The dilemma they face centres around the delicate balance between safety conformity and economic survival.
Preventative maintenance strategies alone are no longer considered a viable solution to the problem because of the inherently high cost of upkeep and the even more significant impact of downtime. Such precautionary measures are increasingly considered an unaffordable luxury.
On the other hand, the attendant risk and commercial consequence of unscheduled stoppages and the increased sensitivity of ecologically driven public opinion are equally onerous and unacceptable burdens.
Non-destructive testing, as a means of establishing 'as-built' quality and 'in-service' condition, has promised for many years to provide operators with the information they need to effectively manage their assets, but so far singularly failed to deliver. Not surprisingly, many of those responsible for the operational safety, reliability and commercial performance of critical industrial installations are wary of the claims made by those advocating the broader use of NDT for this important task.
It is really only in the last few years that the limitations of long established inspection methods have started to be accepted - that radiography, despite its representative hard copy evidence, is not sensitive to critical defects such as cracks and structural degradation; that electromagnetic methods are severely 'depth' restricted and that conventional 'reflectivity base' ultrasonic techniques can be highly subjective and not necessarily accurate in sizing and characterising integrity affecting defects. It is also somewhat ironic that, despite major technological advancement in the field of fault diagnosis and prognosis, visual and surface inspection methods are still almost exclusively relied upon as means of assessing and certifying the safety, reliability and longevity of economically critical plant components such as statutory pressure systems which suffer structural rather than superficial damage as a consequence of the extreme operating conditions they routinely endure.
Perhaps the greatest surprise of all is that until very recently the view was upheld that these methods of assessment could, in isolation, be used as some form of magic oracle to pass judgement on the quality and condition of critical components in fabrication and in service. These somewhat naive beliefs are borne partly out of the false claims practitioners have made about their non-destructive testing capabilities in the past and partly out of industry's over expectation which results from the desperate need to find cost effective solutions to performance affecting problems.
As yet the all embracing panacea has eluded scientists. Increasingly, it is rightly accepted that no particular method holds all the answers but that a combination of complementary inspection methods used in conjuction with process knowledge, metallurgical assessment and operational history provide a more complete picture of condition than could be achieved by any single inspection method and that this integrated approach constitutes the most effective basis for the management of important and valuable industrial assets.
Despite being increasingly recognised as a worthwhile method of determining condition, TOFD remains inadequately prescribed and poorly understood and, despite being generally acclaimed as an accurate defect sizing technique, little is known about its 'screening' capabilities and even less about its applicational shortcomings.
Some of this lack of understanding emanates from the mystique built up by those responsible for its introduction. For many years scientists promoted the technique as a highly specialised 'sizing' tool - so complex that it required their specialist knowledge and sophisticated technology to effectively apply - and unsubstantiated claims were made about what the technique could and could not achieve.
This may have been the case in the very early stages of evolution but TOFD has now been around for more than 25 years, its effectiveness has been proven by performance demonstration (more so than all other inspection methodologies combined) and whatever mystery once surrounded the technique has been completely dispelled by repeated applicational success in the hands of routinely qualified personnel using what is now relatively 'standard' equipment.
The following attempts to explain TOFD in layman's terms, to throw light on its shortcomings and to allay many of the myths that surround what is undoubtedly a very powerful and viable plant asset management tool.
It should perhaps first be restated that TOFD is an ultrasonic technique and, as such, it is governed essentially by the same laws of physics which apply to established reflectivity methods and is therefore subject to many of the same limitations.
Most notably these are the laws of wave propagation at the frequencies in question - typically 5 - 10 MHz. - and consequently an elastic medium is necessary to support the transmission of vibrational energy on which all ultrasonic techniques depend. A coupling medium must therefore be maintained between the transducers and the items under test and this must be adequate to ensure signal transmission at all times. Loss of couplant will result in loss of signal which, in turn, results in lack of coverage. TOFD does however enjoy the benefit of continuously monitoring and recording couplant status so that any shortcomings can be identified and addressed.
