You’ve likely heard about the use of industrial CT scanning and may be wondering if adopting it is the right move for your growing manufacturing business. This whitepaper provides a look at the current state of product inspection, what CT scanning combined with analysis and visualization software brings to the table, and the benefits that moving to this more advanced methodology can provide to you with a wide variety of industry examples.
Did you know that you can download this whitepaper as PDF by filling out the form?
Whether it’s engine blocks or solder joints on printed circuit boards, manufacturers know that whatever they produce, they must also inspect. Vision- and laser-based inspection systems, optical comparators, micrometers and various gauges—there's a broad range of metrology and inspection equipment being used these days. But such tools may not be providing you with everything you should know about your product because they can’t always reach, or even see, specific part features, let alone accurately identify defects within the material itself.
Some non-destructive testing (NDT) approaches (such as eddy current and magnetic particle inspection) can detect shallow material flaws. Destructive testing, through cross-sectioning of a workpiece, is both costly and limited. When your goal is meeting the most exacting specifications for quality control that you can, you need a better way to peer deep inside your parts and extract detailed, accurate data about every structure inside data that can then be used to asses dimensional conformity, analyze structural integrity, predict performance, and perhaps even guide changes to your manufacturing methods.
The technology to achieve this level of quality control exists and it begins with industrial CT scanning. Also known as 3D-computed tomography, this technology works by processing multiple X-ray images to gain a dimensionally accurate 3D model of the object being scanned.
The wavelengths used with industrial CT scanning are both shorter and more energetic than those used to X-ray a broken bone, which means they can easily penetrate dense materials such as wood, plastic, composites, and metal. And where X-ray machines take a one-sided image of the target without depth resolution, industrial CT scanners view all sides of an object and use mathematical methods to reconstruct a volume. This resulting 3D volume offers an interior view that no other technology provides.
Yet CT scanning alone does not deliver everything you need to assure and document product quality. The best scans in the world won’t tell you enough without the right software tools for analyzing and visualizing your scan data.
Advanced analysis and visualization software, such as that provided by Volume Graphics, supplies supporting capabilities that far exceed the native image gathering potential of CT equipment alone: Automated collection and analysis of scan data, defect detection, measurement functions including GD&T, FEM meshing tools—these are just some of the available CT data-analysis software capabilities that will enable you to fully streamline your own quality assurance activities and deliver those high-quality, finished parts.
Regardless of how you choose to access the power of CT for quality assurance, it may be useful to have a brief review of the processes involved.
Key to the CT scan and the critical data analysis that follows is the voxel. When an X-ray beam’s high-energy photons pass through their target, some are absorbed and others scattered, but the remainder strike the detector screen opposite the X-ray source. They then energize the (two dimensional) pixels there in proportion to the density of the material being scanned as well as its thickness and attenuation values, and generate a series of grayscale images.
This is where the computational part of the process comes in. CT-data visualization software reconstructs those millions of pixels into three-dimensional voxels. Then, employing advanced mathematical algorithms, the software creates a full three-dimensional visualization of the exact object that’s been scanned. Next, a rainbow of tools can be employed to analyze the material density, stress patterns, internal defects, design flaws and a host of other attributes of what’s been scanned— which includes machined metal parts, 3D-printed prototypes, plastic injection-molded parts, fiber-reinforced plastics, or anything else that’s scannable.
These software-based analyses and visualizations can be highly valuable to manufacturers in almost any industry, allowing designers and engineers to understand failure modes, validate critical part features, predict how products will perform, and both quantify and qualify manufacturing processes. That’s the beauty of CT scanning when married to analysis and visualization software.
“But my company has already invested millions in our quality department,” you may be thinking. No matter how capable it may be, do you really need yet another system, one that requires X-ray radiation to image a part, and specialized software just to analyze the results? If you don’t want to purchase a system yourself, is it worth it to outsource CT scanning to a reputable firm that has the environment to do it for you?
The answer is often a resounding yes, and for some excellent reasons. As an example, consider a plastic injection-molded connector, like the ones found in cars, refrigerators, or laptop computers. Manufacturers produce these surprisingly detailed and accurate objects via multi-cavity molds, each of which must be validated by carefully inspecting the parts that come out of them.
Measuring the products' visible features is accomplished using one of the metrology tools mentioned earlier. The unseen interiors are a different story, however. To inspect these, manufacturers often have to destroy representative part samples. In this case, a technician must saw the connector into sections, polish the rough edges, and then inspect the now-visible faces. If a defect is found, the part must be made over again. As the mold cavity wears, the process must be repeated. After a shift change, or a new batch of resin, or an equipment malfunction, the repetition may continue.
The bottom line? The status-quo methodology of sectioning manufactured parts, whatever they are, is expensive, time-consuming, and error-prone. Consider all those connectors in between, the ones that weren’t destroyed: Unless the manufacturing process is perfectly stable, which is rarely the case, there’s always a question mark left hanging about the parts or part sections left unchecked.
