A method for detecting the bonding state of a copper-aluminum brazing structure by using an ultrasonic phased array

Abstract

The present application relates to a kind of copper-aluminum brazing structure combination state ultrasonic phased array detection method, belong to nondestructive testing technical field.It includes that ultrasonic phased array detection system is carried out speed calibration and time gain compensation TCG sensitivity setting, aluminum side main inspection and copper side auxiliary verification, using synthetic aperture focusing SAFT to the original scanning data obtained after processing, based on TCG calibration curve and SAFT enhanced image, real-time comparison is carried out to the detection signal and standard reflection signal database to determine the defect type, obtain detection result.The advantage is that the acoustic representation is strong, the defect coverage is comprehensive, supports the complete calibration of double-side detection, the detection precision is high, the coincidence rate with X-ray verification reaches 98.3%, engineering practicability is strong, technical system has the mature conditions of popularization and application, can be used as the common technical specification of detection organization, manufacturing enterprise and supervision organization, meets the sensitivity requirement of IEC 62271 and other international standards.

Description

Technical Field

[0001] This invention belongs to the field of non-destructive testing technology, specifically relating to an ultrasonic phased array testing method for the bonding state of copper-aluminum brazed structures, applicable to the quality inspection and in-service safety testing of brazed copper-aluminum transition equipment clamps. Background Technology

[0002] Brazed copper-aluminum transition clamps (hereinafter referred to as copper-aluminum clamps) are core connectors for connecting copper conductors and aluminum conductors in power systems. They are widely used in critical applications such as substation main busbar connections, GIS switchgear incoming and outgoing lines, transformer bushing downleads, and overhead line tension section jumpers. These clamps are typically made by metallurgically bonding approximately 1 mm thick T2 pure copper plates and approximately 10 mm thick 1060 industrial pure aluminum plates through a vacuum brazing process, achieving copper-aluminum electrical contact through the copper-aluminum bimetallic composite structure.

[0003] Copper and aluminum differ significantly in their physicochemical properties: copper has a coefficient of thermal expansion of 16.8 × 10⁻⁻⁻⁴. 6 / ℃ is approximately 23.8×10⁻ of aluminum. 6 The elastic modulus of copper is approximately 110 GPa, while that of aluminum is approximately 70 GPa. Copper has a melting point of 1083℃, while aluminum has a melting point of 660℃. Copper's lattice constant is 3.61 Å, while aluminum's is 4.05 Å. The copper-aluminum interface is highly susceptible to the following typical defects during brazing and subsequent thermal cycling: ① Incomplete fusion (desoldering): The brazing filler metal fails to adequately wet the interface, leading to localized complete separation; ② Porosity: Gases released during molten pool cooling fail to escape, forming pores; ③ Inclusions: Oxide films or flux residues become embedded in the interface; ④ Microcracks: These form when thermal stress exceeds the material's strength limit. These defects increase contact resistance, cause localized overheating, and propagate during service, ultimately leading to clamp failure and even power outages. Statistics show that approximately 70% of copper-aluminum clamp malfunctions are directly related to incomplete fusion or poor bonding at the interface.

[0004] Existing non-destructive testing methods for copper-aluminum wire clamps have significant shortcomings: radiographic testing (RT) has low sensitivity to area-type unfused defects and requires radiation protection; eddy current testing (ET) has limited penetration capability for deep interface-type defects; traditional pulse-echo ultrasonic testing (UT) uses single-crystal A-scan, which lacks spatial resolution under the complex multiple reflection interference of the copper-aluminum bimetallic interface, making it difficult to achieve accurate localization and quantitative assessment. Phased array ultrasonic testing (PAUT) has electronic focusing and deflection capabilities, enabling the acquisition of multi-angle cross-sectional images in a single scan. However, the unique characteristics of copper-aluminum brazed structures (copper acoustic impedance 41.8 MRayl, aluminum 17.3 MRayl, a difference of 142%; multi-layer thin-walled composite structure) present unique challenges for PAUT testing.

[0005] More importantly, there is currently a lack of standardized comparative test blocks and corresponding calibration methods specifically designed for copper-aluminum brazed composite structures, both domestically and internationally. Existing general-purpose test blocks (such as GB / T 11259 aluminum alloy test blocks and GB / T 4730 steel test blocks) are designed based on homogeneous materials and cannot accurately reflect the acoustic characteristics of the copper-aluminum interface, leading to systematic errors in sensitivity settings. Furthermore, there is no systematic set of dedicated test blocks that simultaneously covers three typical defects: aluminum-side voids, copper-side voids, and interface incomplete fusion, thus limiting the consistency and reliability of test results. To address this issue, Chinese patent CN121558898A discloses a comparative test block set and its calibration method for ultrasonic testing of brazed copper-aluminum transition equipment clamps. This set of comparative test blocks simulates aluminum-side voids, copper-side voids, and interface incomplete fusion defects. A standard reflection signal database is established through calibration, providing assurance for quality inspection and in-service safety testing of brazed copper-aluminum transition equipment clamps. Summary of the Invention

[0006] This invention provides an ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures, aiming to achieve high sensitivity, high precision detection and quantitative evaluation of three typical defects in copper-aluminum wire clamps: aluminum-side holes, copper-side holes, and interface non-fusion.

