Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

NE 5599 Advanced Material Science Non-Destructive Testing Review

Written by: Eric S. Krage Instructors: Dr. Maria Okuniewski & Dr. Sebastien Teysseyre

Page | 1

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Abstract
Non Destructive testing is a set of analysis techniques used by industry and scientists to evaluate and quantify material properties. The use of acoustic emission is a passive technique that takes advantage of waves generated in a material by a microstructural change. The analysis of the generated waveform provides information about what types of changes are occurring; and by the use of multiple sensors the location of the event can be quantified. There are multiple different types of wave that are able to be generated but not limited to longitudinal, shear, and surface waves. Although, not all waves are supported in all types of media such as a longitudinal wave is the only type of wave supported in a liquid. While all types of waves are supported in a continuous solid. The use of surface waves is a way to actively detect surface defects in a material. Surface waves can be guided along at a prescribed depth and subsequently received by a receiver. The effect of a rubber wedge in contact with a metal surface was evaluated close to the transducer and receiver and was found to have different effects on the signal attenuation. Surface waves were also used to show that they can be generated on the more complex geometry of a curved surface and still resolve a surface inclusion. The future of non-destructive testing is vast and has the ability to actively and passively monitor for stress corrosion cracking events that are prevalent in the nuclear industry.

Page | 2

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Background
Non-destructive Testing (NDT) is a wide group of analysis techniques used in science research and industry to evaluate the properties of a material, component or system without causing damage.1 Because NDT does not permanently alter the article being inspected, it is a highly valuable technique that can save both money and time in product evaluation, troubleshooting and research. Common nondestructive testing methods include ultrasonic, magnetic-particle, liquid penetrant, radiographic, and eddy-current testing. The majority of these techniques can be used actively to scan and test the material as well as passively evaluate a material under testing conditions. The focus of this work is to implement active and passive ultrasonic testing to search for stress corrosion cracking events. The ultrasonic testing method was developed by Dr. Floyd Firestone between 1940 and 1944. His original patent pertained to a device for detecting the presence of inhomogeneities of density or elasticity in materials. His device allowed the presence of a flaw to be detected and its position located, even though the flaw lies entirely within a casting and no portion of it extends out of the surface. The general principal consisted of sending high frequency vibrations in the part to be inspected, and determination of the time intervals of arrival of the direct and reflected vibrations at one or more stations on the surface of the part.2 Acoustic emission is the sound waves produce when a material undergoes stress, as a result of an external force applied. The ultrasonic elastic waves generated from such an event occur because of a small surface displacement of a material produced due to stress waves. The subsequent wave generated by the source is of practical interest in methods used to simulate and capture acoustic emissions in a controlled fashion, for study and or use in inspections, quality control, system feedback, and process monitoring.3 Stress corrosion cracking (SCC) is defined as the growth and generation of cracks in a corrosive environment. Stress corrosion cracking can lead too sudden failure of normally ductile metals subjected to a tensile stress, especially at elevated temperature in the case of metals. This failure mechanism is caused by the simultaneous contribution of three factors. The first factor is the environment in which the material is placed such as a chlorinated or ammonia bearing environment. The second contribution to stress corrosion cracking is the susceptible material. The most prominent material is alloyed materials when compared to pure metals with their natural impurities because of their electronic configuration.5 The third contribution to stress corrosion cracking is the tensile stress applied to the material. The stresses present in the material can be the result of crevice loads due to stress concentration, or can be cause by the type of assembly which may result in residual stress, this residual stress then can be subsequently reduced by annealing the material.

Page | 1

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Theoretical Information Acoustic Emission

