William E. Coleman Ph.D.
Photo Stencil, Colorado Springs, CO USA
Transcript
Many of the latest SMT assemblies for hand held devices like cell phones present a challenge to process and manufacturing engineers with the introduction of miniature components such as .3 millimeter CSP and micro BGA devices, as well as microchip component devices.
Printing these miniature devices along with more conventional SMT devices like .5mm QFPs and midsize passives, in addition to RF shields is a challenge. Whereas a 4mil or 5 mil thick stencil provides good paste transfer for the normal SMT devices, stencils with this thickness have very low area ratios for the miniature devices.
This paper examines stencil technologies (including laser and electroform), aperture wall coatings (including Nickel-Teflon coatings and Nano-coatings), and how these parameters influence paste transfer for miniature devices with area ratios less than the standard recommended lower limit.
A matrix of print tests is utilized to compare paste transfer and measure the effectiveness of different stencil configurations.
Summary
Many of the latest SMT assemblies for hand held devices like cell phones present a challenge to process and manufacturing engineers with the introduction of miniature components such as .3 mm CSP and uBGA devices as well as 0201 and 01005 chip component devices.
Printing these miniature devices along with more conventional SMT devices like .5mm QFP's and 0603 and 0805 passives, in addition to RF shields is a challenge. Whereas a 4mil (100 micron) or 5 mil (125 micron) thick stencil provides good paste transfer for the normal SMT devices, stencils with this thickness have very low Area Ratios for the miniature devices. For example a .3mm CSP with a 7.5 mil (190 micron) has a .47 Area Ratio for a 4 mil thick stencil.
This paper will examine stencil technologies (including Laser and Electroform), Aperture Wall coatings (including Nickel-Teflon coatings and Nano-coatings), and how these parameters influence paste transfer for miniature devices with Area Ratios less than the standard recommended lower limit of .5. A matrix of print tests will be utilized to compare paste transfer and measure the effectiveness of the different stencil configurations. Area Ratios ranging from .32 to .68 will be investigated.
Conclusions
Goal of this study: Determine if Special Coatings can improve Paste transfer for Apertures with Area Ratio's less than .5. The Electroform Stencil with Nano-Coat was the only stencil tested able to achieve this goal. This stencil is a good candidate when small (01005 and/or .3mm pitch CSP components are coexistent on the same PCB. Below is a overall summary of the results:
Electroform with Nano-Coat: At .375 Area Ratio this stencil was close to being acceptable, having 77% transfer and 17% Std. Dev. Much better than the other 4 stencils. At .500 Area Ratio an 89% transfer and 8.2% Std. Dev. was achieved. Acceptable down to .42 Area Ratio.
Electroform with no coating: At .500 Area Ratio 83.5% transfer and 8.8% Std. Dev. was achieved, better than the remaining 3 stencils. Acceptable down to .50 Area Ratio.
Electroform and High Precision Chem-Etch stencils with Nickel Teflon coating performed about the same: At .625 Area Ratio both had transfer of about 86% and Std. Dev. of about 8%. Acceptable down to .60 Area Ratio. It was disappointing and surprising that the Electroform Stencil with Nickel-Teflon coating gave lower paste transfer performance than Electroform stencil without any coatings. It was impressive that the High Precision Chem-Etch stencil providing acceptable paste transfer down to a .60 Area Ratio.
Initially Published in the IPC Proceedings
Comments
Good article but there were a couple omissions that would be needed to make a fully informed decision.
One variable omitted was the paste sphere size. Type 3, 4, 5 or 6. My experience shows that the sphere size plays one of the larger roles when it comes to aperture release as aperture size decreases.
I might also question the dismissal of the laser cut stencil. I have used laser cut stencils that outperform E-form. The laser cut stencil needs to be a fine grain alloy, cut with a diode pumped fiber laser and a nano-coating applied.
I will use the article as a guideline for testing but not as proof that E-Form outperforms all other stencils.
