AREA CONSORTIUM REPORTS

1994 REPORTS

Variable and Response Identification for Mass Reflow Component Placement Parameterization
Abstract: A major step in the mass reflow assembly process is the placement of the components. A typical component placement process will include vision locating of global or local fiducials, vacuum pick up of the component, mechanical or vision centering of the component and then placement of the component. Fine pitch components (0.65mm pitch and below) must be vision centered because of the need for higher placement accuracy and repeatability. Fine pitch component leads are also generally too fragile to mechanically center.

The variables of the placement process are numerous and interactive. The three main variables that must be controlled and understood to produce a successful placement process are the printed circuit board, the component and the placement machine. These three main variables contain many variables.

This test plan will address only the placement machine and the variables directly associated with it. The objective of the mass reflow component pick and place parameterization portion of the Ultra Fine Pitch project is to identify, parameterize and optimize all placement variables that are directly associated with the placement machine. Data generated from this testing will also be utilized in the Fine Pitch Placement Modeling and Simulation work being conducted at S.U.N.Y. Binghamton. Placement model verification will also be conducted as part of this test plan. It should be noted that this document only identifies the variables, parameters and responses of the test. Please refer to individual Test Results reports for the methodology of the experiments.

Variable and Response Identification for Stencil Parameterization
Abstract: A major step in the mass reflow assembly process is the application of solder paste to the printed circuit board. Stencil printing is the most common method of applying solder paste to printed circuit boards that contain fine pitch devices. The goal of stencil printing is to apply an accurate and repeatable volume of solder paste, to precise locations on the printed circuit board. The variables of the screen printing process are numerous and interactive. The four main variables that must be controlled and understood to produce a successful screen printing process are: the solder paste, stencil, printed circuit board and the screen printing machine.

These four main variables contain many variables. This test plan will address only the stencil. All of the stencil testing that will be performed under this test plan will be solder paste printing related.

The primary goal of the stencil parameterization portion of the Ultra Fine Pitch project is: to identify, parameterize and optimize all solder paste printing related variables of stencils.

It should be noted that this document only identifies the variables, parameters and responses of the test. Please refer to individual Test Results reports for the methodology of the experiments.

Variable and Response Identification for PCB Parameterization
Abstract: The Ultra Fine Pitch Project that is currently under way at Universal Instruments Corporation will be broken down into two major sections, mass reflow assembly and hot bar assembly. The project will attempt to address all facets of the assembly process for each of the two assembly techniques. One major step in the mass reflow assembly process is the application of solder paste to the printed circuit board.

Stencil printing is the most common method of applying solder paste to printed circuit boards that contain fine pitch devices. The goal of stencil printing is to apply an accurate and repeatable volume of solder paste to precise locations on the printed circuit board. The variables of the screen printing process are numerous and interactive. The four main variables that must be controlled and understood to produce a successful screen printing process are: the solder paste, stencil, printed circuit board and the screen printing machine.

These four main variables contain many variables. This test plan will address only the printed circuit board and the printing related variables directly associated with it. All of the printed circuit board testing dealt with in this outline will be solder paste printing related.

The primary goal of the printed circuit board parameterization portion of the Ultra Fine Pitch project is:

to identify, parameterize and optimize all solder paste printing related variables of printed circuit boards.

It should be noted that this document only identifies the variables, parameters and responses of the test. Please refer to individual Test Results reports for the methodology of the experiments.

Test Results of Vendor Solder Paste Evaluation- Printability and Reflowability
Abstract: Solder paste application is one of the most important and critical steps of an Ultra Fine Pitch Mass Reflow Assembly Process. A large number of interactive variables must be understood, properly set up and controlled to obtain a successful Ultra Fine Pitch printing process. One of the most important variables is the solder paste itself. Selecting the solder paste flux type (solvent clean, water clean, no-clean air reflow, no-clean inert reflow, etc.). Powder size, percent solids and viscosity must be considered. Although any other solder paste variables must be considered when selecting a solder paste, this research will focus only on the printability and the reflow characteristics of Ultra Fine Pitch solder pastes.

The objective of the Vendor Solder Paste evaluation was to determine the best solder pastes for further evaluation for 0.3mm and 0.4mm pitch mass reflow assembly. Water clean, no-clean air reflow and no-clean inert reflowable materials were evaluated. A total of seven vendors submitted twenty-five samples for evaluation. All the solder pastes tested were Sn63/Pb37 alloy and contained powder sizes of -400 mesh or smaller. Only no-clean and water cleanable materials were tested with priority being focused on the no-clean materials. The twenty-five solder pastes were ranked for printability as a whole group and also by flux category. The printability ranking was based on 0.3mm pitch printing.

The solder paste reflow portion of the experiment was conducted by assembling one UFPIII board and then visually inspecting the board for soldering defects and soldering “quality”. Other solder paste related tests that have been conducted during the Ultra Fine Pitch Project for the fluxes are Surface Insulation Resistance, Electro-Chemical Migration and Wetting Balance Performance. Please note that some of the solder pastes that were evaluated are experimental products. Also note that the Koki product that was tested was supplied by one of Principles without product information. The history of this sample is unknown.

Accuracy Evaluation Methodology for Fine Pitch Surface Mount Placement Equipment
Abstract: This report summarizes the information contained in the document “Accuracy Evaluation Methodology for Fine Pitch Surface Mount Placement Equipment” in regard to the data and results concerning UIC 4766 Omniplace Fine Pitch Surface Mount Placement System (Machine A). Testing was performed in a fashion which allowed appropriate data to be collected and used to make statistically valid inferences regarding the accuracy and repeatability of the 4766 Omniplace System. A discussion of the implementation of the accuracy evaluation method follows including the necessary inputs, the process, and the outputs from the process.

Chemical Resistance Testing of Solder Paste Stencils to Solvents
Abstract: This document describes the procedure followed to test the resistance of stencil medium and support adhesives to various commonly used stencil cleaning agents.

