The American Society for Nondestructive Testing   

Marketplace

Certification

Publications

Membership

Meetings and
Events

Local Sections

Links

What's New

Learning

Awards and
Honors

Advertise
 


 

Members Only | Contact Us | ShopASNT | Search   

 

Back to Basics

[ click here for the Back to Basics Archive ]

Loading and Stress in
Acoustic Emission Testing

by Adrian A. Pollock*

 

Here is a rare contribution to "Back to Basics": it discusses some basics of acoustic emission testing (AE) and does so in a readable fashion without mathematics! It explains a lot of the what and why of AE procedures. A really pleasant tutorial paper.

Frank Iddings
Tutorial Projects Editor

 


Figure 1-3
Figure 4-6
Figure 7-9

LOADS, STRESSES AND DISCONTINUITIES 

Acoustic Emission: Testing with Stress
All nondestructive testing (NDT) methods rely on some kind of energy input. It may be the energy of sunlight, as in visual or dye penetrant testing. It may be energy supplied by the NDT equipment, as in radiographic, ultrasonic and eddy current testing. In magnetic particle testing, it is the energy of the magnetic field; in infrared testing, it is heat; in vibration analysis, it is the energy of the motor. In acoustic emission testing (AE), the required energy input is mechanical stress.

In AE, the structure is stressed by an applied load. External forces or internal pressures are used to make discontinuities emit stress waves. These acoustic emissions are like miniature earthquakes. They carry acoustic energy at all frequencies from far below the audible range to far above it. By listening at frequencies between the audible and ultrasonic ranges, we can detect discontinuity growth and other processes related to structural integrity (Pollock, 1989).

The overall process for AE is shown in Figure 1. The applied loading - force, pressure or even a thermal gradient - produces a stress field in the test area. The stress field stimulates discontinuities to generate acoustic emissions. The acoustic emissions are used for nondestructive testing of the structure. The test load and the stress field in AE are just as crucial as the source in radiographic testing or the search unit in ultrasonic testing. In this article, we will first discuss the nature of stress. We will then show how the test load and the loading schedule are designed into various test procedures to get a practical and cost effective nondestructive test.

Because it uses stress as the stimulus, AE has some special capabilities in the world of NDT.

Because it uses stress as the stimulus, AE has some special capabilities in the world of NDT. After all, it is stress that causes cracks to grow and structures to fail. Being intimately connected to stress, AE is also intimately connected to structural failure. The immediacy of this connection gives AE unusual qualifications for predicting and guarding against failure and for directly assessing structural integrity.

 

The Nature of Stress
What exactly is stress? By definition, mechanical stress σ refers to the internal forces F within a body, normalized by area A:

(1)

Equation 1 shows stress in a material that is calculated by dividing the applied force by the area upon which it acts. As an example, Figure 2 shows a compressive force of 400 kN (90 000 lbs), acting on the top face of a 152 mm (6 in.) concrete cube resting on the floor. Let us assume that this force is distributed evenly over the surface area which is 152 by 152 mm = 0.023 m2 (6 by 6 in. = 36 in.2). The stress is then 400 kN/0.023 m2 = 17.4 MPa (90000 lbs/36 in.2 = 2500 lb/in.2).

Note that the measurement unit for stress in SI is pascals (1 Pa = 1 N/m2). In the imperial system, the unit is pounds per square inch (lb/in.2). Thus, stress in a solid is akin to pressure in a liquid or gas - they are measured in the same units. However there is a key difference. Pressure in fluids is the same in all directions, but stress in solids is rich in directionality.

When you stand on a chair, some parts of the chair are compressed, some parts are bent and some parts carry load in shear. Every part stretches, compresses, shears or twists a little, passing the load along in an attempt to establish and maintain equilibrium. We say there is a stress field created by the load on the chair. At every point in the chair, there is a local balance of forces. A microscopic cube inside the material of the chair has elastic forces acting on all its faces, as shown in Figure 3. These forces (divided by the corresponding areas) constitute the local stress field.

In Figure 3 there are 18 distinct forces - three on each of the cube's six faces. But, if we want the cube only to stretch elastically, not to start moving or spinning, there has to be a balance among these forces. We cannot pick every one at will. Analysis shows that there are really only six independent, freely selectable components underlying the 18 forces shown in Figure 3. The diagram shows three pairs of equal and opposite tensile or compression forces (black arrows). It also shows three systems of four equal shear forces (red, blue and magenta arrows).

