For Printed Circuit Board(PCB)surface defect detection,traditional detection methods mostly focus on template matching-based reference method and manual detections,which have the disadvantages of low defect detection ...For Printed Circuit Board(PCB)surface defect detection,traditional detection methods mostly focus on template matching-based reference method and manual detections,which have the disadvantages of low defect detection efficiency,large errors in defect identification and localization,and low versatility of detectionmethods.In order to furthermeet the requirements of high detection accuracy,real-time and interactivity required by the PCB industry in actual production life.In the current work,we improve the Youonly-look-once(YOLOv4)defect detection method to train and detect six types of PCB small target defects.Firstly,the original Cross Stage Partial Darknet53(CSPDarknet53)backbone network is preserved for PCB defect feature information extraction,and secondly,the original multi-layer cascade fusion method is changed to a single-layer feature layer structure to greatly avoid the problem of uneven distribution of priori anchor boxes size in PCB defect detection process.Then,the K-means++clustering method is used to accurately cluster the anchor boxes to obtain the required size requirements for the defect detection,which further improves the recognition and localization of small PCB defects.Finally,the improved YOLOv4 defect detection model is compared and analyzed on PCB dataset with multi-class algorithms.The experimental results show that the average detection accuracy value of the improved defect detection model reaches 99.34%,which has better detection capability,lower leakage rate and false detection rate for PCB defects in comparison with similar defect detection algorithms.展开更多
The quality of printed circuit board(PCB)micro-hole processing directly determines the stability of the inner and outer circuit connections.Micro-hole drilling technology is a typical method for PCB micro-hole process...The quality of printed circuit board(PCB)micro-hole processing directly determines the stability of the inner and outer circuit connections.Micro-hole drilling technology is a typical method for PCB micro-hole processing.The problem of optimal control of its drilling force is one of the main factors affecting the quality of micro-hole machining.To address this problem,the thrust forces and torques in PCB drilling were first modeled and analyzed,and the corresponding prediction models were established.The drilling force analysis was carried out through the micro-hole drilling experiment,the specific cutting energy under different feed rates was calculated,the influence of the size effect was clarified,and the accuracy of the prediction model was verified.The result shows that during the drilling of glass fiber cloth,changes in the material removal mechanism are induced as the feed per revolution is varied.When the feed per revolution is less than the tool edge radius,the glass fiber is not cut by the main cutting edge,but is crushed and broken.When the feed per revolution is greater than the radius of the tool edge,the glass fiber is cut by the main cutting edge.At the same time,the established analytical model can accurately reflect the influence of the size effect on the drilling torque in PCB micro-hole drilling,and the error is within 10%.This method has certain practical application value in controlling PCB micro hole processing quality.展开更多
Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including lo...Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.展开更多
In order to study the role of printed circuit board(PCB)in high-power LED heat dissipation,a simple model of high-power LED lamp was designed.According to this lamp model,some thermal performances such as thermal resi...In order to study the role of printed circuit board(PCB)in high-power LED heat dissipation,a simple model of high-power LED lamp was designed.According to this lamp model,some thermal performances such as thermal resistances of four types of PCB and the changes of LED junction temperature were tested under three different working currents.The obtained results indicate that LED junction temperature can not be lowered significantly with the decreasing thermal resistance of PCB.However,PCB with low thermal resistance can be matched with smaller volume heat sink,so it is hopeful to reduce the size,weight and cost of LED lamp.展开更多
Electronic scrap, especially wasted printed circuit boards (PCBs), is regarded as an environmental challenge. At present, the physical separation is thought to be the environmental friendly and economical method of tr...Electronic scrap, especially wasted printed circuit boards (PCBs), is regarded as an environmental challenge. At present, the physical separation is thought to be the environmental friendly and economical method of treating and reutilizing electronic waste. An effective liberation of metals from non metallic components is a crucial step towards mechanical separation and recycling of wasted PCBs. In this paper, the selective shredding theory and mechanics characteristics of wasted PCBs were analyzed, and the shredded experiments of wasted PCBs by hammer mill were investigated. The result shows that the selective shredding exists in the wasted PCBs shredded process by hammer mill. The shredding velocity of non metallic components is far greater than that of metals in the wasted PCBs shredding, which makes the metals concentrate in the coarser fraction. And the impact force of hammer mill is superior to metal liberation from non metallic components, a satisfied metal liberation degree can be achieved in the wasted PCBs shredding by hammer mill.展开更多