[0007] To achieve the above objectives, the technical solution adopted by the present invention includes the following steps: Step S1: System calibration; Using a dedicated set of comparison test blocks consisting of aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C, the ultrasonic phased array detection system is calibrated for sound velocity and time gain compensation TCG sensitivity is set based on the standard reflection signal database of preset defects in each test block. Step S2, Aluminum Side Inspection: Couple the linear phased array probe to the aluminum side surface of the copper-aluminum wire clamp under inspection, and use E-scan and S-scan dual modes to perform a full-area coverage scan of the aluminum layer, and acquire B-scan and C-scan images in real time to detect volumetric hole defects in the aluminum layer and area-type non-fusion defects at the copper-aluminum brazing interface. Step S3, Copper-side auxiliary inspection: Couple the probe to the copper-side surface and use the S-scan mode to re-inspect the copper layer and the copper-solder interface area to detect small holes in the copper layer and minor delamination defects at the copper-solder interface. Step S4: Image enhancement; The acquired raw scan data is post-processed using Synthetic Aperture Focusing (SAFT), and the multi-angle echo signals are coherently superimposed using a delay superposition algorithm to improve the horizontal resolution of the image. Step S5: Defect Quantification; Based on the TCG calibration curve and SAFT enhanced image, the equivalent area of ​​the area-type non-fusion defect is measured using the relative amplitude method, and the equivalent diameter of the volumetric hole defect is calculated using the equivalent flat-bottom hole area method or the 6dB half-wave height method. The detected signal is then compared with the standard reflection signal database in real time to determine the defect type and obtain the detection result.

[0008] In step S1 of this invention, the acoustic phased array detection system uses a 64-element linear array probe with a center frequency of 10 MHz, an element spacing of 0.6 mm, and a total aperture of 38.4 mm. It is used in conjunction with a polystyrene wedge with a refraction angle of 55°, so that the longitudinal wave angle of the sound beam refracted in the aluminum layer is 38°, and the sound path coverage range in the aluminum layer is 6 to 14 mm.

[0009] The TCG sensitivity setting method in step S1 of the present invention is as follows: the echo amplitude of the Φ8 mm flat-bottom hole in the aluminum side hole defect test block A is adjusted to 80% of the full screen as a reference benchmark, and the signal positions of the Φ4 mm, Φ3 mm, and Φ2 mm holes are gradually scanned to establish a depth-gain compensation curve to eliminate the depth effect caused by the 0.02 dB / mm acoustic attenuation of the aluminum layer.

[0010] The aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C described in step S1 of this invention are all made of 1060 aluminum plate and T2 pure copper plate by vacuum brazing. The aluminum plate has a thickness of 10±0.1mm and a size of (100±0.1)mm×(100±0.1)mm, and the copper plate has a thickness of 1±0.05mm and a size of (100±0.1)mm×(80±0.1)mm. The copper plate and the aluminum plate have one side of the end face coplanar and flush. The surface roughness Ra of the test surface of the three sets of test blocks is ≤6.3 μm. The longitudinal wave velocity of the 1060 aluminum plate is (6280±50)m / s, and the transverse wave velocity is (3040±30)m / s; the longitudinal wave velocity of the T2 pure copper plate is (4700±50)m / s, and the Vickers hardness is 60~80 HV. Aluminum side hole defect test block A: Four flat-bottomed blind holes of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the inspection surface of the aluminum plate. The hole depth is (5±0.1) mm. The hole axis is perpendicular to the surface of the aluminum plate with a perpendicularity error of ≤0.5° and the distance between the centers of adjacent holes is ≥30 mm. Test block B for copper side hole defects: Four through holes of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the copper plate, and the copper-aluminum brazing interface is used as the acoustic reflection bottom surface. The distance between the centers of adjacent holes is ≥30 mm. Test block C for interface non-fusion defects: A rectangular groove with a length of 35 mm, a width of 50 mm, and a depth of (0.2±0.05) mm is machined in a designated area on the bonding surface of the aluminum plate using precision wire cutting. The flatness of the bottom surface of the groove is ≤0.02 mm. After ultrasonic cleaning, it is bonded to the copper plate by vacuum brazing. The vacuum brazing parameters are vacuum degree ≤5×10⁻³ Pa and brazing temperature 720±10℃. A closed air gap is formed in the groove to accurately simulate the completely non-fusion state of the copper-aluminum interface.