Evaluation and Comparison of Solder Paste Stencils
Abstract: Surface mount attachment is a highly complex process involving a vast number of critical parameters each contributing to the success or failure of printed circuit board assembly. One key group of variables is associated with the surface mount solder paste stencil. Stencil material, aperture shape, fabrication method, as well as a great number of other factors, drastically affect final assembly yields and long term product reliability. Factors such as feature accuracy and repeatability can be measured and parameterized preceding assembly and then be used to correlate assembly defects with a given aperture size. By examining and parameterizing each type of stencil used in the ultra fine pitch project, predictions can be made to the final assembly yields of fine pitch surface mount attachment process.

Component Lead Coplanarity
Abstract: This document describes the method for testing fine pitch mass reflow component leads to determine coplanarity of components tested at UIC. All other components will be measured with a Robotics Vision System LS 2000.

Component Lead Deflection Testing
Abstract: This document describes the procedure followed to test the amount of component lead/joint deformation due to a uniaxial load. The test is intended to simulated loads and displacements caused by a mismatch of thermal coefficient of expansion.

Placement Accuracy Evaluation for UIC Model 4682B Fine Pitch Bonder
Abstract: Testing was performed in a fashion which allowed appropriate data to be collected and used to make statistically valid inferences regarding the accuracy and repeatability of the 4682B Fine Pitch Bonder Placement System. A discussion of the implementation of the accuracy evaluation method follows including the necessary inputs, the process, and the outputs from the process. The purpose of this test is to identify the accuracy characteristics of the placement system used for the Hot Bar Attachment portion of the Ultra Fine Pitch Attachment Study. This work was conducted as an extension of that described in “Parameterization of the Fine Pitch Placement Process” by Snyder and Westby.

Evaluation and Comparison of Printed Circuit Boards
Abstract: Surface mount attachment is a highly complex process involving a vast number of critical parameters each contributing to the success or failure of printed circuit board assembly. One key group of variables is associated with the surface mount attachment pad metallurgy. Plating type and thickness, uniformity and shape as well as a great number of other factors drastically affect final assembly yields and long term product reliability. Factors such as plating thickness can be measured and parameterized preceding assembly and then used to correlate assembly defects with a given plating thickness. Other defects such as solderability problems can be examined in such a fashion. By examining and parameterizing each type of printed circuit board attachment pad metallurgy used in the ultra fine pitch project, predictions can be made for final assembly yields of fine pitch surface mount attachment process.

Experimental Load Deflection Tests on UFP Devices
Abstract: Experimental Load-displacement tests were designed and performed on several common ultra-fine pitch mass-reflow and TAB surface mount devices. These tests involve compressing a device onto a smooth glass surface past the lead’s elastic limit so that lead stiffness and yield point could be determined. Load-displacement data was collected for individual leads, half the total number of leads (two opposite sides), and for entire components.

Yield was based on the load necessary to cause a 50% reduction in lead stiffness in the direction normal to the seating plane. These results may be used to determine appropriate placement loads for devices during circuit board assembly. Also, these tests may be used to determine the effects of lead-frame roll direction on the mechanical properties of the leads.

Finite Element Fatigue Modeling Studies of Gull-Wing Leaded Solder Joints
Abstract: The scope of this project encompassed the application of the finite element method for the fatigue life estimation of fine pitch, gull-wing leaded devices. A number of background issues, including solder constitutive modeling, finite element modeling approaches, and failure prediction methods, to name a few, were addressed.

Based upon extensive study of the available literature, the accumulated matrix creep method was chosen as the fatigue failure indicator. Using this method, various leads and solder joints were modeled and their fatigue lives estimated. Parameters such as lead toe angle and solder volume were considered. Guidelines for using the finite element method for solder fatigue prediction were developed.

Six specific lead and solder joint geometries, based on the Consortium test board, were modeled. Their fatigue lives were estimated for the thermal profile adopted by the Consortium. It was found that for these leads, the local thermal mismatch between the solder, lead, and pad, played a dominant role in the fatigue life prediction. A key result was that the traditional relationship between lower lead stiffness and higher fatigue life was not observed for the cases that were dominated by local thermal mismatch.

FEM Analysis of Contact Stresses During Solder Stenciling Operations
Abstract: This work is the result of a pilot study to evaluate stencil stresses during solder paste printing operations. Two-dimensional finite element models were constructed of the stencil, mask, and lead attachment pads so that a parametric study of important variables such as masking height, stencil material properties, and stencil/pad overhang could be performed. These models utilized friction interface elements, elastic-perfectly plastic material properties, and large deflection capabilities.

Perhaps the most significant conclusion drawn from the models is that high yield strength is an important characteristic for a stencil material. Material properties for brass (representing low modulus and yield strength) and nickel alloy 42 (representing high modulus and yield strength) were used in the study. These models may also be used to evaluate other alloys.

Flux Activity Evaluation Through the Use of Wetting Balance Analysis
Abstract: This report identifies an approach and studies the influence of flux activity in the promotion of wetting through the use of wetting balance analysis. Standardized specimens were used along with a set of specified test variables. The flux types included in this study exhibited different performance characteristics in promoting the wetting of solder to the various prepared specimens. This caused the test responses (wetting forces, wetting time etc.) to be different from one type of flux to another. The effect of artificial aging on the surfaces to be soldered and various flux types was studied in this research.

Procedure for Grading Solder Paste Deposit Definition
Abstract: This document defines the equipment and methods that will be used during the Ultra Fine Pitch Project for grading solder paste deposit definition.

Heat Spread Evaluation
Abstract: During the hot bar reflow soldering process, heat is transferred from the thermode to the component leads and ultimately to the printed circuit board attachment pads and surrounding surface. This study investigated the geometric nature of this heat spread and relative thermal thresholds. This procedure may be beneficial for testing individual component sites for thermal attributes and assist in profile generation. The information from this study may also be applied to soldering flux constraints whereas some fluxes require a minimum thermal exposure to render the activators inert. Different types of substrates were used in this study. The results indicate that the area of heat spread on the PCB depends upon the concentration of circuit traces, the dwell time of the thermode, and the peak temperature.

The experiment was designed to identify the area of heat spread resulting from various temperature profiles, thermode configurations, and substrate materials and to construct a statistical model for the heat spread geometry.

