Copyright © 2006 Penton Media, Inc., All rights reserved.
Printing of this document is for personal use only.
Reprints

Compactness and efficiency. Greater functionality at lower costs. These features—mainstays of the component world—are possible thanks to a steady stream of improved materials and manufacturing methods. Displays for large-screen TVs, outdoor signs, and high-definition TV (HDTV) home theaters have set record diagonal sizes, a trend that isn't slowing. Cell phones and other handheld devices are experiencing similarly significant display developments. Yet as prices drop to all-time lows at both ends of the scale, we've seen performance levels (brightness, resolution, clarity, color quality, and power dissipation) jump for all sizes.

Sensors, mostly of the microelectromechnical-system (MEMS) variety, continue their march into mass-market applications like cell phones, consumer electronics, laptops and notebooks, and digital cameras. Key drivers have been CMOS and CMOS-like manufacturing processes, better design tools, higher manufacturing yields, and improved manufacturing standards.

Yet MEMS sensors still lack actuators like switches, relays, valves, and pumps. These functions are inherent to MEMS, a technology borne of its useful mechanical silicon properties. Microfluidics has shown some promising improvements as MEMS actuators, witnessed by a few announcements last year. But manufacturing approaches need development, and there must be higher yields before it can make a significant impact.

Once such things come to pass (and that may be soon), a complete closed-control system-on-a-chip (SoC) will become a reality. The same chip that senses input variables, whether in the ambient environment or within a circuit, will be able to process that signal, analyze it, and deliver a decision-making signal to an actuator.

The greatest sensor gains were made in imaging. Lower-cost, denser, higher-resolution, brighter, and better color image sensors are constantly evolving. Adoption of CMOS manufacturing technology and clever pixel layouts have been crucial.

CMOS image sensors will take over more applications from charge-coupled device (CCD) image sensors, which have had exclusivity in the high-performance realm. Though they still cost more than CMOS image sensors—a price gap that's narrowing—CCD imagers will continue to dominate niche high-performance markets.

Traditional Components Go To Chip Form
Traditional resistor, capacitor, and inductor components continue the trend toward chip form. They're available in small sizes that can handle highest-ever power levels and span the widest range of values.

Advances in metal-electrode face (MELF) thin-film and MEMS technologies have been key enablers. Most discrete components are now available in surface-mount technology (SMT) form for compatibility with modern wave-soldering equipment and to comply with the European Union's Restrictions on Hazardous Substances (RoHS) lead-free initiative.

Calls for denser packages with smaller profiles are creating a need for 3D packages. Various chip- and board-level interconnect schemes are competing for the 3D space, with the hottest two being package on package (PoP) and wafer-level packaging (WLP).

Tighter packing densities of higher-performance ICs creates the inevitable problem of greater heat-dissipation levels. In other words, what do you do with them and how do you remove the heat away from the hot spots on a chip, board, or system? The industry is working on materials that provide better heat transfers, improving heatsinking, and using liquid cooling methods.

High-Speed Interconnects Key To System Scalability
Board- and system-level interconnects such as PCI Express, 10G Ethernet, Serial RapidIO, and InfiniBand all benefit from improved packaging and low-level interconnect schemes. The switch from parallel to serial will be complete this year at the high end, where performance is paramount.

Fabric interconnects are critical to scalability and improved reliability. Pin reduction in these serial interconnects is crucial. Otherwise, parallel alternatives would have fallen down on a number of fronts trying to meet system demands.

See Associated Figure