Where TOFD differs radically from all other ultrasonic based methods is that it relies on the detection of diffracted rather than reflected signals. On initial assessment this may not appear all that significant, but closer analysis shows that the consequences of this have a major bearing on the techniques capabilities and its limitations.
Rather than monitoring the (high) amplitude response of reflected energy and using this to make a comparative assessment of 'equivalent' defect size, TOFD relies on the detection of relatively low amplitude signals diffracted only from the tips of defects which forms a basis for absolute position (and therefore size) measurement - irrespective of amplitude response. This is achieved by using two separate transducers in a directly opposed tandem configuration - both being reasonably well matched short pulse, wide beam probes of the same angle but one acting as transmitter and the other as a discreet receiver. As with conventional ultrasonic testing the majority of transmitted energy is lost through absorption and diffusion by the material under test or it is reflected by any discontinuities falling within its effective envelope - but some is radiated by these discontinuities and it is these very low amplitude diffracted signals on which TOFD relies. By capturing these responses and processing them in a fashion whereby they can be discriminated from background and structural noise it is possible to create an image which, by differentiation, makes it possible to identify the presence and location of defects and to accurately position these with respect to the geometry of the item under test.
Because the technique does not rely on detection of reflected energy it is not amplitude dependant (for defect size measurement) and therefore not so susceptible as 'pulse echo' testing to consistent surface and consequently couplant conditions.
Also, because of the comprehensive coverage afforded by the characteristically wide probe beams used for TOFD the technique is not as dependant as conventional pulse echo ultrasonics to variances in probe position or defect orientation relative to nominal probe angle. This makes TOFD much less subjective in application and more effective as a routine detection method.
TOFD also differs from conventional ultrasonic examination by virtue of the fact that is performed using only compression waves and consequently any mode converted signals which propagate at lower velocity do not confuse the picture.
But perhaps the most significant distinction between TOFD and all other non-invasive volumetric inspection methods is the manner in which these diffracted signals are captured and processed for display in the form of a proportionate sectional image of the item under test.
This is achieved by digitising all 'raw' position related radio-frequency waveforms which are amplitude quantised using a grey scale to provide pattern differentiation when a series of collected A scans are stacked together to create a coherent sectional B or D scan image representing depth through and distance along the material under examination. The geometry or profile of the item under test is also usually 'imaged' by virtue of a near surface lateral wave transmitted between the two transducers which highlights the test surface and signals which are direct compression wave reflections off the far surface. Any defects encountered within the volume of material by the probe beam are consequently displayed in true relative position to the geometry in question.
This means that for the first time in the chequered history of NDT, engineers, production managers, safety responsible personnel and those involved with the commercial aspects of their plants integrity can be provided with absolute and accurate dimensional information presented in a fashion that they can readily understand.
One fundamental problem with TOFD is that these diffracted defects signal responses are exaggerated in length on the on-line display as a consequence of being scanned across by the wide beam envelope. Consequently even an isolated gas pore will appear initially as a linear response on the created image.
However, this problem is readily overcome because it is of a recurrent and known nature - caused specifically by beam geometry. Knowing the beam characteristics, it is possible to rectify this effect either visually (by using a profiled cursor) or electronically using a process referred to as Synthetic Aperture Focusing - but both techniques can only be used off-line to correct pre recorded data. It is likely, however, that the next generation of instruments available to perform TOFD will incorporate effective real time SAFT algorithms to produce corrected on-line images.
The resultant picture is a dimensionally true sectional elevation showing the through wall characteristics and condition of the item under test - in much the same way that medical scanners are now able to illustrate the internal condition of the human body without having to resort to technical jargon and 'industry gobble-de-gook' to explain what is going on.