UFP Hot Bar Bonding Project Experiment One
Abstract: The results obtained from the preliminary experiments and from Experiment One are detailed in this report. The preliminary experiments were designed to optimize the settings of the UIC 4682B fine pitch bonder (such as the thermode temperature profile, the amount of flux dispensed, etc.). This was done prior to the execution of Experiment One.

Experiment One was designed to study the influence of specific factors (solder thickness, pad width, etc.) on solder joint quality. The data collected was analyzed with respect to performance measures (percentage of shorts, bond strength, etc.) using statistical software (E-Chip). Three-dimensional graphs were plotted to visualize the effects of the variables (domain factors) considered on the performance measures.

Bar Reflow Soldering of TAB Components on Bare Copper Attachment Pads
Abstract: This research studied the feasibility of bonding TAB components with 6.0 microns thick solder plating to bare copper attachment pads. The bonded components were peel tested (tensile tested) to evaluate their bond strength. The results obtained were compared to the solder joints formed by bonding the TAB components with 0.63 microns thick solder plated leads to solder deposited attachment pads. The peak tensile strength of the solder joint was used as the gauge for comparing the strength of the solder joints for various scenarios.

Hot Bar Reflow Soldering Yield Study
Abstract: This yield study was performed to validate the process related inferences made from previous experiments conducted during the Hot Bar Attachment Study. TAB and QFP components were attached to the TAB-1 board in this study. The results from this study indicate that the hot bar bonding process for outer lead attachment of TAB and fine pitch QFPs can produce high yields. The response used to evaluate the hot bar soldering process was the occurrence of solder defects such as shorts and opens. This yield study also highlighted reliability problems associated with excise and form tooling.

Fatigue Investigation of Lap Shear Solder Joints
Abstract: The electrical resistance of lap shear (60%Sn-40%Pb) solder joints has been measured while the specimens were undergoing cyclic fatigue testing. The resistance measuring system had a noise level of less than one nano-ohm. An FFT analyzer was used to transform the time varying resistance to the frequency domain, and measurements were made of the average resistance (zero frequency amplitude), amplitude of the resistance change at the mechanical drive frequency, and the amplitudes of the resistance change at the second and third harmonics of the drive frequency.

The average resistance and resistance amplitude at the drive frequency exhibited a systematic behavior as the specimens were cycled toward failure. Initially the average resistance increased, followed by a much stronger decrease, and lastly increased markedly as the specimen fractured. The resistance amplitude at the drive frequency showed an initial region of instability which was followed by a region exhibiting the same strong decrease seen in the average resistance. After this decrease, the resistance amplitude increases as the specimen begins to fracture, and then the amplitude of the resistance change either leveled off or decreased to zero depending on whether the specimen separates or continues to rub together after complete fracture.

Component Lead/Solder Joint Tensile Test
Abstract: This procedure provides a method to accurately quantify leaded component / solder joint tensile strength by measuring load / displacement characteristics through use of an Instron 4502 Materials Testing Machine.

Test Plan for Mass Reflow of Computer Related 0.4mm Pitch Devices
Abstract: Universal Instruments Corporation has been requested by a customer, to investigate a double sided, mass reflow process, for assembly of a computer related printed circuit board. The process requires that the bottom side of the printed circuit board, that contains 0.5mm, 0.65mm, and 50 mil pitch devices, be assembled first. A standard – print solder paste, place components and mass reflow – process will be used.

The top side of the printed circuit board, which will be assembled last, requires that one 160 pin, 0.4mm pitch QFP device, along with 0.5mm, 0.65mm and 50 mil pitch devices be assembled by the same process. The panel to be tested contains two printed circuit boards rotated at 180 degrees from each other. This project will be used as a preliminary investigation for the Ultra Fine Pitch Project.

Modeling and Simulation of the Fine Pitch Placement Process
Abstract: The effectiveness of the placement process in ultra fine pitch surface mount PCB assembly is influenced primarily by the accuracy of the printed circuit boards (PCB), the dimensional stability of the surface mount components (SMCs), and the capability of the pick and place machine used. Engineers in a PCB assembly facility often need to evaluate the capability of their respective manufacturing processes (and more specifically their SMC placement process). Consequently, they need to understand the variations associated with the primary attributes of the PCBs, the SMCs, and the placement equipment and their impact on the yields. This knowledge will guide the manufacturer in determining alternatives to improve process yields.

The objective of this research was to understand how variations in the functional parameters of the PCBs, SMCs and pick and place machines affect the objective function of lead to pad coverage. A system model was formulated to quantitatively determine the impact of each set of inputs on this objective function. Computer based simulation software was designed and developed to estimate lead to pad coverage.

The computer based solution methodology that was developed provides a reasonable estimate of the distribution of the lead to pad coverage based on the specific PCB, SMCs, and placement machine characteristic data.

The placement process simulation software developed is user friendly. The output format used combines textual and graphic modes. This simulation software system has been rigorously tested and verified. Satisfactory results were achieved.

Visual Yield Inspection Procedure for Mass Reflow Assembly
Abstract: This document defines the equipment and methods that will be used during the Ultra Fine Pitch Project for visually inspecting mass reflow assembled boards. The inspection process will determine the assembly yield for any given production run.

Test Results of Palladium Lead Finish Experiment
Abstract: The objective of the Palladium Lead Finish experiment was to determine the effect on Ultra Fine Pitch Assembly, when using component leads plated with palladium. All previous mass reflow assembly experiments used component leads plated with Tin\Lead. The Tin\Lead alloy on these leads ranged from a 60\40 to a 89\11 percentage.

The two devices assembled in this experiment were a 0.4mm pitch, 80 pin PQFP and a 0.5mm pitch, 208 pin PQFP. The leads on the two devices were plated with approximately 3-4 micro inches of palladium over 80 micro inches of nickel. The response from the experiment was assembly defects in the form of open solder joints and solder bridges.

Both devices were assembled on three different attachment pad designs per board. The attachment pad length was fixed for each of the three sites, while attachment pad width was varied at three levels per device type.

Component lead coplanarity and solder paste deposition was not considered in this experiment due to the unavailability of lead coplanarity data. The experiment was conducted using a fixed solder paste print process and fixed nitrogen reflow process.

