Composites Design and Manufacture (Plymouth University teaching support materials) Acoustic Emission (AE).

Lecture PowerPoint

Review papers

Subject Index

This page is under development

Acoustic emission (AE) is the "audible" response (actually a broad band response from Hz-MHz) of a material under stress to active defects.  It is occasionally referred to by the more accurate term Stress Wave Emission (SWE).  Any fracture event creates new surface and a corresponding pulse of energy which propagates as transient elastic waves (stress waves).  Whilst high-energy events can be heard with the human ear, background noise is often a problem.  AE instrumentation normally uses lead (Pb) Zirconium Titanate (PZT) piezoelectric transducers to detect the stress wave at the surface of the material.  Alternative detection techniques include embedded- or surface- mounted optical fibres or the disturbance of a laser beam.  Two forms of transducer exist:

• ring-down transducers with a specific resonant frequency (normally 175 or for noisy situations 375 kHz)
• broadband transducers with a broader frequency range

The electrical signal generated by the transducer can be viewed on an oscilloscope as a dynamic display of voltage against time.  The acoustic emission can be quantified by a count of the number of crossings of a preset voltage threshold.  To reduce the data set, each set of threshold crossings associated with the decay of one pulse can be classified as an event.

A variety of techniques have been proposed for the calibration of acoustic emission instrumentation, including:

• Breaking 0.5 mm 2H propelling pencil lead (known as the Nielsen or Nielsen-Hsu source), adopted as a standard in ASTM E976-84 and E1067-85.
• Fracture of a glass capillary
• Grinding of powders
• Ball drop
• Air abrasive
• Helium gas jet
• Electric spark discharge
• Ultrasonic transducer
• Capacitive transducer
• Pulsed laser
• Martensitic transformation
• Fracture of boron particles in aluminium
• Stress corrosion

The use of acoustic emission sensors can be problematic due to:

• sensor sensitivity
• effect of mounting condition (couplant and mounting pressure)
• degradation (and it's evaluation) of the sensitivity

The AE signals can be analysed by recording:

• the number of counts (or events) against time
• the amplitudes and plotting the histogram of their distribution over discrete time periods
• the rise time and event duration
• for broadband transducers, the frequencies and plotting the spectra

Kaiser and Felicity effects

The first systematic approach to acoustic emission from materials under stress is generally accepted to be the thesis by Josef Kaiser in Munch in 1950 [1].  From single and repeated rising load tests on metals, he concluded that the number of emissions increased with the applied stress and that after unloading there were no acoustic emissions upon reloading until the previous maximum load was exceeded.  This effect is now known as the Kaiser Effect.

However, the Kaiser Effect does not always occur.  Fowler [2] reported that in some cases emissions did occur upon reloading at a specific fraction of the previous maximum load, known initially as the Modified Kaiser Effect, but now usually referred to as the Felicity Effect.  The ratio of the load at which emissions recur to the previous maximum load is expressed as the Felicity ratio or felicity Percentage.  Although the effect has been described in broad terms it does not have a rigorous definition.  The Felicity Effect has been critically reviewed by Hamstad [3] and by Summerscales [4].

CARP codes

The Committee on Acoustic Emission from Reinforced Plastics (CARP) and the American Society for Testing and Materials (ASTM) have published details of recommended practices for the testing of composite pipes [5], tanks and vessels [6] and insulated aerial personnel devices [7] based on testing procedures designed around retrieving the maximum data from Kaiser/Felicity effect in such structures.

Source location philosophies

When more than one transducer is present it is possible to determine the source of each event:

• linear: on a line between two transducers
• areal: in an area within (and possibly outside) the region defined by three or more transducers
• volume: in a volume within (and possibly outside) the region defined by four or more transducers

and a variety of methodologies are used to determine the position of the event:

• zone location
• transducer hit sequence
• triangulation
• time of arrival
• measurement of arrival times
• iteration
• learning algorithms
• correlation functions
• radiogoniometric principles.

By plotting the position of each located event on a map of the component, it is possible to build up a picture of where the majority of events are occurring and hence locate where failure is likely to occur.

References

1. J Kaiser, An investigation into the occurrence of noises in tensile tests or a study of acoustic phenomena in tensile tests, PhD thesis, Technische Hochschule - Munchen, Munich, Germany, 1950.
2. TJ Fowler, Acoustic emission testing of fiber reinforced plastics, Fall Convention, ASCE, San Francisco, October 1977, paper 3092.
3. MA Hamstad, A discussion on the basic understanding of the Felicity Effect in fiber composites, Journal of Acoustic Emission, 1986, 5(2), 95-102.
4. J Summerscales, The Felicity Effect in acoustic emission from composites, Proceedings of the International Symposium on Composite Materials and Structures, Beijing - China, June 1986, pp978-982.
5. CARP code: recommended practice for acoustic emission testing of fiberglass reinforced plastic piping systems, Proceedings of the First International Symposium on Acoustic Emission from Reinforced Composites, AEWG, San Francisco, 18-21 July 1983, session 4, 3:25-4:00.
6. Standard practice for acoustic emission examination of fiberglass reinforced plastic resin (FRP) tanks/vessels, ASTM Designation E1067-85, August 1985.
7. Standard test method for acoustic emission for insulated aerial personnel devices, ASTM Designation F914-85, May 1985.