Bryan Rodgers

Determination of the Anand Viscoplasticity Model Constants for SnAgCu

The major focus of this work was the determination of the nine constants required for Anand’s viscoplastic constitutive model for a lead-free solder alloy, 95.5Sn3.8Ag0.7Cu and to compare them with those for SnPb. The test specimen was a cast dog bone shape based on ASTM E 8M-01, with a diameter of 4mm and a gauge length of 20mm. A series of tensile experiments were carried out: constant displacement tests ranging from 6.5 × 10−5 /s to 1.0 × 10−3 /s at temperatures of 20°C, 75°C, and 125°C; constant load tests at a range of loads from 10MPa to 65MPa, also at temperatures of 20°C, 75°C, and 125°C. A series of non-linear fitting processes was used to determine the model constants. Comparisons were then made with experimental measurements of the stress-plastic strain curves from constant displacement rate tests: it was found that the model matched the experimental data at low strain rates but did not capture the strain hardening effect, especially at high strain rates. A finite element model of the test was also constructed using ANSYS software. This software includes the Anand model as an option for its range of viscoplastic elements, requiring that the nine constants be input. In this case, an 8-noded axisymmetric element (VISCO108) was used to model the test specimen under constant displacement rate loading. The model was then used to predict the stress-plastic strain curve and this was compared to both the experimental measurements and the fitted Anand model. Reasonable agreement was found between the Anand model and the FE predictions at small strain rates. Finally, a BGA device was simulated under accelerated temperature cycling conditions using ANSYS with the fitted Anand for the SnAgCu solder joints. A Morrow-type fatigue life model was applied using empirical constants from two published sources and good agreement was found between experiment and predicted fatigue life.Copyright

Finite element modelling of a BGA package subjected to thermal and power cycling

The finite element techniques of substructuring and submodelling have been applied to a 9/spl times/9 ball grid array in order to estimate the fatigue life of the solder joints under various thermal loading conditions. Darveauxs method, which relates the accumulated viscoplastic strain energy density and crack growth data to fatigue life, has been used in all cases to predict the life of the solder joint. Three types of cycle were considered: (i) isothermal temperature cycling, (ii) isothermal temperature cycling with constant heat generation in the die, and (iii) power cycling (transient heat generation in the die). Results indicate that for the first two cases, the solder joint closest to the centre will fail first and that the superimposed constant heat generation in the die has little effect on fatigue life. In the case of power cycling, the outermost diagonal joint is predicted to fail first. The two finite element techniques examined are shown to produce similar results, however, substructuring is not applied to the power cycling case due to the transient nature of the problem.

Experimental determination and finite element model validation of the Anand viscoplasticity model constants for SnAgCu

The major focus of this work was the experimental determination of the nine constants required for Anands viscoplastic constitutive model for a lead-free solder alloy, 95.5Sn3.8AgO.7Cu. A series of tensile experiments were carried out: constant displacement rate tests with strain rates ranging from 6.0E-5/s to 1.0E-3/s at temperatures of 20/spl deg/C, 75/spl deg/C, and 125/spl deg/C; constant load tests at a range of loads from 8MPa to 64MPa for the same temperature range. The test specimen was a cast dog bone shape based on the ASTM E 8M-01 standard, with a diameter of 4mm and a gauge length of 20mm. Nonlinear least-squares fitting was used to determine the model constants. Comparisons were then made with experimental measurements of the stress-inelastic strain curves: excellent agreement was found. A finite element model of the test was also constructed using ANSYS 8.1 software. This software includes the Anand model as an option for its range of viscoplastic elements, requiring that the nine constants be input. In this case, an 8-noded axisymmetric element (VISCO108) was used to model the test specimen under constant displacement rate loading. The model was then used to predict the stress-inelastic strain curve under constant displacement rate conditions and this was compared to both the experimental measurements and the fitted Anand model. It was found that the Anand model and finite element predictions matched the experimental stress-inelastic strain curves for small strain rates, but that at higher strain rates the strain hardening behaviour of the solder was not captured accurately. The benchmarking of the ANSYS software showed that the Anand model was being implemented as expected. Using the fitted parameters in an FE model of an electronic component undergoing thermal cycling is likely to give acceptable results as the strain rates in this case are comparatively small.

