This presentation will concentrate on thickness measurement applications. Conventionally, a calibration sample of the exact material to be measured is constructed. The thickness of this sample is measured precisely, as is the time of flight of an acoustic pulse through the part thickness. From this information, the acoustic velocity of the material can be computed. Subsequent time of flight measurements can be interpreted in terms of thickness if we assume that the acoustic velocity is identical to that of the calibration sample.
Figure 1 - Acoustic Velocity Measurement and subsequent Thickness Measurement
If the material can be assumed to be isotropic, current computer technology permits a material velocity measurement and thickness measurement to be made about as quickly as a conventional thickness gauge reading. The process is based on a technique proposed by White  in which a subsurface refracted longitudinal wave is propagated between to transducers placed a known distance apart on the surface of the material, as shown in the Figure. The time of flight is measured and the distance between the transducers is known, therefore the acoustic velocity in the transverse direction can be computed. Using this acoustic velocity and a conventional ultrasound thickness transducer, the thickness of the part can be measured.
Although the concept is straightforward, at the time it was proposed it was difficult to implement. One pair of transducers was connected to the instrument to conduct the velocity measurement and then disconnected so the thickness measuring transducers could be used. An alternative is the use of two separate instruments - one dedicated to velocity measurement and one dedicated to thickness measurement. Additionally, the computed velocity must be entered into the thickness measuring system prior to making the thickness measurement.
Calibration is simple, utilizing two velocity standards, one of which whose thickness is known. The velocity standards have values of .1845 inches/µsec (cast iron) and stainless steel at .2240 inches/µsec.
The automatic system, which is based on the USN-52 portable flaw detector, then goes into a two channel mode at the completion of the calibration cycle, and switches back and forth between velocity measurement and thickness measurement. The velocity is updated approximately four times per second. Upon coupling to a material the velocity is measured and the new velocity is installed into the thickness channel data set. A thickness measurement is taken based on the new data. Thickness results obtained under normal calibration conditions are generally within one percent of actual for materials within a velocity range of .1500 inches/µsec to 2500 inches/µsec.
The following table of results is typical:
|Material|| Actual Thk. |
| Auto. Thickness|
It must be emphasized again that the velocity measured is parallel to the surface of the material and is not necessarily the velocity of the material in the thickness direction. Anisotropic materials such as some cast or rolled materials may exhibit relatively strong velocity differences and measurements should be taken when possible to determine the degree to which the readings have been affected by this characteristic of the material.
Other applications of this equipment may include velocity measurement for determination of nodularity of cast iron and materials sorting and results for these applications will be presented at a later time.
For more information see: Focus on Thickness Measurement in UTonline 10/97
|| UTonline ||