiiyDesignations
hte
quired one or more vowels.Thus G is for gravel,C is for clay, SS is for sand and
silt. E,0,and I, of course, are needed vowels. The letter 1might also represent
inorganics, the soil minerals. In the Unified Soil Classification System, 0 is
used to represent organics, and M to indicate silts. EM also stands for Ed
Monahan.
There are many examples illustrating the need for the concept of the
gemicoss. Often a chart or a formula has a principal limitation of being
applicable to a certain soil type, that is, sand or clay. Blowcounts, for example,
l
€3Wnt/u/~Grunulcrr (Or Cohesiw)Sdk Texture that predominates in the
ria
context of dictating the overall behavior of the soil. This terminology is a
natural extension of the notion of a gemicoss. After carefully evaluating the
texture (sizesand plasticity) of the soil, one may be forced to decide whether it
is essentiallygranular or essentiallycohesive inorder tojustify the use of a par-
ticular design chart or formula.
ate
For example, if a soil is composed of 60% highly plastic clay (by weight),
with 40% sand and gravel, the soil would be an ECS,an essentially cohesive
soil, since the sand and gravel particles are, for all practical purposes, merely
isolated or “suspended” in a clay matrix, and their presence is essentially
irrelevant to the overall engineeringbehavior of the soil.Note that this exam-
ple does not fit the classificationof a gemicoss, because the conclusion (ECS)
dM
is fairly evident. A gemicoss might be 40% medium plastic clay, 60% sand and
gravel. Such a mixture raises questions pertaining to contrasting engineer-
ing properties.
As a contrasting example, if a soil is 75% gravel and sand, with 25% clay of
high plasticity, it would be rated an essentially granular soil (EGS),inasmuch
as it is reasonable to assume that the preponderance of granular particles are
hte
U// The soil that is selected for use at the site. It may be from a site excavation
or “imported” from elsewhere. In the latter case, the fa is termed borrow.
U// c)uU//iy lndlces Texture: grain sizes and grain size distribution of co-
hesionless soils, or plasticity (usually plasticity index, or PI) of cohesive
py
soils.
R// Cond/t/on Density (pounds per cubic foot, pcf) or relative density (per-
cent), a measure of potential settlement. For clay soils, potential expansion is
Co
expand when wetted. Conversely, a cohesive (clay) soil can settle unaccep-
tably if not compacted sufficientlybut can expand detrimentallyupon (future)
wetting if overcompacted. The pressures associated with such expansion can
be substantial, causing damage to highway pavements and even heavy struc-
tures. Clays and silts are also susceptible to disturbance by seepage pressures
l
(piping,uplift) because of their relative imperviousness. Silts are also suscept-
ria
ible to frost action.
ate
respects, simulate what will be done to the soil in the field. As one of the more
obvious examples, the method and energy of compaction in the laboratory
compaction test should approximate that of construction rollers used in
the field.
dM
index pIopercfes Indicates the general nature of the subsurface problems
that must be confronted. Such index properties are (or can be) obtained
routinely and at modest expense in a typical subsurface investigation. Figure
1.2 shows, in simple flowchart form, how index properties should be used.
Note that large jobs, requiring the expenditure of larger amounts of money,
would justify the expenditure of larger amounts for field or laboratory testing
to determine engineering properties needed for design. Engineering proper-
hte
ties are defined as those measuring stress, strain, and strain rate on representa-
tive soil samples that enable computation of specific numerical values of
bearing capacity and settlement. Thus, path ABC provides a design that is
more reliable than path AC, but at considerably more cost. Path AC, for small
jobs, where extensive testing is not economicallyjustified, utilizes index prop-
erties directly for design purposes, usually by entering a chart to select allow-
rig
able bearing capacities, rather than the more expensive ABC route. Of
considerable practical importance, moreover, is the fact that index properties,
ifobtained and properly evaluated,serve as the basis for a rational testing pro-
gram (step B). In summary, index properties serve a twofold purpose and are
indispensable tools in geotechnical engineering.
py
BIOw Count, -
, andRelCrtlve Derulty Illustrativeof the foregoing de-
scription of index and engineering properties are the definitions of blow
count, density, and relative density, particularly as related to compaction and
Co
fill control and the stability and bearing capacity of granular, or essentially
granular, soils.
