-
- News
- Books
Featured Books
- design007 Magazine
Latest Issues
Current IssueLevel Up Your Design Skills
This month, our contributors discuss the PCB design classes available at IPC APEX EXPO 2024. As they explain, these courses cover everything from the basics of design through avoiding over-constraining high-speed boards, and so much more!
Opportunities and Challenges
In this issue, our expert contributors discuss the many opportunities and challenges in the PCB design community, and what can be done to grow the numbers of PCB designers—and design instructors.
Embedded Design Techniques
Our expert contributors provide the knowledge this month that designers need to be aware of to make intelligent, educated decisions about embedded design. Many design and manufacturing hurdles can trip up designers who are new to this technology.
- Articles
- Columns
Search Console
- Links
- Events
||| MENU - design007 Magazine
Electronic Interconnection Files No. 7 - Whither the Future of the Electronic Interconnections?
January 9, 2006 |Estimated reading time: 3 minutes
Substantial attention is being given and money being poured into optoelectronics and photonics. There have been numerous suggestions that the future of all electronics resides there. In a recent contributed commentary in Electronic Business[1] there was recalled the memorable scene from early in the movie, The Graduate, where the title character, Ben, during a party thrown by his parents in honor of his graduation, is pulled aside by one of his father's friends who says: "Ben, I am going to say just one word to you about your future... Plastics" Inserted in place of the term, plastics, however, is the word photonics". Although it is abundantly clear that optoelectronics and/or photonics will always have a sizable role to play in long-haul transmission of digital data, copper is proving that it has more life left in it than is commonly assumed for the short haul. Paraphrasing the great wit, Mark Twain, "Rumors of the death of copper circuits are greatly exaggerated."<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />
The article suggested that "the battle between electrons and photons is starting in a logical place: in networking applications targeting 10Gpbs transmission" unstated and presumed is that copper circuits will still work at 5Gbps to 6.25Gbps. The article goes on to state that the end result would be "easy to predict" with photons being the eventual winner. The author made some very salient points noting that photons have less concern relative to attenuation, use less power and generate less heat (presuming no active cooling will be required) at a given data rate. Also importantly noted was that light beams can pass through other light beams without interference which is not the case for electrons. The author finally correctly noted that photonic/optoelectronic components of today are expensive and not as manufacturable as electronic components.
Given this strong sentiment in favor of optoelectronics by some industry pundits, the question is begged: What is the future of electronic interconnections? The answer is that it is not an either/or proposition relative to copper and photonics. Both technologies will play important roles in the future, the question is where will the line be drawn? It appears that it will be some point well beyond the 10Gbps line that some believe exists.
For example, more than two years ago, with the assistance of Agilent, SiliconPipe demonstrated a prospective copper-based interconnection solution for telecomm that met OC-768 performance levels (40 Gbps)[2]. This was done in a single differential pair copper channel over a distance of 30 inches at low power. One needs to recall at this point that most in the datacom sector of the electronics industry presumed and continue to believe that 40Gbps requires an optical solution. The structure was nonstandard in terms of manufacturing but it did demonstrate proof of concept. While there is no need for such performance at present, it appears that within a 1-meter cube, copper will be able to meet the industry's performance needs, including low or no cross talk, for the foreseeable future and do so at low power.
A more recent demonstration by SiliconPipe, this time with the assistance of Aeluros, NexGen and ERNI, of a single-channel, 10-Gbps solution, showed that the system performed flawlessly over a distance to 30 inches through two connectors and did so using less than 2 percent of the power requirement predicted by the chip developer. Obviously, no electron-to-photon-to-electron (E-O-E) conversion was required. Finally, just last month, data was generated showing that the channel used for the 10Gbps was just as well suited for use at 20Gbps when power was increased to normal levels as can be seen in the accompanying eye diagram. Thus it seems fairly clear that the line drawn as to where photonics will be required needs to be pushed out further.
Researchers in both universities and industry will, of course, continue to pursue a viable photonic transistor, and then look for ways to mass produce the same but in its absence, it appears quite clear that with proper architectural design, copper circuits can and will continue to fill the bill for short-reach interconnections at low power and do so very cost effectively.
<?xml:namespace prefix = v ns = "urn:schemas-microsoft-com:vml" />
Eye diagram for a copper interconnection solution at 20Gbps illustrates the potential of copper circuits extends well beyond the often presumed upper limit of 5Gbps limit. (Courtesy HJ Liaw, Aeluros, Inc.)
[1] Krewell, K., "The Coming Photonic Revolution" Electronic Business, October 2005
[2] Allan, R., "Coax Backplane Connector Handles Rates To 40 Gbits/s" Electronic Design October 2003