Search results

A no‐clean mass reflow process for 396‐pin, 324‐pin and 225‐pin over moulded plastic pad array carriers (OMPACs) or plastic ball grid array (BGA) is presented. Emphasis is placed on the OMPAC assembly parameters such as the design, material and process of the packages and printed circuit board (PCB), solder paste, stencil design, printing technology, pick and place, mass re‐flow and inspection. Furthermore, cross‐sections and the ‘popcorn’ effect of the OMPAC assembly are provided and discussed.

The purpose of this paper is to develop a cost effective ball grid array (BGA) workcell for solder ball attachment.

This paper presents the construction of a low‐cost high‐efficiency automatic ball attachment workcell. In fact, it is an economical means of simultaneous placement of all solder balls on BGA substrates containing multiple BGA units as well as singulated substrates. Common industry problems such as the effect of static charges, the solder ball oxidation, the missing ball, the extra ball, the ball alignment, the deformed ball and, etc. will be addressed and critical issues affecting yield will also be discussed in this paper.

BGA is a popular integrated circuit packaging that is often applied in laptop computers and other handheld electronic devices for the provision of a high‐connection count in a relatively small area. However, the cost of a market available BGA solder ball attachment workcell is very expensive and the flexibility in fitting various customized process is usually low.

The developed workcell cost is about half of the market available machines with similar specifications; the yield achieved is within three sigma confidence interval with competitive output rate. The maintenance and troubleshooting are easy since the machine was developed by the in‐house engineering team.

In this paper, an optical inspection method of the ball grid array package(BGA) is proposed using a binocular machine vision system.

The height of each solder ball is calculated based on spatial geometrical size and location obtained from the two CCD cameras capturing range images of a LED illuminated BGA chip at certain orientation.

The structure of this system is simple and the accuracy is 0.02 mm, The experimental results have proved the validity of this system for BGA failure detection.

The developed machine vision system can provide some of the critical factors for BGA quality evaluation, such as the height of solder ball, diameter, pitch and coplanarity.

Compared with other systems, the structure of this system is simple and accurate, which meets the demand of off‐line and on‐line inspection. The limitation of this system is that the margin of field of view (FOV) is fuzzy. Further study could be focused on this problem.

A ball grid array (BGA) is a surface‐mount device and is processed in a standard surface‐mount (SM) assembly line. However, there are some special features which make it different from other assemblies. The peculiarities of PCB layout, screen printing, placement, soldering, and inspection in BGA processing are presented. The failures of BGA assembly during ramp up and series production start‐up are analysed in detail. An assembly quality better than 1dpm was achieved. The scale of repair is determined by the electrical quality of the devices. If small packages with higher pincount and better electrical performance are required, the µBGA is a sound choice. Compared with other high pincount packages with a small formfactor as TAB, flip‐chip and chip on board, the µBGA may be processed in a standard SM assembly line. The results of µBGA‐assembly feasibility studies are discussed.

This paper aims to present experimental and finite volume method (FVM)-based simulation studies on the scaling effect on the capillary contact angle and entrant pressure for a three-dimensional encapsulation process of ball-grid array (BGA).

With the development of various sizes of BGA packages, the scaling effect of BGA model on capillary underfill (CUF) process is investigated together with the influences of different industrial standard solder bump arrangements and dispensing methods used as case study.

The experimental results agree well to the simulation findings with minimal deviation in filling time and similar flow front profiles for all setups. The results revealed that the capillary contact angle of flow front decreases in scale-up model with larger gap height observed and lengthens the encapsulation process. Statistical correlation studies are conducted and accurate regression equations are obtained to relate the gap height to the completion filling time and contact angle. CUF threshold capillary pressures were computed based on Leverett-J function and found to be increasing with the scale size of the package.

These statistical data provide accurate insights into the impact of BGA’s scale sizes to the CUF process that will be benefiting the future design of BGA package. This study provided electronic designers with profound understanding on the scaling effect in CUF process of BGA, which may be extended to the future development of miniature-sized BGA and multi-stack device.

This study relates the flow behaviour of encapsulant to its capillary contact angle and Leverett-J pressure threshold, in the CUF process of different BGA and dispensing conditions. To date, no research has been found to predict the threshold pressure on the gap between the chip and substrate.

The purpose of this paper is to investigate effects of the multiple rework of ball grid array (BGA) components on mechanical strength of BGA balls, as well as any possible intermetallic (IMC) embrittlement, and obtain data correlated with possible estimation on the maximum permitted limits of BGA rework.

In this paper, mechanical strength of BGA components assemblies with multiple numbers of rework operations was evaluated. Mechanical evaluation was conducted using BGA ball shear tests and four‐point bending tests of BGA assemblies. Test samples were prepared under the following conditions: virgin, one, two, three and five BGA reworks. Failure mechanism was evaluated using cross‐section and SEM analysis.

The results show that both ball shearing tests and four‐point bending tests indicates that strength of BGA solder ball itself was not reduced significantly after repair/rework operation from one to five cycles. The IMC structure layer after rework is a mixture of IMC, Sn‐rich and Pb‐rich phases. This mixture layers with thickness even more than 10 μm in thickness does not show reduction of strength of BGA solder balls and do not cause premature embrittlement. However, the bonding strength of the copper pads to the laminates is reduced with rework/repair operation, with the great reduction coming from the first and second rework operation.

