Project topics and materials
Get Complete Project Material File(s) Now! »
Electronic boards
This section introduces the reliability of solder joints taking into account the constrains of aerospace and defense industries. Electronic boards are described in detail taking into account the global structure and its constitutive elements and materials. The Manufacturing process of electronic packages is recounted next and the implication for reliability are highlighted. Finally, the actual challenges for the reliability of solder joints for critical applications are introduced in a critical discussion.
Aeronautic, space and defense critical markets
Electronic boards are nowadays part of numerous daily used products: computer, mobile, car, home automation systems etc. and generate a large number of technological developments. Consumer electronics market, mainly driven by smartphone applications, requires strong and constant miniaturization effort associated to functional performance increase for high volume, low cost and mild environments applications. A large number of new technologies are therefore developed to reach higher interconnection density of electronic boards. Commercially available electronic packages are also mainly developed for this market. These new technologies and packages are interesting for critical applications. However, reliability is a critical aspect for applications such as commercial avionics, transportation, space and defense. Integration of such new packages and technologies requires to qualify their use with the constraints of critical markets: high reliability, harsh environment and long mission profiles. In addition, if required, the need must be highlighted to apply strengthening strategies or to restrict their use to applications in less severe environment.
Products developed for space, defense and avionics industries have low production volume (10-10,000 pieces/year). Due to the high cost of each electronic board and the efficiency requirement, packages also need to be able to be repaired after failure. Furthermore, products lifetime is longer for these applications than for consumer electronics. In this context, it is critical and meaningful to study end-of-life reliability and electronic board mechanical behavior. For example electronic equipment embedded in a satellite needs to resist more than thirty years with a failure acceptable level smaller than 0,001% (see table for illustration Figure 1). The markets classification is based on a proposition from the American industry association IPC (Institute for Electronic Circuit Interconnects and Packaging)-SM-785 standard [1].
Electronic board structure
Electronic boards are used to perform complex functions in larger systems. Electronic boards are basically composed of a Printed Circuit Board (PCB) on which electronic packages are assembled with solder joints (see Figure 2), and are connected to other systems with connectors. The assembled electronic board with its packages is called Printed Circuit Board Assembly (PCBA). Electronic packages are composed of a silicon die, a substrate and the molding. The silicon die is programmed with algorithms to perform specific calculations and is connected to the package substrate. This connection is the First-Level Interconnects. The substrate, which can be assimilated to a PCB, connects outputs of the die to the solder joints. Substrate has a laminate structure with copper layers and laminates, which are composite material with resin and glass fibers. More details about laminates are given in the following paragraph dedicated to PCB materials.
Solder joint ensures the electrical link between electronic packages and the PCB (Second-Level Interconnects). Electronic packages are connected to perform complex functions. Solder joint maintains also mechanically electronic packages on the PCB. Silicon dies heat up during use. Solder joints are also used to dissipate this heat. Solder joint types are generally classified according to the form of their attachments on the board. In fact, these connections types have a direct impact on packages reliability because failure often occurs in interconnections. Figure 3 illustrates this classification. Understanding the packages structure is important to build relevant numerical models of board for simulation purpose.
PCB main goal is to interconnect electronic packages together to create more complex functions. Historically, PCB where monolayer boards: only composed of a dielectric support with a single printed copper layer. PCB are nowadays composed of multiple copper and dielectric layers (laminates) alternatively stacked together. This multilayer structure of the PCB is required to due to the increasing interconnect density of nowadays electronic packages. Laminate layers are composite structure: matrix of resin is reinforced with fiber glasses. Resin content in the laminate is an important parameter to calculate laminate mechanical properties. Different natures of resins can be used as epoxy or polyimide resins. The resins, with Coefficient of Thermal Expansion (CTE) around 20 ppm/K, is reinforced with fiber glasses which gives to the laminate a CTE in the fiber direction around 13 ppm/K. This value must be considered as first order approximation. Calculation of PCB mechanical properties is complex and out of the scope of this thesis. Lot of parameters including PCB stack-up and resin content must be taken into account.
PCB manufacturing process can be decomposed in four main steps. First step is the etching of copper layers to form lines and planes to interconnect packages according to the electrical scheme. Packages footprints are added in the two external layers for the assembly of the packages. Second step is the manufacturing of the multilayer structure by laminating the different copper layers with dielectric in between for insulation. Third, after lamination, the different copper layers are linked together by mechanical or laser drilling and copper plating of vias. This process is repeated several times to obtain the desired PCB stack-up. Finally, finish plating and solder mask are deposited on the outer layers to preserve exposed copper of footprints from oxidation and to help for packages assembly.
