Precise metal part surface detection method

Abstract

The invention relates to the field of surface detection, and particularly discloses a precision metal part surface detection method. The precision metal part surface detection method comprises the following steps: starting a laser to scan the surface of a precision metal part to be detected point by point, exciting an ultrasonic surface wave at each scanning point, receiving a signal in real time by an ultrasonic sensor, digitalizing the signal through a signal acquisition card, transmitting the signal to a computer, and acquiring an ultrasonic surface wave signal for 5-10 times at each scanning point; and the computer filters the collected signals, extracts time domain features from the filtered signals, calculates a peak amplitude mean value, a standard deviation, rise time and fall time of each signal, and judges that defects exist when the peak amplitude mean value is reduced by more than 20% relative to the standard signal, the standard deviation is increased and the rise time is prolonged by more than 10%. The precision metal part surface detection method has the advantages of being high in detection speed and high in detection precision.

Description

Technical Field

[0001] The present invention relates to the field of surface detection, and more specifically, to a method for detecting the surface of precision metal parts. Background Art

[0002] Precision metal parts play a vital role in modern manufacturing. They are widely used in aerospace, automobile, medical equipment and electronic products. These parts usually require high precision, wear resistance and excellent mechanical properties, so surface quality control is particularly important. Common materials include stainless steel, aluminum alloy, titanium alloy, etc., which are precision-processed by CNC machine tools, such as turning, milling and drilling operations, to ensure that the geometry and size of the parts meet the design requirements. Parts may undergo heat treatment processes such as quenching and tempering to adjust the hardness and corrosion resistance of the material.

[0003] The surface quality of precision metal parts is an important reflection of their overall performance. Surface defects such as scratches, cracks, pits, etc. will not only affect the aesthetics of the parts, but more importantly, will weaken their mechanical properties and fatigue life, and may even cause safety accidents. Therefore, it is particularly important to conduct high-precision and high-efficiency inspection of the surface of precision metal parts. Common inspection methods include optical microscopy and non-destructive testing. The advantage of optical microscopy is that it can quickly identify surface defects, cracks and contamination, but there are many limitations on the volume and shape of the inspected object. Non-destructive tests such as ultrasonic testing, eddy current testing, and magnetic particle testing can detect surface and internal defects, but the equipment cost is high, usually requires professional skills to operate, and the inspection speed is slow, which is not suitable for mass production. Summary of the invention

[0004] In order to improve detection accuracy, the present application provides a method for detecting the surface of precision metal parts.

[0005] The present application provides a method for detecting the surface of a precision metal part using the following technical solution: A method for detecting the surface of a precision metal part comprises the following steps: After cleaning the precision metal parts to be inspected, fix them on the inspection platform; Start the laser, scan the surface of the part point by point according to the preset scanning path and angle, stimulate the ultrasonic surface wave at each scanning point, the ultrasonic sensor receives the signal in real time and transmits the signal to the computer after digitizing it through the signal acquisition card, and collects 5-10 ultrasonic surface wave signals at each scanning point; The computer uses signal processing software to filter the collected signal using a bandpass filter and normalize the filtered signal to the range of 0-1; Extract time domain features, and calculate the peak amplitude mean and standard deviation, rise time and fall time of each signal. When the peak amplitude mean decreases by more than 20% relative to the standard signal and the standard deviation increases and the rise time is extended by more than 10%, it is judged that there is a defect.

[0006] Through the above scheme, the whole detection process is non-contact, using the laser to excite ultrasonic surface waves and the ultrasonic sensor to receive signals. Different from the traditional contact detection method (such as using the probe to directly contact the surface of the part for detection), the physical contact between the probe and the surface of the part will not cause secondary damage such as scratches and wear to the originally smooth surface of the precision metal parts, which well protects the surface quality of the parts. It is especially suitable for precision metal parts detection in the fields of aerospace, high-end electronic equipment, etc., which have extremely high requirements for surface accuracy.

