NDE may be used to reveal defects that would be difficult or impossible to detect by visual examination. The techniques are used during manufacture as a quality control tool to determine the quality of the work. The extent of NDE depends upon the application and the criticality of the joint and is generally specified in the relevant application standards or contract speci­fication. It is important for NDE to be included in the planning of the fabrication process as it can require substantial time and resources. Full account of this must be taken if disruption of production and delays to the programme are to be avoided.

The requirement to perform NDE must also be taken into account during the design phase. As with welding, access for NDE must be planned into the component. The implication of this is that both welding engineers and designers must be conversant with the techniques and their limitations if the processes are to be used effectively.

11.3.1 Penetrant examination

This is a technique that is capable of detecting surface breaking defects only. It relies upon a coloured or fluorescent dye, sprayed upon the surface, penetrating these defects. After cleaning the excess from the surface, the dye within the defect is drawn to the surface by spraying on a developer in the case of the colour contrast dye or by exposing the surface to ultra-violet light. The defect is revealed by the dye staining the developer or by fluorescing (Fig. 11.9).

Figure 11.10 is a photograph of a typical penetrant examination indica­tion. The fluorescent dye gives greater sensitivity than the colour contrast dye and does not require the use of a colour contrast developer but does require the use of an ultra-violet light and preferably a darkened room. The cleaners, penetrant dyes and developers can all be obtained in aerosol cans, making the process extremely portable and ideal for site use.

The dye used as a penetrant must be capable of penetrating narrow cracks but must not be removed from more open defects during the clean­ing operation carried out before the application of the developer. The dye must have a high contrast with the developer. It is important that the test piece is thoroughly pre-cleaned — any dirt, oil or water in the crack may prevent the penetrant from entering. Degreasing should be carried out by swabbing or immersing the item in one of the proprietary cleaners, acetone or methanol. Immersion in an ultrasonic cleaning bath is probably the best

11.9 Penetrant examination principles.

11.10 Liquid penetrant test result illustrating staining of the spray-on developer by a defect in the HAZ. Courtesy of TWI Ltd.

method. Wire brushing or grinding should not be used unless it can be fol­lowed by an acid etch as mechanical methods of cleaning can smear over defects and prevent the dye from penetrating.

Inspection in other than the flat position is difficult but penetrants have been developed with a jelly-like consistency that can be used to carry out inspections in the vertical and overhead positions. Automated methods may be used, with the components loaded into baskets and processed on a con­veyor line. The fluorescent dyes are better in this application than colour contrast dyes because of their greater sensitivity.

Sensitivity of the process can be checked using standard test blocks. For the examination of aluminium components, these are available in 2024 alloy heat treated to give real cracks of a standard size. These blocks should be scrupulously cleaned after each check to ensure that the cracks do not become clogged with debris.

Although the technique is simple to use, interpretation of the results can be difficult, particularly if the surface is ‘naturally’ rough or if the dye is trapped in acceptable geometric features. Operatives should therefore be trained and, for many tasks, a qualification in penetrant examination is either an application standard or contract requirement. Both the British Institute of NDT (BINDT) and the American Welding Society (AWS) run accreditation schemes.

There are few health and safety risks involved in using the technique. The cleaners and some of the solvents in which the dye and developer are dis­solved will cause skin irritation if used with unprotected hands, and gloves are strongly recommended. The cleaners and solvent vapours will also need to be controlled if the process is used in confined spaces. Some of these materials are also flammable, so there are fire and explosion risks. Advantages:

• It can be used on both ferrous and non-ferrous metals.

• It is very portable.

• Large areas can be examined very quickly.

• It can be used on small parts with complex geometry.

• It is simple, cheap and easy to use and interpret.


• It will only detect defects open to the surface.

• Careful surface preparation and cleanliness are required.

• It is not possible to retest a component indefinitely.

• There may be health and safety problems with some of the chemicals.

• There are health and safety problems with fumes in confined spaces.

11.3.2 Eddy current examination

Eddy current examination is a process that may be used on any material that will pass an electric current. A coil carrying an alternating current is placed close to the item to be examined, inducing an eddy current in the specimen. Defects in the specimen will interrupt this eddy current flow and these perturbations can be detected by a second, search coil. The coils can be placed either side of a thin plate-like sample or can be wound to give side-by-side coils in a single probe. These may be shaped to fit in the bore or around the outside of pipes and tubes and in these applications the process lends itself to automation. The equipment is calibrated using a defect-free specimen. The accuracy can be affected by metallurgical condition, stand­off and coil dimensions. For these reasons eddy current testing is used only rarely on welded components, although it is excellent in examining contin­uously welded tube from pipe mills. The process has been developed over recent years to make it more portable and simpler to use. Microprocessor — based control and recording units, improved and more tolerant probes and light-weight electronics have enabled the technique to be used on-site for the examination of structures in service, where it is an effective tool for the detection of cracking and corrosion problems. It is also possible for the depth of surface cracks to be determined. It is of limited use for interro­gating welds, however, being most commonly used in the examination of continuously welded tube, and is not covered further in this chapter.

