TM 5-818-4/AFM 88-5, Chap. 5
backfill operations. The success of ensuring required
is being obtained. Walking out means the roller begins
backfill density often determines the functional serv-
bearing on the soil through its feet only-the drum is
ice of the imbedded structure. Good control involves
riding a few inches above the soil surface. If the roller
many techniques. An experienced inspector will not
walks out after only a few passes, the soil is probably
rely on any one technique but from experience will
too dry; if it does not walk out but continues churning
base his control on a combination of techniques. Mois-
up the material after the desired number of passes, the
ture-density control techniques may be grouped into
soil is too wet or the foot contact pressure is too high.
(b) A trained inspector will spend some time in
three categories: rule-of-thumb techniques, and indi-
the field laboratory, performing several compaction
rect and direct moisture density measurements.
(1) Rule-of-thumb methods. Rule-of-thumb tech-
tests on each type of backfill material to become famil-
niques are derived from experience and are based on
iar with the differences in looks, feel, and behavior
visual observations and feel of the material. A rule-of-
and learning to recognize when they are too dry or too
thumb for judging if the water content of a fine-
wet, as well as when they are at optimum water con-
grained, plastic material is near the optimum water
tent.
content consists of rolling the material between the
(2) Indirect methods. Indirect methods of deter-
hands until it forms a thread approximately 1/8 inch in
mining the density and water content involve meas-
diameter. If the material at this stage tends to crack or
urement of the characteristic of the material that has
crumble, it is in the proper water content range for
been previously correlated to the maximum density
compaction. It will be recognized that this method is
and optimum water content. These methods of meas-
similar to the method of determining the plastic limit
uring in-place density and water content can usually
of a soil. The methods are similar because the optimum
effect a more detailed control of a job than can be ac-
water content for compaction of a cohesive soil
complished by direct methods alone because they can
provide quicker determinations. However, no indirect
roughly approximates the plastic limit of the soil.
(a) Another good indication of whether the
method should ever be used without first checking and
proper water content has been obtained can be deter-
calibrating it with results obtained from direct meth-
mined by observing the compacting equipment. When
ods, and periodic checks by direct methods should be
a sheepsfoot roller is being used and the soil sticks to
made during construction. Indirect methods include
the use of the nuclear moisture-density meter, the
the roller to any great extent, the material is being
Proctor penetrometer (often referred to as the "Proc-
rolled too wet for the equipment being used; at opti-
mum water content it may be expected that a few
tor needle"), the hand cone penetrometer, and in the
hands of an experienced inspector even a shovel.
clods will be picked up by the roller but a general stick-
ing will not occur. If the compacted fill does not def-
(a) The nuclear moisture-density method con-
initely spring (noticeable to visual observation) under
ducted in accordance with ASTM D 2922 (for density
hauling and compaction equipment, it is probable that
determination) and ASTM D 3017 (for water content
several lifts of fill have been placed too dry. The roller
determination) is the only indirect control method
should roll evenly over the surface of the backfill if wa-
used for the Corps quality acceptance control. The
ter content is uniform throughout the lift and should
method provides a relatively rapid means for deter-
not ride higher on some portions of the backfill than
mining both moisture content and density. Of the
three methods presented in ASTM D 2922, Method B
on others. If on the first pass of a rubber-tired roller
the tires sink to a depth equal to or greater than one-
- Direct Transmission is the best suited for a com-
half the tire width, if after several passes the soil is
pacted lift thickness exceeding approximately 4
rutting excessively, or if at any time during rolling the
inches. The nuclear moisture-density method is dis-
weaving or undulating (as opposed to normal "spring-
cussed in more detail in section B- 3, appendix B.
ing" of the surface) of the material is taking place
(b) Penetrometers, such as the Proctor and hand
ahead of the roller, either the tire pressure is too high
cone penetrometers, are useful under certain condi-
or the water content of the material is too high. On the
tions for approximating density. However, both meth-
other hand, if the roller tracks only very slightly or not
ods require careful calibration using soils of known
at all and leaves the surface hard and stiff after sev-
density and water content and considerable operating
eral passes, the soil is probably too dry. For most soils
experience. Even then, the results may be questionable
having proper water contents, the roller will track
because nonuniform water content (in fine-grained
evenly on the first pass and the wheels will embed 3 to
material) or a small piece of gravel can affect the pene-
4 inches. Some penetration should be made into soil at
tration resistance. Penetrometers, therefore, are not
its proper water content, though the penetration will
recommended for general use in compaction control;
decrease as the number of passes increases. After sev-
however, they can be a very useful tool in supplement-
era1 passes of a sheepsfoot roller, the roller should
ing the inspector's visual observations and providing a
start walking out if adequate and efficient compaction
general guide for detecting areas of doubtful compac-
7-5
Previous Page
