GB2054146A

GB2054146A - Ultrasonic Testing

Info

Publication number: GB2054146A
Application number: GB8022014A
Authority: GB
Original assignee: Individual
Current assignee: Individual
Priority date: 1979-07-06
Filing date: 1980-07-04
Publication date: 1981-02-11
Also published as: GB2054146B

Abstract

A method of ultrasonic testing includes the steps of applying a pulse input in a first mode (compression, shear or surface) to a test piece using a first transducer, sensing the response to the input using a second transducer in a second mode on the test piece, and spectrum analysing the response. Presence of a predominant frequency indicates a fault, and the frequency is in many cases a function of the length of the fault. Using a third transducer to sense the response to the input, a relationship between the amplitudes of responses from the second and third transducers can be a function of the depth of the fault.

Description

SPECIFICATION Methods of Ultrasonic Non-destructive Testing The present invention relates to methods of ultrasonic non-destructive testing.

There are many instances in modern technology where early detection of faults developing during the use of equipment is vital for safety. Examples are aircraft and pressure vessels.

Consequently there is continuous research into methods of non-destructive testing in the hope of finding a method of testing which is cheap, simple to operate, and fool-proof.

One method of non-destructive testing which is the subject of much work has as its basis the scattering and reflection of ultra-sonic pulses.

However the interpretation of results from this form of testing are frequently a matter of judgement and hence can lead to error.

Pulses can be applied to a test piece in one of three ways, as compression, as shear, or along the surface. For convenience these will be herein referred to as in the compression, shear and surface modes. For the vast majority of test pieces pulses can most conveniently be applied in the -compression mode. Most ultrasonic testing to date has as its basis the application of compression pulses to one face of a test piece and study of the scatter and reflection from the other face.

When pulses are applied to a test piece in the compression mode they will usually have some effect on the shear and surface modes.

According to the present invention a method of ultrasonic testing includes the steps of applying a pulse input in a first mode to a test piece using a first transducer, sensing the response to the input using a second transducer in a second mode on the test piece, and spectrum analysing the response.

Transducers for applying pulses and sensing responses in each of the three modes are wellknown.

When no fault is present in the component the response will be in the form of noise having no predominant frequency content. When the pulses impinge on a fault, however, a periodic disturbance is set up in all modes, the disturbance having a predominant frequency which is dependent on the material of which the test piece is constructed and the nature of the fault. The presence of a predominant frequency is clearly shown by the spectrum analysis, and is evidence of the presence of a fault. In certain circumstances measurement of the frequency can be used to indicate the size of the fault.

In a modification of the invention two transducers are used to sense the response to the input. The two transducers can sense in the same mode as each other or in different modes. By comparing the amplitudes of the responses sensed by the two transducers the orientation of the fault can be determined. The invention also provides apparatus for carrying out the method of testing.

In order that the potential of the invention and means of putting it into effect be better understood the results of some tests will now be described with reference to the accompanying diagrammatic drawings of which: Figure 1 shows the arrangement of the test equipment, Figure 2 shows a Cathode Ray Tube trace of a response to pulse inputs to an unfaulted test piece, Figure 3 shows a spectrum analysis of the response of Figure 2, Figure 4 shows a Cathode Ray Tube trace of a response to pulse inputs impinging on a fault, Figure 5 shows a spectrum analysis of the response of Figure 4, Figures 6 and 7 show the relationship between the wavelength of a surface wave response and a compression wave input for slit depths respectively small and large relative to a test piece depth, Figure 8 shows graphs of f=V,/4x and f=V,"2x where f is the frequency of a surface wave response to a compression wave input, V, is the surface wave velocity, and x is the slit depth, Figure 9 shows an arrangement of transducers having two sensing transducers, Figure 10 shows another arrangement of transducers having two sensing transducers, and Figures 11 a, b show yet more arrangements having two sensing transducers.

An aluminium test piece 10 having a slit 11 of known depth had pulses applied to a surface from a first transducer 12 in the form of a compression wave probe supplied by a pulse generator 13.

A second transducer 14 was positioned to sense surface waves (sometimes referred to as Rayleigh waves) on the other surface of the test piece 10. Responses sensed by the second transducer 14 were passed to a wide band receiver 1 5 having a time gate allowing a portion of the response to be passed to a spectrum analyser 1 6. The analyser 1 6 was connected to an oscilloscope 1 7 to allow the spectrum to be viewed.

When the first transducer was distant from the slit 11 (as indicated in dotted lines at 1 2a in Figure 1) the response sensed by the second transducer 14 consisted of random noise, as illustrated in Figure 2. Spectrum analysis of the signal of Figure 2 gave the spectrum shown in Figure 3, which had no predominant frequency.

When the first transducer 12 was opposite the slit 11, as illustrated in full lines in Figure 1 the response sensed by the second transducer 14 contained a noticeable predominant frequency, as shown in Figure 4, and spectrum analysis allowed this frequency f, to be measured, as shown in Figure 5.