[0007] By carefully processing and analyzing the characteristics of the collected ultrasonic surface wave signals, and using a bandpass filter to remove high-frequency noise and low-frequency interference in the signal, the signal can be effectively purified, the influence of irrelevant interference factors on subsequent judgments can be reduced, and the true state of the part surface can be more accurately reflected. Defect judgment is made when the mean of the peak amplitude decreases by more than 20% relative to the standard signal, the standard deviation increases, and the rise time is extended by more than 10%. Such clear and quantitative standard judgment conditions help to accurately identify subtle abnormalities on the surface of parts, and can detect surface defects with a size of tens of microns, greatly improving the accuracy of defect identification. In addition, each scanning point collects 5-10 ultrasonic surface wave signals, and the accidental error of the signal is further reduced by averaging multiple acquisitions, ensuring that the acquired signal characteristics can better represent the true situation of the scanning point, making the detection results more reliable, and avoiding misjudgment or omission of defects due to fluctuations in single signal acquisition.

[0008] Optionally, the laser irradiation angle is 30-60° with respect to the normal line of the part surface.

[0009] Through the above scheme, within the 30-60° angle range, the laser energy can act more fully on a certain depth range of the surface, allowing the surface particles to produce more regular and sufficiently strong vibrations to form ultrasonic surface waves, thereby ensuring that the signals received by the subsequent ultrasonic sensor have a higher amplitude and good stability, which is convenient for more accurate analysis of signal characteristics to detect defects.

[0010] Optionally, the laser output wavelength is 1064nm, the pulse width is 10ns, and the energy density is 1-10J / cm 2 .

[0011] Through the above scheme, the electrons in the metal have a suitable absorption probability for the laser photons of this wavelength, so that the laser energy can be effectively converted into heat energy, and then based on the thermoelastic effect or ablation effect, a rapid heat-mechanical conversion process is triggered on the surface of the part, and ultrasonic surface waves are efficiently stimulated. Different metal materials have different absorption characteristics of light, but the 1064nm wavelength can show good excitation effects in many commonly used precision metal materials such as aluminum alloys, stainless steel, titanium alloys, etc., ensuring that this detection method is applicable to a variety of metal parts. The combination of a pulse width of 10ns and a wavelength of 1064nm can transfer high energy to the surface of metal parts in a very short time, instantly forming a strong thermal shock, causing the surface material to expand and contract rapidly, thereby generating clear ultrasonic surface waves with sufficient intensity. By adjusting the energy density between 1-10J / cm 2 Within the range, it can avoid irreversible damage to the surface of precision metal parts caused by excessive energy, such as ablation pits, material melting deformation, etc. On the premise of ensuring that ultrasonic surface waves can be effectively excited for detection, the original surface accuracy and quality of the parts are maintained.

[0012] Optionally, when the thickness of the precision metal part is 5-10 mm, the energy density of the laser is controlled at 1-5 J / cm 2 .

[0013] Optionally, the ultrasonic sensors are evenly distributed around the metal part to be measured, and the distance between the ultrasonic sensors and the surface of the metal part to be measured is 2-3 mm.

[0014] Through the above scheme, the ultrasonic sensors are evenly distributed around the metal parts to be measured, which can realize the comprehensive capture of the ultrasonic surface wave signals propagating from all directions on the surface of the parts. No matter where the ultrasonic surface wave is generated and propagated from the surface of the part and at what angle, the sensors distributed around have a greater probability of receiving it, avoiding the detection blind area caused by the limited position of the sensor. For precision metal parts with complex curved surfaces or irregular shapes, the propagation path of ultrasonic surface waves on their surfaces is complex and changeable. Evenly distributed sensors can ensure that the signals propagated from different curved surface areas can be effectively received, thereby obtaining relevant information on the entire surface of the part to the greatest extent. The distance between the ultrasonic sensor and the surface of the metal part to be measured is set at 2-3mm. This distance range helps to receive ultrasonic surface wave signals of appropriate intensity and good quality. If the distance is too close, the ultrasonic surface wave may interfere with its propagation due to the presence of the sensor, affecting its normal reflection, scattering and other characteristics, and may also make the signal received by the sensor too strong and exceed its optimal receiving range, resulting in signal distortion; if the distance is too far, the ultrasonic surface wave will have a large energy attenuation during the propagation process, making the signal strength received by the sensor weaker, and may even be so weak that it is difficult to accurately distinguish the signal characteristics.

[0015] Optionally, the frequency response range of the ultrasonic sensor is 2-8 MHz, and the minimum detection signal amplitude is 5 μV.