The dynamics of multiple pair-wise collisions in a chain for designing optimal shock amplifiers

The major focus of this work is to examine the dynamics of velocity amplification through pair-wise collisions between multiple masses in a chain, in order to develop useful machines. For instance low-cost machines based on this principle could be used for detailed, very-high acceleration shock-testing of MEMS devices. A theoretical basis for determining the number and mass of intermediate stages in such a velocity amplifier, based on simple rigid body mechanics, is proposed. The influence of mass ratios and the coefficient of restitution on the optimisation of the system is identified and investigated. In particular, two cases are examined: in the first, the velocity of the final mass in the chain (that would have the object under test mounted on it) is maximised by defining the ratio of adjacent masses according to a power law relationship; in the second, the energy transfer efficiency of the system is maximised by choosing the mass ratios such that all masses except the final mass come to rest following impact. Comparisons are drawn between both cases and the results are used in proposing design guidelines for optimal shock amplifiers. It is shown that for most practical systems, a shock amplifier with mass ratios based on a power law relationship is optimal and can easily yield velocity amplifications of a factor 5-8 times. A prototype shock testing machine that was made using above principles is briefly introduced.

Experimental and Numerical Evaluation of SnAgCu and SnPb Solders Using a MicroBGA Under Accelerated Temperature Cycling Conditions

A comparative evaluation of the leading lead-free solder candidate (95.5Sn3.8Ag0.7Cu) and traditional tin-lead solder (63Sn37Pb) under thermal cycling conditions was carried out. A test vehicle consisting of four daisy chained 10×10 array 0.8mm pitch plastic micro ball grid arrays (microBGA) mounted on an 8-layer FR4 printed wiring board was designed. The board finish was organic solder preservative (OSP) for the lead-free devices and hot air solder levelled (HASL) in the case of the eutectic devices. An event detector was used to monitor the continuity of each daisy chain during accelerated temperature cycling, where the test vehicles were cycled with a ramp rate of approximately 3°C per minute from −40°C to 125°C, with 10-minute dwells and a total cycle time of 2 hours 10 minutes. Results to date plotted using a Weibull distribution indicate that the SnAgCu solder is more reliable under these conditions. Experiments were also carried out on large-scale lead-free solder specimens to determine the parameters required for the Anand viscoplasticity model. The Anand model was then implemented in finite element analysis using ANSYS® , where the submodelling technique was employed to determine the viscoplastic work per thermal cycle for each solder joint along the package diagonal. Schubert’s fatigue life model was used to predict the number of cycles to failure of each joint, although it should be noted that the necessary model parameters for the may need to be calibrated. Results indicate that the joint under the die edge is likely to fail first and that the SnAgCu solder is more fatigue resistant. The numerical predictions underestimate the fatigue life in both cases.© 2004 ASME

A stress-life methodology for ball grid array lead-free and tin-lead solder interconnects under impact conditions

Portable electronic products are often subject to impact or shock during use which leads to failures of the external housing, internal electronic components, package-to-board interconnects, and liquid crystal display panels. Moreover, the introduction of lead-free solder to the electronics industry will bring additional design implications for future generations of mobile information and communication technology (ICT) applications. In this paper, drop tests performed on printed circuit boards (PCBs) populated with ball grid arrays (BGAs) are reported. During testing, measurements from strain gauges were recorded using a high-speed data acquisition system. Electrical continuity through each package was monitored during the impact events in order to detect failure of package-to-board interconnects. Life distributions were established for both a lead-free and a tin-lead solder for various drop heights. The explicit finite element method (FEM) was employed to approximate the peel stress at the critical solder joint and a stress-life model was then established for both solders. Finally, failure analysis was carried out using microsection techniques, scanning electron microscopy (SEM), and energy dispersive spectroscopy (EDS). It was found that, for board level drop testing, different failure mechanisms can occur for different drop heights and that there is a considerable difference between the lead-free solder characteristic life and the tin-lead solder characteristic life.

A reliability evaluation of lead-free ball grid array (BGA) solder joints through mechanical fatigue testing

Lead is a hazardous substance which, when ingested can be toxic to humans; therefore it has been banned by a European Union directive in an aim to reduce its harmful effects on health and the environment. The ban, which comes into force on July 1/sup st/ 2006, means that electronic manufacturers must transfer from a tin-lead soldering process to a lead-free process. In this paper a reliability evaluation of a tin-silver-copper (SnAgCu) solder is presented with a baseline of tin-lead (SnPb). An experiment was carried out to optimize the surface mount reflow process and the reliability of the resulting solder joints was investigated using a torsional mechanical fatigue test method. The test vehicle comprised of an 8-layer FR4 printed circuit board (PCB) mounted with four ball grid array (BGA) components - each package comprising four daisy-chains. The basic principle of the torsion test was to stress the BGA solder joints repetitively in order to determine the number of cycles to failure. Graphs of the cycle number versus resistance were created and the numbers of cycles to failure were determined. The failure mechanisms were examined using cross-section and scanning electron microscope (SEM) techniques which showed cracking that initiated at the upper corner of the solder joint and propagated in the solder along the upper copper pad. This failure mechanism was observed for both SnAgCu and Snb solder joints. From a comparison of number of cycles to failure, the reliability of SnAgCu BGA solder joints was found to be superior to that of SnPb joints in torsion tests.

