Jump to content

Are 3.5" diskettes worse than I rmember them ?


Recommended Posts

I had a large hoard of 3.5" disks.  These were not optimally stored at all (non ideal humidity, cold temps, hot temps, etc).  While I haven't had many issues with 5.25" disks, even when not ideally stored, I found a lot of issues with these 3.5s.  I have a stack of zip disks from the same lot, and given the success of the 3.5s, I don't think I should waste my time (possibly damaging my zip drives) trying to see if any are ok.


I also have a bunch of 5.25" drives from that lot... I have not started to examine those yet.


Back to the 3.5s....


Some had small 'tendrils' of mildes/mold ? growing on the surface of the disks, others having an odd look to their surface.  If I found mold I tossed 'em and didn't even put them in the drive.


None really read, even when the surfaces of the diskette looked good.


Even diskettes that looked absolutely fine would read... and attempts to format would result in track 0 bad.


I gave up wasting my time on these and I've trashed the whole lot.


I used multiple drives, cleaned the heads multiple times (by hand) (most of the time I did not see any issues, but cleaning them multiple times in case that was part of the issue)..... all to no avail.


Bulk erased, and tried to format.... track 0 bad.


I gave up wasting my time, since most of the lot was 1.44 diskettes and I have tons of those (though I don't know how those will be, as most were unbranded 'cheap' diskettes.... so maybe I don't have anywhere near as many good 1.44 diskettes as a I thought I had.


At this rate, I'll take a 5.25" drive over a 3.5 any day.....  there has been almost no media (so far) that I have not been able to read.


I also have about 20 original 3.5" game disketts from that lot.  I have to presume their junk now too :-(.  PC classics, Marble Madness, Marble Madness II, Thexder, Thexder sequel (I forget the name), etc.


So, what's the deal with 3.5s.... is this not unusual ?  Or am I just horribly unlucky :-).




Link to comment
Share on other sites

I have started to test the 5.25" floppies I don't mind erasing from that massive lot of floppies....  not good.  3 floppies formatted to 360K fine.

Some floppies had evidence of mildew/mold/etc on the media surface (some faint, some more obvious).  A few had damage (dents), a few had scratches.


All in all out of ~40 disks tested, 3 are fine, 10 format with errors on side 2 (so good for single sided 8 bit generation), and the rest had to be disposed of.


I'm going to pitch the zip disks/etc from their too after examining them.... but if the floppies didn't fair well, I can't imagine the zip disks fairing any better.


I'' have to take my microscope to some of these.... see if I can get a good luck at the ones that look fine.... maybe there will be surface contamination I can't see.


I hope I can salvage some of the original games.  I've come up with a process to open the 3.5" disks... I figure if they don't read well, I'll take the media out, clean the media and reassemble then retest the floppies.

It will be work for sure, so probably will only do this to the titles I really want to save the most.  Its a shame too... probably ~ 40-50 or more original PC games... and they probably all have issues.


So, the issue wasn't the 3.5s.... its all of that media, and had to do with the long and poor storage conditions (some of it my fault as I forgot I had the stuff and thought there was just hardware in the boxes/etc).



Link to comment
Share on other sites

3 hours ago, cwilbar said:

So, what's the deal with 3.5s.... is this not unusual ?  Or am I just horribly unlucky :-).

3.5 media was practically the industry's first real big push to increase data density on consumer-level products. Products you bought at the computer superstore, you know.


They didn't really have a handle on it. There were many formulations of oxide/lube/binder that were literally geared toward a brand of head manufacturer and drive density (400K, 720K, 1.44MB.. and others). And to top it it off, each brand of diskette was slightly different even among same-sized drives. All these formulations, companies, densities, sizes, levels of coercivity, track widths, and more, was a real problem. A lot of experimentation was going on and unfortunately the consumer bore the brunt. The poor saps were always on-edge when it came to retrieving info.


Consider that the drives themselves were still stepper-motor based and not closed-loop servo. This meant that disks written by one drive would tend to be readable only in that one drive. With any semblance of reliability that is. While not immediately apparent, the disks would age slightly. Enough so the track and head alignment simply couldn't produce enough flux to reliably read.