The printed circuit boards used in this experiment were supplied from a number of different vendors. Bare copper, tin\lead and gold attachment pad finishes were included in the matrix. Due to availability, the older (less optimized attachment pad designs) UFPIII rev A printed circuit boards were used in this experiment.

The process parameters used during the experiment were derived from the results obtained from the Final Assembly Yield Experiments, Test Results #0037.These parameters were set at their optimum levels based upon previous experimentation.

Post Reflow/TAB Process Solder Joint Strength Study
Abstract: Typically, the attachment process of TAB components to printed circuit boards follows mass soldering processes such as solder paste reflow and/or wave solder operations. This sequence is followed to minimize possible damage to the delicate leads of the TAB components, avoid heat damage to the process materials used in TAB technology (encapsulating materials, heat transfer compounds, etc.) and to avoid contamination entrapment from conventional board assembly cleaning methods.

As a result of the process sequence, the solder alloy of the attachment pads for the TAB components is generally reflowed at least once prior to TAB component attachment. This could affect solder joint bond strength and yield. The purpose of this experiment was to investigate the effect on solder joint bond strength from reflowing the solder on the attachment pads prior to the hot bar operation.

Test Results of Solder Paste Print Experiment
Abstract: This Solder Paste Print Study is one of a series of experimental studies conducted in the SMT Laboratory for the Ultra Fine Pitch Project. The purpose of this study can be summarized as follows:

– To gain a better understanding of the solder paste stenciling process by considering solder paste characteristics along with the various fine pitch components and screen printer process variables.

– To identify the optimum conditions / regions in this window of interest.

Printed Circuit Board Evaluation Test Plan
Abstract: The Ultra Fine Pitch project currently underway at Universal Instruments encompasses many facets of surface mount technology. The evaluation of printed wiring boards is an imperative of this test program. The overall quality and manufacturability of an SMT assembly is dramatically affected by the accuracy and repeatability of the substrate. The goal of this test sequence is to identify the critical parameters, including pad size, that are required to manufacture a substrate suitable for assembling ultra fine pitch components and to identify the PCB construction types that can be produced which meet this criteria.

Ultra Fine Pitch Printed Circuit Board Inspection Criteria
Abstract: The need for accurate and repeatable fine pitch printed circuit boards (PCBs) is greatly increased as ultra fine pitch mass reflow processes are being developed and implemented. Assembly yields are highly dependent on the materials and process used to create the desired PCB. Both additive and subtractive technology can produce usable PCBs but extreme care must be taken to parameterize the characteristics of each.

A wide variety of factors are available to today’s PCB user such as the fabrication method, base material, photolithographic processing as well as attachment pad metallurgy. A test procedure for the examination and parameterization of fine pitch PCBs must be implemented between the vendor and the end user in order to achieve successful fine pitch mass reflow applications. The following is a recommended inspection guideline developed during the Universal Instruments Ultra Fine Pitch Consortium program, and was designed to help end users create an adequate inspection test plan.

The PCB inspection process can be separated into three distinct categories; manufacturing process, materials and metallurgy, and stability and metrology. The fabrication process chosen should be fully parameterized for each vendor prior to manufacture of actual production level PCBs. Once the vendor is qualified for the desired process, periodic evaluation of the process issues should be implemented.

Materials should be consistent from lot to lot provided the vendor does not change the material type, formulation, or properties. An initial evaluation of a specific set of materials should be performed and periodic examination of the material is recommended.

Metrology must be performed on each set of PCBs supplied by a vendor. Full inspection may not always be warranted, but key parameters must be checked to insure adequate performance. Feature size and position accuracy and repeatability must meet predetermined criteria for every PCB produced.

Procedure for Solder Paste Printing Experiments
Abstract: This document defines the equipment and methods that will be used during the Ultra Fine Pitch Project for solder paste printing experiments.

Repeatabilty Analysis of the SVS 7530 Inspection System
Abstract: The Synthetic Vision Systems (SVS 7530) is an inspection system used to measure the volume of solder paste deposited and the accuracy of the placement of components on Printed Circuit Boards. The SVS 7530 system uses laser based technology to perform the inspection. Since the SVS 7530 is an inspection station, it is necessary to insure that it meets the accuracy and repeatability requirements.

The purpose of this preliminary study is to determine the repeatability of the SVS 7530. Factors like the threshold settings and the orientation of the pads were considered during the study. The solder volume on the test board was measured in two directions at different threshold levels. An Analysis of Variance (ANOVA) was done to determine the influence of the factors mentioned above on the solder volume measured. The consistency of the measurements over time was also examined by measuring the solder volume at fixed time intervals.

The results suggest that the average solder volume measurement was independent of the threshold value, but the variations in the measurements depended on the threshold value. As a result of the analysis conducted on UFP I board, the repeatability of the solder paste measurement was found to be 2%. The result of the analysis conducted on two 284 Pin QFP Ultra-Fine Pitch components indicated that the repeatability of the solder paste measurement is 2% for the first device and 4% for the second. Significant differences were observed between the measurements of the vertical and horizontal solder pads of the same device. The volume measured was found to decrease with time. The results of the gage studies performed are presented along with supportive data.

Results for Assembly of 208 Pin 0.5mm Pitch QFP Components
Abstract: In an attempt to parameterize the transition between current standard and fine pitch mass reflow surface mount assembly, an investigation of a 208 pin 0.5 mm pitch mass reflow component has been performed. Although 0.5 mm pitch assembly can be difficult to implement in certain applications, process refinement can easily lead to high yield attachment. Documentation and investigation of the 0.5 mm pitch assembly process will help in parameterization of finer pitch devices. Results of this experiment will help characterize potential difficulties in assembly of the UFP2 and UFP3 printed circuit boards.

Results for Squeegee Testing
Abstract: In an attempt to better understand the characteristics of typical solder paste stencil squeegees, an investigation into the physical properties of both rubber and metal squeegees was performed. Deflective behavior of the material under typical solder paste printing loads has been simulated using various material property tests. Frictional drag forces that occur during the printing process have been examined as well.

