Authored By:
Jimmy Chow, Heather McCormick, Craig Hamilton, Mike Berry, Roden Cortero and Gianni Facchini
Celestica Inc., Toronto, Ontario, Canada
Summary
Handheld portable products such as smartphones are trending toward smaller form factors while simultaneously increasing in functionality to keep up with consumer demands. This is achieved in part by decreasing the size of components and increasing the density of the circuitry. These unique product needs drive different Design for Manufacturing (DFM) recommendations than those that are in use for larger products - while for larger products, reworkability is paramount, for handheld portable products, high first pass yields and fitting the required functionality into an appropriate form factor are of greater concern in many cases.
This paper summarizes a new test vehicle designed to emulate a next-generation smartphone product. One of the goals of this project was to study the effect of pad design and component spacing on assembly yield. The test vehicle includes a representative range of component types including 01005 and 0201 discretes, 0.3mm pitch CSPs, Package-On-Package, QFNs, and RF shields. For selected components, different pad designs were included on the board, allowing a direct comparison of the various options and recommendations for the optimal pad designs.
In addition, a range of component to component spacings were used on the board, ranging from spacings in common use in today‟s products to extremely aggressive spacings that push the limits of the PCB manufacturers. The test vehicles were inspected after assembly, and yields were determined for the various component to component spacings studied to determine what the limitations are and to update DFM rules specific to the needs of extremely dense handheld portable products.
The results of the yield study will be presented along with the analysis of the implications for the DFM rules.
Conclusions
The assembly of the SPTV boards provided an opportunity to study many different process issues for manufacturing leading edge components found in handheld portables. We can establish the following conclusions based on data that has been gathered:
For 01005s, elliptical solder mask defined pads yielded the best results from the variations studied. To place 01005s, there are many critical control requirements such as dust control, exact tolerances on printed circuit boards and stencil apertures. Automated equipment to inspect for the presence of 01005 paste deposits and component bodies must have precise program algorithms to validate such minute features.
For 0201s, Spacing "B" and larger resulted in zero bridging defects in both SMD and NSMD land patterns. Bridging defects were found in the 0201s with the clearance of Spacing "A". 0201 billboarding defects found on this assembly were determined to be caused by pick-up errors during placement.
For 0.3mm pitch CSPs, bridging defects were found on all variations reflowed in air. The large pad variations (in both NSMD and SMD) had less bridging defects than the small pad variations. In nitrogen reflow, no bridging defects were found on the large pad variations.
For discrete to shield, discrete to CSP, shield to shield, CSP to CSP, and CSP to shield component spacing, the lowest level spacing had zero defects.
Initially Published in the IPC Proceedings
|