[0011] The diameter tolerance of the flat-bottomed blind hole in the aluminum side hole defect test block A of the present invention is ±0.05 mm, and the flatness of the hole bottom is ≤0.02 mm.

[0012] The method for fabricating the interface non-fusion defect test block C of the present invention further includes: ultrasonic cleaning of the inside of the groove before brazing to remove wire cutting fluid residue; vacuum brazing cooling rate ≤ 5 ℃ / min; after fabrication, the test block undergoes precision dimensional inspection (tolerance ±0.1 mm), surface roughness inspection (Ra≤6.3 μm), fluorescence penetrant testing, ultrasonic acceptance scanning, and X-ray inspection in sequence to confirm that the air gap size and position meet the design requirements.

[0013] The three sets of test blocks described in this invention simultaneously support system calibration in two directions: aluminum side incident detection and copper side incident detection. The standard reflection signal database established in conjunction with it is classified and stored according to defect type × detection direction × defect size. Each record contains four parameters: echo amplitude dB value, sound path time μs, waveform image, and signal-to-noise ratio dB.

[0014] In step S2 of this invention, the E-scan parameters are: excitation aperture 16 wafer, step 1 wafer, focusing depth 8 mm; the S-scan parameters are: scanning angle range 40°~70°, angle step 0.5°, focusing depth 8 mm; and the scanning grid step is 5 mm×5 mm.

[0015] In step S4 of this invention, the SAFT algorithm improves the horizontal resolution of the image from 2.5 mm to 1.2 mm, and the minimum detectable defect size is ≤ Φ2 mm; the signal-to-noise ratio for detecting unfused areas at the interface exceeds 20 dB, and the signal-to-noise ratio for unfused defects at the interface detected on the aluminum side is 8 to 12 dB higher than that detected on the copper side.

[0016] The specific method of the relative amplitude method in step S5 of the present invention is as follows: move the probe to find the position of the maximum echo of the defect, move it to the surrounding area until the echo amplitude drops to 20% of the maximum value, record it as the boundary point, and connect the boundary points to form the area as the equivalent area of ​​the defect; when quantifying the natural defect, a correction coefficient of 1.1 to 1.3 is introduced to compensate for the underestimation of the effective reflection area by the irregular defect surface.

[0017] The beneficial effects of this invention are: (1) Strong acoustic representativeness: The special test block adopts the same material system and manufacturing process as the actual product, which truly reproduces the acoustic characteristics of the copper-aluminum bimetallic interface and eliminates the systematic error caused by the homogeneous test block.

[0018] (2) Comprehensive defect coverage: The three test block system covers the three main defect modes of copper-aluminum wire clamps: aluminum side holes, copper side holes and interface non-fusion, and supports complete calibration of dual-side detection.

[0019] (3) High detection accuracy: 100% detection rate for holes larger than 2 mm, 100% detection rate for non-fusion defects with an area of ​​more than 10 mm², quantitative error ≤ ±15%, and 98.3% compliance rate with X-ray verification.

[0020] (4) Complementary dual-side strategy: The signal-to-noise ratio of the aluminum side main inspection to the interface non-fusion is 8 to 12 dB higher than that of the copper side, and the overall defect detection rate is improved by about 15% to 20% compared with single-side inspection.

[0021] (5) Strong engineering practicality: The technical system has mature conditions for promotion and application, and can be used as a common technical specification for testing institutions, manufacturing enterprises and supervision institutions, meeting the sensitivity requirements of international standards such as IEC 62271. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the aluminum side hole defect test block A of the present invention; Figure 2 yes Figure 1 The left view; Figure 3 This is a schematic diagram of the structure of the copper-side hole defect test block B of the present invention; Figure 4 yes Figure 3 The left view; Figure 5 This is a schematic diagram of the structure of the interface non-fusion defect test block C of the present invention; Figure 6 yes Figure 5 The left view; Figure 7 This is a perspective view of the interface non-fusion defect test block C of the present invention; Figure 8 yes Figure 7 Enlarged view of Part I; Figure 9 This is a PAUT reference image of the defect-free copper-aluminum brazed sample of the present invention; Figure 10 This is a PAUT image of the poor bonding area at the aluminum side interface of this invention; Figure 11 This is an S-scan imaging image of the copper side of the present invention; Figure 12This is a PAUT image of a small-sized natural defect on the copper side of the present invention; Figure 13 This invention provides PAUT imaging for natural defects on the aluminum side. Figure 1 ; Figure 14 This invention provides PAUT imaging for natural defects on the aluminum side. Figure 2 . Detailed Implementation