The rubber durometer squeegees performed expectedly. The harder material reflected much less than softer materials under the same loading conditions. The metal squeegee showed very interesting results. At low loads the metal squeegee deflected a great amount acting like a very soft rubber compound. At higher loads the metal squeegee exhibits an increased resistance to deflection, although much less than expected. The varied stiffness of the metal squeegee may prove beneficial depending on the application.

Results for Mass Reflow of Computer Rel;ated 0.4mm Pitch Devices
Abstract: This report contains the results of the 0.4mm pitch mass reflow parameterization project. The computer related application utilizes one, 0.4mm pitch, 160 pin QFP device per board. The assembly process required that both the top and bottom side of the printed circuit board be mass reflowed. The focus of this project was directed toward the assembly of the 0.4mm pitch device. A printed circuit board assembled in Italy and one assembled in Japan were also evaluated and included in the project. It should be noted that the background of the Japanese and Italian assembled boards is unknown. The project was based on the following testing and material evaluations.

– Printed circuit board construction evaluation.
– Laser cut stencil design and construction evaluation.
– Component lead measurements.
– Solder paste print study.
– Solder paste volume measurements.
– Solder paste reflow profile generation.
– Assembly test.
– Post solder cleaning.
– Visual yield inspection.
– Component lead pulls.
– Printed circuit board\component cross sections.
– Finite element modeling of 0.4mm pitch solder joints.

Squeegee Testing: Scooping and Step Down Deflections
Abstract: In an attempt to better understand the characteristics of typical solder paste stencil squeegees, an investigation into the scooping properties of rubber squeegees was performed. Scooping behavior of the material under typical solder paste printing loads has been simulated using a compressive type test fixture. The rubber durometer squeegees performed expectedly. The harder material produced much less scooping than softer materials under the same loading conditions. For a given displacement, the resulting squeegee load can be plotted as a function of durometer and aperture width.

The data collected for the scooping phenomenon will be used to help optimize parameters of the solder paste printing operation as well as prediction of solder volume of printed pads. The relationship between the squeegee properties and stencil step down areas will allow the estimation of closest feature location to step down walls. Squeegee pressure requirements can also be estimated for stencils with step down regions.

Variable and Response Identification for Screen Printing Machine Parameterization
Abstract: A major step in the mass reflow assembly process is the application of solder paste to the printed circuit board. Stencil printing is the most common method of applying solder paste to printed circuit boards that contain fine pitch devices. The goal of stencil printing is to apply an accurate and repeatable volume of solder paste, to precise locations, on the printed circuit board.

The variables of the screen printing process are numerous and interactive. The four main variables that must be controlled and understood to produce a successful screen printing process are the solder paste, stencil, printed circuit board and the screen printing machine. These four main variables contain many variables.

This test plan will address only the screen printing machine. The primary goal of the screen printing machine parameterization portion of the Ultra Fine Pitch project is to identify, parameterize and optimize all printing variables that are directly associated with the screen printing machine. It should be noted that this document only identifies the variables, parameters and responses of the test. Please refer to individual Test Results reports for the methodology of the experiments.

SIR and ECM Summary Report
Abstract: Surface Insulation Resistance (SIR) and Electrochemical Migration (ECM) resistance testing are techniques utilized to evaluate the immediate and long term electrical resistance as well as dendritic growth characteristics of fabrication materials. The tests are also used to predict the effect of subjecting fabrication materials to various processes used in the manufacture of electronic assemblies.

These tests provide the greatest value in electronic applications where uninterrupted function, and/or long product life is essential or anywhere that ″highly sensitive″ circuits are incorporated. Some manufacturing materials have an electrical resistive value (printed circuit board laminate, solder mask, fluxes and their residue, conformal coating, legend ink etc.) that could adversely affect functional integrity if not suited for the application.

Six Solder Pastes Assembly Yield Experiment
Abstract: This experiment was carried out to evaluate different solder paste characteristics on assembly yield. The assembly yield was quantified with respect to the number of open and shorted component leads. Six different types of solder pastes were evaluated. The characteristics of the solder paste were quantified in terms of post reflow flux residue percentage (percent solids), and the ranking of flux performance which was developed based upon wetting balance test results of the solder paste flux binders. Two sets of components with three different types of attachment pad geometries were tested. The solder paste deposition volume, average height, and the coplanarity data of the leads for each component was also considered. The complete list of independent variables, and their natural levels are presented under Experiment Results in this report.

One main purpose of this study and the UFP project in general is to estimate the occurrence of the defects, for opens and shorts, in conjunction with an estimation of the contribution of the various independent variables. In addition to the independent variables mentioned above, many different independent variables are formulated and evaluated in order to have a valid probabilistic model. Logistic regression technique is used for this purpose. The results of the logistic regression analysis are provided under Experiment Results. The characteristics of the probabilistic model are presented in the Discussion/Procedure portion of this report. The conclusions reached from this analysis are outlined under Conclusions.

Pareto, box, and scatter charts are extensively used to explore the data and help to identify the defect/cause relationships. Many of these charts are presented in the Experiment Results section of this report.

This experiment like many others in this program should be considered a filter experiment and not as a final yield experiment. The stencil used in this experiment was not optimized at every location and therefore defects were expected and found. A final yield experiment was run using optimized conditions obtained from this experiment and will further enhance and quantify these results.

An Experimental Investigation of Solder Paste Bridging
Abstract: This report presents an experimental investigation of solder bridging, observed primarily in the context of the bar print process. The work explored the effects of solder alloy type, component pitch, metallization coating, substrate surface properties, interdigitation, temperature, stenciling parameters and process atmosphere on bridging incidence. While performing the above parametric study, we observed two phenomena that seemed to influence bridging at least as significantly as many of the above parameters. These phenomena will be discussed in depth. They were:

(1) thermally-driven Marangoni motions; and
(2) segregation of the liquid alloy from the flux vehicle liquid with extensive wetting of the flux over the substrate material.

