Invited keynote lecture and paper, Geotechnical Special Publication No.

124, ASCE,
pp. 19-33. Geo-Institute International Conference on Drilled Foundations, Orlando,
2004.

ZEN AND THE ART OF DRILLED SHAFT CONSTRUCTION:

THE PURSUIT OF QUALITY
Dan A. Brown, P.E., M. ASCE

ABSTRACT: With apologies to author Robert Pirsig (Zen and the Art of Motorcycle
Maintenance), this paper pursues the concept of quality with respect to construction of
drilled shaft foundations. With the evolution of more sophisticated techniques for
integrity testing and load testing, it is possible now to better observe the end result of our
construction activities and make judgments about the effectiveness of techniques and
materials at achieving quality. This paper describes some aspects of construction
techniques and materials which can lead to defects or less than optimal performance for
drilled shaft foundations. Examples are cited of some of the more common problems
encountered in drilled shaft construction in order that lessons can be learned from these
problems. The case is made for designers and contractors to emphasize constructability
in designs, workability in construction materials, and individual responsibility toward
quality on the jobsite.
INTRODUCTION
In the book Zen and the Art of Motorcycle Maintenance, author Robert Pirsig
describes a personal spiritual journey (and a motorcycle trip) in a quest for quality. In
the construction of drilled shaft foundations, developments in integrity testing and load
testing have afforded the industry an improved opportunity to assess the quality of the
end product of our labors in terms of the integrity and load carrying capacity of the
foundation. While this improved vision has often lead to disputes over the definition of
quality that was specified in the contract documents and that is generally provided by
current practice, most engineers and constructors can recognize those attributes of quality
that are desirable in a drilled foundation:
The foundation should consist of a relatively uniform mass of sound concrete,
The concrete should have good bond and load transfer to the bearing
formation,
The reinforcement should be in the intended position and should be bonded to
the concrete.
This paper will address some aspects of design and construction which influence the
tendency to achieve quality in drilled foundations. A brief description of some selected

Page 1

Dan Brown

case histories is included in order that lessons may be learned from these experiences.
The general theme which emerges from these case histories is that designers and
contractors need to consider and emphasize constructability in designs, workability in
construction materials, and individual responsibility toward quality on the jobsite.
KEY ELEMENTS FOR QUALITY IN DRILLED SHAFTS
The writers experiences suggest that the majority of construction problems which
compromise the quality of drilled shafts come from a failure to adequately consider one
or more of the following categories:
Workability of concrete for the duration of the pour
Compatibility of congested rebar and concrete
Control the stability of the hole during excavation and concrete placement,
especially with the use of casing
Drilling fluid which avoids contamination of the bond between the concrete
and bearing material or excessive suspended sediment
To this list can be added a broader category, which is human attentiveness to any or all
of the above. Inattentiveness can be the result not only of carelessness in workmanship,
but also to contractual arrangements that do not encourage attentiveness and to
inadequate resources devoted to inspection and quality control.
WORKABILITY OF CONCRETE
Workability can be defined as the ability of the concrete to readily flow through the
tremie, the rebar cage, and to all places within the hole where it needs to go. With drilled
shaft construction, this must be achieved without the need for external sources of energy
such as a vibrator. Most commonly, slump is the measured property associated with
concrete workability. When concrete has inadequate workability, several problems can
ensue:
1. During tremie placement of concrete, there is a tendency for debris to become
entrapped within the concrete and thus produce flaws in the structural integrity
of the foundation. This can occur as the oldest concrete in the shaft is riding
on top of the rising column of concrete, and as this old concrete becomes stiff
then the fresh concrete can tend to burp through and trap the debris and/or
contaminated concrete on top. Loss of workability can also lead to plugging in
the tremie itself, which may cause the contractor to breach the tremie in order
to get flow going again. The result of the breach would also be to trap debris
and/or contaminated concrete as some concrete would tend to flow through
water and lose cement.
2. Even with placement of concrete into a cased hole without the use of a tremie,
there is a need for concrete workability to be maintained from start to finish.
When the casing is removed, the concrete must have adequate workability to
flow through the rebar cage, displace the water that may be present outside the
casing, and produce lateral stress against the soil or rock so as to provide a
good bond within the bearing stratum. If the concrete workability has been
lost by the time the casing is pulled, it may be very difficult to remove the
casing. The concrete could tend to arch within the casing and be lifted with the
casing, thus forming a neck. Even if the casing is recovered without necking

