Embedded seems to be the word of the day in the components industry.
But while semiconductor suppliers are well entrenched in the technology, the passive-component industry is still trying to figure out whether embedded passive devices can help OEMs achieve higher levels of integration.
"We know already that manufacturers are adopting higher levels of integration-at least 50 billion discrete components were replaced by some kind of chip array or network package this year," said John Rector, segment executive for microelectronics and technology systems at IBM Corp. Mt. Pleasant, S.C., and director of the National Electronics Manufacturing Initiative (NEMI).
But whether the industry will choose embedded passives in the printed-wiring board, or an integrated passive device, array, or network is still unclear, according to Rector.
"How the industry is going to achieve higher levels of integration [in the future] is what NEMI is struggling with," Rector said. "NEMI is looking at all competing technologies, including embedded passives."
Last month, the National Center for Manufacturing Sciences (NCMS), NEMI, the Interconnect Technology Research Institute (ITRI), and a dozen industry players, including Compaq Computer, Delphi Delco Electronics, IBM, and Northern Telecom, formed the Advanced Embedded Passives Technology Consortium.
What NEMI has found so far is that embedded passive devices are "a key enabler for producing the compact, high-reliability, high-performance and low-cost products of the future," particularly for OEMs that have high-density requirements, said NEMI's executive director and chief executive Jim McElroy.
Passives account for up to 90% of component placement in the average manufacturing process, and take up 40% of the space on a printed-wiring board, according to NEMI. By embedding the devices into the PWB, manufacturers can increase product density and simplify board assembly.
As discrete passives continue to decrease in size, board wiring has become more complex and physical placement more difficult. These factors are making embedded passives attractive, McElroy said.
Of all the passive components, resistors seem to be best suited for the technology, industry executives said. Embedded resistor tech- nology has been around for about 10 years, but hasn't taken off in mainstream applications. Now, there is a lot more interest in the technology from manufacturers looking for volume, low-cost solutions, according to McElroy.
"Resistors seem to be a natural fit for the technology," said John Denslinger, vice president of sales for Murata Electronics North America Inc., Smyrna, Ga. "And the best applications for embedded resistors will be at higher frequencies," he added.
Beyond resistors, it's going to take more than technology to match the cost and performance of discrete components. There may be trade offs in terms of cost of the multilayer board vs. the actual component cost, Denslinger noted.
"Embedding the resistance function in the board seems to make a lot of sense," said Sandy Beck, vice president of worldwide marketing at Kemet Electronics Corp., Greenville, S.C. "It will probably make low-power resistors obsolete. But they'll still be a need for wirewound and precision resistors," he added.
Since none of the integrated passive technologies appear to have emerged as a clear choice for the future, Kemet is looking at several technologies, including embedded passives, according to Beck. The company is participating in a consortium to study the embedding of resistors, capacitors, and inductors.
"OEMs will embrace different technologies that are unique to their particular need, and in the long term, all of them are going to be used but none will dominate" Beck said.
If and when embedded passive technology does evolve into mainstream applications, it's likely that companies will form joint ventures to tap into each other's expertise to "put the whole thing together," said Kemet's Beck.