·Table of Contents
Historical review of Technology Development in NDE
Don E. Bray, Ph. D., P. E.,
Don E. Bray Inc. P. O. Box 10315, College Station, Texas 77842-0315 USA, V
Email : firstname.lastname@example.org
Nondestructive Evaluation has its origin with observers and inventors putting their solutions directly into application on existing problems. In these early cases, the time period from
recognition of a problem, perception of a solution, and development of the technique to affect the solution could be very short. Often, the same people or same company would be involved from perception to application. In more recent years, there are more people involved in the technology development process, the various processes have become more specialized, and the time from perception to application has greatly lengthened. This paper will review some
early development patterns for radiography, magnetics and ultrasonics where the time frame from perception to application is less than ten years. More recently, applications such as tip
diffraction techniques in flaw sizing, phased array transducers and nondestructive materials analysis may take twenty-five years or longer to reach the acceptance level. Recent problems affecting this include a proliferation of research which may not have any real chance for application and an application work force that is not readily accepting of new technology. It is felt that better education of engineers in the principles and applications of
nondestructive evaluation will speed along this acceptance of newer technology, and better focus NDE research.
The primary concern here is that the flow rate of NDE research to application is too slow, and that this inhibited flow is restricting growth of the NDE industry, as well as reducing
the benefits of NDE. This view comes from prior work experience with a United States government-funding agency as well as with research and academic institutions. The present report follows the Mehl Honor Lecture delivered at the 1999 Fall Conference of the American Society for Nondestructive Testing (Bray, 2000).
The plan for this presentation is to look at the gestation period of NDE research, from the
time that feasibility appears to the time of acceptance by industry. Four early examples are given where the period is rather short, i. e. less than ten years. More recent examples indicate that twenty years is not an unlikely time for this process to take place, if it occurs at all. It appears that many solutions to problems that are physically sound never actually make it to the market place.
Wells (1996) has observed problems in the flow of NDE research to the market place. He
comments "It is because of the disparity between the amount of R&D that seems to take place and the achievement of an actual commercial result that causes me to give some time to this subject." He further concludes that most research projects begin with little knowledge of the actual commercial market of the technology. This scenario, repeated many times, will reduce the application of new technology.
TECHNOLOGY TRANSFER IN THE EARLY YEARS
Four examples of early NDE developments will be discussed, namely gamma and tube radiography,
magnetic rail flaw detection and ultrasonic inspection, as shown in Table 1. The rule for the discussion to follow is to estimate the time when the technology was shown to be feasible and either an industry standard or a device to apply the technology appears.
Gamma radiography was discovered in 1895, Briggs (1976). However, feasibility would
more likely be in 1930, with the experiments of Charles Briggs. Here, there appears to be a seven-year period from the readiness of the technology until the ASTM standard in 1937 (ASTM, 1937). Tube radiography follows a similar path, with a nine-year period from readiness to acceptance (Becker, 1990).
Early applications of NDE for rail flaw detection moved quickly toward application.
Some early discovery experiments began in 1915, but these were not really successful (Allison, 1968). Dr. Sperry's experiments in 1923 probably were the start of the development of the
technique for rail flaw detection. By 1930, the technique was working, and four inspection cars were built (Fig. 1). Again, we have a seven-year period from when the technology was shown to be practical until a working system was available and accepted by the industry.
The final example is the development of pulse-echo and through transmission ultrasonic
inspection. It is well known that the instrumentation that led to this technology was developed in WW II. The ideas were introduced in several publications in 1945 and 1946 (Erwin, 1945). By 1953, seven to eight years later, a commercial instrument such as the Sperry UR was available and widely purchased by the aircraft and railroad industries (Fig. 2).
