Slashdot Mirror

← Back to Stories (view on slashdot.org)

IBM's Nanotech Drive Research

Posted by Hemos on from the tomorrow-today dept.

cfanjul writes: "IBM seems to be helping nanotech's slow march to end products with magnetic particles that can be made into a storage device with ten times the density of some of today's drives."

2 of 123 comments (clear)

  1. Re:x-late into hours of mp3? by Anonymous Coward · · Score: 5

    Don't be dumb. We're talking about nanotech here, so obviously you could only store small songs. Duh.

  2. Re:IBM stuff in all hard drives.. by Nagumo · · Score: 5

    The heads are a definitely an important product for IBM. And yes, you can find them in other vendor's products. As for the hardest part to make, perhaps, but there is another piece that is just as tough. The flex cable.

    Flex cable is the ribbon that connects the actuator to the electronics. Sounds easy, but you have to remember that this thing is moving (flexing) constantly. Back and forth, back and forth, back and forth, thousands and millions of times. It takes some serious physics to design these parts, while minimizing the costs. Constant movement inside a little oven, and you have to design these things to cost you pennies. Not easy.

    These are just two parts. What else is tough to make and requires significant engineering?

    • Platters which have to be ground super smooth and coated with the magnetic material. The slightest bump will cause all kinds of bad things to happen when the head contacts it.
    • The channel, which is essentially the A/D converter that takes the signal from the head. This is a big piece of R&D, and the specifics are highly guarded between HDD companies.
    • The interface code - controls how the drive interacts with the rest of the world. IDE, SCSI, FC-AL, etc. and of course it has to be tuned to handle the cases where the host adapter companies got the interface wrong.
    • the motor
    • Servo code to control the motor and the actuator. Handles the basics like seeking, and also the more advanced things like load/unload.
    • ESD concerns. The heads are extremely sensitive to static electricity. The electronics are too, but not to the same degree.
    • Electronic board layout. This is a lot of tradeoffs to cut cost.
    • Power and heat concerns. The attention that these two items get is psychotic. The drives today are very efficient machines.
    • Test. Especially when working with the newer interfaces, newer heads (ie. GMR), etc. Lots of work here. The absolutely worst thing a drive can do is return incorrect data and declare it to be correct. Slightly less severe than this is if the drive explodes in a giant fireball. (At least then you know the data is bad.)

    This is just what I can come up with off the top of my head. The HDD world is a great mix of software and hardware (and some really genius R&D people). The cost to enter this market is absolutely enormous. And to remain in the lead requires a constant investment.