The corrosion of metallic structures is an industry- and government-wide maintenance problem that has been rapidly spreading due to the increased amount of infrastructure and military assets that are aging. However, even in the case of newer systems and components, corrosion can be a significant problem because of the harsh operational environments encountered.
Recognition of the severity and the economic impact of the corrosion problem by various industries and government agencies has led to significant effort over the past 50 years to prevent and control corrosion. Nondestructive evaluation (NDE) plays an important role in this effort by enabling the detection of early signs of corrosion so that corrective action can be taken before the damage becomes severe.
Hidden corrosion is a type of electrochemical material degradation that is not readily detectable visually or by any other surface measurement technique.1 It can often be detected and quantified in terms of reduction of wall thickness or structural discontinuities such as pits, flaws and voids. When attempting to detect material degradation due to electrochemical processes, the corrosion products (for example, iron oxides, aluminum oxides, etc.) must be identified so that an appropriate energy source can be selected for detection.
To inspect for hidden corrosion, the detection energy source must be capable of penetrating the material in which the corrosion is hidden.1 If the appropriate source is selected, then the returned signal will contain an evaluation of the entire material, including the physical geometry of the component or system, which may indicate its structural integrity and any hidden corrosion. Thus, the inherent technical challenges are to select the most appropriate interrogation energy source and to recover the signal that identifies the existence of corrosion.
Recovering the desired corrosion data is a mathematical inversion problem. Depending on the energy source used, the characteristics of materials and the corrosion hidden in structural systems, an exact solution of the inversion problem may not be feasible. Therefore, data analysis and information processing, such as the use of neuro-nets, have become key enablers in developing NDE techniques for hidden corrosion.
The U.S. military is considered the primary driver for the development of corrosion detection technology, while the nuclear, chemical, petroleum, and oil and gas pipeline industries are secondary drivers of this technology. This is due in part to the fact that military systems are typically fielded longer, have higher operational cycle rates and operate in more corrosive environments than commercial systems.2
Aging U.S. Army Department of Defense (DoD) assets have exacerbated the problem of corrosion and have increased the need for prevention, hidden corrosion detection and repair. The corrosion battle extends to essentially the entire spectrum of DoD systems, including surface ships, submarines, carrier and land-based aircraft, land vehicles and amphibious landing craft. As systems age, corrosion becomes one of the largest cost drivers in life cycle costs of weapon systems.
An example of this problem is the cables used for elevators on aircraft carriers (Figure 1). These cables are outside the carrier hull and are exposed to the extremely harsh corrosive environment. Due to the unavailability of NDE detection technology, these elevator cables are replaced on a time-based schedule every several years at a cost of hundreds of thousands of dollars per elevator.
Primary NDE Methods for Detecting Hidden Corrosion Guided Ultrasonic Waves
Guided ultrasonic wave NDE offers the potential for a cost-effective methodology for inspection of hidden corrosion in large and sometimes difficult-to-access areas, such as insulated piping. The field of guided waves has reached some degree of maturity, but unfortunately the number of practical applications compared to the number of research papers is rather small.
Guided waves can be used in three regimes, depending on inspection distance:
Short range (less than 1 meter)
Medium range (up to approximately 5 meters)
Long range (up to approximately 100 meters)
The short-range methods include high-frequency surface wave scanning with Rayleigh waves, leaky Lamb waves, and acoustic microscopy in which a leaky surface wave is generated by the lens. The medium-range methods typically use frequencies in the 250-kilohertz (kHz) to 1-megahertz (MHz) range and are applicable to plate, tube and pipe testing.
The long-range method generally uses long-range frequencies below 100 kHz, and the primary advantage is that it allows a large area to be tested from a single transducer location without tedious scanning. Long-range testing, which has the greatest potential utility in field-testing, is usually carried out in the pulse-echo mode.