In recent years, there have been several initial developments that do not appear to have made a quick trip to application, namely tip diffraction flaw sizing, material property analysis
and frequency analysis techniques. Of course, there are many others not shown.
|| Period (years)
|Magnetic rail flaw detection
|UT Tip diffraction flaw sizing
|UT Phased array systems
|UT Material analysis
|UT frequency analysis
|Table 1 Gestation period for significant NDE developments|
||When technology was presented to technical community in a transferable form.
||Appearance of industry standard, code or commercial product.
||To be determined, acceptance not yet widespread.
Fig 1: Early Sperry Magnetic Rail Flaw Detector Cars (From Allison, 1968)
Fig 2: Sperry UR Reflectoscope (From Sperry Technical Bulletin)
Tip diffraction flaw sizing has followed a very slow path to acceptance. The physics and needed instrumentation have been around for some time, yet there appears to have been little interest in developing this technique for flaw sizing. Amongst the earliest work was by Cosgrove (1970), in a private company report. There may have been some work even earlier. Silk and Liddington (1975) reported the same principles in the open technical literature. Cowfer and Hedden (1991) appear to offer the first industry oriented paper describing a comparison of the technique to other more widely accepted flaw-sizing methods. ASME Case Code 2235 (1996), indicates industry recognition of the technique. Here, we see twenty-six years from basic introduction of the technique to acceptance.
The acceptance of nondestructive material analysis techniques has been even slower.
Interestingly, Firestone and Frederick (1946) used their new ultrasonic system to show the effects of polarized sound in analyzing texture in rolled steel plate. The acoustoelastic effect for stress measurement was demonstrated in the 1970's (Becker, 1973). Suitable laboratory equipment appeared shortly afterwards. Yet, there is still no widespread acceptance amongst engineers and materials specialists that either of these techniques has merit. Cutting,
sectioning, microscope investigations, tensile tests, strain gauges and x-ray techniques remain the primary tools for materials investigations.
Phased array transducers offer many advantages in ultrasonic inspection, particularly the
ability to adjust transducer characteristics such as the beam angle or focus to meet the inspection application. Widdington (1977) gave the principles sufficient to apply the technique. And, there were limited applications. However, it was only recently when commercial vendors began making them available.
Transducer characterization is an exception to the slow acceptance assumption of this
presentation. Acceptance, however, probably was driven by the availability of frequency analysis equipment so that purchasers of ultrasonic probes began evaluating the characteristics of those that they had purchased. This feedback to the manufacturers led to a fairly rapid acceptance of this technology.
Frequency analysis techniques in materials analysis have not met with such a kind fate as
transducer characterization. Materials specialists tend to rely in metallographical techniques rather than nondestructive methods. Metallographic techniques involve cutting and sample preparation, which are slow, and severely restrict the amount of material that can be analyzed. Similarly, frequency analysis in the sizing of small flaws is not yet accepted by industry.
Acceptance of either of these processes could greatly improve material analysis
and flaw sizing for critical materials.
EFFECTS OF SLOW NDE TECHNOLOGY TRANSFER
There are many negative effects of our apparent inability to move new developments to application in a timely fashion. Among these negative effects are more costly designs, less energy efficient operating systems and misdirected research.
