Better Load Ratings Through Nondestructive Evaluation

In a report to Congress in May 1997, the Federal Highway Administration (FHWA) reported that more than 182,000 bridges in the United States are recorded in the National Bridge Inventory as either structurally deficient or functionally obsolete.1 And in the Transportation Equity Act for the 21st Century (TEA-21), Congress, recognizing the safety, mobility, national defense, and economic implications of this problem, authorized spending $20.4 billion in fiscal years 1998 through 2003 to replace or rehabilitate eligible bridges on any public road.

The main factor contributing to a bridge being classified as structurally deficient is a low load rating. A deficient bridge is not necessarily unsafe. That is, not every structurally deficient bridge needs to be replaced. "With proper load posting and enforcement, most structurally deficient bridges can continue to serve traffic safety when restricted to the posted maximum loads. ... About 114,332 bridges (19.6 percent) nationwide are or should be load-posted."1 Therefore, the proper evaluation of each bridge's load-carrying capability is essential.

The vast majority of bridges are load-rated using theoretical calculations rather than actual testing; therefore, the real load capacity is not known. Theoretical calculations tend to be very conservative, underestimating the capacity of a structure. The only way to determine the actual capacity is to perform a load test on the bridge.

This type of test is time-consuming and expensive. As a result, very few load tests are conducted, and this lack of actual testing may lead to an inaccurate count of structurally deficient bridges.

This article describes the test and evaluation of two state-of-the-art prototype nondestructive evaluation systems that, in comparison with theoretical calculations, provide a much more accurate measure of a bridge's load-carrying capacity. The testing of these new technologies was conducted this summer in cooperation with the Alaska Department of Transportation and Public Facilities in Juneau, Alaska. These prototypes were developed under the FHWA Research and Development Program in Nondestructive Evaluation.

Load rating can be conducted by measuring various indicators of the bridge reaction, such as the strain (stress) in beams or the mid-span deflection when a truck with a known load is placed at a critical location.

Installation of sensors and instruments to measure these parameters can be time-consuming and intrusive. In the case of strain gauges, surface preparation such as paint removal is usually required. Deflection measurements require a fixed reference point under each measurement location. In either case, access under the main bridge members is required. Because many bridges are over roadways, access under the structure may require the interruption of traffic. Also, most conventional data acquisition systems require wires physically connected to sensors. The need to run wires from a central location to each sensor is time-consuming and costly.

An effort has been made by FHWA to develop the necessary tools that will allow for the quick and efficient load rating of a bridge. Instrumentation that can conveniently measure both strain Calibrated load vehicles on bridge.

and deflection are needed to accomplish this goal.

A wireless data acquisition system has been developed for the quick, efficient measurement of strain. The wireless system can be set up in a matter of seconds at a particular measurement location and does not require wires to connect the sensors to a central data acquisition system. The system can operate with conventional strain gauges that are bonded or welded to a structure. Alternatively, these systems can use a clamp-on strain gauge that can be quickly applied without surface preparation. Data can be monitored in a near real-time fashion while testing is in progress.

Deflection measurements are also possible with the wireless data acquisition system, but the wireless system has the same requirement for reference locations under the bridge as does conventional instrumentation. Consequently, the method used to easily measure deflections remotely is a laser radar system that precisely measures the distance to a remote target.

The laser radar system is placed at one location under the bridge, and a laser beam is scanned over multiple points on the bridge to remotely measure deflections. Neither surface preparation nor the mounting of targets on the bridge is required. The main advantages of this type of instrumentation are the quick setup time, the large number of measurement locations possible, and the elimination of the need for scaffolding or access below the structure. When used together, the wireless data acquisition system and the laser radar system have the potential to rapidly measure the load capacity of bridges, and that could result in a more precise estimate of the number of structurally deficient bridges.

Alaska Field Test
The Alaska Department of Transportation and Public Facilities (ADOT&PF) has recently begun testing bridges in an effort to better determine live-load capacity and load rating. Given the newness of their program, ADOT&PF wanted some means to evaluate their instrumentation setup and test results. This provided an opportunity to test some of FHWA's nondestructive evaluation (NDE) technology under actual field conditions and at the same time evaluate ADOT&PF measurements.

A bridge over Sheep Creek just outside of Juneau, Alaska, was chosen for this test. This structure is a two-span, simply supported bridge constructed of Bulb-Tee prestressed concrete girders. Each span consists of five girders that are approximately 18 meters in length. Because the structure passes directly over water, access to the underside of the bridge for instrumentation purposes is limited.