Other perspectives can be adopted to envision the stress field in other ways. For example, it can be regarded as a combination of volumetric compression/dilatation and specifically oriented shear. A third common way of viewing a stress field is to rotate our viewing angle to align with the natural orientation of the stress field itself. When we take this viewpoint, the shear components disappear and we only see tension/compression components. These are the three so called principal stresses. A good example is the well known analysis of stress in the cylindrical shells of thin walled pressure vessels. We choose our axes parallel to the length and circumference of the vessel. When we do this, the first two principal stresses are the well known hoop and axial stresses. The hoop stress is pr/t and the axial stress is pr/2t, where p is the pressure, r is the radius and t is the thickness of the vessel wall. The third principal stress, directed through the thickness of the vessel wall, approaches zero when the vessel wall is thin compared to the radius.

Stresses cause the material to stretch elastically. This elastic stretching is called strain. In equilibrium, there is a fixed proportionality between stress and strain. This was observed by Robert Hooke in the 17th century and is known as Hooke's law. Internal stresses and strains in a body arrange themselves so as to balance the external forces and the internal pressure. Stresses and strains are pervasive; they are present throughout the loaded structure. In very simple cases, such as the cylindrical shell of a thin walled pressure vessel, the stress field may be the same at every point. In general, however, the stress field varies from place to place, in both magnitude and directionality. Going from the cylindrical shell of our pressure vessel into the closure heads, the stress field changes along with the transition in geometry. In more complex structures such as a steel highway bridge, the stress field has steep gradients and complex variations, especially around joints, welded details and other geometrical discontinuities. The linear relationship between stress and strain extends to all three dimensions: when you pull a wire, it gets not only longer but also thinner.

Discontinuities will emit according to the local three dimensional stress field in their immediate neighborhood. They do not know about the stress field in the distance. In fact, because they are geometric discontinuities, they will often aggravate the stress field. They serve as stress raisers. The stress at the tip of a cracklike discontinuity can easily be 10 to 20 times higher than the nominal stress in the material at that point. This is why cracks and discontinuities emit while nearby material free of discontinuities is silent.

What does this mean to us as practical NDT inspectors? It means that we should practice awareness of the stresses in the structures we are testing, as much as we can. First of all, the magnitude of the local stress will always be proportional to the magnitude of the applied forces (as long as the structure behaves elastically). However, an important tip I want to give here is to pay attention to the directionality, not only to the magnitude of the applied stress. Envision the directions of the forces in terms of tension, compression and shear, and see how these stresses relate to the orientation of actual or possible discontinuities. How will the discontinuities tend to grow? Picturing stress fields will come with practice and a little guidance and discussion - most people have a good intuitive sense for it. Intuitive terms like bending are helpful as well. Bending actually involves tension, compression and shear in a specific combination. Getting a sense for the stresses at work and maybe even undertaking some simple calculations, will add insight and satisfaction to AE.

A very important aspect of this is that in many kinds of structure, there are places where the stress is low or even approaches zero. These unstressed areas will not emit. This is one of the key limitations of AE. We say that with AE, the whole structure is tested in one operation. This is certainly an advantage, as long as we do not forget that we need a sufficiently high stress for structurally significant discontinuities to emit. In many test procedures, a sufficiently high stress is achieved by applying a load that is clearly higher than the service loads. In some other procedures, the presence of a sufficiently high stress is demonstrated by analysis.

 

Stress and Discontinuities
The stress at a cracklike discontinuity depends on its orientation relative to the prevailing stress field. This effect is shown in Figure 4. If the crack lies along the stress direction, the stress concentration is small. If the crack lies across the stress direction, the stress concentration at the tips of the crack can be very large. In fact, the material at the tip of such a crack will yield, bluntening the crack until the load is distributed in a stable manner. This yielding of the material forms a plastic zone. These effects are shown in Figure 4. Formation and growth of the plastic zone is an important source of acoustic emission.

