When considering X-ray inspection technology, the question should not be which type of system is better, but which system is better suited for the specific quality application. Indeed, each system has distinct and differentiating features and advantages.

2-D advantages
Simple 2-D X-ray inspection systems create planar images by viewing the sample in a top-down fashion. The high voltage rating of the system's X-ray tube determines the penetrating power through the sample, which is placed between the X-ray source and image intensifier. The distance between the tube and the image intensifier helps determine magnification (geometric), and the focal spot size determines image resolution. Resulting images show the X and Y dimensions in grayscale.

In rudimentary 2-D inspection systems, the sample and image intensifier cannot be angled, which is why they can only produce top-down views. However, in response to demand for more detail, 2-D inspection systems made today offer some level of oblique-angle viewing by moving the sample with a tilt-and-rotate feature. Although rotating the sample is the most common method of oblique-angle viewing, this technique limits magnification for larger samples. At the same time, the sample congruently moves farther from the X-ray source, creating a distance between the sample and the tube. As this focus-to-object distance increases, the magnification is reduced.

A more effective means of obtaining oblique-angle views for large samples is by movement of the image intensifier. While some systems offer movement of the image intensifier in quadrants or other specific positions, the most advanced systems offer 360-degree rotation and 60-degree angles, providing views from all possible angles without sacrificing magnification. This allows for nearly perfect inspection of hidden structures in complicated assemblies such as those using flip-chips and u-BGAs, or micro ball grid arrays, in which solder connections are located on the underside of the chip or package.

Standard 2-D X-ray inspection systems offer image resolution between 8 and 10 microns, while the most advanced systems can achieve resolution of less than 1 micron. Computer numeric control (CNC) is an optional feature of 2-D systems that aids in automated sample manipulation and image processing. Other standard features include software that allows the user to manipulate, save, e-mail and review images. Some offer five-axes sample manipulation for oblique-angle views. Nonstandard product offerings include larger cabinet size in order to test larger parts, six-axes manipulation, which is an oblique-angle viewing with tilt and rotation of both the image intensifier and the sample, and a granite table, which contributes to measurement accuracy and repeatability.

The key advantage of 2-D X-ray inspection is its speed. The 3-D inspections can take twice as long as 2-D inspections. 2-D X-ray images are in grayscale and require limited computer manipulation to produce. They also tend to be simpler to interpret for typical problem areas such as pad wetting. Additionally, the systems are flexible with respect to the numerous types of applications they can handle, as well as their ability to customize test settings.

On the other hand, standard 2-D systems do have limitations. Some electronics packages are more densely populated than ever before, causing inner structures to be somewhat obscured to 2-D X-ray systems. Even on single-sided printed circuit boards, the shadowing or nonappearance of inner structures can hinder a correct analysis. Standard systems can alleviate some of these problems; however, the most advanced 2-D systems with oblique-angle viewing by movement of the image intensifier can offer a valuable glimpse into the third dimension.

3-D for more details
3-D systems are used for applications that require a more detailed analysis. These systems, offering 3-D inspection of planar components, work in an entirely different way, with the ability to determine a feature's position and identify inner structures unavailable in 2-D X-ray inspection. Images can be taken at an angle to the sample or top-down like standard 2-D systems. From this point, the user can choose to view a 3-D analysis of the sample, created by bringing together multiple 2-D images through software computation and reconstructing it to show the depth dimension (Z axis). The user chooses the height, geometric magnification, and resolution to be used for the model, and can move the 3-D rendering in all directions to precisely reveal the assembly's inner structure.

The key to these systems is creating the initial 2-D images that will be manipulated by the system software to create the 3-D view. Systems form these slice images in different ways. The most common methods are laminography and planar computed tomography (PCT).

Laminography involves synchronized movement between the X-ray source and image intensifier. Multiple angles are used to image a specific layer. This process can be time-consuming when multiple-layer information is needed. All of the views are combined to form a single slice of the sample in which the focal area is prominent and all else is blurred. A disadvantage of this is that the blurred background reduces the overall image quality.

In PCT, the sample and detector are rotated around a fixed source. This involves choosing a number of 2-D projections and a beam angle. The information taken from these views is stored and processed to form the tomosynthesis (layer-by-layer) and algorithmic reconstruction theory, which allows tilting, rotating and slicing of the image.

There are multiple advantages to 3-D systems. They provide depth information--in addition to viewing a void, one can also determine its exact location. Also, the 3-D image includes views of hidden structures within double-sided or dense boards. Finally, the systems typically make use of a software-based process "wizard" that guides the user through the test process. This simplifies system programming and image storage, and lessens the amount of user input required.

Key factors to consider when selecting a 3-D system include image acquisition and reconstruction time, image resolution and system price. Some 3-D systems are specifically designed for production applications. These automated systems allow basic viewing of areas of interest relatively quickly, though not with the speed of a 2-D system. This speed, however, comes with a considerable loss of image resolution. 3-D X-ray inspection systems that are designed for R&D or failure analysis applications produce re-constructed images with much greater resolution and provide the most detailed view into the depth dimension. These systems can often take 10 to 15 minutes for image reconstruction because they are the result of computations based on much more detailed sample data.

How to choose
Arguments can be made for both 2-D and 3-D X-ray inspection systems. 2-D systems are best for production control functions across industry barriers, proving themselves in automotive, aerospace, microelectronics, surface mount technology, semiconductor and medical applications--all of which involve testing numerous samples throughout the manufacturing process.

3-D systems are useful tools for more intricate processes and products. These systems are best suited for failure analysis and R&D projects in which slower image development is not a major consideration. More specifically, 3-D X-ray inspection works best for determining the position of internal structures of electronic devices, and assemblies using technologies such as flip-chip, ball-grid arrays, uBGAs, multichip modules, chip-on-board and double-sided boards.

An X-ray system is only as good as its ability to meet the customer's needs. The best rule of thumb in choosing X-ray inspection systems is to know what they offer, how they operate, and which type is best for an application.

TECH TIPS

• Standard 2-D X-ray inspection systems offer image resolution between 8 and 10 microns. The most advanced systems have resolution of less than 1 micron.
  
• The key advantage to 2-D X-ray inspection is its speed.
  
• 2-D images tend to be simpler to interpret for typical problem areas such as pad wetting.
  
• 3-D systems are used for applications that require a more detailed analysis. Images can be taken at an angle to the sample or top-down like standard 2-D systems.
  
• 3-D systems work by constructing 2-D views or "slices" into 3-D images. The most common methods for doing this are laminography and planar computed tomography (PCT).
  
• Laminography involves synchronized movement between the X-ray source and image intensifier.
  
• In PCT, the sample and detector are typically rotated around a fixed source. The information taken from these views is stored and processed to form the tomosynthesis (layer-by-layer) and algorithmic reconstruction theory 3-D views.