ToFD 'D' SCAN IMAGES
|Through wall linear welding defect||Indication of lack of interun fusion||Images of poor quality welding depicting linear and point defects|
ToFD with its inherent capability of imaging the metallurgical structure of areas examined as well as the speed with which data can be collected, which is a direct result of the single pass data collection technique employed, is the ideal tool for defect detection (screening), critical sizing, defect characterisation and propagation monitoring. Not only are on line hard copy print out's of the areas inspected produced, as evidenced above, but all the ultrasonic data is digitised and stored to magnetic medium (These archives can be maintained for the life of the plant). The stored data is then available for further off line processing and forms a fingerprint on which the assessment of future material degradation can be based. Because the results of inspections produce accurate measurements of the defects detected the results can be used to carry out fracture mechanic studies with confidence.
Because it is fast, efficient. 'sees' everything and records all raw data for presentation in a proportionate and representative fashion, TOFD is an ideal detection tool which provides an accurate and invaluable 'fingerprint' of condition as a quality control function at the time of construction. Because the process is highly repeatable it also constitutes the most reliable means of propagation monitoring and, because of its inherent accuracy, this provides plant operators with invaluable management tool.
These defect detection capabilities have been validated by numerous independently conducted performance demonstration trials under the auspices of such organisations as TW1, DDT, EPRI, Lloyds, etc. And proven by literally thousands of routine applications where physical assessment of defect condition has subsequently been endorsed the finding of TOFD results.
From a commercial point of view, one of the major advantages of TOFD is that both detection and sizing can usually be performed instantaneously from the same source data without any need to recalibrate and rescan using additional or alternate techniques. This clearly has a radical bearing on time and on cost.
Statement like '2dBs above grass', 'cluster of porosity' and 'intermittent LOIF' do little to instil confidence and even less as a reliable basis for structural integrity assessment.
Unlike virtually all other inspection techniques TOFD does not rely on comparative assessment to quantify the significance of a detected defect.
Not only is it capable of visualising virtually every material or structured anomaly in characteristic fashion, it is also capable of giving such discontinuities true dimension and location to an order of accuracy and repeatability that is unprecedented in the field of NDT.
These are measurements related in millimetres, not dB's or 'equivalent' scales of response. The dimensions are absolute and unambiguous. Evasive descriptive jargon is not necessary when absolute size and accurate classification can be readily derived; as is the case in all but the most complex application.
The proven levels of accuracy attainable are often to within ± 0,1mm in terms of (critical) through wall extent and ± 0,1 - 0,5mm in terms of horizontal and vertical dimensional extent. Position is usually established to within 0,5mm and angular dispositions can be resolved to within a few degrees when appropriate scan procedures are used. For the first time accurate and reliable defect data can be used as a basis for fracture mechanics assessment without resorting to self-defeating methods entailing destructive intervention and physical measurement.
Other inspection techniques have been claimed to achieve similar dimensional and positional accuracy and to offer consistency - but have any of these been repeatedly proven capable of producing these results across the spectrum of conditions which prevail in industrial application and do any of the practitioners of these techniques have the confidence to quote actual defect size in their final reports? Unless applied under controlled laboratory conditions - without consideration to time, cost and effort - the answer is an emphatic NO!
With TOFD all relevant parameters are accurately recorded at very high resolution. These are stored digitally and can be retrieved and redisplayed at any time. All data is position related so that location can be reliably identified and results from recurrent inspection can be directly compared for change and propagation monitoring. The fact that the data is so comprehensive and inherently accurate means that sophisticated analysis techniques can be used for this purpose - including pattern recognition processes which are capable of eliminating any spurious factors such as couplant variation and datum displacement.
As previously explained, TOFD relies upon the reception of angled compression ultrasonic waves generated by a discreet transmitter and diffracted by the tips of any defects falling within its envelope of coverage. Because the two transducers used to achieve this condition are configured to face each other, an element of the beam is detected which travels direct from the transmit to the receive crystal just beneath the material surface. This is referred to as the lateral wave which, in effect, is a relatively short pulse/low amplitude 'standing' signal occurring at a fixed position along the timebase, dictated by probe separation.
