OT- repairing internal HDDs, the mechanical side - Page 3
Close
Login to Your Account
Page 3 of 3 FirstFirst 123
Results 41 to 46 of 46
  1. #41
    Join Date
    Sep 2002
    Location
    People's Republic
    Posts
    5,953
    Post Thanks / Like
    Likes (Given)
    710
    Likes (Received)
    3710

    Default

    Quote Originally Posted by Joe Gwinn View Post


    Banging away isn't needed.





    But it is fun

  2. #42
    Join Date
    Nov 2009
    Country
    UNITED STATES
    State/Province
    Massachusetts
    Posts
    603
    Post Thanks / Like
    Likes (Given)
    0
    Likes (Received)
    202

    Default

    Quote Originally Posted by EmanuelGoldstein View Post
    But in the old days, you could take the platter out and resurface it. They were either copper or plated copper. 8" disk, as I remember. They also had external pulleys to drive them - to change from 60hz to 50hz you changed the drive pulley. Shugarts, K&T's with the C and some D controls had them. Later D's went to scsi, a magnificent improvement
    The platter I have hanging on the wall is 14" in diameter, and came out of one of the early Winchester disk drives with removable platter pack. I fixed that beast a number of times, back in the day when it wasn't that different from working on an automobile. But a gouged platter could not be fixed.

    War story: This drive was the size of a washing machine, with a glass window in the top, which hinged up to permit the disk pack to be removed. As I recall, the drive held something like 100 MBytes, which was very impressive in the day. My job was to write the I/O Driver for it. And things were not going well. I kept getting random read errors that correlated with nothing, and nothing I did made any difference. As desperation grew, a crazy thought came to me: I wanted to see the platter in motion, and given the glass top, this could be done with a strobe light. Which we happened to have, an old General Radio strobotac (?) unit. The disk platters spun at 2400 rpm. When the motion was frozen by the strobe, all seemed well. But what is that tic - every so often, the stopped image jumped five or ten degrees.

    Hmm. What the hell could do that? The disk pack has immense rotational inertia, and is driven by a flat belt which isn't nearly stiff enough to impart enough torque to cause such a jump, even if the motor tried and gave its all. Only the main bearing can do this.

    Call the manufacturer, Control Data Corp. Who sighed and offered my boss a deal - CDC would fix it for free, if we would pay for travel and lodging. Agreed. Replaced the main bearing assembly. Turns out he could not get the drive back together, because the instructions on how to align the voice coil actuator were not correct. A co-worker and I, who both worked on cars, knew exactly what to do: Use ordinary paper (0.003") as a shim to hold the voice coil away from the magnet assy, clamp everything, so now it can move radially without hitting or rubbing anywhere.
    Last edited by Joe Gwinn; 04-20-2021 at 05:47 PM. Reason: fix typos; add missing details

  3. Likes JDennis, Screwmachine, Tbartoletta liked this post
  4. #43
    Join Date
    Apr 2016
    Country
    UNITED STATES
    State/Province
    Texas
    Posts
    421
    Post Thanks / Like
    Likes (Given)
    167
    Likes (Received)
    175

    Default

    having taken a bunch of hard drives apart (magnets are really handy and the kids liked the platters) there is no way on earth you are going to take a platter (or 5) out, stick it back in the drive and have it work. No way.

    Now, if the problem is electronic, as others have said you could buy the same make and model of drive then swap out the electronics board on the outside. That's pretty feasible and I have heard of people doing that successfully.

    But the platters? Not going to happen.

  5. #44
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    10,289
    Post Thanks / Like
    Likes (Given)
    591
    Likes (Received)
    8385

    Default

    This is not the old days.
    If you can get to spin and the tracking to get up some good maybe.
    No not the same, interchange impossible.
    If needed I see the worry of data recovery and all that kind of so silly.
    Do not know it you have a black, grey or white hat but this worry .....
    Why did i not keep some CDC platters as wall hangings?

    What I see here is a concern about data theft.

  6. #45
    Join Date
    Oct 2017
    Country
    UNITED STATES
    State/Province
    California
    Posts
    1,267
    Post Thanks / Like
    Likes (Given)
    396
    Likes (Received)
    435

    Default

    Maybe it's been mentioned but modern hard drive platters are sectored during manufacture and then installed in the drive. The sectoring data is stored on an eeprom in the control circuit. You need a donor drive of the same make, model, and board revision to transplant the platters into. Then you need to desolder the eeproms and swap them between boards. If the eeprom is dead your data cannot be recovered. I forget the specifics of what is stored on it that makes it so critical, but it has something to do with how hard drive read heads aren't accurate enough to position themselves in a way that allows them to write their own sectors without tracking information.

    Swapping read heads is possible for a hobbyist if you make a decent laminar flow work station. If you have physically damaged disks that will take out the read head then it's still possible to recover some data but it gets difficult and expensive, fast.

    There are machines with accurate enough positioning to read a platter of any format. Those machines are very expensive, and similar to what is used to sector the disks in the first place. I'm also pretty sure whoever makes them won't sell them all willy nilly.

  7. #46
    Join Date
    Jan 2007
    Location
    Flushing/Flint, Michigan
    Posts
    10,289
    Post Thanks / Like
    Likes (Given)
    591
    Likes (Received)
    8385

    Default

    Quote Originally Posted by Strostkovy View Post
    Maybe it's been mentioned but modern hard drive platters are sectored during manufacture and then installed in the drive. The sectoring data is stored on an eeprom in the control circuit. You need a donor drive of the same make, model, and board revision to transplant the platters into. Then you need to desolder the eeproms and swap them between boards. If the eeprom is dead your data cannot be recovered. I forget the specifics of what is stored on it that makes it so critical, .
    All this makes it see like outsourcing to people with the million dollar equipment makes sense after trying the spin up, frezzer and wack on the table methods.
    On the other hand these hack methods may destroy some data you want.
    One method of killing a hard drive is shoot it with a 22 or drill a hole. Amazing what can still be recovered nowadays.
    Compared to the old 3340 or CDC platter drives today's hard drive are light years in function and density.


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •