x
Comms chips

Hot Components Research Rewarded

This week, a couple of optical startups got a lucky break from the U.S. government. As recipients of awards from the Department of Commerce's Advanced Technology Program (ATP), they'll be receiving grants of up to $2 million over two years.

With venture capital drying up, the awards couldn't have come at a more timely moment for Luxtera Inc. and SiWave Inc. A third company, Corning IntelliSense Corp., also won an award.

According to the blurb on the program's Website, ATP is a highly-competitive award scheme aimed at encouraging "high-risk research and development projects that have the potential to spark important, broad-based economic or social benefits for the United States." It's designed to fund research before it reaches the product development stage.

Last year's photonics developments didn't attract the ATP judges: Out of 43 awards, there were none for optical components companies. Possibly the amount of venture capital flowing into the industry put optical vendors out of the running.

This year, however, optical components are back on the agenda, with three out of 40 awards going to photonics firms. Oddly enough, all three companies are developing all-optical switching subsystems. Here are some more details:

Corning IntelliSense Corp.
Ultrahigh-Density Scaleable Digital Control of Micro-electro-mechanical Systems (MEMS)
  • Project duration: 2 years
  • Total project (est.): $5.666 million
  • Requested ATP funds: $2 million
  • Project aim: MEMS are analogue -- they must be moved through very precise angles in order to direct light through an optical switch. Controlling these angles precisely is widely recognized as one of the fundamental challenges in MEMS design. The company plans to develop robust control algorithms that will be implemented on a control circuit connected vertically to the MEMS chips. It will then make this technology available to others through its foundry business (see Big Vendors Acquire MEMS Makers). Luxtera Inc.
    Nanophotonic Integrated Circuits for Telecommunications and Computing
  • Project duration: 2 years
  • Total project (est.): $4.194 million
  • Requested ATP funds: $2 million
  • Project aim: Luxtera plans to shrink entire optical switching systems down to just a few microchips using what it calls "nanophotonic integrated circuit (npIC) platform technology." Exactly how it is going to do this is rather vague, and few clues are elsewhere available. What is known is about Luxtera is that it was founded by Eli Yablonovitch, who is widely regarded as the inventor of photonic crystals -- miniscule structures that can act like perfect mirrors (see Twisty Crystal and The Hole Thing for more details). Luxtera says its project is too futuristic to win venture capital funding -- no, really? SiWave Inc.
    Low Cost, Highly Scaleable Optical Switches Using Digital MEMS Mirror Arrays
  • Project duration: 2 years
  • Total project (est.): $4.066 million
  • Requested ATP funds: $1.978 million
  • Project aim: SiWave is developing what it calls "digital 3D MEMS," comprising pairs of digitally controlled mirrors to direct light in hundreds to thousands of different directions inside an optical switch. In other words, it has found a way to sidestep the problem of analogue mirror control that Corning Intellisense is trying to solve. According to its project proposal, it plans to do this using precise mechanical positioning to eliminate costly feedback circuitry. The upshot, it claims, is a much simpler and cheaper design. In fact, SiWave goes as far as claiming that digital MEMS would be "10 times cheaper, 20 times smaller, require less power, and be more reliable than other 3D MEMS-based approaches to optical switching." The ATP funds will be used to develop and demonstrate a high-capacity switch with 256 input and output ports. It's worth noting, however, that other startups also claim to have invented digital MEMS (see Umachines Claims 2D MEMS Advance). — Pauline Rigby, Senior Editor, Light Reading
    www.lightreading.com Want to know more? The big cheeses of the optical networking industry will be discussing optical switches at Lightspeed Europe. Check it out at Lightspeed Europe 02.

  • Page 1 / 4   >   >>
    FastSwitching 12/4/2012 | 9:34:22 PM
    re: Hot Components Research Rewarded With the death of the LambdaRouter 3D MEMS should have vanished from the planet. Instead the government is throwing my tax dollars after the junk to solve already solved problems. Like SEEDS MEMS has had it's chance and the technology FAILED. Lucky the government didn't get in the way back then. The problem to solve is to find an (all) optical switching technology that performs better than electrical(OEO) switches.
    CogswellCogs 12/4/2012 | 9:34:20 PM
    re: Hot Components Research Rewarded "The problem to solve is to find an (all) optical switching technology that performs better than electrical(OEO) switches."

    Go to www.corvis.com for the answer to this problem. Already deployed. Working.

    Cogs
    FastSwitching 12/4/2012 | 9:34:20 PM
    re: Hot Components Research Rewarded ""The problem to solve is to find an (all) optical switching technology that performs better than electrical(OEO) switches."