In some respect this phenomena is useful in that, with known velocity and fixed probe separation, it defines the inspection surface and creates an important datum for defect depth positioning.
However, the fact that it is always present means that the very near surface - (typically 3 - 5mm) will always contain a signal which is often construed as a negative factor affecting near surface resolution - sometimes described by the ill-informed as a 'dead zone'.
But this description is not strictly valid. Any defects occurring within this region can still be observed (albeit at low amplitude) and their response will be out of phase from and superimposed above the lateral wave signal - making detection possible but rendering sizing and characterisation less reliable near the surface than through the rest of the depth range.
By using very high energy, short pulse (shock wave) transducers whose beam envelope is concentrated in the main area of interest, it is possible to reduce this lateral wave effect to 2 - 3mm in steel and to suppress its amplitude effect to just a few dB's - making defect discrimination more effective.
Because the lateral wave is a consistent signal occurring in a predictable and recurrent position it is possible to electronically (and visually) nullify its effect in much the same way that unwanted noise is eliminated in audio equipment.
Even when the regularity of the lateral wave is disturbed by uneven surface or by unstable couplant conditions, it is possible to 'process out' these anomalies by 'electronically straightening' the signal trace to further improve defect discrimination.
So in other words - Yes - TOFD does suffer from a near surface effect caused by its inherent lateral wave but his is not a serious problem unless very near surface sizing is called for. Being pragmatic, very few near surface (included) defects can be considered integrity critical and it is debatable whether the 'near field' characteristics of single compression probes and the inherent 'dead zone' effects of twin probes could provide better resolution using conventional reflectivity methods.
While radiography may be a little more sensitive at detecting such minor inclusions it would not provide any depth or worthwhile through wall positional information.
With a TOFD tandem array straddling a 50mm thick weld it is possible, by virtue of the divergent beam, to inspect the full volume of weld material and HAZs at speeds which cannot be comtemplated by conventional reflectivity ultrasonics methods which rely on comprehensive raster scan coverage by a number of relatively focused probes of different angle. TOFD has also been demonstrated to be faster than field radiography on all but the thinnest materials. Because of the simplicity of the transducer configuration and its ease of application, set up and calibration are also far more efficient than with alternate methods of ultrasonic testing.
The coverage rates attainable are restricted only by the practicalities of scanning and production rate of 100 -150mm/sec are not out of the question using manual deployment methods and even higher rates of coverage can be achieved when automated scanning is used. It should be remembered that these statistics relate to the length of weld volumetrically inspected in a single 'pass' of the transducer array and not just the scanning speed of the probes.
Many people are still of the impression that ultrasonics is as range limited as radiography and that TOFD, being dependant upon the detection of low amplitude signals, is even less effective than conventional probe echo methods on thicker material sections.
More importantly, from a commercial point of view, there is no time (and therefore cost) premium to pay in achieving such depth or volumetric coverage as there is for instance with angle shear wave pulse echo testing and with radiography. Very often a single 'pass' is sufficient to adequately cover the volume of material under test and scanning is so efficient that even when multiple passes are required there is no real time penalty.
Clearly at very long ranges or on coarse materials the effects of signal attenuation apply; but TOFD has been proven to be highly effective at reliably reporting code critical defects (missed by other methods) on 250mm thick reactor vessels for the nuclear and chemical industries - including over extended 'ranges' caused by complex geometry in the vicinity of flanges, nozzles and penetrations.
Austenitically clad ferritic components are often assumed to be beyond the scope of TOFD but again this is a misconception. Whereas radiographic and reflective ultrasonic methods are virtually worthless at detecting and depth reporting cracks propagating from such metallic interfaces, TOFD is perfectly suited to imaging both the interface (indicating bond integrity) and even minor (micro) cracks emanating from the bi-metallic fusion face.
