Featured Video
This Week in Quality Digest Live
Metrology Features
NIST
Vernon Dantzler wrote one of the earliest spreadsheet programs
Ryan E. Day @ Quality Digest
ASNA overcomes shop-floor challenges
Dirk Dusharme @ Quality Digest
Amazon Go, productivity hacking, questioning consensus, and Kennedy vs. Trump
Robin Materese
NIST’s list for what’s out and what’s in this year
Ryan E. Day @ Quality Digest
For Manitowoc Cranes, seeing ROI is believing

More Features

Metrology News
Portable measurement experts urged to submit abstracts for presentation and publication opportunities
For measurement versatility on up to three orthogonal axes
Integrates sensors in smartphone, IoT, automotive, and other product designs
Ergonomic, safety, and reliability enhancements comply with the latest European standards
Marking and cutting lasers used in manufacturing endoscopic devices and catheters
Technique provides about a sixfold improvement over regular microscopes

More News

  •  

  •  

  •  

  •  

     

     

  • SUBSCRIBE

Pat Carlson

Metrology

Inspecting Additive-Manufactured Components Using Computed Tomography

A nondestructive option for detecting flaws and anomalies

Published: Thursday, December 8, 2016 - 00:00

From aerospace applications to simple one-off projects built at home, additive manufacturing (AM) has gained incredible interest in all industry facets. Its rapid expansion into production manufacturing is due to the technology's immense versatility and use.

With additive manufacturing, objects are built by adding layer upon layer of material. A wide variety of materials can be used with the technology, including metals, plastics, and amalgams. Given this variety, and knowing that many of these components will be used for critical applications in the aerospace, medical, or automotive industries, the challenge becomes one of quality assurance. How does a manufacturer inspect these products to ensure their safety and longevity? In many instances, traditional inspection methods aren't adequate.

Computed tomography (CT) is a unique and powerful tool that allows users to inspect a component for defects or anomalies, take internal and external metrology measurements, and create a model or drawing from the component. A key benefit of this technology is that it's a nondestructive inspection method, i.e., one that doesn't destroy the component in order to extract data. This allows for both inspection of components and the ability to use end products as intended.

How CT works

Computed tomography begins by taking an image of the component and gathering a specific number or 2D radiographs, typically done in 360° of rotation. These are then processed through software that generates a CT volume of the data. These data are displayed as 3D voxel data, which produce a 3D image that can be manipulated (see figure 1).


Figure 1: CT scan of a full part made of titanium additive

The manipulation can include rotation, flipping, and displaying specific angles and clip planes that allow the operator to cut through the sample in multiple directions for evaluation. From the volume clip plane, a 2D image can be generated that allows the user to further enhance the area of interest, and also take measurements.

As an example, a component made of titanium was created and imaged at two different voxel resolutions, in order to show the capability of locating areas of concern. A key term to understand at this point is “resolution,” which refers to the single voxel size of data captured from the imaging device's (i.e., digital detector array) 2D pixels. The more magnification that can be applied to the area of interest or object, the smaller or better the resolution becomes. Thus, a 10-µm scan will resolve and display indications much better than a 50-µm scan.

The one challenge here is that, depending on part size or area of interest, you may only be able to achieve a certain scan resolution on a specific size of part. Physical limitations include moving the part as close as possible to the tube, or moving the detector back as far as possible.

Another issue concerning resolution is understanding probability of detection, to ensure you have enough voxel data covering a known indication. For example, if you are attempting to find a 45-µm indication, I would try to achieve 15-µm scan resolution or better.

At this resolution you know you have a minimum of three (15 µm) voxels covering a 45-µm indication in each direction of a cube. This scenario generates a higher probability of detection when compared to a single voxel size of 15 µm and a required flaw size of 15 µm.

Many questions must be addressed prior to scanning to ensure that what you are asking for can in fact be accomplished.

Measuring with CT

Let's review the data from the titanium additive sample. Two scans were performed to review and compare data about indication detection. Scan No. 1 was performed of the full sample, and this yielded a 45-µm resolution of scan data (see figure 2). These data were reconstructed and viewed for different types of indications. Indications of voiding, or less dense inclusions and high-density inclusions, were present.


Figure 2: 45-µm scan resolution of a titanum additive part

From these data an isolated area was chosen for scan No. 2. This second scan of the specific area, with increased voxel data, achieved a scan resolution of 15 µm (see figure 3).


Figure 3: 15-µm scan resolution of a specific area of the part

Under higher magnification, the scan now yields a smaller region of interest. Image detail is improved, along with a higher probability of detection. The 45-µm resolution of scan data on the left in figure 4 shows less-dense indications appearing near the inner edge; they are visible but appear blurry with low contrast. Indications shown in the 15-µm resolution on the right in figure 4 now appear more sharp and well defined.


Figure 4: A 2D image shown at a 45-µm scan resolution (left) vs. a 15-µm scan resolution (right)

Computed tomography additionally has the ability to take measurements internally and externally through your component. When viewing specific features on your component, such as wall thickness or feature to feature, the improved image quality from higher-resolution scans will also make measurements more accurate. To further enhance measurement capability, a polygonal wire mesh can be applied to the surface, defining edges and features to allow for increased measurements accuracy (see figure 5). The measurement data are stored in different point cloud data formats, e.g., .PLY,  and .STL. These are then used for measurements or CAD comparison.


Figure 5: Polygonal wire mesh for surface data

To conclude, computed tomography has obvious uses for additive-manufactured components, especially for those used in aerospace manufacturing. It's a tool that can be used for defect detection, including both high- and low-density anomalies. Computed tomography can be used to take measurements where conventional tools may not, especially with internal parts or areas. By comparing CAD data from a known model to the production parts being made, it's possible to create surface data by using the wire mesh and generating point-cloud data for comparison.

Discuss

About The Author

Pat Carlson

Pat Carlson is an applications and training specialist at North Star Imaging.