x
Optical components

Picolight Picks Up $13M

Declaring itself a survivor of the optical drought, module vendor Picolight Inc. today announced its sixth round of funding, claiming profitability is in sight.

The $13 million round was led by Coral Capital Management and included prior investors BA Venture Partners and Vesbridge Partners LLC. Coral managing director Todd Ortberg has joined Picolight's board. Picolight expects to extend the round to include another $4 million to $7 million in "cushion money" aimed at any expansion the company might need, CEO Steve Hane says.

As a sign that the components market is improving, Picolight says its valuation increased with the extra funding. The company claims business for the first quarter of 2005 was up 125 percent from the same quarter last year.

The $13 million announced today is intended to drive Picolight to profitability, hopefully "by the end of this year," Hane says.

Picolight has now raised more than $90 million since being launched during the dotcom bubble. The company was an early entrant in the Vertical Cavity Surface Emitting Lasers (VCSELs) market, trying to put lasers into a smaller and more easily manufacturable form. Its prospects probably aren't as lofty as they appeared during the bubble, but the company has a solid technical pedigree, including CTO Jack Jewell, who helped pioneer VCSEL research outside of Japan.

"Any startup that's lasted through the last six or seven years and survived all this gets a grade of a 'B' at least, I'd say," says analyst Jeff Montgomery of ElectroniCast Corp. "We've been impressed with them."

VCSELs were a new idea when Picolight started, but, like tunable lasers, they've gained accepance even through the downturn. "There's a higher level of integration going into transceivers, and that certainly fits well with VCSELs," Montgomery says.

Picolight's problem was that it tried to expand beyond VCSELs. Like most optical firms, Picolight believed the key to better revenues was better product breadth, and that meant building such things as small-form pluggable (SFP) modules using other companies' lasers. Those products, which weren't very successful and had lower margins than the VCSEL-based modules, have been jettisoned, Hane says.

Picolight continues to focus on storage, data center links including 10-Gbit/s connections, and parallel optics (see Picolight Preps Parallel Push). In carrier circles, the company has decided to avoid the Sonet market and concentrate on IP networks instead. "Sonet and SDH are places where market share changes hands more slowly," Hane says.

— Craig Matsumoto, Senior Editor, Light Reading

redface 12/5/2012 | 3:19:44 AM
re: Picolight Picks Up $13M I heard that long wavelength, single mode VCSELs are highly sensitive to back reflection. Considering that LC/PC connectors are so widely used in datacom transceivers which gives -14dB return loss when not mated with another connector, is there any chance that LW (long Wavelength)VCSELs can be packaged without the expensive isolator? If an isolator is unavoidable, how would it impact the low cost advantage of LW VCSELs compared to 1310 FP lasers?
DarkWriting 12/5/2012 | 3:19:42 AM
re: Picolight Picks Up $13M From the PR...

**Picolight's problem was that it tried to expand beyond VCSELs. Like most optical firms, Picolight believed the key to better revenues was better product breadth, and that meant building such things as small-form pluggable (SFP) modules using other companies' lasers. Those products, which weren't very successful and had lower margins than the VCSEL-based modules, have been jettisoned, Hane says.**

They still show 1550nm SFPs on their web site. As far as I know, nobody makes a 1550nm VCSEL. What's the truth here? If 1310nm VCSELs become cheap enough, won't they cannibalize the 850nm market?

DW
SIVROCX 12/5/2012 | 3:19:42 AM
re: Picolight Picks Up $13M Might be right if the 1310 can be produced at the same cost as 850's ( not yet clear to me). The eye safety issue would be easier to contain at the connector end for 1310 vs 850. Something about the NA God used to make the eye and with parallel optics you've got even more power. +£
laserburn 12/5/2012 | 3:19:42 AM
re: Picolight Picks Up $13M Different fibers, different applications:

850 nm is for multimode fiber and short distances.
1310 nm is generally for singlemode fiber and longer distances.

The imminent demise of shortwave has been predicted for years, in fact it keeps on winning due to a) cheaper packaging, laser and assembly and b) most links are under 100m.

lb
BlueWater66 12/5/2012 | 3:19:41 AM
re: Picolight Picks Up $13M The 1.3um VCSELs don't drive profitability. 90% of the cost of an aligned laser is in the packaging. Plus, 90% of the total cost of a module level product is elsewhere. There are inexpensive FPs, VCSELs, DFBs, DBRs and other laser die in the market from established players.

The key question is... why would the VCs put more money into this? PicoLight is in a profit constrained industry. ALL the public company competitors are focused on cash conservation and product margins. They aren't spending money on acquisitions.... especially ones like PicoLight.

That leaves PicoLight in a strange place. They have to grow internally and try to IPO. Given that almost all the value of an integrated module (SFP/XFP/X2) is composed of packaging, circuits and module level testing, why would the VCs want to maintain a bunch of MBE reactors and laser die capabilities? That is a huge capital equipment and cost overhead, which eats into their slim profit. Feeding the die fab also requires a wide range of technical professionals that don't offer any module level value (they know epi, chemicals and small features). Picolight is a module company.

I don't get it. Especially in the commodity SFP/XFP/VCSEL Ethernet space. There might be some nice exits (eventually) for WSS vendors who offer something original.... But probably not this.
whyiswhy 12/5/2012 | 3:19:40 AM
re: Picolight Picks Up $13M You gotta be kidding!

ROTFLMAO!

-Why
DarkWriting 12/5/2012 | 3:19:38 AM
re: Picolight Picks Up $13M **Given that almost all the value of an integrated module (SFP/XFP/X2) is composed of packaging, circuits and module level testing, why would the VCs want to maintain a bunch of MBE reactors and laser die capabilities? **

Why did Finisar buy Honeywell's VCSEL business? Picolight isn't just a module company, check out their web site. Also, for companies that don't make hermetic OSAs, the OSA cost (Rx and Tx) can be a significant portion of the cost, at least 30%. Don't disagree that this company will not be an easy sell in any case.

DW
dodo 12/5/2012 | 3:19:38 AM
re: Picolight Picks Up $13M naughty,naughty Why :-)
optoslob 12/5/2012 | 3:19:37 AM
re: Picolight Picks Up $13M "The 1.3um VCSELs don't drive profitability. 90% of the cost of an aligned laser is in the packaging."

I thought that cheap packaging and cheap electronics was the whole point of using VCSEL's, so am I missing something? I know I saw a presentation a few years ago where the fibers were inserted directly into the VCSEL opening making it "self alinged". I also remember a lot of talk about how it was easier to do 10G VCSEL than a 10G FP because of relaxation oscillation issues on the FP laser.

Does anyone know the real facts?



HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE