![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image1.jpg)
Detecting flaws is a challenge even in the best of times. Interpreting complex patterns on NDT instrument displays can be difficult, and a part's size, geometry and material makeup can add to the problem. Often, it takes a certified inspector to weed out the good from the bad.
For example, an East Coast railcar wheel inspector had a problem. The inspector needed to inspect the wheels to meet required industry standards. The product is a wheel about 10 feet in circumference and has a 4-inch wide "tread" area inspected for defects from the surface of the tread down to 3 inches deep. It is a solid piece of steel with some geometry to it and it will undergo intense pressures that can turn a defect such as a crack along the grain boundary into a catastrophic failure.
![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image2.jpg)
In the past, an operator would paint the outside of the wheel with a gel couplant and manually scan the part's circumference with a handheld, digital ultrasonic flaw detector that uses a 1/2-inch diameter, single-element transducer. The operator takes 1/2-inch readings across a 1-inch swath and then moves to the next swath until the entire 4-inch by 10-foot circumference is inspected.
But, their solution had problems. The sound waves generated by the flaw detector were capable of detecting the internal flaws, but it was not easy for the operator to distinguish between flaws and occasional geometric features of the wheel. The test could be influenced by the operator and was slow.
A better solution was a phased-array ultrasonic flaw detector. A handheld detector featuring a single transducer was replaced by a semi-automated system using an electronically controlled array containing 128 transducers. The phased-array inspection fixture was custom-conformed to the shape of the wheel and the array could inspect a 4-inch wide area.
Instead of using a gel-type couplant, a water bath was used. The transducer with a magnetic inspection fixture was attached to the outer surface of the wheel and water was pumped into the fixture to couple the ultrasonic energy into the wheel. While the phased-array system remained still, the wheel was rotated by automated machine-handling equipment. As it rotated, the array scanned the wheel, capturing the location and condition of inspection points. The whole process can be completed in about 30 seconds.
In the manual application, the operator slides the array and fixture around the wheel, scanning as he goes. "Even using manual scanning methods we could inspect the whole part in a minute or two," says Dane E. Hackenberger, leader of global applications for GE Inspection Technologies (Lewistown, PA). "We did the whole part in about 45 seconds. In that time, we can scan the entire circumference of the wheel and inspect it with about 100,000 data points."
On the display, each of these inspection points is assigned a color. The colored points are arranged in a 4- by 10-foot pattern that matches their origin on the tread of the wheel and creates a "map" of the part that is easier to interpret, he says.
The improved cycle times and ease-of-use that this inspector enjoyed are two reasons that phased-array systems are gaining in popularity.
![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image3.jpg)
Gaining widespread usage
This wheel manufacturer is not alone in migrating its inspection to phased-array ultrasonics. "Phased-array ultrasonics is gaining more widespread use," says Sharon I. Vukelich, senior research engineer at the University of Dayton Research Institute (Dayton, OH), and current president of the American Society for Nondestructive Testing. "The reason for this can be attributed to the fact that phased-array ultrasonic inspection results in high-inspection speeds through electronic beam scanning, allows variable focusing of the ultrasonic beam within a component using a single probe and creates the capability of steering the ultrasonic beam for reaching limited-access regions within the component."Price is a factor, however. "A drawback to this technology is the cost of the phased array probes versus conventional ultrasonics probes," she says.
Dan Carnevale, president of Danatronics Corp. (Danvers, MA) agrees that cost is a factor. Phased-array ultrasonic systems can cost $60,000, while portable digital ultrasonic flaw detectors can cost $6,000. That can scare users off. Lack of knowledge about the technology is also a reason for phased-array systems' slow growth.
"I think that its use is in the early stages," he says. "A lot of people think of it as black magic. I liken the example to what Panametrics went through in the mid-1980s, when they introduced the world's first digital flaw detector. People were saying, ‘Why do you have all these funny symbols on the key board?' Or, on the Krautkramer USD 10, ‘Why do you have a tortoise and a hare?' It took a lot of hand holding and convincing of end users and I would say that phase array is in that stage right now."
According to a recent study, that attitude maybe changing. Frost & Sullivan, a market analysis firm, released a study in February that concluded that phase-array ultrasonics is being used more often by manufacturers and that usage is set to grow and cause a "sea change" in the nondestructive testing (NDT) industry.
![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image4.jpg)
Operator involvement
Operators are drawn to phased-array ultrasonic's speed and flexibility as well as new electronics such as improved display technology that helps operators find flaws, the experts say. It can be programmed to perform a variety of tasks, creating images of multiple angles and depths. Using these views, an operator has a greater chance of detecting flaws, some of which maybe hidden and hard to identify; some of which are predictable, while others random.This puts a lot of burden on the skilled operator. "One of the major contributors to the reliability of any nondestructive testing method is the human factor," according to Phased Array Technical Guidelines published by R/D Tech, a subsidiary of Olympus NDT (Waltham, MA). "The personnel involved in the phased array ultrasonic inspection must be trained and certified."