11.3.3 Ultrasonic examination

The ultrasonic examination of welds uses the same principles as when sonar is used for the detection of submarines. A ‘sound’ wave emitted from a transmitter is bounced off an object and this reflection captured by a receiver. The direction and distance of the object can be determined by measuring the elapsed time between transmission and detection of the ‘echo’. In welded components this is usually done by moving a small probe, containing both transmitter and receiver, over the item to be examined and displaying the echo on an oscilloscope screen. The probe transmits a beam of ultrasound that passes through the metal and is reflected back from any defects, much like shining a torch at a mirror, in principle with the same rules applying to the reflection of the beam. This is illustrated in Fig. 11.11. Deeply buried defects such as lack of fusion, lack of penetration and cracks in addition to volumetric defects such as slag entrapment and porosity are all easily detected.

The success of the technique depends upon the use of trained, experi­enced operators who know precisely the characteristics of the metal being examined, the beam direction, its amplitude and frequency and the weld geometry. It is recommended that operators should be approved to one of the certification schemes such as those operated by the BINDT or the ASNT.

The frequency of the ultrasonic waves is generally in the range of 2­5 MHz, the lower frequencies being used for the examination of coarse­grained material and on rough surfaces. The higher-frequency probes are used for the detection of fine defects such as cracks, non-metallic inclusions,

Transmission pulse indication Horizontal CRT beam deflection




Defect indications

Vertical CRT beam deflection

for pulse initiation Signal from reflected___________

pulse (echo signal)



/ *


Signal base from time

Ultrasonic Test Set



Reflected pulses from defects

11.11 Principles of ultrasonic testing of metals. Courtesy of TWI Ltd.

lack of fusion and voids. The beams are transmitted as either compression waves or shear waves and, ideally, a defect should be oriented normal to the wave to give the maximum reflection. Projecting the beam at a glanc­ing angle at a planar defect can result in the beam being reflected away from the receiver and lost — remember the analogy of the torch and the mirror. The probe angle should be selected to optimise the reflection of the sound beam. Probes that project the beam into the test piece at an angle normal to the plate surface are ideally suited to the detection of laminar defects, i. e. those lying parallel to the plate surface and for determining the plate thickness (Fig. 11.12).

Probes can be obtained that project the beam into the test piece at an angle, the most common being 45°, 60° and 70°. The angled probes are best suited for the detection of defects at an angle to the plate surface such as lack of sidewall fusion. Here the defect is at the angle of the original weld preparation and as illustrated in Fig. 11.13 is easiest to detect by a probe of an appropriate angle. Note that the beam may be ‘skipped’ along the interior of a plate, enabling defects a long distance from the probe to be found.

Before commencing the examination some preparation work is neces­sary. Data on material and heat treatment, welding process and procedure and weld preparation design are necessary if accurate determinations of defect types, orientations and sizes are to be made. The normal inspection

11.12 Compression wave examination. Courtesy of TWI Ltd.

11.13 Angle probe examination of a weld. Courtesy of TWI Ltd.

method is to scan the probe on the surface of the parent metal adjacent to the weld. To do this the surface must be free of scale, spatter and roughness and the parent metal should ideally be free of laminations and excessive inclusions. A couplant, generally water, oil, grease or glycerine, is applied to form a film on the surface of the test piece. This aids the transmission of the beam into the sample.

To ensure that all of the defects in both the weld and the HAZ are detected the probe must be scanned over the full cross-section and the full length of the weld. Accurate sizing and positioning of any defects relies upon accurate marking out of the weld. Flaws that lie parallel to the beam may be missed and to ensure that this does not occur it is necessary to scan in two directions at 90° to each other. Interpretation of the reflections from regions such as root penetration beads, backing straps and fillet weld roots can be very difficult, leading to incorrect defect sizing and sentencing. For this reason the root area is frequently excluded from the area to be ultra- sonically examined.


• It is very good for the detection of planar defects and cracks.

• It can easily determine defect depth.

• It is readily portable.

• Access is required to one side only.

• There are none of the health and safety problems associated with the radiographic technique.


• Very skilled operators are required.

• Surface breaking defects are difficult to detect.

• Accurate sizing of small (<3 mm) defects is difficult or impossible.

• The geometry of the joint can restrict the scanning pattern and prevent accurate interpretation.

• No permanent objective record is available.

• The process can be slow and laborious.

11.3.4 Radiographic examination

Electromagnetic radiation has properties that are useful for industrial radi­ography purposes. The rays travel in straight lines and cannot be deflected or reflected by mirrors or lenses; they have wavelengths that enable the radiation to penetrate many materials, including most metals. They will, however, damage living tissue and therefore present some health and safety problems.

The radiation, either X-rays from a suitable source or gamma rays from a radioactive isotope, is absorbed as it passes through the material. This absorption increases as the density of the material increases so that if a pho­tographic film is placed on the side opposite the radiation source, any less dense areas will appear as darker areas on the film (Figs. 11.14 and 11.15), to give a shadow picture of the internal features of the test sample once the film has been processed. Thus voids, porosity, slag, cracks and defects of

11.14 Principles of radiographic examination of a weld. Courtesy of TWI Ltd.



11.15 Radiograph of aluminium welds.

(a) Longitudinal cracking at the weld start. Close square TIG butt weld in 7000 series alloy. (b) Gross porosity in TIG butt weld in 4000 series alloy.

geometry can all be identified, although planar defects normal to the beam may not be detected.

To radiograph a welded joint a suitable source of radiation, a film in a light-proof cassette and some method of processing the film are required. This latter generally requires a dark room where the film can be developed, fixed, washed, dried and viewed. The radiation can be produced from an X — ray tube, the energy generally being described by the voltage and current at which the tube is operated. These may vary from 20kV to 30MV and 10 to 30 mA, although the normal limit for the commonly available industrial units is around 400 kV. A 400 kV unit is capable of penetrating up to 100 mm of steel and 200 mm of aluminium. Gamma radiation is produced by the decay of a naturally occurring or manufactured radioactive isotope. The isotopes decay over a period of time, a measure of the longevity of the source being the half life, the length of time taken for the source to decay to half of its initial intensity. The most common isotopes are cobalt-60, half life 5.3 years; caesium-137, half life 30 years; iridium-192, half life 74 days; thulium-170, half life 127 days and ytterbium-169, half life 31 days. The strength of the source is expressed as curies or becquerels. As a gamma ray source cannot be switched off the isotope is stored in a special container equipped with either a port that can be opened remotely to expose the source or from which it can be wound out when required (Fig. 11.16).

11.16 Radioactive isotope projection system. Courtesy of TWI Ltd.

Neutron and electron guns are also used to produce high-energy beams. These can be used for interrogating materials in the same way as X — and gamma radiation. This equipment is not as readily available but has its uses in industry, particularly for very thick components where long exposure times would be required using conventional lower energy sources.

The quality of the radiograph is affected by the source to film distance

— the greater this is the sharper the image; the size of the radiation source

— the smaller the source the sharper the image; the beam energy — the higher the energy the less sharp the image; the film grain size and quality and the correct film processing. To enable the radiographic quality to be determined an image quality indicator (IQI) is used. This comprises a number of wires of different diameters or a stepped wedge with varying diameters of holes drilled in the steps. The IQI is placed on the source side of the test piece and adjacent to or across the weld so that its image can be seen on the radi­ograph after processing. The diameter of the thinnest wire or the smallest diameter hole that can be seen is then expressed as a percentage of the specimen thickness — the percentage sensitivity of the radiograph. The other quality control measure is the density of the radiograph which may be mea­sured easily with a densitometer. Ideally the density should be between 1.8 and 2.5. Radiographs produced using X-radiation are generally of better quality than those produced using gamma radiation. Variations in sample thickness will result in variations in density which may make parts of the film either too dark or insufficiently dense for accurate defect detection.

Real time radiographic equipment is now being more widely used. This uses a fluoroscopic screen and a video camera, enabling the image to be stored, retrieved, and automatically judged almost instantly. This has obvious benefits with respect to the speed of identifying and correcting welding faults.

Radiographic interpretation should be entrusted to well-trained experi­enced radiographers and should be performed in a darkened viewing room on a viewer designed for the task.


• A permanent record is available.

• Both buried and surface defects can be detected and the technique is particularly good for finding volumetric defects such as slag and porosity.

• The equipment is portable, particularly the gamma ray sources.

• All materials can be examined.


• The capital cost of equipment, which will need to include the process­ing and viewing facilities.

• Health and safety considerations — large areas may need to be closed off during radiography or enclosures must be provided in which the radiography is carried out. Radiographers must also be monitored for exposure to radiation.

• Access is required to both sides of the component, the source on one side, the film on the other.

• There are problems in detecting planar defects and fine cracks if these are normal to the beam.

• There is a limitation on the thickness that can be radiographed and defects easily detected.

• Skilled and experienced radiographers are required.

• The depth and through thickness dimension of a defect is very difficult to determine.