[0023] Includes the following steps: Step S1: System calibration; Using a dedicated set of comparison test blocks consisting of aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C, the ultrasonic phased array detection system is calibrated for sound velocity and time gain compensation TCG sensitivity is set based on the standard reflection signal database of preset defects in each test block. Step S2, Aluminum Side Inspection: Couple the linear phased array probe to the aluminum side surface of the copper-aluminum wire clamp under inspection, and use E-scan and S-scan dual modes to perform a full-area coverage scan of the aluminum layer, and acquire B-scan and C-scan images in real time to detect volumetric hole defects in the aluminum layer and area-type non-fusion defects at the copper-aluminum brazing interface. Step S3, Copper-side auxiliary inspection: Couple the probe to the copper-side surface and use the S-scan mode to re-inspect the copper layer and the copper-solder interface area to detect small holes in the copper layer and minor delamination defects at the copper-solder interface. Step S4: Image enhancement; The acquired raw scan data is post-processed using Synthetic Aperture Focusing (SAFT), and the multi-angle echo signals are coherently superimposed using a delay superposition algorithm to improve the horizontal resolution of the image. Step S5: Defect Quantification; Based on the TCG calibration curve and SAFT enhanced image, the equivalent area of ​​the area-type non-fusion defect is measured using the relative amplitude method, and the equivalent diameter of the volumetric hole defect is calculated using the equivalent flat-bottom hole area method or the 6dB half-wave height method. The detected signal is then compared with the standard reflection signal database in real time to determine the defect type and obtain the detection result.

[0024] In step S1 of this invention, the acoustic phased array detection system uses a 64-element linear array probe with a center frequency of 10 MHz, an element spacing of 0.6 mm, and a total aperture of 38.4 mm. It is used in conjunction with a polystyrene wedge with a refraction angle of 55°, so that the longitudinal wave angle of the sound beam refracted in the aluminum layer is 38°, and the sound path coverage range in the aluminum layer is 6 to 14 mm.

[0025] The TCG sensitivity setting method in step S1 of the present invention is as follows: the echo amplitude of the Φ8 mm flat-bottom hole in the aluminum side hole defect test block A is adjusted to 80% of the full screen as a reference benchmark, and the signal positions of the Φ4 mm, Φ3 mm, and Φ2 mm holes are gradually scanned to establish a depth-gain compensation curve to eliminate the depth effect caused by the 0.02 dB / mm acoustic attenuation of the aluminum layer.

[0026] See Figures 1-8 The aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C described in step S1 of this invention are all made of 1060 aluminum plate and T2 pure copper plate by vacuum brazing. The aluminum plate has a thickness of 10±0.1mm and a size of (100±0.1)mm×(100±0.1)mm, and the copper plate has a thickness of 1±0.05mm and a size of (100±0.1)mm×(80±0.1)mm. The copper plate and the aluminum plate have one side face coplanar and flush. The surface roughness Ra of the test surface of the three sets of test blocks is ≤6.3 μm. The longitudinal wave velocity of the 1060 aluminum plate is (6280±50)m / s, and the transverse wave velocity is (3040±30)m / s; the longitudinal wave velocity of the T2 pure copper plate is (4700±50)m / s, and the Vickers hardness is 60~80 HV. Aluminum side hole defect test block A: Four flat-bottomed blind holes of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the inspection surface of the aluminum plate. The hole depth is (5±0.1) mm. The hole axis is perpendicular to the surface of the aluminum plate with a perpendicularity error of ≤0.5° and the distance between the centers of adjacent holes is ≥30 mm. Test block B for copper side hole defects: Four through holes of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the copper plate, and the copper-aluminum brazing interface is used as the acoustic reflection bottom surface. The distance between the centers of adjacent holes is ≥30 mm. Test block C for interface non-fusion defects: A rectangular groove with a length of 35 mm, a width of 50 mm, and a depth of (0.2±0.05) mm is machined in a designated area on the bonding surface of the aluminum plate using precision wire cutting. The flatness of the bottom surface of the groove is ≤0.02 mm. After ultrasonic cleaning, it is bonded to the copper plate by vacuum brazing. The vacuum brazing parameters are vacuum degree ≤5×10⁻³ Pa and brazing temperature 720±10℃. A closed air gap is formed in the groove to accurately simulate the completely non-fusion state of the copper-aluminum interface.

[0027] The diameter tolerance of the flat-bottomed blind hole in the aluminum side hole defect test block A of the present invention is ±0.05 mm, and the flatness of the hole bottom is ≤0.02 mm.

