Macintosh Hard Drive Repair

The Macintosh II was a popular computer in the era before Apple dominated the coffee shop user market, but for those of us still using our Mac II’s you may find that your SCSI hard drive isn’t performing the way that it should. Since this computer is somewhat of a relic and information on them is scarce, [TheKhakinator] posted his own hard drive repair procedure for these classic computers.

The root of the problem is that the Quantum SCSI hard drives that came with these computers use a rubber-style bump stop for the head, which becomes “gloopy” after some time. These computers are in the range of 28 years old, so “some time” is relative. The fix involves removing the magnets in the hard drive, which in [TheKhakinator]’s case was difficult because of an uncooperative screw, and removing the rubber bump stops. In this video, they were replaced with PVC, but [TheKhakinator] is open to suggestions if anyone knows of a better material choice.

This video is very informative and, if you’ve never seen the inside of a hard drive, is a pretty good instructional video about the internals. If you own one of these machines and are having the same problems, hopefully you can get your System 6 computer up and running now! Once you do, be sure to head over to the retro page and let us know how you did!

57 thoughts on “Macintosh Hard Drive Repair

    1. That bothered you more than his pushing the head arm with his finger? I could almost hear the -scritch!- of the head digging into the platter.

      Anyway, why is it that so many people feel compelled to comment redundantly on what every bloody one of us “knows” about the drive getting contaminated? If your choice is a dead drive or an iffy drive, you wouldn’t choose iffy? You aren’t going to spend big bucks sending this old drive out for repair, so what do you have to lose?

      1. Oh yeah the head-pushing was pretty awful also, but it running without a top was the first thing I saw so it surprised me more I guess.

        Also, I was genuinely wondering whether it was as big of a deal as I thought. Some people go apeshit over ESD, yet I’ve never in the past 15 years broken any part of my PC through ESD even when using a vacuum cleaner on the inside of it. I thought maybe the same held true for hard drives and contamination. From some of the other comments I get the idea that it is in fact a matter of 80s hard drives being significantly less sensitive, which makes a lot of sense.

        1. For reference, I had a pc break after vacuum it. Tried moving all contacts, switching power supplies, etc. nothing worked. It did start working a few month later, but I’m more careful now. I now clean out the dust outside, with the help of a soft brush and blowing damp air.

          1. Static. Don’t vacuum electronics without a properly-grounded, anti-static vacuum. The movement of dry air can create discharges. Anti-static vacuums aren’t cheap, though.

            Good, cheap results can be had with an air compressor (any of which automatically create moist air), and a metal nozzle. Hold the nozzle in one hand, and touch the (grounded!) case with the other.

            It doesn’t vacuum, so best to do this outside.

            It eventually works out cheaper than the dozens of cans of R-134a (modern “canned air”) that we’ll all go sooner or later, though, and the compressor has a lot of other applications around the office/shop/garage/whatever. (Tires low on the car/bike/lawnmower? Air compressor. Dusty hair after a day playing with wood? Air compressor. Inexpensive nailers for all manner of household things? Air compressor. Cheap impact wrench for changing tires and other automotive stuff? Air compressor. etc.)

          2. First: I reported comment from “cutandpaste” by accident – I wanted to click a – nonexistent – “reply” button.

            Second: I Like using compressed air, but it can give you bad static discharges. One time I had to clean my vaccum cleaner (burst dust-bag), I did it outside on a freezing-cold (dry) day and suddenly I saw a 10-12mm spark jumpping from to nozzle to the electronics (TRIAC control). Luckily, it survived.

      2. The head pushing certainly seemed like a bad idea until I reversed a bit and realized he was pushing on the rigid section of the arm. While this is still never a smart idea, it does turn dead into iffy at least long enough to get your data from the drive to a reliable storage setup. Working at a datacenter I can’t count the number of times I’ve placed a drive in either a freezer or a hot oven to free either a spindle or a head long enough to recover precious customer data to save the day. Harrowing tasks indeed. Props to TheKhakinator indeed for getting the system to start. I hope he immediately backed the drive up to another unit.

        1. Yeah, as everyone here touched on, if it’s already dead, no matter what you do you can’t make it worse.

          I’m actually using it as the Mac II’s main drive and it’s held up for 2 weeks or so now. As I’m just running old software and not actually creating any real content or anything of value on there, it doesn’t matter if I lose the lot. If this had been a data recovery operation then it would have been essential to back up the data as soon as I got it running.

