|
Technological advancement of component and PCB technology from through-hole to surface mount (SMT) is a major factor in the miniaturization of today’s electronics. Smaller and smaller component sizes and more densely packed PCBs lead to more powerful designs in much smaller product packages. With advancement, however, comes a new set of challenges in building these smaller, more complex assemblies. This is the challenge original equipment manufacturers (OEMs) and contract manufacturers (CMs) face today. One of these challenges is the stencil printing process. Many of today’s designs incorporate a mix of miniature and much larger components. Manufacturing engineers are faced with the dilemma of choosing a thinner stencil foil to ensure solder paste release for the miniature components or a thicker foil to ensure sufficient solder volume for the larger components. With a standard laser-cut stencil using 300 series stainless steel, one would have to make that difficult choice. An electroformed stencil gives more options in balancing release for miniature components and volume for larger ones due to its ability to successfully print smaller components without reducing the foil thickness. However, many have difficulty justifying the 3-4 times cost increase and added schedule delay for an electroformed stencil, especially with more and more companies moving to a low volume, high mix array of jobs. Faced with these two options, is electroformed technology the right solution or have technological advancements allowed new developments in stencil technology? Developments in stencil laser technology
Stencil laser technology has seen continuous advancement over the past 10 years. The majority of advancement has been in linear motor technology, leading to improvements in the cutting speed of stencil lasers. Until recently, the source of the laser beam has remained the same with reliance on lamp pumped technology. The lamp pumped technology is comprised of flash lamps, YAG rods, mirrors, and focal lenses. With this technology, the smallest diameter laser beam possible was approximately 40μm. While this diameter beam is fine for the majority of stencil designs, the energy density with a 40μm beam diameter is not high enough to produce the smoothest aperture wallswhen cutting stencil apertures for miniature components. In the past two years, there has been a major leap forward in laser technology. The most significant development is the introduction of the single mode CW Ytterbium fiber laser (“fiber laser” for short). The new fiber lasers produce shorter pulse widths, higher frequencies, and have a fully programmable pulse/pause ratio. In addition, they produce a smaller laser beam diameter of 19 micron with a corresponding 4x increase in energy density. The 4x increase in energy density significantly increases the laser beam’s ability to cut through the metal and the result is amuch smoother aperture wall (see figure 1). 
Developments in stencil material technology
Along with advancements in laser technology, there also have been advancements in stencil material technology. For many years, laser-cut stencils used either 300 series stainless steel or a higher nickel alloy (Invar alloy 36, alloy 42) for the stencil foil material. These are good solutions for the majority of assemblies, but their paste release performance reduces considerably when printing apertures with surface area ratios below 0.66. As a result, one would have to either increase, “overprint,” the aperture sizes when selecting a thicker foil or reduce the foil thickness for acceptableprints. Overprinting miniature components, however, is not always a guaranteed solution since the crucial surface area in the surface area ratio formula is the common metallic surface area between the SMT pad and the stencil aperture. If a PCB has a CSP component with a 0.010” diameter pad and the stencil overprints with a 0.012” diameter aperture, the common metallic surface is still limited to the 0.010” diameter of the SMT pad. The additional paste beyond the 0.010” limit of the SMT pad is not in contact with the metallic surface and, therefore, does not contribute to pulling the paste from the stencil. Advancements in stencil material technology include new stencil materials specifically designed for stencil lasercutting such as Fine Grain material which has a much finer grain structure (see figure 2) when compared to standard 300 series stainless steel and alloys, and contain smaller and fewer voids in the material. With smaller and fewer voids, the solder paste does not adhere as easily to the stencil walls. This is primarily due to the micro size of the voids (in some cases smaller than the particle sizes in the solder paste) that makes it more difficult for the solder paste particles to get a grip on the stencil walls. When the solder paste is pulled from the stencil as the PCB drops, release is easier and less paste residue is retained in the stencil. The easier release allows for the printing of smaller stencil apertures, without reductions in foil thickness, and the reduction in paste residue allows for an increase in the number of prints before having to clean the stencil. 
In addition to improved paste release for smaller apertures and much cleaner paste release throughout the entire stencil, the finer grain structure of these new materials also produces a more defined aperture edge (see figure 3) when cut with a properly tuned laser beam. As the aperture size decreases, the importance of repeatable and accurate solder paste release rises. With miniature components, small fluctuations in solder volume have a much larger impact on solder joint reliability due to the minimal solder volume required. A more defined aperture edge, along with improved paste release, leads to more repeatable and accurate solder paste release. 
The new materials are a stainless steel composition and are rolled so thickness tolerances are extremely tight. They also have improved thermal conductivity as well as similar mechanical and corrosion resistant properties when compared to standard 300 series stainless steel. Stencil life and durability are similar to standard 300 series stainless steel stencils. Performance
Technological developments in component and PCB design are beginning to outpace current stencil technology. Do these significant advancements in stencil laser and material technologies provide the current and future solutions the electronics assembly industry requires? That question is best answered through a design of experiments (DOE) comparing the new laser and material technologies with the standard stencil technologies available today. • Solder paste volume
Results in figure 4 illustrate the print performanceof the various stencil technologies over the entire range of solder paste types tested. All were laser-cut on thenew LPKF Multicut fiber laser, except the electroformedstencil which utilized traditional electroform technology.The electroformed stencil was the performance baselinewith acceptable paste volume percentage at a surface arearatio of 0.5. The laser-cut electroformed nickel sheet hadacceptable paste volume percentage down to 0.45, but itsprint performance quickly flattened out compared to theFine Grain and electroformed stencils. The Fine Grain stencilhad acceptable paste volume percentage at 0.45 and itsprint performance continued to outperform electroformedas the surface area ratio increased. The performance increasedown to a surface area ratio of 0.45 allows the printing of evensmaller components without a corresponding reduction inthe stencil foil thickness. The result is additional solderpaste volume for the non-miniature components, resultingin less rework and improved solder joint reliability. 
• Aperture registration
As component pad sizes continue to decrease, alignment accuracy between the stencil and PCB is becoming more critical. The adhesion of the solder paste to the SMT pad is the sole force involved in pulling the paste from the stencil. Since PCBs will tend to shrink during the manufacturing process, SMT pad locations tend to be slightly short of their expected locations. Long PCBs, of course, will havesignificantly greater shrinkage than short PCBs. In addition to PCB shrinkage, the electroformed stencil process uses Mylar film to create the stencil image. The film is dimensionally unstable due to its susceptibility to temperature and humidity fluctuations. Without tight temperature and humidity controls in the manufacturing area, shifts in aperture locations can occur during plotting of the Mylar film and during its use. Since the electroform process only produces the electroformed foil, it typically has to be mounted into a stencil frame. During the electroform process, no tension is applied to the electroformed foil. When mounted into a stencil frame, tension is applied to electroformed foil by the stencil frame’s polyester mesh. This tension pulls on the foil causing slight shifts in the locations of the stencil apertures. In most cases, the electroformed stencil aperture locations will be long, or further away from their expected locations. If the PCB has SMT pad locations that are short of expected locations and the electroformed stencil has aperture locations further away than expected, there can be a significant shift, or misalignment, between the stencil apertures and PCB pads. A shift between the stencil aperture and PCB pad reduces the amount of solder paste in contact with the surface of the PCB pad. This lowers the adhesive force between the solder paste and PCB pad, effectively reducing the ability of the board to pull the paste from the stencil. Miniature components already have very low surface area ratios. The lower the surface area ratio, the more critical the alignment between the stencil aperture and PCB pad. The Fine Grain stencil in this DOE was cut in the frame on the new LPKF high power, short pulse fiber laser. The intent was to minimize stencil aperture registration errors, thereby increasing the alignment accuracy between the stencil and PCB. The results (27 position errors for the Fine Grain stencil and 2,307 position errors for the electroformed stencil) in figure 5 show a marked improvement in apertureregistration when compared to an electroformed stencil. Conclusion
As advancements continue in component and PCB technologies, will the stencil technology of today provide current and future solutions to the challenging assembly issues faced by OEMs and CMs? Is electroformed technologythe right solution or have new developments in stencil laser and material technologies caught up with and surpassedthe electroformed technology of today? The answer to these important questions is in our view an unequivocal “yes.” Stencil laser and material technologies have advanced to the point where laser-cut stencil performance is beyond that of current electroformed technology. Using the new LPKF high power, short pulse fiber laser technology and the new fine grain material, stencil performance is significantly improved over electroformed, especially when printing miniature components. Improvements in stencil laser and material technologies have lead to significant improvements in solder paste release down to a surface area ratio of 0.45 as well as improved aperture registration accuracy. These improvements are critical to meeting future requirements when printing miniature components like 01005s. The technology summary is shown in table 1. 
At a cost savings of 30-50 percent compared to electroformed, the ability to produce multi-thickness (step) stencils, and the option of same day turn times, Fine Grain stencils, cut with the new fiber lasers, are a marked improvement compared to the high performance stencil solutions available today. OEMs and CMs can get the performance they need while reducing costs and meeting critical delivery schedules. The new stencil laser and material technologies available today give stencil manufacturers the tools and materials needed to supply an ever-changing industry formany years to come. About the authors
Robert F Dervaes is VP Technology and Engineering at Fine Line Stencil; Jeff Poulos isVP Manufacturing and Sales at Alternative Solutions; and Scott Williams is Product/Account Manager at Ed Fagan Inc. Acknowledgement
The authors would like to thank Stephan Schmidt and Sebastian Gerberding of LPKFLaser Electronics (www.lpkfusa.com) for their contribution to this article. | 