While the question mark can be relatively small in the case of molded components and similar high-volume products, the same can't be said about mission-critical, costly products such as turbine blades, engine blocks, high-end electromechanical assemblies, and the newest darlings of the manufacturing industry, 3D-printed parts. In such cases, major manufacturers worldwide have already set up in-house CT scanning facilities while others with more limited financial or bandwidth resources are turning to CT-scanning service providers.
For a deeper understanding of how this methodology can help you create better tooling, improve part quality, increase product lifespan, and reduce costs, let’s take another industrial example, this time metal casting. This is a mature and well-understood process, used for automotive powertrain components, pumps and valves, turbine blades, and practically anywhere robust, yet often complex, metal parts are needed.
There’s just one problem (and it’s quite similar to that faced by plastic injection molders). Because the molten material used with casting and molding is made to flow through a series of cavities, the likelihood of air entrapment is high. As a result, voids and porosity are an all-too-common occurrence, and the only way to detect such flaws short of sectioning the part and thus destroying it is with high-end CT scan-data analysis and visualization software.
There’s more to it than finding a bunch of bubbles in a metal part: There’s also the critical aspect of shrinkage and warpage to consider, as well as wall thickness, inclusions (trapped foreign material), draft angles, as-cast to as-designed comparisons—never mind the need to measure and report dimensional information on hidden part details. These are just a few of the roles that industrial CT-data analysis and visualization software is now beginning to take on in casting houses everywhere.
Such information gives toolmakers the knowledge they need to make intelligent decisions about mold design. Since they now have a better handle on the unseen world, they can identify what steps should be taken to accommodate thermal effects and material flow within their tools.
Here again, a capable CT-scan-data analysis and visualization software package simplifies this process by generating meaningful views of the workpiece (and/or the tool itself). Manufacturing-geometry-correction tools within the software reduce the number of design iterations needed to obtain an optimized tool, and enable the export of values to the shop’s CAD/CAM software for use in the manufacturing process.
This scenario applies equally to forgings and stamped, bent, or formed metal parts; anywhere a machined tool is used to shape metal, CT-scan-data analysis can be used to make it better and less expensive to produce.
The CT-scanning, analysis and visualization process is fast, sensitive, reliable, non-contact and nondestructive. Its powerful capabilities for inspection and metrology provide knowledge far beyond traditional optics or tactile metrology. This advanced methodology can be applied at any stage of production, from prototyping to even testing in-line on the shop floor. The variety of potential applications for industrial CT-data analysis is broad and growing.
One of the newest fields in which CT is being applied is additive manufacturing, a.k.a. 3D printing. An ever-growing number of product designers, MRO suppliers, and entrepreneurs are taking advantage of this relative newcomer to the manufacturing industry for use with prototype and low- to mid-volume production of metal, polymer, and composite parts. But, as with cast and molded parts, concerns continue to exist over material integrity. This is why a large percentage of additively-manufactured parts, especially those made of metal or intended for flight-critical use, are inspected using CT scanning.
CT scanning, analysis and visualization isn’t limited to metals and polymer materials. Composites, used in a wide variety of industries to manufacture parts for wind-turbines, boats, helicopters, cars, other vehicle bodies and more can be analyzed in depth for fiber orientation, part delamination, undulations from process errors, and gluing integrity.
Hundreds of such applications exist, running the gamut from car-key fobs to industrial robots to rotor blades to carbon-fiber racing bikes. In each example, after scanning is done, CT-integrated software helps manufacturers analyze, visualize, measure, and document workpieces inside and out. The result is continuous product improvement, reduced engineering time and costs, and fewer failures in the field.
As discussed earlier, destructive sectioning is clearly limited (and costly). Consider once more our injection-molded connector. You might get lucky and cut the part in two at precisely the right place to find an air bubble or void, but chances are just as likely that defects you can’t see will lie above or below the cutting plane. CT-data analysis and visualization software, on the other hand, allows the quality engineer to take a virtual flight through the entire part from any direction. Nothing remains hidden, no matter how tiny it might be.
Better yet, the right CT-data analysis and visualization software eliminates much of the tedium associated with such part exploration. It supports high-quality renderings and exploded views of parts and assemblies. It makes possible the analyses of cracks, porosity, cellular structures, and fiber alignment, edge detection and surface determination, unrolling or flattening of cylindrical objects and more. Inspection plans are easily generated and executed, part features and geometries measured, potential design or manufacturing failures identified.
The software also automates much of this work, with integrated engineering workflows and other tools to streamline the analysis processes, making the often-huge data sets from CT scanning easier to manage, merge, and analyze.
Industrial CT-scan data analysis and visualization software can make what would otherwise be a highly complex process perfectly suitable and valuable to business profitability whether used on the factory floor, by a metrology service provider, or integrated with the research laboratory of a university.
