Solder joint reliability has been a key issue for electronic assemblies and microelectronic packaging for many years. Many different factors can affect the solder joint reliability,such as package construction,mechanical properties of solder joints,external mechanical and thermal stresses,and environmental change. In applications of portable electrical devices,the mechanical impact such as shock,bending and twisting plays an important role in the product’s reliability. When an electronic assembly experiences repeatable mechanical stresses,failure of the solder joint may be induced. In this study,we carried out a series of experiments to understand the failure mode of Sn-Cu solder joints under mechanical bending at the product level. By comparing two different designs with button/switches,the bending induced strain was found to be the key factor that makes the solder joint fail. Low strain-rate cyclic button push tests were performed with PCB board strain measured simultaneously by strain gage. A number of analytical methods were used to study the failure modes under different button push conditions,such as dye and pry,cross section,and X-ray. The black pad defect was observed with the Electroless Nickel Immersion Gold (ENIG) surface finish between the component and board pad,which contributed to the brittle fracture in the intermetallic region. Ductile fractures were observed inside the bulk solder,which was attributed to the high strain at low strain-rate cyclic loading during operations. To eliminate the solder joint crack and improve the interconnection reliability,alternative surface finishes and methods to reduce the board strain were suggested.

Storage time allowed between bare printed circuit board manufacturing and assembly is quite limited. Based on an interpretation of IPC standards,shelf-life of the surface finish is the limiting factor. A lot of boards are scrapped worldwide because they have exceeded their shelf-life. Size of manufacturing lots is limited due to the risk of overpassing shelf-life of stored PCB1. A three-year study involving PCB manufacturers,EMS2and OEMs3has been conducted to evaluate the possibility to extend the shelf-life. Details of the study will be described in this paper. More than five thousand samples with different finishes coming from different PCB suppliers have been aged in normal and accelerated conditions with or without additional packaging (none,anticorrosion paper,dry pack). Following ageing,solder ability has been evaluated by two methods. Solder spread test have been realized by an EMS involved in automotive market on its production line. Wetting balance test have been realized by an independent laboratory. Results summarized in this paper clearly indicate that actual limitations are very conservative,and that shelf-life in normal ambient conditions can beat least doubled,whatever the packaging conditions. New recommendations for storage conditions and limitation will be issued from this study. They will contribute to significant improvement of the worldwide electronic industry efficiency and costs.

Effective Roughness Dielectric (ERD) is a homogeneous lossy dielectric layer of certain thickness with effective (averaged) dielectric parameters. The ERD layer is used to model copper foil roughness in printed circuit board (PCB) interconnects by being placed on a smooth conductor surface to substitute an inhomogeneous transition layer between a conductor and laminate substrate dielectric. This work derives the ERD parameters based on the understanding that there is a gradual variation of concentration of metallic inclusions in the transition layer between the dielectric and foil. The gradual variation can be structured as thin layers that are obtained using the equivalent capacitance approach. The concentration profile is extracted from scanning electron microscopy (SEM) or high-resolution optical microscopy. As the concentration of metallic particles increases along the axis normal to the laminate dielectric and foil boundary,two regions can be discerned: an insulating (pre-percolation) region and a conducting (percolation)region. The rates of increase in effective loss (or corresponding conductivity) in these two regions differ significantly. The proposed model of equivalent capacitance with gradient dielectric is applied to STD,VLP,and HVLP foils. The frequency-dependent dielectric parameters of the homogenized ERD are calculated from the equivalent capacitance. The results are validated using 3D numerical electromagnetic simulations. There are two types of numerical models: with homogeneous ERD parameters,as well as layered. Both models show excellent agreement with measurements.

Imbalanced weakly and strongly edge-coupled differential pairs on printed circuit boards (PCBs),both microstrip (MS) and stripline (SL),are studied under different conditions using mixed-mode S-parameters. The r ate of coupling between the lines influences both signal integrity (SI) and electromagnetic compatibility (EMC) of the PCB design. Weakly coupled lines are preferable for SI,but this is not always the case for EMI. Common-mode and mode conversion that negatively affect EMC are typically higher in the weakly coupled cases than in the corresponding strongly-coupled. This is due to technological factors such as the difference in lengths of lines in a differential pair; trapezoid cross-section of signal traces; copper foil roughness; solder mask over microstrip lines; and presence of an epoxy-resin pocket between the stripline traces. In this work,results of 3D full-wave numerical electromagnetic modeling,taking into account these various technological features,are compared with the measured results on the designed test fixtures.