          1. A dead drive is better than a drive in production that can fail any minute. Unless your only purpose is to get it backed up. But given this is a 25 years old Mac, I suppose whatever that needs backing up must have been done already.

            Given that you can retrieve down to the bare drive, why not just replace the drive? Is SCSI HDD that hard to find? Wouldn’t it be worth while to make/use some sort of SCSI to ATA bridge?

          2. @HackJack: There are a few options.

            An SCA or 68 pin SCSI drive connected with a passive adapter is common, but sometimes require more power and come in (relatively) gigantic sizes. (Mac OS before OS 8.1 used plain old HFS, with a fixed number of blocks per partition. If you formatted an entire 1 gigabyte drive as one HFS volume, an empty text document would be 32k.)

            SCSI to IDE adapters are expensive, but will let you use commodity PC drives, CompactFlash cards, or even small SSDs. (Some Macs made before 1996 have native IDE controllers, typically entry-level or home market machines. After that they started moving away from SCSI, and the original G3s used commodity hard drives and PC100 RAM.)

            There are also a couple products out there that will present a MicroSD card as a SCSI block device. The ‘SCSI2SD’ board is especially popular, and can run off the SCSI bus termination power.

    1. But if the drive is already “dead” what do you lose by going inside to free up a stuck part?
      Once it gets spinning/seeking again, back it up! You know you’ve only bought a tiny bit of time.
      Another tactic is to replace the PCB on the outside of the drive with a known good one. It may not have the
      proper parameters but it might be all you need to get the data.

    1. Luckily these are not today’s ultra sensitive regarding contamination. Only a few hundred megabytes per platter, the head hovers much higher than todays drives.
      I remember you had to format these in the orientation which you will use them. Meaning if you formatted them laying flat and installed in a standing case vertically they would not work. Turn them back to horizontal position and magically all your data was back.

      1. Even back when consumer hard drives were only a few tens of megabytes, opening them was a complete no-no. In fact, other than a big floor-standing DEC drive I saw demonstrated when I was a kid that had user-swappable removable platter packs (!), it’s always been the common notion that one does not open a hard drive outside of a clean room.

        Hard drives, including that of the Mac in TFA, have always been said to be super-high-tech, zero-tolerance devices.

        Something about the size of a particle of dust being smaller than the gap between the head and the media, combined with the Bernoulli effect.

        But I digress. I’ve opened hard drives for the purposes of recovering the data, and never has it made the problem the drive was experiencing worse.

        I once got a tens-of-gigabytes iTunes collection off of a drive that had been submerged in river water for a week, after trying to dry it the old-fashioned way (a week on a shelf). I opened it up, let it dry completely, and despite the traces of river silt plainly visible on the platters, it worked long enough to recover the data.

        So, no, one should not open hard drives, no matter their vintage: Yes the density is much lower on an old drive, but so is the sensitivity of the electronics that read the data of that old drive.

        In the case of TFA, I’d do the following: Open the drive, remove the goopy rubber bits (and I’d insulate any ferrous machining with a layer of outwardly-sticky tape and maybe my own small neodymium magnet as a getter), and read the data onto a modern PC. Having done that, I’d write the data onto a different, non-broken, still-sealed drive that the old PC can grok.

        And then I’d go ahead and do my vintage-computing experiments. But I’d never trust a drive to live very long that has been opened.

        Maybe I’m alone here, but I’ve still got a few 4.5GB and 9GB 7200rpm IBM 9ES SCSI discs kicking around that were ridiculously expensive at the time, and which are completely backward-compatibile with any SCSI device of the time given proper cabling and termination. They’re in a pile of drives waiting for “precious metals recycling” once I get enough to fill a USPS flat-rate box and hawk them on Ebay for enough to buy a new SSD, but I’ve got them, I’ve taken care of them, they worked fine when I retired them, and I’m reasonably certain they’d still be just fine today.