CT scanning has evolved into a mainstream process. When equipped with capable software, automakers can leverage the largest scanning machines to image complete vehicles for design analyses, while 3D-printing service bureaus and equipment manufacturers can use smaller, less-expensive machines to validate material integrity or measure internal part features. Sitting between these two extremes are those who produce wooden furniture, foam insulation, metal castings, plastic bottle caps, and everything else imaginable— any manufacturer that wants to know what exactly is happening inside their products.
And thanks to the industrial CT-scanning industry’s overwhelming adoption of advanced software from Volume Graphics (which has 80 percent market share), they can see, and understand, precisely what is going on. The company offers solutions to fit every industrial need, from material and geometry analyses to volume meshing, structural mechanics, simulation and more. All are available as easy-to-use modules that offer highly accurate, full-breadth analysis capabilities, giving manufacturers every tool needed to improve their products, no matter what they’re making.
Simply put, industrial CT scanning eliminates doubt. There are no question marks left hanging over any scanned part following analysis with the advanced software tools available from Volume Graphics. Aside from the cost and efficiency benefits made possible by this increasingly essential, advanced technology, industrial CT scanning also reduces risk—so product certification is assured, enhancing corporate reputation.
Whatever it is you’re making, industrial CT scanning— when combined with data analysis and visualization with Volume Graphics software—provides the final word on quality assurance and delivery of competitive products to your target marketplace.
How does CT scanning actually work? And what should a machine shop, bicycle maker, or 3D-printing service bureau know before reaching out to a CT scanning laboratory, or investing in its own equipment and supporting software?
The process starts by placing the workpiece on a rotating turntable, an X-ray source sitting on one side, a digital detector panel on the other. Larger objects might be scanned by simultaneously rotating the X-ray source and detector, not unlike the CT scanner used by medical facilities.
Whatever the approach, the X-ray device is then energized with anywhere between 40 kilovolts to six megavolts or more, depending on a number of factors that we’ll discuss in a moment. As the object rotates, it weakens X-ray radiation in proportion to its density. The X-rays that aren’t absorbed or scattered pass through to the detector panel, which continuously captures the resulting image. Based on the type of machine, the beam shape, and the size of the workpiece, the turntable (or beam) can also be "translated" (moved vertically) to capture the entire part geometry.
The beam itself might be shaped like a pencil, a cone, or a flattened fan, each of which has a direct effect on the speed and quality of the resulting image. Other factors play a role as well, including the target material, the part thickness, production requirements, desired precision, and image resolution.
Perhaps most relevant of these is the material. It might seem obvious that the denser it is, the greater the amount of power needed to penetrate it. This is true, but as with most things in life, more is not always better. While higher power usually correlates to improved image contrast due to less noise, higher voltage leads to less noise as well, but may reduce the contrast in some cases if no countermeasure is taken. It's not possible to simply crank up the volume, so to speak, on very large or very dense parts. It's for this reason that CT scanning is usually limited to suitcase-sized and smaller parts, although there are exceptions.
What does all this talk of power, scattering, and absorption mean? Without getting too far into the electromagnetic weeds, X-rays are generated by applying current to a heated metal filament (the cathode) situated at one end of a tube, not unlike how a light bulb works. Instead of visible light, though, the filament emits free electrons which then are accelerated towards a metal target sitting at the opposite end of the tube (the anode). Electrically- charged coils are used to focus the electrons on a small area, the so-called focal spot.
The target is typically made of copper coated with tungsten, although molybdenum and other super-hard metal coatings might also be used. The electrons penetrate the target material, are decelerated, and emit X-rays. Unfortunately, X-ray generation is not terribly efficient. In fact, only 1% or so of a free electron’s kinetic energy is actually converted into X-rays, with the rest creating waste heat that must be removed lest it destroy the tube producing it.
It’s not necessary to describe how this removal is accomplished—what’s more important to know is that, due to this heat, the minimum size of the focal spot is limited in order not to melt the target. Knowing this, one has to consider the spot size’s effect on image quality.
To illustrate this, grab one of those adjustable spot-size camping flashlights, take it into a dark room, and make shadow puppets. You’ll soon learn that you can affect the size and clarity of the shadow by modifying the diameter of the beam together with the distance between your hands and the wall. A large but still clear image is created by using a small spot size and placing your hands far from the wall (the detector); this correlates to high magnification and low power. Now, twist the lens on your flashlight to make the beam larger (i.e., apply more voltage and/or current), and watch as the edges grow fuzzy. Moving your hand closer to the wall solves the problem, although the image might become much smaller than you’d like. This correlates to low magnification and high power (large spot size).
As you can see, lower power is the often way to go with CT scanning, unless of course you want to inspect parts very quickly on a production line. In these instances, you'll need as much power as you can get; just understand that there might be trade-offs in image quality. Or, you might use a lower power device and simply give the CT scanner more time to complete the imaging process (within reason).