Finding bad guys and product defects

22 Jun 2015 9min read

X-ray vision has been a fascination for me ever since I first watched the original 1978 Superman film.

He could look through solid brick walls and peer into hidden crannies to find the bad guys, but alas, even Superman’s vision had some limitations, most notably when Lois Lane challenged him to look through her dress and confirm the colour of her underwear when unfortunately his vision was blocked by a lead plant pot.

I know there are a few scientific ‘suspicions’ regarding Superman’s Kryptonian X-ray vision. For example, it is fairly well known that X-ray radiation is dangerous, causing the generation of free radicals that play havoc with cell DNA and lead to the formation of cancerous cells. Whilst Superman’s eyes beamed out his X-rays, the naive population of Metropolis were probably developing an assortment of curious tumours and growths. The other issue is that X-ray images cannot show real colour. Some systems, such as those used in airports, show a contrived colour broadly based on the density or chemical composition of the object, but no matter how much effort Superman spent trying to ogle Lois’s knickers he would (probably) have seen them as grey. However, who am I to question the powers of Superman?

More recently, therefore, my curiosity in X-ray technology has become a little less fictional and a bit more practical. About ten years ago I read an article on 3D Computed Tomography (CT) Scanning, specifically how designers and engineers were using it for industrial research and metrology (measurement).

In the 1970s, CT evolved from conventional X-ray technology thanks to the independent efforts of Sir Godfrey Hounsfield and Allan Cormack. CT or CAT (Computed Axial Tomography) scanning for medical analysis is a well-known process for creating detailed views of internal organs, large blood vessels and bones, but its use in other sectors outside of the hospital is relatively recent.

In 1895, the first X-ray image (or radiograph) showing the internal bones of a hand, was published. A few months later, Major John Hall-Edwards was the first person to use X-rays under clinical conditions to observe a needle stuck into the hand of his associate. Unfortunately for Hall-Edwards, ten years later he had his left arm amputated because of severe radiation burns caused by unprotected self-experimentation but despite this, the field of radiology had been recognised and happily, the risks are now well understood.

X-ray slices

X-ray imaging is the attenuation of X-rays as they pass through a material; the varying densities of different materials or organs determine the amount of X-ray that reaches the detector giving a contrast between the denser structures such as bone and less dense surroundings such as air and soft tissue. During conventional medical imaging, the projected X-rays interact with a detection device (X-ray film or other image receptor) to provide a 2D image of the tissues within the body. CT scanning works on the principle of taking lots of X-ray ‘slices’ rotating around the object (or patient) and using software to process them into a 3D volume rendering.

There are many potential uses for CT scanning in industrial analysis such as measurement reporting, comparing real parts to 3D CAD, inspection of hidden features, defect analysis and reverse engineering (to name a few). The application of CT scanning for industrial metrology has only really become accessible in the last 15 years as computing power and scanning speeds have increased and the costs have decreased. Companies such as Carl Zeiss GmbH and Nikon (which are perhaps better known for their photographic lenses) have invested heavily in the development of industrial CT scanners making the technology more available for engineers.

CT scanning of components and assemblies for metrology can provide good accuracy and a high degree of repeatability. A typical resolution of ~10μm (0.01mm) is normal, which compares reasonably well to most standard touch probe, or coordinate measurement machine (CMM), methods.

Such measurement repeatability is achieved largely because there is little user or machine error (as specific measurements are made digitally in the software post-scan), and because a single CT scan shows all part surfaces, eliminating the need to run multiple metrology programs and jigs for opposing features as with CMM methods. The physical fixture is usually a simple block of cheap polystyrene, a material of a lower density than the object and therefore invisible to scan. Scanning speed is not generally an issue as individual scans can be performed in a matter of a few minutes. One drawback however, is that only a single component (or object) can be placed in the scanner, which means that multiple scans are needed to measure more than one component.


The capability for comparing real parts to the starting 3D CAD has been a particularly useful asset in the process validation stages to check components for possible surface deviations and tooling errors that may have been missed by CMM or visual inspections. To generate a CAD comparison, a 3D scan of a real component is overlaid with the native 3D CAD and any deviations between the two geometries are shown chromatically by a surface colour map clearly highlighting errors. I have worked on many projects with large medical device manufacturers who, for the most part, have been largely sceptical of CT scan technology in their sector. However, in recent years they have been so impressed with its advantages that several companies have purchased their own CT scanners and incorporated scan metrology into their production procedures. This acceptance of the technology by device manufacturers is a good indication of the potential for future production inspection methods.

A digital record of production

A CT scan of a component is an excellent digital record of the production process at any given time. It can be stored without any risk of deterioration (unlike plastic components that degrade over time) and used later to check equivalence with future production runs, or if the tooling is dismantled for maintenance or repair.

CT scanning can also help engineers verify that the design intent for the device has been met. Just as medical imaging is used to see inside the body, industrial CT scanning can be used to see inside devices and observe the individual components in their functioning location, as well as enabling measurements to be taken of previously hidden features. Scans can also show very high levels of detail such as minor voids and cracks within the actual structure of a part. The same scanned data-set can be used to perform a defect analysis, for example the software can generate a report on the porosity and size of air inclusions within a component’s composition. CT scanning enables a detailed and non-destructive analysis to be performed very quickly. A CT scan from a contract service typically costs between £100 – £300 depending on the level of information required, which is very reasonable when you consider the time and effort needed to perform alternative methods, such as ‘potting’ products in resin and physically cutting them up. There are, however, some current limitations to the CT process:

It is not always possible to get a good image if the assembly contains materials with very different densities such as metals and plastics. Typically, a scan of mixed materials will contain some blurring or streaking around the denser materials where the X-rays have slowed.

  • Scanners have limitations on the maximum size of the sample; typically, as the available size increases the resolution decreases.
  • Scans are like a photograph, not a video — they are a snap-shot in time and cannot record the function of a mechanism.
  • File size can be large, in the gigabyte range for a full step-through scan of a device.

Despite these limitations, CT scan technology has proven itself to be very useful in all stages of the development process. The ability to actually see through product assemblies and discover what goes on when the ‘light goes out’ has really enhanced our analytical understanding of how and why a product functions (or not). The fact that a component or device can be analysed without the need for expensive fixtures or any physical contact from a CMM probe is very appealing.

I encourage our project teams to make the most of CT scan metrology to support both the development phases of a project and to help with the industrialisation of a product. As the technology develops, I’m convinced that CT scanners will become smaller, faster and more accurate, and I predict their eventual inclusion on high-speed production lines to deliver 100% inspection of all critical-to-life products. There is a real possibility that CT scanning may eventually replace the need for destructive testing, assumptive sampling regimens or conventional CMM metrology for certain applications.

And just in case you are still wondering, Lois Lane’s underwear was pink. Superman caught a sly glimpse when she stepped out from behind the plant pot.

This article was taken from issue 8 of Insight magazine. Get your free copy of the latest issue here.

About Brennan
Brennan is a senior consultant and has worked on a range of surgical and drug delivery products including dry powder inhalers, injectors and ophthalmic devices.

Join the conversation

Looking for industry insights? Click below to get our opinions and thoughts into the world of
medical devices and healthcare.