When a driller is sampling a soil that is expected to be an essentially
granular soil, a sampling device is driven into the soil and a blow count N
(blows per foot) is obtained. Clearly,the blow count (with any appropriate cor-
rections) is an indicator of density: the higher the blow count, the higher
the density.
A. Sampling and
index testing
l
ria
ate
C. Design B. Laboratow
or
Legend: NF= Field blow count, standard penetration test field testing
N, = Design blow cwnt (corrections to NF) (determination
IL = Liquidity index (clay soils) of engineering
9, = Unconfined compression test (cohesive soils) properties)
dM
Rgun +aDesign fbwchart.
Thus, a soil of 100% relative density would have a negligible potential for
future settlement. Furthermore, given the thickness of the soil, a knowledge of
the relative density would allow for the computation of a specific, numerical
settlement estimate. A judgment could then be made as to whether the com-
puted settlement would be tolerable.
In this example, the blow count is an indicator (or index property) signify-
rig
ing the general condition of the soil (e.g., very loose, loose, medium-dense,
dense, very dense), whereas the relative density (involving the added expense
of laboratory testing) is ap engineering property.
With respect to fills, Ifnoclvledge,of the relative density and the compacted
density-which the designer can control by appropriate specifications-
py
If the texture and blow coiints of a soil being considered as potential borrow
are determined from exploratorydrilling and sampling operations,what prac-
tical infomation can be derived from such index data?
1A.1 case1
Case 1 involves a well-graded gravelly sand, with approximately 25% fines of
significant plasticity. (PI on m i n u s 4 fraction is determined to be 30. Liquid
limit (LL) is 65.) Blow count range, 35-60.
l
~ ~ ~ “ ~ r m , ~ r w l r rFines ( . all soils passing the No. 200 sieve, thus
K ) )are
ria
silts or clays. If fines are judged by simple and quick field tests to be signifi-
cantly plastic (clayey rather than silty), Atterberg Limits tests are ordered on
that portion of the soil passing the No.40sieve:the m i n u s 4 fraction. Result is
PI = 30; LL = 65. Such fines would be rated “high plasticity,” CH in the
ate
Unified Soil Classification System. In summary, this soil is a dense to very
dense,well-graded,essentiallygranular soil, but containing cohesive clay bin-
der. In a real situation, the locale of the potential borrow area would, of course,
be known. Thus, soil maps, geologic maps, and perhaps personal knowledge
of the area (geologic, topographic, and land-use, for example) would be avail-
able to augment the data from the exploratory program. The soil described is
dM
typical of a glacial till, soil deposited directly by a glacier in a mechanical
fashion as opposed to alluvial deposition (by say, glacial meltwater). This
accounts for the wide range of sizes, as contrasted to uniform sizes associated
with the sorting action of flowing water. Thus, one could reasonably expect to
encounter boulders in the area. Confirmation could be obtained by recon-
naissance, particularly by inspection of road cuts in the vicinity.
Before proceeding with a listing of the practical and potentially very valu-
hte
able information that can begleaned from the foregoing, I feel it is necessary to
add some commentary here regarding a very important subject: soil (and
rock) descriptions as they pertain to formal classification systems. There are
several such systems. The most commonly used, in the sense of broadest
acceptance in the United States, are the Unified system and the AASHTO sys-
tem (the American Association of State Highway and Transportation Offi-
rig
The description given for Case 1, although it will serve its purpose here, is
not in accordance with any formal system. Thus, in professional practice, it
would be of little value, since “well-graded,”“essentially granular,” and “con-
taining clay binder” have no well-defined meaning to other potential users. A
soil classificationsystem, then, is seen as a universally agreed upon language,
l
wherein all words and symbols used have specific definitions. When people
ria
describe soils without adopting a recognized system, no one, save themselves,
can be sure of what they mean. Unfortunately, it is all too common to see soil
descriptions in practice, generally written by one of our nonspecialists, that
are ambiguous, vague, and often visually useless because of usage of ter-
minology not belonging to any recognized system.
ate
For a brief yet excellent treatment of the Unified and AASHTO systems
(and some others) refer to fitter and Paquette ( l W ) , Chapter 6. For a com-
plete description of the Burmister system, see Burmister (1953,1955,1958).
