Loading ...
Sorry, an error occurred while loading the content.

MFM Hard Drive Repair - Update!

Expand Messages
  • Kyle Owen
    I spent the last day or so taking apart and putting back together my 40 MB hard drive from my PC/AT. I found that one head had come unglued from the actuator
    Message 1 of 11 , May 1, 2012
    • 0 Attachment
      I spent the last day or so taking apart and putting back together my 40 MB hard drive from my PC/AT. I found that one head had come unglued from the actuator arm (which may have been from removing the actuator assembly from the case), so I carefully reattached it with a tiny drop of nail polish. I was afraid of using super glue, as the cyanoacrylate fumes go *everywhere*. The last thing I'd want is for the fumes to coat the disk. Clear nail polish seems to have worked well.

      One of two bearings was shot in the actuator axis bearing assembly. After carefully removing the top (bad) bearing, I sprayed it with some penetrating oil, hoping it would loosen it up some. It did, but it was still very rough. I found that if I pressed on one side, it would rotate much more smoothly. It turns out the spring inside the assembly keeps some pressure on the bearing, which I used to my advantage by turning the bearing over to use the better surface. 

      After centering the axis assembly, I proceeded to put all of the actuator arms back onto the actuator assembly. This was a very delicate process, but it went without a hitch, thankfully. I used nuts to separate the heads when getting the heads back over the platters. The nuts would be pushed out by the edge of the three platters, after the heads were fully resting over the platters. The solenoid lock was secured back in place. After using my best judgement on centering the lock and actuator assembly, I put the top cover back on. The circuit boards were next to be installed, and finally the whole thing was back together, some 20 or so screws later. 

      I plugged it back into the AT, and to my utter astonishment, it went straight into GW-BASIC without reporting a single error (like 1701). The hard drive was spinning away, and no clunking was heard (from the actuator smashing back and forth). 

      Making an Advanced Diagnostics Disk for the AT was quite...fun. I had to dig out the Compaq Deskpro 386/20, which has both a 3.5" and 5.25" drive (combo). After making the 5.25" disk from the files on the 3.5" floppy, I proceeded to unconditionally format the drive. It made it through without any problems. Surface scanning revealed some number of bad cylinders, but nothing too problematic. After all was said and done, FDISK and FORMAT were run, and though I ran into some bad allocation units, it seemed to complete properly. DOS 4.01 was installed without any issues, and by golly, it boots!

      All this to say, I am still so surprised that the hard drive was able to be repaired. I normally have the worst luck with these kinds of repairs, but I guess the stars were aligned well enough to get it working again. 

      Also, does anyone have some spare PC/AT keys? I am missing the lower half of my numpad, for whatever reason. It'd be lovely to have an authentic keyboard with a down arrow and delete key, for instance. Thanks!

      Kyle
    • Dave McGuire
      ... Wow. That is damn impressive. Nice work. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
      Message 2 of 11 , May 1, 2012
      • 0 Attachment
        On 05/02/2012 12:48 AM, Kyle Owen wrote:
        > I spent the last day or so taking apart and putting back together my 40
        > MB hard drive from my PC/AT. I found that one head had come unglued from
        > the actuator arm (which may have been from removing the actuator
        > assembly from the case), so I carefully reattached it with a tiny drop
        > of nail polish. I was afraid of using super glue, as the cyanoacrylate
        > fumes go *everywhere*. The last thing I'd want is for the fumes to coat
        > the disk. Clear nail polish seems to have worked well.
        >
        > One of two bearings was shot in the actuator axis bearing assembly.
        > After carefully removing the top (bad) bearing, I sprayed it with some
        > penetrating oil, hoping it would loosen it up some. It did, but it was
        > still very rough. I found that if I pressed on one side, it would rotate
        > much more smoothly. It turns out the spring inside the assembly keeps
        > some pressure on the bearing, which I used to my advantage by turning
        > the bearing over to use the better surface.
        >
        > After centering the axis assembly, I proceeded to put all of the
        > actuator arms back onto the actuator assembly. This was a very delicate
        > process, but it went without a hitch, thankfully. I used nuts to
        > separate the heads when getting the heads back over the platters. The
        > nuts would be pushed out by the edge of the three platters, after the
        > heads were fully resting over the platters. The solenoid lock was
        > secured back in place. After using my best judgement on centering the
        > lock and actuator assembly, I put the top cover back on. The circuit
        > boards were next to be installed, and finally the whole thing was back
        > together, some 20 or so screws later.
        >
        > I plugged it back into the AT, and to my utter astonishment, it went
        > straight into GW-BASIC without reporting a single error (like 1701). The
        > hard drive was spinning away, and no clunking was heard (from the
        > actuator smashing back and forth).