All of the successful applications of long-range guided wave testing to date have been on structures with low feature density. This means that the coherent noise produced by multiple reflections between different features is modest. However, it would be valuable to be able to test more complex structures, such as aircraft fuselages, where the spacing between stiffeners is typically less than 300 millimeters (mm). Some initial work has been done on propagation in this type of complex structure which indicates that subtraction algorithms may be helpful.3, 4
Guided ultrasonic waves have been applied for inspecting pipes by using an array of transducers around a pipe to focus energy at particular positions on the pipe circumference at chosen axial locations.5 Although good results were obtained in a laboratory environment, this technique was not as successful in actual shipboard tests of piping where pipe elbows, hangers, flanges, valves and welds add significant complexity. However, this approach still has the potential to be applied to more complex structures if coherent noise can be controlled and algorithms can be developed to interpret waveforms.
Prior work has shown that guided waves also have some potential for inspection of multilayer aircraft structures for hidden corrosion and cracking.6 Figure 2 (a, b, c) shows guided wave waveforms from a three-layer aircraft structure of aluminum-adhesive aluminum. The first signal is from a noncorroded region (Figure 2a), the second is from a corroded region in the bottom aluminum plate (Figure 2b), and the third is from a corroded region in the top aluminum plate (Figure 2c).
Noncontact air-coupled transducers can be used to apply guided waves to the inspection of thinning in aluminum plates.7 In this application, a pair of micromachined gas (air)-coupled capacitive transducers is used for the generation and detection of guided plate modes.
Features in the dispersive behavior of selected guided wave modes were used to detect plate thinning. Mode cutoff measurements provided a qualitative detection of plate thinning, while frequency shift measurements were able to provide a quantitative measure of plate thinning. The experimental setup with air-coupled transducers is shown in Figure 3.
Noncontacting electromagnetic acoustic transducers (EMAT) can also be used to generate and detect shear horizontal (SH) guide waves for inspection and mapping of corrosion in pipe walls and plates.8 The SH waves have a pure shear-motion parallel to the surfaces and perpendicular to the plane of incidence.
SH guided waves have a unique feature in contrast to guided waves with inplane polarization; the lowest order mode has no dispersion and the dispersion of the higher order modes is much weaker than modes with polarization in the plane of incidence. As a result, SH guided waves could be economically and reliably used to detect and map corrosion in plates and pipes. Couplant-free excitation and the resultant simplified waveforms add to the versatility and usefulness of the technique.
A shorter range wave technique that utilizes creeping (or lateral waves) and head waves in parallel or near-parallel walled metal structures has been commercialized in the United Kingdom.9 The technique utilizes a transducer at the critical angle to generate creeping and head waves, and a second receiving transducer is placed up to one meter away.
The unique way in which the waves propagate provides complete isonification of the plate or pipe with little attenuation. This allows the probes to be well-separated compared to traditional creeping wave inspection. The technique is also applicable to corrosion in pipes under pipe supports, and the instrumentation has been field proven.
Confidence in ultrasonic inspection to detect and quantify corrosion in field applications has often required the disassembly of systems and testing in water baths. Results of various tests have shown that the detection of hidden corrosion on various aluminum alloys of varying thickness was useful above 10 percent metal loss, but the technique was not applicable for metal loss below 10 percent.
To improve the ability to detect hidden corrosion, there have been continued efforts to apply the dripless bubbler ultrasonic scanner, which is an ultrasonic technique that does not require a water bath and disassembly.10 This technique was selected as a primary candidate by the Air Force Logistics Center in Oklahoma City (OCALC) for the detection and quantification of intergranular corrosion prior to the onset of exfoliation around wing skin fasteners. This is a major inspection problem for aging aircraft. Figure 4 shows the results of tests for corrosion detection around wing skin fasteners.
In a similar nondestructive evaluation study, a novel ultrasonic pulse-echo technique was developed to detect intergranular corrosion around fastener holes in aluminum wing skins before the exfoliation stage.11 In this case, a focused transducer with a special fixture was used to overcome the typical problems: not enough couplant, transducer not perpendicular to the part, and varying transducer pressure. In general, there was agreement between the ultrasonic results and the results from the mechanical rework of the wing skin and dissection of the fastener hole.
Ultrasonics can also be applied without a water bath by using laser ultrasonics.12 Laser ultrasonics have been applied for the inspection of painted metal skin, and aircraft lap joints. When lap joint corrosion reaches a specific level, normally 10 percent of the nominal skin thickness, the section of the lap joint must be replaced. Visual inspection of the pillowing of the surface has been used to detect this type of corrosion, but it cannot supply quantitative information.
By using spectral analysis of the laser-ultrasonic waveforms, the residual metal skin thickness of the top skin of the lap joint could be determined. Results from standard samples with flat-bottom holes showed that the technique could detect metal loss below one percent of the nominal thickness of the metal skin. Comparison of the laser-ultrasonic measurements to X-ray images showed that the laser-ultrasonics has the same accuracy as the X-ray imaging (metal loss below one percent).
However, laser-ultrasonics as compared to X-ray imaging does not require disassembly of the structure, and the inspection could be carried out during routine maintenance.
Other ultrasonic approaches for detecting hidden corrosion involve ultrasonic imaging using dry-coupled probes.13 In addition, a commercial instrument that essentially operates as an ultrasonic camcorder can produce C-scan images from ultrasound signals, which are introduced into the sample with a large, unfocused commercial transducer.14 The implementation can be either in transmission or reflection. A water squirter or ultrasonic couplant on the surface is required for the application of this technique.
By combining obliquely backscattered ultrasonic signals with the sensor array real-time charge-coupled imaging system used in the ultrasonic camcorder, a rapid technique for detecting corrosion in aircraft skins without the need for paint removal has been developed.15 The system can produce 30 frames per second, and the unit can be programmed to examine time-of-flight bounds and, thus, produce 3-D images of material slices.
Figure 5 shows a schematic of the technique and Figure 6 shows a pulse-echo image of a corroded area in an aluminum plate. The combination of obliquely backscattered ultrasonic signals and the charged-coupled ultrasonic imager produces a viable aircraft corrosion inspection technique.
Eddy current NDE is a prime method for detecting hidden corrosion in electrically conducting materials. The method is based on generating a localized alternating current field in the sample using a probe coil. The same probe coil or another detector then measures the material's response to the induced eddy currents.
Defects cause a perturbation in the eddy currents, and this can be detected by a change in impedance or phase variation in the detecting circuit. In the past, eddy current NDE was limited in detecting hidden corrosion to shallow depths by the lack of penetrating capability of the eddy currents. However, over the last several years, there have been several notable advances in eddy current NDE stemming from research to detect hidden corrosion and other types of defects.
When combined with magnetoresistive sensors, eddy currents can be used to detect corrosion in second and third layers in aircraft lap joints.16 High-performance magnetoresistive sensors are more sensitive than standard pickup coils used in conventional eddy current NDE, and they operate at much lower frequencies. Their low-frequency range and linear response to frequency allows for greater depth of inspection than with conventional eddy current NDE that operates at higher frequencies.
Pulsed eddy current NDE technology has also made recent advances.17 Its robustness to variations in geometry, paint thickness, rivet heads, surface warping and sensor liftoff allows this technology to become a major method for detection of hidden corrosion.
Conventional eddy current instruments measure the impedance and reactance of the detection coil, while the pulsed eddy current instrument measures the transient voltage signal with a frequency spectral content from direct current to 100 KHz or higher. This broadband characteristic provides the ability to conduct digital signal analysis and to extract subtle differences in waveforms associated with small geometry variations.
Hence, it provides the ability to quantify a large number of parameters. For example, material loss at or near the surface will have more of an effect on the early time portion of the total transient interval than on the late-time portion. The opposite will be true for deep defects. Remote field eddy current is another variation of conventional eddy current that has found widespread use in detecting hidden corrosion.
