This post is also available in: Español
There are currently three main competing methods of non-destructive inspection (NDI) for infrastructure:
— Gamma rays from cobalt-60 sources
— Hard X-rays generated by linear accelerators (linacs)
All have been well understood for years. However, all have serious drawbacks that prevent them from being very effective, and except for cobalt-60, from entering into widespread use.
Gamma rays. Scanning with cobalt-60 sources, which emit gamma rays at 11 and 13 MeV, is the most common method of infrastructural inspection, and there are many suppliers for the cobalt-60 sources. Sources are placed on one side of the structure to be inspected, with a film or other detector on the other side. The main advantage of the approach is that it is cheap, requiring only the purchase of radioactive material and a detector (or just film). In addition, the gamma rays are highly penetrating. However, the sources are quite weak, producing typically 6×1010 photons per second. This limits achievable resolution. As a result, incipient cracks and voids can be missed.
The most important disadvantage of cobalt-60 sources is that they require the detector or film to be placed on the opposite side of the structure to be scanned. For above-ground pipes, this is not much of a problem, but for buried lines and bridges, positioning the detectors can be difficult or impossible. Roadway scanning is impossible.
X-rays from linacs. The principal alternative method to cobalt-60 for producing penetrating radiation for non-destructive testing and inspection employs a linear accelerator to produce an electron beam. The beam, colliding with a solid target, then generates the X-rays. Varian Corp. is a leading producer of such devices, whose main use is for medical radiation treatment. Individual machines produce hard X-rays ranging in energy from 1-15 MeV. Pulses, lasting approximately a microsecond, are generated at a 300 Hz rate. Devices sell for $400,000 and up.
Linacs used for infrastructure inspection share the same drawback as cobalt-60 in that they require the source and detectors to be on opposite sides of the object being inspected. In addition, linac-based devices are too expensive for most end users. As a result, Varian reports that only about 150 of the devices are in use worldwide, and most of these are used by US Dept. of Defense for inspection of tanks and other armored vehicles, not infrastructure. On the other hand, linacs have the significant advantage over cobalt-60 that they produce far more photons per second, approximately 5,000 times more. This increases resolution, which can approach a couple of mm.
Radar. In this method, used for road inspection, radar signals are reflected off of sub-surface layers, and the return signal is analyzed to detect road layer thickness. A number of small companies, such as Geophysical Survey Systems, Inc. supply truck-mounted systems for highway inspection. Radar does not penetrate steel, so cannot be used for inspection of metal structures such as bridges and pipelines. However, for roadway inspection, radar has significant advantages. It is very fast, with scans being produced at highway speeds. In addition, of course, detector and source are on the same side of the object being scanned. The main disadvantage of the radar system for road inspection is that the return signals are difficult to interpret. Highway departments that have tested such systems in an effort to detect roadway separation and surface voids have found that many conditions, such as changes in material composition, or just minor dampness, change the reflection of signals, and interpretation requires a high level of skill and long training. As a result, only a few dozen of these inspection systems are in use.
In addition, radar does not provide high resolution, with typical resolutions being 3 cm.
Incipient faults can thus be overlooked when they are easiest to fix.
X-Scan advantages. The X-Scan hard X-ray source overcomes the limitations of competing techniques due to its power, adjustability and low cost. At its intended output of 100 J at 300 keV photon energy, the source will have sufficient capability to scan deeply into structures using Compton backscattering. The return signal is backscattered, so that both detector and source are on the same side of the structure. This overcomes the principle limitation of both cobalt-60 and linacs. X-Scan has sufficient power for one-sided scanning of roadways, bridge structures, reinforced concrete girders and other large infrastructure components. In contrast, existing portable X-ray sources simply have too few photons per pulse for such a technique. Even if longer exposures are made, with multiple pulses, backscattered radiation falls below background levels with conventional sources and is not observable. But the powerful and fast DPF pulses far exceed background levels. The high per-pulse energy of the DPF source will generate 10,000 times more photons per pulse than conventional linac sources. With far shorter pulses than linacs, there will be ten million more photons per unit time.
X-Scan will have less energy per photon than linacs, and thus each photon will have only a third the penetrating power of linac-generated photons. However, the far greater number of the X-Scan photons completely overwhelms this difference. The net result is that X-Scan has enough photons returned from a scanned object to use Compton scattering, while linacs do not.
Unlike radar, which cannot penetrate metal, the hard X-rays our source will produce can scan through concrete and metal. In addition, the scans produced will be very straightforward to interpret, in sharp contrast to the difficulty in interpreting radar signals. Since voids created by corrosion, cracking or layer separation will backscatter essentially no X-rays, such voids will show up clearly as shadows in the final image. X-Scan, with 2mm resolution, matches the resolution of linacs and far exceeds that of cobalt-60 or radar.
Using our unique and adjustable approach, the DPF source will be able to scan an object with different energy X-rays, ranging down to as low as 30 keV, which will allow more depth information and detailed scans of near-surface features. Adjustability of the photon energy is extremely useful for inspection of structures, since it allows the inspector to focus in on small surface features once deep corrosion has been detected with higher energy. It also allows the user to select the energy appropriate for the structural part being inspected–higher energy for large parts, lower for smaller ones, maximizing resolution at each step. Linacs cannot produce the adjustable output of X-Scan. Equally important, the DPF X-ray source will make possible a more economical NDI device, reducing costs and allowing agencies responsible for infrastructure inspection to purchase them. Interviews with highway inspection managers indicate that prices at around $100,000 per unit would be acceptable, in contrast to the much higher linac prices of $400,000, which are unaffordable.