[0028] The method for fabricating the interface non-fusion defect test block C of the present invention further includes: ultrasonic cleaning of the inside of the groove before brazing to remove wire cutting fluid residue; vacuum brazing cooling rate ≤ 5 ℃ / min; after fabrication, the test block undergoes precision dimensional inspection (tolerance ±0.1 mm), surface roughness inspection (Ra≤6.3 μm), fluorescence penetrant testing, ultrasonic acceptance scanning, and X-ray inspection in sequence to confirm that the air gap size and position meet the design requirements.

[0029] The three sets of test blocks described in this invention simultaneously support system calibration in two directions: aluminum side incident detection and copper side incident detection. The standard reflection signal database established in conjunction with it is classified and stored according to defect type × detection direction × defect size. Each record contains four parameters: echo amplitude dB value, sound path time μs, waveform image, and signal-to-noise ratio dB.

[0030] In step S2 of this invention, the E-scan parameters are: excitation aperture 16 wafer, step 1 wafer, focusing depth 8 mm; the S-scan parameters are: scanning angle range 40°~70°, angle step 0.5°, focusing depth 8 mm; and the scanning grid step is 5 mm×5 mm.

[0031] In step S4 of this invention, the SAFT algorithm improves the horizontal resolution of the image from 2.5 mm to 1.2 mm, and the minimum detectable defect size is ≤ Φ2 mm; the signal-to-noise ratio for detecting unfused areas at the interface exceeds 20 dB, and the signal-to-noise ratio for unfused defects at the interface detected on the aluminum side is 8 to 12 dB higher than that detected on the copper side.

[0032] The specific method of the relative amplitude method in step S5 of the present invention is as follows: move the probe to find the position of the maximum echo of the defect, move it to the surrounding area until the echo amplitude drops to 20% of the maximum value (a decrease of about 14 dB), record it as the boundary point, and connect the boundary points to form the area as the equivalent area of ​​the defect; when quantifying the natural defect, a correction coefficient of 1.1 to 1.3 is introduced to compensate for the underestimation of the effective reflection area by the irregular defect surface.

[0033] The present invention will be further illustrated by the following experimental examples.

[0034] Preparation of the Special Comparative Test Block Set in Experiment Example 1 (1) Raw material preparation and pretreatment 1060 aluminum plate, thickness (10±0.1) mm, dimensions (100±0.1) mm × (50±0.1) mm, Vickers hardness 25~35HV, longitudinal wave velocity (6280±50) m / s, transverse wave velocity (3040±30) m / s; T2 pure copper plate, thickness (1±0.05) mm, dimensions (80±0.1) mm × (50±0.1) mm, Vickers hardness 60~80 HV, longitudinal wave velocity (4700±50) m / s. All raw materials must have material certificates and acoustic property test reports. The measured sound velocity should be within ±1% of the theoretical value.

[0035] Pretreatment of aluminum plate surfaces to be brazed: Grind with 600-grit sandpaper until Ra≤1.6 μm, then sequentially degrease with acetone, wash with 10% NaOH for 30 s, rinse with water, and clean and dry with anhydrous ethanol. Pretreatment of copper plate surfaces to be brazed: After grinding and degreasing, sequentially pickle with 10% hydrochloric acid for 30 s, rinse with water, and clean and dry with anhydrous ethanol to ensure that the brazing surface is free of oxide film and contaminants.

[0036] (2) Preparation of test A for aluminum side hole defects On the aluminum plate inspection surface, four flat-bottomed blind holes of Φ8 mm, Φ4 mm, Φ3 mm, and Φ2 mm are machined using a precision CNC machine tool with low feed rate. The hole depth is (5±0.1) mm, the hole axis is perpendicular to the aluminum plate surface (perpendicularity error ≤0.5°), the center of each hole is ≥20 mm from the edge of the copper plate layer, the distance between the centers of adjacent holes is ≥30 mm, the diameter tolerance of the flat-bottomed hole is ±0.05 mm, and the flatness of the hole bottom is ≤0.02 mm. After machining, the aluminum plate and copper plate are treated according to the above pretreatment process and then vacuum brazed. The vacuum degree is ≤5×10⁻³ Pa, the brazing temperature is 720±10℃, the holding time is 5 min, and the cooling rate is ≤5℃ / min. One side end face of the copper plate and the aluminum plate are coplanar and flush, and a 20 mm operation allowance is left on the suspended side of the copper plate.

[0037] (3) Fabrication of copper side hole defect test block B For the copper-side hole defect test block B, the aluminum plate and copper plate are first bonded together using a vacuum brazing process. After the brazing is completed and cooled, four through holes of Φ8 mm, Φ4 mm, Φ3 mm, and Φ2 mm (penetrating the copper layer to the copper-aluminum interface) are machined in the copper layer on one side of the copper plate using precision machining. During machining, a low feed rate and high speed cutting are used to prevent machining vibration from causing the copper-aluminum interface to peel off. The through holes use the brazing interface as the acoustic reflection bottom surface, and the copper layer defect detection is equivalently transformed into acoustic response analysis at the interface.