Its capabilities have recently been expanded and automated.18 The remote field eddy current technique is based on the measurement of the voltage induced in a pickup coil by flux which has passed twice through the test piece (as shown in Figure 7). Measurement of the phase change caused by a discontinuity has a linear relationship with changes in thickness of the pipe wall.
Remote field eddy current is not sensitive to liftoff, but an ultra-sensitive eddy current system is required to handle the low amplitude signals obtained from remote field eddy current probes.
Magneto-optical imaging of eddy currents is another innovative advance in eddy currents applied to detection of hidden corrosion.20 Magneto-optical imaging of the eddy currents generates real-time images of large surface areas to quickly detect subsurface corrosion and cracking. Although eddy current techniques do not have as much sensitivity and spatial resolution as ultrasonic NDE techniques, the ability to detect corrosion in multiple layer structures and at greater depths is advantageous.
Thermography provides a fast, noncontact technique for detection of subsurface corrosion. Thermography is based on using the thermal differences between a material and material defects to image discontinuities such as corrosion. It has demonstrated ability to detect corrosion under paint without paint removal. In addition, advances in commercial instrumentation and algorithms, particularly for thermal wave techniques21, have substantially enhanced field applications. Moving linear heat sources have also enhanced thermographic capability.
Thermal wave systems operate by sending a pulse of heat from the surface into the material, where it undergoes thermal reflection at either the rear surface or at thermal impedance changes (for example, corrosion). The effect of these thermal reflections is that it modifies the local cooling rate at the surface. The cooling rate, in turn, is monitored through its effect on the IR radiation from the surface. The IR radiation is detected by an IR camera and processed as a sequence of images by the control computer.
A number of other heat source techniques applied to thermography to detect corrosion have been demonstrated. Hot air guns as heat sources for thermography (named fan thermography) have been shown to detect corrosion in aircraft structures.22 A quartz line-shaped heat source (3,000-watt quartz lamp with an elliptical reflector) moving at speeds of 2.5 to 12.2 centimeters per second has also been demonstrated to provide a back surface profile.23]thickness could be determined to within five percent of the actual thickness. Thermographic NDE techniques are not currently in widespread use. However, the rapid inspection times for broad areas of coverage, availability of instrumentation, and ease of use should quickly lead to a wider use in the field.
In the past, NDE of hidden corrosion was a challenge. However, in the last decade, substantial advances in NDE methodology and software have made the challenge much more tractable. For example, eddy current can now detect corrosion in second and third layers of aircraft lap joints. Pulsed eddy current NDE, low-frequency eddy current NDE with magnetoresisitive sensors, remote field eddy current NDE at low frequencies using Hall probes, and magneto-optical imaging of eddy currents have greatly increased the tools available to detect hidden corrosion.
Thermography has also undergone great advances in the past decade, and it is now widely used to detect corrosion under paint. Advances in commercial instrumentation and software, particularly for thermal wave imaging, have enhanced field applications. Moving linear heat sources and algorithms have also broadened the field applications of conventional thermography. These advances in thermography, coupled with wide field of view and rapid inspection times, will result in the continued expansion of field applications.
In the past, ultrasonic techniques were limited mainly to inspecting disassembled systems in a water bath. However, advances in ultrasonic methods (for example, the dripless bubbler) have now expanded the applications for detecting corrosion. In addition, laser ultrasonics should continue to find niche field applications as it has over the past decade, but the rate of market penetration will most likely remain modest due to the complexity of the technique.
Guided wave ultrasonics has been shown in a number of research settings to be capable of inspecting large areas for hidden corrosion in difficult-to-access areas. However, transition of this capability, with its complex analysis, to commercial applications is expected to continue to proceed slowly.
Other NDE techniques, such as X-ray radiology and optical techniques, should continue to find niche field applications. In summary, the future looks bright for existing techniques and for improving field NDE capability to detect hidden corrosion.
This article is reprinted with permission from AMMTIAC and the American Institute of Physics. It was originally published in The AMMTIAC Quarterly, Volume 2, Number 2.
* Shadow refers to the shadow of the fastener countersink.