Even early hard drives were like that. But they had the advantage of being "tuned to their media". And you could use SpinRite to redo the tracks every few years.


That's why (those multiple reasons) 3.5 drives were always on the fringe with reliability. Factory-pressed disks didn't have that problem as much. They were written with more powerful heads on more durable media so as to avoid customer returns. There was a time when I would write a 3.5 disk and make note of which computer did it. I had few if any problems. But I would not trust the things today beyond transfers to vintage rig without USB or something.


5.25 had similar issues, but they were beefy and brawny compared to 3.5 and thus would tolerate all kinds of slop and variances in the chain.


The Iomega Zip Drive was different in that the media formulation was more tightly controlled. You either made it to this specification or you didn't. the different branding was just that, branding inserts & labels. I think Fuji was the prime maker. The Zip (I believe) also used a form of closed-loop positioning. If one track was a few microns out of position the circuitry hunted around and found it. This closed-loop servo meant that one drive would not have a problem with another drive's disks. A third thing advantage of the Zip was that all disks had a permanent set of factory embedded guide tracks. Or reference tracks. This helped maintain law and order.


All modern IDE style disks from the late 80's and early 90's till today use some of those techniques. And today they allow for 500,000tpi and higher if you do dual-stage actuation.


With contaminated 5.25 and 3.5 disks I've often split the housing and washed them down with various combos of water, alcohol, and hand soap. Works good for things growing on the surface and has allowed me to save many otherwise unreadable disks.


It is also important to note that a fungus infected disk can spread it to another if you don't do a head cleaning in-between swapping disks. These sorts of issues are becoming more prominent as time goes on. Lucky they lasted this long. Though a properly cared for diskette ( can and will outlive today's flash media. Believe it or not!






  • Like 1
Link to comment
Share on other sites

I don't think I have any 3.5" disks that don't work. I have a lot of 5.25" disks that don't work. So, anecdotes.


That said, this is all super-old media that was never intended to last longer than about 10 years. I'm pretty sure that's a real thing that I am not just retroactively applying. I remember reading that. Even CD-Rs had a 10-15 year quoted lifespan.


Link to comment
Share on other sites

There are many factors in determining media longevity. Storage, manufacture, usage.


The times I had 3.5 fail on me were when I mis-matched densities. They worked for a few weeks, months even, then simply became unreadable in certain drives.


The IBM PC 5.25 format had similar problems with 360K and 1.2MB disks. Though for whatever reason unbeknownst to me, people seemed to match the 2 different densities' disks & drives so much better. And therefore reliability was greater than 3.5.


Consumer CD-Rs from what I recall were made to last 5-8 years. In practice they go a little longer. the more expensive ones with stable dye might make 10-15 years. Many people think they are permanent. Not true at all. Not unless you get something like M-Disc. Then the discs are likely to outlive the methods and tech needed to read them.


I've got HHDs from the 80's and 90's which are working just fine have retained data beautifully. So. If I was to give a consumer advice about data preservation, any plan would include 2 copies on media that can be migrated from/to over the years. M-DISC and HDD are current choices. And spare equipment to read the stuff.

Link to comment
Share on other sites

I buy floppy disks when ever I find them at thrift store and sometimes they are even still in the box... I've had so many 3.5" disks fail to format. Now I mostly use them with a E-MU digital sampler for music production so they need to be specially formatted, but many of these disks fail the check at the end. 

Link to comment
Share on other sites

9 hours ago, cwilbar said:

These were not optimally stored at all (non ideal humidity, cold temps, hot temps, etc).


That's your problem.  I have well over 1,000 floppies from 3½, 5¼ and 8 inch, used in my TRS-80, Atari and PCs.  If I don't get a perfect format (zero bad sectors), I consider the disk unreliable and either toss it or use it as a test disk.  I'd estimate 1 out of 50 might list a bad sector or two after formatting.


It's very rare that I get a disk that won't format and it's usually obvious before even trying.  Several years ago I bought a large lot of 5¼ disks off eBay and when I opened the package, the smell about knocked me out.  Every disk had mold on the media.  I sent the seller photos, he refunded my money and I threw them all in the trash.


