3D AI AOI Equipment Achieves Remarkable Results in BGA Void Detection

Currently, intelligent manufacturing has become an important development trend in the electronics/PCBA industry, putting forward higher requirements for the accuracy and efficiency of detection technology. WeLinkirt provides an effective way to solve the problem of BGA void detection with its innovative 3D AI AOI equipment.

Against the backdrop of the booming intelligent manufacturing, the electronics/PCBA industry is in a critical period of transformation and upgrading. The market has increasingly strict requirements for the quality and performance of electronic products, which makes the inspection process in the production process particularly important. Take a leading electronics manufacturer as an example. Its PCBA production line is mainly responsible for the production of printed circuit boards for various electronic products. In the BGA (Ball Grid Array) soldering process, the quality of BGA solder joints directly affects the performance and reliability of the circuit board. Therefore, accurate detection of void defects in BGA solder joints has become a key link to ensure product quality.

Pain Points: Why Is It Difficult?

Traditional 2D optical inspection methods face many difficulties when detecting BGA voids. From a quantitative perspective, the false-negative rate is relatively high, about 3%. This means that some defective products may flow into subsequent processes, increasing the product rework rate and production cost. For example, if a batch of 1000 circuit boards is produced, according to a 3% false-negative rate, 30 circuit boards with BGA void defects will enter the next process, and the subsequent rework will consume a lot of manpower and material resources.

The false-positive rate is also a major pain point of traditional inspection methods. The false-positive rate of traditional 2D optical inspection reaches 20%. A large number of false-positive information needs to be re-checked manually, which not only wastes a lot of manpower and time but also reduces production efficiency. Suppose a testing station needs to test 500 products per day. According to a 20% false-positive rate, 100 products need to be re-checked manually, which undoubtedly brings great work pressure to quality inspection personnel.

In addition, when changing product models, traditional inspection equipment also shows obvious deficiencies. The programming and debugging time is relatively long, about 30 minutes. In the era of pursuing high-efficiency production, this 30-minute time waste seriously affects the overall efficiency of the production line. The root cause of these problems is that the 2D optical inspection method can only obtain two-dimensional image information of the object surface and cannot accurately reflect the three-dimensional structure and depth information of the object. It is difficult to accurately identify the void defects hidden inside the solder joints.

Technical Principle

WeLinkirt's 3D AI AOI equipment uses a self-developed 3D camera for imaging, and its core principle is based on the structured light three-dimensional measurement technology. This technology projects a specific structured light pattern onto the surface of the detection object, and then uses the camera to capture the deformation of the reflected light. According to the triangulation principle, the equipment can calculate the three-dimensional coordinates of each point on the object surface, thereby realizing three-dimensional shape reconstruction and generating point cloud data.

Compared with traditional 2D optical inspection methods, 3D data can provide more abundant information. It can observe the solder joints from multiple angles, accurately judge the existence and size of voids, and effectively overcome the limitations of 2D optical inspection. At the same time, the equipment uses 2D-3D fusion technology, which combines the texture information of 2D images and the geometric information of 3D point clouds to further improve the accuracy of detection. At the algorithm level, the equipment uses deep learning algorithms to train a large amount of sample data, continuously optimizing the feature recognition ability of the model so that it can quickly and accurately identify defects such as BGA voids.

Typical Application Scenarios

• BGA Void Detection: This is one of the main application scenarios of the equipment. During the detection process, the equipment obtains information about the solder joints from multiple angles through 3D imaging and 2D-3D fusion technology, and accurately identifies the location and size of the voids. The difficulty lies in that the voids may be hidden inside the solder joints, which are difficult to find by 2D optical inspection. The 3D AI AOI equipment can effectively solve this problem with its advanced technology.
• Hidden Solder Joint Detection: Some solder joints may be blocked by other components, and traditional 2D detection methods cannot detect them. The 3D AI AOI equipment can observe the solder joints from different angles through three-dimensional shape reconstruction and find hidden solder joint defects. The difficulty lies in how to accurately identify the characteristics of the blocked solder joints. The equipment improves the recognition accuracy through deep learning algorithms by learning a large number of samples.
• Coplanarity Detection: The coplanarity of BGA solder joints has an important impact on the performance of the circuit board. The equipment calculates the height information of the solder joints through 3D point cloud data to determine whether they meet the coplanarity requirements. The difficulty lies in the need for high-precision measurement and calculation. The self-developed 3D camera and advanced algorithms of the equipment can ensure the accuracy of detection.
• Micron - Level Morphology Detection: Traditional inspection methods are difficult to detect some tiny morphological defects. The 3D AI AOI equipment can detect micron-level morphological changes and accurately identify tiny defects through the analysis of 3D point cloud data. The difficulty lies in the recognition and differentiation of tiny features. The equipment improves the detection ability of tiny defects through fine-grained label learning.
• Porosity Detection: Porosity in solder joints will affect the electrical performance of the circuit board. The equipment uses 3D imaging technology to observe the inside of the solder joints and detect the existence of pores. The difficulty lies in that the pores may be small and irregular. The equipment improves the detection accuracy by learning the characteristics of pores through deep learning algorithms.

Implementation Case

A large-scale electronics manufacturing enterprise is responsible for the production of a large number of printed circuit boards for electronic products on its PCBA production line. In order to improve product quality and production efficiency, the enterprise introduced WeLinkirt's 3D AI AOI equipment. During the implementation process, WeLinkirt's professional technical team installed and debugged the equipment and provided systematic training for the enterprise's operators. After a period of trial operation, the equipment was officially put into use.

After using WeLinkirt's 3D AI AOI equipment, the BGA void detection effect of the enterprise has been significantly improved, and the production efficiency has been greatly increased.

WeLinkirt's Solution and Products

Centered around the 3D AI AOI equipment, WeLinkirt provides a complete inspection solution for electronics manufacturers. The equipment has high-precision detection ability and can accurately detect BGA voids. Its self-developed 3D camera and three-dimensional shape reconstruction technology can effectively overcome the limitations of 2D optical inspection and detect hidden defects.

During the implementation process, the DaoAI AI AOI software system is used in conjunction. The system supports 5-minute zero-code automatic programming for one good product. Through APDT positive sample/few-sample learning (only 1-20 good products are required), it can quickly adapt to the inspection requirements of different products, greatly shortening the model-change time. At the same time, the semantic false-positive filtering function of the system can effectively reduce the false-positive rate.

Quantitative Results

After using WeLinkirt's 3D AI AOI equipment, the BGA void detection rate of the electronics manufacturer has increased to 99.2%, and the false-negative rate has decreased to <0.8%. The false-positive rate has been reduced by -65%, greatly reducing the workload of manual re-inspection. The product model-change time has been shortened from the original 30 minutes to 5 minutes, and the production efficiency has been significantly improved.

This article was generated by AI. Customer cases are simulated scenarios based on real product capabilities and figures are illustrative; see product pages for official benchmarks.