JUICE ANALYSIS OF SURFACE DEFECTS
Patent Information
- Application Number
- DE502014016938
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2013-06-13
- Filing Date
- 2014-01-16
- Publication Date
- 2025-07-10
- Estimated Expiration
- 2034-01-16
AI Technical Summary
Conventional SAFT analysis methods produce poor results near the surface of test objects due to the spatial blurring of defect indications, leading to inadequate defect localization, separation, and signal-to-noise ratio.
The method involves direct recording and use of sound propagation times dependent on the test head during SAFT analysis, which are determined by evaluating a sound field simulation of the probe for the respective material of the test object, taking into account the exact shape and aperture configuration of the transducer.
This approach significantly improves defect localization, separation, and signal-to-noise ratio near the test surface, enhancing the testing of objects with small transverse dimensions and near-surface defects, especially when using large or focused probes.
Description
[0001] The present invention relates to a method and a device for ultrasonic testing of test objects, in particular metallic components, by means of SAFT analysis.
[0002] Ultrasonic testing is used to detect defects, particularly in metallic components, such as forged products. The SAFT analysis technique is well-known for its improved localization and separation of defects.
[0003] SAFT (Synthetic Aperture Focusing Technique) is a process in which a B-scan is mathematically obtained from previously recorded, digitized, and stored A-scans. A probe with the largest possible aperture angle is moved along a line. A-scans are digitized at intervals and stored in a computer. Depending on the probe position, echoes from a defect with corresponding time-of-flight differences occur. For each voxel in the test volume of a workpiece, the expected sound travel times are conventionally calculated from geometric relationships for each probe position, and the echo amplitudes corresponding to these travel times are searched for and added in the stored A-scans.This results in precise time-of-fault compensation at the location of a defect, so that the defect echoes from all probe positions are superimposed in phase, resulting in a correspondingly high amplitude. This focuses the sound intensity on the respective voxel. If the obtained amplitude value is assigned to each voxel during this synthetic focusing, a focused volume data set is obtained.
[0004] The inspection is performed in the conventional manner, but the RF data and precise position information are recorded. During the subsequent SAFT analysis of the measurement data, amplitude sums are calculated from numerous time-domain signals, known as A-scans, for small elements of the volume to be inspected. Such elements of the volume to be inspected are referred to as voxels. Depending on the respective distance between the voxel and the measurement point, which is the position of the probe, the amplitudes are summed at the times corresponding to the respective distance between the voxel and the measurement point.
[0005] Due to the aperture angle of the sound beam emitted by the probe, defect indications are spatially blurred, so that in the cross-sectional view, the so-called B-scan, point-like defects become crescent-shaped indications. Using SAFT analysis, these crescent-shaped indications are refocused into point-like indications.
[0006] However, this only works in the so-called far field; near the test object surface, other readings are generated due to the probe sound field. Therefore, conventional SAFT evaluation produces poor results near the test surface.
[0007] Various variants of the SAFT algorithm are known, but they address different problems and do not provide any improvement in the near field. For example, the FT-SAFT analysis is known to accelerate calculations. This method provides a significant speed advantage for flat test surfaces through the use of Fourier transformation. Furthermore, the HAFT-SAFT method, which takes the direction-dependent propagation velocity into account, is known for testing homogeneously anisotropic materials. Instead of a spherical wave, the energy velocity surfaces of a point wave are used, depending on the respective materials.
[0008] DE 10 2005 051 783 A1 discloses a method and a device for imaging ultrasonic testing of a three-dimensional workpiece.
[0009] EP 2 469 276 A1 discloses a method and a device for non-destructive material testing of a test object using ultrasonic waves.
[0010] DE 10 2010 040856 A1 discloses a method and a device for determining an orientation of a defect existing within a mechanical component.
[0011] The scientific paper "Synthetic aperture focusing of ultrasonic inspection data to enhance the probability of detection of defects in strongly attenuating materials," by Spies et al., NDT&E International, published July 1, 2010, describes the determination of the size of defects in strongly attenuating materials and the determination of the probability of detection (POD) from ultrasonic data. Duplex stainless steels are used as an example, as their ferritic-austenitic microstructure causes significant attenuation of ultrasonic waves. Therefore, the synthetic aperture focusing technique (SAFT) was applied, which reduces microstructural noise signals and thus improves defect detection. Based on the ultrasonic RF data acquired on a test bench with model defects, a versus approach was performed to determine the POD for the tested duplex specimen according to MIL-HDBK-1823.
