Part One of this article looked at Quinta Corp’s attempt to prolong the life of today’s Winchester technology with its Optically Assisted Winchester. Quinta has the backing of its parent Seagate Technology Inc. But another new technology is being touted by California-based start-up TeraStor. Like Quinta, TeraStor has not demonstrated its NFR Near field recording […]
Part One of this article looked at Quinta Corp’s attempt to prolong the life of today’s Winchester technology with its Optically Assisted Winchester. Quinta has the backing of its parent Seagate Technology Inc. But another new technology is being touted by California-based start-up TeraStor. Like Quinta, TeraStor has not demonstrated its NFR Near field recording technology publicly or announced any specifications but it too has attracted significant attention.
By Nat Tunbridge
TeraStor has received $85m in funding from a variety of venture capital firms and companies, one of which is Quantum. Like OAW, NFR uses lasers to enhance traditional magnetic recording techniques. A laser beam is shot along the actuator arm and into the head, where it hits two different lenses which magnify it. It then passes through a magnetic write coil and onto the media. A key difference, however, lies in the nature of the second lens, which the company is calling a solid immersion lens (SIL). TeraStor claims it can focus the laser beam into an incredibly small spot on the recording surface of the disk, allowing for more data to be concentrated into a smaller space. The technology’s title, near field recording, is derived from another of its attributes. At the bottom of the SIL is a flat surface where the laser beam would stop and, if left to its own devices, reflect back on itself. But by bringing the drive head to within a quarter of a wavelength of light (150 nanometers) from the disk, TeraStor makes use of a quantum phenomena called evanescent coupling. This means that the energy of the laser can be transferred to the surface of the media without them having to touch, hence the description NFR. At present, TeraStor is using NFR to produce products for the removable storage market, taking on Iomega and Syquest, rather than joining Quinta in the hard drive market, populated by Seagate, Quantum and others. The fixed drive market is very competitive, says Dr Gordon Knight, chief technical officer of TeraStor. You have to have the fastest access time, the highest capacity, the cheapest box. We felt the removable market was easier to penetrate. Knight claims that the finished NFR cartridge will have 20 GB of capacity and be competitively priced against rivals such as the 2 GB Iomega Jaz, which is expected to retail for about $560. But he admits that the project is already four months behind schedule. Originally planned for release in May, the pilot product is now not due until the fall. The major hitch is getting it into production, says Joan Pinder, executive director of the International Disk Drive Equipment and Materials Association (IDDEMA). A whole new system of delivering light onto disk is being implemented. Trying to do that reliably, repeatedly is going to be a challenge. Both OAW and NFR promise to use patterned media which, although not new, has not yet been put into widescale production because of the technical issues involved. Traditionally, the media in a disk drive is blank and when the drive is manufactured the head has to go through a lengthy process called servo-writing. This involves imprinting a template showing where information can be stored on the media. By using a photolithography technique to premap this pattern onto the media before manufacturing, a lot of time can be saved. Much of the increase in disk density that TeraStor and Quinta are claiming for their products, however, comes from a new recording technique called crescent recording. This involves achieving greater density by overlapping the bit cells that hold the data. Usually these tiny dots are lined up with their edges touching, and the drive reads the whole of each one. But the binary data stored on a disk can only exist in one of two states – one or zero. To determine which state a particular cell is in, the drive needs to read only the edge of the dot, not the whole thing. Accordingly, data cells can be laid on top of each other with only a crescent of the dot showing, allowing 15 cells to fit into the space where six lay previously. Another innovation which is already being implemented is load/unload technology. Currently when the read/write head in a drive is not in use it lands on an area in the inside rim of the disk, which is roughly patterned to hold it in place. But because heads are flying closer and closer to the surface of the disk it has become harder to make this landing zone smooth enough to not damage the head but still rough enough to keep the head still. Load/unload solves this problem by directing the head off the disk completely, onto a little ramp where it is held in place. This also provides greater security if the drives are moved, which is why the technology’s initial implementation has been in the portable arena. Virtually all portable drive makers are implementing load/unload technology, and over the next two to three years the technology is expected to appear in fixed drives as well.
Computer Business Review. á