As previously discussed an acoustic emission can result from deformation of a material under stress or strain. Crack growth, slip and dislocations movements as well as twinning and phase transformations in metals can generate an acoustic event. When a stress is impinged on a material, a strain is subsequently introduced into the material as well. The magnitude of the stress applied and the property of the material, the object may return to its original dimensions or be permanently deformed after the stress is removed. These two types of conditions are known as elastic and plastic deformation, respectively. Other effects during subsequent plastic deformations can change the threshold to which an acoustic event will be generated. The most evident acoustic emissions take place when loaded material undergoes plastic deformation or when a material is loaded at or near its yield stress. The yield stress is defined by the amount of stress applied to a material that causes a permanent deformation of the material. In observation of plastic deformation at the microscopic level atomic planes slip past one another through the movement of the dislocations. The atomic-scale deformations release energy in the form of elastic waves which are essentially naturally generated ultrasonic waves that travel through the object. When cracks exist in a crystalline material, the stress levels present in front of the crack tip can be several times higher than the surrounding area. This difference will cause acoustic emission activity to be observed when the material of the crack tip undergoes plastic deformation otherwise thought of as micro-yielding. Two sources of fatigue cracks also cause acoustic emissions. The first is emissive particles such as non-metallic inclusions at the origin of the crack tip. Since these particles are less ductile than the surrounding material, they tend to break more readily when the material is strained which results in an acoustic emission signal. The second source is the propagation of the crack tip that occurs through the movement of dislocations and small-scale cleavage produced by triaxial stresses. The energy released by an acoustic emission and the amplitude of the resolved waveform are directly related to the magnitude and velocity of the source event. The amplitude of the emission is proportional to the velocity of crack propagations and amount of surface area created. Large, discrete crack jumps will produce large acoustic emission signals than cracks that propagate slowly over the same distance. The detection and conversion of these elastic waves to electrical signals that can be analyzed is the complicated part of the acoustic process. In depth analysis of the signals produced yield information regarding the origin and importance of a discontinuity in a material. Acoustic emission events can be generated by different loading situations and can provide information regarding the structural integrity of the material under evaluation. Loading levels that have been previously observed by the material do not produce acoustic emission activity. Therefore, the discontinuities created in a material do not expand or move until the previous stress has been exceeded by the subsequent stress known as the Kaiser effect. The Kaiser effect can be seen in the load versus acoustic emission plot shown in the figure below. As loading is applied on the material acoustic emission events occur in a linear fashion represented in the figure by segment AB. The segment shown in C represents loading being removed and reapplied which exhibits no effect until the previous loading potential is exceeded. As the load applied to the material is increased from the region of BD acoustic events are generated and stop when the load is removed. However there is an exception to the rule at which can be seen at F where the load is high enough to cause significant emissions even though the previous maximum load at

Page | 2

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

point D had been reached called the Felicity effect. The Felicity effect is used in terms of the Felicity ration which is described as the load where a considerable increase in acoustic emissions resume divided by the maximum applied load (F/D). Using information gained from the Felicity and Kaiser Effect during material testing and evaluation can be used to determine if significant material defects are present. To measure a constant load is applied to the material with respect to the specific material design and application and continuously search for acoustic emission events. The acoustic emissions continue when the prescribed loading tension is held as shown by section GH of the curve. In this region it can be assumed that substantial structural defects are present in the material. The reapplication of such a load may cause critical failures of the material.

Figure 1: Loading Potential vs. Acoustic Emission Event Rate.

Some known issues with acoustic emission testing for stress corrosion cracking evaluation is that it is a passive measurement. So, in other words it is not useful to actively look for a crack that has already started but not currently propagating. This leads to issues with implementation in real world settings that were not originally fitted with acoustic emission listening devices. Other such shortfalls occur when hydrogen embrittlement occur in the material as it causes poor support for waves to propagate through the medium. This occurs because the crystal structure breaks down and causes the shear modulus to change in such a way that significant dampening of the acoustic event occurs in the material. Other known problems that occur are noise from external sources. As with most testing the signal to noise ratio dominates the performance of the device and as the ratio increases the confidence in the value measured increases. These undesirable signals

Page | 3

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

detected by the sensors include; frictions sources (loose bolts or moving parts), impact sources (falling objects or particulate fluid flow). Other mechanical vibrations from external components of the system such as motors and fans may be seen in the acoustic background. To mitigate the effects of background noise electronics improvements can be made to the sensor signal processing. The addition of coincidence circuits that are in coincidence with a noise source along with signal decomposition to eliminate elements of them are some of the more complex ways to analyze the signals. The simplest method to reduce noise when searching for acoustic emission events is to move the sensors as far away from the sources as possible. One of the key advantages of acoustic emission testing over other nondestructive test methods is the fact that it detects signals in real time that are emanating from the materials themselves. Upon fracture due to the loading potential an acoustic emission stress wave travels from the source through the structure, and can be detected by an acoustic emission senor placed upon the structure. The sensors location may be located near or far from the source and still detect the signal after taking into account the possible self-attenuation in the sample medium. If multiple sensors are placed on the same structure, one may be able to determine the location of the source by analyzing the time difference of arrivals from multiple sensors that have been in place and solving the triangulation calculations. Determining the location in one dimension described by a line a minimum of two sensor arrivals are required to pinpoint the signals origin. While in two dimensions that can be described as a plane and or surface a minimum of three sensors are required. The three dimensional case require a signal reaching a minimum of four sensors to be analyzed. The severity of an event can be analyzed also by monitoring multiple acoustic emission events in the same area and applying a clustered layout of detectors. The advantage to this is large structures can be monitored with a minimal number of acoustic sensors. This is advantages when used in vessels, especially when they are insulated, since few access holes are needed for placement of acoustic emission sensors to determine the structural integrity of the vessel in question. Other nondestructive testing techniques require the complete removal of all the insulation and material around the containment vessel to complete inspections which requires more time to complete while increasing operating costs.