[0012] The objective is to provide a method and a device for ultrasonic testing of test objects, in particular metallic components, using an analysis, in particular SAFT analysis, such that, compared to the prior art, better results of the SAFT calculation near the surface of a test object are achieved, in particular better defect localization, stronger defect separation, and a higher signal-to-noise ratio. The aim is to achieve improved testing of test objects with small transverse dimensions, improved analysis of near-surface defects, and improved evaluation when using large or focused probes.
[0013] The problem is solved by a method according to the features of the main claim and a device according to the features of the dependent claim.
[0014] According to a first aspect, a method for ultrasonic testing of test objects, in particular metallic components, by means of a test head is proposed, wherein in an analysis, in particular SAFT analysis, a calculation of amplitude sums is carried out for each element of a volume of the test object to be tested from a number of A-image time signals, wherein in order to determine times of the amplitudes to be summed in the A-image time signals, a sound propagation time dependent on the test head from the test head positioned at a measuring point to the location of the element of the volume to be tested is directly recorded and used.
[0015] The sound propagation time is determined by evaluating a sound field simulation of the probe for the respective material of the test object. By taking the sound field shape of the probe(s) used into account during the SAFT analysis, the evaluation near the test surface is improved.
[0016] The sound field simulation takes into account the exact shape and aperture configuration of a transducer in the probe.
[0017] The sound field simulation takes into account the focusing of the probe.
[0018] In phased array sound field simulations, a delay is taken into account according to the delay law. This delay law is also referred to as the "delay law."
[0019] The sound field simulation is monochromatic or polychromatic.
[0020] The sound field simulation is performed using point source synthesis or spatially discrete methods. Examples include FE, EFIT, and the like. FE is a conventional finite element method. EFIT is a conventional method known as the elastodynamic finite integration technique.
[0021] According to a second aspect, a device for ultrasonic testing of test objects, in particular metallic components, by means of a test head is proposed, wherein, by means of a computing device, amplitude sums are calculated from a number of A-scan time signals for each element of a volume of the test object to be tested during an analysis, in particular a SAFT analysis, wherein the computing device directly records and uses a sound propagation time dependent on the test head from the test head positioned at a measuring point to the location of the element of the volume to be tested to determine times of the amplitudes to be summed in the A-scan time signals, and the computing device determines the sound propagation time by evaluating a sound field simulation of the test head for the respective material of the test object,wherein the computer device takes into account an exact shape and an aperture occupancy of a transducer of the probe during the sound field simulation, and the computer device takes into account a focusing of the probe during the sound field simulation, wherein the computer device takes into account a delay according to a delay law during the sound field simulation for phased arrays, wherein the computer device executes the sound field simulation monochromatically or polychromatically, wherein the computer device executes the sound field simulation by means of a point source synthesis or spatially discretely.
[0022] Ultrasonic testing, particularly using imaging techniques, generally encompasses all graphical representations of a recorded amplitude as a function of time or location. These range from simple RF image representations using a simple oscilloscope to A-scans and D-scans. These ultrasound images vary in their informative value depending on the representation. The A-scan (less so than the RF image) is particularly important due to its relatively easy-to-interpret display. For physical reasons, ultrasonic testing is usually performed as a single-point test, with the signals processed as an A-scan, which can, for example, be a non-rectified RF image. A-scans can be non-rectified or rectified. Only later, in the history of development, did line scans (B-scans) and area scans (C / D-scans) become technically usable.The scan is a raster scan because the A-scans are combined into a line-like or area-like image and scaled (e.g., according to color or grayscale). During a measurement, the RF signals are recorded as a function of location and time and analyzed for signal propagation time and / or amplitude or attenuation.
[0023] In contrast to the conventional use of the distance between the measurement point and the voxel, SAFT analysis considers the probe-dependent sound propagation time from a probe placed at the measurement point to the location of the voxel to determine the time points of the amplitudes to be summed in the A-scan time signals. The probe-dependent sound propagation time can be determined in different ways.