Adding to the challenge is that inspectors are working with more and more complex parts. "Many parts have complex geometry including corners, protrusions and tapped holes. These features can hide defects and also reflect ultrasonic energy just like defects," Hackenberger says.
That is why the capability to examine a part from multiple angles and depths is so important. New hardware and powerful software create images with multiple views of the part and provide better feedback about the size and location of the flaws.
For example, if a clear, 1-inch thick plate has holes drilled into it from the bottom of the plate, those holes will look different-and give different data-depending on the angle depicted. From the top, the drill holes would look like circles, a view that is called a C-scan. If looked at from the side of the plate, however, the operator would see two pillars running up through the plate. This is a B-scan.
An additional complication for inspectors is the emergence of composites, according to Frost & Sullivan. The study pointed specifically to delamination problems in composite materials that are frequently used in aerospace and other industries.
Carnevale says that operators need to be mindful of material differences. Steel is easy to examine with ultrasonics, while multilayered composites are more difficult. "Sound travels through the gel and the fibers at different speeds and you need to be able to be aware that the timing of these echoes may just be a function of the physics, as opposed to an error or defect," he says.
Frequency levels for the different levels are important as well. Typically, low frequency is used for materials that absorb sound, a process called attenuating. High frequency is used for near surface resolution or smaller defects. "A low frequency, such as 1 megahertz, is used on parts where material structure is coarse and hard to penetrate with sound, something like cast iron," Carnevale says. "If you were to use a higher megahertz transducer on a cast iron part, the grain boundaries would come back looking like flaws or you would likely not receive any returning echoes because of scattering."
Higher frequencies are used on materials with tight grain boundaries. For example, fine ceramic would probably use a higher frequency such as a 15-megahertz transducer that can find small flaws.
The real key to phased-array technology is what is done with all of the data, adds Dave Jankowski, general manager of ultrasonic and eddy current products, GE Inspection Technologies (Lewistown, PA). Scans run the gamut from a simple A-scan, those seen on oscilloscopes, on to scans that combine the features of various scans to create 2-D and even pseudo 3-D images. "Phased-array probes give the operator the advantages of speed in collecting data from more complex geometry and image creation from that data. The image is presented to the operator, so they can discern the quality of the component."
![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image5-sm.jpg)
The process
Phased-array technology allows the user to ultrasonically scan an area without having to move the instrument. Applied first by the Armed Forces, who used phased-array technology in antennas to electronically steer radar and sonar "beams," the concept employs electronically "synchronized" smaller beams of sound waves that are grouped together to sweep a larger area and which can be more tuned for defects.While traditional ultrasonic flaw detection maps a part's internal structure by using one transducer element, producing a single "beam" of sonic waves that can be physically moved across the structure to form an ultrasonic image (called a C-Scan), phased-array technology uses many transducer elements that are electronically steered into the structure-producing a similar image. This means increased coverage area with each pass; that in turn increases productivity by reducing inspection time.
The analogies describing phased-array ultrasonics are many and include a ladder, pebbles dropped into a pond and a flashlight. If thought of as a flashlight, a conventional UT system would be a single beam of light; whereas a phased-array flashlight is made up of many smaller beams of light. And, each of these beams can be fired or turned on independently so they light one at a time, in groups or all at once. By adjusting the timing, along with choosing the appropriate bulb size and wattage, the resulting beam of light can be controlled. It can be moved from straight-ahead to left or right, up or down, brightening a swath of a room without having to move the flashlight.
"Phased array is a term used to indicate special ultrasonic transducer usage," says Jankowski. "That allows us to essentially phase the excitation of the different elements and control how and where the part is inspected."
A conventional inspection process is performed with a single type (typically angle and focal length) of ultrasonic transducer. However, for every angle or focal point desired, a separate transducer must be used. If a part needs to be inspected at a 45-degree angle, a 45-degree transducer is required.
"All the phased-array system is doing is what a human would do by hand, or in an automated scanner, using motion, but it is doing it at electronic speed," says Richard Kazares, VP and general manager of NDT Automation (Princeton Junction, NJ). "But, many inspections are done with sound that is on an angle, rather than straight. You can have that same angle with a single transducer. You could move it back and forth and sweep it by hand, which is called scrubbing, while the operator looks at a screen looking for flaws."
On the screen is the image of the part's internal features. "When you do C-scan imaging using a single transducer, the operator, or an automated system, is moving the transducer across the part to create a grid," says Kazares. "If done manually, the operator would stop the transducer and capture the data. If performed automatically, the data is put into a grid format (C-Scan) that corresponds to its position.
"With phased array, the pixels are electronically moved in one, or more, directions and electronically ‘fill the grid,'" Kazares adds. "With phased-array technology, if the transducer array is oriented horizontally, it would still have to be moved-mechanically-in the vertical direction to fill the grid pattern. With a single transducer, the same result can be accomplished but it has to be physically moved in both directions-a more time-consuming process."