It is common for PCBs used for high frequency RF or high speed digital applications,to be tested for an impedance value prior to shipping the board. The controlled impedance board is typically specified to a nominal impedance value with a given tolerance. Some time ago a tolerance of ± 10% was considered acceptable however recently and with more demanding applications,the impedance tolerance is often much narrower. Understanding the many aspects of impedance is beneficial to the PCB fabricator and designer,who are often required to make a judgment for product shipment based on the impedance measurement. There are several different types of impedance however characteristic impedance is normally specified for a controlled impedance PCB. This circuit property has many variables and some variables are related to the PCB manufacturing process,some are associated with material properties and some variables are due to measurement techniques. Additionally,some of these variables are more or less dominate for a particular type of circuit design and / or construction. This paper will give an overview of the basic theory for impedance,with an emphasis on characteristic impedance for circuits of different design types. Microstrip,grounded coplanar waveguide (GCPW) and stripline structures will be discussed with their unique impedance attributes. PCB fabrication related variables,as well as material variables,will be illustrated using modeling software and verified with measured results. The variables associated with impedance measurement are many and details will be given for several related issues. One variable,often not recognized,is masking and that is how the impedance value of a circuit can be altered due to an impedance spike which is located prior to the body of the circuit. Masking can cause inaccuracies for impedance measurements and there are ways to minimize this concern,which will be illustrated. The impact on impedance resolution and accuracy due to rise time will also be demonstrated with measured examples.

A novel lead-free solder alloy 90.6Sn3.2Ag0.7Cu5.5Sb (SACSb),was developed targeted for high reliability with a wide service temperature capability. The alloy exhibited a melting temperature range of 223 to 232?C,reflowable at profile with peak temperature 245?C and 255?C,with ambient temperature Yield Stress 60MPa,UTS 77 MPa,and ductility 28%,and a higher stress than both SAC305 and 90.9Sn3.8Ag0.7Cu3Bi1.45Sb0.15Ni (SACBSbN),the latter two alloys were used as controls. When tested at 140?C and 165?C,the die shear stress of SACSb was comparable with SACBSbN but higher than SAC305,and the ductility was higher than both SACBSbN and SAC305,with SACBSbN exhibiting distinct brittle behavior. When aged at 125?C and 175?C,the die shear strength of SACSb was comparable or higher than both controls. When pretreated with a harsh condition,a temperature-shock test (-55?C/155?C) for 3000 cycles,the die shear strength of SACSb was 8 times of that of SACBSbN and SAC305. When pre-conditioned using a temperature-cycling test (-40?C/175?C) for 3000 cycles,the die shear strength of SACSb was 11 to 20 times higher than that of SACBSbN and SAC305,depending on the flux type used. Both SACSb and SACBSbN are alloys based on SnAgCu,but reinforced with precipitate hardening and solution hardening,with the use of additives including Sb,Ni,and Bi. SACSb exhibited a finer microstructure with less particles dispersed,while SACBSbN exhibited more particles with some blocky Ag3Sn plates or rods. SACSb is rigid and ductile,while SACBSbN is rigid but brittle. Under the harsh test condition where ?T was high,the dimension mismatch between parts and substrate became very significant due to CTE mismatch. This significant dimension mismatch would cause a brittle joint to crack quickly,as seen on SACBSbN. The challenge was more tolerable for a ductile joint,as shown by SACSb. Accordingly SACSb showed a much better reliability than SACBSbN under harsh conditions,including high testing temperature and large ?T. Overall,to achieve high reliability under a wide service temperature environment,a balanced ductility and rigidity for solder alloy is critical for success.

Electronics manufacturers are searching for new lead-free solders that can improve upon SAC305 voiding performance and that exceed the current thermal cycle performance of this solder in harsh environments,all the while being processed at or near current typical SAC305 peak temperatures. This paper compares -40ºC to 125ºC,thermal cycle test (TCT) results of boards built with SAC305 and a newSAC347+Bi/Sb/Ni/Co solder paste,which were assembled to intentionally contain three levels of voiding,ranging from 0%to 0.5%,5% to 20% and higher than 20%,in order to not only observe variation in TCT performance but find any correlation between voiding levels of up to 30% and the corresponding thermal cycle reliability of the solder joint.