[0024] The sound propagation times are not derived indirectly from geometric quantities, but after their direct measurement are directly used to determine the amplitudes to be summed in the amplitude-time curves of the A-scan time signals.
[0025] Further advantageous embodiments are claimed in conjunction with the subclaims.
[0026] According to a further advantageous embodiment, the surface shape of the test object can be taken into account in the sound field simulation.
[0027] According to a further advantageous embodiment, sound propagation times of the sound field simulation can be determined in advance, tabulated and stored in a retrievable manner.
[0028] The sound field simulation can be carried out in advance and the results can then be tabulated.
[0029] According to a further advantageous embodiment, sound propagation times can be recorded in advance, tabulated and stored in a retrievable manner by measuring a sound field of the probe for the respective material of the probe.
[0030] The sound field can be determined by measurement and the results can then be tabulated.
[0031] According to a further advantageous embodiment, tabulated sound propagation times for a plurality of probes can be stored in a probe library in a retrievable manner.
[0032] The sound field simulation can be carried out for many probes and the sound propagation times can be stored in a probe library.
[0033] According to a further advantageous embodiment, respective sound propagation times can be interpolated from tabulated propagation times. The sound propagation time from the measurement point to the voxel can be interpolated from the tabulated propagation times.
[0034] The invention is described in more detail using exemplary embodiments in conjunction with the figures. They show: Figure 1 shows an embodiment of a method according to the invention; Figure 2 shows an embodiment of a device according to the invention.
[0035] Figure 1shows an embodiment of a method according to the invention. The figure shows wavefronts in the near field of a probe 1, where transit times Δt are specified as multiples of a basic time duration or period T. According to the present invention, better results of the SAFT calculation are achieved near the test surface, in particular better defect localization, stronger defect separation, and larger NNR. This results in improved testing of objects 3 with small transverse dimensions, improved analysis of near-surface defects 5 due to direct sounding and shorter sound paths, as well as improved evaluation when using large probes 1, for example to introduce more signal energy into the test object 3, or when using focused probes. Fig. 1Pulsed ultrasonic waves propagating from the probe 1 are shown, each generating a wavefront at whole multiples of a period T in the near field of the probe 1. In this example, the position of a defect 5 can be specified and defined as a sound propagation time Δt = 3.6 x T. In general, Δt depends on all coordinates u, v, w. Basically, Δt(u, v, w) applies. In contrast to conventional methods, the position of the defect is expressed directly as a sound propagation time Δt, and this is used to define an amplitude of an A-scan time signal intended for summation, particularly in a SAFT analysis.
[0036] Figure 2 shows an embodiment of a device according to the invention which implements a method according to Fig. 1 By means of a probe 1 and a measuring device, sound propagation times Δt can be recorded in advance by measuring a sound field of the probe 1 for the respective material of the test object 3.
[0037] By means of a computer device 7, a calculation of amplitude sums "+" can be carried out for each voxel of the test object 3 from a number of A-image time signals in a SAFT analysis. The computer device 7 uses a sound propagation time Δt, dependent on the probe 1, from the probe 1 positioned at a respective measuring point i(xi, yi, zi) to the location of the voxel or defect 5 to determine times Δt of the amplitudes ai to be summed in the measured A-image time signals ai. By means of the computer device 7, the sound propagation times Δt can be determined, for example, by evaluating a sound field simulation of the probe 1 for the respective material of the test object 3. The computer device 7 can record sound propagation times Δt in tabular form, and the resulting tables can be stored for retrieval by means of a memory device 9. Fig. 2An outer loop Sa is represented over reconstructed positions x, y, z, and an inner loop Si over measurement positions i. A coordinate system transformation KS into u, v, and w coordinates is performed. Using the sound field SF of probe 1, corresponding sound propagation times Δt are generated from the u, v, and w coordinates. Fig. 1 illustrates the representation of the u-coordinate of defect 5 in object 3 as sound travel time Δt(u).
