Printable Version

New Angle on Pipe Inspection

Approximately one-third of U.S. natural gas pipelines cannot be inspected by traditional methods. SwRI researchers have developed a technology to inspect these pipelines

By Gary Burkhardt

Pipelines are used extensively to carry natural gas to destinations throughout the world. Large high-pressure transmission lines may extend hundreds of miles cross country while smaller, lower-pressure distribution lines are used to deliver gas to homes and businesses. In the United States, there are more than 1.1 million miles of natural gas pipelines.

Because pipelines are typically buried underground, they are in contact with the soil and subject to corrosion, where the steel pipe wall oxidizes, effectively reducing wall thickness. Although it’s less common, corrosion also can occur on the inside surface of the pipe. Extensive efforts are made to mitigate corrosion; these include the addition of protective coatings to the pipe outside diameter and the use of cathodic protection systems — an electrical method that retards corrosion development on metal structures.

Nevertheless, corrosion still occurs, which reduces the strength of the pipe. If corrosion goes undetected and becomes severe, the pipe can leak and, in rare cases, fail catastrophically. In August 2000, corrosion resulted in failure of a pipeline near Carlsbad, N.M., where 12 people were killed.

Add to this the factor of aging infrastructure, which is of concern to both industry and the government. About 60 percent of U.S. gas transmission pipelines are 40 to 50 years old and in many cases have outlived their design life.


Gary Burkhardt, a staff scientist in the Mechanical Engineering Division, led a team in developing the RFEC inspection system. His expertise includes the development of electromagnetic techniques and sensors for nondestructive evaluation (NDE) of components and structures. Burkhardt holds 10 U.S. patents and has authored
more than 60 publications in the NDE field.


To replace hundreds of thousands of miles of gas transmission pipelines would be both cost-prohibitive and impractical from the standpoint of supply disruptions amid the nation’s growing energy needs. Industry is looking at ways to inspect those pipeline systems, to predict remaining life and to do maintenance where appropriate.

Pipelines are typically inspected for corrosion using a device called a “pig” that is inserted into the pipeline and carried along by the gas flow. The most common type of inspection pig uses a sensing technology called magnetic flux leakage (MFL), where strong magnets magnetize the pipe wall and sensors detect changes in the magnetization caused by corrosion. However, many lines, particularly those in the small-diameter range, have internal restrictions and/or low pressure or flow rates. Therefore, conventional MFL pigs cannot be used because the drag forces from the magnets may be so large that the gas flow will not adequately propel the pig, and the large magnets needed cannot readily be transitioned through restricted areas. About one-third of the pipelines in the United States, or approximately 300,000 miles of pipelines, currently are not able to be inspected.

To address the need for inspecting “unpiggable” pipelines, a tool was developed that couples a remote-field eddy current (RFEC) inspection system, developed by engineers at Southwest Research Institute (SwRI), with the Explorer II robotic transport tool developed by the National Robotics Engineering Consortium at Carnegie Mellon University.

RFEC can detect and characterize pipe wall loss from corrosion defects. Unlike conventional MFL, RFEC does not require large magnets, and the RFEC system could be designed to accommodate the robot’s capability to expand for inspection of unrestricted areas, yet retract to a smaller diameter to pass through obstructions.

Another characteristic of the RFEC system is that the drag forces are small, thus allowing the robot to be driven by battery-powered electric motors so that it does not require propulsion by the gas flow and can be used in pipelines with low pressure and flow rates.


These photos depict typical corrosion damage in buried pipelines of the type targeted by the RFEC inspection system. The corrosion product has been removed to reveal the underlying material loss.


The combined Explorer II/RFEC system consists of 11 modules that are linked to form a self-propelled device that can travel untethered through a pipe. It is launched into the pipe through a “hot tap,” a connection that is made to the pipe without taking the pipeline out of service. The system can inspect 6-inch to 8-inch diameter pipelines containing tight elbows and tee joints, and it can be used while the pipeline is in operation.

Remote-field Eddy Current Technology

The RFEC inspection approach was developed in the 1980s for inspecting pipe and tubing. RFEC inspection technology is based on the use of an excitation coil placed in the pipe with its axis oriented along the pipe axis. The coil is driven with alternating current, and the associated alternating magnetic field produces the flow of electrical currents, or eddy currents, in the circumferential direction in the pipe wall in much the same way as the primary winding of a transformer induces current flow in the secondary winding. The excitation frequency is low enough (typically on the order of 200 Hz or less) that the generated magnetic field will penetrate through the pipe wall and not be severely attenuated by the electromagnetic skin effect. Sensors are placed adjacent to the pipe wall at a distance of approximately 2.5 to 3 pipe diameters away from the exciter. At this “remote-field” location, the magnetic field from the excitation coil is very small (about 4 to 5 orders of magnitude less than immediately adjacent to the coil), and therefore direct coupling from it into the sensors is minimal. This allows the sensors to have greater sensitivity to the magnetic field from the eddy currents induced into the pipe wall by the excitation coil. Since components of this field have penetrated through the pipe wall from the outside diameter (OD) to the ID, the detected signal is sensitive to material-loss defects such as corrosion.