[0016] Through the above scheme, the frequency response range of 2-8MHz can ensure that the sensor can capture the ultrasonic surface wave signals generated in these different situations, avoiding the omission of certain key signals due to frequency mismatch, thereby improving the universality of surface detection of various types of precision metal parts. The minimum detection signal amplitude of 5μV gives the ultrasonic sensor a higher sensitivity. The sensor can capture the weak signal changes caused by tiny defects, making it possible to detect tiny defects.

[0017] Optionally, the sampling frequency of the signal acquisition card is 80 MHz, and the resolution of the signal acquisition card is 16 bits.

[0018] Through the above scheme, the combination of high sampling frequency and high resolution enables the signal acquisition card to comprehensively and accurately collect ultrasonic surface wave signals, thereby more realistically reflecting the surface conditions of precision metal parts.

[0019] Optionally, the bandpass filter has a low cutoff frequency of 1.5 MHz and a high cutoff frequency of 8.5 MHz.

[0020] Through the above scheme, the low cutoff frequency of the bandpass filter is set to 1.5MHz and the high cutoff frequency is set to 8.5MHz, which can accurately screen out the main frequency range of the ultrasonic surface wave signal, and effectively remove low-frequency interference below 1.5MHz (such as low-frequency fluctuations caused by environmental vibration, power supply frequency interference, etc.) and high-frequency noise above 8.5MHz (such as high-frequency thermal noise generated by electronic components themselves). In this way, the ultrasonic surface wave signal is greatly purified, so that the subsequent analysis of the signal can focus on the effective information that is truly related to the surface condition of the part, reducing the impact of irrelevant interference factors on the detection results.

[0021] In summary, this application has the following beneficial effects: 1. Since the present application uses a laser to scan the surface of the part point by point to stimulate ultrasonic surface waves for surface detection, the method adopts non-contact detection, avoiding secondary damage such as scratches and wear on the surface of the part caused by traditional contact detection, which is beneficial to protecting the surface quality of precision metal parts in fields such as aerospace and high-end electronic equipment, and greatly improves the accuracy of defect identification.

[0022] 2. In this application, the laser irradiation angle is preferably 30-60° with the normal line of the part surface, so that the laser energy can fully act on a certain depth range of the surface, so that the surface particles produce regular and strong enough vibration to form ultrasonic surface waves with high amplitude and good stability, which is convenient for accurate analysis of signal characteristics to detect defects; and the laser output wavelength of 1064nm can allow the electrons in the metal to effectively absorb laser photons and convert them into thermal energy, and efficiently excite ultrasonic surface waves based on the thermoelastic effect or ablation effect, and is suitable for a variety of commonly used precision metal materials. The 10ns pulse width can instantly form a strong thermal shock to produce clear and strong enough ultrasonic surface waves, which can be adjusted to 1-10J / cm 2 The energy density can effectively stimulate ultrasonic surface waves for detection, while avoiding damage to the surface of parts and maintaining their original accuracy and quality.

[0023] 3. The method of the present application is applicable to surface defect detection of various metal parts, is not limited by the appearance and shape of the metal parts, and has high universality. DETAILED DESCRIPTION

[0024] The present application is further described in detail below in conjunction with the examples. It is particularly noted that if no specific conditions are specified in the following examples, the experiments are carried out according to conventional conditions or the conditions recommended by the manufacturer. Unless otherwise specified, the raw materials used in the following examples can be obtained from common commercial sources. Example

[0025] Example 1 A method for detecting the surface of precision metal parts: Laser Selection and Setup The Nd:YAG laser is selected, with an output wavelength of 1064nm and a pulse width of 10ns. The laser is mounted on an optical platform that can precisely adjust the angle and position. Through the computer control system, the laser beam can be irradiated at different positions on the surface of the part at a specific angle to achieve comprehensive scanning.

[0026] Choose a piezoelectric ultrasonic sensor with a frequency response range of 2-8MHz, which can effectively receive ultrasonic surface wave signals generated on the surface of metal parts. The sensor should have high sensitivity and a minimum detectable signal amplitude of 5μV to ensure that weak signal changes can be captured.

[0027] Arrange sensors according to the shape and size of the parts. For rectangular metal parts, install a sensor at the midpoint of each of its four sides; for circular parts, arrange 4 sensors evenly in the circumferential direction. The sensor is fixed at a distance of 2-3mm from the surface of the part by a high-precision fixture to ensure stable signal reception without affecting the propagation of ultrasonic surface waves.

