TWI Industrial Member Report Summary 998/2011
By P Catton and K Thornicroft
Background
Long Range Ultrasonic Testing (LRUT) is complicated by the existence of a great number of possible wave modes that may occur at a single frequency. This complexity is further exacerbated by the highly frequency dependent velocities of the modes, which leads to distorted pulse shapes and unpredictable arrival times of test signals. To mitigate this effect, efforts are taken to reduce the frequency bandwidth of the transmitted pulse (Wilcox, 2003). This is achieved by using a window function (such as the Hann window) which facilitates an operator to conduct an LRUT inspection at a specified nominal central frequency, with minimal transmission of energy at other frequencies. Therefore, the velocity variations over frequency are minimised. A comprehensive review of the technique can be found in J Rose's text on the subject found in the bibliography at the end of this document.
Excitation methods presented in this report present a distinct shift from this 'discrete frequency' mentality. Whereas the current Teletest® (Gan, Jackson, France, 2009) procedure aims to operate over a very narrow bandwidth, requiring separate tests for each frequency of interest. It is predicted that using a so called 'chirp signal' a broad range of frequencies can be excited in a single input pulse. This has been used in other fields such as RADAR, but has never before been used with guided waves and has the potential to drastically reduce the number of tests (and therefore the overall operation time) of LRUT inspection.
Objectives
While the potential of broadband excitation is wide ranging, the aim of this initial research was to prove that the concept can be applied to LRUT. The specific benefit investigated here is the time saving aspect of conducting a single chirp (broadband) test instead of several narrowband tests.
Specific objectives were:
- Devise a procedure for conducting chirp tests under laboratory conditions.
- Carry out a comparison of chirp tests with conventional 'discrete frequency' tests.
- Make recommendations for future exploitation and implementation of chirp approaches.