In addition to detecting corrosion, RFEC can also characterize its extent. The electrical phase of the RFEC signal, which is the phase of the sensed signal relative to the excitation coil current, varies with the wall thickness of the pipe. Therefore, the wall loss from corrosion (depth) can be determined by measuring the RFEC signal phase. The length of a corrosion pit can be determined from the length of the signal in the axial direction (the direction of travel of the RFEC inspection tool). The width of the corrosion pit can be determined from the circumferential extent of the signal measured by multiple RFEC sensors spaced circumferentially around the pipe.


The RFEC inspection technology, which can detect and characterize pipeline defects such as corrosion, is fitted into a robotic transport tool. The National Robotics and Engineering Consortium at Carnegie Mellon University developed Explorer II to inspect 6- to 8-inch-diameter pipelines that are not inspectable with current technology.


Initial RFEC Developments

The current RFEC project consisted of three phases. The first phase was a small project directed at developing and demonstrating concepts for an RFEC excitation coil that could retract to pass through obstructions in the pipe and then expand to perform an inspection. At this point, the approach was not tied to a particular transport vehicle such as Explorer; it was only intended to demonstrate a unique coil with foldable segments. As part of this project, a laboratory breadboard RFEC system with a foldable coil and single sensor was developed to test feasibility. SwRI, along with other organizations with competing inspection technologies, participated in a “blind” demonstration at Battelle Memorial Institute where tests were performed on pipes containing flaws that were hidden from view. The SwRI inspection system successfully detected and characterized the flaws.

These good results led to the second phase, which involved the development and demonstration of a design concept for an RFEC system that would meet the requirements for the Explorer II robot. Among other constraints, the entire RFEC system had to fit within two modules measuring 4 inches in diameter and 5 inches long, with a maximum total weight of less than 10 pounds and a maximum power draw of 25 watts. The team overcame many difficult challenges, both electronic and mechanical, to meet these and other demanding requirements. Within an eight- month period, SwRI developed a design concept, participated in preliminary and final design reviews, fabricated a laboratory breadboard demonstration system and performed another blind test demonstration. Because of the size constraints imposed by the robot, the foldable segmented coil developed in Phase 1 could not be used; however, test results showed that project goals could be met with a small fixed exciter coil.

The blind tests using the breadboard system were performed on three 8-inch-diameter, 0.188-inch-wall pipes, with the goal of accurately determining the length, width and depth of the defects. Again, other competing organizations also participated in the testing. The pipes contained 32 simulated external corrosion defects having complex geometries — all defects were hidden from view and defect characteristics were not revealed until after test results were submitted. Overall, good agreement between RFEC predicted and actual values were obtained. Based on our performance on these tests, SwRI was selected for the third phase of the project to develop the final inspection system that would be integrated with the Explorer II robot.


The schematic figure shows the SwRI-developed RFEC modules. The system consists of two modules, one containing an array of sensors (left) and an excitation coil (right) with robot centering module in-between. The sensor may expand to a 6- or 8-inch diameter for different pipeline sizes (bottom view), and it collapses to 4 inches for traversing bends and for launching (inset).


RFEC Inspection System

The sensor and exciter modules contain all of the electronic systems for signal acquisition, processing and storage; for the communications interface to Explorer II; and for generation of the exciter drive signal. The signals from the sensors are amplified, multiplexed and digitized. Signal processing is performed by implementing a lock-in amplifier (phase-sensitive detector) function on the digitized signals for all 48 channels using a digital signal processor (DSP), a microchip designed for high-speed scientific calculations. The DSP also controls onboard storage of the processed data in flash memory and communication with the Explorer robot. Use of the DSP allows the electronic circuitry to be greatly simplified so that the circuitry can fit in the small space available.

The RFEC system is controlled by an operator using the robot’s wireless communications link and its internal communications bus. The operator can control deployment of the sensor arms to the desired pipe diameter, as well as RFEC operational parameters such as excitation frequency, exciter coil drive level and signal amplifier gain. RFEC inspection signals are stored on internal memory and are also transferred to the operator’s computer in real time where they can be viewed on the computer display. Detailed analysis of the signals is usually performed on stored signals after the inspection is completed.

Conclusion

The RFEC system allows inspection of many natural gas pipelines that cannot be inspected with conventional technology, thus improving the reliability and safety of pipeline infrastructure in both the United States, as well as the world. Plans are under way for commercialization of the technology so that routine pipeline inspections can be performed on a commercial basis by licensed contractors.

Acknowledgments
The author would like to acknowledge Senior Research Engineer Albert Parvin, Principal Engineer Ronald Peterson, Senior Research Scientists Todd Goyen and Richard Tennis, Staff Technicians Gary Asher and David Jones, Senior Designer Rolf Glauser and SwRI Technical Advisor Al Crouch for their contributions to the design, fabrication, and testing of the RFEC system. The author is grateful to James Merritt and Robert Smith of the U.S. Department of Transportation, Pipeline and Hazardous Materials Safety Administration, for funding the SwRI project. Appreciation is also expressed to Dr. George Vradis and Daphne D’zurko of Northeast Gas Association, as well as Dr. Hagen Schempf of the National Robotics Engineering Consortium at Carnegie Mellon University, and the U.S. Department of Energy, National Energy Technology Laboratory for development of the Explorer II robot and support of system integration and testing.

 

Published in the Winter 2007 issue of Technology Today®, published by Southwest Research Institute. For more information, contact Joe Fohn.

Winter 2007 Technology Today
SwRI Publications SwRI Home