NDE is an integral component in a total failure prevention process. Efficient designers
will characterize the expected load environment for a part, select the best material and establish a suitable inspection system. Often, several iterations will occur in this process, with feedback from all segments. Mechanical design first assumes a fatigue (S-N) curve for a component. Recognizing the uncertainty around the curve, a Factor of Safety is used to move the actual design point downward. If a part is expected to see N cycles of service at a
specified load P, the stress s is established by selecting the load area A in the part (s = P/A). Thus, the designer can minimize the weight of the part by choosing the smallest possible area allowable by the yield stress of the material. If you need a factor of safety of 4, for example, then the area and amount of material will be greater than for a factor
of safety of 1. Supposing that a satisfactory design is created with a factor
of safety of 4 and an expected life of N2, and it operates for some period (less than N2) without failure. Then, if someone asks, "Can I increase the load on this part and expect it to operate safely?" The answer may be yes, since the operating conditions of the part are in the factor of safety zone. However, there is risk. A proper answer to the question may be "Yes, provided we increase the inspections on the part." These are engineering
Decisions on frequency of inspection are an important contribution to design and
maintenance. Engineers make these decisions. For example, Figure 3 shows a typical failure rate curve. Failure rate is on the vertical axis and number of cycles (time) is on the
horizontal axis. Toward the right side, we see the constant failure (flat) line suddenly
take an upward turn at an inflection point Tf. This is the onset of fatigue failure. A proper inspection scheme will recognize that no NDE system has 100% POD (probability of detection). Therefore, an inspection plan will start the inspection before Tf, at TM. This approach will enable several repeat inspections before final fatigue failure is expected, assuring that fatigue defects will be found before they enter the
critical region. Commercial aircraft maintenance is based on full knowledge of this fatigue and failure curve (Norris, 1988). For aircraft maintenance, the damage tolerance region is where the inspection occurs, and this corresponds to Tf and TM above. Engineers make decisions on the frequency of inspection as well as the type of inspection.
If the information relating fatigue and inspection is well known, then why did the
Aloha Airlines 737 blow off the top of its fuselage over the Pacific Ocean in Hawaii in 1988 (Carey and Duffy, 1988, NTSB/AAR-89/03). Here one stewardess was sucked out of the airplane to her death, and the passengers completed their flight peering out on the Pacific Ocean through the large opening in the side and top of the airplane. The pilots were unaware of the condition of the aircraft, and didn't realize that they were flying suspended over the Pacific on a fragile cantilever beam
The reason for the event just described may be related to the fact was that there was no
engineer in the employ of the operating airline in the periods before the accident. The inspector in charge of performing the eddy current inspection on the aircraft could not demonstrate this process to FAA inspectors after the accident. The flaw that caused the final failure was so large that a passenger saw it upon boarding the aircraft. She assumed that the airline would not ask her to fly on an unsafe airplane.
As is often the case, the response to this catastrophe was a call for more research to find
smaller cracks on aircraft. Rather than finding smaller cracks, the real need was for more engineering involvement in aircraft inspection and maintenance and the development of better, more efficient inspection using existing techniques.
In response to the Aloha failure, one industry expert said "...80% of the cracks are found by visual inspection." The importance of visual inspection also was demonstrated in a survey of offshore rigs where twenty percent of the flaws were 4 inches (100 mm) or greater, easily in the visual regime (Cole, et al. 1987). Many in the range less than 4 inches (100 mm) also were visible to the naked eye.
An example of the typical information gap between research and application is expressed by an
anonymous NDE researcher. First, the individual commented "I found it hard to believe how little we knew about the NDE of electronic materials." Further, he stated, "The "Ignorance" on the customer side can be overcome by using the expert systems approach. That is a real challenge for an NDE expert, though, when the
potential user does not have the knowledge (nor any desire to get this knowledge) of the basic principles of your technique." Here the researcher acknowledges his lacking in the fundamental knowledge of the application of his work. However, he feels that the potential users of his technology have little interest in learning about the basics of these new developments. This inhibits his ability as a researcher to learn about the application, and limits the understanding of the technique by the potential user.
Fig 3: Failure rate curve showing inspection interval Tm to Tf.
ENGINEERING EDUCATION AND THE NDE INFORMATION
What is the cause of this information gap in NDE? Probably, it is engineering education. A
review of the annual education surveys published in Materials Evaluation shows that for the year 1998, only 82 senior colleges report any presence of NDE (ASNT, 1999). The 82 senior colleges represent a mere 23% of the 350 engineering colleges. And, this does not say that the engineers graduating from these colleges know anything about NDE. It simply says that NDE is known, somewhere, at that college.