Active Ultrasonic Testing

Wave Propagation Fundamentals

In order to describe waves traveling through media we must first describe the fundamental propagation of a wave on a string. These waves are governed by the wave equation for wave propagation in a taught spring expressed in Equation 1. The equation of motion in the direction of the string is a second-order hyperbolic partial differential equation (PDE) that is homogeneous. This one dimensional, homogeneous, simple wave equation can solve for the velocity of wave propagation.

Use of Newtons second law and assuming small deflections in the string describe the following.

Page | 4

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Which allows us to solve for co which is the velocity of wave propagation?

u = displacement, F = Tension, q(x, t) = body force or external loading per unit length, and = mass density / length Understanding how a wave travels across a taunt string we can apply this information to more complex systems. This development of more complex wave propagation will be these equations of motion for isotropic media are defined by the Naviers governing equations.

Longitudinal Wave Velocity

In order to examine the longitudinal velocity of steel block a 3.5MHz 0.25 diameter. Longitudinal transducer was used. The pulsar/receiver mode was used and set to 3.5MHz to correspond to the signal required by the transducer applying no filter. Using the following equation the longitudinal wave velocity was found to be 6012 m/s where d = 15.40mm.

Shear Wave Velocity

Now the need arises to experimentally measure and calculate the velocity in a steel specimen using the shear wave transducer 5MHz 0.25 dia. The signal was sent and measured using the pulsar/receiver mode with a nominal velocity of 3306 m/s as calculated by the following equation.

Surface Wave Velocity

The surface wave velocity was determined experimentally implementing a 0.5 * 1.0 1MHz angle beam transducer used on pitch and catch mode. The pitch and catch mode causes a short burst of acoustic energy in the sample and waits for a response back until it sends another signal. The signal was received by the longitudinal transducer used in the longitudinal wave velocity measurements. The second transducer was inserted into an angle block to act as a second angle beam transducer to receive the signal. The angle beam transducer was set to generate surface waves and the time between the initial pulse and the received pulse was recorded using the computer display. Using the following equations we calculated the surface wave velocity and critical angle.6 Poissons ratio is found using the following equation. We can now rearrange the equation to solve for Poissons ratio.

Page | 5

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

( (

) )

( (

) )

Now we need solve for using Viktorovs approximation which is less than 1% as shown in homework #3.

We now will calculate for the velocity of the surface wave using the following equation.

We now need to solve for the critical angle in which we generate surface waves. ( )

To experimentally measure the velocity of the surface wave the angle of the angle beam transducer was set to that of the critical angle for surface wave generation. Figure 2 exhibits the experimental procedure and allows for us to utilize two transducers one a transmitter and the other a receiver.7

Figure 2: Surface Wave Velocity Measurement Setup6

Current Research and Uses Acoustic Emission

Currently acoustic emission nondestructive evaluation is used for crack detection, weld analysis, vessel inspection, leak detection, and nuclear lift rig examination. Crack detection is one of the most successful applications and has been accepted and used the longest; with the ability to detect any type of cracking in various materials during bonding, forming, or working the material in some way. When a crack occurs, the acoustic emission system provides a failure output for
Page | 6