Although solder bridging defects tend to be both process specific and random in occurrence, the bar print process provides a reasonable industrial framework from which to study the problem. The bar print process is a soldering technique that employs a simple stenciling procedure in which solder is deposited in the form of a continuous bar of paste spanning an entire row of pads on a printed circuit board (See Figure 1, Schematic of the Bar Deposition Process and Figure 2, Schematic of the Bar Soldering Process). The bars are deposited through an array of single slot screens with a width, length and thickness that are dictated by the particular technology to be soldered.

After the board is populated with SMT components, it is subjected to a heating cycle during which the solder paste undergoes a phase transformation resulting in a liquid metal bridge instantaneously spanning the entire row of pads. If the process is to be successful, e.g. produce a soldered board which is defect-free, the bridges between adjacent pads must break. Since bridge breaking occurs with the solder in its molten state, the dynamics of this phenomenon are clearly determined by the laws of fluid physics and the discussions presented in this report are developed from this viewpoint.

Conclusions are drawn from results of experiments involving the bar print process and a modification of the bar print process in which only two or more adjacent pads are connected by a solder paste bar. Additionally, where unexpected results were encountered, attempts were made to conduct experiments that would elucidate those results.

Attachment Pad Thickness to Solder Mask Thickness Print Experiment
Abstract: There are five main attributes that are key to a successful solder paste printing process. They are solder paste, screen printer, printing environment, stencil and printed circuit board. Within each main attribute there are many interactive variables.

Printed circuit board construction variables that affect the solder paste printing process are, flatness of the attachment pad, flatness of the printed circuit board, surface texture of the attachment pad and the relationship between the height of the attachment pads relative to the height of the solder mask.

Solder Paste Deposit Definition Inspection Sheet

Test Results for Ultra Fine Pitch Mass Reflow Solder Paste
Abstract: Correct solder paste selection is a fundamental requirement of the mass reflow assembly process for surface mount components. As the component pitch decreases the reliance on high quality solder paste with decreased particle size increases. The flux systems are becoming more varied as the no clean and water clean systems join RMA solvent clean pastes in main stream production environments. This analysis of pastes includes RMA, water clean, and no clean systems and has served as the basis for the selection of the pastes to be used in the UFP study.

“Variable and Response Identification for Solder Paste Parameterization, Printing Related”
Abstract: A major step in the mass reflow assembly process is the application of solder paste to the printed circuit board. Stencil printing is the most common method of applying solder paste to printed circuit boards that contain fine pitch devices. The goal of stencil printing is to apply an accurate and repeatable volume of solder paste, to precise locations, on the printed circuit board. The variables of the screen printing process are numerous and interactive. The four main variables that must be controlled and understood to produce a successful screen printing process is the solder paste, stencil, printed circuit board and the screen printing machine. These four main variables contain many variables.

This test plan will address only solder paste and the printing related variables directly associated with it. The primary goal of the solder paste parameterization portion of the Ultra Fine Pitch project is to identify, parameterize and optimize all printing related variables of solder paste. It should be noted that this document only identifies the variables, parameters and responses of the test. Please refer to individual Test Results report for the methodology of the experiments.

Solder Paste Printing Experiment Check List

Solder Paste Stencil Evaluation Visual Inspection
Abstract: This document describes the procedure and methods for the evaluation of solder paste stencils by visual inspection.

Solder Paste Volume Calculation
Abstract: This document describes the procedure and mathematics supporting the calculation of solder paste volume deposited by stencil squeegees.

Solder Stencil Squeegee Deflection Testing
Abstract: This document describes the procedure followed to test the deflective properties of solder stencil squeegees.

Solderability Testing and Evaluation by Wetting Balance Analysis
Abstract: The increased use of Surface Mount Technology (SMT) in Printed Circuit Board (PCB) assembly is a result of the need for reduced weight, volume and cost in electronic packaging. While traditional electronic (through hole) PCB assembly technology has essentially reached its limits, SMT has offered a possible solution to achieve the needs of the market with respect to cost, weight, volume and reliability.

Solderability is one of the most important factors to be considered during the conversion from conventional through hole technology to surface mount PCB assembly. Conventional technology inserts components through holes in the board and the clinched leads provide high mechanical strength. In contrast, surface mount electronic components are mounted on the surface of PCBs. Even though solderability is important for conventional technology, the solderability requirement of components and boards becomes even more critical for surface mount PCB assembly because the solder joint needs to provide both the electrical and mechanical connection. Solderability evaluation for surface mount components and the identification and study of factors that influence it are of primary importance in the surface mount manufacture of PCBs.

The wetting balance test method is performed as a quantitative investigation of solderability. This research interprets issues that concern wetting balance solderability evaluation and the generation of a validated test procedure for various types of surface mount components. This research provides a method for quantizing flux activity through the use of the wetting balance test procedure. The results obtained can serve as a bench mark for the candidate fluxes that might be considered for a manufacturing process.

“Squeegee Drag Friction Testing, Process and Testing Specifications”
Abstract: This document describes the procedure followed to test the frictional drag forces resulting from the movement of solder stencil squeegees across a typical stencil surface.

“Squeegee Scooping Deflection Testing, Process and Testing Specifications”
Abstract: This document describes the procedure followed to test the amount of stencil squeegee scooping into fine pitch apertures.

The Influence of Steam Aging Solderable Surfaces and the Hot Bar Bonding Process
Abstract: The primary objective of this research (experiment two) was to study the effect of oxidation on solderable surfaces (attachment pads and component leads) generated from “steam” aging. Solder joint attributes were evaluated. Two component lead plating materials (SnPb and Au), two no clean fluxes, and two attachment pad solder thicknesses (300 and 600 micro inches) were used. The boards and components were aged in a Thermotron environmental chamber for eight hours. After the components were bonded, the peel (tensile) test was used to evaluate solder joint strength. The data collected was analyzed using a statistical software package (E-Chip).

Stencil Aperture Taper and Radius Print Experiment
Abstract: The Stencil Aperture Taper & Radius Experiment is designed to investigate effects on the solder paste printing process when using different stencil aperture geometries. The variations in the stencil aperture geometry are created by various taper and radius levels for a set of pre-selected devices. The devices range between 0.5 mm and 0.3 mm pitch.