Further insight into the problems with engineering education are seen in the recent US News and World Report survey ranking U. S. colleges and universities (Wildavsky, 1999). The USNWR report divides colleges and universities into two groups, major research institutions granting Ph. D. degrees, and teaching schools granting a Bachelor's or Master's as the highest degree. Comparing these schools with those found in the ASNT survey in Materials Evaluation, we see that 34% of the research universities show some NDE. But, again, this does not indicate significant NDE exposure to the general population of engineering college graduates. More significantly, only 6% of those schools whose primary degree is a BS or MS show any knowledge of NDE. Thus, NDE is sorely lacking in engineering education.
It appears also that the education of United States colleges and universities does not
cover the topics needed by the first jobs taken by recent engineering college graduates. What first jobs do young, graduating engineers take in industry? The Sloan Foundation/American Society of Mechanical Engineers (1999) reports that almost as many graduates take jobs where NDE is used (Plant operations, maintenance, manufacturing etc.) as take jobs where design is done. The education of our graduates is not balanced toward this job distribution. Where we have heavy emphasis in design in the engineering colleges, we have very little in NDE
related areas such as plant operations, maintenance and manufacturing. Perhaps there is a need to alter the course of engineering education, and add courses that relate to maintenance, operations and manufacturing.
NDE and GLOBAL WARMING
Our lack of NDE education apparently has affected the response of the ASME to the Kyoto
Protocol, the treaty dealing with carbon emissions (ASME, 1999). The ASME position paper emphasizes research to improve energy efficiency through improved combustion, seals etc. Also, many of the proposed responses to the Kyoto Protocol are administrative. For example, trading power and emissions amongst various plants is a way to achieve an average level, which might be
acceptable, without an actual reduction in emissions at individual power plants.
The potentially significant role of NDE in reducing global worming is not mentioned
in the ASME report. There is some mention of the need for life extensions in existing power generating plants. However, no one has thought about "plugging the holes" to actually reduce
energy losses in power generating stations. More, and improved NDE, could have a significant affect on the plant emissions in the US. Estimates of potential energy saving from reduced leakage ranges from 1% to 24% (Doggart, 2000, Pruitt & Lebsack, 1994). The 24% energy saving with NDE for one plant is not typical, but it does flag this as an area that must be investigated. Recent reports indicate that engineers in industry are applying leak detection
techniques to reduce emissions (Carey, 2000). Recent engineering graduates are not well acquainted with these principles, however.
The process for bringing NDE research to the market place is described by Papadakis (1992),
and shown in Fig. 4. In the initial stages of basic research, the risk is high. As the initial idea moves down the development slope, the risk decreases. Capital investment is required as the value added increases and the new, viable process emerges at the bottom of the slope. The engineer plays a crucial role in the technology transfer at mid-slope. If the engineer at the technology transfer stage is not knowledgeable of NDE, the process will, most likely,
Fig 4: Value Added in Technology Transfer (Papadakis, 1992).
Despite the unique advantages of nondestructive evaluation in improving safety and
operating efficiency in engineering systems, many engineers are either untrained in the field, or they have only minimal capability. The reason for this is that engineering colleges do not include NDE as part of their curriculum. When mentioned, NDE is often treated as testing, omitting the engineering planning and analysis that should be part of a full and effective
application. More effective research, better designs, safer operations, and fewer emissions through energy losses are possible with effective NDE. Better educated engineers will lead to quicker acceptance of NDE research. Further, with increased application of NDE, equipment sales also will increase. The NDE community must become activist in order to achieve these goals.
- Allison, Arthur B., "Sperry Rail Service," The Bulletin of the National Railway Historical Society, Vol. 33, No. 6, 1968, pp. 7-21.
- Case Code 2235 - "Use of Ultrasonic Examination in Lieu of Radiography Section VIII." American Society of Mechanical Engineers, New York, New York, Dec. 1996.
- American Society of Mechanical Engineers, "General Position Paper of ASME International on technology Implications for the US of the Kyoto Protocol Carbon Emissions Goal, ASME International, Washington, D. C., Feb 1999.