There is another important source of acoustic emission: the forward movement of the crack tip itself. Here again, crack orientation is a significant factor. Three modes of crack growth have been defined in the literature, as shown in Figure 5. They are known as the opening, sliding and tearing modes. All three modes are important in practice. If one's thinking is limited to the opening mode, many field situations are puzzling. Enlightenment often comes when you realize that the crack is growing in one of the other modes. To understand a crack, try to envision the directionality of the stress field and figure out what is the mode of crack growth.

 

ACOUSTIC EMISSION TEST LOADINGS

Variety of Loading Techniques
Acoustic emission test loads are applied to structures in various ways, including:

  • hydrostatic/hydraulic/pneumatic loading: pressure vessels are loaded by applying internal pressure. Storage tanks are loaded by filling with liquid. Sometimes, storage tanks have a small internal pressure as well.
  • mechanical loading: bucket trucks are loaded by pulling down on the bucket or points near it. Railroad tank cars are loaded by jacking them up at specified points in carefully designed ways. Aircraft structures during structural design tests are loaded with elaborate arrangements of hydraulic jacks and "wiffle trees."
  • thermal loading: some of the largest acoustic emission field tests, using upwards of 1000 sensors, have been performed by monitoring refinery plants during cool down for scheduled outage. Temperature gradients produce time dependent stress/strain fields that are governed also by structural geometry and mechanical constraints.
  • normal in service loading: the structures are monitored during normal or service conditions. The cool down tests mentioned in the previous paragraph can also be considered a special case of this kind of loading.

Four aspects of test loading that have to be considered in setting up an acoustic emission test procedure are:

  • the operating condition and configuration of the test structure during loading
  • the kind of loading and the loading connections
  • the magnitude of loading (the test load)
  • the timeframe and time sequence of loading (the load schedule).

The loading procedure design has to be practical, cost effective and supportable in terms of the principles of acoustic emission source behavior. Ultimately, it has to be proven against the touchstone of practical results.

When possible, the test loads applied to the structure should be of the same general form as the loads experienced in service (these are presumably the loads that are causing the discontinuities to grow in the first place). A discontinuity stressed in the wrong direction might not emit. Ideally, the test stress field will match the service stress field in directionality and exceed it in magnitude. This is the scenario that will give the right stimulus to the discontinuities we are looking for. Figure 6 shows a load line being attached near the end of a bucket truck upper boom. Loading at this point will realistically test the boom, the elbow and the other load bearing components all the way to the attachment to the chassis. However, this loading does not test the attachment of the bucket to the upper boom. For that, a test load must be applied to the bucket itself.

Acoustic emission test procedures can be divided into controlled and uncontrolled load categories. These will be discussed in the upcoming sections. In controlled load tests, it is very important for the acoustic emission inspector to take responsibility and ensure that the loading is carried out according to procedure. This is one of the most important onsite responsibilities of the acoustic emission operator. Incorrect loading can compromise interpretation and testing and, in the worst case, can even make the test worthless.

Loading schedules for six well established controlled load tests are shown in Figure 7. These graphs show applied load versus time. Where the load line is dotted, it means that acoustic emissions are not monitored during that portion of the loading. The variety of loading schedules may be surprising at first sight. Loading schedules vary for several reasons: practicality, noise considerations and considerations of material behavior. Material behavior will now be discussed in general terms before going on to describe test loads and schedules for the particular test types.

 

Underlying Materials Behavior
Several principles of material behavior underlie the choice of loading technique (Spanner et al., 1987). The first of these is the Kaiser principle (or effect), which states, "materials emit only under unprecedented stress" (that is, they emit only when the previous maximum stress is exceeded). This is a good first approximation to real material behavior. However, three other phenomena that run counter to the Kaiser principle are also of great importance in AE. First, when the load is raised and then held at a constant high level, emission often continues for a while. This shows that the material is taking time to stabilize. This behavior is an indicator of high stress and discontinuities. Second, emission will sometimes partially repeat when the load is removed and then reapplied a second time (felicity effect). Like emission during hold, this can be a consequence of material taking time to stabilize or it can be a consequence of rubbing at damage sites within the material. The third phenomenon that runs counter to Kaiser's concept is fatigue, that is, progressive crack growth as a result of cyclic loading. During fatigue, acoustic emissions are generated both by rubbing of the crack surfaces and by the fracture of locally brittle microstructures, at or near the crack front as it advances.

