AXI and Lead-Free Process Characterization
Lead-free solders and processes are introducing new defects to PCB manufacturing. The measurement capabilities of automated X-ray inspection (AXI) are ideally suited to this changing environment. AXI technology is an excellent tool for characterizing lead-free processes and materials, allowing you to optimize the lead-free process and compensate as much as possible for shifting solder properties and reduced margins.
Where We've Been, Where We're Going
Tin/lead (Sn/Pb) solder has been the dominant process in electronics manufacturing for a variety of reasons:
- It conducts electricity well.
- It's strong enough to reliably bond components to PC boards yet soft enough to withstand mechanical and thermal stresses without cracking or breaking the parts or boards.
- Its melting point is high enough that products won’t come unsoldered in use or in storage, and low enough that components and boards aren’t damaged during processing.
Tin/lead solders also have the following desirable characteristics related to strong wetting forces. These characteristics reduce defects in production manufacturing environments:
- Misaligned parts tend to self-align.
- Solder wicks up and makes good contact with many bent leads.
- Pads with insufficient paste still form acceptable joints.
- Many solder bridges clear themselves.
In the lead-free world, new tin-silver-copper (SAC) solders are quickly gaining prominence as a popular alternative to tin-lead solder. However, SAC solder has lower wetting forces caused by surface tension, which provides some process-oriented disadvantages compared to tin-lead. With SAC solder, misaligned parts tend not to straighten out as well; solder wick-up is less pronounced, so SAC solder is less likely to make good contact with bent leads; pads with insufficient paste are less likely to form acceptable joints; and fewer solder bridges clear themselves.
Even with a well-characterized lead-free process, defect rates can be expected to be higher with SAC solder compared to a comparable Sn/Pb process. To optimize the lead-free process and compensate as much as possible for these reduced margins, the lead-free solder process must be fully characterized.
Making the transition
Full characterization of lead-free processes and solders is a prerequisite to a successful lead-free transition. Here are the process and material changes that can be characterized with SAC alloys:
- Reflow Temperature Profile: The reflow temperature of most SAC lead-free solders (for Sn-Ag-Cu: tin-silver-copper) is about 217 degrees C. This higher temperature means that the thermal profiles of reflow ovens must be recharacterized to obtain the proper heating and cooling rates for the components and boards in order to assure low defect rates while not damaging the assemblies. This is usually a trial-and-error process.
- Solder Paste Formulation: Higher temperatures are incompatible with some of the chemicals used in some solder pastes. Among other problems, flux components can break down and solvents can cause voids in the finished solder joints.
- PC Board Finish: A standard PCB finish is tin-lead HASL (hot air solder leveling). New lead-free HASL processes are only now under development. HASL has the disadvantage of not providing planar pads which are desirable for fine-pitch components. Most manufacturers are moving to HASL alternatives such as bare copper with OSP (organic solderability protection), electroless nickel/immersion gold, electroless tin, and immersion silver. Each new metallurgy adds another layer of process variability that must be characterized.
A multivariable experiment
An experiment with a limited number of carefully selected variables can be an effective way to characterize a lead-free soldering process. For example, an experiment can be designed with three different reflow profiles, three brands or formulations of solder paste, and two different board finishes. This provides 18 (3x3x2) combinations or cells which can be evaluated in a single experiment. This is a fairly large number of combinations for a single experiment, but much can be learned at one time with this type of multivariable experiment.
Commonly, attributes are used as a measure of merit for experiments: Good vs. Bad, Go vs. No Go, etc. Defects typically occur in assembly processes at a rate of only a few hundred defects per million opportunities (DPMO.) A very large number of boards would need to be assembled in the experiment to generate a sufficient quantity of defects to be statistically meaningful. Running this large number of boards under well-controlled conditions in a production environment is costly and often impractical. It can take several days of production to generate enough defects, and the chance of introducing other variables such as machine set-up and differences between shifts is very high.
One method of reducing the number of boards required in an experiment is to use a continuous variable (e.g. solder thickness) as a measure of merit instead of an attribute. Generally, the wetting performance of the solder in the solder joints is a good indicator of the quality of the process. Better wetting means fewer defects. In a soldering experiment such as this, a method of generating continuous variable data that indicates the wetting performance of the process is needed.
Using AXI for process characterization
AXI can provide information on solder features that reflect the wetting performance of the process. Measurements such as solder fillet thicknesses, solder joint fillet lengths, and other measurements on selected components can provide this data. These indirect measurements of wetting performance will correlate to the performance of each variable. Experiments similar to this for process characterization have been successfully run with a limited number of boards in each cell, allowing all of the boards in the entire experiment to be built in a few hours instead of over several days.
To learn more about how Agilent's AXI technology can ease your transition to lead-free manufacturing, please contact your Agilent representative.