& cplSiteName &

Nortel and Agility in Tiff Over Lasers

Light Reading
News Analysis
Light Reading
4/30/2002

Nortel Networks Corp. (NYSE/Toronto: NT) whipped up something of a storm in a teacup today by proclaiming the general availability of "the industry's first Telcordia-qualified widely tunable laser" (see Nortel Unveils Tunable Laser).

Why? Because Agility Communications Inc. announced exactly the same thing just over a month ago (see Agility Tunable Laser Certified).

When questioned about this, Tom Dudley, Nortel's marketing manager for its ML-20 laser, cast doubt on Agility's claims. "Some of our competitors recently made announcements, but frankly, the wording was somewhat nebulous," he says. Agility's March 28 press release states that its tunable laser "technology architecture" had met and exceeded reliability assurance requirements in accordance with "the intent" of the Telcordia Technologies GR-468-CORE.

"We are announcing qualification of a generally available product, not just an architecture or technology," crows Patrick Walsh, VP and product line manager for optical components in Nortel's own press release.

But Agility says Nortel's got it wrong. "It's a product that we qualified, no question about that," insists Kevin Affolter, Agility's director of marketing. "It happens to be our first-generation tunable laser product, not the high-power device, but this will follow on next quarter."

In fact, Agility says it chose the wording in its press release to try to avoid confusion that might arise because there isn't actually a standard for testing tunable lasers. Telcordia GR-468-CORE was originally devised for fixed-wavelength lasers, and on its own it isn't a sufficient reliability qualification for tunables, Affolter contends.

"GR-468 is mostly about mechanical and thermal properties, like showing you can hit the laser with a hammer and it will still work," he says. "But, it doesn't talk about things like wavelength drift, which, frankly, are incredibly important. The laser is designed to tune, so you need to make sure it only tunes when you want it to."

In the future, there may be a standard specifically for tunables -- Agility, Nortel, and other tunable laser vendors in the Optical Internetworking Forum (OIF) are working on one right now. Both Nortel and Agility say the testing they've done goes way beyond the requirements of GR-468-CORE, but because their lasers are based on different technologies, they've focused on different areas of testing. A true standard would be technology agnostic.

In Nortel's case, its ML-20 tunable laser, which it first announced back in March, is based on a VCSEL (vertical cavity surface emitting laser), with a MEMS (micro-electro-mechanical system) mirror on top -- the technology it acquired from Coretek (see Nortel Gambles $1.43 Billion On Tunable Lasers). So Nortel has been particularly keen to show that the presence of a small moving part does not detract from reliability. It's tested the MEMS mirror on its own and is claiming a FIT (failure in time) rate of less than one, meaning one failure during a billion hours of device operation (that's extremely low).

Nortel has several tunables in its portfolio, including some narrowband tunables. The ML-20 can tune to any wavelength in the entire C- or L-bands, with an output power of 20 mW. It achieves this by copackaging the laser with a semiconductor optical amplifier.

As noted, Agility's qualified device is its first-generation product, which has an output power of 4 mW (see Agility Launches First Product). However, it has also demonstrated a higher-power device, offering 10 mW, which should be generally available -- and qualified -- next quarter (see Agility Packs Three Into One and Agility Unveils Long-Haul Laser).

The bottom line is that the appearance of two qualified tunable lasers is good news for the tunable laser industry as a whole. "Finally we have a laser that network planners can have confidence in," says Nortel's Dudley. Agility's Affolter has this to say: "We need more than Agility to have a qualified product, before there can be a market for us all."

— Pauline Rigby, Senior Editor, Light Reading
http://www.lightreading.com

(17)  | 
Comment  | 
Print  | 
Related Stories
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 2   >   >>
WhiteKnight_2
WhiteKnight_2
12/5/2012 | 12:49:24 AM
re: Nortel and Agility in Tiff Over Lasers
IMHO Agility will be history before you know it. Their entire business case is based on a single device that has no market application, it is extremely complicated to fabricate (four segments with different layer structure, processing techniques, separate electrodes, etc.), the yield is extremely low (DFB yield is about 40% with only one section, imagine what it is for four segments with various gratings and functionality), and a nightmare to control electrically. At the end, it is a single device for a single wavelength. Any prospects? forgetaboutit. Like many other companies (remember the late Nanovation) it was founded on the ignorance of greedy investor and the naivety of overzealous university professors.
Titanic Optics
Titanic Optics
12/4/2012 | 10:29:53 PM
re: Nortel and Agility in Tiff Over Lasers
What will be successfully adopted by OEMs and carriers? When?
simontemplar
simontemplar
12/4/2012 | 10:29:44 PM
re: Nortel and Agility in Tiff Over Lasers
Fujitsu claims to have shipped +4000 units already.