With these subtleties of materials emissivity in play, it is not surprising that several different loading strategies for AE have been introduced and successfully applied as the technology has matured.

 

Examples: Controlled Load Strategies for Pressure Vessels and Tanks
The first widely used strategy for acoustic emission field test loadings was based on the Dunegan corollary to the Kaiser effect (Spanner et al., 1987). Under this concept, an acoustic emission overload test will reveal damage that occurred in service subsequent to the last overload test. Thus, an effective acoustic emission test on a pressure vessel can be obtained by taking it, say, 10% higher than the working pressure. Throughout the vessel, the stress will be higher than its service levels. Any cracks or discontinuities that had grown in service will now be subject to unprecedentedly high stress and will produce acoustic emission.

This strategy had much success at the time of its introduction and became the conceptual basis for many further developments. The actual amounts of overload varied. The acoustic emission test planners might have asked for higher overload margins than the vessel owners were willing to provide and compromises would be negotiated, typically in the range of 5% to 20% in industrial storage tanks, pressure vessels and piping. Sometimes there would be a regulatory requalification test and the acoustic emission monitoring would be conducted in conjunction with that.

Figure 7a shows a classic, simple loading schedule used to test high pressure "jumbo" tubes in the compressed gas industry per ASTM E 1419 (ASTM, 2000). The tube is monitored while filling with product, to a test pressure 10% above the normal fill pressure (Green et al., 1987). The test is conducted at high sensitivity and at low pressures there is much noise from the gas rushing into the vessel. Therefore, acoustic emission monitoring only starts at about 40% of the test pressure. The filling of the 9.1 m (30 ft) long tube to 20 MPa (2900 lb/in.2) takes several hours, so there is plenty of time for the material to equilibrate. Including the hold period, the total time above the service pressure is on the order of 1 h.

Tanks and vessels have relatively simple stress fields and their service load histories can normally be determined quite well from the owner's records. As acoustic emission technology matured, detailed loading schedules were laid down within test procedures such as the Committee on Acoustic Emission from Reinforced Plastics procedure for fiberglass reinforced plastic vessels (Green, 1987; ASTM, 1996) and the procedure for metal vessels described by Fowler et al. (1989). These procedures stated what levels of overload were needed relative to service load histories and design parameters. Coordinated with specified equipment setups and specified test techniques, these test loads give effective tests on the structures that fall within the scope of the procedures.

Figures 7b and 7c show the ASME loading schedules for acoustic emission tests on fiberglass and metal pressure vessels, respectively. These schedules are carefully designed in the light of commonly observed acoustic emission behavior of these structures. In the case of fiberglass vessels, the schedule comprises a series of rising up and down steps with hold periods. This allows the acoustic emission during hold and the Felicity ratio (Green et al., 1987) to be measured and factored into the test. These effects are very prominent in fiberglass. In metal vessels, however, a more important structural behavior is the stress relieving of weldments (if there has been no thermal stress relief before the acoustic emission test). This creates much acoustic emission even when the weldments are good. It may, therefore, be necessary to perform a second load cycle in which the stress relieved good weldments will be quiet but any structurally significant discontinuities will emit again.

In one way, pressure vessels are easier to test than storage tanks. Most pressure vessels operate at essentially constant pressure or between well prescribed pressure levels, so it is easy to determine the appropriate test pressure. Storage tanks, however, are likely to have gone through a quite complex history of filling and emptying. Test planning for storage tanks must include consideration of these historic load levels, of the resources available for filling and of the level to which the tank is being qualified. These factors need to be worked out carefully before the acoustic emission test, in accordance with the test procedure being used.

 

Examples: Controlled Load Strategies for Other Structures
In many types of structures, it is not possible to apply the idea of a definite percentage overload above normal working conditions. A typical example is found in railroad tank cars. Here, fatigue and corrosion are common, while the actual loads and stresses seen during service are largely unknown. Still, several generations of acoustic emission test procedures have been developed and applied very successfully (Association of American Railroads, 1999a; 1999b; 2002). Available procedures include a pressure test and several kinds of mechanical loading. The introduction of different kinds of loading to meet the stressing needs of different parts on the same structure was an important innovation. The loading schedules are shown in Figures 7d and 7e. In the pressure test, which tests primarily the tank, the pressure is ramped up with holds at 50% of the test pressure (for 10 min) and 100% of the test pressure (for 30 min). In the mechanical loading tests, which test primarily the underframe and sill, the car is jacked rapidly to the final position and then held. In all these tests, only the emission during load holds is recorded and tested. This design choice maximized the structural significance of the test data. It is important to have adequate jacking speed and good synchronization between the end of the jacking and the start of monitoring. A jacking test is illustrated in Figure 8.

