x
Optical components

Bandwidth9 Scores a Coup

ATLANTA -- Supercomm 2001 -- Bandwidth9 Inc. made a splash at Supercomm 2001 this week by having its tunable laser incorporated in a demonstration of metro equipment by Nortel Networks Corp. (NYSE/Toronto: NT).

The demo shows how low-cost tunable lasers can enable service providers to set up and tear down wavelengths on demand in metro networks, from a remote control panel. This promises to slash provisioning times and eliminate truck rolls -- having to send craftsmen into the field to reconfigure equipment manually.

The key words here are "low cost." The idea won’t fly in metro networks unless tunable lasers cost more or less the same as the fixed wavelength lasers used in current metro equipment. And that’s where Bandwidth9 comes in.

Bandwidth9 has a particularly low-cost tunable laser for metro applications. In a nutshell, it’s taken a laser technology that’s already stolen the market for 850 nanometer, short-range fiber optics –- vertical cavity surface emitting lasers (VCSELs) -- and found a way of making it work at 1550 nanometers, the wavelength used in telecom networks (see Bandwidth9 Claims Laser Breakthrough).

Bandwidth9 says that its VCSEL will cost about the same as a fixed-wavelength distributed feedback laser (DFB) and gives three reasons for managing to get the price down to that level.

First, VCSELs are intrinsically low cost because light comes out of them vertically, so they can be tested while they’re still on a wafer. This means that duds can be weeded out early on, before a lot more money is spent on packaging for them (see Laser Blazers).

Second, Bandwidth9’s lasers can be directly modulated. There’s no need for a separate modulator to interrupt the beam of light coming out of the VCSEL to create the stream of light pulses that carry data.

Third, Bandwidth9 plans to make its VCSELs in very large volumes. While it’s been developing the laser itself it’s also been developing the process for making them. It’s built a facility with "very substantial capacity -- tens of thousands of devices a month," says Tim Richardson, Bandwidth9’s executive vice president, business development.

Other startups -- notably Agility Communications Inc. -- make similar claims for other types of laser. Agility is developing a side emitting laser (a form of distributed Bragg grating -- see Tune In!) and says that it can also bring costs down to those of today’s fixed wavelength lasers. It’s also aiming to make an integrated laser and modulator and is building a high-capacity production plant.

Nortel, however, says it’s sold on tunable VCSELs becoming the predominant laser technology in telecom equipment. That’s why it acquired another tunable VCSEL startup, Coretek, last year (see Nortel Gambles $1.43 Billion On Tunable Lasers). Coretek’s tunable VCSEL targets a different market than Bandwidth9’s. It’s much more powerful and much more expensive and is a key part of Nortel’s plans to deliver 40-gig long-haul transmission equipment.

Bandwidth9’s tunable laser pumps out around 100 microwatts of power and switches from one wavelength to another in 200 microseconds, according to Richardson. In the Supercomm demo, it was being used to switch among 11 wavelengths.

“It’s a very low-cost solution for metro access,” according to Peter Evans, vice president of marketing in Nortel’s metro optical division. Evans expects tunable VCSELs like Bandwidth9’s to become “a standard offering” in Nortel’s metro platform.

- Peter Heywood, Founding Editor, Light Reading
http://www.lightreading.comFor more information on Supercomm 2001, please visit the Light Reading Supercomm 2001 Site.

Page 1 / 2   >   >>
RazorDude 12/4/2012 | 8:17:27 PM
re: Bandwidth9 Scores a Coup Check out their web site (http://www.bw9.com).

I don't see any volume manufacturing capabilities in there flow charts. If anything, it seems slow and labor intensive. I guess they have some work to do...

Also looks like they're having trouble meeting their own test specs...
ownstock 12/4/2012 | 8:17:24 PM
re: Bandwidth9 Scores a Coup Labor intensive just means volume manufacturing goes overseas...as to specs, that's another issue...but working with Nortel says they have been delivering enough to their customer entertained...good stuff...too bad it took so long, it would have been IPO time if it had happened last year this time...
ownstock 12/4/2012 | 8:17:23 PM
re: Bandwidth9 Scores a Coup Labor intensive just means volume manufacturing goes overseas...as to specs, that's another issue...but working with Nortel says they have been delivering enough to keep their customer entertained...good stuff...too bad it took so long, it would have been IPO time if it had happened last year this time...
time2tune 12/4/2012 | 8:17:17 PM
re: Bandwidth9 Scores a Coup Good Stuff?

You call 100uW and 11 measily channels good stuff? There are companies out there (Agility, Iolon, New Focus, etc) that are currently in the 7-10mW range covering the entire C-Band? It sounds as if Nortel wasn't sure what was going and couldn't get their Coretek Lasers (MEMS based as well - which is rumored to be having similar problems as well) to work and they just threw something together extremely fast just for SuperComm. 2002 will be the year of tunable lasers and I can pretty much guess that Bandwidth 9 will not be running with the forerunners! As they seem to already be many months behind the above companies.