[0028] Connect the sensor to the signal acquisition card through a low-noise, well-shielded cable to reduce the impact of external electromagnetic interference on the signal.

[0029] A high-speed signal acquisition card is used, and its sampling frequency is set to 80MHz, which can meet the requirements of ultrasonic surface wave signal acquisition. The resolution of the acquisition card is 16 bits, which can accurately quantify the signal amplitude.

[0030] A USB 3.2 high-speed data transmission interface is used between the signal acquisition card and the computer. At the same time, a special signal acquisition software is installed in the computer for real-time acquisition, display and storage of ultrasonic surface wave signals.

[0031] Testing process Clean the precision metal parts to be tested to remove oil, dust and other impurities on the surface to ensure the surface is clean and avoid interference with the test results.

[0032] Place the parts on the testing platform and fix them with mechanical clamps to ensure that the parts are in a stable position and will not move during the testing process.

[0033] Start the laser, and scan the surface of the part point by point at a 45° angle between the laser irradiation angle and the normal line of the part surface. The laser energy density is 5J / cm 2 At each scan point, the laser pulse excites the generation of ultrasonic surface waves.

[0034] The ultrasonic sensor receives the ultrasonic surface wave signal in real time, and transmits the signal to the computer after digitizing it through the signal acquisition card. During the acquisition process, the ultrasonic surface wave signal is collected 8 times for each scanning point.

[0035] In the computer, the collected signal is filtered using signal processing software. A bandpass filter is used with the low cutoff frequency set to 1.5MHz and the high cutoff frequency set to 8.5MHz to remove high-frequency noise and low-frequency interference in the signal.

[0036] The filtered signal is normalized and the signal amplitude is normalized to the range of 0-1 using the maximum-minimum normalization method.

[0037] Extract time domain features and analyze the peak amplitude, rise time, and fall time of each scan point signal. Calculate the peak amplitude mean and standard deviation of each signal. If the peak amplitude mean drops by more than 20% relative to the defect-free standard signal and the standard deviation increases, there is a defect. At the same time, record the changes in rise time and fall time. If the rise time is extended by more than 10%, it may indicate an abnormality.

[0038] The extracted signal features are compared with the pre-established defect feature database. The database contains the ultrasonic surface wave signal features of different types of defects on the metal part obtained through experiments. Based on the comparison results, the type of defect is determined.

[0039] Example 2 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser irradiation angle is 30° to the normal line of the part surface.

[0040] Example 3 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser irradiation angle is 60° to the normal line of the part surface.

[0041] Example 4 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser output energy density is 1 J / cm 2 .

[0042] Example 5 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser output energy density is 10 J / cm 2 .

[0043] Example 6 A method for detecting the surface of precision metal parts: The difference from Example 1 is that ultrasonic surface wave signals are collected 5 times at each scanning point.

[0044] Example 7 A method for detecting the surface of precision metal parts: The difference from Example 1 is that ultrasonic surface wave signals are collected 10 times at each scanning point.

[0045] Comparative Example 1 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser irradiation angle is 20° with the normal line of the part surface.

[0046] Comparative Example 2 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser irradiation angle is 70° to the normal line of the part surface.

[0047] Comparative Example 3 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the laser output energy density is 12 J / cm 2 .

[0048] Comparative Example 4 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the distance between the ultrasonic sensor and the surface of the metal part to be detected is 1-2 mm.

[0049] Comparative Example 5 A method for detecting the surface of a precision metal part: The difference from Example 1 is that the distance between the ultrasonic sensor and the surface of the metal part to be detected is 3-4 mm.

[0050] Detection Methods A precision-machined shaft with a diameter of 120 mm*1250 mm was used as a test sample. The surface defect data of the sample was obtained by magnetic particle testing. The data was used as a reference standard and then retested using the testing methods of the embodiment and the comparative example to compare the data difference obtained from the two tests.