Since knowledge of the actual service stresses in tank cars is limited, it is a real challenge to validate the mechanical loads used to stimulate the car underframes. The test procedure called for jacking the car to specified deflections. Finite element analysis and strain gages were used to verify that these deflections produced high enough stresses relative to material strength. This analysis gave assurance that structurally significant discontinuities in the test area would emit. This assurance contributed significantly to regulatory acceptance of the acoustic emission method for tank car testing.

In the electric utility industry, acoustic emission testing is widely used as part of the mandatory annual testing of the bucket trucks that are used by linemen to maintain overhead power lines (Hutton et al., 1987). The actual service loads seen by bucket trucks are more or less unknown, but their load capacities are well defined by the manufacturers. Acoustic emission test loads are specified as multiples of these load capacities. The pertinent ASTM standards use factors of 1.5 or 2, much larger than the factors of about 1.1 that are common in pressure vessel and tank testing. These high test loads are viable for bucket trucks because the safety factors used for designing these devices are much larger still. Testing at these relatively high test loads, the immediate effects of recent service history are somewhat abated. However, care must be taken. The inspector has to read and interpret the capacity charts correctly and not damage the truck by applying too high a load.

The bucket truck test is quite quick once the setup is complete. The actual loading takes 20 to 25 min and follows the schedule shown in Figure 7f. Like most schedules, it begins with a period monitoring at zero or low load to check that the background noise level is acceptable. Then, there are two loadings with an intervening pause for the truck to relax. Acoustic emissions are monitored throughout the whole process. With this schedule, the operator has good visibility of emissions during holds, Kaiser/Felicity behavior and frictional sources during unloading. Interestingly, data interpretation and testing have come to rely largely on the second loading. The first loading can almost be conceived as a preload to standardize the condition of trucks that have seen widely varying loads in service.

 

Examples: Uncontrolled Load Test Strategies
For some structures, the application of a controlled overload is very difficult. An obvious example is highway bridges. Although there have been projects in which an extra heavy load has been driven over a bridge to test it, this is not the preferred test mode. Instead, monitoring during normal traffic conditions became the normal approach after it was found that this gave valid and useful results. One specific test procedure involves monitoring the area of interest for just one hour. To develop this procedure, spatial filtering and guard sensor techniques were used in a series of field demonstrations on a number of bridges to monitor an assortment of cracks, retrofits and ultrasonic indications (see Figure 9). Acoustic emission activity from these areas of interest correlated well with the state of the discontinuities being monitored. The procedure includes a comparison table that can help a bridge engineer assess the significance of a questionable discontinuity.

This test is based on the principles of acoustic emission monitoring during fatigue crack growth. Fatigue crack growth produces acoustic emission both from friction at the crack surfaces and, in smaller quantities, from incremental crack growth. The acoustic emission from friction is especially strong in the sliding and tearing modes (see Figure 5). Because it indicates movement, acoustic emission is taken as a bad indication regardless of whether it comes from friction or new crack growth. Extra emission is observed when heavy loads go over the bridge. This corresponds to the fact that the heavy loads are the ones that cause structural damage.

Another case of uncontrolled loading is the in service testing of above ground storage tanks. The leading technique utilizes acoustic emission from active corrosion to assess the condition of the tank floor (Mathers, 1998). It is remarkable that the stress is produced by the corrosion process itself - oxidation of steel produces expansion and the expansion generates stress. This is an uncontrolled process; no external loading is needed to detect it. As with highway bridges, the key is to be monitoring while the stepwise, progressive damage is actually taking place. Maintenance cost savings run to many millions of dollars.