    Go to www.corvis.com for the answer to this problem. Already deployed. Working.

    Cogs"

    Cogs, you didn't read my post. The LR was deployed and working. It didn't sell and got canned. The Corvis switch isn't selling if you don't count trials with stock holders. It should be canned. The government shouldn't fund either technology as forward looking work. The question still stands..................

    Fast
    dave77777 12/4/2012 | 9:34:18 PM
    re: Hot Components Research Rewarded If LR did work, it wouldn't have been canned, just shelved until recovery. To give up on it, there had to be some unresolved technical problems. No OOO MEMS switch has ever been commercially deployed anywhere.

    BRW and WCG continued to order OOO eq after they no longer had any equity interest in Corv. And whether they were shareholders or not, the swtches are the only commercially deployed and working OOO switches anywhere.
    FastSwitching 12/4/2012 | 9:34:16 PM
    re: Hot Components Research Rewarded Cool, I started a fight with Corvis people. YOUR TOTAL INCOME LAST QUARTER WAS $90 MILLION.

    Your assumption about the LR is wrong. In 2002 people are actually cancelling products that have no customers.

    I'll go to the mats not counting BRW and WCG as real customers for Corvis. There is so much cross polination.

    To stay on message: If something is deployed, don't spend my tax dollars on it for forward looking work. Use Dave's money first. Fund truelly revolutionary stuff. I'd rather have a discussion about what that is.

    Fast
    pro zack 12/4/2012 | 9:34:15 PM
    re: Hot Components Research Rewarded "The problem to solve is to find an (all) optical switching technology that performs better than electrical(OEO) switches."

    3D MEMS based switches are not all-optical as they claim to be. Have evaluated one from a startup and it's the funniest thing I've even seen: itGÇÖs got a laser, a detector and more hardware and software to just point the mirror, and thatGÇÖs called all optical? LOL! No wonder OEO switches are kicking butts.
    straight shooter 12/4/2012 | 9:34:15 PM
    re: Hot Components Research Rewarded Pro zack,

    Really displaying your lack of knowledge in this area...

    Calient switch is/has been all-optical for some time. Yes, software for GMPLS signaling, management control, et. al., but all-optical data path.
    grateful photon 12/4/2012 | 9:34:13 PM
    re: Hot Components Research Rewarded
    while i agree that control algorithms for MEMS mirror arrays would not be my choice for forward-looking work in the spirit of the ATP awards, you might consider the luxtera project. if eli yablonovich is in it, it almost certainly contains PBG work. that would also agree with the descriptor nano-optical. photonic bandgap materials (PBG's) are the best shot at the holy grail for photonic integration that would lead to the silicon-like manufacturing/economics behind moore's law. wafers of PBG's could be patterned via photolithography, etched by plasma processing, and metallized to produce nearly lossless waveguides, mirrors, and thresholdless lasers. using PBG's, one could ultimately pattern devices well below half a wavelength (1550/2 nm), hence the 'nano-' reference, and thereby increase density of optical integration beyond the diffraction limit. call it the silicon wafer of integrated optical circuits...now that is worth investigating and investing in.

    gratefully yours...
    dave77777 12/4/2012 | 9:34:12 PM
    re: Hot Components Research Rewarded LOL Actually, Corv income was around negative 50 million this quarter. Revenue was only 3 million. But on the plus side, they have cash for five years. Good luck making your debt payments over there at LU.

    Real customers or not, it's still the only commercial deployment of an OOO switch. All the MEMS-based optical switches at all the major vendors were canned because they were too big, expensive, and unreliable. Not one survived into a commercial deployment despite many field trials.

    A thousand tiny moving mirrors means a thousand moving parts to break down. Maybe the big advance that MEMS really needs is completely frictionless nanotechnology (don't hold your breath).


    think_small 12/4/2012 | 9:34:11 PM
    re: Hot Components Research Rewarded Each DLP projector (e.g. from InFocus) has almost 1 million tiny mirrors that are individually controlled, albeit in a somewhat simpler manner than a 3D switch.

    These Texas Instruments mirrors are already frictionless, as they use flexural suspensions, and are similar (but smaller than) mirrors used in the Lucent and XROS devices.

    The reliability of these and similar MEMS mirrors can be extremely high, but that does not necesarily mean that the business case exists for large scale optical cross connects in this environment.
    Page 1 / 4   >   >>
    HOME
    Sign In
    SEARCH
    CLOSE
    MORE
    CLOSE