In steel production processes, on-line automatic ultrasonic testing, one of the main methods of quality control and guarantee of steel products, is required to have higher reliability based on a better signal-to-noise ratio. This is due to the fact that quality requirements for steel products are becoming severe. The conventional systems have tried to obtain better signal-to-noise ratio using narrow band transducers that prevent reduction in sensitivity caused by the high attenuation of the material and by the low efficiency of the ultrasonic transducers. It would be possible by using broad band transducers to achieve high resolution and to apply the various kinds of time-frequency domain signal processing techniques, which distinguish types of detected flaws. However, the application of the broad band transducers causes degradation of the signal-to-noise ratio. This is due to harmful noise intrusion associated with the broad band frequency range. From this view point, chirp pulse compression is a useful technique that yields both of higher signal-to-noise ratio and higher resolution. Although the pulse compression has been applied to ultrasonic testing [1], the previous works have not concerned with the unsatisfied effect of the pulse compression caused by additional sidelobes. This is due to the ultrasonic propagation characteristics of the material and the transducers. This paper describes an optimum modification of the chirp pulse wave form. The chirp wave form is so modified taking account of the ultrasonic propagation characteristics of the material and the transducers that the compressed pulse wave form be less distorted to have less sidelobes. In order to apply this modification to on-line automatic ultrasonic testing, the authors developed a real-time digital signal processing system in which the pulse compression is processed within the dead time between the repeated pulses. As a result of field applications of the new system for flaw detection in steel plates, good performance in terms of high signal-to-noise ratio and high resolution of flaw detection has been achieved. REFERENCE
- M. Pollakowski, H. Ermert, L. von. Bernus and T. Schmeidl, "The optimum bandwidth of chirp signals in ultrasonic applications", Ultrasonics, Vol. 31, No. 6, pp417-420, 1993
NDTnet