[0051] Table 1 Test data It can be seen from Examples 1-3 and Comparative Examples 1-2 in combination with Table 1 that within the 30-60° angle range, the laser energy can more fully act on a certain depth range of the surface, allowing the surface particles to produce more regular and strong enough vibrations to form ultrasonic surface waves, thereby ensuring that the signals received by the subsequent ultrasonic sensor have a higher amplitude and good stability, which is convenient for more accurate analysis of signal characteristics to detect defects. When the ultrasonic surface wave propagates in a direction at an angle of 30-60° to the normal and interacts with the defect, the reflected, scattered and other changed signals are received by multiple ultrasonic sensors, and combined with the propagation characteristics and geometric relationships of the ultrasonic surface wave at a specific angle, the position of the defect on the surface of the part can be more accurately calculated.

[0052] It can be seen from Examples 1, 4, 5 and Comparative Example 3 in combination with Table 1 that by adjusting the energy density to 1-10 J / cm 2 Within the range, it can avoid irreversible damage to the surface of precision metal parts caused by excessive energy, such as ablation pits, material melting deformation, etc. On the premise of ensuring that ultrasonic surface waves can be effectively excited for detection, the original surface accuracy and quality of the parts are maintained.

[0053] It can be seen from Example 1 and Comparative Examples 4-5 in combination with Table 1 that the distance between the ultrasonic sensor and the surface of the metal part being measured is set at 2-3mm. This distance range is helpful for receiving ultrasonic surface wave signals of suitable intensity and good quality. If the distance is too close, the ultrasonic surface wave may interfere with its propagation due to the presence of the sensor, affecting its normal reflection, scattering and other characteristics. At the same time, it may also make the signal received by the sensor too strong and exceed its optimal receiving range, resulting in signal distortion; while the distance is too far, the ultrasonic surface wave will have a large energy attenuation during the propagation process, making the signal strength received by the sensor weak, and it may even be so weak that it is difficult to accurately distinguish the signal characteristics. Maintaining a distance of 2-3mm allows the ultrasonic surface wave to propagate to the sensor in a relatively ideal state, ensuring that the signal received by the sensor has sufficient amplitude, clear waveform, etc., which is convenient for subsequent accurate analysis and processing of the signal.

[0054] It can be seen from Example 1 and Examples 6-7 in combination with Table 1 that the accidental error of the signal is further reduced by taking an average through multiple acquisitions, ensuring that the acquired signal characteristics can better represent the actual situation of the scanning point, making the detection result more reliable, and avoiding misjudgment or omission of defects due to fluctuations in a single signal acquisition.

[0055] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed. However, as long as it is within the scope of the claims of the present application, it shall be protected by the patent law.

Claims

1. A method for detecting the surface of a precision metal part, characterized in that: The following steps are involved: After cleaning the precision metal parts to be inspected, fix them on the inspection platform; Start the laser, scan the surface of the part point by point according to the preset scanning path and angle, stimulate the ultrasonic surface wave at each scanning point, the ultrasonic sensor receives the signal in real time and transmits the signal to the computer after digitizing it through the signal acquisition card, and collects 5-10 ultrasonic surface wave signals at each scanning point; The computer uses signal processing software to filter the collected signal using a bandpass filter and normalize the filtered signal to the range of 0-1; The time domain features are extracted, and the peak amplitude mean and standard deviation, rise time and fall time of each signal are calculated. When the peak amplitude mean decreases by more than 20% relative to the standard signal and the standard deviation increases and the rise time is prolonged by more than 10%, it is judged that a defect exists.

2. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The laser irradiation angle is 30-60° with the normal line of the part surface.

3. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The laser output wavelength is 1064nm, the pulse width is 10ns, and the energy density is 1-10J / cm².

4. The method for detecting the surface of precision metal parts according to claim 3, characterized in that: When the thickness of the precision metal parts is 5-10 mm, the energy density of the laser is controlled at 1-5 J / cm².

5. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The ultrasonic sensors are evenly distributed around the metal part to be measured, and the distance between the ultrasonic sensors and the surface of the metal part to be measured is 2-3 mm.

6. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The frequency response range of the ultrasonic sensor is 2-8 MHz, and the minimum detection signal amplitude is 5 μV.

7. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The sampling frequency of the signal acquisition card is 80 MHz, and the resolution of the signal acquisition card is 16 bits.

8. The method for detecting the surface of precision metal parts according to claim 1, characterized in that: The bandpass filter has a low cutoff frequency of 1.5 MHz and a high cutoff frequency of 8.5 MHz.

Images