Page 2

Dan Brown

the concrete, a column of stiff concrete that has been slip-formed into an
oversized hole is not likely to provide good bond to the bearing formation.
The presence of a heavy rebar cage can complicate the problem, as the lateral
concrete flow after removal of the casing will be restricted by the cage.
The FHWA guidelines (ONeill and Reese, 1999) for drilled shaft concrete suggest
that a slump of around 200 mm (8 inches) should be used for tremie placed concrete.
Many state DOTs use specifications which routinely call for a slump of at least 100 mm
(4 inches) to be maintained for a period of 4 hours after batching. It is the opinion of the
writer that a 100 mm (4 inch) slump is probably not adequate for most conditions. If
concrete with 200 mm slump is being placed into concrete which now has 100 mm
slump, there will be two dissimilar fluids interacting within the hole with potentially
undesirable consequences.
Rather, it is suggested that the concrete mix be designed to have a very high
workability (slump loss of no more than 50 mm, or 2 inches) for the duration of the
period required for placement, whatever that period may be. These days, it is quite
possible to use admixtures to retard concrete for many hours. The concrete mix design
should have workability and the time required for the construction sequence as a primary
component of the mix design process.
Example 1 Surface Flaws in Concrete
Observation
Single drilled shafts were used to support individual columns for a bridge over a lake
in the southwest. The shafts utilized casing extending through the lake and the relatively
thin alluvial soil overlying rock. The upper portion of the shafts were formed using a
removable casing so that no permanent steel casing would be visible within the zone of
water fluctuation. The shafts were drilled using water only as a drilling fluid and the
contractor appeared to do an excellent job of cleaning the hole. The rebar cage was not
particularly restrictive to concrete flow, with relatively wide openings of at least 200 mm
between longitudinal and transverse rebars.
The project was a 45 minute drive from the concrete plant. Upon arrival at the jobsite,
the concrete trucks were placed on a barge and ferried to the foundation location.
Concrete was placed using a tremie, with concrete delivered from the truck to the tremie
by using a bucket. The tremie was maintained at least 2 m below the surface of the
concrete in the hole at all times. Each shaft required 5 to 6 concrete trucks to complete
the pour and the concrete placement took approximately 4 to 6 hours from start to finish.
Upon removal of the forms, the inspectors noted the presence of pockets of weak
concrete which could be chipped away quite easily with a hammer. These pockets
appeared to be a weak, cemented grout-like material that had no aggregates within it.
Photographs of this material, along with a typical pattern after the weak material had
been chipped away, are provided on Figure 1.

Page 3

Dan Brown

Figure 1 Surface flaws in concrete (left), after removal for repair (right)
Explanation
At first it was suspected that either bleed water was contributing to this problem, or
the contractors removal forms were somehow not sealing and water was mixing with
concrete during placement. However, further investigation revealed that the concrete mix
did not maintain sufficient workability for the duration of the pour during the hot summer
months. As tremie placement of concrete continued, the tremie would be lifted from the
bottom of the shaft but always maintaining the base of the tremie at least 2 m below the
surface of the concrete. As the initial charge of concrete (now riding on top, above the
tremie) started to lose workability, the freshly place concrete entering the shaft through
the tremie below the surface of this now-stiff concrete tended to erupt through this old,
stiff concrete like a volcano rather than lifting the entire surface upwards. As this fresh,
fluid concrete vents through the surface of older, less fluid concrete, the latent
cement/water mixture on the surface of the rising concrete plug flowed to the lower
surface outside the cage and became trapped below the fresh concrete. This flow pattern
was visibly revealed in one of the shafts when the contractor pumped off the water above
the concrete after the concrete was well up into the casing, and the placement continued
using the tremie. The inspector reported the volcano of fresh concrete with the lateral
displacement and subsequent trapping of latent water/concrete mixture that was present
on top of the old concrete.
Implications for Performance
The concrete outside the rebar cage serves primarily to transfer load to the soil and as
cover to protect the rebar from corrosion. In this case, the concern is for the long term
durability of this cover. Discussions with a bridge inspection diver with the Texas DOT
indicates that at least one bridge (Lake Houston) built in 1988 using similar construction
techniques is now suffering spalling of concrete from the surface of the shafts.
Lessons Learned
It is critical that the concrete mix have sufficient retarder that the concrete maintain its
workability for the duration of the pour. In this case, that was at least 6 hours from the
time of batching because of the slow delivery of concrete to the foundation location.

