The Elegance of Line Scan Technology for AOI

Similar documents
Mirtec MV-3 Series Desktop AOI Systems Now with 5 and 10 MP camera!

Eight Tips for Optimal Machine Vision Lighting

True 2 ½ D Solder Paste Inspection

because inspection matters

Figure 1 HDR image fusion example

because inspection matters

XM: The AOI camera technology of the future

INTRODUCTION TO VISION SENSORS The Case for Automation with Machine Vision. AUTOMATION a division of HTE Technologies

BF-X2. In-line 3D automated X-ray inspection system for Semiconductor, Power module inspection

because inspection matters

An Introduction to Automatic Optical Inspection (AOI)

Machine Vision for the Life Sciences

Practical Image and Video Processing Using MATLAB

Process Control & Inspection, Assembly Programming, Reverse Engineering & Design. Products for the Electronics Industry

FSI Machine Vision Training Programs

Optimizing throughput with Machine Vision Lighting. Whitepaper

Optical Inspection Systems

Applying Automated Optical Inspection Ben Dawson, DALSA Coreco Inc., ipd Group (987)

Parallel Mode Confocal System for Wafer Bump Inspection

WHITE PAPER. Methods for Measuring Flat Panel Display Defects and Mura as Correlated to Human Visual Perception

Solve it with the problem solvers.

TR7500 SIII 3D SERIES. AutomAted optical InsPeCtIon

CircumSpect TM 360 Degree Label Verification and Inspection Technology

ORIFICE MEASUREMENT VERISENS APPLICATION DESCRIPTION: REQUIREMENTS APPLICATION CONSIDERATIONS RESOLUTION/ MEASUREMENT ACCURACY. Vision Technologies

because inspection matters

Using Optics to Optimize Your Machine Vision Application

flexible lighting technology

Real-Time Face Detection and Tracking for High Resolution Smart Camera System

Versatile Camera Machine Vision Lab

Camera Overview. Digital Microscope Cameras for Material Science: Clear Images, Precise Analysis. Digital Cameras for Microscopy

Patents of eye tracking system- a survey

Records the location of the circuit board fiducials.

NELA Brüder Neumeister GmbH

The Fastest, Easiest, Most Accurate Way To Compare Parts To Their CAD Data

Systematic Workflow via Intuitive GUI. Easy operation accomplishes your goals faster than ever.

Camera Overview. Digital Microscope Cameras for Material Science: Clear Images, Precise Analysis. Digital Cameras for Microscopy

Imaging Optics Fundamentals

The History and Future of Measurement Technology in Sumitomo Electric

What is really inside your AOI?

FLASH LiDAR KEY BENEFITS

WHITE PAPER. Sensor Comparison: Are All IMXs Equal? Contents. 1. The sensors in the Pregius series

MML-High Resolution 5M Series

Introduction. Lighting

Camera Overview. Digital Microscope Cameras for Material Science: Clear Images, Precise Analysis. Digital Cameras for Microscopy

APPLICATIONS OF HIGH RESOLUTION MEASUREMENT

SMART LASER SENSORS SIMPLIFY TIRE AND RUBBER INSPECTION

Speed and Image Brightness uniformity of telecentric lenses

Low Cost Earth Sensor based on Oxygen Airglow

Smart vision and optical solutions for the food and beverage industry

Highly Versatile Laser System for the Production of Printed Circuit Boards

Bringing Answers to the Surface

VERSAPRINT 2 The next generation

TECHNICAL SUPPLEMENT. PlateScope. Measurement Method, Process and Integrity

Advanced Optical Line Scanners for Web Inspection in Vacuum Processes Tichawa Vision GmbH

Opto Engineering S.r.l.

Vision Lighting Seminar

On spatial resolution

ARTIFICIAL INTELLIGENCE - ROBOTICS

Application Note 5026

High Performance Imaging Using Large Camera Arrays

ROBOT VISION. Dr.M.Madhavi, MED, MVSREC

USING SIGNATURE IDENTIFICATION FOR RAPID AND EFFECTIVE X-RAY INSPECTION OF BALL GRID ARRAYS

Machine Vision Basics

DECISION NUMBER FOURTEEN TO THE TREATY ON OPEN SKIES

Exercise questions for Machine vision

Correlation of Wafer Backside Defects to Photolithography Hot Spots Using Advanced Macro Inspection

Make Machine Vision Lighting Work for You

Observational Astronomy

Our Vision... Tomorrow s Innovation

IMAGE FUSION. How to Best Utilize Dual Cameras for Enhanced Image Quality. Corephotonics White Paper

Improved sensitivity high-definition interline CCD using the KODAK TRUESENSE Color Filter Pattern

Prepared by Qian Ouyang. March 2, 2013

ECEN 4606, UNDERGRADUATE OPTICS LAB

SUPRA Optix 3D Optical Profiler

DIRECT PART MARKING THE NEXT GENERATION OF DIRECT PART MARKING (DPM)

Registration performance on EUV masks using high-resolution registration metrology

Chapters 1 & 2. Definitions and applications Conceptual basis of photogrammetric processing

How interference filters can outperform colored glass filters in automated vision applications

PRODUCT GUIDE Vision software from the world leader.

A machine vision system for scanner-based laser welding of polymers

Capabilities of Flip Chip Defects Inspection Method by Using Laser Techniques

APPLICATION OF COMPUTER VISION FOR DETERMINATION OF SYMMETRICAL OBJECT POSITION IN THREE DIMENSIONAL SPACE

UNIVERSITY OF WATERLOO Physics 360/460 Experiment #2 ATOMIC FORCE MICROSCOPY

Selective Soldering How it has evolved to become an Integral. Method in Todays Manufacturing Processes

High-Speed 3D Sensor with Micrometer Resolution Ready for the Production Floor

CyberOptics Award-Winning Systems Portfolio

FACE RECOGNITION BY PIXEL INTENSITY

SONAR THEORY AND APPLICATIONS

Comparators. Contour Projector Technologies. Compare the Quality. Precision for People

ECC419 IMAGE PROCESSING

Phase One 190MP Aerial System

Applied Machine Vision

A Handheld Image Analysis System for Portable and Objective Print Quality Analysis

LPR SETUP AND FIELD INSTALLATION GUIDE

ScrappiX. Visual inspection equipment for dimensional. and surface defects control

Revolutionizing 2D measurement. Maximizing longevity. Challenging expectations. R2100 Multi-Ray LED Scanner

APPLICATIONS FOR TELECENTRIC LIGHTING

hurryscan, hurryscan II

COLOR CONTRAST AND LUMINESCENCE SENSORS

Understanding Imaging System Specifications for Pixel-Level Measurement of Displays

Transcription:

By Mike Riddle, AOI Product Manager ASC International More is better? There seems to be a trend in the AOI market: more is better. On the surface this trend seems logical, because how can just one single downward- looking FOV (field-of-view) camera circled by a ring light detect production errors in solder quality, lifted leads head-in-pillow defects? And the answer seems simple enough: employ additional cameras, surrounding the downward-looking principle camera, and position them at oblique angles so that they can provide a side view of the region of interest, capturing images of defects that seem undetectable from overhead. But drilling deeper into this solution uncovers several questions: 1) There are some AOI systems on the market that tout using up to eight cameras. So, how many cameras are enough? 2) How does an AOI system process the images generated by these multiple cameras to produce useable inspection data? Typically, most AOI systems with side view cameras perform only image comparison on side view data. In other words, they use a golden image generated from one or multiple samples as a model and compare that golden image to the current area of interest. But what happens if the area of interest is good, but there is a difference in color or lighting, or board-to-board, lot-to-lot variations? Will the same golden image work on the same component package type placed anywhere on the PCB in any orientation? In other words, can one use the same component library no matter where the component is located, or what taller components surround it (shadowing issues)? Does a multiple camera system take more time to program and debug? Does increasing the number of cameras increase the number of false failures or false calls? 3) How sensitive is the calibration and alignment of the optical system on an AOI using multiple cameras? 4) How reliable is an AOI system using additional cameras? Does attaching more sensitive hardware make it more prone to mechanical and electrical problems in a production environment? 5) How does adding more cameras (and more images to process) affect the inspection speed? 6) What does the additional hardware - as well as the software to support it - add to the price and operating cost of the AOI system? Line Scan Technology may offer an elegant alternative to multiple camera AOI systems. Since the mid-1990s, SAKI Corporation has employed this reliable yet flexible machine vision concept in over 7,000 AOI installations worldwide, mostly in high-production environments across Asia. This paper will provide an explanation of Line Scan Technology and how it s being applied in AOI to solve difficult inspection problems such as soldering and lifted lead defects without the use of additional imaging hardware. 1

