Skip to Content
Uncategorized

Detecting "Rail Cancer"

November 1, 1997

The progressive growth of hidden cracks within railroad tracks, known as “rail cancer,” is a major safety hazard in the railroad industry. Crosswise cracks that begin with tiny flaws inside the steel rails can grow into surface cracks and eventually cause the rails to break. In the effort to prevent derailments, service cars equipped with special sensors travel the nation’s railroads, seeking out signs of rail cancer and other defects. But because the sensors require direct contact with the rails, the cars have to creep along at speeds averaging less than 15 miles per hour, forcing commercial train traffic in the area to shut down.

Finding a more efficient, cost-effective method of track maintenance and inspection is a high priority for the railroad industry, according to Jim Lundgren, head of the American Association of Railroads’ (AAR) Technology Scanning Program for identifying and addressing critical areas of railroad-related research. “We need the ability to do more frequent and thorough inspections that do not interfere with revenue-generating traffic,” he says. Now an MIT team led by civil and environmental engineering professor Shi-Chang Wooh is investigating ways of detecting potentially lethal cracks at high speeds while the nation’s trains are running.

Early rail-defect detection techniques involved passing electric currents through the tracks and looking for changes in direction or drops in voltage that would indicate the presence of a crack. For the last 30 years, how ever, operators have relied largely on ultrasound, sending an ultrasonic signal directly into the rail and measuring the time it takes to bounce back. A crack prevents the signal from reaching the base of the rail, so it bounces back more quickly.

The goal of Wooh’s team is to use ultrasonic detection in a way that does not require contact between the sensors and the tracks. An expert in ultrasonics and materials evaluation, Wooh proposes a system that would rely on the Doppler effect-the shift in the frequency of sound waves that occurs as a moving object, such as a whistling train or wailing ambulance, approaches and then recedes from a stationary object. As the distance between the two objects narrows, the sound waves pile up and the frequency and pitch become higher; when the distance increases, the waves spread out again and the frequency and pitch become lower.

Similarly, a high-speed monitoring system that continuously directs an ultra-sonic beam downward at the rail as it passes by would register a slight Doppler shift in the frequency of the response signal over a defect, such as a surface crack or rail break, according to Wooh.

Wooh suggests that once a detector identified a problem area, it could broadcast its location to a Global Positioning System (GPS) satellite. A ground-based Graphical Information System could then obtain the area’s position from the GPS satellite and provide maintenance personnel with a map showing the range of track with problems and the closest rail service station.

With funding from the AAR’s Technology Scanning Program, Wooh and his team tested out this hypothesis in the laboratory earlier this year. “We limited ourselves to [testing for] cracks on the surface,” Wooh explains. To simulate the motion of detectors along flawed tracks, the team beamed an acoustical signal from a piezoelectric transducer-a vibrating device that converts electrical power to acoustic power and vice versa-toward the rim of a spinning metal wheel. The rim of the wheel, which represented a moving rail, was marred by a notch, which represented a crack in the rail. A second transducer picked up the signal after it bounced off the rim, and converted it into digital data. To test different train speeds, the researchers spun the wheel at various rates.

Using software to analyze signal responses collected as the wheel spun, Wooh’s team determined that the sound-wave frequency obtained above the notch was higher than that found over the rest of the wheel. According to Wooh, similar detectors mounted on a train running at 60 miles per hour could send and receive ultrasound signals and a signal-processing chip could evaluate the responses immediately.

While this laboratory experiment demonstrated the viability of this approach for detecting surface cracks at high speeds, a much more challenging problem remains: detecting cracks hidden inside the rails before they reach the surface.

The key difficulty in applying the same method to interior cracks is that sound transmitted through air will bounce off the rails but not penetrate them. So Wooh is experimenting with techniques that will generate ultrasound signals from inside the rail rather than from above. The most promising technique, he believes, is laser-based ultrasound. A device mounted on the train will emit a laser beam that creates a warm spot on the track surface. As the warm spot expands against resistant, colder metal, the resulting tension will produce an acoustic stress wave within the track. When the sound waves generated in the rail pass through a transverse interior crack, they will pile up, causing an increase in wave frequency.

Wooh points out that industries from aerospace to piping have used laser-based ultrasound to detect flawed materials. However, these techniques have not been used for high-speed monitoring and have relied on intermittent rather than continuous laser beams, which would provide the most thorough coverage. With limited funding from the AAR, Wooh’s team is now designing experiments to test the use of both intermittent and continuous laser-based ultrasound to detect and monitor subsurface cracks. According to AAR principal investigator Richard Reiff, if laboratory tests prove successful and Wooh can develop a promising prototype for rail-defect detection, he can test it on the Gauntlet, a stretch of track containing more than 50 defects at the AAR’s Transportation Technology Center in Pueblo, Colo.

Wooh argues that the investment in new hardware required for high-speed detection will be more than offset by the cost savings it entails. In an era when freight and passenger railroad traffic is growing, notes Federal Railroad Administration Track Safety Specialist Al MacDowell, “the window to get out on the tracks for inspection is getting smaller and smaller.”

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

Google’s Gemini is now in everything. Here’s how you can try it out.

Gmail, Docs, and more will now come with Gemini baked in. But Europeans will have to wait before they can download the app.

This baby with a head camera helped teach an AI how kids learn language

A neural network trained on the experiences of a single young child managed to learn one of the core components of language: how to match words to the objects they represent.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.