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

part rejections. Many acoustic emission detection systems currently operating provide continual real time monitoring. Acoustic emission has also been successfully applied to welding applications for many years. Acoustic emission systems have addressed solutions to real time weld quality, analysis, and weld control, including resistance spot welding. Every burst of an acoustic emission received from the welding process has a logical source and meaning. By focusing in on the expected occurrence of given weld phenomena through time, frequency and feature discrimination. Utilizing this feedback the welding process can be modified to meet the required or requesting weld requirements. Acoustic emission has been able to detect oxidation burn off, nugget formation, hot cracking, expulsion, and post weld tracing. All of these phenomena are very important in weld quality analysis and feedback control for decision making in stopping the weld at the appropriate time.14,15 The welding acoustic emission setups are commonly used in industry in conjunction with robotic welders to be able to reduce errors and further manage the quality control and quality assurance of the facilities. Another very successful application of acoustic emission testing is in vessel inspection for the chemical industry in both metal and composite vessels. Vessel testing is performed every day by many different service companies around the world. Sensors are placed on the vessel in an array to monitor the entire pressure boundary. The vessel is then subjected to pressures typically 10 percent above previous operating levels still below the operating temperature. The test pressures the vessel is being subjected to be done in an increase pressure hold increase and hold again to the peak stress level, while simultaneously monitoring the acoustic emission activity during each of these pressurization segments. This method is the most cost effective and sensitive way to test containment vessels and adheres to the standards set forth by the American Society of Testing and Materials and the American Society of Mechanical Engineers.10 Using acoustic emission instrumentation acoustic events may be detected from turbulent or cavitational flow through a crack, valve, seal or hole. The energy is transmitted as a longitudinal wave through the fluid, and through the air or the structure to the piezo electric sensor mounted on the material. The signal is subsequently processed filtered and compared to a leak profile then located using triangulation techniques. Existing installations include monitoring of pipelines in nuclear and chemical plants as wells as boilers vessels and through valves. These systems offer the capability to connect and monitor multiple sensors throughout a plant the systems can be operated in a standalone mode, interfaced to programmable controllers or tied into plant-wide distribution control systems. They also offer ability to plot plant piping and overlay acoustic event information on top of them to improve repair time.12 One of the most significant nuclear applications is the use of acoustic emission testing on nuclear lift rig examinations in pressure water reactors in containment lift rigs that are utilized during refueling and inspection outages. The internal lift rig is often contaminated so that exposure time to the radioactive material must be minimized which gives rise to an advantage over active forms of testing. The reduction of potential hazard is evident in the time spent in contact with the rig to just place the sensors or to actually physically make measurements in the presence of the potential contamination.

Page | 7

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Using this fundamental idea it is possible to actively search for defects between the transducer and receiver and the setup previously discussed. Since the waves are traveling on the surface to observe an interruption a rubber dampening wedge was placed on the sample block. The wedge was placed in two locations on the sample block one close to the transmitter and the other close to the receiver. A decrease was noticed in the received amplitude of the signal in both cases. However, a bigger decrease was observed when the wedge was placed closer to the receiver as depicted in Figure 3 and 4 which can be prepared to observe this phenomenon we can use Figure 4 as the control image. This phenomenon can be explained that the waves have not fully propagate into the surface traveling as deeper Raleigh waves and thus are more easily attenuated at the transmitter side of the apparatus.

Figure 3: Wedge near the transmitter note the amplitude ~ 40

Page | 8

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Figure 4: Wedge near the receiver note the amplitude ~ 60 Understanding this phenomenon of surface waves between two transducers allows its application in more complex geometries. That can be used for finding defects or obstructions on plate surfaces by scanning between the two surfaces. Or on a cylinder that is representative of a pipe.

Experimental Parameters Sample Description

The test specimen used was a curved aluminum cylinder, shown in Figure 5. No quantitative calculations were done for this specimen; it was used solely to test if we could see surface waves on a curved surface refracted by the flaw.

Page | 9

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Figure 5: Curved aluminum block contains a surface flaw Using the Olympus Omni Scan MX the shear, longitudinal, and adjustable angle wave transducers were used at 5MHz, 3.5MHz, and 1MHz respectively. The angle beam transducer was placed at the critical angle for the propagation of surface waves calculated in the theory section.

Generating Surface Waves on a Curved Surface

Using the longitudinal beam transducer with the wedge placed against the curved surface to see if a surface wave is able to be generated on a curved object with a flaw in it and reflected back. Figure 6 shows the reflection of the surface wave back from the defect. The distance from the surface wave to the minor crack is proportional to the time from the main bang to the start of the second reflection that partially overlaps the initial pulse. So in conclusion we are able to generate surface waves along a curved surface just as we can generate they can be generated in a more simple situation such as a plate.