[0038] (4) Fabrication of test block C with interface non-fusion defect In the centrally located area on one side of the aluminum plate bonding surface (retaining a sufficient defect-free reference area), a rectangular groove is machined using precision wire electrical discharge machining (WEDM): 35 mm long, 50 mm wide, and (0.2±0.05) mm deep. The flatness of the groove bottom surface is ≤0.02 mm, the corners are kept right angles, and the verticality error of the groove wall is ≤0.5°. After ultrasonic cleaning to remove the cutting fluid residue inside the groove, copper-aluminum composite is carried out using a vacuum brazing process. During the brazing process, the gas inside the groove is sealed by the surrounding brazing filler metal. After cooling, a precise and controllable closed air gap (0.1~0.3 mm in height) is formed, accurately simulating the completely unfused interface state.

[0039] (5) Final inspection of each test block After the three sets of test blocks were fabricated, the following procedures were carried out in sequence: ① Precision dimensional inspection (vernier calipers + depth gauge to confirm that all tolerances are within acceptable limits); ② Surface roughness inspection (Ra≤6.3 μm); ③ Fluorescence penetrant testing (to confirm that there are no surface opening defects); ④ Ultrasonic acceptance scanning (to confirm that the signals of each preset defect are clearly identifiable, and the bottom wave amplitude of the intact area is ≥50% of the full screen); ⑤ Test block C was also inspected with X-rays to confirm that the size and position of the air zone meet the design requirements.

[0040] Experiment Example 2: Establishing a Standard Reflection Signal Database and Configuring the PAUT Detection System (1) The calibration equipment used was an Olympus OmniScan X3 multi-channel ultrasonic phased array detector (or equivalent instrument), equipped with a high-frequency probe with a center frequency of 5-15 MHz, and the coupling agent was glycerol or deionized water (uniform water film layer of 0.5 mm). Calibration environment: temperature (20±5)℃, relative humidity ≤80%, and the test surface was placed horizontally.

[0041] First, the defect-free areas of the three test blocks were scanned from the aluminum side and the copper side respectively, and the baseline A scan waveform of the intact bonding area was collected: the echo amplitude of the bottom surface of the copper-aluminum interface on the aluminum side was adjusted to 80% of the full screen as the baseline, and the multiple echo sequences with equal spacing and attenuation on the copper side were recorded as fingerprint features of the intact area.

[0042] Then, the preset defect areas of the three test blocks were scanned in sequence, and the A-scan waveforms of various defects under incident light from the aluminum and copper sides were collected respectively. The echo amplitude (dB value), sound path position (μs) and waveform shape were recorded. A standard reflection signal database was established according to defect type × detection direction × defect size. The database was stored in electronic document format and accompanied by waveform images and comments.

[0043] (2) PAUT testing system configuration The testing instrument used was an Olympus OmniScan X3 fully digital phased array ultrasonic testing instrument (128 channels, sampling frequency 100 MHz). Linear array probe parameters: center frequency 10 MHz, 64 crystals, crystal spacing 0.6 mm, total aperture 38.4 mm; coupled with a polystyrene wedge with a refraction angle of approximately 55° (the longitudinal wave refraction angle in aluminum is approximately 38°), deionized water was used as the coupling agent, and a water film thickness control device ensured coupling stability (water film thickness 0.5 mm).

[0044] The selection of a 10 MHz frequency is based on the fact that the wavelength in aluminum is approximately 0.63 mm, reliably detecting Φ2 mm defects (approximately 3 times the wavelength) and providing a depth resolution of approximately 0.32 mm. At a low frequency of 5 MHz, the lateral resolution decreases by approximately 50%, making it impossible to distinguish Φ2 mm small holes. At a high frequency of 20 MHz, multiple reflections significantly increase noise, leading to a deterioration in the signal-to-noise ratio. The selection of a 55° wedge is based on the fact that the sound beam coverage range is approximately 6–14 mm, covering the full thickness of the aluminum layer and the area near the interface, representing the optimal balance between detection sensitivity and sound beam penetration depth.

[0045] Experiment Example 3: Product Testing Taking the factory quality inspection of 60 500 kV copper-aluminum transition clamps for a substation as an example, the detection sensitivity is set as follows: the echo of the Φ4 mm flat-bottomed hole is adjusted to 80% of the full screen reference; a full-coverage scan of the aluminum side of the workpiece is performed according to a horizontal step of 5 mm × vertical step of 5 mm grid scheme; suspicious signals with amplitudes exceeding the echo amplitude of the Φ2 mm flat-bottomed hole (40% of the reference line) are manually and meticulously reviewed, and compared with waveforms in the standard database in real time. The method of this invention is used for verification, including: (1) System calibration: Sound velocity measurement and wedge delay calibration were performed in the defect-free area; TCG sensitivity was set using test block A, and gain compensation curves at each depth were established by adjusting the echo of the Φ8 mm flat bottom hole to 80% of the full screen to eliminate the depth effect of approximately 0.02 dB / mm sound attenuation in the aluminum layer.