In fact, the nuts and bolts of the ultrasonic inspection do not change whether the tester has a single element transducer or an array of transducers. If the application calls for a 5-megahertz, 1/2-inch transducer using a conventional UT, so too would a phased array. The difference is that the transducer would be made up of the array of elements that add up to equivalent performance, and would also have the additional benefit of allowing the part to be inspected from many different angles.
![](/NDT/2006/04/Files/Images/ndt0406-ft3-flaw-image6.jpg)
Virtually probing
The act of inspecting a section of a part is called a virtual probe. The probe occurs when an array of elements are fired. Arrays are set up in octo groups made up crystal elements that range in groups of 8, 16, 32, 64, 128 and the latest varieties of 256 elements. Each array has its own pulser and receiver. The arrays are typically used is a linear configuration, although the array can be many shapes and sizes.These elements can be fired as a single group creating a single, straight-ahead pulse; they can be fired in smaller groupings, to get specific angles and depths. They can also be fired one at a time with a deliberate time delay, measured in nano-seconds, between pulses that are created to create sonic waves at predetermined angles. This process is called beam steering and is considered the number-one strength of phased array ultrasonics.
For example, if a specification calls for a 45-degree sweep to the left and 45 degrees to the right, then element 1 would be fired, there would be a delay of several nanoseconds and then element 2 would be fired. This would continue until element 16 is fired and the entire 45-degree virtual probe to the left is accomplish. To do the right, the elements would be fired in reverse order, 16 to 1.
It is in this state that the elements pulse constructively and destructively collide with other pulses allowing the beam to be steered.
Trends ahead
Phased-array ultrasonics has followed the line of improvement that most other electronic-driven tools have followed. It has improved its electronics, its software and its scan-and-display capabilities. The suppliers say that these improvements will continue. In-line automation will grow, software algorithms will get more powerful and automatic defect recognition will occur."A system," says Hackenberger, "might be programmed to recognize that if two adjacent points are above a certain amplitude, then the customer wants to call that a defect. If they are not adjacent, then those two flaws may be acceptable. The systems will make that ‘decision' based the parameters that have been set."
A preprogrammed decision tree? Could neural net be far behind?
For more information on the companies mentioned in this article, visit their Web sites:
- American Society of Nondestructive Testing, www.asnt.org;
- Danatronics Corp., www.danatronics.com;
- Frost & Sullivan, www.frost.com;
- GE Inspection Technologies Co., www.geinspectiontechnologies.com;
- NDT Automation, www.ndtautomation.com;
- Olympus NDT Corp., www.olympusndt.com/en; and
- University of Dayton Research Institute, www.udri.udayton.edu.
SIDEBAR: PHASED-ARRAY READY TO MAKE A 'SEA CHANGE'
A decades-old technology, used infrequently by manufacturers, may be ready to explode into industry's collective consciousness, according to new market analysis.A study released in February by Frost & Sullivan, a market analysis firm, says that phased-array ultrasonics tools are "about to bring a sea change in the NDT industry."
Phased-array ultrasonics scans a part, looking for hidden defects, without having to move the inspection instrument, creating productivity gains and better defect detectability.
Array technology has migrated from radar- and sonar-antenna uses by the Armed Forces, who discovered the phenomena during WWII and used the technology extensively in the Korean conflict. This migration has led to the consumer market where it is used in cars and mobile phones, and principally, the medical industry, which remains one of the more enthusiastic users of phased-array ultrasonic technology.
For the last several years, phased-array ultrasonic imaging has been used more frequently in industries such as automotive, aerospace, nuclear and power generation industries.
While this technology may have been on the fringes in years past, taking a back seat to the less-expensive single-element, ultrasonic transducer technology that has dominated the marketplace, that is about to change. Despite its price tag in the $60,000 range, compared to $6,000 for a portable, digital ultrasonic flaw detector, the marketplace for phased-array ultrasonics appears to be growing. Buyers, according to industry watchers, seem to be hearing the message of ultrasonic technology: it is fast, accurate and reliable.
Research analyst Vijay Shankar Murthy, says that the phased array technology will reduce the time for inspection and gradually help in reducing overall cost of the final product. He points to new materials as one reason:
"The emergence of composite structures as a material of choice for aircraft manufacturers has also brought with its sophistication of testing techniques used to inspect them," Murthy says. "The transition from X-ray inspection to non-film digital methods of structures within the aerospace industry is a direct result of the use of new and cost effect methods for manufacturing."
SIDEBAR: Phased-array ultrasonic technology is based on the following technical features:
- Multiplexing a large number of identical crystals as a single probe.
- Control of the focal depth
- Control of the steering angle
- Control of the beam width
- Program of the virtual probe aperture
- Display of the ultrasonic data in a generic view called S-scan
Source: R/D Tech