State-of-the-art systems currently in commercial deployment for corrosion detection.
Current state-of-the-art (SoA) NDT techniques for detecting the presence of corrosion on metallic surfaces, often suffer losses in effectiveness, or are even rendered completely ineffective, by the presence of protective coatings such as paint. Even though the NDT techniques employed in detecting CUP are mature, they still tend to be highly labour intensive and require extensive preparation beforehand. There is therefore a continuing drive to increase the speed of inspection, to reduce the preparation required and, if possible, to inspect without the need in some cases, to shut down operation. With some notable exceptions, it is possible to find most types of defect. However, it is frequently not cost effective to do the inspection, so driving down overall inspection costs is a major research and development aim in the NDT industry. In the SoA overview below, the suitability for detection of corrosion under paint is evaluated for each of the current inspection techniques.
1. Visual NDT. Visual and optical testing is by far the most common NDT method of corrosion detection. An experienced inspector equipped with basic tools as a magnifying glass and a hand-held torch can conduct a effective inspection. For difficult to access areas, a boroscope (miniature camera attached to a fibre-optic cable) can also be used. In maintenance operations, visual testing and optical inspection is usually the first line of defence. After initial detection, secondary follow up techniques include Electronic Speckle Pattern Interference (ESPI) for early detection of de-cohesion of paint coatings and optical surface topography systems such as laser interferometry can be used. Such enhanced visual/optical methods mainly detect and measure deformations on surfaces which are caused by sub-surface corrosion damage such as blistering and pitting. Suitability for CUP Detection: Visual NDT inspection does not use complicated equipment and is the usual first step in the detection of CUP by the location of porosity, bead contouring and blistering. The limiting factor is the resolution ability of the human eye which has difficulty determining porosity diameters less than 0.25 mm or cracks less than 0.025 mm wide. Also, objects nearer than 150 mm - 250 mm are hard to focus on and lighting conditions must also be adequate for good visibility. Only surface discontinuities can be detected with this method and another major disadvantage is that the surface must be cleaned of contaminants to prevent masking of those surface discontinuities.
2. Ultrasonic Inspection. Ultrasonic NDT is a very broad area but essentially most ultrasonic inspection techniques use frequencies in the range (1-10) MHz, and as the propagation speeds of various wave modes are a function of the material’s various elastic moduli and density, this make ultrasound a very powerful technique for materials characterisation. Strong reflections occur at boundaries where material properties change make the technique highly appropriate for thickness measurements and crack detection, both of which could be manifestations of corrosion.
Suitability for CUP Detection: Near surface ultrasonic inspection methods (such as Lamb or Rayleigh) could work in most instances but usually require full intimate contact and impedance-matching coupling with the inspected surface which is a very time-consuming process . Moreover, reasonable specification ultrasonic flaw detectors can cost anything upto €20K and require many days of expensive operator and/or contractors time to cover even small areas.
3. Magnetic Particle Inspection (MPI). A magnetic field is applied to the painted metal surface using a permanent magnet, electromagnet, flexible cables or hand-held probes. If the material is homogenous then, most of the magnetic flux is concentrated below the material's surface. However, if any material discontinuity is present, the magnetic flux is distorted and locally 'leaks' from the surface in the immediate region of the defect. Fine magnetic particles (usually powder) which are applied to the surface of the specimen are attracted to the area of flux leakage producing a visible indication of the flaw. The powders are usually black iron particles and red /yellow iron oxides. In some instances, the iron particles are coated with a fluorescent material to allow them to be viewed under a UV lamp for even easer detection.
Suitability for CUP Detection: magnetic field-gradients associated with corrosion pitting on the surface of steel are usually far smaller than the magnetic field gradients associated with cracks and it is a well-observed empirical fact that even with cracks, magnetic particle methods are of little use when said surface cracks are covered in paint thicker than about 75 microns and most methods need a supply of electricity and requires a skilled operator for interpretation of most measurements.
4. Traditional Eddy Current Techniques. Eddy current testing (ET), uses alternating currents applied to a conducting coil held in close proximity to the object under inspection. In accordance with Lenz’s Law, the inspected object generates eddy currents to oppose the alternating current in the coil. The eddy currents are then detected by the same coil, separate coils, or magnetic field-sensors. Changes in the induced eddy currents may be caused by changes in a material’s electromagnetic properties, variations in material thickness or the sharp discontinuities caused by the presence of cracks. Again, these are all phenomena which may also be associated with the presence of CUP. Current state-of-the-art ET apparatus is now extremely portable and ET It is the second most common method specified for the routine NDT of aircraft. Recent technological advances in this field which have rapidly increased both efficiency and speed of detection, are multichannel sensors and the replacement of search coils with Giant Magneto-resistive (GMR) magnetic field sensors (originally developed for computer hard drives).
