Twin roles of NDT

The fundamentals of non destructive testing

Non-destructive testing (NDT) is the process by which materials, components or assemblies can be inspected without affecting their ultimate serviceability. This offers some important advantages over destructive testing processes such as mechanical testing. A destructive test must necessarily be on a sample basis. This is acceptable in many cases: to check that a batch of material is within specification, for example, or as a sample check on a large number of identical components. But a sample destructive test may not be feasible for a small number of high value components and in critical cases a 100% inspection may be required. Because of this, one key role of NDT is in assuring the quality of many manufacturing processes.

NDT can also be used to test components which have been subjected to service conditions. A welded structure can be inspected to detect fatigue cracking for example, or the wall thickness of a pipe can be measured to check for internal corrosion or erosion. So a second key role of NDT is to monitor for in-service degradation. In this, the NDT results can further be used to aid decisions on the need for and timing of repairs. Suitable techniques, correctly applied, can be used to measure the depth of a fatigue crack for example. This, coupled with a knowledge of the system stresses, the toughness of the material and the fatigue crack growth rate, can be used to determine if a cracked component can be left in service until the next planned overhaul period.

These twin roles of NDT ensure that the technology will continue to play a fundamental role in materials testing. This article describes the main methods used in NDT and describes some typical applications.

Radiography
The radiographic method is probably the oldest of all the NDT methods. It is analogous to medical radiography in that the image is formed by variations in the absorption of penetrating radiation within the subject. Engineering materials, however, do not in general exhibit the wide contrast found in the human body and this limits the usefulness of the method. Volumetric defects such as slag and porosity in welds are revealed well but the detection and sizing of cracks is more difficult. Because of this, the main use of radiography is in checking the quality of components or fabrications, or revealing the internal structure of components such as valves (Figure 1). 

Fig 1:A portable flaw detector being used for the ultrasonic inspection of pipework.

Radiography is carried out using X-radiation or, more usually in site situations, gamma-radiation from radioactive isotopes. For many years, film was the only medium for recording and storing radiographs but nowadays the technology is available to record, view and store radiographs in digital form.

Surface defect detection
Surface inspection methods are similarly long-established forms of NDT. Liquid penetrant inspection is the simplest. A coloured or fluorescent dye is sprayed onto the surface of the component and is drawn into surface defects by capillary action. An absorbent developer is then sprayed onto the surface and this draws the dye out to reveal the defects. On a ferromagnetic component, magnetic particle inspection is possible. Here, the component is magnetised using a coil, electric current or an electromagnet and a ‘magnetic ink’ is sprayed onto the surface. This contains coloured or fluorescent magnetic particles which are attracted to surface discontinuities.

Both of these methods are used extensively in automated or semi-automated forms for the testing of large numbers of components in the aerospace or automobile industry for example. Magnetic particle inspection is very widely used for the in-service inspection of power stations and chemical plants where many of the components are made from ferritic steels and where fatigue cracking from the surface is a possible defect of concern.

Ultrasonic testing
Radiography and the surface inspection methods probably account for the majority of industrial NDT carried out today. But they lack the ability to detect and measure the size of planar defects such as cracks within the bulk of a component. That deficiency is met mainly by the ultrasonic method. Here a beam of ultrasound, with a frequency typically between 1 and 5MHz, is directed into the material from a piezoelectric transducer. Discontinuities within the material reflect the ultrasound back to the transducer in rather the same way that sonar is used to detect a submarine.

Much ultrasonic testing is carried out by manual techniques on industrial plant, often under difficult conditions. The ‘flaw detectors’ used to generate and display the ultrasonic signals have become progressively smaller and lighter as the technology has developed, with digital processing and displays now common (Figure 2).

Figure 2:An automated ultrasonic system is used to inspect a power station component.

The simplest display shows signal against time (which can be calibrated to give range). Interpretation is a skilled process, with the operator having to build up a mental picture of a possible defect as he interrogates it with ultrasonic beams from a variety of directions.

Increasing demands for reliable ultrasonic inspection, mainly from the nuclear and aerospace industries, have accelerated developments in ultrasonic testing. Automated inspection systems are now common, in which several ultrasonic transducers are scanned across a component.