I've found 3½ disks to be more reliable back in the day and and today due to the hard shell and sliding door.  I have a 3½ USB drive on my current desktop and have no issues when I put programs on them to use on something like my old IBM P70s.


Also, if a used disk looks good and doesn't format, it may have been formatted on a misaligned drive at one time.  An electric bulk eraser will make it like new again.

Link to comment
Share on other sites

So, in the lot I've been working on (that started this thread), I've faired much better in the 5.25" diskette arena.  I've had to pitch ones with mold/mildew.... but I also had a decent sized lot of disks (most from the same shareware vender) that developed no mold/mildew, ran quiet in the drives and all byt 1 formatted w/o error.  There are still 3.5s left to go through, but so far only 2-3 are fine and format w/o error.


But, then I remembered.....


I have a massive lot of C64 disks (non originals) that I don't even remember where they came from (this is in addition to the diskettes discussed above where I know where they came from, and I know some of their (bad) storage history).  I don't remember where these came from or the condition when I obtained them.  They were also not stored in a good place at some point in their history (garage, shed, damp basement, something).  I've been going through them little by little (have 300+ more to go), and I've found mold/mildew on half ?


The interesting thing is I'm finding patterns in the manufacturers in this lot.  Half of these disks are generic, so I have no idea on those.... but when they are branded diskettes.... I've found a pattern.

Almost all Maxells developed no mold/mildew.  BASF faired well as well.  Memorex... mixed results.  BONUS.... ALL mold/mildew.... not a single Bonus branded diskette was spared.  I've thrown out 150+ diskettes so far.  Only retaining ones with interesting sounding games such that if I feel so inclined in the future, I can clean the diskette and recover the contents (which will likely never happen).  The ones without mold/mildew that did not contain anything (by label) of interest will be double checked, and test formatted.


I did find the manufacturer results of this unintended experiment quite interesting.  Also interesting is the degree in which the colors, surface textures, etc differ between different brands.  You can tell the cheap ones.... probably like sandpaper at the microscopic level ?



Link to comment
Share on other sites

I bought a USB floppy drive a short while back. Wouldn't work on any Windows system. So I hooked it up to the linux box. Went through about 30 old 3.5 disks. Only one was able to be read. I couldn't even format it. 


These disks were nearly 30+ years old, looks like their time has come to toss 'em.


My very first box of 5.25 disks were of the BONUS brand. ALL of them failed. 

Link to comment
Share on other sites

I have had more trouble with 5.25" disks failing than 3.5" disk.


But is it the disk that failed, or is the drive not working so well anymore?   Hard to tell because I have far more 3.5" drive mechanisms than I do 5.25" drive mechanisms,  so if one is having trouble reading 3.5" disks, I can try a different drive,  I can't do that with 5.25" disks.

Link to comment
Share on other sites

Once a disk starts making noise it's this " to shedding. Or has already shedded a track. That material builds up in the head. Not on it, but in it. And you need a wet cleaning to fix that.


Too bad about the moldy disks. Many times they can be washed for reading. But of course they aren't consistently reliable in the future.


Link to comment
Share on other sites

Thought of 'inventing' a disk washer.... using an old drive with a dead head.... but then reality stepped in.... most of these diskettes aren't worth it.  Therefore, any that were 'infected' that had data I might want to salvage (old C64 games) were stored in zip lock in case I ever get to it.... the rest trashed.


If I accumulate enough to clean and recover data from... maybe I'll see if I can come up with something to make the process easier.... but I'll have to gather enough to be worth the effort of not hand cleaning them.


Link to comment
Share on other sites

Oh, in the process of going through these I've been fouled (on 3.5 and 5.25) by dirty head syndrome....  it doesn't take much to foul them up.  On the 5.25" drive, one speck stuck to the head can screw it up.


And drive cleaning diskettes....  might be good for something light... smoke maybe, dust.... but any transfer from the coating to the head doesn't seem to like to come off with those.


So for this mass work, I'm using my old Tandy 2500 SX/33 with the top removed and a floppy cable extended out the front.  The drive sits on top (and in the case of the 3.5" drive with the drive top cover removed.  This allows me to check and clean the heads.  If a disk doesn't run silently, I check the heads.  I've found if the drive runs quiet in the drive, it generally does not cause a problem.