Suitability for CUP Detection: On aluminium surfaces for instance, using conventional scanned eddy-current transducers is time consuming and ineffective as variations in paint thickness overlaid on a corrosion-free section often produce “lift-off” signals which are essentially indistinguishable from signals produced from paint layers of uniform thickness which do conceal areas of corrosion on the underlying surface. Large surface areas are also a problem and need some kind of area scanning device which can be expensive and also very limiting in terms of speed of scan. Also the more complicated the geometry of the surface becomes, the more difficult it is to distinguish between real defects and corrosion.
Suitability for CUP Detection: In general, eddy current techniques do not have as much sensitivity and spatial resolution as say ultrasonic NDE techniques and can only inspect relatively small scan areas. Although remote-field eddy current scanning is not sensitive to liftoff - the downside is that an ultra-sensitive eddy current system is also required to handle the low amplitude signals obtained from remote field eddy current probes. Magneto-optical imaging of eddy currents shows great promise but the current state-of-the-art with this approach is that it is far more sensitive to cracks below paint than the actual surface corrosion itself Another disadvantage is that these sensors are based on a rigid substrate and impossible to bend and so cannot be used very easily to detect corrosion and cracks under paint on structures having curved surfaces which makes it unsuitable for inspection of structures such as bridges and piping.
6. Thermography. Thermal imaging cameras are the most common sensing devices used for this technique. However, passive imaging will not normally reveal areas of CUP if there are no leakages or other localised sources of heat and the area under inspection is in a state of uniform thermal equilibrium with its surroundings, meaning that no hot or cold spots are visible. In order to be able to detect CUP, external heating or cooling is applied in the form of the short energy pulses. Flash thermography techniques usually use quartz lamps and thermal perturbation is then followed by a differential time-resolved infrared image analysis. Coating defects such as blistering and sub
surface corrosion spots are detected in infrared images due to the differences in the thermal diffusivity between the defective and non-defective areas. A temperature difference of a fraction of a degree is sufficient for reliable detection of CUP. Another significant recent advancement in thermal imaging to detect CUP is Thermal Stress Analysis (TSA) which uses mechanical energy (in the form of vibration) to stimulate very localised heating at sub-surface discontinuities, such as corrosion-induced cracks in metals. This has opened up a new field of application for the IR method.
Suitability for CUP Detection: Thermographic inspection has the advantage that it has the potential for large areas to be inspected in real-time and thermal wave imaging techniques are a very promising prospect for the detection of CUP. However, if water is present under a layer of paint which is not actually covering corrosion, false indications occur. Major limitations are that if no “hot spots” or leakages are present then the temperature differences that produce thermal image contrast are too small and any CUP features can become indistinguishable from good regions. To overcome this, it is necessary to induce localised temperature differentials and the accessories required for flash thermography (halogen lamps and hot air guns) are relatively large and cumbersome, restrict inspection to relatively small areas and can also restrict access in difficult to reach places. The technique is quite costly and the purchase price of thermal imaging equipment also tends to be relatively high and often outside the reach of all but large organisations and government agencies.
7. Radiography. Current state-of-the-art radiography used in NDT falls into two broad categories: x-ray radiography and isotope radiography. Both types of inspection have the advantage that they can detect both surface and hidden flaws and that results can be obtained with the minimum amount of sample preparation. The range of detectable flaws extends to some of those surface features associated with corrosion and which may also be covered by layers of paint, such as pores, inclusions and cracks.
Suitability for CUP Detection: X-ray radiography requires bulky and expensive equipment, presents safety hazards to the users and other personnel and is highly directional and sensitive to flaw orientation. Moreover, it takes a high degree of skill and training on the part of the operator. Isotope radiography is less expensive than x rays and inherently more portable but its use is heavily legislated and requires a high degree of safety and legal compliance and also requires highly trained and experienced operators. Another major disadvantage with regards to detection of CUP is that neither of these two radiographic techniques can be readily used in the back-scattered (reflectance) mode which is necessary for single sided inspection of painted surfaces. Cost is another issue portable x-ray equipment can be costly and spare parts such as x-ray tubes can cost several thousand euros at a time to replace.
