WeLinkirt's 3D AI AOI equipment (self-developed 3D camera + 3D morphology reconstruction/point cloud, detecting hidden solder joints/coplanarity/micron-level morphology/voids and other 2D optical blind spot defects, with 2D-3D fusion), through on-premises private deployment, addressed a leading electronics manufacturer's challenge of balancing data security and precision in BGA void detection, reducing false positive rates by −45% and ensuring sensitive production data remains within the factory, significantly enhancing autonomous control and efficiency of production line inspection.
In the rapidly evolving electronics manufacturing industry, particularly in PCBA production, Ball Grid Array (BGA) packages are widely used due to their high density and performance. However, voids within BGA solder joints have consistently been a critical factor affecting product reliability. These voids not only weaken the mechanical strength and electrical conductivity of the solder joints but can also lead to product failure in extreme environments. For small and medium-sized enterprises (SMEs), balancing improved inspection efficiency with ensuring the security of core process and quality data, preventing data leaks or cloud-related risks, is a primary consideration when adopting advanced inspection technologies. WeLinkirt's 3D AI AOI equipment, deployed on-premises, is specifically designed to meet these stringent data security requirements, providing exceptional defect detection capabilities while safeguarding data sovereignty.
Pain Points: Why This Hurdle Is Difficult to Overcome
The challenges of detecting internal BGA solder voids manifest in several dimensions. Firstly, traditional 2D optical inspection methods are completely ineffective for internal defects, only capable of inspecting the solder joint surface morphology, leading to up to 70% of internal void defects being missed. Secondly, while traditional inspection equipment like X-ray can visualize internal structures, image interpretation is highly dependent on human experience, resulting in low interpretation efficiency and susceptibility to operator subjectivity, with false positive rates typically exceeding 10%. For SMEs, the acquisition and maintenance costs of expensive X-ray equipment are prohibitive, and the image resolution and automated analysis capabilities of X-ray devices often fall short of the precision required for micron-level defect identification. Furthermore, the manual review process is extremely time-consuming, often accounting for over 40% of the total inspection time, significantly slowing down production cycles. The most critical pain point lies in data security: many advanced AI inspection solutions rely on cloud computing power or data storage, which is an unacceptable risk for companies with strict confidentiality requirements for intellectual property and production data. They require a solution that can be 100% privately deployed on-premises, ensuring all sensitive data remains within the factory.
The root cause of these difficulties lies in the three-dimensional complexity of BGA solder joints. Voids can appear anywhere within the solder ball, vary in shape, and range in size from tens to hundreds of microns. Traditional 2D images lack depth information, making it impossible to distinguish between surface depressions and internal voids. While X-ray imaging can penetrate, its 2D projection still presents ambiguity regarding the shape, size, and location of 3D voids, especially when solder joints are densely packed, leading to increased interpretation difficulty due to overlapping images. At the same time, for AI smart cameras in SME manufacturing inspection, if easy deployment and localized solutions are not provided, data security concerns will be the biggest obstacle to their adoption. WeLinkirt recognizes that only by fundamentally addressing data security can more enterprises confidently embrace the advantages of AI vision inspection.
Technical Principles
The core of WeLinkirt's 3D AI AOI equipment lies in its self-developed high-precision 3D camera and advanced three-dimensional morphology reconstruction technology. Unlike traditional 2D vision systems, our 3D camera employs structured light or laser triangulation principles to rapidly acquire point cloud data from the surface of the object under inspection. This point cloud data contains precise three-dimensional coordinate information for each pixel, enabling complete reconstruction of the BGA solder joint's 3D morphology. For BGA void detection, the equipment first obtains precise contours and internal structural information of the solder joint surface through multi-angle, high-resolution 3D scanning. Subsequently, the DaoAI AI AOI software system utilizes deep learning algorithms to analyze the reconstructed 3D point cloud data. This algorithm, trained on a large dataset of real BGA void samples, can accurately identify tiny voids, bubbles, cracks, and other defects within the solder joint, and quantify them, such as void volume, position, and shape parameters. This 3D data-based analysis completely eliminates 2D vision blind spots, achieving comprehensive detection of hidden defects.