As with interface cracking TOFD is perfectly suited to resolving intrusive root defects (including lack of penetration and cracking) in a very efficient 'single pass' fashion. Mismatch or HI/LO can also be detected but additional scans are required if transverse position is required.
TOFD is also particularly suited to the resolution of surface breaking circumfereneial cracks such as those caused by fatigue and prevalent in cyclically loaded rotating machinery.
With very tight cracks, or where discontinuities are contaminated with material deposits capable of supporting the passage of ultrasound, The phenomena of through transmission can sometimes occur. The resultant image appears as a ghostly translucent 'shadow' made up of processed signals diffracted through the body of the defect and these can often highlight microscopic details such as 'beachmarks' where cyclic fatigue crack propagation has arrested during its destructive progression.
These can be serious integrity affecting conditions which are beyond the scope of detection, let along quantification, of the majority of available NDT methods, and which cost industry billions each year in maintenance, repair and downtime.
Where TOFD scores over alternate inspection techniques is in the speed at which it can be applied and the method by which it captures its data. One metre of 50mm thick weldment can be volumetrically covered and all data captured for off-line analysis in less than 10 seconds. This equates to more than 10,000 high resolution waveforms, each of which represents every event with the body of the weld and both HAZ'S over the full depth of interest. The composite image is immediately available for initial assessment and each individual waveform can be retrieved and analysed off-line in an environment more conducive to decision making than in the glaring heat radiating from an on stream pressure vessel.
In more recent years this expertise has been adapted for non nuclear applications including vessels for the chemical/process industries, complex forgings and castings (eg turbine discs) and nodal configurations on tubular structures.
However, the scan procedures, calibration, data presentation and analysis processes involved are often as complex as the geometry itself and the cost of implementation is usually proportionately high.
In terms of overall safety it could be argued and demonstrated that the improved defect reporting accuracy and reliability of TOFD is a very powerful tool in the risk assessment equation and that, as such, it is in itself an important safety device.
This is not problematic when applied to relatively 'clean' or refined material where major anomalies are to be reported but - in coarse material, poor fabrication or where the true extent of problems is not desirable information - the techniques sensitivity can be construed as a hindrance and, in certain circumstances, can make interpretation and sentencing a less than straight forward task.
However, once again, this must be viewed in the context of how other available techniques would deal with such problems. The same or even greater difficulties would be encountered under these circumstances with reflectivity based ultrasonic testing (with arguably less reliable results), and radiographic examination would probably prove totally ineffective save for reporting non critical defects such as porosity and slag.
One of the real advantages of TOFD, in this respect, is that because all hi-resolution position related data is captured and stored, very detailed off-line analysis can be performed and a wide range of validated analytical tools have been developed which aid and improve this process.
The logic behind this is something of an enigma and it can only be concluded that the inertia of the regulatory bodies and the vested interests in some areas of the NDT fraternity are the attributable reasons for this prolonged gestation.
The broader acceptance and specification of TOFD has not been assisted by the fact that most acceptance criteria in use today relates to radiographic and reflectivity based ultrasonic techniques prescribed as Fabrication Quality control methods during the construction boom of the 60's and 70's.
Despite major advancements in material science, improved knowledge of degradation processes, better fabrication techniques, more refine operating processes and revolutionary progress in inspection technology, many of these often archaic standards are still in use today and are regularly proposed for in service condition assessment applications to which they hold no relevance.
It is really only over the last 2-3 years when these legacies have started to be eroded away. TOFD is now 'accepted' by Lloyds, it carries its own British and (pending) European standards and it is under review by all major international bodies including ASME and API. Perhaps more importantly, industry itself has recognised the technical and commercial benefits the technique holds and major client organisations across every sector of the market are now specifying the use of TOFD on an increasingly routine basis.