8. Microwave NDT. State-of-the-art microwave NDT inspection is currently a niche market and theoretically microwave methods should offer several advantages for detection and evaluation of corrosion under paint: they can penetrate low-loss dielectric materials and are sensitive to changes associated with variations in dielectric properties and boundary interfaces – all of which would suggest that they are suitable for detecting the presence of layers such as corrosion under paint. Moreover, microwave techniques are non-contact, one-sided (i.e. reflection) and relatively fast as compared with other NDT techniques.
Suitability for CUP Detection: Despite great early promise, it was found that detection of CUP based upon comparison of the amplitude and/or phase of a reflected microwave signal with an initial calibration value was highly susceptible to errors caused by variations in paint thickness, and sensor drift. The sensitivity to these errors is because the mode of operation is based on the pressing up of the horn antenna against (and in direct physical contact with) the painted surface and the formation of a standing wave between the microwave transceiver and metal surface being inspected. The presence of dielectric materials such as corrosion or paint shifts the standing wave, causing a change in the amplitude and phase of the response. Unfortunately, shifts can also result from changes in the effective altitude above the metal surface caused by variations in paint layer thickness and the effects of dielectric layers such as corrosion and paint, and altitude above the metal surface are difficult to separate. Also, being a relatively low-volume product, units can be difficult to source and tend to be costly as compared to the more common types of NDT equipment.
The Capacitance Imager overcomes the shortcomings of existing SoA technologies and is essentially unaffected by any of the factors which inhibit and limit the performance of all the existing NDT technologies currently used in the field today.
This new NDT system represents a truly disruptive technology which goes well beyond the state-of the-art in its ability to readily detect the presence of both deep sub-surface hidden corrosion as well as even minor occurrences of surface scaling and corrosion under paint. Moreover, the system operates at area scanning rates which are orders of magnitude more rapid than existing state-of-the-art and at a fraction of the cost of existing systems. (It is even possible that one embodiment of this CI new system could also work in a stand off fashion underwater and would for instance, do away with the expensive and time consuming need for ocean-going vessels to go into dry docks to be inspected).
The overall innovation in this system is that it is unaffected by factors inhibiting and limiting existing NDT technologies with scanning rates which are orders of magnitude faster than scanning than current state-of-the-art and unlike ultrasound, eddy currents and radiography etc, does not require intimate contact/ fixed distance with the surface itself. Example
For example, one of the fastest non-contact scanning rates for the detection of corrosion under paint is Eddy Current Array (ECA) scanning. One such device is an Olympus “ECA SAA-112-005-032”. This is a 32 element array probe with one of the largest commercially available scanning heads (112 mm wide). It scans at 5mm distant from the surface at an average linear scanning rate of 30 mm per second. This equates to scanning a 1m2 area every five minutes (306 seconds) .
In comparison, a CI unit is immune to small variations in stand-off distance and irregular surfaces and scans at a rate at least 10 times faster. frame every two seconds.
Overview of Operation Front-sided Capacitance Imaging
The basic configuration is to “open up” a standard parallel-plate capacitor so that both electrodes are now in the same plane i.e. co-planar capacitor. An AC voltage is then applied and an electric field distribution is established. An object between the electrodes will affect this electric field pattern and any localised change in sample properties will change the field distribution pattern and modulate the output signal. Correlating these changes in signal with localised surface information will therefore produce an image as the probe is moved over the surface of the object. The major practical advantage of this approach is that it is non-invasive, non-contact and only requires single sided access to the object being examined.
Co-planer copper electrodes are simple to fabricate via standard printed circuit board (PCB) techniques. Aside from geometry considerations, probe performance is also determined by other factors such as the gap between the electrodes and stand-off distance between the probe and the surface of the inspected object. Optimum probe design for any particular application is derived from both analytical models and the results of experimental trials. Overall, and as a general rule, it is found that field penetration depth is determined mainly by the geometry of the electrodes and that there is a trade-off between probe size and image resolution. And the major practical advantage for NDT applications is that it is completely non-contact and requires access to only one side of the inspected object.
Electric field distribution for a metal sample covered with an insulating layer of paint. (b) a metal surface containing a notch or deep crack, and (c) a simulated delamination of the insulator from the metal of a type typically associated with corrosion.