Volume 5, Number 3		July 2006

 

 

Unique applications originating from the Columbia space shuttle catastrophe required the development and implementation of new radiographic technology. A very large CMOS digital system incorporating the parallax method was developed to view the reinforced carbon carbon panels that shield the space shuttle.

New Technology Needed

On January 16th, 2003, the space shuttle Columbia took off into the Florida sky on what would be its last flight. One and a half minutes into the launch, insulating foam disengaged from the central booster tank and fell — impacting the leading edge of the left shuttle wing and damaging the reinforced carbon-carbon (RCC) composite panels that protected the shuttle from the heat generated by atmospheric friction. During re-entry, the unprotected surface allowed heat to penetrate into the wing structure causing the structural members of the wing to fail, and the shuttle to disintegrate at supersonic speed.

In addition to determining the cause of the crash, the investigation conducted by the Columbia Accident Investigation Board (CAIB) concluded that more inspections were needed for the shuttle program and that new technology was needed to assay the integrity of the RCC panels. Each wing of the shuttle has twenty-two U-shaped RCC panels; one side of the U is flat, the other composed of a continuously variable curve. New technology utilizing the parallax method and incorporating a complementary metal oxide semiconductor (CMOS) three-dimensional digital imager was developed specifically for the inspection task.

Computer-Generated Tomography

Before discussing the advanced imaging technology of the parallax method, it is helpful to have a basic understanding of the computer-generated tomography it is similar to. The word tomography is derived from the Greek word tomos meaning a piece cut off or a cross-sectional view. Tomography therefore is a technique in which a single selected plane or slice can be imaged. Through the use of software, the outlines of structures in all other planes can be eliminated. Alternatively, many sections can be viewed simultaneously from a single perspective to show the volume of a particular area or the entire structure of the item being imaged.

Tomography has been around the industrial and medical fields for approximately thirty years. Most of us are probably more familiar with medical tomography applications in which the patient is positioned on a flat table that is indexed horizontally through a ring-housing containing an imager and an X-ray device. The radiographic source and the imaging panel are situated directly opposite each other within the ring-housing and are rotated around the patient. This rotation, in conjunction with the indexing of the table captures a series of cross-sectional planar images that are helical in structure.

Industrial tomography varies from medical tomography in that the X-ray device and the imaging panel are securely mounted and are stationary during the test. The component to be imaged is placed on a turntable that is situated in-line between the source and the imager. The turntable is rotated (indexed) completely around, stopping at designated increments to capture a predetermined number of images.

The obvious advantage of tomography over film radiography is that the component can be viewed three-dimensionally rather than as a flat two-dimensional image. This allows the determination of depth and thereby the accurate location of objects or discontinuities within the component. In addition, the ability to isolate individual layers of an object allows the observer to look at areas of interest without the superimposed images of the rest of the component.

The disadvantage of industrial tomography is that the helical data acquisition technique creates massive data files that typically take many hours or even days to process, depending on the sensitivity required. The required sensitivity determines the number of images the system must acquire. The diameter of the smallest discontinuity to be detected is the minimum rotational increment. The following formula may be used to determine the minimum number of images required to detect a discontinuity of a specified size:

C ÷ D_min = X

Where:
C = circumference of imaged object
D_min = minimum defect size
X = number of images required

Using this formula, if one wanted to detect a 0.05 in. (1.27 mm) discontinuity in an object that was 30 in. (76.2 cm) in diameter, 600 planar images would have to be acquired. For further discussion of computed tomography, the reader is referred to additional documentation.^1-3

Parallax Method

What does parallax mean? Parallax is the apparent change in the position of an object resulting from the change in the direction or position from which it is viewed. The significant advantage of the parallax method is that by employing basic geometry, it can greatly reduce the number of images required to render a component, thereby increasing the number of specimens that can be processed in an equivalent amount of time. Typically, a digital system using the parallax method will only require the acquisition of 4, 8, 16, or 32 images. It is how the images are taken and processed that distinguishes the parallax method from ordinary tomography.