The system will log the ultrasonic signals and the transducer positions. A data analysis program is then used to plot the results effectively on an engineering drawing of the component, with the capability to view the results from different directions, take ‘slices’ through the component and zoom in on interesting areas. These automated systems offer permanent storage of all the inspection data, unlike manual techniques, and they provide high reliability in coverage of the component and offer the possibility of accurate repeat inspections. Skilled personnel are still required but the skill now lies in the interpretation of the data.
There are also developments in transducer technology, with the increasing use of phased arrays being an important feature. A phased array transducer contains an array of piezoelectric elements, rather than a single one, and the system can be programmed to apply varying delays to individual elements. This allows the ultrasonic beam to be scanned, steered or focussed. This reduces the number of transducers needed for an automated inspection and permits the rapid scanning of large areas (Figure 3).

Figure 3: Image of a carbon fibre aircraft component produced by a large area ultrasonic scan using a phased array system

Electrical methods
If an electric current is induced in a conducting component, its amplitude and distribution will be affected by discontinuities in the component, geometry and material properties. This is the basis of a number of NDT methods, the best known being the eddy current method. In this method a coil carrying an alternating current with a frequency typically of a few hundred kilohertz is placed adjacent to the component under test. Eddy currents are induced in the component and the impedance of the coil varies as the eddy currents vary. The variation in the eddy currents may be caused by discontinuities in the component (offering the possibility of crack detection), by variations in the electrical properties of the component (offering the possibility of materials sorting or detection of variations in heat treatment) and by the spacing of the coil from the surface (offering the possibility of measuring the thickness of non-conductive coatings). The main problem in carrying out eddy current testing lies in separating these competing effects and that is done by careful design of the coil and the test equipment. The equipment usually displays the impedance as a vector plot, allowing the effects of phase and amplitude to be separated. Because the effects usually vary differently with frequency, the equipment may be capable of mixing two or more frequencies in a way which cancels out the unwanted effects.

Because of the skin effect, the eddy currents are usually confined to a surface layer of a few mm at most, making this a surface or near-surface technique. Typical uses of the technique are for the detection of cracks around fasteners in aircraft structures or, using circumferential coils, for the detection of defects in tubing, either during manufacture or in service, for example for the in-service testing of condensers or steam generators on power stations. Eddy current systems are also widely used for the in-process inspection of components (Figure 4).

Figure 4:Automated eddy current equipment used for the inspection of aircraft wheels

Other electrical techniques have been developed. In one example, a proprietary instrument makes use of what is termed the ACFM (for ‘alternating current field measurement’) technique. An AC field is induced in the surface of the component and this in turn induces a magnetic field above the surface of the component. The horizontal and vertical components of the magnetic field are measured and used to determine the length and depth of surface-breaking cracks. To do this, the instrument incorporates a mathematical model of the electrical and magnetic fields around a crack.

Other methods
The methods described above are the most common but there are many others. Visual inspection, with or without optical aids, is an obvious but often overlooked method. Acoustic emission is a method used to analyse structures, in which crack growth, local yielding or the fracture of corrosion products may result in the generation of elastic waves which are detected and located by an array of transducers. Leak testing is another important method, utilising a variety of different techniques. Laser techniques may be used to analyse strains and detect discontinuities in structures. This list is certainly not exhaustive.

Research and development
Research and development is clearly an important contributor to NDT, both in developing new methods and in improving the reliability and usability of existing ones. Equipment manufacturers are active in this area, of course, but there is a role for universities in developing the more advanced ideas. Examples are improvements in transducer design, signal processing to extract useful information from noisy data and the use of neural networks which can be trained to carry out the automatic recognition of defects. The UK Engineering and Physical Sciences Research Council recognised the role of universities in NDT research and development when, in 1993, they funded the creation of a Research Centre in NDE to link university research groups and industrial NDE users. This comprises a consortium of universities and a management board including industrial and university members. Further details may be found on www.rcnde.ac.uk.

Further information
This short article can only give an overview of the wide field of NDT. A good starting point for further information is the web site of the British Institute of Non-destructive Testing (www.bindt.org).

By Dr Robin Shipp, Firecrest Consulting, Bristol
enquiry number 05102

Figure captions
1. X-ray equipment being set up to inspect a large engine component.
GE Inspection Technologies.

2. A portable flaw detector being used for the ultrasonic inspection of pipework.
GE Inspection Technologies.

3. An automated ultrasonic system is used to inspect a power station component.

4. Image of a carbon fibre aircraft component produced by a large area ultrasonic scan using a phased array system.
NDT Solutions Ltd.

5. Dual-frequency eddy current equipment being used to inspect fasteners on an aircraft structure.
Staveley NDT Technologies.

6. Automated eddy current equipment used for the inspection of aircraft wheels.
GE Inspection Technologies.