ATLANTA -- OFC 2003 -- Simon Cao, the former CTO of Avanex Corp. (Nasdaq: AVNX) and founder of Oplink Communications Inc. (Nasdaq: OPLK) has popped up at the OFC Conference as the president and CEO of a startup called Arasor Corp.

Arasor is, in fact, a new manifestation of Corlux, a company that was founded by a bunch of people from Fujitsu Ltd.’s active components operations in early 2001. Cao joined Corlux about a year later, and soon after that Light Reading tracked him down and ran an article speculating on what Corlux was up to (see Tracking Simon Cao's Secret Startup).

Now we know. Cao told Light Reading yesterday that Corlux was making external modulators using lithium niobate (LiNbO₃) -- the sort of modulators used for long-haul transmission networks. Some of them are on display at Arasor’s booth here (no. 3564).

The change of name, from Corlux to Arasor, reflects the fact that there’s not much of a market for such modulators right now, so Cao and team have adapted their technology for array processing applications.

In essence, Arasor is developing widgets that improve the performance of things like wireless antennas, scanning radar, and cable TV broadcast equipment. Cao says Arasor is already working with a big system vendor to develop the first product, which will go into field trials at the end of this year.

All of this makes use of the original technology developed by Corlux, which eliminates the need for expensive amplifiers when using lithium niobate modulators. Conventional LiNbO₃ modulators need an input voltage between 5 and 6 volts, according to Cao. Corlux successfully reduced this to 0.7 V -- eliminating the need to amplify power coming from CMOS circuitry.

Arasor achieves this using 3D circuitry in lithium niobate. “It forces the electric field into a very confined region so it has the maximum contact with optics,” says Cao.

This not only improves the efficiency of the modulator but also shrinks its size enormously, according to Cao. “It increases density by a factor of 100. Usually, people get 50 modulators out of a four-inch wafer. We get 5,000.”

That, of course, means that Arasor might only need to produce one wafer every year or so, in the current market conditions, which explains why it's shifted product strategy.

In order to get into the array processing business, Arasor is buying in electronic chips and flip-chip bonding them to LiNbO₃ devices it makes.Although the telecom recession has forced Cao to change course, he says it’s nice to have time to do real development work. “A couple of years ago, all the R&D was done in the marketing department,” he quips.

Arasor has a staff of only 17 -- seven in Japan, where the wafer fab is, and 10 in Fremont, Calif., where packaging is carried out. So far, the company has raised $10.5 million from business angels. Cao says Arasor will close its first round of VC funding soon. He declines to give further details but says the amount is “definitely less than $10 million.”

— Peter Heywood, Founding Editor, Light Reading

For up-to-date information about the OFC Conference, please visit Light Reading’s Unauthorized OFC Preview Site.