Page 4

Dan Brown

Also, the slump life of the concrete mix varies with the temperature, and increased
retarder dosage is required during hot weather.
Example 2 Poor Bond in Rock Socket
Observation
A drilled shaft was installed through about 12 m of soil and socketed approximately 3
m into an underlying rock formation. In order to allow downhole visual inspection of the
bottom of the shaft, the contractor was required to case the hole for the full length. Once
satisfied with the inspection, a load test shaft was constructed using an Osterberg cell
placed at the base of the socket. After placement of the O-cell and rebar, concrete was
placed within the rock socket and the casing subsequently removed. The O-cell was
found to mobilize only less than 0.5 MN of side shear resistance, a small fraction of the
amount which had been expected.
Another test shaft was constructed, only this time a wet hole method was used with
tremie placement of concrete and without casing into the rock. In this case the O-cell test
indicated over 10 MN of side resistance in the socket.
Explanation
The amount of time required after concrete placement to extract the casing allowed the
concrete workability to diminish to the point that the shaft was almost like a slip-formed
column within the rock socket. Because of the lack of lateral pressure between the
concrete and the rock, the side shearing resistance of the socket was very low. There may
have been some additional detrimental effects of using the casing, such as trapping of
debris behind the casing which contaminated the bond between concrete and rock.
Lessons Learned
Even though the casing provides a dry hole the need for concrete workability for the
entire duration of the construction process remains. And although a shaft is constructed
without any observable structural defects in the concrete, it may not be a quality shaft.
COMPATIBILITY OF CONGESTED REBAR AND CONCRETE
In recent years, it seems that contractors have become more well equipped to construct
very large diameter drilled shafts and so engineers have become more prone to design
and specify very large diameter drilled shafts. Large shafts have some compelling
advantages for structures such as highway bridges, where large lateral and overturning
forces are produced by design conditions for seismic, vessel impact, wind, etc. And a
single large diameter shaft can have a smaller footprint than a pile footing, an advantage
when working on congested sites or nearby existing structures. However, with the use of
large diameter shafts designed for large bending moments, the rebar cages can become
quite dense. Added to the rebar is the frequent addition into the cage a number of access
tubes for integrity testing.
Problems can arise from restrictive rebar cages in the following ways:
1. If the lateral flow of concrete is significantly impeded, then there is an
increased likelihood that debris will become trapped in the annular space
outside the cage. This trapping of debris can result from the fact that the rising
column of concrete inside the cage tends to be at a higher elevation than the