Figure 6: Surface Wave on Curved Surface with a Surface Defect Reflection

Page | 10

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

Conclusions
In conclusion it can be ascertained that the use of ultrasonic waves is a good method for detection of stress corrosion cracking events both passively and actively. The acoustic emission events generated by a cracking event are able to be quantitatively identified and located. This type of application has been prominently used in the petrochemical industry and many specific material defects and stress corrosion cracking events have been identified; although, the nuclear industry is behind in this regard. Future research needs to focus on improving the technology for hydrogen bubbles that cause a loss of signal in a material as a specific example. More recent studies show that by further analysis of the signal received from the acoustic emission event is allowing the ability to detect hydrogen penetration inside metallic structures.13

Conclusions of Wedge Touch Test

Since the waves are traveling on the surface to observe an interruption a wedge was placed on the sample block. The wedge was subsequently placed in two locations on the sample block one close to the transmitter and the other close to the receiver. A noticeable decrease in the received amplitude of the signal in both cases was observed. However, a bigger decrease is observed when the wedge was placed closer to the receiver as depicted in Figure 3 and 4. This phenomenon can be explained in the sense that the waves have not fully propagated into the surface traveling as deeper Raleigh waves and thus, are more easily attenuated at the transmitter side of the apparatus.

Detecting Surface Defects Using Surface Waves

High frequency waves were used to locate surface and subsurface defects in the metal material. Ultrasonic testing is superior to many other methods for testing for flaws near the surface of a material, however, in order to take advantage of the ultrasonic testing, the surface must be accessible to probe and a couplant.8 In the experiment, this wasn't a problem, due to having a small material that could be easily manipulated, however, in some instances direct access to the surface may not be so accessible and another method should be used. The surface waves move in an elliptical pattern, and span a wide frequency range. When applying this to a real life application such as the pipe segment depicted in Figure 5 it can be shown that these types of surface waves may be generated on multiple geometries.

Page | 11

Idaho Dakota State University Pocatello ID, 83201

Krage Bengal ID# 827943

References
1 Cartz, Louis (1995). Nondestructive Testing. A S M International. ISBN 978-0-87170-517-4.

2 U.S. Patent 3,260,105 for Ultrasonic Testing Apparatus and Method to James F. McNulty 3 Sotirios J. Vahaviolos (1999). Acoustic Emission: Standards and Technology Update. STP-1353. Philadelphia, PA: ASTM
International (publishing). ISBN 0-8031-2498-8.

4 ASM International, Metals Handbook (Desk Edition), American Society for Metals, (1997) 5 Metals Handbook, 10th Ed., v.2, ASM Handbook Committee, American Society for Metals, USA, 1990. 6 Rose, Joseph L. Ultrasonic Waves in Solid Media. Cambridge [ u.a.: Cambridge Univ. 2004] 7 D. E. Chimenti "Guided waves in plates and their use in materials characterization", Appl. Mech. Rev., vol. 50, no. 5, pp.247 -284 1997 8 I. A. Viktorov Rayleigh and Lamb Waves: Physical Theory and Applications, 1967 :Plenum 9 P. Fromme and M. B. Sayir "Measurement of the scattering of a Lamb wave by a through hole in a plate", J. Acoust. Soc. Amer., vol. 111, no. 3, pp.1165 -1170 2002 10 NDT Education Resource Center, 2001-2012, The Collaboration for NDT Education, Iowa State University, www.ndt-ed.org 11. S. Q. Qang, D. K. Zhang, D. G. Wang , L. M. Xul and S. R. Ge, Int J. Electrochemical Science, 7 2012 12 Mazille H and Rotha R, The use of acoustic emission for the study and monitoring of localized corrosion phenomena. In: K.R. Tretheway, P.R. Roberge, Modelling aqueous corrosion, Kluwer Academic Publishers, Netherlands, (1994). 13. Djeddi, R. Khelif, International Journal of Electrochemical Science, 8 2013 14 D. Fang and A. Berkovits, "Fatigue Design Model Based on Damage Mechanisms Revealed by Acoustic Emission Measurements," Trans. of ASME, 117 (1995) 15 A. Berkovits and D. Fang, "Study of Fatigue Crack Characteristics by Acoustic Emission," Eng. Fracture Mech., 51 (3) (1995

Page | 12