[0038] The invention relates to a method and a device for ultrasonic testing of test objects by means of a test head, wherein a sound propagation time dependent on the test head from the test head positioned at a measuring point to the location of the voxel is directly recorded and used by means of a computer device in a SAFT analysis to determine times of the amplitudes to be summed in the A-scan time signals.
Claims
1. Method for ultrasonic testing of test objects, in particular metallic components, by means of a test head (1), wherein in an SAFT analysis, a calculation of amplitude summations is executed in each case for elements of a volume to be tested of the test object (3) from a number of A-image time signals, wherein to determine points in time of the amplitudes to be summed in the A-image time signals, a sound propagation time (Δt), which is dependent on the test head, from the test head (1) positioned at a measurement point to the location of the element of the volume to be tested is directly acquired, and the sound propagation time is used by means of analysis of a sound field simulation of the test head for the respective material of the test object, characterized in that a shape and an aperture allocation of an oscillator of the test head are taken into consideration in the sound field simulation, and a focusing of the test head is taken into consideration in the sound field simulation, wherein a delay according to a delay law is taken into consideration in the sound field simulation in the case of a phased array test head, wherein the sound field simulation is monochromatic or polychromatic, and the sound field simulation is executed by means of a point source synthesis or in a spatially discrete manner.
2. Method according to Claim 1, characterized in that the surface shape of the test object is taken into consideration in the sound field simulation.
3. Method according to either of the preceding claims, characterized in that sound propagation times of the sound field simulation are determined beforehand, tabulated, and stored in retrievable form.
4. Method according to Claim 1, characterized in that sound propagation times are acquired beforehand, tabulated, and stored in retrievable form by means of measurement of a sound field of the test head for the respective material of the test object.
5. Method according to either of the preceding Claims 3 and 4, characterized in that tabulated sound propagation times for a plurality of test heads are stored in retrievable form in a test head library.
6. Method according to any of the preceding Claims 3 to 5, characterized in that respective sound propagation times are interpolated from tabulated propagation times.
7. Apparatus configured to carry out a method according to any of Claims 1 to 6, comprising a computer unit and a test head, for ultrasonic testing of test objects, in particular metallic components, by means of the test head, wherein, in an SAFT analysis, the computer unit is designed to execute a calculation of amplitude summations in each case for elements of a volume to be tested of the test object from a number of A-image time signals, wherein, in order to determine points in time of the amplitudes to be summed in the A-image time signals, the computer unit is configured to directly acquire and use a sound propagation time, which is dependent on the test head, from the test head positioned at a measurement point to the location of the element of the volume to be tested, and the computer unit is configured to determine the sound propagation time by means of analysis of a sound field simulation of the test head for the respective material of the test object, characterized in that the computer unit is configured to take a shape and an aperture allocation of an oscillator of the test head into consideration in the sound field simulation, and the computer unit is configured to take a focusing of the test head into consideration in the sound field simulation, wherein the computer unit is configured to take a delay according to a delay law into consideration in the sound field simulation in the case of a phased array test head, wherein the computer unit is configured to execute the sound field simulation in monochromatic or polychromatic fashion, and wherein the computer unit is configured to execute the sound field simulation by means of a point source synthesis or in a spatially discrete manner.
8. Apparatus according to Claim 7, characterized in that the computer unit is configured to take the surface shape of the test object into consideration in the sound field simulation.
9. Apparatus according to either of the preceding Claims 7 and 8, characterized in that the computer unit is configured to determine sound propagation times of the sound field simulation beforehand and to tabulate them in a table and to store the table in a retrievable manner by means of a storage unit of the computer unit.
10. Apparatus according to Claim 7, characterized in that the apparatus comprises a measuring unit, the measuring unit is configured to acquire sound propagation times beforehand by means of measurement of a sound field of the test head for the respective material of the test object, and the computer unit is configured to tabulate the measured sound propagation times in a table and to store the table in retrievable form in a storage unit of the computer unit.
11. Apparatus according to either of the preceding Claims 9 and 10, characterized in that the storage unit comprises a test head library, and the storage unit is configured to store the tabulated sound propagation times for a plurality of test heads in retrievable form in a test head library.
12. Apparatus according to any of the preceding Claims 9 to 11, characterized in that the computer unit is configured to interpolate respective sound propagation times from tabulated propagation times.