  1. Mimimimi dust in your drive kills it mimimi all bullshit
    ive fixed several drives in a non clean area, they still works. when they spin up they throw the particles into a filter (with cloth thing feels like silica gel bags)

    1. I’ve also fixed a few drives in my time. Even had some working with cover removed, just for the fun of it.
      My first fix, was a drive with a bad bearing. Made the worst noise whenit spun up entered some failure mode i.e. dead on arrival.. Opened it up, added some oil to the ball bearing, booted successfully, and extracted all the data on the drive.
      Might not work with the very latest helium filled drives ;-)

  2. oh my god.
    this is the worst repair job i have ever seen.
    i used to work for a datarecovery comany as a cleanroom engineer disassembling harddrives all day.

    things i noticed:
    the environment is dirty
    no gloves? -> fingerprints all over the platter
    heads stuck to the surface? do NOT move the heads or they will rip off
    heads are moving outside off the platter but there is no parking position ramp? -> do not let it happen or the heads will stick together, bend and break.
    the black stuff he pulls out is also there for a reason
    and what?! did g really boot the system from that vey drive again after opening the disk?! how can you?!

    all in all 1/5 stars on this job. the 1 star is only because it seems like the drive did survive that torture somehow.

    1. A lot of these old drives are very resilient. Not sure why, but they are (might have something to do with, as the other commenters have said, the lower data density and greater head gap). I’ve booted up Mac LC’s that have spent well over a decade outdoors, covered in possum excreta. Sometimes the motherboards even have a bit of corrosion and the things still works. They’re a bit like the old Keith Richards cliche.

  3. I remember using an old printer acoustic hood as a make shift clean room, to allow us to un-stick heads on drives of this vintage. The drives are pretty robust, and we only needed to recover the data, so we were not particularly concerned with long term problems. A quick tweak with a make shift paper clip head lifter and a metric twiddle of the platters and off they spun (most of the time). We used to fix Sagate Archive 2060S and Exabyte tape drives with air dusters and cotton buds. Saved us a fortune in replacement drives.

    1. I have used this when opening camera lenses. The water vapour condenses on the dust particles, and when large enough, fall to the floor and stick to it. As long as there is no ventilation there will be very little dust. be sure to have all the tools you may need, opening a door would quickly replace the clean air with dusty air.

  4. as a partner for a big data recovery company i want to share one important point. If you want your data restored, DO NOT OPEN THE HARD-DRIVE OUTSIDE A CLEANROOM. If they see the disk has been opened they refuse to handle it further !
    first of all they don’t want to risk contaminating their very expensive(!!) equipment, and second, stats have shown the success rate in restoring any data from opened drives is almost nihil.

    You often hear ppl going through all sorts of trouble trying to physically “secure-wipe” their HDD by the most insane means.
    But just opening the disk and putting a single scratch on a platter will, even for forensic services, make it almost impossible to recover any data. Nobody will even bother trying, unless maybe national interests or intelligence matters are at stake, which gives you an idea of the cost involved in even trying.

    1. Yeah, I do hope people watching this video realise that it is only applicable to old drives that you just want to get working again, not modern drives or drives you want to recover data from.

  5. To throw another monkey wrench into Macintosh hard drive repairs: all 680×0 Macintosh (except for the first few) and early PowerPC uses SCSI technology and terminator had to be done right or it won’t work. On internal hard drives it is usually a couple resistor SIPP that needs to be removed if the hard drive is in mid chain, or present if it’s at the end of the chain. (FWIW Mac IIfx were bigger pain in the butt, they used different termination and couldn’t work with simple resistor SIPP)

      1. I’ve had a few macs come through that surprisingly resilient to termination issues. It’s either a lucky combination of software and hardware or the drives I’ve used have some kind of automatic internal termination I hadn’t noticed.

  6. I commented on the youtube vid, but thought it would be a good mention here. Look into getting a set of JIS (Japanese Industrial Standard)bits and drivers. Spendy, but you will love how much easier it makes opening electronics, particularly laptops. Phillips is designed t cam out to prevent thread stripping or heads snapping off. What it usually does is damage the head. JIS have a straight profile, and bite better. Really nice ones have a laser engraved flat surface that acts like teeth to grab.

  7. I have a lot of Conner Hard Drives from 1990-1994 which all have this exact same problem or a similar one (some have gooey rubber, others can’t overcome the land-zone magnet, but the same solution applies to both). But I’m cheap, I just put a strip of adhesive tape in between.

  8. If everyone thinks THIS is cringeworthy, they should have seen the operation I did to completely transplant the platters from a damaged Micropolis 1375 140mb 5.25″ hard drive to a working Micropolis 1375. The data density on even 3.5″ ~700mb Quantums is still so low that I wouldn’t worry too much about damaging the platters if you keep your fingers off of them.

  9. Is it possible to replace such a hard drive with a solid state drive? I’ve got a Macintosh SE with the same problem. I always wondered if I could replace the stuck hdd with a modern sdd with system 6 installed.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.