I'm learning more than expected about old floppies.


Conversely to all these disks, I have my original Atari diskettes.... and other than SubLogic Flight Simulator II starting to shed during imaging, all of my disks imaged fine.  A few (possibly due to slightly dirty heads or simply due to cheaper/aged media) are showing circular patterns but they do still read, and they are not scratched (visibly).  But the worst experience those disks had was lots of use years ago, and storage in the finished part of my basement for ~20 years (cool but not damp).


Don't know the life story of these hundreds of C64 disks I'm going through now, but somebody had a gigantic collection of C64 stuff.... far dwarfing my own Atari collection of the day. 


I don't think I mentioned it, but I took apart and checked the zip disks.... mold/mildew growth on the media....  I trashed 'em all.


Oddly a few backup tapes that were in there.... no signs of mold/mildew on the exposed media there.  Some formulations, surface treatments, etc are definitely more resistant to mold/mildew than others.



Link to comment
Share on other sites

Oh, one last note.... it seems much easier to interfere with proper operation of 3.5" disks.... that probably makes sense due to the increase of density, the smaller heads, etc....


Depending on the 3.5" drive, those heads can be difficult to clean by hand ?

Some 5.25" drives too (the one I'm using had a shield above the top head.... I had too remove a screw and (temporarily a PCB) and take the shield off so I can do frequent head inspections/cleaning while working with these.


 Curious if anyone has ever found certain drive make/models to be more robust than others for data recovery/etc.  I've found little difference in 1.44 dives so far, but my quantity is limited, as the Tandy has to have a 3.5" drive set to DS0.  It will not worth with a DS1 drive even with a twist cable. 


For 5.25" work, the Panasonic JU-455-5AAB I'm using on the tandy has been great... (minus having to remove that upper shield for frequent head inspections/cleanings).  I have a different 5.25" drive on my supercard pro rig that has also been great.... but I forget what that one is (it is a 360K, possibly YE data).



Link to comment
Share on other sites

The type of contaminant on the head determines what's going to clean it. Alcohol, water, detergent. Sometimes one alone isn't enough. The correct cleaner for the task.


Alcohol would clean sharpie marks. Whereas water and detergent wouldn't touch it. And you'd want water to clean salt deposits from disks coming from storage located on  seaboards/coasts. Alcohol would just let the it stay there and grind. Water would temporarily dissolve and lube it allowing an absorbent material to wick it all away. This on both heads and diskette surfaces.


Edited by Keatah
Link to comment
Share on other sites

@Keatah... regarding the shedding... definitely something to watch out for... i.e. I think it is old Wabash diskettes (at least 8") if my memory is correct.  I have some of those for DEC... and if you use your fingernail you can w/o any pressure shed material.


However, even back in the day, the noise level of disks differed between brands.... their binders, the 'grain?' of the particles, the coatings/etc... these all made a difference in sound of operation.  Possibly even the cleaning liner made a difference in sound.  The noisier (less silent) ones in my testing are not shedding... but the surface of the media is not as smooth (visually) as the quieter media.


So far the only issue I've ran into in dirty heads is contamination from the media... either a bad disk that shed a bit, or there was contaminant freed up (most of these C64 disks were flippies... who knew which way they were rotating last in the jacket ? ).  So thankfully I haven't needed anything other than a Q-tip and alcohol.... so far ?


It has been an interesting experience.... I've learned from it.  Ultimately the time I've spent on going through these was probably not the best use of my time.... but I hate throwing potentially useful stuff out ?


I did a bit more last night.... finally format tested some of the name brand disks.  The Maxells faired best.


Someone else also mentioned bulk erasing... and I had a couple of disks that had initially formatted with a few errors... after bulk erasing and reformatting they formatted fine.


Further format tests will be preceeded by a bulk erase so that if a disk has errors on formatting I know it is not because of stray or off track magnetic interference.



Link to comment
Share on other sites

Most of the noise comes from the surface rubbing against the liner. I've had all kinds noisy and quiet ones. Doesn't seem to affect lifespan.


When noise happens because of the head, you're in trouble. In Apple drives it's about 300Hz. Give or take as the bunched up shedding oxide rolls around. Once it bunches up, the track is threadbare and you get a warbling tone that's clearly audible.