The parallax system images the component at an angle. The angular cross-section represents the component from top to bottom and includes aspects of the component’s width in the image. Picture the first image as a diagonal cross-section of the component. Using correlating data from another diagonal cross section, dimensions and densities are projected vertically and horizontally. By correlating the different perspectives together, the computer generates the different vertical and horizontal planes. One plane may image the front of an object accurately while another images the side. When this data is correlated with a top or bottom view, we gain a three-dimensional image. This technique greatly reduces the affects of geometric unsharpness. Additional information regarding the topics of parallax and stereo radiography can be found in other resources.⁴

The parallax system developed to inspect the RCC panels for the space shuttle program is permanently installed in a vault that is 15 ft. X 15 ft. X 20 ft. (4.6 m X 4.6 m X 6 m). The wall behind the linear array is lined with 1.75 in. (4.4 cm) of lead shielding to protect personnel. Wall-mounted emergency shut-off switches and a safety interlock on the entrance door are additional safety requirements. A 450 kilovolt (kV), horizontally mounted, X-ray device is utilized (Fig. 1a). The linear CMOS array is mounted horizontally between two uprights masts (Fig. 1b). The array may be moved to any elevation on the masts.

Also mounted on two vertical masts in the center of the system is a mechanized turntable. The six-foot diameter round table has motion control in two axes. The imager and X-ray tube remain stationary and the component is rotated. The component does not rotate during data acquisition. Through the use of gear reduction motors, the table is tilted to 45 degrees before processing begins and is held at that angle for the duration of the inspection by two locking rotors. High strength aramid fiber straps secure the test component to the turntable. The restraining straps are relatively transparent during imaging and will not adversely affect the procedure.

Image quality indicators (IQIs) can be used during the course of the inspection. Traditional penetrameters or wire-type penetrameters can also be incorporated around the component to ensure sensitivity. However, for accurate linear measurements of a component, the most important indicators are round metal discs of known diameters. These serve as points of reference when the images are correlated and provide a known dimension within the completed images as well. Step wedges may also be placed within the viewing area for the same purpose.

With the proper X-ray procedure for imaging the material of interest established, the parallax system is now ready to begin data acquisition. The required voltage (V) and amperage (A) are programmed into the X-ray machine and the tube is turned on. The linear array takes a snapshot of the photon energy penetrating through the component. When the image is complete, the X-ray tube is turned off. For most materials, the image is acquired in less than one second. A computer communicates with a motion control system that drives the rotation of the table and the table is indexed the predetermined number of degrees with feedback from encoders. When in position for the next perspective, the X-ray tube is reactivated, and another image is captured. This process will repeat itself until the required numbers of images have been completed.

The individual images are then processed by the computer to form a three-dimensional representation of the component. This aspect of parallax imaging process is the most time consuming part, taking approximately an hour.

The round metal disks that were placed around the component are now used to calibrate an accurate virtual caliper. Precise measurements can be made by clicking a mouse where the measurement is to start and dragging the cursor to the point where the measurement is to end. Discontinuities can be sized and evaluated by precise measurements in all three major axes. The three-dimensional representation is scalable and can be viewed in any plane with sharp, distinct edges.

Detecting Panel Discontinuities

The procedure to detect discontinuities on the RCC panels requires the X-ray radiation to penetrate the material at a line tangent to the discontinuity. One of the primary areas of interest in each of the forty-four shuttle panels is the point at which the leading edge blends into the straight section of the panel. Using a tomographic system, it would be necessary to image through approximately 24 in. (61 cm) of material. This is because the panel sits flat on the turntable as the horizontal images are acquired. At certain points, some areas of interest cannot be viewed because the material is too thick to be imaged.
Alternatively, the parallax method tilts the panel 45 degrees for imaging and all of the images are acquired while in this plane. This means that if an area is too thick to view in one plane, there are at least two other image planes that bisect the area of interest at different angles. The data from those two planes can be projected to reconstruct the area of interest. Thus, the parallax method increases the effective volume that can be imaged when compared to tomography.

Conclusion

Similar to computed tomography in that it also allows for the construction of a three-dimensional representation of the component being examined, parallax imaging has the significant advantage of requiring far fewer images to do so. This results in a proportional increase in the number of components that can be processed in an equivalent amount of time. In addition to the ability to render scalable images with distinct delineation, parallax imaging also allows accurate depth (location) and linear measurements and can image unique part geometry not possible with traditional tomography. TNT

References

Raymond R. Shepard is the Training Manager/Quality Specialist for Kakivik Asset Management LLC in Anchorage, Alaska. He is a former Professor of welding and NDT at the University of Alaska, Anchorage. (907) 770-9418, rshepard@kakivik.com.

 

[ The NDT Technician ]

 

 

---

 
Copyright © 2008 by the American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement or recommendation of ASNT.

IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc.

ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.