Long Term Reliability of Quantum Drives

[Scott Baret]

I just refurbished my Color Classic tonight. Overall it's looking wonderful--the new display I put in is free of the geometry faults that were present in the display that came with the computer, the case is in great shape, and the keyboard and mouse are the originals.

However, I have one problem with it--the 80MB Quantum drives I keep putting in are bad! The first one was a ProDrive ELS. The second was an LPS. Neither worked. I'm going to put a 40MB LPS in now, which I know works. Not too thrilled about having to use a 40MB drive, but I don't want to put my bigger drives in this little guy--the IIci is more their style.

I've noticed Quantum's 1990+ 80MB drives (both ELS and LPS) seem more prone to failure than other capacities of Quantums from the same time period. ELS drives seem to fail the quickest, but my 80MB LPS drives have been going bad too lately. Out of five LPS drives, only two work now. Three of my four ELS drives failed.

Then there are my 40MB drives. I have six of them, none have failed completely, and only one has trouble spinning up. For some reason they seem more reliable than 80MB drives from the same manufacturer. All of my 40MBs are LPS drives; I don't think they made a 40MB EPS (at least I've never seen one).

I also have a 52MB that works fine as well as a few other larger capacity drives of both LPS and EPS varieties.

Does anyone else know if the 80MBs are more prone to failure than other capacities? Was their some sort of design flaw with them?

Off to put in my little 40MB drive...at least it will run the IIe Card software (the reason I fixed the CC up in the first place--so much more convenient to use than a big, bulky Apple IIe).

 

[phreakout]

All mechanical hard drives, regardless of manufacturer, are prone for failure. That is the sad fact to realize. As long as they are using a spinning platter to store data, the integrity of these drives will be at risk for failure.

These drives use two motors: one to spin the disc platters and one to maneuver the read/write heads back and forth across the disc surfaces. These can fail by either heat, the coils of wire having a sudden short created and the ball bearings wearing out.

There's also a known problem on older drives that we call "stiction" (pronounced "stick-shun"). Some drive manufacturers placed a layer of material on top of the platters that acted as a lubricant. This allowed the heads to gently be riding above the disc along with a cushion of air, instead of coming in contact with the platters and gouging the surfaces. If the read/write heads are "parked" on a spot of the disk where data isn't stored for long periods of time, this lubricant can fuse to the heads and get stuck together like glue. The best way to temporarily fix this is to hold the drive in your hand (not connected to the computer) and twist with your wrist to break up the stiction. The permanent solution is to replace the drive. By the late 1990s, most hard drive manufacturers used a different material that was less prone to stiction.

The drives, in addition, can be damaged by a head crash. This is where the read/write heads are jolted enough to come in contact with the surface of the platters, gouging them and damaging to the point of drive failure. The chances of getting data back due to head crash are 50-50.

Heat is another common problem with mechanical hard drives. Too much heat can burn out the controller card underneath the drive and also cause metal parts to expand and shift out of alignment. So you'll need to make sure you have adequate ventilation around the drive to get heat away from it.

And finally, you have data image corruption as probably the most common problem. This happens at random, but more often when you are writing new data on top of old data on the drive. Your boot tracks and sectors on the drive could be at risk of being recorded over. Sometimes if you are changing system settings or the like, any changes you make can pose a risk. Sometimes the drive just dumps too much information in one spot, that it gets confused and loses track of what data is where.

Luckily the new solid state drives don't suffer most of the problems that mechanical drives do. You'll risk data corruption as possible problems, but not head crashes, motor failures or stiction for that matter. Once the prices start dropping and these drives are featured more often, then I think the reliability of them will be much better. We can only wait and see what holds for the future.

73s de Phreakout. :rambo:

 

[ClassicHasClass]

All of my "Quanta" are doing well, as are my Seagates.

I distrust Maxtors, but other than a single abused external drive I have purged my entire network of them.

The hardest working, longest living drive in my stable is the Seagate Hawk in my IIci server, which has been running since 2000 continuously and is certainly much, much older than that. I do have an old Rodime 20 Plus for my Plus, but it is used very lightly.

 

[Osgeld]

Most of my quantum drives have died off, in all fairness most any other drive that age has also with a few exceptions, including a ibm, a fijutsu, a few maxtors, and a lot of seagate drives

I have the worst luck with Western Digital drives of all ages, old drives Conner

 

[Dog Cow]

The remains of my Quantum 80 MB HD: http://macgui.com/gallery/showphoto.php?pic_id=1286

That was my first hard drive, used in the SE/30. It just got harder and harder to spin up as the years went by. xx(

 

[Scott Baret]

My most reliable drives have been IBMs (of all types--SCSI, IDE, 3.5", 2.5", and also floppy drives) followed by Conner desktop drives (not laptop drives) and Quantums of capacities other than 80MB.

I distrust Toshiba (all of mine have died, ranging in vintage from 1987-2005) and Maxtor.

A thought--have MTBF figures been accurate on these drives over the years? Those of us who have owned Macs since the day they were purchased should know about how long we have used our machines and, especially on older Macs where we can't spin down the drives by hand, should have some idea how close the drive is to that MBTF interval.

(I wish hard drive manufacturers would give us the standard deviation of that MTBF number as well as the median and mode, as I feel it could help us in the long run--the mean may be accepted by most as the de facto standard of central tendency figures, but it isn't always the best or only option).

Does anyone know the MTBF on these Quantums, or, for that matter, any of Apple's stock drives (MiniScribe, Conner, etc)?

 

[LCGuy]

For what its worth, I actually had an LPS die recently...I knew it was on its way out, for a good couple of months it was VERY LOUD due to the bearings wearing out. Now its dead...oh well, nearly 20 years out of a hard drive is pretty damn good, really.