Compared to traditional X-ray inspection, WeLinkirt's 3D AI AOI offers significant improvements in both efficiency and accuracy. X-ray images are 2D projections, requiring experienced engineers for manual interpretation, which is prone to misjudgment and missed detections. In contrast, our 3D AI AOI equipment directly acquires and analyzes 3D data, automatically identifying and quantifying defects through AI algorithms, elevating the objectivity and consistency of inspection to a new level. Concurrently, by integrating 2D-3D fusion technology, WeLinkirt's equipment can also address traditional 2D defects on the solder joint surface (such as short circuits, open circuits, etc.), achieving comprehensive, one-stop inspection. Crucially, WeLinkirt's DaoAI AI AOI software system supports 100% on-premises private deployment, with all data processing and model inference performed on the client's local servers, ensuring absolute security of customer production data and meeting stringent data compliance requirements. This is vital for manufacturing enterprises with extremely high data security demands.
Typical Application Scenarios
- **Internal BGA Solder Void Detection:** This is one of the core applications of WeLinkirt's 3D AI AOI. Through 3D morphology reconstruction, the equipment can penetrate the solder ball surface to precisely identify internal air bubbles, voids, and cracks, quantifying their size, position, and proportion, overcoming the limitations of 2D vision and traditional X-ray interpretation.
- **BGA Solder Joint Coplanarity Inspection:** Ensuring all solder balls are on the same plane is critical for BGA solder quality. WeLinkirt's 3D camera can accurately measure the height of each solder ball, calculate overall coplanarity, and identify warped or collapsed solder joints, preventing open circuits or short circuits.
- **Micron-level Morphological Defect Detection:** For tiny scratches, dents, foreign objects, or deformations on the BGA solder joint surface, high-precision 3D scanning can capture subtle morphological features that are difficult for traditional 2D cameras to distinguish, ensuring product aesthetic and functional perfection.
- **Hidden Solder Joint Detection for Packages like QFN/LGA:** For leadless packages such as QFN (Quad Flat No-lead) and LGA (Land Grid Array), solder joints are located at the bottom of the device and are completely invisible to 2D vision. WeLinkirt's 3D AI AOI equipment, with its unique imaging and algorithms, can effectively inspect the soldering quality of these hidden solder joints, such as open circuits, missing solder, or insufficient solder volume, filling a gap in traditional inspection methods.
- **PCB Warpage and Component Height Measurement:** Beyond solder joint inspection, 3D AOI can precisely measure the overall warpage of PCB boards and the exact height of various components on the board, which is crucial for assembly precision and structural integrity.
Deployment Case Study
A leading Tier-1 electronics manufacturing supplier, whose production line primarily focuses on the assembly and testing of high-end PCBA products, had extremely high requirements for BGA solder joint quality and strict internal audit and compliance requirements for production data security. Previously, the company relied mainly on manual visual inspection combined with a small amount of X-ray sampling for BGA void detection, but faced severe missed detection and false positive issues, leading to high rework rates. The efficiency bottleneck of manual X-ray image interpretation became increasingly prominent. More concerning was that several cloud-based AI vision solutions they had tried could not meet their on-premises private deployment requirements, making data security risks the biggest obstacle to adopting advanced technologies. After in-depth research, they chose WeLinkirt's 3D AI AOI equipment for pilot deployment.