Page 5

Dan Brown

concrete outside the cage, so there would be a natural tendency for any
accumulated sediment on top of the concrete to slough off toward the side.
Even a small accumulation outside the cage can be detrimental to the bond in
the bearing formation.
2. Even with a clean slurry, the concrete can be impeded to such a degree that
voids form outside the cage or the lateral stress at the concrete/rock/soil
interface is diminished.
The FHWA guidelines (ONeill and Reese, 1999) recommend that the clear space
between bars be at least 5 times the size of the maximum aggregate. The writer has seen
this guideline routinely violated in practice. In particular where seismic loads are
important, there is a tendency for designers to use spiral confinement with a 90 mm pitch
(3.5 inches), leaving only about 75 mm (3 inches) or less clear between spirals. The
FHWA guidelines would suggest a mix design using a pea-gravel size aggregate for this
case. Some state DOTs are using such a mix with success. Workability of the concrete
is enhanced in such severe cases if the aggregate is specified to be a rounded gravel rather
than a crushed stone. It would also be prudent for designers to consider the implications
of the use of such tight spiral reinforcement, and consider if the needed confinement of
the interior concrete can be provided in a way which is more easily constructed.
It should also be noted that it is not sufficient for an agency to ALLOW the use of a
pea gravel mix, and then place the burden entirely upon the contractor. Because a pea
gravel concrete mix is more expensive on a materials basis, the result of such practice is
that the winning bid on the job goes to the contractor who uses the least expensive mix
allowed by the project specifications rather than the one which is needed. Subsequent
problems can lead to poor quality, disputes about who is responsible, and claims.
Example 3 Observations on the Flow of Tremie-Placed Concrete
Observations
During the winter of 2002-2003, several drilled shafts were cast at the Auburn
University National Geotechnical Experimentation Site (NGES) at Spring Villa, Alabama
for the purpose of closely observing concrete flow during tremie placement. Concrete
consistent with the Alabama DOT standard mix was used, with a #57 crushed aggregate
(19 mm ( inch) maximum size) and approximately 200 mm (8 inch) slump was placed
using a tremie within a 1.1 m diameter hole and the process filmed using a downhole
camera. The hole was dry so that the concrete could be observed, but the placement was
conducted as if in a wet hole environment. Rebar included longitudinal bars with about
200 mm (8 inches) clear between bars and hoops with approximately 125 mm (5 inches)
clear between hoops.
The concrete was observed to flow up alongside the tremie pipe and produce a rolling
action of the top of the rising concrete column, with the concrete surface rolling from the
center of the vent near the tremie radially towards the perimeter of the shaft. There was
typically about m difference in head between the concrete within the rebar cage and the
concrete in the annular space outside the cage. The concrete could be seen to cascade
over the hoop steel to fill the annular space outside the cage.
Another shaft was constructed similarly and using a similar slump, but with a gravel
aggregate of 12 mm ( inch) maximum size. There was less noticeable rolling of the

Page 6

Dan Brown

Figure 2 Surface of the Shaft with #57 Stone

concrete from the center of the shaft towards the perimeter and the difference in head
between the interior and the annular space was significantly less.
Subsequently, the shafts were exhumed and the surface of the concrete examined (see
Figure 2). The surface texture of the shaft with the pea gravel aggregate appeared less
permeable. There was no trapped debris, since both shafts were constructed using tremie
placement in the dry.
Lesson Learned
Concrete does not flow upward above the tremie as a nice homogeneous mass.
Rather, the surface is a rolling, boiling three dimensional thing which could easily entrap
any debris which resides on top and which can build up a head differential between the
interior and the exterior of the rebar cage.
Example 4 Extreme Rebar Cage
Observations
A bridge on the west coast was designed to utilize a single 4 m diameter shaft to
support an oval shaped column. In order to mate the rebar cage in the shaft to the shape
of the column, a double rebar cage was utilize in which two cages overlap in a sort of fat
figure 8 configuration. As a result, concrete was forced to flow through not one, but two
rebar cages, both of which were had extremely congested cages due primarily to the
closely spaced transverse steel. The shafts were socketed through an overlying sandy
alluvial soil into sandstone. The concrete mix had an aggregate which used crushed stone
with a 25 mm (1 inch) maximum size. The contractor attempted to case through the
overlying sand with a temporary casing; due to aesthetic reasons, permanent casing to the
ground surface was not permitted.
Integrity testing indicated the presence of voids around the perimeter of several shafts.
Access shafts were constructed to examine the problem and pockets of sand were
discovered within the zone where the two cages overlap, as illustrated on Figure 3.
Page 7

