The Federal Highway Administration (FHWA) often develops collaborative partnerships with professional and trade organizations to serve the highway community's best interests. Such is the case with FHWA's long-standing relationship with ADSC: The International Association of Foundation Drilling.

Through this partnership, drilled shaft technology has advanced by leaps and bounds. Drilled shaft use in highways has doubled over the past decade. Departments of transportation and other public and private sector owners are reaping the benefits.

Drilled shafts often make superlative bridge foundations. They can carry huge vertical loads. They effectively carry large lateral and seismic loads. For many soil conditions, they are easier to install than driven piles and do not create ground vibrations. They can be readily installed offshore and can be made highly resistant to scour. And they are economical.

One perceived disadvantage of drilled shafts is that because the shaft is built in-place (as opposed to formed in a casting yard as with piles), it is difficult to monitor the shaft's constructed integrity. But as the level of comfort in using drilled shafts as foundations has increased over the years, so has the sophistication in testing their properties.

"In the early 1990s, states started routinely performing nondestructive testing on their drilled shafts and noticed some anomalies such as potential voids or discontinuities in the concrete," explains Albert DiMillio, head of the FHWA Geotechnical Research Team.

"As testing has become more commonplace," says DiMillio, "databases of observations have grown, leading to more questions about their performance. That's one of the main areas on which we're focusing our research efforts."

A steel reinforcing cage is lowered into the hole, and the concrete is placed. If the hole is filled with slurry, the concrete is placed by using a "tremie" pipe or by pumping from the bottom of the hole upward, thereby displacing the slurry.

In an open or cased hole, free-fall of concrete is the preferred placement method. If the hole is cased, the casing is then pulled out before the concrete hardens.

Despite care and skill in construction, voids or gaps in the concrete can occur for many reasons. For example, even when using slurry, soil can cave into the concrete. Sediment can be caught in the slurry as the concrete is placed. The reinforcing cage can be out-of-plumb. When extracting the casing, the concrete can adhere to the casing and cause gaps.

In research led by Dr. Magued Iskander, licensed professional engineer and professor at Polytechnic University, artificial anomalies were installed in six full-scale drilled shafts installed at the National Geotechnical Experimentation Site (NGES) located at the University of Massachusetts at Amherst.

Items such as 1- to 10-gallon (3.78- to 37.8-liter) plastic pails, foam insulation, 9- to 13-inch- (230- to 330-millimeter-) diameter cardboard construction tubes, and 4-inch (102-millimeter) flexible drain pipe were secured to the reinforcing cage of 3-foot- (0.91-meter-) diameter drilled shafts. FHWA provided funding for instrumenting the shafts and helped facilitate the work. Seven testing organizations, including two universities, tried their hand at detecting these anomalies using various nondestructive testing techniques in a Class-A prediction symposium.

"Although the data are still being analyzed, preliminary results indicate that most large anomalies were located by all detectors," says Iskander, "but there were also a number of 'false positives' showing an anomaly where none actually existed.

Similar research conducted by Dr. Michael O'Neill and others at the University of Houston, also funded by FHWA and ADSC, indicates that the largest void that would go undetected by a well-conceived testing program is one that occupies about 15 percent of the gross cross-sectional area of the shaft. For a 3-foot-diameter shaft, this equates to a void the size of a 1-gallon paint can. The results were consistent with earlier studies conducted for and by FHWA.

How Do Anomalies Impact Performance?

Finding anomalies is important, but the compelling question is how these anomalies affect the vertical and lateral load-carrying capacity of the drilled shaft. The two current FHWA/ADSC research projects mentioned above are exploring this issue.

Research at the Polytechnic University involves axial load tests on the shafts used for the prediction symposium. The research is ongoing, so data are not yet available.

The University of Houston load-testing program is further along and results are providing good insight into the structural performance of drilled shafts with anomalies. At the NGES at the University of Houston, six full-scale drilled shafts, each three feet in diameter, were constructed. Five shafts had pre-installed minor anomalies, while a sixth shaft that had no anomalies served as a reference. The shafts were loaded laterally until they approached structural failure.

Initial results showed that the ultimate lateral load capacity was reduced by a maximum of 10 percent. However, after the test shafts were exhumed, inspected, and measured, researchers factored in the reduction in shear strength caused by shaft geometry.

"We found that the maximum reduction in capacity was actually 23 percent," says O'Neill. "But we still weren't satisfied that we fully understood the whole picture. For example, we believed that the anomaly's position within the shaft may also impact its load-carrying capacity."

O'Neill's team tested 19 scaled shafts to simulate nine or 10 anomalies that can be caused by normal construction techniques.

"We determined that an anomaly outside of the reinforcing cage has much less impact than one inside the cage," says O'Neill.

Two full-scale tests were conducted at the Structures Laboratory at FHWA's Turner-Fairbank Highway Research Center on specimens that were identical to the small-scale specimens to evaluate further the effects of scale.

"At the outset, we thought that the lab tests would not be representative because the lab shaft is not confined by the soil," explains O'Neill. "It turns out that the soil-confining pressure was not important after all when compared with the confinement afforded by the transverse steel in the reinforcing cage."

From the laboratory testing, O'Neill's team could then develop and calibrate a computer model to evaluate other anomaly sizes, geometries, and combinations.

The computer model showed that for the unlikely case of three simultaneous anomalies occurring in a critical section of a drilled shaft, the pure axial capacity would be reduced by approximately 33 percent and the pure flexural capacity would be reduced by 47 percent from the theoretical values for a perfect section.

Intuitively, one expects that the occurrence of three simultaneous anomalies at the critical section along the shaft has a very small probability. Therefore, the next step in the research will be to apply probabilistic studies to refine the deterministic capacity-reduction factors previously calculated.

"The results of our research," says FHWA's DiMillio, "will give designers, owners, and contractors confidence that they are building safe economical drilled shafts that take into consideration the imperfect state of even the most carefully constructed shafts."

Sybil Hatch is a technical and marketing communications consultant, specializing in the engineering and construction industry. She has a bachelor's degree in civil engineering and a master's degree in geotechnical engineering, both from Virginia Tech. She is a registered professional engineer in California and has been practicing civil and geoenvironmental engineering for 14 years.