With definitions,discussion,and commentaryregardingthe importance of
careful soil classification, the following list briefly gives some representative
dM
examples of valuable information pertaining to Case 1.
lA.2 case2
Co
Case 2 involves a sand with coarser sizes predominant, fine gravel present,
and no significant silty materials evident. (Sieve test showed fines fraction to
be 7%; dilatancy test on fines confirms no plasticity-NP.) Blow count range,
15-25.
*Dlo size, also called Hazen’s effective size, can be used to estimate soil permeability.
l
Geologically,this select fill is probably of alluvial origin. Judging by its low
ria
blow counts, it is a recent alluvium (mapped as AR on soil maps published by
SCS,the Soil Conservation Service, a division of the U.S. Department of
Agriculture) and has the following characteristics.
ate
areas) and is thus a valuable “commodity”; it should therefore be re-
served for special purposes and used sparingly.Typical special purposes
are trench bottoms to support pipelines and graded filters in earth-
dam construction.
2. Free-draining. Thus, not susceptibleto seepagepressures, frost action. If
dM
properly compacted, it is permanently stable (except Richter 8’s).
3. Easily excavated (low blow counts, no cohesion).
4. No laboratory compaction test is necessary (or even sensible). Write
compaction specificationson basis of percent relative density.Compact
by flooding.
5. Best compactor would be a heavy vibratory roller.*
hte
1.5 GLogsARy
repeatedly with the other hand, while one observes the surface changes (if
any) of the wafer-shaped soil. A ready and quick shiny, glassy change in
appearance (i.e., within seconds) constitutes a positive dilatancy test, and
means that the soil is nonplastic(NP). Ifone squeezes the hand with the pat,
the surface will immediately become dull in appearance; resumed tapping
Co
will quickly recreate the shiny appearance. (The squeezing shears the soil
and allows the surface fluid to quickly drain away, thus assuming the dull
appearance. The tapping consolidatesthe soil, and the water film reappears
on the surface to yield the shiny appearance.) The quickly positive result
identifies the soil as a “clean” silt (as opposed to, say, a clayey silt) or a fine
sandhilt mixture. The longer it takes to produce any reaction (called di-
latancy), the higher is the degree of plasticity. This and other field tests are
used by experienced soils people to assessthe plasticity rating ofthe soil in a
qualitative way. For confirmation, the PI is determined by l a w t o r y
l
analysis for the liquid limit and the plastic limit, both ofwhich are moisture
ria
contents. The numerical difference is the PI. (For details concerning field
testing or laboratory testing, see Burmister (1953, 1955,1958) and Lambe
(1951), respectively.)
Dlo size This means that 10%of the soil (by weight) is finer than that size.
Thus, it may be thought of as an indicator of the finenessof a soil; an impor-
ate
tant index in assessing the general seepage quality of a soil. (See also Sec-
tion 5.1.2.)
Frost action (frost heave) The seasonal heaving (winter) and recession
(spring breakup) of frost susceptible soils, notably silty soils. Aserious pro-
blem in highway and other route design. (See also Section 6.2.1.)
dM
Graded filter Asoil filter that provides a gradual (graded)transition for water
flow, to prevent piping, for example. (See also Section 6.1.)
Granular (cohesionless) Coarser particles of soil (gravel, sand, “clean” silts)
that exhibit no plasticity or cohesiveness.
Piping The dislodging and movement of finer soil particles under the in-
fluence of seepageof water through the soil, starting usually as a slow proc-
ess, and leading eventually to greater distress and, possibly, failure
hte
which the soil acts in this way is called the plasticity index (PI). All soils
exhibitinga measureablePI are cohesive, meaning simplythat the particles
“stick together” naturally by other than (or in addition to) capillary forces.
Clays, clay mixtures (clay with other soil types), and organic soils are
cohesive.
py
Richter 8 The Richter scale is used to rate the intensity of earthquakes. Data
obtained from seismographs located throughout the world results in the
rating, which is on a logarithmic scale (base 10);thus a Richter 6 is 10times
more intense than a Richter 5, and so forth. A Richter 8, used somewhat
Co
facetiouslyin the text, would be a rare and very major earthquake ofdevas-
tating effect.
Unconfined compressive strength The compressivestress at failure on a cylin-
drical specimen,usually expressed in tons per square foot (tsf); commonly
performed on sampled, natural soils of a cohesive nature (e.g., clays). There
is no reason, however, that such a test cannot be performed on compacted
samples of suitable cylindrical dimension.