Otherwise I hadn't really noted much head noise if any.

Link to comment
Share on other sites

On 6/7/2020 at 1:20 AM, spacecadet said:

That said, this is all super-old media that was never intended to last longer than about 10 years. I'm pretty sure that's a real thing that I am not just retroactively applying. I remember reading that. Even CD-Rs had a 10-15 year quoted lifespan.


Deep in storage, I have a set of CD-Rs from Spring 2001 that include a full backup of a website and online database project that I was once working on. (The project was put on hiatus, I changed jobs, and it was never actually finished.) For the 20th anniversary next year, I will have to dig them out and see if they are still readable. They have always been kept in dry, temperature controlled conditions, so this will be a good test.  


I don't think I have any old floppy disks anymore, though there may be some in the depths of my storage unit. 


Link to comment
Share on other sites

I'm getting close to the end of going through these old diskettes of unknown origin (massive C64 collection).  Last container is a plastic toolbox with ~300 disks in it.  I've been through a bit less than half of them.


126 'saved' to recover contents... no mildew, or little, and spot cleaned

64 'saved' will need more extensive cleaning, media to be trashed once recovered

31 discarded, contents not worth recovering, mold/mildew

25 diskettes formatted, most DSDD, some SSDD (second side formatted with errors)


138 disks remain to be gone through


Maxell continues to appear the most mold/mildew resistant.  I think I found one with a tiny spot out of all the Maxell.  Others have been more mixed.  BONUS was a complete failure till this last batch where many were mold free.  The media looked slightly different, so they must have changed their formulation/supplier/process at some point.  Almost every FUJI with a colored jacket has had mold, though some with black jackets had none. 


It would have been more interesting if the diskettes had all been name brands, but many were generic, or had their labels removed, etc.  So no easy way to make any correlations on mold/mildew resistance.


I had completely forgotten I had this many C64 disks.


Link to comment
Share on other sites

I've worked on the Commodore horde....  between non mold/mildew disks and ones with very little that I cleaned up through the head window in the diskette jacket, I have 332 diskettes for imaging/recovery (298 have games (front and back on almost all of them), 22 have GEOS related stuff, and 12 are other/utility).


I have an additional 61 (games) that have mold... half or more enough that I'd want to slice open the jacket to clean.  Most of these are titles I may not bother with (based on the hand written labels).


In the same basement closet (finished portion of basement) I found 20 'brand new' Kodak diskettes.  These were in a plastic bag taped shut.  I believe I bought these (w/o sleeves) years ago at a local computer show.  I never used them.  I found mold on many of these.  So, unless my memory is faulty, at least some of this mold came from while they were in storage.  That closet is rarely opened, so it is a cool dark place.  I do have a few more boxes of PC diskettes in there that I'm going to have to check now too.


The winning brand for mold resistance is Maxell.  hands down the winner.  I think I found one small mold spot on one Maxell disk.  All the others were mold/mildew free.  3-4 different generations of Maxell product.

The losing brand was BONUS and generic.  Hard to grade generic, as you really don't know who made them.  Some faired well, many had mold.  But BONUS... all but one batch of BONUS diskettes grew the most mold/mildew above and beyond even most of the generic disks.  The one group of BONUS disks that did not probably came from the same lot, and clearly were different from the others.  So, BONUS probably had diskettes made my multiple vendors. 


I have many other diskettes from that basement too.... but they were not stored in the dark closet.... and so far I have not found mold/mildew on any of them.


Out of the Maxell disks.... I like the later generation of 5.25" (small almost square label), and the 'mid' generation of disks, with the black/white label.  There were a number of the old red/white label ones too.  All did well on mold/mildew resistance.... but the B&W and the later small black/red square label disks have a smoother surface than the earlier generation.


So far, very few 3.5" diskettes that were in my first mold lot are mold/mildew free.  I do still have all the original game disks to go through from that very first lot (PC software lot).  Those I'll want to check.  Any with mold hopefully can be cleaned and imaged.  Fun old titles like Thexder and its sequal along with other Sierra On-Line titles from back in the day (among others too).



Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Recently Browsing   0 members

    • No registered users viewing this page.
  • Create New...