A technique for ultrasonic characterization of plates has been extended to tubes and curved structures in general. In this technique, as explained in more detail below, one performs measurements that yield a thickness-independent value of local through-the-thickness speed of sound in a specimen. From such measurements at numerous locations across the specimen, one can construct a map of velocity as a function of location. The gradients of velocity indicated by such a map indicate through-the-thickness-averaged microstructural parameters that affect the speed of sound. Such parameters include the pore volume fraction, mass density, fiber volume fraction (in the case of a composite material), and chemical composition.

Figure 1. Ultrasonic-Pulse/Echo Time Intervals are measured in the absence and in the presence of the specimen. These intervals are used to calculate a thickness-independent value of through-the-thickness speed of sound in the specimen.
Figure 1 schematically depicts the technique as applied to a plate specimen. An ultrasonic transducer is placed in a tank of water at a fixed distance above a horizontal reflector plate. The transducer is operated in a pulse/echo mode, and the round-trip travel times for ultrasonic pulses are determined from the intervals between transmitted pulses and received echoes. At first, the pulse/echo interval for the first echo from the reflector plate is measured without a specimen present in the tank.

Next, a plate specimen is placed in the tank, approximately parallel to the reflector plate. The plate can be moved horizontally to obtain measurements at various surface locations. At each location, one measures the interval (2τ) between the first echo from the top surface and the first echo from the bottom surface of the specimen, as well as the pulse/echo interval for the first echo from the back surface of the specimen. These measurements can then be used to calculate the local through-the-thickness speed of sound (V) in the specimen from the thickness-independent right side of an equation derived from the basic equations for the pulse/echo intervals. The equation is V = c[(Δt/2τ)+1], where c is the known speed of sound in water and δt is the difference between the reflector-plate pulse/ echo intervals without and with the specimen present.

In an apparatus used to apply the technique to a tubular specimen, the specimen is mounted on a horizontal turntable in a water tank, with its axis vertical and coincident with the turntable axis. A machined metal reflector plate narrow enough to fit within the inner diameter of the specimen is suspended vertically from above and positioned inside the specimen about 1 cm from the inner tube wall. A horizontally oriented ultrasonic transducer is positioned outside the specimen, facing the reflector plate. Pulse/echo measurements are taken in basically the same manner as for plate specimens. The transducer is translated vertically to obtain measurements at various axial positions (e.g., increments of 1 mm) and the turntable is rotated to obtain measurements at various azimuthal positions (e.g., increments of 1°).

Figure 2. Non-Thickness-Independent Versus Thickness-Independent Velocity Maps for silicon nitride tube.

The technique has been demonstrated in experiments on tubular specimens of mullite (silica/alumina), a polymer-matrix composite, a composite of SiC fibers in an SiC matrix, and a high-temperature-structural grade of silicon nitride. Although the turntable, specimen, reflector plate, and transducer should be aligned as nearly perfectly as possible and the specimen should approximate a perfect round tube, it was observed that in general, some misalignment and out-of-roundness can be tolerated; this is an advantage over peak-amplitude-based ultrasonic techniques in which measurements are altered drastically by refractive effects associated with out-of-roundness. The present technique made it possible to eliminate most of the effects of variations in tube-wall thicknesses upon velocity maps (through-the-thickness velocities as functions of axial and azimuthal positions), except that edge effects associated with discontinuous changes in thickness were not eliminated completely. In the case of the silicon nitride tube (see Figure 2), differences between velocities at different locations were found to be correlated with differences between densities and pore volume fractions revealed by x-radiography and destructive metallographic analysis at those locations.

This work was done by Don J. Roth, Dorothy V. Carney, and George Y. Baaklini of Glenn Research Center and James R. Bodis and Richard W. Rauser of Cleveland State University. For further information, access the Technical Support Package (TSP) free on-line at www.nasatech.com/tsp  under the Physical Sciences category.

Inquiries concerning rights for the commercial use of this invention should be addressed to:

NASA Glenn Research Center
Commercial Technology Office
Attn: Steve Fedor
Mail Stop 4 - 8
21000 Brookpark Road,
Cleveland, Ohio 44135.

Refer to LEW-16693.