ST
belas_knap
belas_knap
12/4/2012 | 10:29:43 PM
re: Nortel and Agility in Tiff Over Lasers
Marketing 101- Depends on how cheap they become & the advantage in terms of cost saving to the consumer. There are advantages for equipment manufacturers (higher volumes of a single component = bigger discounts, lower manufacturing overheads in terms of production) and carriers/enterprises (single spare card(s) to cover all the wavelengths instead of multiple spare cards to cover each card equipped with different wavelengths means lower overheads) but the price of this must show cost savings over what the equivalent process is today. If it's cheaper for a carrier to buy/store/maintain 20 cards of different fixed wavelength than it is to buy one/two cards with tunable laser, then they will keep the same process.
Titanic Optics
Titanic Optics
12/4/2012 | 10:29:40 PM
re: Nortel and Agility in Tiff Over Lasers
>>Marketing 101- Depends on how cheap they become & the advantage in terms of cost saving to the consumer...but the price of this must show cost savings over what the equivalent process is today.<<

And also maintain equivalent performance. While the "future" plans for many tunable vendors show the POTENTIAL of a roadmap to near-equal performance. Things like power, chirp, linewidth, signal-to-noise, etc. must not be comprormised. And, any cost discussions must take into account that, while a tunable laser may have a roadmap for a price reduction over the next two years, traditional source lasers will continue to fall in price--the world isn't a static place that waits for the godsend of tunable laser technology to come along and just sits still.

In this vein, a few of the current offerings that I have seen incorporate an amplifier, or they rely on a ridiculously low price for the amplifier to make the laser cost competitive in their future "next generation" designs they like to tout.

Anybody care to give out any hard evidence of where ASPs are and what the performance trade-offs are?
belas_knap
belas_knap
12/4/2012 | 10:29:39 PM
re: Nortel and Agility in Tiff Over Lasers

Quite right - should have made this point also.


>> And also maintain equivalent performance. While the "future" plans for many tunable vendors show the POTENTIAL of a roadmap to near-equal performance. Things like power, chirp, linewidth, signal-to-noise, etc. must not be comprormised. And, any cost discussions must take into account that, while a tunable laser may have a roadmap for a price reduction over the next two years, traditional source lasers will continue to fall in price--the world isn't a static place that waits for the godsend of tunable laser technology to come along and just sits still.
nobollox
nobollox
12/4/2012 | 10:29:38 PM
re: Nortel and Agility in Tiff Over Lasers
Agility and Nortel - which is better? which is qualified?

I visited Agility's booth at OFC and was impressed with their technology.

Plus, I have never been a big fan on the MEMs architecture that Nortel uses.

I don't know what kind of hammer they used to hit the mirror, but I wouldn't build a house with it.

sridude
sridude
12/4/2012 | 10:29:34 PM
re: Nortel and Agility in Tiff Over Lasers
What do you mean by a "MEMS Architecture"?

RE: Hammers, Telcordia is pretty specific on the type of hammer required.

- sd
zweisel
zweisel
12/4/2012 | 10:29:34 PM
re: Nortel and Agility in Tiff Over Lasers
... you were impressed by the technology demonstarted at a tradeshow? you have got to be kidding I hope ...
clear brain
clear brain
12/4/2012 | 10:29:32 PM
re: Nortel and Agility in Tiff Over Lasers
The only time these tunable lasers will sell in volume is when tunable lasers are used more than just as DWDM source. The telecommunication industry will need to wake up to the fact that (widely) tunable laser will "never" be as cheap as regular DFB laser. Universal back up has a very small market. Replacing current linecards with tunable linecards will always be expensive. Without new and innovative applications, no one can afford tunable lasers.
Page 1 / 2   >   >>
Featured Video
Upcoming Live Events
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events