Along with advancements in the electronics industry, modern PCBs have become more and more compact. With reducing size comes increased complexity; BGA (Ball Grid Array), and WLCSP (Wafer Level Chip Scale Package) became more and more common in PCBs. Even relatively smaller boards have thousands of soldered joints because of complex design, and these are where most problems are found. Hence manual inspection of PCB is no longer as effective as it was before. More complexity of recent PCBA has made the job of PCB inspection more tiresome, making inspectors or technicians tired which led to missed defects in PCB. There are enhanced PCB training courses for people interested in learning more. In the modern era, where high quality and high quantity production has become a necessity, automated process of inspecting PCB is essential.
Automated optical inspection, or AOI, is an essential technique for eliminating potential errors in printed circuit board manufacturing. It is an automated process of detecting defects in PCB, which provides the PCB manufacturing and rework industry faster, cheaper, and more efficient production of PCBs. Since in the case of modern miniaturized PCB traditional method of detecting errors and defects on PCB is no longer effective or efficient, AOI is becoming essentially important for PCB manufacturing and reworking, Traditional methods of examining PCB such as visual inspection by technicians is prone to human error and time consuming. Hence introducing automation provides better results.
Faulty or damaged printed circuit boards from a manufacturer can cost organizations a significant amount of time and money. So, it is vital to find such issues in earlier stages of production. If resolution of such issues is not found then it can even be hazardous to an end user. In this blog, we investigate automated optical inspection, and how it aids to provide high-quality printed circuit boards.
What is Automated Optical Inspection (AOI)?
AOI is an automated visual inspection methodology used primarily to test for defects in printed circuit board assemblies (PCBAs). Automated Optical Inspection and manual inspection are fundamentally similar. Only difference is that the formal one is coordinated by computer, whereas the later one is by human. AOI autonomously conducts scans of PCBAs with cameras. With AOI, two types of failures can be identified: 1. catastrophic failures, such as missing components, and 2. quality failures, such as misshapen components that are skewed.
When should I use Automated Optical Inspection (AOI)?
Automated optical inspection systems can be placed into the production line just after the solder process. It is ideal to catch problems early in the production process since it costs more to fix the faults further along the production process. Moreover, process problems in the solder and assembly area can be seen early in the production process and resolution for such issues are used to send feedback to earlier stages. A rapid response can ensure that problems are recognised quickly and rectified before too many boards are built with the same problem.
What are the attributes of Automated Optical Inspection (AOI)
AOI method can identify a variety of surface feature defects (nodules, scratches, and stains) and dimensional defects (Open circuit, shorts, thinning of the solder) and incorrect, missing components. It includes performing visual scanning of the whole board. Various sources of light and high-definition cameras are used to assist the AOI system to paint pictures of the PCB. The AOI system uses the following techniques to identify defects inside PCBs.
- Template matching: In this process automated optical inspection system compares the captured image with the image gathered from known good PCBA.
- Pattern matching: With these techniques the AOI system stores information of both good and bad PCB assemblies to analyze patterns from previous defects. Afterwards it processes to match the obtained image to these calculated patterns.
- Statistical pattern matching: This approach is very similar to pattern matching, but the major distinct part of this process is the introduction of a statistically developed method of identifying problems from previously stored data such as number of distinct PCBs, and major types of failures.
Another key element of the AOI system is the image-capturing system. AOI consists of an image capture system that serves various applications and has different levels of complexity and cost. These image-capturing systems accommodate one or more cameras with the possibility of 3D image-capturing capabilities. Essentially there are two kinds of image-capturing system is used in practice:
- Streaming video system: In this system, the camera takes streaming video, from which complete frames are taken. Each frame afterward is converted into a still image to be used in signal processing. Although this approach is not accurate enough, it is processed with high speed.
- Still image camera system: These kinds of systems need to be placed close to the PCB under inspection due to the requirements of good lighting. In this method, the camera is moved by an AI program.
Lightening is a vital aspect of the AOI system. Accurate lighting source is significant to highlight various types of defects with ease. A variety of lightening featured in AOI systems are following:
- Fluorescent lighting: Since fluorescent light is effective for identifying defects on PCB, it is a favorable option as a source of lighting for AOI. However, the main problem with fluorescent lighting is that the lamps degrade with time. So the AOI system is subjected to constantly adjusting intensity of light.
- LED lighting: It is considered a modern source of lighting for AOI systems, as the level of intensity for LEDs can easily be controlled by changing current. It is also considered a more stable source of lighting. Only disadvantage of LED lighting is that it suffers from reduction of intensity with time.
- Infra-red or ultraviolet: When defects can not be visually captured, infrared or UV lights are used to examine hidden layers of PCBs.
Along with the type of lighting, position of lighting is also equally important. For different PCBA, different positions work best.
For any AOI system to work effectively, they need to be programmed correctly. There are various programming methods for AOI.
- Use of known good PCBA: In this method, the AOI system is provided with a known good PCB for examination. Once analyzing that PCBA, AOI systems learn and store relevant attributes such as shape of components, solder profile of each joint, connectivity along PCB. AOI system is delivered with more than one good PCBA to inspect and collect more data.
- Algorithm based programming: PCB data is provided to the system, and it then generates its own profile for the board. This scheme will also require real boards, but fewer are generally required.
- AI programming: In recent trends AOI systems powered by AI programs perform inspection with best accuracy. These powerful AI can be trained and perform inference to catch defects, flawed parts, and other abnormalities, making sure every product leaving a production line meets all quality criteria.