 

Structural Health Monitoring
Structural health monitoring can be described as a new way of doing NDT, in which the sensors are permanently installed on or in the structure (Beral and Speckmann, 2003). Especially in the aircraft and spacecraft industries, structural health monitoring is widely regarded as a key way to improve competitive performance by reducing maintenance and operational costs. Structural health monitoring technology considers both active and passive sensors and both online and offline systems.

It may be a surprise to find that structural health monitoring by means of acoustic emission has been undertaken in both online and offline modes, using every one of the loading categories discussed in the previous sections. What first comes to mind is continuous, online monitoring under planned, controlled service loading. This has been undertaken for leak detection in chemical and nuclear plants and for crack growth detection during numerous structural fatigue tests of costly aerospace test articles, both metal and composite.

We can also find examples of online structural health monitoring/acoustic emission using normal but uncontrolled service loading. The first program of this type (in the early 1980s) was a simple, safety oriented AE system for bucket trucks. A single sensor was permanently mounted on each truck at a safety critical point. Today, more complex systems are in use, with more than 50 acoustic emission sensors permanently installed for structural health monitoring on an offshore platform and on a suspension bridge. Acoustic emission data have even been acquired during flight testing of a reusable launch vehicle technology demonstrator (Goggin et al., 2003).

An example of offline (on demand) structural health monitoring/AE is the 96 sensors (piezoelectric acoustic emission transducers and cables) permanently mounted on an ammonia storage tank in the mid 1980s. These sensors were installed for use during plant outages, when the structure would be monitored during a controlled overfill. In a similar but more recent program ongoing in Europe, acoustic emission sensors are permanently installed in large, buried storage tanks to permit acoustic emission tests for periodic requalification.

Offline structural health monitoring using controlled loading is further illustrated by electric utility programs where acoustic emission sensors are permanently installed on hot reheat lines. Structural health monitoring is routinely conducted using these sensors to acquire acoustic emission test data during plant startup, cool down and controlled overpressure. These same sensors are sometimes used to take data during a few days of normal online operation. During these tests, the operating load fluctuates according to production demands, so it is essentially uncontrolled from the test standpoint.

These examples show that there is a significant body of experience in structural health monitoring/AE, going back 20 years, even though these projects were not always conceived in structural health monitoring terms. They show also that loading techniques for structural health monitoring/AE vary just as widely as loading techniques for the acoustic emission tests that are done by field test crews using temporarily mounted sensors.

 

SUMMARY AND CONCLUSIONS
Stress is the cause of acoustic emission indications. It is as important as the radioactive source in RT or the magnetic field in MT. Loading of the structure (mechanical, hydraulic/pneumatic or thermal) produces a stress field which is further aggravated by discontinuities, causing them to emit.

The type and timing of the loading for an acoustic emission test is determined on the basis of material characteristics, service loading conditions and practical feasibility. It is common to use a test loading that mimics the service loading but is a little more stressful. There are predetermined loading schedules to be followed. These schedules are planned in light of acoustic emission behavioral characteristics such as emission during hold and the Kaiser and Felicity effects.

The magnitude of the test load is often specified as a multiple of the service loading. The underlying thinking here is to overcome the effects of recent service history and to stimulate discontinuities that have been growing in service. In structures like bridges, however, controlled loading may not be practical. Such structures are monitored under normal operating conditions, looking for indications of fatigue or corrosion.

Several important variants in loading technique have been discussed and many examples given. The loading technique has to be harmonized with the test setup and evaluation criteria to give an effective and accurate test. Good validated test procedures and competent execution of the loading by the acoustic emission test technicians are keys to test success.

Sometimes the acoustic emission data are recorded with permanently installed sensors. This is the domain of structural health monitoring. Examples of structural health monitoring/AE applications were collected. It was found that they exhibited the same range of loading techniques as the acoustic emission applications using temporarily installed sensors.

The inspector's increasing responsibility for the structural loading has been one of the major growth vectors in acoustic emission technology. In the early days, acoustic emission pioneers had no control over the loading process. They were only allowed to put their sensors on and monitor whatever was going on. Now, AE is well recognized in everyday NDT. It is a commonplace for test loads, loading rates, load holds and repeat cycles to be controlled in accordance with acoustic emission test procedures, often by the inspector directly. As we have seen in this article, this flows logically from the technical basics of the method.