What is Line Scan Technology? SAKI s Line Scan Technology (LST) drives a high-resolution linear CCD array sensor over the PCB surface rapidly capturing the complete PCB image. Scanning a 460 x 550 mm (18 x 20 ) PCB takes 10 seconds. The scanning speed is unaffected by the number of components on the PCB, and smaller PCBs take less time to scan. The lighting system for SAKI s LST contains over 3000 LEDs, arranged in multiple banks, generating Toplight, Sidelight and Lowlight Illumination. During each scan, the LED lighting is modulated several thousand times to produce over twenty different lighting schemes, providing remarkable flexibility for finding just the right contrast for any inspection problem. FOV vs. LST The principle sensor used in conventional AOI system design whether or not side cameras are included is a downward-looking large area array CCD camera. Typically, this camera is surrounded by ring lighting projected at an oblique angle. In order to inspect a PCB, the camera is positioned over multiple areas of interest, gathering snapshots. Depending on the camera s Field of View (FOV), the size of the PCB and its component density, data acquisition can take anywhere from a few seconds (small, sparsely populated assemblies), to several minutes (large, densely populated assemblies). In contrast, LST scans at a constant speed, collecting the complete PCB image in one single, rapid motion. Co-axial Toplight is born Since different components have different heights and some are much taller than their neighbors SAKI realized that the best illumination for solder inspection was from a light source projected directly perpendicular to the PCB surface. Any other off-axis illumination would lend itself to creating shadows, as the heights of taller components blocked the light from reaching the smaller ones, which could increase the number false calls, creating more programming and de-bugging work. SAKI also discovered that good solder fillets reflected overhead lighting or Co-axial Toplight much differently than cold solder or no solder, and that the contrasting reflections could be used to clearly distinguish Good vs. No Good solder. Further refinement in SAKI s Coaxial Toplight concept introduced telecentric optics, increasing the sensor s Depth of Field, and removing parallax associated with other highly focused optical 2

designs. This virtually eliminated shadowing and optical distortion altogether, enabling the use of flexible component inspection libraries that could be deployed anywhere on the PCB surface, without regard to the heights of adjacent components a problem that can still plague a number of AOI system designs. Which lighting is best? While Co-axial Toplight has proven to be the best for solder fillet inspection, there are other inspection tasks that require different projection angles and illumination color to obtain the highest contrast between G and NG conditions. Tasks such as Optical Character Recognition (OCR) and Optical Character Verification (OCV) may require different projection angles as well as data from multiple projection angles to accurately determine that the correct part has been placed. In some cases, the color of the component is used to detect presence/absence, or the correct part. Color inspection usually requires another unique illumination setting. In some cases, the contrast of color and/or reflectivity between the component and the substrate is very subtle. This may require a special, user-defined lighting scenario to highlight the differences that will provide repeatable results. To enhance its Co-axial Toplight concept, SAKI added banks of LEDs to project off-axis Sidelight, which has been found best for color inspection and Lowlight, which uses a very acute projection angle, and is ideal for OCR and OCV inspection tasks. Further testing demonstrated that sometimes a combination such as Toplight-Sidelight or Sidelight-Lowlight produced superior results. Still, SAKI discovered other inspection tasks requiring even more lighting options. Rather than add more lighting hardware and creating more servicing issues, SAKI decided to solve the problem using software. Today, with each scan of the PCB, SAKI AOI systems generate over twenty different lighting schemes. Any one of those lighting schemes plus a user-defined option that allows the user to decide the best Toplight/Sidelight/Lowlight blend can be applied to any inspection window and algorithm used in any component library. This lighting flexibility unique to the industry employs a very simple electro-mechanical 3

