Optical components

Princeton Optronics Takes On Coretek

What could turn out to be a significant milestone in the development of 1550 nanometer vertical cavity surface emitting lasers (VCSELs) is being cooked up by Princeton Optronics Inc., a startup that’s planning to announce a $25 million second round of funding next Monday (August 21).

The startup is developing a tunable VCSEL that appears to compete head-on with Coretek -- the startup acquired by Nortel Networks Corp. (NYSE/Toronto: NT) in March 2000 for stock then worth $1.43 billion (see Nortel Gambles $1.43 Billion On Tunable Lasers).

Princeton Optronics says its device will deliver the same power as Coretek’s laser -- 20 milliwatts – and will have an equivalent tuning range. Its manufacturing costs, however, will be five times lower, according to Tim Hays, Princeton Optronics’ VP of marketing.

It’s tough to evaluate this claim, as developers of tunable lasers simply won’t say how much their products cost. In a session on tunable lasers at the Opticon 2001 conference yesterday, a bunch of executives from five manufacturers wouldn’t go further than saying that their products would cost up to 20 percent more than the equivalent fixed-wavelength lasers.

Parvis Tayebati, Coretek’s founder, says it's tough to comment on Princeton's claims without having more details. "On the face of it, it looks like a structure that's more difficult to manufacture, so I don't know how it could be cheaper," he says. "If they can do it, I'll have to congratulate them."

It's also worth noting that the cost of manufacturing the laser itself typically represents only a small fraction of the overall cost of the unit sold to customers. Packaging typically represents the bulk of the cost -- around 80 percent according to Tayebati -- which suggests that Princeton's fivefold reduction doesn't refer to the price paid by customers.

Tayebati also contends that Princeton itself won't have a clear idea of its costs until it starts shipping samples.

Princeton says that it’s been developing its tunable VCSEL for six or seven years and will ship first samples in the next few weeks.

Like Coretek, Princeton achieves relatively high powers by pumping light from a second laser into the VCSEL’s cavity -- although Princeton uses a second, high-power 980 nm VCSEL to do this, while Coretek uses a 1310 nm edge emitting laser.

A great deal of the startup's intellectual property lies in the production of the 980 nm laser, says Hays.

Also like Coretek, Princeton Optronics tunes its laser using MEMS (micro-electro-mechanical systems). But instead of being integrated, the MEMS device is a separate part, which forms an external cavity.

"The critical step is getting the pump, the cavity, and the fiber all aligned," says Princeton's Hays. Right now, this process has to be done “actively” -- adjusting the alignment to optimize performance in an iterative process, with the laser actually operating. However, Princeton hopes to be able to move to “passive” alignment, where the elements are automatically placed in the optimum position first time, without the laser operating.

Of course Coretek isn't the only competition to be reckoned with. New Focus Inc. (Nasdaq: NUFO), Iolon Inc., and Blue Sky Research all claim to have external cavity laser designs that reach 20mW of output power (see New Focus, New Laser, Corning Backs Laser Startup, and Blue Sky Scores Four in a Row).

"In the case of New Focus, we think our five-times [cost] advantage is more like eight times," says Hays. He also points out that New Focus is dependent on a supplier for its laser chip, whereas Princeton Optronics has no supply issues. It buys in wafers -- a commodity -- and then does its own processing and dicing.

At lower powers, Princeton’s competition includes another startup that’s recently unveiled a 1550 nm tunable VCSEL development, Applied Optoelectronics Inc. (AOI) (see Startup Makes Waves in VCSEL Market).

Princeton Optronics’ second round of financing, to be announced on Monday, is led by Stockholm-based Investor AB. It also includes St. Paul Venture Capital and Technology Venture Partners LP, which joined forces with first-round investors Intel Capital and Novak Biddle Venture Partners.

Princeton Optronics shouldn’t be confused with Princeton Lightwave Inc. (PLI), a developer of distributed feedback lasers.

— Pauline Rigby, Senior Editor, Light Reading
ownstock 12/4/2012 | 7:55:59 PM
re: Princeton Optronics Takes On Coretek Good follow-up. Now check out this URL for OPSEL laughs:


There's a real popular product for ya'...

Humm, 1 AMP threshold current, wha' thas voltage now? Wowee, that's reel eeeefffishent Jethro! Problee goes with great rel-billitee too. Har har! Gimmee bunch o' those ritenow! Thaze gonna make my netwurk reel good wurkin!

Aram is laughing all the way to the bank! Twice!

Peter Heywood 12/4/2012 | 7:56:07 PM
re: Princeton Optronics Takes On Coretek Just wanted to draw your attention to the fact that Coretek's founder, Parviz Tayebati, got back to us with some comments yesterday. As a result, we've updated the story.

Tayebati gave the current status of developments in the panel on tunable lasers at Opticon.

ownstock 12/4/2012 | 7:56:18 PM
re: Princeton Optronics Takes On Coretek Wim:

I have a pretty good idea, but to protect the innocent, I will make the following argument:

BW9 and Connie have been working on a tunable VCSEL for some time now. Maybe five years, maybe a lot more, depending on how much you put into the experience of the founders and key technical people. They have gotten the thing to Alpha or maybe Beta...and have been scrambling to get a few into the hands of Nortel and the like, so there is obviously no massive amount of life test data per se. But, in their favor, the device is low power...so they should be able to point to a body of literature on VCSEL reliability as some feel-good, it-can-be-done support.

I would simply say that you can generally put a figure of five years to critical mass (alpha, beta stage) on a significant new device....and that is after the definitive thesis is written (and after the seed round is closed, etc). Also, it will cost at least $20 Million. And if they make it that far (5 and 20), then they only have to worry if it will prove reliable (another two years and $20 Million).

So use those as factors.

wimchatta 12/4/2012 | 7:56:20 PM
re: Princeton Optronics Takes On Coretek Ownstock,

Convincing argument there about this technology being neither high reliability nor low cost. Is it possible that they have some silver bullet?

Would you know where Coretek stands with its qualification and product performance?

ownstock 12/4/2012 | 7:56:32 PM
re: Princeton Optronics Takes On Coretek When you hear them spinning cheap and tunable and high power, you know you've got to have your hip boots on and pulled up high! Trying to raise a $25 Million round or die? If you can't smell it now, you never will!

Fact: There is no body of reliability for optically pumped tunable VCSELs in the literature. But, regardless, cascade reliability of optically pumped tunable device: pump rel x device rel x tuner structure rel x package rel x control circuit rel = really low rel.

Even if they did not suffer cascade manufacturing yield of devices, and high packaging costs on top of that, and really could make and sell it cheap; selling it cheap only makes it seem cheap.

After you have to install two or more to get the same rel you had with one conventional fixed wavelength EEL...it blows the claimed savings...and that is just at initial provisioning.

Key concept: a reliable high power OPSEL, tunable to boot, and cheap = free lunch.

Try agin...ain't nosech ani-mule Jeth-ro!

As to the round, let's see, their equation is: no customer value = no investor value.

This one is hot-dang! I wanna eenvest rinow!

But hey, that's JMHO...

calpole 12/4/2012 | 7:56:35 PM
re: Princeton Optronics Takes On Coretek I guess Optronics Tunable lasers are equipped
with an wavelength locker and still they are
1/5th of the market price..i.e.$1.4K (@13dBm output
power) And still they are for real??? It can stabilize wavlengths within +-20pm?

I belive the price but I don't think they have
all the desired qualities.

Can anyone from Optronics can clarify this point????
Sign In