Automated machine vision techniques ensure an accurate and repeatable PCB inspection process and a smoother PCB manufacturing process.
The PCB manufacturing process is becoming more intricate, as technological innovations are driving the growth of compact electronic devices with maximum functionalities. Miniature discrete packages like 0201 and 01005 are increasing the level of complexity for the PCB inspection process, also. To achieve the best PCB quality, manufacturers are using a combination of inspection and testing technologies during PCB assembly.
Advanced technologies like automatic optical inspection (AOI), automated X-ray inspection (AXI), and the latest machine vision (MV) method are used to speed up the inspection process and identify the defects in the early stages of PCB manufacturing. Also, the inspection technologies can provide feedbacks to identify fault patterns in the assembly process. A timely correction included in the procedure can reduce the fault occurrence and can save additional debugging time.
Manual visual inspection (MVI) is the simple, low-cost inspection technique used as the first step in the PCB inspection process. A trained staff manually goes through a checklist of items and verifies each assembled PCB to ensure the board thickness and dimensions, any physical damages, missing components, etc. This inspection method is suitable for prototypes and simple circuits. Soldering issues beneath surface mount devices (SMD) will not be captured in the MVI method.
The solder paste inspection (SPI) method uses a laser scanner to capture the image of the applied solder paste. CAD files are used as a reference and high-resolution image processing is done to inspect the shape, quantity, and coverage of the solder paste dispensed on the SMD pads. The automated SPI system can also indicate if the solder application machine needs cleaning or maintenance.
In the AOI method, an error-free PCB (golden PCB) image is captured with sufficient lighting using a still or video camera. A complete picture of the golden PCB is stored as a standard reference. Further, all other PCBs are inspected against the reference image for any missing or misaligned components, solder joint errors like opens, bridges, etc. AOI is also used for bare board inspection to check surface feature issues like scratches, nodules, and stains. 3D AOI machines can capture the component’s height accurately. But AOI is limited to line of sight and less effective on densely populated boards.
Automated laser test (ALT) measurement uses a laser beam to inspect the PCB and take measurements like solder joint shapes and solder paste depositions. The measured value is verified against the standard specifications to detect any faults on the board.
In the visual inspection techniques discussed so far, we have seen that the solder joints below SMDs are not visible easily. With the increased usage of IC packages like Ball grid arrays (BGAs) in the circuit design, the requirement of an inspection technique like AXI has become mandatory. The X-ray inspection method is best suited for complex designs that require a detailed inspection of the solder joints under the SMD leads. Solder paste absorbs more x-ray due to its heavy elements like lead and tin. This distinguishes the solder paste from other components during the inspection process, highlighting any possible bubble inside the solder joint. 2D X-ray inspection is used to inspect components for alignment and solder joint defects. Based on the complexity and quality requirements, 3D X-ray inspection methods can be used during the PCB assembly.
Machine vision inspection is one of the emerging inspection technologies that has achieved greater efficiency in fault detection. A sequence of images captured from multiple high-resolution cameras is used to extract the required data and compared with a standard reference image. It can be used before PCB assembly to verify correct component placement and optimal solder paste application on the component pads. Also, machine vision technology is used to inspect assembled boards to be error-free and as per the quality standards.
Machine vision checks for any of the solder paste issues leading to faults like bridging, peaking, etc. In component placement, machine vision is used to inspect lead pitch and alignment. Broken, bent, or missing pins of ICs are also identified in this inspection method. Thus, machine vision inspection covers all three stages of PCB assembly including solder paste inspection, component placement, and post-reflow inspection.
The PCB etching alone costs around 70% of the total PCB fabrication process. Any defect in this stage has to be spotted to reduce the count of faulty bare boards. The commonly found faults on bare PCBs are classified as fatal and potential faults. The short in the tracks or an opened conductor line is considered fatal, as it may lead to the rejection of the PCB itself. The less severe defects like pinhole or breakout are classified as potential defects that compromise the PCB functionality. Identifying these defects before the etching process is possible by using algorithms and machine vision equipment.
Image subtraction is one of the latest PCB inspection methods used to detect faults and improve PCB quality. The process of image subtraction includes a selection of the reference image and the inspection image. The subtraction method compares both the images pixel-by-pixel using XOR logic. Any defects present in the inspection image are captured during the process as a positive or negative image. In binary logic, a white pixel is represented as 1 and a black pixel is denoted as 0. The XOR logic works as follows:
White pixel (binary 1) – Black pixel (binary 0) = positive pixel image
Black pixel (binary 0) – White pixel (binary 1) = negative pixel image
The binary output image is used for fault detection and measurement. The method offers the advantage of verifying overall defects in the board but may not capture problems like color variation, reflectivity variation, etc.
Assembled PCBs undergo a combination of inspection techniques followed by testing procedures like In-circuit tests (ICT) and functional tests. After successful completion of all the inspection and test procedures, the assembled PCBs are cleared for shipment. Usually, the contract manufacturers offer these additional facilities of PCB inspection and testing along with the PCB assembly. It is always recommended to outsource these procedures to an experienced CM. The automated machine vision techniques ensure an accurate and repeatable PCB inspection process and a smoother PCB manufacturing altogether.
About the Author
Ken G is a production manager at Technotronix Inc.