Deflection and strain measurements were made on one span of the bridge after it was subjected to a calibrated load. Two trucks weighing approximately 25 metric tons each were used as the load vehicles. These vehicles were strategically placed during the load test to determine the load distribution for the bridge under several different loading conditions.

The deflections of the bridge girders were measured using the laser system. To make a measurement, the system sends a laser signal from the instrument to an object. This signal reflects off the object and returns to the instrument. It is not necessary for the object to be highly reflective to make a valid distance measurement. Typical structural materials, such as steel (both painted and unpainted) and concrete, produce good results. A mechanical scanner is incorporated to direct the laser to multiple measurement points from a single instrument scanner location.

The laser system is capable of sub-millimeter measurements over a maximum range of about 30 meters. This instrument differs from other remote, distance measurement devices in that it can resolve sub-millimeter measurements without requiring a special retroreflective target.

The laser system can make measurements only along its line of sight. To position the instrument for measurements on all five girders of one span on the Sheep Creek Bridge, it was necessary to place the laser system in the water under the bridge. To accomplish this, a small platform was used to support the laser scanner.

This platform placed the laser system at the mid-span of the bridge (approximately along the centerline), allowing the laser to measure locations on all five girders over the majority of their 18-meter length. Measurements were taken at 11 points on each of the five girders — a total of 55 discrete deflection measurements on the span under test. Setup of the laser scanning pattern took approximately 20 minutes, and obtaining the data during one scan of all 55 measurement locations took approximately 13 minutes.

Deflection sensor and wireless transceiver.

The bridge was also instrumented with FHWA's wireless data acquisition system to measure strain and deflection. The wireless data acquisition system is designed to be a low-cost, portable instrument in which wireless transceivers communicate with a base station using a 902- to 928-MHz direct sequence spread-spectrum radio link. This radio system is non-licensed and relatively immune to noise. One base station can communicate with up to eight transceiver modules, each having one sensor. The transceivers were designed to work with strain gauges, tilt sensors, accelerometers, and any transducer compatible with the input/output requirements of the transceiver. The data acquisition system is set up, operated, and downloaded with a laptop computer.

Three deflection measurements and one strain measurement were made on the girders at the mid-span of the structure. The installation of these sensors and the configuration of the wireless data acquisition system were completed in approximately one hour.

Conclusions
The results obtained through this field test illustrate several advantages of the laser system over traditional instrumentation for load rating. First, multiple locations on the structure can be measured from a single instrument position. A greater number of measurement points are possible with the laser system than would ever be possible with conventional instrumentation. Second, because no targets are required for the laser system, access to difficult-to-reach areas under the bridge is not necessary. Third, the system is quick to set up, and results are produced immediately after a measurement.

The test also illustrated advantages of the wireless data aquisition system. Because it was not necessary to install a hard-wired data acquisition system on the bridge, the installation and configuration of the system was very fast. This field test provided much useful information on the operation of the new NDE instruments on actual bridge structures. The results of this type of testing will eventually determine the most effective way to use these technologies to aid the bridge community.

These technologies have the potential to provide more accurate load ratings, which will reduce the number of bridges considered to be deficient and will save many commercial vehicle operators a great deal of time and money.

The authors thank ADOT&PF for providing a test bridge and load vehicles. The field test went very well, largely due to the help and cooperation of the staff of ADOT&PF. In particular, the authors acknowledge the contributions of chief bridge engineer Steve Bradford, Loren Gehring, Elmer Marx, Keith Carlson, and Gary Scarbrough. Another key individual was Drew Sielbach, structural engineer in FHWA's Alaska Division, who coordinated the field test.

The Status of the Nation's Highway Bridges: Highway Bridge Replacement and Rehabilitation Program and National Bridge Inventory, Thirteenth Report to the United States Congress, Federal Highway Administration, Washington, D.C., May 1997.

Glenn Washer is a research structural engineer in the Special Projects and Engineering Division of FHWA's Office of Engineering Research and Development (R&D) at the Turner-Fairbank Highway Research Center in McLean, Va. He is the special projects manager for FHWA's Nondestructive Evaluation R&D Program. He has a master's degree from the University of Maryland, and he is a licensed professional engineer.

Dr. Paul Fuchs is a consultant to the Special Projects and Engineering Division. He works on the design, development, and testing of instrumentation related to the nondestructive testing of bridges and other civil structures. He received his doctorate in electrical engineering from West Virginia University.