From RoHS to BGAs, the technology fills an inspection void.

Over the past few years, the use of x-ray to analyze solder joint structures has not only accelerated but has offered a wholly new view on joint inspection. (While the use of x-ray in electronics is a relatively recent phenomenon, the technology has been used for years in medical, metal casting and welding applications and for general metallurgical inspection.)

The impetus for using x-ray technology in electronics came when BGAs ventured into regular use. How could one determine if all the joints were formed correctly? The first systems were often adapted from medical x-ray systems and suffered from definition less acute than available today. It was difficult to interpret images and so the development of intelligent software routines accelerated. Early 2-D systems became 2.5-D systems that could tilt and twist the sample, or the x-ray tube, so that different perspectives of the same image could be determined and BGA joints could be seen with a high degree of confidence. For example, Figure 1 shows a typical oblique view of a set of BGA joints, and it can be seen that the corner balls have not made good contact with their pads - possibly due to thermo-mechanical stresses within the BGA's structure.

X-ray parameters can now be fine-tuned to show a clear picture of the target image. Modern high power and microfocus tubes permit a more varied resolution and materials with differing densities can be more easily seen in detail. Hence it is possible to analyze voids in detail.

Joints have always had voids, but they could be seen only under microsectioning. Microsectioning is a destructive exercise but we now have a nondestructive tool to identify voids.

Objects with strong contrasts, such as dense solder joints and copper traces within PCBs, appear cleanly at a good resolution and rapid interpretation of faults can be made.

But electronics assemblies almost always end up in some form of casing or rack system and the need is growing to see what may be in said casing or in a less dense material as well as what has happened to the PCB. We also need to see a true 3-D view of the assembly to hunt for defects that may pass all initial test criteria but threaten longer-term product life. For example, if the assembly is conformally coated or totally encapsulated, fault-finding a field return almost always results in destructive analysis; this destructive element can sometimes introduce faults or misleading information not present in the first place.

Materials used for coating or encapsulating are far less dense than the metallic elements found in assemblies and we now need x-ray systems to offer a range of abilities to suppress reflective images from dense materials while also permitting good views of dense materials.

Medical imaging has used CT (computer tomography) techniques for years to permit the varying densities of human tissues to be evaluated. In essence, the technique requires many x-ray images slices to be taken across the sample and the software then recombines all the slice images into a 3-D picture. The process is complex and the software is usually vast and requires huge amounts of computer memory.

A typical CT system used for electronics will probably take between 400 and 1600 slices. The tradeoff is the time required versus the ultimate resolution. Density variation across the slice will likely be large and so the focus spot size of the tube and its power rating are likely to be 130KV and 5 µm or so. But, we cannot talk just in terms of the spot size because the recompilation of the images is 3-D and the terminology used in this technology is VOXEL, or volume pixel. A 90KV tube power is likely to offer a 50 µm voxel or 50 µm cube element. Larger power ratings, different detectors and the sample size all factor in the ultimate result.

Confused? It's not surprising. It took the medical world years to work out the optimum parameters for analyzing a human body and we are only now finding sensible parameters for industrial use. Figures 2 and 3 are individual slices from a CT scan of a seed and a Rotring drafting pen, respectively. Both have low but varying densities of material and they emphasize the level of the technology available by showing a good degree of contrast. One can clearly see what they are and the images are not "fuzzy" as they once were.

We cannot state here what is correct for all applications nor can we state the power or focus needs for everyone but it is clear that the analytical tools at our disposal are much improved.

Advances in CT technology are permitting looks beyond voids in joints to voids or other defects in the entire assembly. An encapsulated product may have tiny voids, or blow holes, anywhere in its plastic encapsulation, or it may not, depending on the process control of encapsulation. CT x-ray with high power and high resolution offers the chance to nondestructively check this at any stage during production or later during field return analysis.

As with any test equipment, x-ray is a non-value add tool and is less easy to justify than a production tool. But, considering our earlier example of the encapsulated product, the cost of destroying a product just to find a fault or introduce one that may not have been there in the first place is a burden on company finances. CT x-ray can be cost-effective.

The XRF answer. XRF (x-ray fluorescence) is now used to check the core substances of materials, often for compliance with RoHS laws. XRF is not the complete answer to finding banned substances as it has limitations in determining compounds. Individual elements can be found but brominated flame-retardants show up as bromine. However, as screening tools, XRF systems offer a good initial gauge on the likelihood of a substance causing trouble later.

The principle of XRF is as follows: High energy photons emitted by an x-ray tube interact with the sample target. The high energy photons are absorbed by an electron of the target atom. This electron is accelerated and forced to leave the atom. The "hole" thereby created in the structure of the electron shell is filled up by an electron of higher energy. The difference in energy between the ejected electrons and the newly arrived electrons may leave the atom as a photon of defined energy, or as an electron. If a photon, the process is called x-ray fluorescence and the energy of the ejected photon is characteristic for this atom and therefore for the element. This also happens in any x-ray system but the detectors are set to look for different parameters.

X-ray is playing an increasingly important role in electronics manufacturing and the pressures of traceability and product confidence mean that its use is unlikely to be transitory. The growth in the use of SiP and array packages will require a regular ability to check the construction of the package or the joints and CT x-ray is likely to play a large role in this. Even after the frenzy of RoHS activity has calmed, the need to check material content will remain, as exempt manufacturers are likely to want to check if lead is present.

 

Peter Grundy is director of P G Engineering (Sussex) Ltd. and ITM Consulting (itmconsulting.org); peter.grundy2@btinternet.com. His column appears semimonthly.