- Committee E-7 on Radiographic Testing (Nondestructive Testing), ASTM, 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959
- American Society for Nondestructive Testing, "Directory of Educational Institutions Offering Courses in Nondestructive Testing," Materials Evaluation, Vol. 57, No. 2, February 1999, pp. 218-223.
- Becker, F. L., "Ultrasonic Determination of Residual Stress," Battelle-Pacific
Northwest Laboratories, Richland, WA, January 1973.
- Becker, George L., Radiographic NDT, Du Pont NDT Systems, 1990, 492 pp.
- Bray, Don E., "Fixing the Impedance Mismatch in US Technical Education," Mehl Honor Lecture, Materials Evaluation, Vol. 58, No. 3, pp. 394-395, 397-400.
- Briggs, Charles W., "Developments in Gamma Ray Radiography 1928-1941," Materials Evaluation, Vol. 34, No. 3, March 1976, pp. 14A- 20A.
- Carey, John, "A Free Market Cure for Global Warming," Business Week, May 15, 2000, pp. 167-169.
- Cary, Peter, and Brian Duffy, "When to Junk Geriatric Jets," U.S. News & World Report, Vol. 116, No. 20, May 16, 1988, pp. 16-17.
- Cole, M.W., T.F. Marucci, and D.G. Taft, "MODU Marine Safety: Structural Inspection and Readiness Surveys," Journal of Petroleum Technology, Vol. 39, No. 11, November 1987, pp. 1437-1440.
- Cosgrove, D. G., "Check-out of the Delta Technique," Report No. DP-70-001, General Dynamics, Convair Division, 19 October 1970. A70-45572.
- Cowfer, C. D. and Owen Hedden, "Overview of the ASME Section XI Code Relating to NDE and Ultrasonic Examination Performance Demonstration," J. Pressure Vessel Technology, Vol. 113, May 1991, pp. 170-173.
- Doggart, Robert R., Private email, 01/03/2000.
- Erwin, W. S., "Vital Measurements of Aircraft Parts Made by Supersonic Measurement, Steel, Vol. 16, No. 10, March 5, 1945, pp. 131, 188, 190, 192.
- Firestone, F. A., and J. R. Frederick, " Refinements in Supersonic Reflectoscopy. Polarized Sound," The Journal of the Acoustical Society of America, Vol. 18, No. 1, July 1946, pp. 200-211.
- Norris, T.H., "European Cooperation in NDT," Reliability in Nondestructive Testing, Eds. C. Brook and P. Hanstead, "The President's Honor Lecture, Proceedings of the 27th Annual British Conference on NDT, Portsmouth, UK, 12-15 September 1988, Oxford, UK, Pergamon Press.
- Papadakis, E., "Research and Real-World Relationships," Materials Evaluation, Vol. 50, No. 3, March 1992, p. 352.
- Pruitt, D. and Lebsack, S., "Return on Investment from NDE and Maintenance Expenses - A Case History," Materials Evaluation, Vol. 52, No. 11, November 1994, pp. 1279-1281.
- Silk, M. G., and B. H. Liddington, "The Potential of Diffracted Ultrasound in the Determination of Crack Depth," Nondestructive Testing, Vol.8. June 1975, pp. 146-151.
- The Alfred P. Sloan Foundation/American Society of Mechanical Engineers, Mechanical Engineering, 1996.
- Wells, C.D., "The Commercial Role of NDT in the Context of A Changing World" Insight, Vol. 36, No. 5, May 1994, pp. 334-341.
- Whittington, K. R., "Electronic Steering and Focusing of Ultrasound for Tube Testing,"
Chapter 4, in Research Techniques in Nondestructive Testing, Vol. III, R. S. Sharpe, Ed., 1977, pp. 135-173.
- Wildavsky, Ben, "America's Best Colleges - Business and Engineering," U.S. News & World Report, Vol. 127, No. 8, August 30, 1999, p. 105.