While most tunable laser vendors are hibernating in the telecom winter, startup Santur Corp. is trying to buck the trend.
The company announced an extra $23 million in funding today, which it plans to use to ramp its product to volume production (see Santur Grabs $23M). Furthermore, it has reportedly landed Corvis Corp. (Nasdaq: CORV), and Innovance Networks as customers, say sources close to the company.
Although Santur won't officially confirm this, it appears that both Corvis and Innovance were using its tunable lasers in demonstrations at the recent Supercomm show (see Vendors Aim to Cut Costs in Core). Both Corvis and Innovance plan to announce commercial availability of their platforms in the next few months, the vendors said before the show.
Whether this puts Santur on a path to success remains to be seen. Other tunable laser vendors have found that customers have reneged on their commitments to use tunable lasers, so there's no guarantee that the alleged Corvis or Innovance deals will turn into significant revenues (see Scattered Signals for Tunable Lasers). And industry analysts, including Jeffrey Morgan at J.P. Morgan H&Q Equity Research, generally believe that volume deployment of tunable lasers could still be a year or more away.
However, Santur seems better placed than many startups. For starters, it has money in the bank. Its funding today is only the first closing of the round, says Gurindar Parhar, Santur's VP of business development. It expects to announce a further closing in the next few weeks.
The lead investor in today's round was Thomas Weisel Partners, with existing investors Menlo Ventures and Sequoia Capital also taking part. JDS Uniphase Corp. (Nasdaq: JDSU; Toronto: JDU) also took part as a strategic investor, although Parhar points out that this doesn't mean JDSU is neglecting its in-house tunable laser development.
This brings Santur's total funding to around $43 million in equity, in addition to a lease line of $10 million. Although Parhar doesn't want to say exactly how many employees the company has, he does say it is fewer than 100.
Santur's key selling point, he says, is the fact that its laser offers a wide tuning range and the optical quality of Distributed Feedback (DFB) Lasers -- the type of laser in widespread use in DWDM systems today. In other words, Santur's laser is high power, stable, reliable, and low cost, it claims. Other types of tunable laser don't necessarily meet those requirements, says Parhar, because the technology is radically different from traditional DFB laser technology (see Tune In!).
"The concept of tunable lasers has been around a long time," he says. "What's been missing is one that meets the needs of carriers." Carriers are comfortable with Santur's laser technology, he adds, because it gives the same performance as the lasers they are used to dealing with.
Santur's laser is DFB-like because, in fact, it contains DFBs -- four of them in an array, to provide wavelengths from different parts of the spectrum -- integrated in a chip barely much bigger than a standard DFB chip. A MEMS (micro-electro-mechanical system) mirror couples the output from one of the lasers to the output of the chip. The whole thing is tuned (slowly) by heating and cooling. The product was announced back in March (see Santur Debuts With Laser).
A potential sticking point for Santur, however, is that its laser isn't yet Telcordia Technologies Inc. qualified. Santur is just embarking on this process, which it hopes to have completed by the end of the year. Given that any potential customers will need to have qualified components before they can ship their systems, this means that Santur will not book revenues this year.
Corvis and Innovance could not be reached by press time.
I hear they still have more employees than any other tunable group, even after ditching a bunch of employees during various downsizings!
Nortel is laying off manufacturing employees at coretek left and right. ADC just shut sweden down. BW9 is down to R&D people only. iolon has dumped their reliability and manufacturing people and is looking to be saved by an acquisition (Enron maybe?) Santur is pushing MEMS and DFB arrays? Good luck!?
[Ref message #3] Don't include Agility in the list of strong players--they have all but turned off the lights on their manufacturing due to yield issues and Telcordia qual issues.
I can second the statements from Gratefulphoton: From what I heard couple of months ago, former CEO of Coretek has left Nortel early this year or late last year. The Coretek guys developed the tunableVCSELs were about to leave together (separate from the CEO) ... -- from lastyears' posting http://www.lightreading.com/boards/message.asp?msg_id=11761
>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 Axsun’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 coretek’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 BW9’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
as stated, the coretek disposition is based on what i have heard from colleagues. no specific evidence or announcement that i have seen.
i will diverge with your assessment of the ease of manufacture of the arrayed DFB with free-space switched coupling via MEMS. although i agree the more difficult part is the mirror/fiber coupling, making DFBs yield is still not simple. yield-killer defects are probabilistic, and thus making 4 of them yield on the same chip is at least 4 times less probable than yielding a single element. if you want to know mirror/alignment issues, talk to an ex-OMM employee, for example. it's doable, but tough.
one reason i like the VCSEL-based devices is the on-chip testing which allows you to sort and mark the devices earlier in the process, thereby eliminating costly post-processing on non-yielding devices. that's a big advantage. fiber coupling is more straightforward with circular modes that are not astigmatic (unlike edge-emitters). they are smaller and cost is yield is real estate on the wafer.
Correct me if I am wrong, but isn't Innovance supposedly manufacturing L-band systems and using Iolon tunable lasers for this. Recent news releases from Iolon and Innovance indicate a partnership between both companies, of siginificant revenues. If this is so Innovance is just testing components from Santur and other suppliers to test technology.
I'm going to stick my neck out here and say that going by what I am hearing, the leaders in tunable lasers are Fujitsu and Nortel. Odd that they are among the few remaining vendors with captive components divisions...
Out of the startups making tunable lasers, the leaders of the pack appear to be iolon, Bandwidth9, and maybe Santur.
Although Santur is starting out later as a company, its founders were working on the DFB array technology when they were still at SDL, some two and a half years ago. So they have had time to iron out some of the production problems, I bet. And at the end of the day, yields are the main thing. The structure can't be that difficult to make since, with the exception of the MEMS mirror, is pretty much identical to normal DFBs.
I believe Fujitsu has something similar (a tunable DFB laser array), although I haven't managed to find anybody at the company to tell me about it directly. If anyone from Fujitsu is reading this, can you put me in touch with the tunable laser division?
grateful photon, if you know for sure that Coretek really is dissolved, and you can provide evidence, then please contact me directly.
the only thing you are missing is the koolaid these guys are drinking. 4 integrated DFBs coupled by a MEMS mirror into an output waveguide/fiber isn't cheap to produce or simple to pass carrier-class product integrity. there are at least 4 strong players with tunable laser offerings (i would count agility, BW9, altitun, and fujitsu as the leaders). the standard question: how many winners will the market support?
coretek was a huge crater for nortel. the promise of a c-band tunable VCSEL was a seductive sell (nearly 1B$!) but not realizable by that group. thus it was an expensive and not very useful acquisition that is now dissolved, last i heard. HOWEVER, IF a group was able to realistically develop such a device, it would be much more interesting than another variant on a longitudinal laser scheme such as santur. i hear that BW9 has struggles with realizing such a device, but it is the right direction for actually getting 'low cost' and 'tunability' in the same product (IMHO).
According to the guys in the CO, Fujitsu has been shipping 4 lambda DFB tunable transponders for a year and a half, Telcordia blessed, carrying live traffic with no problems. At SuperComm02 they announced shipments of 22 lambda tunable transponders. A laser component startup hoping to sell to Corvis! and to run the Telcordia gauntlet by the end of the year is not "bucking the trend", Santur is behind the curve.
The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Digg
Del.icio.us
Reddit