Dan Brown

Figure
Figure
3 Defect
2 Surface
in Shaft
of thewith
Shaft
Double
with #57
Rebar
Stone
Cages
Coring was performed from the top of the shaft and showed good concrete of consistent
compressive strength within the center of the shaft but very erratic compressive strengths
around the perimeter.
Lesson Learned
A simple cage with considerations of constructability is needed in order to make
drilled shafts less vulnerable to defects and to promote quality construction. A permanent
casing socketed into the underlying rock could also have reduced the likelihood of
defects in this case because the hole could have been kept dry, although the difficulty
with the rebar and concrete would still be present. The concrete mix in the case of
severely congested rebar should incorporate good flow characteristics with small rounded
gravel aggregates.
Example 5 Concrete Placement in a Large Shaft with Tight Spiral Reinforcement
Observation
During construction of the test shafts for a new bridge in South Carolina, careful
observations were made of the concrete behavior during tremie placement (Camp et al,
2002). The shafts were 2.4 m diameter (8 feet) and up to 50 m deep (160 feet), with a
rebar cage which included a heavy longitudinal reinforcement tied with a spiral
transverse rebar at a 90 mm (3.5 inch) pitch. The concrete used a small crushed stone
maximum aggregate less than 12 mm size ( inch) and 200 mm slump (8 inch). In spite
of the small aggregate and high slump, measurements indicated as much as a 1.4 m
difference in head between the concrete levels on the interior and exterior of the cage.
Lesson Learned
If any adverse performance of the test shafts resulted from the tight spiral
reinforcement in the cage, it was masked by other factors. Nevertheless, the observations
indicate that the potential exists for debris to be trapped on the outside of the cage under
Page 8

Dan Brown

these circumstances, and designers and contractors were alerted to the need to take extra
precaution to maintain a clean slurry during construction of the production shafts.
CONTROL THE STABILITY OF THE HOLE
The successful installation of a drilled shaft is predicated on the ability of the
contractor to maintain a stable hole in order that the foundation can be cast-in-place with
quality materials and workmanship. However, it is not sufficient to only gain stability of
the hole prior to concrete placement. Quality foundation construction requires that the
stability of the hole be maintained at all times in order to preserve the integrity of the
bearing formation and to avoid defects resulting from voids and irregularities in the
overlying strata.
In wet hole construction, it is essential that a positive pressure be maintained against
the sides of the hole at all times. Groundwater should not be allowed to seep from
surrounding soil into the hole, as cave-ins and sidewall sloughing can occur. Even if
sloughing does not occur, the surrounding soil can become loosened and lateral stresses
can be reduced around the shaft and around nearby structures. Problems with ground
subsidence are an all too common occurrence with augered cast-in-place piles (although
not the subject of this paper) in water-bearing sands. An unstable and loosened sidewall
could result in sloughing during concrete placement which would result in defects within
the concrete.
With temporary casing used to provide stability of the hole, it is often attempted to
complete the shaft excavation and place concrete in the dry. However, seepage from the
bearing formation below the casing can result in softening of the bearing materials and
the accumulation of debris on the base of the hole. In shales or fractured rock materials,
there may not be a large quantity of seepage observed, but the unbalanced fluid pressures
and the stress relief provided by the hole may be sufficient to produce softening. If
seepage is observed, it is often preferable to complete the excavation in the wet to avoid
uncontrolled seepage into the hole.
Where temporary casing is sealed into the top of a relatively impervious formation, it
is important that the seal be successful so as to avoid seepage into the hole around the
base of the casing. Such seepage could result in a large cavity forming around the
outside of the casing, and large cavities can result in large concrete overruns and lead to
potential defects in the concrete. As illustrated in Figure 4, a large loss of concrete
volume into a hole around the casing could result in the head of concrete within the
casing becoming less than the head of water on the outside of the casing. If this condition
occurs, there will be flow of water into the casing, potentially displacing or mixing with
the concrete.
Example 6 Cavity Around Casing
Observation
A drilled shaft was installed at a site in Florida by vibro-driving a casing through the
overlying water-bearing sand into the top of limerock. The shaft excavation was
completed within the limerock and concrete placed to near the top of the casing. The
casing was pulled upwards approximately 0.6 m (2 feet) and the concrete level within the
casing dropped to near the bottom of the casing. Although the contractor attempted to

Page 9

Dan Brown

casing
concrete
cavity
(waterfilled)

Figure 4 Loss of concrete into cavity during casing extraction

place additional concrete with a pump, the hole eventually was abandoned. No
significant inflow of sand had been noted during the excavation into the limerock.
Explanation
Although the contractor had used care to avoid causing a cavity around the casing
during excavation, the limerock in this area had natural cavities formed by solution
activity. These cavities tend to be found most often near the top of the formation. The
contractor used good practice when withdrawing the casing only by a small amount so
that additional concrete could be added to maintain the head, but the volume lost was so
large that the head of concrete dropped below that of the surrounding groundwater and
the structural integrity of the shaft was compromised.
Lesson Learned
Because this cavity was a natural occurrence, the design of the shaft needs to
accommodate the possibility of large cavities as a construction consideration. This
condition would be more effectively handled with the use of a permanent casing.
A CLEAN SLURRY HELPS AVOID CONTAMINATION
The use of drilling slurry can be a very effective means to maintain control over the
stability of the hole. However, it is important that the slurry properties be controlled in
order to avoid potential contamination of the bond between the shaft and surrounding soil
or rock, or within the concrete due to excessive suspended sediment.
Several recent comparative tests have demonstrated that polymer slurries can
significantly outperform bentonite in terms of side shearing resistance, as illustrated by
the data on Figure 5 (Brown et al, 2002). This trend appears most notably in granular
soils where there is some fluid loss into the surrounding formation and a bentonite filter
cake is likely to form (Brown, 2002; Majano et al, 1994; Meyers, 1996). Where slurry is
used in a relatively cohesive formation and there is little opportunity for fluid loss,
differences in unit side shear have been insignificant (Camp et al, 2002). Although
current design procedures do not generally delineate values used for design on the basis
of construction technique, it is clear that the construction procedure used can have a
major effect on performance. Where bentonite slurry is used within the bearing
formation, the quality of the bond is enhanced by minimizing the exposure time.

Page 10

Dan Brown

Displacement (mm)

Polymer
Shaft

-10
-20
-30
-40

Bentonite
Shaft

-50
-60
0

6
8
Load (MN)

10

12

14

Figure 5 Comparison of Two Identical Shafts in Cemented

Fine Sand with Different Slurry, Brown et al (2002)
The quality of the shaft also depends on the control of the amount of suspended
sediment in the slurry. Years ago, contractors were routinely permitted to have as much
as 10% sand within bentonite slurries. In recent years, most state DOTs have adopted a
4% criterion for suspended sand within the slurry. However, the key concern is not so
much the amount of sand in suspension, but the amount of sediment which can settle out
during concrete placement. As drilled shafts have become larger diameter and deeper,
the time required to place concrete has increased and thus the time opportunity for
sediment to settle out of suspension has increased. Because of considerations described
previously of the rolling surface and the potential for differential head across the top of
the rising concrete column, and sediment which occurs during concrete placement may
be subject to become included within the concrete. As shafts become larger and deeper,
the allowable sand content within bentonite slurry will likely need to be reduced.
With polymer slurries, there is a tendency for a false sense of security about suspended
sediment, since sand particles tend to settle out quickly. However, silt sized particles
(smaller than the #200 sieve) and fine sands have been observed to remain in suspension
for extended periods and result in contamination atop the concrete column. There have
been numerous cases of projects with polymer slurry where removal of silts and fine
sands from the slurry has been very difficult. On one recent project near the Atlantic
coast, a contractor constructing shafts using a polymer slurry was forced to overpour the
shafts by over 2 m (6 feet) in order to remove concrete contaminated with silt. On
another project at a site in Florida with very fine silty sands, the base of several shafts
were inspected using a downhole camera to evaluate the condition of the base of the shaft
excavation prior to concrete placement. The camera revealed the bottom 0.6 m (2 feet) of
the slurry to be heavily laden with silt in an appearance which resembled that of a gel.
These and many other anecdotal experiences with polymer slurries in silty fine sands
suggest that engineers and contractors should exercise caution in the use of these
materials.

Page 11

Dan Brown

Example 7 You Broke My Shaft!

Observation
A small diameter drilled shaft was constructed for a load test at a site with sand and
clay overlying limerock. Temporary casing was seated into the rock, and the shaft drilled
into the stable limerock formation in the wet. The limerock was stable, but dewatering
the excavation was not attempted due to the potential for seepage. Since the rock was
stable and the overburden was cased, the slurry was not given much attention; water was
used, although some mixing with the native materials occurred. The shaft was completed
and the casing was pulled. No integrity testing was performed.
When the load testing was performed, the shaft failed suddenly at approximately
150% of the design load. Strain gauges within the shaft suggested that very little load
was reaching the lower portions of the limestone socket. Upon excavating the upper 2 m
of the shaft (6 feet), the shaft was observed to have suffered structural failure at a load
corresponding to less than 14 Mpa (2000 psi) over the theoretical cross sectional area of
the shaft (see Figure 5). The shaft was cut off cleanly at about 2 m below grade, a new
top formed, and the load test was successfully carried out to over 2 times the design load
without achieving failure (either geotechnical or structural).

Figure 6 Broken Top of Shaft after Load Test

Lesson Learned
The shaft was constructed with a false sense of security regarding concrete placement
because of the casing and the stable limerock formation. Even though only water was
used during drilling, the properties of the resulting slurry are important because of the
potential effect on the concrete during placement. It appears that the concrete integrity
was compromised within the upper 2 m of the shaft, resulting in structural failure when
loaded to high stresses. The lesson learned from this is example is that any wet pour has
the potential to impact concrete integrity and the drilling fluid must be clean and the
concrete must have good workability.

Page 12

Dan Brown

CONTRACTUAL AND ORGANIZATIONAL FACTORS

In order to achieve quality construction, the work must be organized so that all parties
involved have incentive to achieve quality. This aspect of quality is perhaps the most
difficult to achieve. If only the structural integrity of the concrete between the crosshole
sonic logging tubes is tested and observed, the contractors workers often have little
concern for other aspects of quality, such as preservation of the integrity of the bearing
stratum. Engineers may sometimes focus so much on the optimal arrangement of
reinforcement for bending forces that constructability issues are overlooked.
Quality construction of drilled shafts requires that:
The design engineer should be knowledgeable regarding constructability issues
and would produce a design for which ease of construction is a key element,
The general contractor should appreciate the need for a qualified sub and
provide the resources and support needed to ensure that this critical part of the
work is performed without interruptions,
The drilling subcontractor should be conscientious and genuinely interested in
producing a quality product, and would have well trained workers who are
properly equipped for the job,
The inspector should be well trained and knowledgeable regarding drilled shaft
construction, the critical aspects of the design, and the geologic conditions at
the site,
The project should include provisions for measuring quality using the latest
techniques for inspection, non-destructive testing, load testing, and test
installations where necessary.
The entire project team must work together to achieve quality, with each party
accepting individual responsibility for their own role in the process. Responsible team
members expect and demand quality from other members of the team while at the same
time working cooperatively to resolve difficulties. It is too often the case that the various
parties adopt and adversarial position early in the project in order to position themselves
for the anticipated battles (claims, damages, disputes over problems) with other team
members. These expectations for disputes seem to be self-fulfilling.
There is no magic solution to resolve problems in these areas. Quality construction
requires preparation and persistent attention to details. Just as an army succeeds because
of extensive training for its mission, designers, inspectors, and constructors can succeed
by preparing themselves with extensive training. And a final key ingredient is that
quality be measured in a timely fashion so that adjustments to construction techniques
can be adopted before small problems become major ones.
Example 8 I Didnt Do It, Nobody Saw Me Do It, You Cant Prove a Thing
Observation
A bridge project in the Midwest was constructed using a group of drilled shafts
installed into a rock formation through water and overlying soil using permanent casing.
The general contractor hired a drilled shaft subcontractor to provide services only
consisting of drilling the holes. In order to plan their work efficiently, the general
contractor had the sub drill the entire bent (with up to 12 shafts), after which the concrete
would be placed and the bent completed. Although the bridge was over water, the holes
were expected to be dry due to the permanent casing.

