摘要:

An integral transform method is employed to predict the normal surface displacement resulting from the elastic waves initiated by a variety of line sources embedded within plates and laminates. Results are derived for a monolithic isotropic plate, a symmetric isotropic laminate consisting of an inner core of one elastic material sandwiched between two outer layers of a different elastic material, and an anisotropic laminate formed of four layers of a fibre composite material. The sources consist of line couples and line double forces, located at a quarter the overall laminate depth below the surface. Receiver signals are simulated at distances of two, four and ten plate depths from the plane of action of the source. The results bring out the effects of reflections from the internal interfaces as well as from the traction free bounding surfaces of the laminates.

ISSN:1058-9759    

DOI:10.1080/10589759808953040

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

The finite geometry of a laboratory specimen changes a measured acoustic emission waveform because of reflections, transmissions and mode conversions at the interfaces and boundaries of the specimen, thus making it difficult to interpret the measured signal. This paper develops a transfer function that removes these geometric effects from measured acoustic emission signals. This transfer function is developed using a repeatable, broad band (synthetic) acoustic emission source, a pulsed laser, and a broad band, high-fidelity sensor, a laser interferometer. The steps in the development of the transfer function are as follows: an acoustic emission signal is generated and detected in a specimen whose geometry is being characterized; this experiment (same source) is repeated in a “geometry-less specimen,” one that contains no geometric features; and the “geometry-less” signal is divided by the specimen signal (in the frequency domain) to form the transfer function. This transfer function can operate on an acoustic emission signal, measured in the same specimen, but caused by a different source. The accuracy of this transfer function is demonstrated by operating on different acoustic emission sources such as a pencil lead break and removing the unwanted geometric features.

ISSN:1058-9759    

DOI:10.1080/10589759808953041

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

This paper presents the results of a detailed study of the micromechanisms of room-temperature fatigue damage in a metastable beta Ti-15V-3Cr-3Al-3Sn composite reinforced with SiC (SCS-6) fibers. Mechanistically based fracture mechanics/acoustic emission models are then applied to the prediction of the total number of AE counts to failure. The predicted number of AE counts to failure are in good agreement with the experimental measured number of counts to failure. The implications of the results are also discussed for potential applications of acoustic emission techniques in prognostic and diagnostic non-destructive evaluation devices.

ISSN:1058-9759    

DOI:10.1080/10589759808953042

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

The objective of this study was to investigate and develop/adapt acoustic emission (AE) methods for detection and characterization of fatigue damage growth in metals. The materials investigated were 2024-T3 aluminum and 4340 steel. Edge-notch and compact tension specimens were tested under cyclic tensile loading. The investigation consisted of signal/noise discrimination, direct crack growth monitoring, acquisition of acoustic emission data, analysis of AE data and correlation of AE data and damage growth. In the case of aluminum the AE output consists of three characteristic parts corresponding to three stages of crack propagation; the first stage of high but decreasing AE rate, the second stage with a low and nearly constant rate and extending over 80% of the specimen lifetime and the third stage with increasing AE rate up to failure. The rate of AE output in the second stage can be described by a power law in terms of the stress intensity factor range, analogous to the Paris law for crack growth rate. In the case of steel AE results from crack extension, notch tip plasticity and closure, and plasticity dominated rapid crack propagation. The various sources of AE activity were analyzed based on the phase of the loading cycle at which they occur. A high rise in AE activity in the first half of the fatigue life has been attributed to a transition from plane strain to plane stress crack propagation.

ISSN:1058-9759    

DOI:10.1080/10589759808953043

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor

摘要:

It is possible to transfer results from laboratory specimens to field tests when the AE waves are measured and understood according to sound theoretical principles. There are several other advantages to measuring the waves. Various types of noise, ever present in testing situations, can be distinguished from the wave modes of interest by examining the characteristics of measured signals. This paper provides some basic wave propagation examples which equip the reader with the knowledge necessary to understand how to approach making a transition from all the database material already collected in certain field testing applications and put such testing on a firmer physical basis. For example, typical measurements of signal strength may be grossly in error. A way to prevent this is described. Reflections severely influence parameter data from laboratory specimens and must be eliminated in order to compare with field data where reflections are not as prevalent. In both laboratory and field tests it is the direct part of the wave which should be measured. To do this, it is necessary to set the instrument so that the signal strengths or energies of the direct waves are measured. Even when looking at the direct waves, care must be taken in analyzing signal strengths. Measuring the signal strength with an envelope beginning with the extensional mode and ending after the flexural mode would be grossly in error since the space in between the modes doesn't count. The envelope must take the two modes into account separately. Thus, simply checking an instrument with a function generator using a single burst is inadequate. A dual burst closely spaced in time, say 20 to 100 us apart, would be a better test.

ISSN:1058-9759    

DOI:10.1080/10589759808953044

出版商:Taylor & Francis Group    

年代:1998

数据来源: Taylor