Brian Pavlakovic

DISPERSE: A GENERAL PURPOSE PROGRAM FOR CREATING DISPERSION CURVES

The application of guided waves in NDT can be hampered by the lack of readily available dispersion curves for complex structures. To overcome this hindrance, we have developed a general purpose program that can create dispersion curves for a very wide range of systems and then effectively communicate the information contained within those curves. The program uses the global matrix method to handle multi-layered Cartesian and cylindrical systems. The solution routines cover both leaky and non-leaky cases and remain robust for systems which are known to be difficult, such as large frequency-thicknesses and thin layers embedded in much thicker layers. Elastic and visco-elastic isotropic materials are fully supported; anisotropic materials are also covered, but are currently limited to the elastic, non-leaky, Cartesian case.

A General Purpose Approach to Calculating the Longitudinal and Flexural Modes of Multi-Layered, Embedded, Transversely Isotropic Cylinders

Over a number of years, the authors have been developing a general purpose program for predicting the properties of guided elastic waves. In addition to the well known solutions such as Lamb modes, the program can model waves in complicated structures, including structures that are comprised of multiple layers, have flat or cylindrical geometries, and are immersed in a fluid or embedded in a solid. The methodology for the program, ‘Disperse’, which is based on a global matrix algorithm, has previously been presented at this meeting and elsewhere [1,2]. This paper presents the extension of the model to include transversely isotropic cylinders. Generality is pursued by allowing arbitrary numbers of layers, axially symmetric or non-axially-symmetric (nonzero circumferential order) modes, and free, immersed, or embedded structures.

The design of a vibration transducer to monitor the integrity of dental implants

Abstract Bone-anchored titanium implants are being used increasingly to provide support for prostheses replacing missing teeth in edentulous and partially dentate patients. A technique is required to monitor bone formation at the implant-tissue interface during healing, and also to check whether there has been bone loss from around the top of the implant. One possible method is to screw a beam into the implanted fixture and to measure the first flexural resonance frequency of the resulting system. This resonance frequency is affected by both the exposed length of fixture and the stiffness of the interface between the implant and the bone. This paper describes the design of a beam-like transducer for clinical trials of the technique. The sensitivity of the transducer resonance frequency to the changes of interest is dependent on the thickness and length of the beam element. However, the choice of these dimensions is constrained by the need to avoid closely spaced resonances. The performance of different transducer shapes and the influence of the thickness and length of the beam element in the transducer has been studied. The results have been used to finalize a transducer design for the clinical trials.

Long Range Inspection of Rail Using Guided Waves

Ultrasonic techniques have been used for many years for the inspection of rail. However large transverse cracks of the type likely to cause catastrophic failure can be detected but these are often masked by small non-critical cracks at shallow angles close to the running surface of the rail. Also, alumino-thermic welds are difficult to inspect due to the attenuation of the weld material at ultrasonic frequencies. A new technique for inspecting rail that makes use of guided acoustic waves is described. These waves travel along the rail, for tens or hundreds of metres, and are partially reflected by any defects that are present. They are particularly sensitive to transverse defects and because they are used at relatively low frequency they are not significantly attenuated by weld material. The guided wave modes that can exist in a rail are found using a 2-dimensional finite element (FE) technique. Experimental measurements have been used to identify a number of candidate modes that are suitable for long range testing. The interaction of these candidate modes with a wide variety of defect geometries is investigated with 3-dimensional time-marching FE models. This study enables characteristic mode conversion signatures for various defects and features to be obtained. A prototype transducer rig has been developed and tested on a variety of rail samples containing artificial defects and also on in-service rails. Results from these tests are presented and compared with FE predictions. For the covering abstract see ITRD E122683.

Prediction of Reflection Coefficients from Defects in Embedded Bars

There is a recognized need worldwide for improved methods for the detection of corrosion of the tendons in post-tensioned concrete bridges [1,2]. Post-tensioning is used to construct light, strong bridges with the possibility of long spans. The technique involves constructing the concrete spans, leaving hollow tubes, called ducts, in place in theformwork while the concrete cures. Steel bars or strands, collectively called tendons, are then fed through the ducts and tensioned to force the concrete into compression. Finally the ducts are filled with grout to provide corrosion protection for the steel tendons. However voids which can form during the grouting process can allow water to collect in contact with the tendons, promoting corrosion. The detection of the corrosion of the tendons is very difficult because they are embedded deep within the bridge structure and are shielded by the ducts. Current inspection is primarily visual, involving drilling through to the ducts from the exterior of the bridge [1].

A Resonance Frequency Technique to Monitor the Integrity of Dental Implants

Titanium implants are being used increasingly to provide support for prostheses replacing missing teeth in edentulous and partially dentate patients. There is good histological evidence to show that in satisfactory implants bone forms in intimate contact with the implant surface during the healing process following fixture placement [1]. Currently, fixtures are left unloaded for a period of 3–6 months following placement in order to allow this healing process to occur. It would be very valuable to be able to monitor the healing process non-invasively in order to determine more accurately when it is safe to load the implant. Similarly, during service, it would be beneficial to be able to detect the onset of problems such as an increase in the mobility of the implant due to infection, or to a decrease in the height of the bone surrounding the implant.