[0046] (2) Defect scanning: The inspected workpiece is scanned in both E-scan and S-scan modes for full coverage, with a grid step of 5 mm × 5 mm, and the complete original waveform dataset is recorded. Typical characteristics of the intact area on the aluminum side: bottom surface echo amplitude ≥ 50% of the full screen, and interface positive reflection amplitude not exceeding 15% of the full screen; Typical characteristics of the interface desoldering area: a single strong interface reflection peak appears on the aluminum side, the bottom surface echo completely disappears, and the amplitude change of multiple reflection sequences on the copper side deviates from the exponential decay law.

[0047] (3) Image processing: Image enhancement was performed using the SAFT algorithm to generate high-resolution B-scan and C-scan images. Defect identification criteria: Signals with amplitudes exceeding the Φ2 mm flat-bottom hole baseline in the B-scan image after TCG calibration were considered suspected defects; areas where the interface bottom wave completely disappeared and the interface echo exceeded the baseline value were identified as non-fusion defects; the threshold for natural defects was set at 80% of the baseline amplitude (down by approximately 2 dB), see [reference]. Figures 9-14 .

[0048] (4) Defect quantification: Area-type non-fusion defects are quantified using the relative amplitude method (the area enclosed by the boundary point where the echo drops to 20% of the maximum value, i.e., about -14dB); point-like hole defects are quantified using the equivalent flat-bottom hole area method or the 6dB half-wave height method; when quantifying natural defects, a correction factor of 1.1 to 1.3 is introduced to compensate for the underestimation of the effective reflection area by the irregular defect surface; the suspected signal is compared with the waveform in the standard database in real time to determine the defect type and generate a formatted test report.

[0049] The verification results show that the detection rate of holes larger than 2 mm is 100%, the detection rate of non-fusion defects with an area of ​​more than 10 mm² is 100%, the measurement error is within ±15%, and the conformity rate between the detection results and the subsequent X-ray verification is 98.3%. It meets the sensitivity requirements of international standards such as IEC 62271 and has mature conditions for promotion and application in the quality control and in-service safety inspection of copper and aluminum clamps in power systems.

Claims

1. A method for ultrasonic phased array detection of the bonding state of copper-aluminum brazed structures, characterized in that, Includes the following steps: Step S1: System calibration; Using a dedicated set of comparison test blocks consisting of aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C, the ultrasonic phased array detection system is calibrated for sound velocity and time gain compensation TCG sensitivity is set based on the standard reflection signal database of preset defects in each test block. Step S2, Aluminum Side Inspection: Couple the linear phased array probe to the aluminum side surface of the copper-aluminum wire clamp under inspection, and use E-scan and S-scan dual modes to perform a full-area coverage scan of the aluminum layer, and acquire B-scan and C-scan images in real time to detect volumetric hole defects in the aluminum layer and area-type non-fusion defects at the copper-aluminum brazing interface. Step S3, Copper-side auxiliary inspection: Couple the probe to the copper-side surface and use the S-scan mode to re-inspect the copper layer and the copper-solder interface area to detect small holes in the copper layer and minor delamination defects at the copper-solder interface. Step S4: Image enhancement; The acquired raw scan data is post-processed using Synthetic Aperture Focusing (SAFT), and the multi-angle echo signals are coherently superimposed using a delay superposition algorithm to improve the horizontal resolution of the image. Step S5: Defect Quantification; Based on the TCG calibration curve and SAFT enhanced image, the equivalent area of ​​the area-type non-fusion defect is measured using the relative amplitude method, and the equivalent diameter of the volumetric hole defect is calculated using the equivalent flat-bottom hole area method or the 6dB half-wave height method. The detected signal is then compared with the standard reflection signal database in real time to determine the defect type and obtain the detection result.

2. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: In step S1, the acoustic phased array detection system uses a 64-element linear array probe with a center frequency of 10 MHz, an element spacing of 0.6 mm, and a total aperture of 38.4 mm. It is used in conjunction with a polystyrene wedge with a refraction angle of 55°, so that the longitudinal wave angle of the sound beam is refracted in the aluminum layer to be 38°, and the sound path coverage range in the aluminum layer is 6 to 14 mm.

3. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: The TCG sensitivity setting method in step S1 is as follows: the echo amplitude of the Φ8 mm flat-bottom hole in the aluminum side hole defect test block A is adjusted to 80% of the full screen as a reference benchmark, and the signal positions of the Φ4 mm, Φ3 mm, and Φ2 mm holes are gradually scanned to establish a depth-gain compensation curve, thereby eliminating the depth effect caused by the 0.02 dB / mm acoustic attenuation of the aluminum layer.

4. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: The aluminum-side hole defect test block A, copper-side hole defect test block B, and interface non-fusion defect test block C mentioned in step S1 are all made of 1060 aluminum plate and T2 pure copper plate by vacuum brazing. The aluminum plate is 10±0.1mm thick and (100±0.1)mm×(100±0.1)mm in size. The copper plate is 1±0.05mm thick and (100±0.1)mm×(80±0.1)mm in size. The copper plate and aluminum plate are coplanar and flush on one side. The surface roughness Ra of the test surface of the three sets of test blocks is ≤6.3 μm. The longitudinal wave velocity of the 1060 aluminum plate is (6280±50)m / s, and the transverse wave velocity is (3040±30)m / s. The longitudinal wave velocity of the T2 pure copper plate is (4700±50)m / s, and the Vickers hardness is 60~80 HV. Aluminum side hole defect test block A: Four flat-bottomed blind holes with diameters of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the inspection surface of the aluminum plate. The hole depth is (5±0.1) mm. The hole axis is perpendicular to the surface of the aluminum plate with a perpendicularity error of ≤0.5° and the distance between the centers of adjacent holes is ≥30 mm. Test block B for copper side hole defects: Four through holes of Φ8 mm, Φ4 mm, Φ3 mm and Φ2 mm are machined on the copper plate, and the copper-aluminum brazing interface is used as the acoustic reflection bottom surface. The distance between the centers of adjacent holes is ≥30 mm. Test block C for interface non-fusion defects: A rectangular groove with a length of 35 mm, a width of 50 mm, and a depth of (0.2±0.05) mm is precision wire-cut in a designated area of ​​the aluminum plate bonding surface. The flatness of the bottom surface of the groove is ≤0.02 mm. After ultrasonic cleaning, it is bonded to the copper plate by vacuum brazing. The vacuum brazing parameters are vacuum degree ≤5×10⁻³ Pa and brazing temperature 720±10℃. A closed air gap is formed in the groove to accurately simulate the completely non-fusion state of the copper-aluminum interface.

5. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 4, characterized in that: The diameter tolerance of the flat-bottomed blind hole in the aluminum side hole defect test block A is ±0.05 mm, and the flatness of the hole bottom is ≤0.02 mm.

6. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 4, characterized in that: The method for fabricating the interface non-fusion defect test block C further includes: ultrasonic cleaning of the inside of the groove before brazing to remove wire cutting fluid residue; vacuum brazing cooling rate ≤ 5 ℃ / min; after fabrication, the air gap size and position are confirmed to meet the design requirements by sequentially performing precision dimensional inspection (tolerance ±0.1 mm), surface roughness inspection (Ra≤6.3 μm), fluorescence penetrant testing, ultrasonic acceptance scanning, and X-ray inspection.

7. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 4, characterized in that: The three sets of test blocks simultaneously support system calibration in both aluminum-side incident detection and copper-side incident detection directions. The standard reflection signal database established in conjunction with the test blocks is classified and stored according to defect type × detection direction × defect size. Each record contains four parameters: echo amplitude dB value, sound path time μs, waveform image, and signal-to-noise ratio dB.

8. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: In step S2, the E-scan parameters are: excitation aperture 16 wafers, step 1 wafer, focusing depth 8 mm; the S-scan parameters are: scanning angle range 40°~70°, angle step 0.5°, focusing depth 8 mm; and the scanning grid step is 5 mm×5 mm.

9. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: In step S4, the SAFT algorithm improves the horizontal resolution of the image from 2.5 mm to 1.2 mm, and the minimum detectable defect size is ≤ Φ2 mm. The signal-to-noise ratio for detecting unfused areas at the interface exceeds 20 dB, and the signal-to-noise ratio for unfused defects at the interface detected on the aluminum side is 8 to 12 dB higher than that detected on the copper side.

10. The ultrasonic phased array detection method for the bonding state of copper-aluminum brazed structures according to claim 1, characterized in that: The specific method of the relative amplitude method in step S5 is as follows: move the probe to find the position of the maximum echo of the defect, move it to the surrounding area until the echo amplitude drops to 20% of the maximum value, record it as the boundary point, and connect the boundary points to form the area as the equivalent area of ​​the defect; when quantifying the natural defect, a correction coefficient of 1.1 to 1.3 is introduced to compensate for the underestimation of the effective reflection area by the irregular defect surface.

Images