Electronic board assembly
Solder joints are formed during the assembly of the electronic board. A schema with the principal steps and the temperature profile of the PCB during the assembly in presented in Figure 5. More precisely the assembly process consists of the following steps:
(i) Solder paste is deposited on PCB footprints by a screen-printing process,
(ii) Package are deposited on top side with a pick and place loading arm.
(iii) The board is heated up in a reflow oven following a dedicated thermal profile.
The thermal profile reaches a peak temperature above the melting point of the solder and presents specific heating and cooling ramps to minimize thermal residual stresses. For double-sided PCB architectures, the process is repeated in order to assembly an additional series of packages on the other side of the PCB. The process is ended by a series of inspections and electrical tests to verify the final package assembly. A protective layer can be applied by conformal coating on the complete board, without adding further temperature constraints on the packages.
Let us further remark that several process parameters have a direct impact on the reliability of the solder joints and consequently on the entire PCB assembly. Residual stresses are an outcome of the different coefficients of thermal expansion and the temperature changes during the assembly as they can facilitate the opening or closure of cracks if they have a tensile or compressive character. Moreover, cooling speeds will determine the precise crystallization process of solder joints and influence as such their joint microstructure.
Assemblies from two distinct processes will be studied in the following chapters, (a) SAC305 using a lead-free solder joint and (b) Backward SnPb+ using a lead-containing alloy SnPb. If lead-free joints are the standard norm in the consumer electronics, the use of lead is still tolerated in critical applications as aerospace and defense. In the lead-free SAC305 assembly process, solder balls and solder paste are made out of the SAC305 alloy. The SAC305 alloy is chemically composed of 3 wt.% of silver, 0.5 wt.% of copper and 96.5 wt.% of tin. The backward SnPb+ process, uses a solder paste made out of a leaded Sn37Pb alloy (37 wt.% of lead and 63 wt.% of tin), whereas the balls of the electronic package are made out of SAC305. Let us further notice that the solder joints of the Wafer Level Package (WLP) which will be discussed in chapter 2 contains 7 wt.% of lead after backward SnPb+ process. This value has been computed from the deposited quantity of solder paste. Up to date, aerospace and defense applications are not in the scope of environmental directives restricting the use of Lead.
Heterogeneity of assembly processes is a difficult and complex task from an industrial point of view. A low volume of production increases the relative cost of development and manufacturing and can justify the mixing of assembly technologies, which drives in itself the increasing cost. An example of the complexity of the assembly is the size of the package pitch, defined as the distance between two solder joints. Current electronic boards have packages with pitches in the range of 0.4 to 1.27
mm. Have this variety on an assembly line requires the existence of stencils with different thicknesses in different areas during the assembly process. As a consequence, the qualification of such an assembly process becomes very difficult and expensive.
Reliability of solder joints
Failure is a state of inability to perform a normal function. Reliability is the ability of a product to perform a under given conditions and for a specified period of time with an acceptable failure risk. Miniaturization and densification of electronic boards increase the risk of solder joint failure due for example to the induced reduction of solder joint sizes. Electronic boards structure and solder joint have been described in the previous section. The definition of reliability of solder joints will be now introduced. Mission profiles and induced stresses for solder joints will be presented. Accelerated reliability tests performed by companies to guarantee the reliability of solder joints will be detailed. Finally, some details about the use of simulation to predict cycles to failure in the context of solder joint will be given. Simulation is in fact an interesting solution to help reliability tests, understand failure mechanism and increase results from experimental tests.
Failure rate in time analysis
Let us now consider at a fixed time instant the instantaneous failure rate of a device. The evolution of the later with increasing time has proven to be a “bathtub” curve as described in [4]. In this representation one can identify three distinct phases characterized as (i) infant mortality, (ii) useful life also denoted as the “random steady-state” and (iii) end of wear or wear out (see Figure 6 for schematic representation).
The “infant mortality” corresponds to early failure and is associated to weakness created during or due to the manufacturing process. The reduction of this initial failure rate is generally harnessed through improvements in the quality control of the manufacturing process manufacturing and is not within the purpose of this work.
The “random steady-state” period characterizes the useful lifetime. In this period experience shows that failure occurs mostly random due to exceptional accidental loadings and has a low rate. Moreover, no evidence exists that would incriminate solder joints for the failures in this region (see [1] for a detailed discussion of this topic).
The “wear-out” region corresponds to the end-of-lifetime of the device under standard service loadings. It is characterized by a high increase of the failure rate and is a direct consequence of accumulating damage on different components. The detailed understanding of the subsequent damage mechanism and the precise distribution lifetimes corresponding to each failure mode are important for several reasons. On the one hand side they define the reliable life-time of the components and define inspect periods and non-destructive failure detection techniques and on the other hand side they permit to improve the component design and the assembly process by eliminating failure mechanism or by increasing lifetime through optimal design. It is obvious that in order to guarantee lifetime of the products engineers has to understand both service and exceptional loadings, or what is denoted in aerospace and defense applications the mission profile.