Creep corrosion on printed circuit boards (PCBs) is the corrosion of copper metallization and the creeping of the copper corrosion products across the PCB surfaces to the extent that they may electrically short circuit neighboring features on the PCBs. The paper will report the results of a round robin evaluation of the modified iNEMI creep corrosion qualification test. The earlier versions of the test involved testing in a specially designed flower of sulfur (FoS) chamber for 10 days at nominally 81% humidity level provided by KCl saturated salt solution. The modification to the iNEMI test was the change to three 5-day test runs,the first 5 days at 31% relative humidity provided by MgCl2saturated salt solution,the 2nd5 days at 47% relative humidity provided by NH4NO3saturated salt solution and the 3rd5 days at 81% relative humidity provided by KCl saturated salt solution. The rest of the test procedure remained unchanged,with the iNEMI designed setup providing somewhat controlled and reproducible concentration of chlorine gas and a tray of sulfur for providing sulfur vapors. As usual the test was run at a constant 50oC. The paper will compare the round robin test results from three companies on immersion silver (ImAg),electroless nickel immersion gold (ENIG) and organic solderability preservative (OSP) finished PCBs soldered using organic acid (OR) and rosin (RO) fluxes. As expected,the ENIG finished PCBs suffered the most creep corrosion while the OSP finished PCBs suffered the least creep corrosion. The paper will also list the copper and silver corrosion rates experienced during the test runs and discuss means of better control of these corrosion rates.

The European Union RoHS directive took effect in 2006,and of the 6 restricted materials,the elimination of lead from electronic devices took the most development effort and had the worst degrading effect on hardware reliability. One negative impact was the brittleness of the lead-free solder alloys that replaced the industry favorite,ductile Sn-Pb eutectic alloy. Another was the unexpected occurrence of creep corrosion on printed circuit boards using alternative PCB surface finishes. Along with the implementation of RoHS,the miniaturization of circuits,the expansion of IT markets in developing countries with high-levels of sulfur-bearing gaseous pollution,and the trend towards energy saving by resorting to free-air cooling,have all led to increased rates of corrosion-related hardware failures associated with particulate and gaseous contamination. The IT industry has taken a two-pronged approach to mitigating these failures: (1) by making the products more robust against contamination and high humidity levels; and (2) by gaining better understanding of the allowable levels of contamination,temperature and humidity under which IT equipment can operate reliably. Additionally,many points along the supply chain have been identified where corrosion can form,and the additive effects may or may not be detected by testing or manifest themselves before delivery to the end-user. Failures at this point may be due to the cumulative effect of numerous “micro-failures” generated throughout the supply chain. However,what remains most frequent are product failures resulting from exposure to elevated pollutant levels and inadequate environmental controls at manufacturing locations. The result is an operating environment that does not meet current manufacturers’ warranty requirements – requirements that have been put into place since the implementation of RoHS. This paper will describe the common modes of corrosion-related hardware failures in the past 10 years,the actions taken to make the products more robust,the understanding of the role played by contamination,and the means of negating their detrimental effects. The case will also be presented for environmental monitoring at various points along the supply chain and the addition of enhanced air cleaning for those locations that do not meet the air quality requirements of the finished devices. Data will be presented that highlight the need for air quality assessments of manufacturing facilities,where enhanced air cleaning is indicated,and the benefit of establishing an ongoing real-time air monitoring program to assure compliance with air quality specifications and warranty requirements.

Over the past few years,studies have shown that Nano-coatings can improve solder paste release and reduce underside cleaning in the print process. Many of these studies have focused on print volume and the improvement of transfer efficiency in small component printing. This paper investigates whether Nano-coated stencils improve repeatability and uniformity in the print process for a range of components sizes. Repeatability and uniformity were defined as how tightly controlled print deposits were from print to print over time. Solder paste inspection (SPI) data was collected and analyzed for the following component types: 01005 up to 1206 Imperial chip components; 0.5mm pitch micro BGA components; 0.4mm pitch up to 1.25mm pitch QFP components and 0.4mm pitch up to 0.6mm pitch QFN components.