During the deployment process, the WeLinkirt team fully deployed the 3D AI AOI equipment within the client's internal network environment. All data acquisition, model training, and inference were completed on the client's own servers, ensuring that no data left the factory. Initially, the WeLinkirt DaoAI AI AOI software system, leveraging APDT few-shot learning technology, completed the rapid training and optimization of the BGA void detection model with only 15 good samples, significantly shortening the modeling cycle. After going live, compared with the original X-ray manual interpretation results, WeLinkirt's 3D AI AOI equipment demonstrated excellent performance: while achieving a detection rate of 99.6%, the false positive rate was reduced from the original 12% to <6.6%, and the overall rework volume decreased by −45%. More importantly, by achieving complete local deployment, the client's core production data was strictly protected, eliminating data security concerns and paving the way for larger-scale adoption of AI vision technology.
Through WeLinkirt's 3D AI AOI on-premises private deployment, we not only solved the challenge of precise BGA void detection but, more importantly, completely eliminated data security risks, keeping our core production data firmly in our own hands.
WeLinkirt Solutions and Products
WeLinkirt's 3D AI AOI solution for the electronics/PCBA industry, centered on its self-developed 3D camera and powerful DaoAI AI AOI software system, focuses on addressing the detection blind spots and efficiency bottlenecks of traditional 2D vision and X-ray. Our 3D AI AOI equipment can perform in-depth analysis of BGA solder joints through high-precision 3D morphology reconstruction, accurately identifying internal voids, micron-level morphological defects, and hidden solder joints in packages like QFN. For deployment, WeLinkirt offers highly flexible on-premises private deployment options, supporting various integration methods such as SDK / API / Docker, ensuring that all customer production data and AI models run on the enterprise's internal network, achieving data-out-of-factory prevention and meeting the highest levels of data security and compliance requirements. The modeling process is extremely simple; the DaoAI AI AOI software system supports APDT positive/few-shot learning, requiring only 1–20 good sample images to complete model training, reducing new product changeover downtime to 5min, greatly enhancing production line flexibility and efficiency.
In addition to the core 3D AI AOI equipment, WeLinkirt also provides the DaoAI AI AOI software system, whose visual foundation model possesses strong feature recognition capabilities, coupled with semantic false alarm filtering, to further optimize detection results. For enterprises requiring broader intelligent upgrades, our DaoAI World model serves as a unified foundation, enabling semantic understanding, cross-scenario generalization, and continuous learning from production line feedback, constantly improving detection accuracy and automation. Through the on-premises private deployment of WeLinkirt's 3D AI AOI equipment, customers not only gain high-precision defect detection capabilities but, more importantly, achieve complete control over core production data, reducing false positive rates by −45%, effectively reducing manual re-inspection workload, improving overall production efficiency and product quality, and ensuring the enterprise maintains a leading edge in fierce market competition.
FAQ
How does WeLinkirt's 3D AI AOI equipment ensure data security for BGA void detection?
WeLinkirt's 3D AI AOI equipment supports 100% on-premises private deployment. All data acquisition, AI model training, and inference are conducted within the client's internal servers and network environment, ensuring that sensitive production data never leaves the factory. This fundamentally eliminates cloud-based data leakage risks and meets stringent enterprise data compliance requirements.
What are the advantages of WeLinkirt's 3D AI AOI for BGA void detection compared to traditional X-ray?
Traditional X-ray relies on manual interpretation of 2D projected images, which is inefficient and prone to subjective errors. WeLinkirt's 3D AI AOI uses a self-developed 3D camera for 3D morphology reconstruction, automatically analyzing point cloud data with AI algorithms. This provides higher detection accuracy, precise quantification of void volume and position, faster inspection speed, reduced reliance on human expertise, and avoids radiation risks associated with X-ray operation.
How does WeLinkirt's 3D AI AOI quickly adapt to new products or defect types?
WeLinkirt's DaoAI AI AOI software system integrates APDT few-shot learning technology, requiring only 1–20 good samples to quickly train and deploy models for new products or defect types, significantly reducing changeover time. Furthermore, its visual foundation model possesses strong generalization capabilities, allowing it to rapidly adapt to production line changes for efficient and flexible manufacturing.