Axsun Gets $111 Million for Subsystems
It’s not always the promise of big leaps forward in technology that encourages venture capital companies to cough up big sums of money, to judge by the $111 million third round of funding announced by Axsun Technologies earlier this week (see Axsun Closes $111M Third Round).
Axsun is making optical subsystems, but it’s not venturing down the path taken by many other startups, of trying to develop the optical equivalent of integrated circuits.
Instead, Axsun is making subsystems the traditional way – by assembling components on a platform, and guiding the light among them through free space using lenses. However, it’s automated the manufacturing process and it’s found a way of miniaturizing everything.
This may not sound as sexy as cramming everything on to a single chip, but it’s a lot more practical. It addresses two crucial issues for vendors wanting to use subsystems to shorten development times. First, Axsun’s automation makes it easier to ramp up production quickly should demand soar. Second, Axsun’s miniaturization efforts help pack more functionality into smaller boxes.
Axsun’s automation comes courtesy of little metal cradles in which individual components are placed. The cradles ensure very precise alignment of each piece and are made to sub-micron tolerances by a micromachining technique known as LIGA -- a process also used to make micro-electro-mechanical systems (MEMS). (For more on LIGA, see http://daytona.ca.sandia.gov/LIGA/index1.html).
The miniaturization comes from using ultra-thin lenses between components. The lenses are about 100 times thinner than those used in traditional subsystems, according to Axsun’s CEO, Dale Flanders. A typical multifunction subsystem needs around 10 lenses, so reducing their thickness 100-fold has a big influence on overall size, he notes.
Axsun has already launched a spectrum analyzer that’s not much bigger than a credit card (see Axsun Launches Tiny Optical Analyzer). Existing products offering equivalent functions are shoe-box size, according to Flanders. The upshot is that vendors are incorporating spectrum analyzers in their equipment for the first time. Axsun has sold them to a dozen customers, including most of the big system vendors, according to Flanders.
Right now, Axsun doesn’t appear to have much competition, although at least one other startup -- Cenix Inc. -- is making similar claims for OC192 interface cards (see Cenix Backed by Cisco, Kleiner Perkins).
All of this probably explains why Axsun didn’t have much trouble raising its third round of funding – which comprises $101 million in equity finance and $10 million in debt - despite the unfavorable market conditions. Flanders reckons the round would have been “trivial” if he’d started negotiations a month earlier. “As it was, we scaled back our valuation by about 30 percent, compared to prior to the meltdown -– which doesn’t seem too bad to me,” he says.
Axsun plans to use the money to expand its manufacturing facilities from two to five production lines. It expects to use one third of the capacity to make subsystems for specific customers. The remaining capacity will be used for making Axsun's own products. The process is well suited to making dynamic gain equalizers and optical amplifiers, according to Flanders.
Flanders doesn’t expect Axsun’s way of making subsystems to get superceded by developments in optical integrated circuits. He points out that different optical devices are best made in different materials, and right now that embraces something like 17 different materials. As a result, integration may reduce the number of chips in subsystems, but it won’t eliminate the need for automated assembly processes altogether.
Flanders says that current plans envision staging an IPO “towards the end of this year.”
-- Peter Heywood, international editor, Light Reading http://www.lightreading.com