In general, the industry recommends two rework cycles for BGA components on the same spot. This study indicates that further rework (up to five) causes little degradation, therefore there is room to increase the total rework cycle limit beyond recommended two for plastic BGA components.

Presented test results shows that in most cases industry overestimates risks associated with increased embritlement of the BGA solder joints due to the intermetallics growth after multiple BGA rework operations. Strength reduction of BGA assemblies is mostly associated with reduction of bonding strength of the copper pads to the laminates is reduced with rework/repair operation and number of reworks could be increased in most cases.

Plastic ball grid arrays (PBGAs) have begun to gain popularity in the electronic packaging market because of their advantages over peripheral leaded devices. In addition to good electrical performance, the PBGAs offer advantages to the assembler. The development of a successful assembly process for PBGAs requires an understanding of the PBGA itself and the assembly process. The key attributes required for high yield manufacturing of PBGAs are discussed. The emphasis is placed on the PBGA assembly parameters including proper PBGA handling, printing , stencil design, placement, reflow and inspection. This paper is intended to increase the understanding of the plastic ball grid array manufacturing process.

The purpose of this paper is to develop a thermal coupling method of a ball grid array (BGA) assembly during a forced convection reflow soldering process.

The reflow oven was modeled in computational fluid dynamic (CFD) software (FLUENT 6.3.26) while the structural heating BGA package simulation was done using finite element method (FEM) software (ABAQUS 6.9). Both software applications were coupled bi‐directionally using the code coupling software MpCCI.

The convective heat transfer coefficient (h) simulated during the reflow process showed a sufficient view of the changing h in the BGA assembly of each reflow oven. The solder joints were found to experience phase change from solid to liquid during heating and liquid to solid during cooling. These phase changes were present at the melting temperature of the solder joint. The effect of the phase transition point was to cause a large range of temperature difference within the BGA assembly. This situation runs the risk of a skewing defect of components. The simulation results were compared with the experimental results and found to be in good conformity. In addition, the maximum thermal stress from simulation results was trapped in the interfaces between the solder joints and substrate, which tended to form the nucleation of initial crack.

The current study provides a methodology for designing a thermal profile for reflow soldering production.

The findings provide new guidelines for the thermal coupling method. This guideline is very useful for the accurate control of temperature distributions within components and printed circuit boards, which is one of major requirements for achieving high reliability in electronic assemblies.

Because of growing demand for slim, thin and cheap handheld devices, reduced-volume solder interconnects like land grid array (LGA) are becoming attractive and popular choices over the traditional ball grid array (BGA) packages. This study aims to investigate the mechanical shock and impact reliability of various solder alloys and BGA/LGA interconnect configurations.

Therefore, this paper uses drop testing experiments and numerical finite element simulations to evaluate and compare the reliability performance of both LGA and BGA components when exposed to drop and impact loadings. Additionally, three common solder alloys, including 63Sn37Pb, SAC305 and Innolot, are discussed.

The results of this study showed that electronic packages’ drop and impact reliability is strongly driven by the solder configuration and the alloy type. Particularly, the combination of stiff solder alloy and shorter joint, LGA’s assembled with SAC305, results in highly improved drop reliability. Moreover, the BGA packages’ performance can be considerably enhanced by using ductile and compliant solder alloys, that is, 63Sn37Pb. Finally, this paper discussed the failure mode of the various solder configurations and used simulation results to explain the crack and failure situations.

In literature, there is a lack of published work on the drop and impact reliability evaluation and comparison of LGA and BGA solders. This paper provides quantitative analysis on the reliability of lead-based and lead-free solders when assembled with LGA and BGA interconnects.

Reflow soldering is one of the most significant factors in determining solder joint defect rate. This study aims to introduce an innovative approach for optimizing the multiple performances of the reflow soldering process.

This study aims to minimize the solder joint defect rate of a ball grid array (BGA) package by using the grey‐based Taguchi method. The entropy measurement method was employed together with the grey‐based Taguchi method to compute for the weights of each quality characteristic. The Taguchi L18 orthogonal array was performed, and the optimal parameter settings were determined. Various factors, such as slope, temperature, and reflow profile time, as well as two extreme noise factors, were considered. The thermal stress, peak temperature, reflow time, board‐ and package‐level temperature uniformity were selected as the quality characteristics. These quality characteristics were determined using the numerical method. The numerical method comprises the internal computational flow that models the reflow oven coupled with the structural heating and cooling models of the BGA assembly. The Multi‐physics Code Coupling Interface was used as the coupling software.

The analysis of variance results reveals that the cooling slope was the most influential factor among the multiple quality characteristics, followed by the soaking temperature and the peak temperature. Experimental confirmation test results show that the performance characteristics improved significantly during the reflow soldering process.

The proposed approach greatly reduces solder joint defects and enhances solutions to lead‐free reliability issues in the electronics manufacturing industry.

The findings provide new guidelines to the optimization method which are very useful for the accurate control of the solder joint defect rate within components and printed circuit board (PCB) which is one of the major requirements to achieve high reliability of electronic assemblies.