Search results

Statistical modeling has been successfully applied to integrated circuit (IC) solder joint inspection. However, there are some inherent problems in previous statistical modeling methods. This paper aims to propose an adaptive statistical modeling method to further improve the inspection performance for IC solder joints.

First, different pixels in the IC solder joint image were modeled by different templates, each of which was composed of the hue value of the pixel and a proposed template significance factor. Then, the potential defect image was obtained by adaptive template matching and the potential defect threshold for each pixel. It was noted that the number of templates, matching distance threshold, potential defect threshold and updating rate were adaptively updated during model training. Finally, the trained statistical model was used to inspect the IC solder joints by means of defect degree.

Experimental results indicated that the proposed adaptive schemes greatly contributed to the inspection performance of statistical modeling. Also, the proposed inspection method achieved better performance compared with some state-of-the-art inspection methods.

The proposed method offers a promising approach for IC solder joint inspection, which establishes different numbers of templates constructed by pixel values and template significance factors for different pixels. Also, some important parameters were adaptively updated with the updating of the model, which contributed to the inspection performance of the model.

Automatic optical inspection (AOI) systems have been widely used in many fields to evaluate the qualities of products at the end of the production line. The purpose of this paper is to propose a local-to-global ensemble learning method for the AOI system based on to inspect integrated circuit (IC) solder joints defects.

In the proposed method, the locally statistically modeling stage and the globally ensemble learning stage are involved to tackle the inspection problem. At the former stage, the improved visual background extraction–based algorithm is used for locally statistically modeling to grasp tiny appearance differences between the IC solder joints to achieve potential defect images for the subsequent stage. At the latter stage, mean unqualified probability is introduced based on a novel ensemble learning, in which an adaptive weighted strategy is proposed for revealing different contributions of the base classifier to the inspection performance.

Experimental results demonstrate that the proposed method achieves better inspection performance with an acceptable inspection time compared with some state-of-the-art methods.

The approach is a promising method for IC solder joint inspection, which can simultaneously grasp the local characteristics of IC solder joints and reveal inherently global relationships between IC solder joints.

This paper aims to inspect solder joint defects of integrated circuit (IC) components on printed circuit boards. Here, an IC solder joint inspection algorithm is developed based on a Gaussian mixture model (GMM).

First, the authors train a GMM using numerous qualified IC solder joints. Then, the authors compare the IC solder joint images with the trained model to inspect the potential defects. Finally, the authors introduce a frequency map and define a metric termed as normalized defect degree to evaluate qualities of the tested IC solder joints.

Experimental results indicate that the proposed method is superior to the state-of-the-art methods on IC solder joint inspection.

The approach is a promising method for IC solder joint inspection, which is quite different from the traditional classifier-based methods.

Solder joint failures in military missile systems and commercial communication products were investigated. The study was initiated after it had been determined that visual PWBA solder defect inspection for external surface attributes of solder joints such as brightness, smoothness, evenness, solder quantity and PTH plugs and other attributes (including those of present IPC and DoD‐STD‐2000 criteria) was neither effective nor efficient. It was determined that visual inspection is subjective and varies to a large degree between inspectors.

To accommodate increasing levels of device integration at the chip level, circuit line densities in electronic packages are continually increasing. Greater circuit line density, in turn, necessitates a corresponding increase in package‐to‐board interconnection density, with I/O counts expected to reach over 600 by 1995. In conjunction with the upward trend in I/O counts are a complementary upward trend in clock speed and an opposing downward trend in package sizes driven by the need to provide more functionality in less space, particularly in notebooks and PCMCIA cards. To satisfy the requirements of increased I/O counts and clock speed, and reduced package sizes, various package‐to‐board interconnection technologies are being developed, such as flip chip attach (FCA) using C4 joints. However, FCA interconnections have a disadvantage of being very difficult, if not impossible, to visually inspect. Though automatic test equipment (ATE) can determine whether the package is functional, it cannot determine the quality and reliability of FCA interconnections. Of the possible inspection techniques available to assess the quality of FCA interconnections — differential laser thermal analysis, acoustic microscopy and cross‐sectional X‐ray radiography — only cross‐sectional X‐ray radiography is capable of accurate, automated inspection of production volumes. This paper will first examine the requirements for inspecting FCA joints and will then describe the various inspection alternatives, outlining their advantages and disadvantages. Having described the potential advantage of one particular cross‐sectional X‐ray technique, digital tomosynthesis, the paper will conclude with some cross‐sectional images of FCA and SMT joints taken by a digital tomosynthesis system being developed for the inspection of FCA joints.