A series of tests was carried out with slits 11 of various depths, and it was noted that as the depth varied, so did the frequency. For all waves the velocity, frequency and wavelength are related by the formula V=fZ. It might be expected that when a slit 11 depth x is small relative to test piece 12 depth, as shown in Figure 6a surface wave responses to a compression wave input might have a relationship between slit depth and wavelength of x=A/4 as shown in Figures 6b, 6c.

Similarly for a slit 11 depth x large relative to the test piece 12 depth (Figure 7a) the relationship might be expected to be x=j12 (Figures 7b, 7c).

Values f=V,"4x) V, being the surface, or Rayleigh, wave velocity) and f=VV2x are plotted in Figure 8.

Test results were found to fall in the shaded area 1 8 on this plot.

In using the method of the invention for testing a piece of equipment, if the depth of any faults is important, the test apparatus can be calibrated against slits of known depth in material of the same composition and thickness as that of the equipment.

It will be realised that whilst the tests described above used a compression wave probe as the first transducer 12 and a surface wave sensor as the second transducer 14, and the transducers 12, 14, were on opposite sides of the test piece 10, this arrangement may not in practice always be the most convenient or practical. The invention will, however, work with any combination of two transducers selected for compression, shear, and surface operation. The invention will also work with both transducers on the same surface. Selection of the best test methods in any case will depend on the test conditions.

In a modification of the invention two transducers are used to sense the response to the input. Examples of this arrangement are illustrated in Figure 9 where a compression input from a transducer 20 is sensed by surface wave transducers 21,22; in Figure 10 where a compression input from a transducer 30 is sensed by a surface wave transducer 31 and a shear wave transducer 32; and Figures 11 a, b where a shear wave input from a transducer 40 is sensed by surface wave transducers 41, 42.

Spectrum analysis of the responses, as described with reference to Figure 5, allows the length (shown as 1 in Figure 9) of a fault 24 to be estimated from a calibration such as that illustrated in Figure 8. Comparison of the amplitudes of the responses (as shown in Figure 4) of transducers 21, 22, 31, 32 or 41, 42 respectively allow the depth (shown as d in Figure 9) to be estimated from a calibration similar to that illustrated in Figure 8.

As an alternative to using calibrations such as that illustrated in Figure 8 a microprocessor may be included in a circuit based on that of Figure 1, and may be programmed to read out fault characteristics directly.

Claims

1. A method of ultrasonic testing including the steps of applying a pulse input in a first mode to a test piece using a first transducer, sensing the response to the input using a second transducer in a second mode on the test piece, and spectrum analysing the response.

2. A method of ultrasonic testing as claimed in Claim 1 including the further step of comparing a predominant frequency with a calibration to ascertain the length of a fault.

3. A method of ultrasonic testing as claimed in Claim 2 wherein the calibration is in the form of a mini-computer programme.

4. A method of ultrasonic testing including the steps of applying a pulse input in a first mode to a test piece using a first transducer, sensing the response to the input using second and third transducers each in a mode other than the first mode, and spectrum analysing the response.

5. A method of ultrasonic testing as claimed in Claim 4 wherein the second and third transducers sense in the same mode.

6. A method of ultrasonic testing as claimed in Claim 4 or in Claim 5 including the further steps of comparing a predominant frequency with a calibration to ascertain the length of a fault and comparing relative amplitudes of responses from the second and third transducers to ascertain the depth of the fault.

7. A method of ultrasonic testing as claimed in Claim 6 wherein the calibration is in the form of a mini-computer programme.

8. Ultrasonic testing apparatus including a first transducer operating in a first mode, a second transducer operating in a second mode, a pulse generator supplying pulses to the first transducer, a wide band receiver having a time gate, and a spectrum analyser and an oscilloscope receiving signals from the second transducer via the wide band receiver.

9. Ultrasonic testing apparatus as claimed in Claim 8 including a third transducer operating in a mode other than the first mode.

10. A method of ultrasonic testing as claimed in any one of Claims 1 to 7 wherein the first mode is the compression mode.

11. A method of ultrasonic testing as claimed in any one of Claims 1 to 7 wherein the first mode is the shear mode.

12. A method of ultrasonic testing as claimed in any one of Claims 1 to 7 wherein the first mode is the surface mode.

13. Apparatus as claimed in Claim 9 or in Claim 10 wherein the first mode is the compression mode.

14. Apparatus as claimed in Claim 9 or in Claim 10 wherein the first mode is the shear mode.

15. Apparatus as claimed in Claim 9 or in Claim 10 wherein the first mode is the surface mode.

1 6. A method of ultrasonic testing as herein described with reference to the accompanying Figures 1 to 9.

1 7. Apparatus for ultrasonic testing as herein described with reference to Figure 1.