design to accomplish what other systems strive for using complex hardware. Furthermore, producing these numerous lighting schemes generates no impact on scanning time. Why Not Inspect the Whole Board? In conventional AOI systems employing FOV - or FOV with side cameras the regions for inspection are only where the camera(s) are positioned on the X/Y plane of the PCB. Usually, these regions are confined to where the PCB has been populated, since taking multiple FOV snapshots to cover the whole PCB area increases tact time. Limiting inspection to only populated regions can mean that large areas of the PCB can pass through without scrutiny. Foreign materials such as loose chips, solder balls and other debris could go undetected, causing production problems downstream. 4 With LST, each inspection captures an image of the entire PCB in one rapid, continuous scan. To ensure no foreign material on the PCB passes to the next process, SAKI provides an Extra Component Detection (ECD) algorithm that compares a database comprised of ten known good sample images to the inspection image, looking for irregularities across the entire PCB surface. Built-in flexibility enables the user to select size boundaries so that only variations considered problematic are flagged, minimizing false calls. This unique feature enables SAKI AOI with LST to provide a thorough and comprehensive inspection of the entire board without affecting production beat rates. An Elegant Alternative to Side View Cameras As discussed above, mounting side view cameras on an otherwise downward looking large FOV camera seems like a logical solution to detecting defects like lifted leads, dryjoint solder, and head-in-pillow non-solder problems. But what if there were solutions to these issues that didn t require additional sensitive hardware and complex software, as well as considerably longer programming and de-bugging times? SAKI has come across its share of these defects and through many years of research and thorough testing the Company has implemented smart algorithms in its inspection libraries to accurately detect and report these production errors using the power of its LST and flexible lighting technology. Below are a few examples of how these algorithms work: Solder Quality With Co-axial Toplight, solder quality of components as small as 01005s can be inspected based on simple principles. When a fillet is correctly formed, light projected perpendicular to the PCB

will reflect off at an angle, generating a low brightness condition, and produce a dark area where the fillet is formed. On the other hand, when there is no fillet - as in cases where there is no component or insufficient solder light will reflect in an equal but opposite direction, as in a mirror, creating a high brightness. In SAKI s inspection libraries, high and low thresholds for brightness can be set for each inspection window. In addition, several windows can be used for one inspection task, and configured with conditional jumps. In this way, one component library, 0805smc in the above example, can be used for part numbers with the same component package size and type, even though the component color, PCB color, solder brightness or surrounding reflectivity levels may vary. This produces faster program and de-bug times, in addition to lower false call rates. Component Lead Inspection The necessity to detect lifted lead conditions is probably the driving force behind implementation of side view cameras in AOI. That being said, SAKI s LST can accurately detect lifted leads using a number of principles that work in conjunction without the additional hardware, software and operational expense of using side view cameras. Lead Length From a perpendicular, overhead view, a lifted lead will exhibit a different length than its neighbors. By making a comparison either using a measured length value as benchmark, or comparing the length of an individual lead to the length of its neighbors the length of a component lead is a symptom of its condition, either OK or NG. The length of the lead is measured from the body to the tip of the lead foot. As with other SAKI algorithms, acceptable thresholds are user-selectable, providing a robust inspection while reducing false calls. Lead Tip/Pad Solder Quality Another symptom of a lifted lead condition is the quality of the reflowed solder on the pad. As shown below and when using Toplight illumination, the reflective characteristics of the remaining pad of a good solder fillet and that of a pad with no lead attached exhibit very different brightness conditions. 5

Therefore, taking stock of the lead length, as well as the reflectivity of the solder fillet and reflectivity of pad end, a lot can be learned about the condition of a component lead solder fillet. Conclusion: The Elegance of Line Scan Technology In scientific circles, the word elegant is used to describe an idea or concept that combines simplicity, power and a certain ineffable grace of design. SAKI s Line Scan Technology is an example of an elegant yet comprehensive approach to AOI design, without relying on the use of multiple cameras and sensors that add significant hardware expense and increased programming time, while providing fewer opportunities for reliability issues. This paper has provided only a broad conceptual glimpse into the capabilities of LST, which in practice uses a combination of these as well as other smart algorithms to provide accurate, highspeed inspection capability. In today s rough and tumble manufacturing environments, production tools that are proven to be powerful yet simple are keys to efficiency and economy. ASC International is the North American Distributor for SAKI AOI/AXI. www.ascinternational.com About the Author: Since the mid-1980s, Mike Riddle has worked in electronics inspection and test, helping companies with leading edge technologies find their paths to market leadership. Customer-focused and growth-oriented, Mike has held various roles in business development, marketing, sales and international distribution. He is currently serving as AOI Product Manager for ASC International.You can learn more about Mike by visiting: http://www.linkedin.com/pub/mike-riddle/12/491/710 6