Electronics manufacturing throughout the world now uses an increasing percentage of Surface Mount Technology (SMT). The compact and light‐weight surface‐mounted components offer a number of advantages to manufacturers. Unfortunately, however, these same beneficial characteristics make the quality of the product difficult to guarantee. As miniaturisation continues, the inspection problem becomes worse, and so advanced methods of inspection are required. Automatic inspection systems already exist, although an effective, inexpensive and reliable system has yet to be found. Recent work carried out within the Coherent and Electro‐Optics Research Group at Liverpool Polytechnic has looked at the feasibility of applying some of its established inspection methods to the problem of solder joint inspection. Extensive development must still take place; however, the methods employed have shown promise. The system uses structured light techniques to add height information to an image of the solder joint under inspection. In this way a 3‐D image of the joint may be built up, digitised and processed in a computer at high speed in order to determine its quality.

Solder quality is a critical issue in circuit board production. However, the manufacturing process of today has few built‐in controls to ensure good solder quality. Instead, attention is only paid when a defect is found. The solution lies in the integration of inspection, test and process control. Now a new method, Scanned‐Beam Laminography, takes 3‐dimensional X‐ray slices of solder joints, enabling thorough inspection of each joint. Special software algorithms measure and report the solder joint conditions, creating a system which provides real‐time process control.

The initial excitement over increased solder joint densities, higher manufacturing throughput, and superior electrical performance brought forth by surface mount technology (SMT) has been replaced by frustrations over lower yields and the inherent difficulties of inspecting hidden solder joints. In the plated through hole (PTH) process, rework and inspection tasks were not only relatively easier tasks, but also less costly. The high cost of inspecting and reworking SMT assemblies dictates a rethinking of the assembly process. Increasing first time yields becomes the key to reducing SMT inspection and rework costs. In a high volume facility, a 100% visual inspection process is not feasible because of the high cost of inspection and rework. However, if a company intends to remain competitive, inspection and rework must be reduced without a sacrifice to final product quality. Realising that it is not possible to ‘inspect’ quality into a product, improved yield must result from a controlled process environment. By maintaining a controlled environment, one will be provided with lower inspection costs, lower rework costs, lower scrap and, in the final analysis, improved product quality. At the heart of any process control environment should be a real‐time process control system designed specifically to accommodate SMT process defects. Process monitoring is accomplished by locating and identifying SMT process flaws. These flaws will then be reported to a host system for statistical analysis. These are statistical data used to make timely adjustments to the various stages of the assembly process in a real‐time manner. Being able to monitor the production process objectively in real time, and detect hidden flaws accurately, are the keys to having a successful process inspection system. Automated X‐ray Inspection is gaining acceptance as a viable process monitoring tool, capable of detecting and reporting SMT process flaws, including those hidden flaws not reported with typical visual inspection systems. The purpose of this paper is to show how an Automated X‐ray Inspection system can be integrated into the SMT production process as a cost‐effective method for improving SMT yield.

This paper aims to describe and document the application of commonly utilized solder joint failure analysis techniques to lead‐free solder joints.

Traditional failure analysis techniques, including visual inspection, X‐ray radiography, mechanical strength testing, dye and pry, metallography, microscopy and photomicrography, are reviewed. These techniques are demonstrated as applied to lead‐free and tin lead solder joints. Common failure modes observed in lead‐free and tin lead solder joints are described and compared.

It is shown that the traditional failure analysis techniques previously utilized for tin lead solder joints are widely applicable to the analysis of lead‐free solder joints. The changes required to effectively apply these techniques to the analysis of lead‐free solder joints are described.

This paper will be instrumental to the process, quality, reliability and failure analysis engineering disciplines in furthering understanding of the application of failure analysis techniques of both tin lead and lead‐free solder joints.

The purpose of this paper is to evaluate the application of an acoustic micro‐imaging (AMI) inspection technique in monitoring solder joints through lifetime performance and demonstrate the robustness of the monitoring through analysis of AMI data.

Accelerated thermal cycling (ATC) test data on a flip chip test board were collected through AMI imaging. Subsequently, informative features and parameters of solder joints in acoustic images were measured and analysed. Through analysing histogram distance, mean intensity and grey area of the solder joints in acoustic images, cracks between the solder bump and chip interface were tracked and monitored. The results are in accord with associated Finite Element (FE) prediction.

At defective bumps, the formation of a crack causes a larger acoustic impedance mismatch which provides a stronger ultrasound reflection. The intensity of solder joints in the acoustic image increase according to the level of damage during the ATC tests. By analysing the variation of intensity and area, solder joint fatigue failure was monitored. A failure distribution plot shows a normal distribution pattern, where corner joints have the lowest reliability and are more likely to fail first. A strong agreement between AMI monitoring test data and FE prediction was observed, demonstrating the feasibility of through lifetime monitoring of solder joints using AMI.

The paper indicates the feasibility of the novel application of AMI inspection to monitor solder joint through lifetime performance non‐destructively. Solder joints' real life conditions can be tracked by an AMI technique, hence solder joint fatigue failure cycles during the ATC tests can be monitored and analysed non‐destructively.