DE2701054A1 - Ultrasonic surface fault detection - monitoring sound pressure changes whilst interference pattern is varied - Google Patents

Abstract

Workpieces are tested by irradiating them with h.f. ultrasonic radiation, varying the interference pattern of the coherent background radiation by changes in the position of measurement and/or angle of incident radiation and/or frequency of incident radiation and averaging measurements of sound pressure amplitude. In a practical arrangement, the ultrasonic transmit-receive transducer is coupled to a probe head and an impedance matching unit. The received analogue signal is converted to digital form at a max. frequency of 100 MHz in an A/D converter. The digital signal is averaged in an adding circuit and the resultant data may be stored on punched cards, transmitted by teleprinter or further processed in a calculator circuit. For positional averaging, the transducer has a moving probe head, and for frequency averaging a sweep-frequency generator is employed.

Description

Procedure for the determination of defects in materials

The invention relates to a method according to the preamble of the claim. The increasing use of austenitic materials, especially those higher Wall thickness poses major problems in ultrascball error checking. The useful displays are accompanied by scatter indicators from the structure.

The amplitudes of this interference structure can, depending on the size of the Defects to be detected that reach or even exceed the amplitude of the useful signal.

In the course of earlier investigations, the useful signal was used Using methods of Naohrichtenteohnik to raise it against the litter substrate. These are correlation and inverse filtering. The correlation can be a known useful signal can be raised compared to a white signal, in the case of the inverse The filtering is time-overlapping signals based on the arrival time and amplitude separated.

Neither method yielded any significant success because the structure was the same is coherent noise, i.e. the same frequency content as the useful signal Has.

Ultrasound scattering If you give a US pulse into a metallic one Sample, the energy of the wave increases with the path through absorption and Scatter consumed. The crystallites are in the microstructural area covered by the sound beam stimulated to vibrations that superimpose one another and move in all directions spread. The scattered energy can also be measured by the transmitting probe as a function of time be received.

The usual measuring arrangement when determining the structure by means of US scattering and the general course of the curves obtained is described in DBP 2,511,750.

Measurement setup The equipment used to carry out the measurements are shown in Figures 1 and 2 as a block diagram slide.

Figure 1 shows the general basic structure. The sender / receiver unit with the test head PK is via an impedance matching with the data processing system tied together. The analog signals are transmitted in an AD converter ADC with a maximum conversion rate of 100 MEz digitized and summed up in an averager.

The resulting signal can be digitized on paper tape brought or read out via a gel type. There is also the possibility to pick up the signal analogously and display it on a plotter or, after digitizing again, to process in the connected computer.

In the case of location averaging, an ultrasonic pulse device with analog outputs is used used for the signal in high frequency and equidirectional representation. The sample is moved under the test head via a sohritt motor-controlled slide.

With frequency averaging, a sine wave is generated in a frequency generator, its frequency in the attached wobber in preselectable sohritt widths around the set Center frequency is varied. About the close-up burst generator, power amplifier and the probe, the US pulse reaches the sample, which in this case is fixed. The signals from the receiving amplifier are fed into the data processing system outlined above fed in.

Determining the test conditions A central problem poses the Selection of the appropriate test parameters. In the case of location averaging, the probe position must can be varied across the reflector of interest; in frequency averaging, however is the excitation frequency to suitably change the center frequency.

The parameters - height x of the stroke) r of the angle or

Width t f of the frequency interval - can from the corresponding Echodynamic curves are obtained (Figure 3).

The useful signal drop is plotted as a function of the probe position relative to the error center or the excitation frequency or the switch-on angle.

If one considers the possible oscillation ranges, e.g. with a 6 dB drop, the dependency of the lifting height on the depth of the defect must also be observed. As dictated by the sound field structure, the lifting height is for the smaller defect depth smaller than for the error of greater depth.

The relationships in frequency averaging are very similar.

The width of the excitation pulse comes here as an additional parameter added; as the pulse width increases, the dynamic curves become narrower.

Measurements The display of the measurement results for location averaging shows Figure 4 for testing coarse-grained austenitic materials.

The only duroh summing up the interference pattern for two different ones The curve determined for the probe positions is still identical to the first display. A fourfold addition shows no significant improvements.

However, if one goes over to 32-fold or 64-fold summation, then the back wall echo sequence can already be clearly displayed up to the 4th Eoho.

Claims (1)

Method for fault detection in materials with high US scatter by measuring the Sobali pressure amplitude as a function of the Time of flight of the Sobali waves in the sample and improvement of the signal-structure noise ratio, d u r c h e k e n n z e i c h n e t that the interference pattern of the coherent Litter background due to local change and / or change in angle of incidence of the sounder and / or changing the one ball frequency is changed and at least two measured Sobali pressure amplitudes are summed up.