The remains of my Quantum 80 MB HD: http://macgui.com/gallery/showphoto.php?pic_id=1286
That was my first hard drive, used in the SE/30. It just got harder and harder to spin up as the years went by. xx(

Hah, I had an 80MB ProDrive ELS as my very first hard drive (the original drive in my very first LCIII) too. Mine is also dead, though it died in 1999.

For what its worth, I actually had a 160MB ProDrive ELS for a short time last year that actually spun up. I remember saying that I was fairly sure it was the only ELS in my entire local area to still spin up. Now its dead.

 

[tmtomh]

I've come across one or two dead Quantums, but that's gotta be out of about 100 or more. In my experience they're extraordinarily reliable.

I also agree that Maxtors are relatively unreliable. Apparently - and I'd never heard this before about any other drive - they have a tendency to get their firmware corrupted, resulting in a drive that seems to spin up but is unreadable. I've experienced this with a 40GB IDE model. What a waste.

 

[Dog Cow]

The remains of my Quantum 80 MB HD: http://macgui.com/gallery/showphoto.php?pic_id=1286
That was my first hard drive, used in the SE/30. It just got harder and harder to spin up as the years went by. xx(

Hah, I had an 80MB ProDrive ELS as my very first hard drive (the original drive in my very first LCIII) too. Mine is also dead, though it died in 1999.

That photo is from Feb 2010, but the last time I successfully booted from that disk was sometime in 2009. No idea when. The sticker on the drive says EPROM 1992. I have reason to believe that it wasn't the original HD in the SE/30, which I think came from ISU.

I have the startup sound of that HD burned into my brain. I first got the SE/30 in 1998 and had it until about 2001 or so. I continued using the drive in a Quadra 605 which was given to me sans-HD.

 

[Scott Baret]

The earlier Quantums (pre-1990) were somewhat noisy, slow, and unreliable. Apple used these in most SE/30s (some examples from 1991 may have a newer drive). These were also common in most Macs aside from the 20MB SE until the ProDrive LPS and ELS came about. I remember hearing about Apple drives being unreliable back when these were fairly current. Quantums became more dependable in the early 1990s but the long term reliability, of course, seems to be a toss-up based on what everyone has written here so far.

It was also common to pull a drive from an LC in the late 1990s and stick it inside another Mac. Many recommended this about ten years ago when LCs were common and inexpensive/free on the used market, especially because the drives were better performers than anything the SE/30s, SEs, and some Mac IIs contained.

I only have two pre-1990 Quantums--a 40MB and a 105MB. (Only the 40MB is Apple-branded). I've had a few die on me, and most have been 80MB in capacity. Of course, there could be extraneous factors here too (namely the former life of an SE/30 or Mac II as a server). Additionally, it could be that a Mac originally had a 40MB or 80MB and someone replaced it with a larger drive.

Most Apple-branded ProDrive LPS/EPS seem to have a ©1990 EPROM sticker on them. I think they were this way until sometime in 1993. The ProDrives were also common as non-Apple drives, and any older LaCie unit will contain a Quantum, as LaCie used to have corporate relations with Quantum (a manual from a drive I got in 1993 says "LaCie: A Quantum Company" on its cover).

 

[Nathan]

Who knows, the quantum in my LC II worked last time I tried. I suspect it got light use, but only time will tell when it will fail. As for modern hard drives I'd be inclined to stick to western digital (wd), especially since my web server is running on an old 9GB Western Digital drive. I half-expect it to die any time now, but it's been working fine for the last 10 years maybe and before that it was in the family computer for probably a couple years.

 

[0xF2]

Had two Quantum LPS 240 drives fail in a row in a LaCie Tsunami enclosure lately. Horrible (bearing?) rattle sounds. I wonder if I am not shutting them down correctly, as in needing to park heads or some such, or it is just age catching up with them. LaCie's enclosure running them on the side probably also does not help.

 

[kkritsilas]

Running hard drives on their sides should be fine. Most hard drive vendors have only one restriction on position, and that is to not run them "nose down" (nose being the front of the drive, or the opposite end of where the drive connector is). Drive manufacturers will deny warranty coverage for that. Some drive vendors had restrictions (in the past) about operating drive upside down (i.e. the drive electronics board being away from the ground), but that seems to have gone away more recently.

Heads haven't needed to be parked for a long, long time, and I am pretty sure that SCSI drives, being higher end drives, really never did have to have their heads parked. They automatically park their heads on power down, unless the drive is already damaged. The very early hard drives that did need to have their heads parked, used a stepper motor to position the heads, so on power down, the heads would be sitting on, or just above the magnetic surface. The voice coil positioning head positioning system used with SCSI, and most higher end drives automatically retracts the heads on power down, and locks them so they are not above the magnetic surface.

Without opening your drives, it is speculation as to what is wrong with them. The horrible sounds that you are hearing may be the center shaft bearings, or thickened/dried out lubricant making the drive bearings sound that way. It could also be a head crash, which is the by product of something getting between the heads and magnetic surface (it can be magnetic material, usually caused by the heads contacting the magnetic surface and gouging out magnetic material from the rotating platter). I suppose it could also be material that got into the drive if the seals on the drive were compromised. The material doesn't have to be very big, but even so, will cause a head crash. The leading cause of head crashes is moving the drive, or exposing it to vibration, while in operation. Age alone will cause lubricants to dry out, and not allow a drive to spin up to normal operating RPM, so the drive never puts out a "ready signal" or the heads won't unlock and move over to the magnetic surfaces.

 

[3lectr1cPPC]

The main thing that kills quantums is the rubber bumpers on the head assembly - they turn to sticky goo and the heads get stuck. Some can be fixed by taping over the rubber or replacing it, but many have the rubber under the platters. If you have one of those drives, it's game over.