In this Study, the devices with 0.3 mm pitch are 100 pin components. Similarly, 0.4 mm and 0.5 mm pitch devices are called 296 pin and 208 pin devices respectively. Two stencils are designed to investigate the effect of opposite taper geometries. Three taper and slot corner radius levels are evaluated for each of these stencils. The levels of these factors and the design matrices for the stencils are given under the Results section of this report.

The response variables used to derive the conclusions are the measured volume of the solder paste deposit, the measured average height of the solder paste deposit as well as the percent volume and percent average height of the solder paste deposits. Multiple linear regression was the primary statistical analysis tool applied.

The response surfaces are also primarily used to visualize the complex relationships between variables. The residual analysis is carried out to check the model adequacies. The response surface plots and regression analysis results are presented under Results. The model adequacy checking and the other technical information regarding the experiment is presented under Discussion and Procedure.

The conclusions reached from the statistical analysis are presented as section three of this report.

Universal Instruments Corp. would like to thank IRI for supplying the stencils and chem etching the steps.

Universal Instruments would also like to thank Lumonics, Ltd for laser cutting the stencils.

Stencil Material Optimization: Olin Brass Stencil Material
Abstract: One of the most critical parameters in the application of solder paste to printed circuit board attachment pads, is the physical characteristics of the stencil material. An improperly selected stencil material can greatly contribute to premature stencil failure. A variety of defects including coining, excessive wear, corner rounding, aperture material failures and others can all be attributed to improper stencil material selection.

In an effort to optimize stencil materials for the application of solder paste to ultra fine pitch surface mount devices, a joint development program was undertaken with Olin Brass and UIC-UFP consortium. The main objective of this collaboration is the development, testing, and implementation of superior stencil materials for future industry wide use.

Material selection includes considerations for both chemically etched and laser cut stencil applications. Print experiments were performed on all materials provided. Three dimensional laser profilometry, scanning electron microscopy and cross section metallography were used to determine the suitability of the Olin Brass materials for implementation to solder paste stencil applications.

Test Results of Step Stencil Solder Paste Print Experiment
Abstract: The development of a solder paste printing process that utilizes a “step” stencil design requires the stencil design consideration of two variables not involved when designing single thickness stencils. These two variables are “step” thickness change (the amount of stencil thickness change from one level of the stencil to the next level of the stencil) and the distance from the edge of the step window to the stencil aperture openings within the step window. The use of step stencils increases the complexity of stencil design and constrains the type of squeegee material that can be used for the solder paste printing process.

Influence of Substrate Thickness on the Hot Bar Bonding Process
Abstract: This research studied the influence of substrate thickness on the formation and strength of solder joints from the hot bar bonding process. The thermal influence from variation of the substrate thickness was first investigated using temperature sensitive paints. The color of the paints change when they reach specific temperatures. In addition to the thermal dissipation, the thermode temperature profiles resulting from different substrate thicknesses were also studied. Components were bonded to substrates of different thicknesses. Also, the peel test was used to evaluate the difference in solder joint strength (tensile strength).

Surface Insulation Resistance Testing
Abstract: The objective of this research project is to systematically evaluate the SIR (surface insulation resistance) of processed fluxes used for the solder attachment of ultra fine pitch surface mount components to printed circuit boards. This study will be used in conjunction with other evaluation experiments to analyze, screen, and choose fluxes for subsequent use in the UFP research project. Eighteen no-clean fluxes and six water-soluble fluxes have been tested based on the Bellcore SIR test specification. The SIR values of three water clean fluxes were studied after cleaning with R.O. (Reverse Osmosis) water versus cleaning with D.I. (De-ionized) water.

Evaluation and Comparison of TAB Componentry
Abstract: Surface mount hot bar attachment is a highly complex process involving a vast number of critical parameters each contributing to the success or failure of printed circuit board assembly. One key group of variables is associated with the surface mount hot bar component. Plating type and thickness, lead material and shape as well as a great number of other factors, drastically affect final assembly yields and long term product reliability. Factors such as component lead solderability can be measured and parameterized preceding assembly and then used to correlate assembly defects with a given solderability level. Other issues such as plating type can be examined in a similar fashion. By examining and parameterizing each type of component used in the ultra fine pitch project, predictions can be made for the final assembly yields of fine pitch surface mount attachment process.

Test Plan for UFP Mass Reflow and TAB Components
Abstract: The Ultra Fine Pitch project at Universal Instruments encompasses many facets of surface mount technology. The evaluation of the fine pitch component is an important segment of the UFP program. The overall quality and manufacturability of an SMT assembly is dramatically affected by the metallurgical and mechanical properties of the component. The goal of this test sequence is to identify the critical parameters that are required to solder ultra fine pitch components.

Evaluation and Comparison of Mass Reflow UFP Componentry
Abstract: Surface mount attachment is a highly complex process involving a vast number of critical parameters each contributing to the success or failure of printed circuit board assembly. One key group of variables is associated with the surface mount component. Plating type and thickness, lead material and shape as well as a great number of other factors, drastically affect final assembly yields and long term product reliability.

Factors such as component lead coplanarity can be measured and parameterized preceding assembly and then be used to correlate assembly defects with a given coplanarity level. Other defects such as solderability problems can be examined in such a fashion. By examining and parameterizing each type of component used in the ultra fine pitch project, predictions can be made for the final assembly yields of fine pitch surface mount attachment process.

Ultra Fine Pitch Stencil Inspection Criteria
Abstract: The need for accurate and repeatable fine pitch solder paste stencils is greatly increased as ultra fine pitch mass reflow processes are being developed and implemented. Solder paste stencil performance is highly dependent on the materials and process used to create the desired solder paste stencil. Both additive and subtractive technology can produce usable solder paste stencils but extreme care must be taken to parameterize the characteristics of each.

Test Results of UFP-II Solder Paste Print Experiment
Abstract: UFPII print study is one of the series of experimental studies conducted at the SMT lab. The purpose of this study is to gain a better understanding of the solder paste stencil printing process for ultra fine pitch components as well as for other standard surface mount components. The objective of this experiment is to determine the effects of squeegee durometer and print gaps for 0.3mm, 0.4mm, and 0.5mm pitch components. This experiment is also suitable to study the effects of print direction (front squeegee vs back squeegee), aperture direction (vertical vs horizontal apertures), clean vs unclean stencil aperture, and run numbers (board number) on the response variables.