Stresses in solder joints during mission profile
Electronic boards of space, defense and avionics products are used in harsh environment. Except manufacturing defects, corrosion and other chemical damage, there are three main sources of failure for solder joints of electronic boards: punctual mechanical shocks (during landing or operation for example), vibrations and large thermal variations [3]. Other environmental contributions, as corrosion or oxidation, induce damage mechanisms in solder joints but are not considered in the thesis. Manufacturers of critical markets need to guarantee long lifetime reliabilities in harsh environment for their products. If any failure occurs in the system, the consequences can be human casualties (for example in transport) or a huge financial losses (in satellite or space applications).
Mission profile analysis
Mission profile is the description of the product environment and the loadings submitted during the service. The environment loadings are described with physical quantities as temperature changes, vibration levels and relative humidity. An example of the daily temperature variations for two cities in different regions of the United States are represented in Figure 7 [5]. Possible mission profile definition for these two cities is highlighted by the two dotted lines. Red lines represent the maximum and minimum temperatures and the blue line the most frequent cyclic stress. This analysis is presented as an example to illustrate how temperature variations from the environment can be evaluated. These temperature variations can describe for example the mission profile of a system which will be used in these cities. However, more investigations are required to evaluate the induced mission profile of the electronic boards of this hypothetical system.
Reliability of solder joint is qualified by manufacturers according to the product mission profile. However, it is not possible to perform 30 years of test before using technologies. Accelerated reliability tests are performed in order to evaluate the reliability of a product by subjecting it to conditions in excess of its normal service parameters. The equivalent lifetime in the field is predicted with accelerated test results and an acceleration factor (see Figure 8). These acceleration factors are estimated by engineers with the knowledges of fatigue properties of the considered component.
Table of contents :
General introduction
Chapter I. Context and Objectives
1. Electronic boards
1.a. Aeronautic, space and defense critical markets
1.b. Electronic board structure
1.c. Electronic board assembly
2. Reliability of solder joints
2.a. Failure rate in time analysis
2.b. Stresses in solder joints during mission profile
2.c. Reliability challenges for critical applications
2.d. Prediction of solder joint failure with simulation
3. Mechanical properties of solder joints
3.a. Lead-Free solder joint microstructure
3.b. Elastic and plastic mechanical properties
3.c. Viscosity properties
3.d. Fatigue properties
4. Conclusions of the literature review and perspectives
5. Objectives
Chapter II. Experimental Setup
1. Innovative shear test bench
1.a. Shear test bench description
1.b. Samples of the shear test bench
1.c. Schematic representation of the shear test bench
1.d. Experimental results post-processing
1.e. Test in temperature
2. Accuracies of force and displacement measurements
2.a. Accuracy of the force sensor
2.b. Accuracy of the local displacement
2.c. Accuracy of tests in temperature
3. Monotonic tests
3.a. Results of the mechanical characterization
3.b. Comparison of the monotonic test results with the literature
3.c. Residual stress for test in temperature
4. Step-Stress Results
4.a. Step-stress test
4.b. Step-Stress Results
4.c. Analysis of step-stress results
5. Conclusions
Chapter III. Failure Definition
1. Mechanical and electrical failures
1.a. Mechanical failure
1.b. Electrical failure
1.c. Specific configuration in this work
2. Criterion and mechanism of failure
2.a. Force and electrical resistance monitoring
2.b. Initial inspection
2.c. Failure mechanism during fatigue tests
2.d. Homogeneity of the loading
3. Mechanical parameters extraction
3.a. Methodology
3.b. Mechanical parameters for fatigue tests without dwell time
3.c. Mechanical parameters for fatigue tests with dwell time
4. Conclusions
Chapter IV. Fatigue Results
1. Fatigue tests
2. Test controls
2.a. Validation of experimental results based on selection criteria
2.b. Disturbance detected during the test
2.c. Disturbance detected after post-processing
2.d. Test control synthesis
3. Results and discussions
3.a. Weibull distribution
3.b. Mechanical response
3.c. Damage indicators
3.d. Discussions
4. Conclusions
Chapter V. Cyclic damage analysis
1. Fatigue of solder joints
1.a. Fatigue regimes
1.b. Importance of creep-fatigue interaction for solder joint alloys
2. Analysis of the results of the fatigue tests plan
2.a. Shear stress and strain
2.b. Inelastic strains
2.c. Simulation of the sample mechanical response
3. Creep-fatigue-interaction law for solder joint
3.a. Damage indicators synthesis
3.b. Damage law for solder joint material
3.c. Duration of dwell time
4. Conclusions
General conclusion
Bibliography