        Wow. That is damn impressive. Nice work.

        -Dave

        --
        Dave McGuire, AK4HZ
        New Kensington, PA
      • Kyle Owen
        Thanks Dave! And your large list of freebies really makes me regret not being able to make it to VCF...those Sun servers sound awesome! Kyle
        Message 3 of 11 , May 1, 2012
        • 0 Attachment
          Thanks Dave! And your large list of freebies really makes me regret not being able to make it to VCF...those Sun servers sound awesome! 

          Kyle

          On Tue, May 1, 2012 at 11:51 PM, Dave McGuire <Mcguire@...> wrote:
           

          On 05/02/2012 12:48 AM, Kyle Owen wrote:
          > I spent the last day or so taking apart and putting back together my 40
          > MB hard drive from my PC/AT. I found that one head had come unglued from
          > the actuator arm (which may have been from removing the actuator
          > assembly from the case), so I carefully reattached it with a tiny drop
          > of nail polish. I was afraid of using super glue, as the cyanoacrylate
          > fumes go *everywhere*. The last thing I'd want is for the fumes to coat
          > the disk. Clear nail polish seems to have worked well.
          >
          > One of two bearings was shot in the actuator axis bearing assembly.
          > After carefully removing the top (bad) bearing, I sprayed it with some
          > penetrating oil, hoping it would loosen it up some. It did, but it was
          > still very rough. I found that if I pressed on one side, it would rotate
          > much more smoothly. It turns out the spring inside the assembly keeps
          > some pressure on the bearing, which I used to my advantage by turning
          > the bearing over to use the better surface.
          >
          > After centering the axis assembly, I proceeded to put all of the
          > actuator arms back onto the actuator assembly. This was a very delicate
          > process, but it went without a hitch, thankfully. I used nuts to
          > separate the heads when getting the heads back over the platters. The
          > nuts would be pushed out by the edge of the three platters, after the
          > heads were fully resting over the platters. The solenoid lock was
          > secured back in place. After using my best judgement on centering the
          > lock and actuator assembly, I put the top cover back on. The circuit
          > boards were next to be installed, and finally the whole thing was back
          > together, some 20 or so screws later.
          >
          > I plugged it back into the AT, and to my utter astonishment, it went
          > straight into GW-BASIC without reporting a single error (like 1701). The
          > hard drive was spinning away, and no clunking was heard (from the
          > actuator smashing back and forth).

          Wow. That is damn impressive. Nice work.

          -Dave

          --
          Dave McGuire, AK4HZ
          New Kensington, PA


        • Dave McGuire
          ... You can always come here to get some.. -Dave -- Dave McGuire, AK4HZ New Kensington, PA
          Message 4 of 11 , May 1, 2012
          • 0 Attachment
            On 05/02/2012 01:05 AM, Kyle Owen wrote:
            > Thanks Dave! And your large list of freebies really makes me regret not
            > being able to make it to VCF...those Sun servers sound awesome!

            You can always come here to get some..

            -Dave

            --
            Dave McGuire, AK4HZ
            New Kensington, PA
          • s100doctor
            ... This is awesome work, speaking as a digital engineer,. and a good description. I hope you took more than a few photos. I encourage you to put this work and
            Message 5 of 11 , May 2, 2012
            • 0 Attachment
              Kyle Owen <kylevowen@...> wrote:
              >
              > I spent the last day or so taking apart and putting back together my 40 MB
              > hard drive from my PC/AT.

              This is awesome work, speaking as a digital engineer,. and a good description. I hope you took more than a few photos. I encourage you to put this work and description on a Web site; I'd be glad to make a Web page of it on my site, with other vintage repair and restoration work. Let me know accordingly.