 

REFERENCES
ASTM International, E 1067-96: Standard Practice for Acoustic Emission Examination of Fiberglass Reinforced Plastic Resin (FRP) Tanks/Vessels, West Conshohocken, Pennsylvania, ASTM International, 1996.

ASTM International, E 1419-00: Standard Test Method for Examination of Seamless, Gas-filled, Pressure Vessels Using Acoustic Emission, West Conshohocken, Pennsylvania, ASTM International, 2000.

Association of American Railroads, Procedure for Acoustic Emission Evaluation of Tank Cars and IM101 Tanks, Issue 8, Washington, DC, Association of American Railroads, October 1999a.

Association of American Railroads, Annex Z to the Procedure for Acoustic Emission Evaluation of Tank Cars and IM101 Tanks, Issue 6, Washington, DC, Association of American Railroads, October 1999b.

Association of American Railroads, Procedure for Structural Integrity Inspection of Tank Cars Using Acoustic Emission, Issue 1, Revision 2, Washington, DC, Association of American Railroads, August 2002.

Beral, B. and H. Speckmann, "Structural Health Monitoring (SHM) for Aircraft Structures: A Challenge for System Developers and Aircraft Manufacturers," 2003 International Workshop on Structural Health Monitoring, Stanford University, September 2003.

Fowler, T.J., J.A. Blessing, P.J. Conlisk and T.L. Swanson, "The MONPAC Procedure," Journal of Acoustic Emission, Vol. 8, No. 3, 1989, pp. 1-8.

Goggin, P., J. Huang, E. White and E. Haugse, "Challenges for SHM Transition to Future Aerospace Systems," 2003 International Workshop on Structural Health Monitoring, Stanford University, September 2003.

Green, A., P.R. Blackburn, B. Craig, N.O. Cross, M. Ferdinand, T. Fowler and D. Robinson, "Acoustic Emission Applications in the Petroleum and Chemical Industries," Nondestructive Testing Handbook, second edition: Volume 5, Acoustic Emission Testing, P. McIntire and R.K. Miller, eds., Columbus, Ohio, American Society for Nondestructive Testing, 1987, pp. 155-224.

Hutton, P.H., J.A. Baron, C.E. Coleman, T. Kishi, H. Nakasa and P. Ying, "Acoustic Emission Applications in the Nuclear and Utilities Industries," Nondestructive Testing Handbook, second edition: Volume 5, Acoustic Emission Testing, P. McIntire and R.K. Miller, eds., Columbus, Ohio, American Society for Nondestructive Testing, 1987, pp. 225-274.

Mathers, P., "Million Dollar Savings: You Just Have to Listen," Bulk Distributor, Vol. 10, No. 5, June 1998, p. 22.

Pollock, A.A., "Acoustic Emission Inspection," Metals Handbook, ninth edition, Vol. 17, Materials Park, Ohio, ASM International, 1989, pp. 278-294.

Spanner, J.C., Sr., A. Brown, D.R. Hay, V. Mustafa, K. Notvest and A. Pollock, "Fundamentals of Acoustic Emission Testing," Nondestructive Testing Handbook, second edition: Volume 5, Acoustic Emission Testing, P. McIntire and R.K. Miller, eds., Columbus, Ohio, American Society for Nondestructive Testing, 1987, pp. 11-44.

 

* Physical Acoustics Corporation, 195 Clarksville Rd., Princeton, NJ 08550; (609) 716-4000; fax (609) 716-4057; e-mail <apollock@pacndt.com>.

 

Copyright © 2004 by the American Society for Nondestructive Testing, Inc. All rights reserved.

 
 
Copyright © 2008 by the American Society for Nondestructive Testing, Inc. ASNT is not responsible for the authenticity or accuracy of information herein. Published opinions and statements do not necessarily reflect the opinion of ASNT. Products or services that are advertised or mentioned do not carry the endorsement or recommendation of ASNT.

IRRSP, NDT Handbook, The NDT Technician and www.asnt.org are trademarks of the American Society for Nondestructive Testing, Inc. ACCP, ASNT, Level III Study Guide, Materials Evaluation, Nondestructive Testing Handbook, Research in Nondestructive Evaluation and RNDE are registered trademarks of the American Society for Nondestructive Testing, Inc.

ASNT exists to create a safer world by promoting the profession and technologies of nondestructive testing.