Other recognised players fought hard to discourage the broader acceptance of TOFD because it was in commercial conflict with their established businesses. Under these circumstances many influential bodies (eg. Corporate research establishments, technical support organisations and regulatory 'institutions') fought shy of endorsing the routine application of TOFD because of industry inertia. A classic example of this is the attitude of certain personnel training and certification bodies who have done everything within their power to perpetrate the myths that their particular brand of NDT (ie. that in which they have a vested commercial interest) is, without question, the most effective and - more importantly - the one 'endorsed' by them.
However, over the last few years, it would appear than common sense has prevailed.
The cost of TOFD is now more affordable; industrial awareness has improved and an increasing number of reputable vendors have adequately resourced to offer the service.
These currently number five or six established companies with international representation, an equal number of new players who have geared up to support their domestic market needs and one or tow new organisation set up specifically to provide competent and cost effective TOFD services on a world-wide scale.
Whereas the former factor is totally invalid the latter, until the early 1990's, was certainly true.
The few commercially available instruments capable of performing the technique, typically retailed for in excess of $200k. They were logistically cumbersome (and therefore expensive to ship) and required the attention of highly skilled and equally expensive operatives.
In response to this situation, a number of organisations have been tempted to try to 'cheapen' the technique by attempting to perform TOFD with conventional 'manual UT sets' and traditionally certified personnel - with disastrous effects which caused them commercial embarrassment and did nothing to improve the credibility of the technique self.
However, in response to increasing market recognition, a number of manufacturers have now developed instrumentation which is substantially less expensive, more portable and more reliable, and which is de-sophisticated to the point where it is user friendly in the hands of conventionally skilled personnel.
Recent commercial applications where this new generation technology has been deployed have shown that production rates of up to 250m/shift can be achieved by a single operator using automatically actuated scanning on repetitive tasks. Under construction site conditions, where access and logistics are less conducive. Production rates of 100m/shift have been routinely achieve by a 2 man team. On a 24 hour day cover basis, this equates to inspection costs as low as $4/m under ideal 'production' conditions and less than $10/m 'in the field' where reasonable access and good continuity are assured. These viable production costs do not however prevent a few vendors from believing that, because TOFD is so powerful, they can still command excessive premiums for the service.
Not withstanding the rates charged for implementation of the technique, the main economic benefits of TOFD do not lie simply in the reduced cost of inspection.
Increased productivity afforded by reduced 'scheduled' downtime and improved availabitly as a consequence of fewer 'unscheduled' stoppages are the real commercial bonuses. The factors - combined with increased confidence, improved plant reliability and safety, and therefore better (asset) management control that ensues from improved awareness of condition - constitute a powerful argument for the more expansive use of TOFD.
It is here, in the area of consequential benefits, where the real value of TOFD lies - despite its few inherent limitations.
None of these limitations can be considered as serious and most can be quantified and adequately resolved or compensated for.
The technique is faster, more reliable, more accurate, more comprehensive in terms of coverage and more cost effective than any alternate available methodology for quantifying the integrity of industrial plant in construction and in service.
It is safer and more sensitive to integrity critical defects than any form of radiography.
It has been largely misunderstood, often badly presented and has historically been over priced - but these commercial inadequacies no longer apply.
It has also been restricted by availability and by prescription. Availability has significantly improved in recent years and most end users are increasingly aware of and comfortable with its capabilities - but isn't it a pity than 'learned' organisations have vacillate for many years and the cost of this indecision to industry is of enormous economic proportion.
It has been jealously ignored or attacked by those equally learned organisations who failed to spot the obvious or who considered it a threat to their particular thread of direction research, commercial bias or perceived lifeline for survival.
But, in final analysis and despite these evolutionary difficulties, TOFD has now 'come of age'.
No longer can its attributes be ignored, no longer can its effectiveness be avoided and no longer will industry continue to be deprived of an integrity assessment technique so crucial to the management of its assets that its value should be measured on what real benefits it can offer - nor on what it is perceived to cost.
For more information see: TOFD in UTonline 09/97
|| UTonline ||