The response variables measured are the percent solder deposit volume (percent fill of the aperture), percent average height (percent height of stencil thickness), and observed area. The percent volume and average height are the primary response variables. The observed area is used as a secondary variable, mostly for confirmation purposes.

The technical information related to the solder paste, stencil, PCB, printer settings, and the conditions under which the printing process is conducted, is presented under Discussion/Procedure section of this report. The experimental and statistical results are presented under UFP II Print Study Results. The explanations are summarized Conclusions. Some statistical issues regarding the models employed, and a comparison between the theoretical scooping model and experimental model is also presented under Discussion/Procedure.

The primary statistical analysis tool was the multiple linear regression technique. The Means Comparisons and ANOVA procedures were also used. Box Plots were widely used to obtain visual presentations of the physical processes.

UFP-II Assembly Yield Experiment (Attachment Pad Length)
Abstract: The purpose of this experiment is to determine how attachment pad length effects the assembly process of 0.5, 0.4, and 0.3mm pitch mass reflow devices and help define optimum attachment pad lengths. The experiment also included the use of two different solder paste print processes using 80 & 92 durometer squeegees. The two processes will help determine the relationship between solder paste deposit volume\height and the yield generated from the two different processes. The response variables for the experiment will be assembly yield (shorts & opens) and solder joint volume, shape and appearance. This experiment will also identify process steps that will be focused on for future experiments. UFP Test Results #0021 contains the results from the UFPII solder paste print experiment which contained stencil aperture length as one of the variables. The effect of stencil aperture length relative to the solder paste print process is defined in this document.

UFP-III Attachment Pad Finish Assembly Experiment (Part II)
Abstract: In this experiment, the effect of different board finishes on the assembly yield is evaluated. Three types of board finish used are Gold, Copper, and Tin/Lead. The same type of solder paste was used to assemble all the boards. The assembly yield defects observed are solder shorts and open joints. The percent of solder volume transferred through the stencil aperture for each lead is also used as a variable, along with other variables to estimate the occurrence of assembly defects.

The purpose of this experiment is to identify attachment pad finish types that are acceptable for Ultra Fine Pitch Assembly by establishing the link between defects and attachment pad finishes so that the further assembly yield experiments can be focused on the best attachment pad finish. In Results, the possible causes of the defects are evaluated. The Logistic regression analysis is carried out for this purpose and the result of this analysis is presented under Results. Pareto charts are extensively used to illustrate the relationships between significant variables and defects.

Conclusions contains the conclusion reached from this experiment. The logistic regression model characteristics and the process related issues are presented under Discussion/Procedure.

UFP-III Attachment Pad Finish Assembly Experiment (Part I)
Abstract: This report contains the results from a process assembly experiment that was performed on UFP board #III. The experiment consisted of assembling UFP #III boards, manufactured with five different attachment pad coatings. Two boards, from each of the five board finishes were assembled using a low activity (R flux) no clean, nitrogen reflowable solder paste. All boards were assembled using exactly the same process for each board finish. A second experiment was conducted on a pre-solder coated board using the same process with the exception of replacing the solder paste with a no clean (RMA) tacky flux. Only one board was available for assembly using the tacky flux process.

Final Mass Reflow Assembly Yield Experiments
Abstract: The objective of the Ultra Fine Pitch Mass Reflow Project was to identify the significant variables used in the assembly processes of Ultra Fine Pitch devices, determine the optimum settings of the variables for each process, perform assembly experiments to determine the quality and assembly yields from each process, and todocument the findings. The objective of this experiment is to determine the assembly yield and quality of the Ultra Fine Pitch Mass Reflow process. Six major attributes of Ultra Fine Pitch Mass Reflow Assembly, solder paste, stencils, screen printer process parameters, printed circuit boards, component placement and solder paste reflow were studied. Many other Ultra Fine Pitch attributes were studied during the project, but they did not directly affect the mass reflow assembly process.

UIC 4682B Fine Pitch Bonder and Thermode Characterization
Abstract: This report depicts the evaluation procedures and test apparatus, along with the resulting data and conclusions from the characterization of the UIC 4682B Fine Pitch Bonder and associated thermode assemblies. The purpose of the characterization is to identify the performance attributes of machine variables for subsequent experimentation in support of the Hot Bar Attachment portion of the Ultra Fine Pitch Attachment Study. The machine characterization included factors such as thermode force, force of the component pick-up / placement spindle, and compliance characteristics of the thermode suspension system. The thermal performance of various thermode assemblies as well as a comparison of molybdenum versus titanium alloy thermodes is discussed.

Test Results of UIC Demonstration Board Print Experiment
Abstract: The Show Board print study is one of the series of experimental studies conducted in the SMT lab. The purpose of this study is two fold. The first is to gain a better understanding of the solder paste stenciling process for a broad group of component types. The second is to identify the optimum conditions / regions in that window of interest. The window of interest consists of various components ranging from a 208 pin, 0.5mm pitch PQFP to 12/10 devices. The print process variables that will be researched include print gap, print speed, squeegee types, and squeegee pressure. The aperture directions are also included in the analysis of the experiment. A brief explanation of the Design Matrices for this planned experiment is given under Results.

The response variables measured are percent solder deposit volume, percent average height, and observed area. The percent volume and percent average height are the primary response variables used for the statistical inference.

The technical information related to the solder paste, stencil, and PCB board used, as well as printer settings and the conditions under which the printing process was conducted are presented in the Discussion/Procedure of this report. The experimental and statistical results are presented under Results. The Conclusions are summarized. Some statistical issues regarding the models employed are also presented with the Discussion/Procedure.

The primary statistical analysis tool used is the multiple linear regression technique. Box Plots, Contour Plots and Response Surface Plots are widely used to realize the objectives of this experiment.

Validation and Testing of Solder Paste Evaluation Procedures
Abstract: Solder paste affects almost every facet of the reflowed PCB assembly process. The proposed research would systematically determine the validity of test procedures used to evaluate solder pastes.