Save. Share. Connect.
Friday, July 1, 2016
VOLUME - NUMBER
PCB and Test
Test and Assembly
SMT and Assembly
Assembly and Production
PCB and Production
Assembly and Production
PCB and Assembly
Assembly and Packaging
PCB and Manufacturing
SMT and Production
Test and Measurement
Components and Distribution
Production and Packaging
HOME / CURRENT ISSUE
Test and Measurement
Add Message Board
Dual Technology Automated Optical Inspection
Sample image Nordson YESTECH FX-SL system with Fusion light.
By Giancarlo de la Garza and Andy Bonner, Nordson YESTECH
The single inspection approach does not provide all of the answers that a production engineer needs to keep the production line running at maximum yield. With components ranging from the tiny 01005 up to long SMT or THT connectors, the AOI programmer needs superb image resolution, a comprehensive "toolkit", to reliably carry out inspection, as well as very flexible software that allows for the most comprehensive programming with ease of use. Programmers must contend with identifying parts of different shape, size and color, with text of all varieties and printing techniques. Lighting, image size and algorithm selection must all be considered, along with the speed of inspection and ease of generating a program.
Dual technology inspections on an IC component.
AOI systems with single algorithm-based processing inspections are known to perform only when highly tuned. However, the systems that utilize dual technology, combining both algorithm and image-based processing, are not only easier to program, but offer an easier-to-use approach that is essential in today's changing manufacturing environments. Single technology systems have also demonstrated to be harder to program and have some overall performance shortcomings. Dual technology do much better in detection, programming speed and availability of data that can be obtained, without compromising reliability or the inspection and process control that is the AOI tool's purpose in life.
Two Different Approaches
Image-based AOI methods utilize a historical database of images, which can contain a series of both good and defective components. Part of the setup process is learning the variables within a group of images, and applying those learned variables against all future component inspections. By comparison, algorithm-based inspection utilizes contrast location mechanisms that look for variations in light and dark grayscale or color patterns to identify the component and the surrounding area.
Dual Inspection for color LED component.
Today's technology has helped manufacturers of AOI equipment offer systems with a dual technology approach, with flexible hardware and robust software inspection solutions, employing a combination of both methods in one system, and providing powerful and comprehensive inspection.
The combination of both methods of inspection offer significant benefits including:
Faster setup in high mix environments, while also providing comprehensive parameter control for high volume production environments.
Analysis of both color and grayscale image information.
Library-based software enables fast setup times and copy-exact program portability in multi-system installations.
Intuitive interface for operators, technicians and engineers.
Comprehensive collection of SPC, data, and image-based reports enable real-time visibility and optimization of the manufacturing process.
Consider a commonly used 56-pin quad flat pack (QFP). There are many potential areas for fault on such a device. A few of these may present incorrect position, incorrect orientation, lifted leads, lead bridge, missing solder, or even the wrong part altogether. A single technology approach to inspection, with only pattern match or rule-based inspection, would struggle to address all of these fault conditions. It is much more effective to use the appropriate type of inspection for the fault that you are trying to detect.
Angle Camera Inspection for LED Coplanarity.
Dual Technology systems also allow the programmer to choose a pattern matching or OCR algorithm to check for the correct part, and then pattern match or utilize a rule-based edge detection algorithm for part position — whichever is best for the combination of part and layout position for that component. This also includes the use of "on-demand" angle camera technology for hard-to-inspect components. The programmer can then select rule-based analysis for lead bridge detection, pattern and color or rule-based inspection for the solder and lifted lead situations. Systems with a single approach, rule or template matching, simply cannot provide the inspection capabilities and flexibility possible with a dual technology system with multiple inspection capabilities.
All of these rules and algorithms can only be applied if the image quality is sufficient to be able to apply them. As a simple rule of thumb, if an image cannot be seen clearly on the screen, it is very unlikely that any inspection method will be successful. When looking at an image captured by the system, operators must be able to clearly read both laser etch and printed part labels. If they are not seen clearly, then there is the likely risk of allowing wrong or incorrectly oriented parts escape detection during the inspection process. Further, if lead tips of a device or spaces between the leads are not clearly visible, it is again unlikely that lifted leads and bridges will be detected. This image quality and repeatability are vital to the inspection process.
One more technology advance using the dual technology inspection approach is the use of angled cameras in High Resolution Flexible systems. These systems optimize the field of view (FOV) of the camera, to the spread of the lighting array, such that there is a consistent light level across the image. The use of this technology for the "on-demand" multiple cameras allows for complete coverage of the most difficult-to-see defects.
Cameras and Lighting
More pixels do not necessarily mean that you have a bigger, faster or better system. The image capture mechanism has to be taken as a whole. The camera also has to be capable of resolving the image that it sees. The capability of a camera to resolve close-together objects, like component leads, will depend on the number of line pairs per millimeter that the imaging device can resolve. This, in turn, can also depend on the transfer ratio of the lens. It does not matter how many megapixels a camera has; if it does not also have the imaging capabilities required for accurate inspection, it simply won't provide the results that you need.
Table summarizing single vs. dual technology AOI systems capabilities.
Until recently, development of higher resolution automated optical inspection technology was hindered by the limited capabilities of computers and rudimentary cameras. Another factor was lighting that constrained the quality of images that cameras could collect and process in a timely manner. With the expanding capability of today's computers, cameras and LED lighting, cutting-edge AOI systems directly process the images provided by the higher-resolution cameras, and then perform a quick transfer to a high-power computer, allowing for ultra-high-speed inspection to match manufacturing lane bit rates.
Nordson YESTECH systems are a good example of these capabilities and come standard with high-speed, 5 megapixel cameras, providing excellent color and grayscale response. These cameras work in conjunction with a megapixel transfer lens, and line separation to meet the tightest inspection requirements. The resolutions obtained with this configuration range from 3 to 25µ pixel size. The resolution that is chosen depends on the minimum defect size detection that is needed.
Innovations in inspection techniques, lighting, multiple camera technology and processing, both at maximum speed, are still only part of the story. Data also needs to be entered into the system to generate the program, with results being clearly presented to the operator, providing a detailed statistics log, and then fed back to the line. Boards have to be handled and recorded — whether thousands or millions — and all seamlessly, day after day on the production line. Today's AOI systems must offer a software package that provides the comprehensive input, output and monitoring facilities required, and can be updated over the Internet, as new software solutions are released.
A new AOI development is the Nordson YESTECH FX-SL AOI system, which offers an innovative technology approach to high-speed inspection. It does this by keeping the camera and lighting array in constant motion to optimize the speed at which the head travels across the board. Images are captured at a very high frame rate and analyzed while the camera keeps moving. This technique provides up to 3X increase in speed over standard AOI systems.
There are many types of AOI systems available in today's market — from benchtop machines to fully inline systems — and both platforms benefit from the latest technology developments. In general terms, the use of a dual-technology inspection approach and the latest advances in computer power, color high-resolution cameras, and LED lighting, have simplified the process of adapting a fully complementary AOI system into the manufacturing line of these days. Ranging from high volume and low mix to high mix and low-to-medium volume applications, the dual technology systems have demonstrated their ability to fit entirely into today's manufacturing environment, providing the best of both inspection techniques.
Contact: Nordson YESTECH, 2762 Loker Avenue West, Carlsbad, CA 92010
760-918-8471 fax: 760-918-8472 Web:
© 2015 USTECH. All Rights Reserved. |
Contact Us: 610-783-6100 | firstname.lastname@example.org
powered by GIM