Do some homework on volume manufacturing and come back when you learn that domestic automation practices will beat foreign manual labor in terms of quality & yields & profits every time!
trojanlight 12/4/2012 | 8:17:09 PM
re: Bandwidth9 Scores a Coup I guess the rumors have some truth to them

After Xros not being able to deliver, rumors were indicating that Coretek was having difficulty in meeting the specs for manufacturable devices and the program was on the verge of being canceled.

It must be very embarrassing for coretek guys to see their archenemy delivering the tunable lasers to the mothership before they did...
TJ

itistime 12/4/2012 | 8:17:09 PM
re: Bandwidth9 Scores a Coup I am not an expert, but who else can deliver a low cost metro solution.
time2tune 12/4/2012 | 8:17:05 PM
re: Bandwidth9 Scores a Coup Finally someone on LightReading that will admit they don't know everything about every part of the optical world...for that I admire you!

Now as for low-cost tunable lasers. I have to admit that the only way tunable lasers will catch on (Cisco, Nortel, Lucent, etc) is for them to be totally cost effective especially with the recent down turn of the industry.

As I mentioned in previous emails companies such as Agility and Iolon (Both Top 10 Private Companies on here) as well as ADC and New Focus are in a race with some other minor players to be designed into these new generation system systems.

Remember different technologies Tunable Lasers vary in power (including measly 100uW all the way to 20+mW). They do vary the wavelength and some have mentioned about integrating other components into them as well. Tunable Modulators, filters, and amplifiers are going to go hand in hand with these laser. The company that utilize these advances into their design will be the winners!

Watch out for 2002 - Year of the Tunable Laser!
time2tune 12/4/2012 | 8:17:04 PM
re: Bandwidth9 Scores a Coup The rumors do sound like they have some truth behind them.

My question is whether or not it is the company failing or the technology (VCSEL/MEMS)? With Coretek and Xros not delivery and Bandwidth9 coming out so far behind the semiconductor guys (Agility and Iolon) in power and tuning, there might be some rudamentary issues behind that technology. (Or Agility and Iolon have the right team to design and assemble the product - only time will tell).

What IS embarrassing is that Nortel put such a lower power/low tuning channel device in their product and then displayed it.

I have to question Light Reading's comment on Nortel being totally sold on the idea of VCSELs. It seems that BW9 has a long way to go to get to where it can deliver large amounts of product and get their demo to actually meet their PUBLISHED Spec Sheet! Good Luck with that!
trojanlight 12/4/2012 | 8:17:01 PM
re: Bandwidth9 Scores a Coup >The rumors do sound like they have some truth behind them.
>My question is whether or not it is the company failing or the technology (VCSEL/MEMS)?
------------------------
The main problem with Coretek is that they have trivialized the technology such that on PowerPoint presentations of device manufacturability seemed extremely easy. Note that Coretek is using an optical pump to power MEMS VCSEL Chip. Even though coupling from a VCSEL is relatively easy compared to DFBs, Coretek first have to couple the Pump light into the VCSEL chip on one side and couple the1550 nm light into fiber on the other side. Automation of this task is not straightforward (see AxsunG«÷s Technology). In addition to the low yield packaging a pump laser and VCSEL into single butterfly package, yield of their MEMS structure is not high. Thus combination of these two factors alone probably forced Coretek to delay delivery of the manufacturable device technology.

After Xros failure, this must have had a cold shower effect on NT execs who decided to spend $1.4B on Coretek. I guess they are finally biting the bullet and decided to go to coretekG«÷s archenemy BW9. I wonder if this is signal to Coretek that mothership will no longer wait for the promise and may send coretek product plans to black hole.

As for BW9G«÷s technology, it is very power limited. The output power from electrically pumped VCSELs will be at most limited to 1 mW for reliable- long life devices.
TJ
time2tune 12/4/2012 | 8:16:58 PM
re: Bandwidth9 Scores a Coup
>Thus combination of these two factors alone >probably forced Coretek to delay delivery of the >manufacturable device technology.

>As for BW9G«÷s technology, it is very power >limited. The output power from electrically >pumped VCSELs will be at most limited to 1 mW >for reliable- long life devices.
>TJ
------------------------------------------
Great article, I have to admit I really don't the world about VCSEL Technology, but that certainly cleared some of it up for me.

One question I would have is how easy is it to integrate items such as Ampifiers or Modulators to a VCSEL device? That would definitely solve their power issues.

Stating the two comments above, would it be safe to assume that the semiconductor tunable laser that certain companies are developing (Agility, Iolon-ADC, and I think New Focus) are a more stable and reliable at this point of time? If so it seems to me that these companies are going to be the sucessful ones, due to the fact that VCSEL will just be ramping (In terms of manufacturability as well as specifications) up when the design wins are pouring in for these other companies!

T2T
Page 1 / 2   >   >>
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE