K2 Claims Laser Record

Startup K2 Optronics Inc. is preparing to do battle with heavyweight components vendors at the NFOEC next week. It plans to demonstrate a direct modulated laser that can send 2.5-Gbit/s signals over distances up to 300 kilometers -- about twice as far as other lasers of this type.

In its booth demo, K2 intends to show a "direct comparison" of its laser with one from a competitor. A 42-inch plasma screen will display real-time traces of the two lasers' linewidth -- an indication of how far their light pulses travel before they become indecipherable.

The logo of "Brand X," the competitor's laser, will be tastefully blanked out, says Davinder Basuita, K2's executive VP of sales and marketing: "We don't want to upset anybody." It would be unfair to identify Brand X, he adds, because K2 has conducted similar comparisons with lasers from at least three other competitors with similar results.

Whose logo will be inked out? It could be Agere Systems (NYSE: AGR), Alcatel Optronics (Nasdaq: ALAO; Paris: CGO.PA), Corning Inc. (NYSE: GLW), ExceLight Communications, Fujitsu Microelectronics Inc., or Nortel Networks Corp. (NYSE/Toronto: NT). All of them make the sort of directly modulated lasers that might be Brand X in K2's demo.

Basuita says that K2's laser technology will offer significant cost savings, particularly in metro networks. "Right now, most devices in the metro are long-haul lasers," he contends. "In many cases that's overkill for the application."

Direct modulation -- where the signals are encoded on the laser beam by turning the drive current to the laser on and off -- is a cost-effective option, because it eliminates the need for an external modulator. However, most lasers of this type are only specified out to 80 kilometers. A small subset of vendors offer devices that reach out to 170km, but some of these are specified over modern optical fiber, which is specially designed to reduce the signal degradation -- and most carriers won't have this type of fiber in their networks, says Basuita.

What limits the reach of a direct modulated laser? Chirp, he explains. In most direct modulated lasers, the wavelength-selective element is a grating written on top of the chip. When a drive current passes through the chip, the density of electrons inside the chip goes up, so the refractive index shifts slightly, and changes the wavelength reflected by the grating. The end result: "The wavelength wobbles in sympathy with the data." That means the signal is more prone to dispersion.

K2 tackled the chirp problem by taking the grating off the chip, and putting it into the fiber pigtail, to create an external cavity laser. "It's not a new technology," Basuita notes, "But we've solved the packaging and production problems that stopped people commercializing it earlier."

One worry people have had with external cavity lasers was the fact that they are prone to mode-hopping -- sudden jumps in the laser's wavelength. To prevent this, the laser cavity needs to be extremely stable, both thermally and mechanically. Basuita claims that K2 has solved the problem completely and hopes that potential customers will give his company a chance to prove it.

Another potential drawback is the fact that it's more difficult to do high-speed modulation with long laser cavities. So far, K2 has developed an OC48 (2.5 Gbit/s) product. Extending the technology to OC192 (10 Gbit/s) could prove tricky.

If the laser works as advertised, then the cost savings could be huge. The part itself will not necessarily be cheaper, but it will bring down the cost of the overall solution. In the first place, there's no need for an external modulator or a wavelength locker. Other savings depend on the application. For example, thanks to the high performance of the laser, a customer might be able to do away with the need for dispersion compensating fiber, resulting in significant savings.

There's one other key advantage, says Basuita: K2 should be able to respond to orders very quickly. "The laser is color blind," he notes. "No wavelength is assigned until late in the manufacturing process. In fact we don't have to stock the finished laser. Instead, when the order comes in, we attach the pigtail [which contains the grating], and then the device gets shipped out."

K2 is hoping to launch its first product at the ECOC exhibition in September.

— Pauline Rigby, Senior Editor, Light Reading
http://www.lightreading.com For more information on NFOEC, please visit the Light Reading NFOEC Site.

opticaltoys 12/4/2012 | 8:08:21 PM
re: K2 Claims Laser Record Where is the innovation. Is it patentable?
redface 12/4/2012 | 8:08:13 PM
re: K2 Claims Laser Record "Where is the innovation. Is it patentable?"

This technology is not new. I doubt it is patentable. I have seen similar results from Marconi's research lab in Italy and British Telecom before, see R. Paoletti, et al, "15-GHz modulation bandwidth, ultralow-chirp 1.55 -¦m directly modulated hybrid distributed Bragg reflector (HDBR) laser source", IEEE Photonics Technology Letters, Vol. 10, No. 12, 1998, pp1691-1693. Timofeev published a series of papers on it as well.

But the fact that this company is commercializing it can be exciting.

One major difficulty with this laser is definitely the mode hopping, as pointed out in the article. Typically the FBG is pretty long, so the mode spacing is very narrow. It should hard to eliminate mode hopping.
Sign In