Printed-circuit board (PCB) elliptic antennas with useful bandwidth exceeding 10:1 ratio are suitable for wideband radar, wireless ultra wideband (UWB) and other wireless communication applications. We present wideban...Printed-circuit board (PCB) elliptic antennas with useful bandwidth exceeding 10:1 ratio are suitable for wideband radar, wireless ultra wideband (UWB) and other wireless communication applications. We present wideband PCB elliptic dipole antennas which are capable of achieving the bandwidth requirements for all the applications. A set of elliptic dipole antennas with varying eccentricities have been fabricated for demonstration. We find one specific size (specific eccentricity) dipole that can yield an impressive 1.5-16 bandwidth exceeding the currently available. A couple of elliptic dipole antennas suitable for UWB application have been presented. We have measured swept frequency response, impedance and radiation patterns of all dipoles. An empirical formula is given for calculating the starting resonant frequency within the operating band. The calculated values are found in good agreement with measured results.展开更多
Printed Circuit Boards(PCBs)are materials used to connect components to one another to form a working circuit.PCBs play a crucial role in modern electronics by connecting various components.The trend of integrating mo...Printed Circuit Boards(PCBs)are materials used to connect components to one another to form a working circuit.PCBs play a crucial role in modern electronics by connecting various components.The trend of integrating more components onto PCBs is becoming increasingly common,which presents significant challenges for quality control processes.Given the potential impact that even minute defects can have on signal traces,the surface inspection of PCB remains pivotal in ensuring the overall system integrity.To address the limitations associated with manual inspection,this research endeavors to automate the inspection process using the YOLOv8 deep learning algorithm for real-time fault detection in PCBs.Specifically,we explore the effectiveness of two variants of the YOLOv8 architecture:YOLOv8 Small and YOLOv8 Nano.Through rigorous experimentation and evaluation of our dataset which was acquired from Peking University’s Human-Robot Interaction Lab,we aim to assess the suitability of these models for improving fault detection accuracy within the PCB manufacturing process.Our results reveal the remarkable capabilities of YOLOv8 Small models in accurately identifying and classifying PCB faults.The model achieved a precision of 98.7%,a recall of 99%,an accuracy of 98.6%,and an F1 score of 0.98.These findings highlight the potential of the YOLOv8 Small model to significantly improve the quality control processes in PCB manufacturing by providing a reliable and efficient solution for fault detection.展开更多
基金This work was funded by the Natural Science Research Project of Higher Education Institutions in Jiangsu Province(No.20KJA520007)Min Zhang receives the grant and the URLs to sponsors’websites are http://jyt.jiangsu.gov.cn/.
文摘For Printed Circuit Board(PCB)surface defect detection,traditional detection methods mostly focus on template matching-based reference method and manual detections,which have the disadvantages of low defect detection efficiency,large errors in defect identification and localization,and low versatility of detectionmethods.In order to furthermeet the requirements of high detection accuracy,real-time and interactivity required by the PCB industry in actual production life.In the current work,we improve the Youonly-look-once(YOLOv4)defect detection method to train and detect six types of PCB small target defects.Firstly,the original Cross Stage Partial Darknet53(CSPDarknet53)backbone network is preserved for PCB defect feature information extraction,and secondly,the original multi-layer cascade fusion method is changed to a single-layer feature layer structure to greatly avoid the problem of uneven distribution of priori anchor boxes size in PCB defect detection process.Then,the K-means++clustering method is used to accurately cluster the anchor boxes to obtain the required size requirements for the defect detection,which further improves the recognition and localization of small PCB defects.Finally,the improved YOLOv4 defect detection model is compared and analyzed on PCB dataset with multi-class algorithms.The experimental results show that the average detection accuracy value of the improved defect detection model reaches 99.34%,which has better detection capability,lower leakage rate and false detection rate for PCB defects in comparison with similar defect detection algorithms.
基金National Natural Science Foundation of China(No.51805079)Fundamental Research Funds for the Central Universities,China(No.2232021D-15)Shanghai Science and Technology Program(No.20DZ2251400)。
文摘The quality of printed circuit board(PCB)micro-hole processing directly determines the stability of the inner and outer circuit connections.Micro-hole drilling technology is a typical method for PCB micro-hole processing.The problem of optimal control of its drilling force is one of the main factors affecting the quality of micro-hole machining.To address this problem,the thrust forces and torques in PCB drilling were first modeled and analyzed,and the corresponding prediction models were established.The drilling force analysis was carried out through the micro-hole drilling experiment,the specific cutting energy under different feed rates was calculated,the influence of the size effect was clarified,and the accuracy of the prediction model was verified.The result shows that during the drilling of glass fiber cloth,changes in the material removal mechanism are induced as the feed per revolution is varied.When the feed per revolution is less than the tool edge radius,the glass fiber is not cut by the main cutting edge,but is crushed and broken.When the feed per revolution is greater than the radius of the tool edge,the glass fiber is cut by the main cutting edge.At the same time,the established analytical model can accurately reflect the influence of the size effect on the drilling torque in PCB micro-hole drilling,and the error is within 10%.This method has certain practical application value in controlling PCB micro hole processing quality.
