Table of Contents ECNDT '98 Session: Materials Characterization | Investigation of the Austenitic Structures by Ultrasonic Spectral AnalysisPetre Petculescu, Ovidius University ConstantaRazvan Ciocan, Institute for Nuclear Research Pitesti Vasilios Bokas, T.B.A. S.R.L. Corresponding Author Contact: Petre Petculescu Ovidius University, Mamaia Ave. 124, 8700 Constanta, Romania, 04041-550064, 04041-672488 Email: petculescu@ovidius.ct.ro |
TABLE OF CONTENTS |
Ultrasonic spectroscopy denotes the methods used to analyze the frequency components of ultrasonic signals that are wideband pulses. When an ultrasonic signal traverses a medium, the frequency components associated with the input signal are altered. To produce the frequency domain information a discrete Fourier transform can be used. It is possible to study the effect of material properties on the alteration of the input signal subjecting to frequency-spectrum analysis the echo or transmitted pulse for different materials. A given input impulse will emerge having different frequency spectra after traversing dissimilar materials. Spectrum analysis essentially gives an ultrasonic"signature" that is generic to the particular microstructure type concerned.
An important application of the frequency-spectrum analysis involves the differential analysis of successive echoes. Each successive time domain echo will be more attenuated than its predecessor. In the frequency domain the ratio of the amplitudes for a series of frequency components represents the basis for determining a functional relation between the attenuation coefficient and frequency, (f).
We have performed an ultrasonic system for the investigation of the material structure (USIS). The ultrasonic data are acquisited from US instrument via RS-232 interface. The program performs the saving of the radiofrequency ( RF) signals and the storing of all the investigation parameters.
The USIS program performs the following operations:
The output of the USIS program consists of the following information:
In order to evaluate the performances of the USIS program we have investigated the samples from steels of three kinds:
We have compared the grain sizes evaluated of each sample from methalographical images (ASTM E112) and those evaluated using the data given by USIS program. Each sample was investigated in different positions. The velocities determined on each sample are showed in figure 2.
Fig. 2: The velocity values obtained at the investigation of the three samples in different points |
The grain size evaluation using the ultrasonic data is based on the Lifshits and Parkhomovstkii theory [1]. The attenuation coefficient for cubic crystallites in Rayleigh scattering range can be written in the following form:
_{s}^{L}(mm ^{-1}) = _{s}^{L,L} + _{s}^{L,T} = (8 ^{2}/375) (T f^{ 4} A^{2})/ ( _{o}^{2 }v_{L}^{3}) ( 2/ v_{L}^{5} + 3 / v_{T}^{5}) =S D^{3} f^{ 4} | (1) |
where we have used the following notations:
f - the frequency;
A- the anisotropy factor , A= c_{11}-c_{12}-2c_{44} , (c_{11}, c_{12} , c_{44} are the elastic moduli for cubic materials );
v_{L} , v_{T} are the longitudinal and transverse wave velocities
D - the average grain size;
T- the average grain volume ;
S- proportionality factor.
The relation (1) was written taking into account the mode conversion at scattering process too, for an incident longitudinal wave. The grain size was obtained by fitting the experimental data with the following dependence [2]:
( f ) = c_{1}f + c_{4} f^{4} | (2) |
In this relation we have considered supplementary and the elastic hysteresis loss - c_{1} f. The coefficients c_{1} and c_{4} have been determined by a least squares fit algorithm implemented in the USIS program. The grain size was determined by equating the coefficient of f^{4} from equations (1) and (2). The S factor has been calculated assuming that the elastic coefficients of the three considered kinds of steel have the values equal with that for 316 steel [3]. The 316 steel has been studied intensively and therefore its elastic coefficients have been available. We have presented in figure 3 the values of attenuation obtained for each sample. The values are calculated including also diffraction corrections calculated as presented in [4]
Fig. 3: The attenuation values obtained at the investigation of the three samples in different points |
| Grain Size obtained by ultrasonic spectroscopy(µm) | |
We performed a new system based on ultrasonic spectroscopy to investigate the material structure ( USIS). The experimental results obtained with USIS in grain size evaluation have been compared with those obtained by methalographical analysis (see table I) . The good agreement between them prove that USIS can be used in nondestructive evaluation of grain size for austenitic materials . It is necessary to point out that this evaluation can be performed automatically too. We can also solve with USIS the reverse problem: using the determined grain size we can evaluate the elastic coefficients.
NDT.net |