Optimising lead-free screen-printing and reflow process parameters

Introducing a lead-free solder replacement requires studies to be conducted by electronic assembly manufacturers in order to determine process alteration requirements and suitability of current equipment. This paper presents the results of an investigation of the screen-printing and reflow steps of the surface mount technology (SMT) manufacturing process. Experiments were conducted to investigate these two processes using a tin-silver-copper (95.5Sn3.8Ag0.7Cu) solder and a baseline of standard tin-lead (63Sn37Pb). 10/spl times/10 array micro Ball Grid Arrays (BGAs) mounted on 8-layer FR4 printed wiring boards (PWBs) were used with an organic solderability preservative (OSP) finish for use with lead-free components, and a hot air solder level (HASL) finish for use with tin-lead components. The screen-printing experiment investigated the deposition of the solder paste on the board. The parameters used in the investigation were print speed, squeegee pressure, snap-off distance, separation speed and cleaning interval with the responses being measurements of paste height and volume. Optimum screen-printer settings were determined which give adequate paste volume and height and a good print definition. The reflow experiment investigated the following parameters of the temperature profile; preheat, soak and reflow temperatures, and conveyor speed. The solder joints were examined using cross-section analysis and X-ray techniques in order to determine the presence of defects. The outcome of the investigation is a set of optimum settings for the screen-printer and reflow oven for use with SnAgCu lead-free solder.

The Response of a Miniature Scale Cantilever Beam to High-G Impact Stimuli

Many contemporary innovations in MEMS devices result from the miniaturization of existing macro-scale systems, exploiting changes in physical phenomena with scale. In terms of a system subject to mechanical stimuli, the response of the system changes significantly as scale decreases: in particular, natural frequencies increase and the system can sustain higher acceleration levels without damage. The objective of this paper is to investigate the response of a miniature scale cantilever beam to high-G impact stimuli in order to gain an understanding of its response. The test model is a machined aluminium cantilever-beam simply supported at one end. The beam cross section is 400 × 200 microns and has lengths of 4mm, 10mm and 20mm. The test bed is an Instron Dynatup 9250HV drop table. The beam response under impact is monitored using an IDT X-Stream XS-4 high-speed camera fitted with a telecentric 10x lenses. Theoretical and computer simulated models using ANSYS and LS-DYNA software are developed and compared with experimentally measured data to verify the accuracy of the techniques used to analyse the structural behaviour of the cantilever beams. The use of high-speed imaging when testing devices for short duration events proves to be beneficial for obtaining several data sets not achievable by post processing techniques.Copyright

SnAgCu Micro-Ball Grid Array (BGA) Solder Joint Evaluation Using a Torsion Mechanical Fatigue Test Method

Due to the hazard which lead poses to health and the environment the EU is banning its use in electrical and electronic equipment from July 2006. This ban along with the market drive to more environmentally friendly products means that tin-lead solders must be replaced with lead-free alternatives. This paper presents the results of an experimental investigation of the mechanical fatigue properties of tin-silver-copper (SnAgCu) solder joints with a baseline of tin-lead (SnPb). The test vehicle comprised of an 8-layer FR4 printed circuit board (PCB) mounted with four micro-ball grid array (BGA) components — each with a total of 100 solder balls in a 10×10 array. The solder joints were formed using surface mount reflow processes optimised for both solder types. A torsion mechanical fatigue test was employed to evaluate the solder joints — the principle of which was to stress the solder joints repetitively in order to determine the number of cycles to failure. The BGA components were daisy-chained — the resistance across each daisy-chain was monitored continuously during the cyclic defection of the test board. A profile of the increase in resistance with cycle number was established and the number of cycles to failure determined. The failure mechanism induced by the cycling was examined using cross-section and scanning electron microscopy (SEM) techniques. The results for SnAgCu joints show a superior performance during torsion mechanical fatigue testing than SnPb joints; giving a greater number of cycles to failure. The results from the tests presented in this paper show that the torsion test method provides a viable alternative to ATC as a qualification method for solder joints, while also providing substantial time savings — taking weeks rather than months to complete.© 2005 ASME