基金supported in part by the IoT Intelligent Microsystem Center of Tsinghua University-China Mobile Joint Research Institute.
文摘Automated optical inspection(AOI)is a significant process in printed circuit board assembly(PCBA)production lines which aims to detect tiny defects in PCBAs.Existing AOI equipment has several deficiencies including low throughput,large computation cost,high latency,and poor flexibility,which limits the efficiency of online PCBA inspection.In this paper,a novel PCBA defect detection method based on a lightweight deep convolution neural network is proposed.In this method,the semantic segmentation model is combined with a rule-based defect recognition algorithm to build up a defect detection frame-work.To improve the performance of the model,extensive real PCBA images are collected from production lines as datasets.Some optimization methods have been applied in the model according to production demand and enable integration in lightweight computing devices.Experiment results show that the production line using our method realizes a throughput more than three times higher than traditional methods.Our method can be integrated into a lightweight inference system and pro-mote the flexibility of AOI.The proposed method builds up a general paradigm and excellent example for model design and optimization oriented towards industrial requirements.
基金Special Fund Project of Science and Technology Innovation of Dongli District(21090302)Research Projectof Applied Basic and Front Technologies of Tianjin(10JCZDJC15400)
文摘In order to study the role of printed circuit board(PCB)in high-power LED heat dissipation,a simple model of high-power LED lamp was designed.According to this lamp model,some thermal performances such as thermal resistances of four types of PCB and the changes of LED junction temperature were tested under three different working currents.The obtained results indicate that LED junction temperature can not be lowered significantly with the decreasing thermal resistance of PCB.However,PCB with low thermal resistance can be matched with smaller volume heat sink,so it is hopeful to reduce the size,weight and cost of LED lamp.
文摘Electronic scrap, especially wasted printed circuit boards (PCBs), is regarded as an environmental challenge. At present, the physical separation is thought to be the environmental friendly and economical method of treating and reutilizing electronic waste. An effective liberation of metals from non metallic components is a crucial step towards mechanical separation and recycling of wasted PCBs. In this paper, the selective shredding theory and mechanics characteristics of wasted PCBs were analyzed, and the shredded experiments of wasted PCBs by hammer mill were investigated. The result shows that the selective shredding exists in the wasted PCBs shredded process by hammer mill. The shredding velocity of non metallic components is far greater than that of metals in the wasted PCBs shredding, which makes the metals concentrate in the coarser fraction. And the impact force of hammer mill is superior to metal liberation from non metallic components, a satisfied metal liberation degree can be achieved in the wasted PCBs shredding by hammer mill.
文摘Printed-circuit board (PCB) elliptic antennas with useful bandwidth exceeding 10:1 ratio are suitable for wideband radar, wireless ultra wideband (UWB) and other wireless communication applications. We present wideband PCB elliptic dipole antennas which are capable of achieving the bandwidth requirements for all the applications. A set of elliptic dipole antennas with varying eccentricities have been fabricated for demonstration. We find one specific size (specific eccentricity) dipole that can yield an impressive 1.5-16 bandwidth exceeding the currently available. A couple of elliptic dipole antennas suitable for UWB application have been presented. We have measured swept frequency response, impedance and radiation patterns of all dipoles. An empirical formula is given for calculating the starting resonant frequency within the operating band. The calculated values are found in good agreement with measured results.
文摘Printed Circuit Boards(PCBs)are materials used to connect components to one another to form a working circuit.PCBs play a crucial role in modern electronics by connecting various components.The trend of integrating more components onto PCBs is becoming increasingly common,which presents significant challenges for quality control processes.Given the potential impact that even minute defects can have on signal traces,the surface inspection of PCB remains pivotal in ensuring the overall system integrity.To address the limitations associated with manual inspection,this research endeavors to automate the inspection process using the YOLOv8 deep learning algorithm for real-time fault detection in PCBs.Specifically,we explore the effectiveness of two variants of the YOLOv8 architecture:YOLOv8 Small and YOLOv8 Nano.Through rigorous experimentation and evaluation of our dataset which was acquired from Peking University’s Human-Robot Interaction Lab,we aim to assess the suitability of these models for improving fault detection accuracy within the PCB manufacturing process.Our results reveal the remarkable capabilities of YOLOv8 Small models in accurately identifying and classifying PCB faults.The model achieved a precision of 98.7%,a recall of 99%,an accuracy of 98.6%,and an F1 score of 0.98.These findings highlight the potential of the YOLOv8 Small model to significantly improve the quality control processes in PCB manufacturing by providing a reliable and efficient solution for fault detection.
