Pushing out the limits of copper - Components

Electronic News, Oct 21, 2002 by Joseph Fjelstad, Belgacem Haba, Para Segaram

Remember copper?

Some people think copper interconnects and other copper and wire componentry are passe for high-speed backplanes, and optoelectronics is ready to take over the transmission of most high-speed data.

But wait, not so fast.

Sure, the late 1990s witnessed a surge of interest and investment in optoelectronics for high-speed data transfer. The driving force was largely telecommunications, with component and system companies pressured to provide greater bandwidth to satisfy skyrocketing Internet businesses.

The Internet superhighway strained to handle the traffic already generated by the ingenious products and services offered or expected in the immediate future.

Many pundits believed that copper technology had run its course. There was no way that copper could deliver the bandwidth necessary for next-generation routers, or so they thought.

But we all know the Internet superhighway has hit a few bumps since then, and optoelectronics has run into some technical, and cost, hurdles as well. Maybe there is still time for more copper componentry after all.

Optical backplanes may represent the future, but developers need solutions today. A number of companies are revisiting copper technology to improve their performance and buy time to allow optoelectronic solutions and reliability to get fully mastered.

Various new high-speed copper interconnect solutions are in the works throughout the industry in an effort to push data along at ever-increasing speeds. Silicon Pipe and others are developing low-loss and low-dielectric constant solutions to the problem. While many industry thrusts involve the use of relatively costly low-loss and low-Dk materials such as fluropolymers, Silicon Pipe's approach is focused on the use and modification of traditional FR-4 technology to take advantage of the existing global infrastructure and years of manufacturing experience.

Renewed interest in copper has resulted in some new industry alliances, such as the recently formed High Speed Backplane Initiative, which seeks to transmit signals at data rates from 5Gbits/sec. to 6.5Gbits/sec. over a distance of 30 inches, including two connectors. Watch for more such industry moves.

Copper is still an excellent medium for high-speed signal transmission. In theory, it should be capable of signal transmission near the speed of light. Structures and materials have been created that have demonstrated signal propagation in copper at nearly 80 percent of the speed of light and development continues.

One key advantage of copper is that the infrastructure is fully in place and already properly geared for production. The only remaining question is how long it will take to turn it on and prepare for ramping.

Optoelectronics technology does offer unique advantages in signal transmission, and it is unexcelled in many ways. High-speed, low-loss, excellent signal integrity over long distances, security and nonexistent electromagnetic emissions (thus virtually no cross-talk) are among the most commonly cited advantages.

However, the technology is no panacea. Lacking a photonic transistor, traditional electron-based transistor technology must be used in signal generation and processing. Thus, when a signal is transmitted from system to system, it is necessary to convert the signal electrons into photons and then back to electrons. Over long distances this is a minor inconvenience, but over relatively short distance it can be a problem.

Using current technology, this process is expensive, energy-intensive and increases system latency or delay. It has been estimated that it requires up to 1.5 kilowatts of energy just to light up a half-rack backplane in a high-performance router. This represents a significant portion of the present energy budget of the router.

Thus optoelectronics technology, while excellent and the default choice for long-haul signal transmission, is proving, so far, to be less than ideal for short range transmissions such as backplanes and box-to-box signal transmission, with nagging and often costly problems that await effective solution.

On the other hand, copper wire and cable is tried-and-true technology for backplanes used in high-end routers and switches. It has proven capable of handling from hundreds of gigabits per second to terabits per second of bandwidth while demonstrating superior reliability. Copper backplane technology is largely a passive technology and thus lacks the complexity and expense of optics. Moreover, optical components represent potential reliability problems because of their active nature and nagging heat management issues.

Even so, copper technology does have limits and that is the reason that optical backplane technology was first proposed. But the full development of that Internet superhighway still has a long way to go. There is still plenty of time for copper.

Joseph Fjelstad, Belgacem Haba and Para Segaram work for San Jose-based Silicon Pipe Inc.

[GRAPH OMITTED]