              Herb Johnson
              retrotechnology.com
            • Kyle Owen
              Hi Herb, Thanks! To be honest, I had no clue I d have even remotely close to this much success. That being said, I did not think to photographically document
              Message 6 of 11 , May 2, 2012
              • 0 Attachment
                Hi Herb,

                Thanks! To be honest, I had no clue I'd have even remotely close to this much success. That being said, I did not think to photographically document my progress. Darn! I figured I'd take the drive apart, screw something up more, put it back together to make it look nice, and maybe hang onto it for parts. But to have a working drive...I still can't get over that!

                I might take you up on the website offer still, if it's okay. I intend to do quite some stuff with my SWTPC 6800 this summer. One of my next projects for sure is making an 8k ROM board to take the place of the (4) 2k ROMs on the main board, as well as a loading utility to go with it (think about not having to wait five minutes to load BASIC!). The pins will be spaced such that no modification of the actual board will be required; just plug it in and go. With some modification to the board, you could instead have RAM installed at $C000, but I'm a purist: no traces shall be cut on my boards! 

                I've not made any significant progress on the 32k RAM board yet. Exams are all this week, so I've been trying to keep distractions to a minimum. However, I expect I will have the RAM card finalized by the end of the month. I am also working on designing a sound card, one using a 4-channel 8-bit DAC, and another using the popular AY-3-8910. The former will be able to fit on an SS-30 card, whereas the latter will likely be built into my RAM card, assuming I can fit it. I've not played much with the AY-3-8910, but it looks entertaining. As for the DAC sound card, I have to figure out a good way of mixing the channels appropriately. I was thinking pots, but it would be a nuisance to have to reach into the case to adjust the levels manually. Any thoughts?

                Is an 8k ROM card or a sound card (FM or sampled) something anyone would be interested in for the SS-50/30 systems? The ROM card would likely be SWTPC-only, as I am not sure if any others used the same spacing, or even a similar configuration of ROM.

                Kyle

                On Wed, May 2, 2012 at 11:31 AM, s100doctor <hjohnson@...> wrote:
                 

                Kyle Owen <kylevowen@...> wrote:
                >
                > I spent the last day or so taking apart and putting back together my 40 MB
                > hard drive from my PC/AT.

                This is awesome work, speaking as a digital engineer,. and a good description. I hope you took more than a few photos. I encourage you to put this work and description on a Web site; I'd be glad to make a Web page of it on my site, with other vintage repair and restoration work. Let me know accordingly.

                Herb Johnson
                retrotechnology.com


              • joshbensadon
                Wow, nice job! Thanks for the great narrative on the repair! Wish there were some pictures to go with it. But I still felt like I was right there with you
                Message 7 of 11 , May 2, 2012
                • 0 Attachment
                  Wow, nice job! Thanks for the great narrative on the repair! Wish there were some pictures to go with it. But I still felt like I was right there with you rolling that bearing with a little pressure.

                  Cheers!
                  Josh
                • Dave McGuire
                  ... Sorry for jumping in...how about digipots? There are ones with tap-up/tap-down control interfaces that d likely be easy to drive with just a decoded I/O
                  Message 8 of 11 , May 2, 2012
                  • 0 Attachment
                    On 05/02/2012 08:55 PM, Kyle Owen wrote:
                    > As for the DAC sound card, I
                    > have to figure out a good way of mixing the channels appropriately. I
                    > was thinking pots, but it would be a nuisance to have to reach into the
                    > case to adjust the levels manually. Any thoughts?

                    Sorry for jumping in...how about digipots? There are ones with
                    tap-up/tap-down control interfaces that'd likely be easy to drive with
                    just a decoded I/O chip select line.

