Bryan Kenzie and Julian Speck
Bryan Kenzie is a principal NDT engineer at who undertakes a range of inspection development projects and leads TWI's participation in the TOFDPROOF project.
Julian Speck is TWI's structural integrity department manager, responsible for FFS and RBI projects and development activities.
Paper published in Inspectioneering Journal July/August 2005.
TOFD inspection technology
This article is intended to provide a summary of the issues surrounding the use of the ultrasonic Time of Flight Diffraction (TOFD) technique. TOFD was developed for the UK nuclear industry during the 1970s to provide a method formeasuring the height of planar flaws. TOFD is now generally recognised as the most accurate ultrasonic technique for measuring the height of embedded planar flaws (eg. cracks, lack of fusion, etc.) that lie perpendicular to vessel orpipe surfaces.
Commercial TOFD systems (eg. MICROPLUS) have been available for some time from a range of suppliers. (An independent buyers' guide can be found on the website of the British Institute of Non-Destructive Testing at www.bindt.org). At present, national standards for the application of TOFD exist. However, no standardised acceptance criteria have been agreed upon for pre-service (fabrication) inspection.Therefore, for pre-service inspection, TOFD is commonly used with flaw acceptance criteria derived by engineering critical assessments (ECA) as described in BS7910.
How does TOFD work?
Most ultrasonic techniques rely on receiving reflections from flaws, even if only from particular facets of the flaws. TOFD detects cracks using the signals diffracted from the flaw's extremities (tips). Two angled compression waveprobes (typically between 2 to 10MHz frequency) are used in transmit-receive mode, one probe each side of the weld. The beam divergence is such that the majority of the thickness is inspected, although, for thicker components, morethan one probe separation may be required. When the sound strikes the tip of a crack, this acts as a secondary emitter that scatters sound out in all directions (some in the direction of the receiving probe), Fig.1.
Fig. 1. Basic TOFD two-probe configuration
A 'lateral wave' travelling at the same velocity as the compression waves, travels directly from the transmitter to the receiver. The time difference between the lateral wave and the diffracted signal from the flaw provides a measure of its distance from the scanned surface. If the flaw is large enough in the through-wall (height) dimension, it may be possible to resolve the tip diffracted signals from its top and bottom, thereby allowing the through wall height of the flaw to be measured.
TOFD signal interpretation
Due to the low amplitude of the diffracted signals, TOFD is usually carried out using a preamplifier and hardware designed to improve signal-to-noise performance. As the probes are scanned along the weld, the A-Scan signals aredigitised, Fig.2.
Fig. 2. Typical A-scan showing responses from an embedded planar flaw
The signals are displayed as a grey-scale image with flaws as alternating white and black fringes, Fig.3. The diffracted signals from the extremities appear as signals arriving at different times at the receiver. By carrying out geometric calculations an estimation of the through thickness dimension of a flaw can be obtained.
Fig. 3. Probe movement and image orientation
This results in a series of images of a weld showing the flaw positions in B-scan view (from end of weld, or cross section), C-scan view (looking through weld from cap side, or plan view), and D-scan view (from side, showing weld length and height).
TOFD can also utilise synthetic aperture focusing and/or beam modelling software to minimise the effects of beam divergence. This can provide more accurate location and sizing information.
In-service inspection with TOFD
TOFD is a very rapid method of inspecting whole volumes. It is most widely used in the nuclear and oil refining and petrochemical industries, for the in-service inspection of butt welds in pressure vessels, process pipework, etc.The results are then commonly used in fitness for service (FFS) assessments, in accordance with, eg. API RP579. (For an FFS assessment of a known flaw, information about flaw length and height for embedded and surface flaws, is required. It is also important to determine the component thickness and the flaw orientation with respect to the principal stress direction, and whether or not the flaw cross-section is planar).
In the past, collaborative UT trials carried out at TWI and elsewhere in the UK found that typical errors on planar flaw height measurement, using the 20dB-drop technique by operators on 'the shop floor' and in 'the laboratory',ranged from a mean error of -1.0mm (undersize) with a standard deviation of 3.1mm, to a mean error of -2.7mm with a standard deviation of 3.0mm, Fig.4.
Fig. 4. The standard 20dB-drop UT technique generally undersizes, compared to TOFD
The TOFD sizing technique provides a better correlation between measured and actual flaw heights than the 20dB-drop technique, Fig.4. The collaborative trials also found that typical errors on planar flaw height measurement using TOFD had a mean of +0.5mm (oversize) with a standard deviation of 1.8mm. The actual flaw heights investigated in the studyranged from 1.5 to 30mm.
Some important shortcomings of TOFD
FFS engineers should be aware that the technique becomes less reliable when:
- The material contains scattered inclusions (eg. inclusions in 'vintage' steels);
- There is a high density of defects due to, eg. hydrogen damage; and
- Inspecting coarser grained materials (austenitic stainless steel weldments).
The detection of small flaws near the scan surface, can be more difficult due to the presence of the lateral wave response which often occupies several millimetres of the depth axis on images. TOFD is restricted where signal timedifferences are small so that the different signals cannot be resolved in time, eg. for small flaws close to the backwall (far surface).
The roots of a single-vee pipe butt weld have specific difficulties for UT inspection in general, due to the fact that the root geometry itself can be a source of ultrasonic signals. Another difficulty occurs when there is high-lowin butt welded joints (due to thickness variation and/or misalignment). This can mean small flaws are hidden by corner echoes and there is a potential for false calls.
A special case: In-service wet H 2S damage
Carbons steels in wet H2S service may be susceptible to a range of damage mechanisms that give rise to surface and embedded cracks, ie. sulphide stress corrosion cracking (SSCC) and stress oriented hydrogen induced cracking (SOHIC).API RP571 provides a comprehensive discussion on these wet H 2 S damage mechanisms, including methods for monitoring and inspection. RP571 suggests ultrasonic testing for the detection and sizing of cracks in wet H 2 S service.
The TOFD technique is often considered for this application but it has some limitations; the advantage of using TOFD for initial flaw detection is not so certain. For example, it is best suited for the detection of flaws that do notlie close to, or break, a surface which can be important when trying to detect internal surface cracks during external (non-intrusive) TOFD inspection. It is therefore recommended that pulse-echo shear wave UT is used for initialdetection scans, unless a TOFD procedure has been validated for the detection of all H 2 S flaw types and sizes of interest. TOFD should then be used to specifically target any such flaws in an attempt to determine their size.
Due to image interpretation difficulties, it should also be expected that TOFD scans may be incapable of reliably sizing SOHIC (particularly embedded cracks, during the earliest stages of crack development), Fig.5. SOHIC is commonly observed in weldments in the base metal adjacent to the heat affected zone (HAZ), oriented in the through-thickness direction. SOHIC describes an array of cracks, aligned perpendicular to the weldingresidual stress, that are formed by the link-up of small hydrogen induced cracks in steel. The worst SOHIC case would be through-wall connected (stepwise) cracks, perpendicular to one of the principal membrane stresses from, eg.pressure loading.
Fig. 5. Schematic of SOHIC damage at a weld, that may or not be connected
TOFD relies heavily on the correct interpretation of data images, which can sometimes be difficult. In the case of in-service inspection for the detection of SOHIC, the indications from numerous harmless steel inclusions (typical ofa poor quality steels susceptible to wet H 2 S damage) can make the identification of more serious flaws (connected SOHIC cracks) extremely difficult. Inspection engineers should be aware that this may lead to conservative false calls and unnecessaryrepairs!
TOFD technology development in Europe
Although standards exist for the application of TOFD, specific acceptance criteria for weld defects detected and measured during pre-service (fabrication) inspection have yet to be established. (In the US, several ASME Code Caseshave been issued that attempt to address acceptance criteria for welding flaws). The European Commission (EC) in collaboration with EPERC (European Pressure Vessel Research Council), is therefore sponsoring the TOFDPROOF project thataims at producing a coherent package of procedures for applying TOFD, with related specific acceptance criteria and recommendations for training and certification. The € 1.5million project is focussed on all aspects allowing the effective application of the TOFD as a stand-alone method for the weld inspection during manufacture ofpressure equipment.
In TOFDPROOF, a comparison is being made of TOFD performance with conventional NDT applied in accordance with standards for testing pressure vessels during the manufacturing stage. The comparison will be performed using probabilityof detection (PoD) curves and statistical analysis of flaw sizing data. A high degree of operator skill and experience is required to correctly interpret and report the findings of a TOFD inspection. Guidelines for training andcertification of TOFD operators will also be written and distributed to relevant NDT societies. An interactive guideline for interpreting TOFD results with a data bank of typical images recorded on flaws will also be available, in an'interactive training area' on the TOFDPROOF website.
Some of the TOFDPROOF deliverables approved for public access are already on-line at www.mpa-lifetech.de/TOFD/. These deliverables will provide the framework for the development of standards for the use TOFD for pre-service inspection. The benefits of the project areexpected to be increased quality assurance and improved safety of welded vessels. Pressure vessel manufacturers also expect savings in inspection costs, amounting to about 10% of all manufacturing costs for high integrity equipment,eg. thick-walled reactors.