Page 13

Dan Brown

Fortunately for the owner, there was at least post-construction integrity testing
performed using crosshole sonic logging. These logs indicated that several shafts had
zones of very poor quality concrete. Coring indicated that there was weak concrete in
some areas and some zones had washed aggregate in places where sound concrete should
be present. The shafts required extensive grouting and underpinning with micropiles.
Explanation and Lessons Learned
Because the holes had been expected to be dry, the concrete was placed using free-fall
into the hole. Some of the holes were open for weeks between drilling and completion.
In fact, the rock had some small fractures that resulted in seepage into the holes. In some
holes, concrete was dropped through water resulting in defective zones of concrete within
these shafts.
The divided responsibility for completion of the drilled shaft foundations is a very
undesirable arrangement. The work of constructing a drilled shaft cannot be easily
subdivided into drilling the hole and placing the concrete as if they were two independent
operations; once the hole is drilled to the required depth, it is important that the shaft be
completed in a timely fashion with a single point of responsibility for making this
happen.
Although the inspection failed to prevent concrete placement through water, it was
fortunate that the designers had included crosshole testing, which allowed the defects to
be discovered. Many state DOTs require integrity testing only where tremie placement
of concrete is used. The effects on the axial capacity from the seepage and extended
period of open holes is not determined, but the conditions at this site are not conducive to
quality in this respect.
CONCLUSIONS
Like a motorcycle trip across America, the pursuit of quality in drilled shaft
construction is a long and arduous journey. More powerful drilling equipment and better
techniques have lead designers to utilize drilled shaft foundations in larger diameters and
to greater depths than ever before. The challenges for quality in construction have
increased. However, improvements in technology with respect to integrity testing and
load testing have made it possible to measure quality in ways that were never before
possible. Improvements in construction materials, including better concrete and more
sophisticated slurry products, have made it possible to construct quality drilled shafts in
more difficult conditions. Improved quality in drilled shafts is being realized.
But perhaps the most influential component in the process and the most difficult to
control is the human element. All of the examples cited in this paper in which quality
was compromised could have been avoided or corrected by engineers, constructors, and
inspectors who were knowledgeable of their craft and attentive in their work. The
challenge remains to put into place systems of training (for all members of the team) and
jobsite control to encourage and emphasize quality in construction.
ACKNOWLEDGEMENTS
The author would like to acknowledge the contributions of Mike Muchard of Applied
Foundation Testing, Jack Hayes of Loadtest, Inc., and Larry Olson of Olson Engineering.
The author would especially like to acknowledge the many years of mentoring and wise
council of the late Mike ONeill, to whom the author will be forever indebted.

Page 14

Dan Brown

REFERENCES
Brown, D. A. (2002). The Effect of Construction on Axial Capacity of Drilled
Foundations in Piedmont Soils, J. of Geotechnical and Geoenvironmental
Engineering, 128(12), pp967-973.
Brown, D., Muchard, M., and Khouri, B. (2002). The Effect of Drilling Fluid on
Axial Capacity, Cape Fear River, NC. Proceedings Deep Foundation Institute
Annual Conference, San Diego, CA, Oct.
Camp, W.M., Brown, D.A., and Mayne, P.W. (2002). Construction Method Effects
on Axial Drilled Shaft Performance Geotechnical Special Publication No. 116,
ASCE, pp. 193-208.
Majano, R.E., ONeill, M.W., and Hassan, K.M. (1994). Perimeter Load Transfer
in Model Drilled Shafts Formed Under Slurry, Journal of Geotechnical
Engineering, ASCE, Vol. 120, No. 12, pp. 2136-2154.
Meyers, B. (1996). A Comparison of Two Shafts: Between Polymer and Bentonite
Slurry Construction and Between Conventional and Osterberg Cell Load
Testing, Paper Presented at the Southwest Regional FHWA Geotechnical Conf.,
Little Rock, AR, April.
ONeill, M.W. and L.C. Reese (1999) Drilled Shafts: Construction Procedures and
Design Methods, Technical Manual Prepared for Federal Highway
Administration, 758 p.
Pirsig, R. M. (1974) Zen and the art of motorcycle maintenance: an inquiry into
values, Morrow, New York, 412 p.

Page 15

Dan Brown