                    -Dave

                    --
                    Dave McGuire, AK4HZ
                    New Kensington, PA
                  • Kyle Owen
                    No worries...I wouldn t be posting to the group if I didn t want people jumping in! :) The only issue I see with using digital pots with my current design is
                    Message 9 of 11 , May 2, 2012
                    • 0 Attachment
                      No worries...I wouldn't be posting to the group if I didn't want people jumping in! :)

                      The only issue I see with using digital pots with my current design is that the chip I'm using is has (4) channels, and each SS-30 card is only given 2-bits worth of address space. If I want to address all four channels individually, that won't leave me with any room for controlling the digital pots. Alternatively, I could have three audio wires coming off directly, and make an optional mixer board with a few op-amps. That would also reduce costs if you already have another mixer, or are only needing a channel or two, etc.

                      I've been looking over the AY-3-8910 a bit, and it sounds like it could actually fit on the SS-30, with some extra decoding most likely.

                      Any ideas as to what else would work nicely on a 32k RAM board? Currently, it's very bare, with only (4) ICs. And it's a big board for just a few ICs: 5.5"x9". 

                      Kyle

                      On Wed, May 2, 2012 at 11:27 PM, Dave McGuire <Mcguire@...> wrote:
                      On 05/02/2012 08:55 PM, Kyle Owen wrote:
                      > As for the DAC sound card, I
                      > have to figure out a good way of mixing the channels appropriately. I
                      > was thinking pots, but it would be a nuisance to have to reach into the
                      > case to adjust the levels manually. Any thoughts?

                       Sorry for jumping in...how about digipots?  There are ones with
                      tap-up/tap-down control interfaces that'd likely be easy to drive with
                      just a decoded I/O chip select line.

                              -Dave

                      --
                      Dave McGuire, AK4HZ
                      New Kensington, PA


                      ------------------------------------

                      Yahoo! Groups Links

                      <*> To visit your group on the web, go to:
                         http://groups.yahoo.com/group/midatlanticretro/

                      <*> Your email settings:
                         Individual Email | Traditional

                      <*> To change settings online go to:
                         http://groups.yahoo.com/group/midatlanticretro/join
                         (Yahoo! ID required)

                      <*> To change settings via email:
                         midatlanticretro-digest@yahoogroups.com
                         midatlanticretro-fullfeatured@yahoogroups.com

                      <*> To unsubscribe from this group, send an email to:
                         midatlanticretro-unsubscribe@yahoogroups.com

                      <*> Your use of Yahoo! Groups is subject to:
                         http://docs.yahoo.com/info/terms/


                    • David Gesswein
                      ... The standard way around that is you make one of the address an on board address register which then selects which device on the board another address
                      Message 10 of 11 , May 3, 2012
                      • 0 Attachment
                        On Wed, May 02, 2012 at 11:57:42PM -0500, Kyle Owen wrote:
                        >
                        > The only issue I see with using digital pots with my current design is that
                        > the chip I'm using is has (4) channels, and each SS-30 card is only given
                        > 2-bits worth of address space. If I want to address all four channels
                        > individually, that won't leave me with any room for controlling the digital
                        > pots.
                        >
                        The standard way around that is you make one of the address an on board
                        address register which then selects which device on the board another
                        address accesses. Does slow things down since extra writes are needed
                        when you wish to talk to a different device. If one of the devices doesn't
                        use all the data bits you can take the extra bit for a pot/other chip
                        select.
                      • s100doctor
                        ... I ll contact you offline from the MARCH list. I had forgotten we had been in email contact about your SWTPC 32K RAM design. I ll be curious to know how
                        Message 11 of 11 , May 3, 2012
                        • 0 Attachment
                          Kyle Owen <kylevowen@...> wrote:
                          >
                          > Hi Herb,
                          >
                          > Thanks! To be honest, I had no clue I'd have even remotely close to this
                          > much success. That being said, I did not think to photographically document
                          > my progress. Darn! I figured I'd take the drive apart, screw something up
                          > more, put it back together to make it look nice, and maybe hang onto it for
                          > parts. But to have a working drive...I still can't get over that!
                          >
                          > I might take you up on the website offer still, if it's okay.
                          > >

                          I'll contact you offline from the MARCH list. I had forgotten we had been in email contact about your SWTPC 32K RAM design. I'll be curious to know how many MARCH members have those systems - I know a few, Bill Degnan showed one last year at VCF.

                          Herb
                        Your message has been successfully submitted and would be delivered to recipients shortly.