F-Prot was the last virus scanner I ever use. I think I have a floppy with MacAfee AV on it somewhere.

Maybe your hub screw is slipping.

I think I might have solved it, one of the R/W heads snapped off of one side of the mount during the install of MASM and I think that particular disk was relying on that side of the disk to read directory. Will have to try and realign/resolder the head back into place and somehow align it using IMD. I've attempted to solder it but the azimuth of the head must be JUST off, because tracks 22-40 won't read data reliably when the head is aligned for tracks 0-20. Quite disappointing because I had JUST finished finally aligning and tuning the drive just before this happened.

EDIT: when testing alignment, the good (fixed position) R/W head (Head 0) reads perfectly on all tracks and all sectors. The head that has broken off the mount is Head1 (top head).

Arkanix wrote on 2021-07-21, 06:33:

EDIT: when testing alignment, the good (fixed position) R/W head (Head 0) reads perfectly on all tracks and all sectors. The head that has broken off the mount is Head1 (top head).

FYI it's not only the head position and azimuth - and keep in mind there's also a bit of side to side play, which might look like azimuth issue. There's also the pressure of the head, that's very important especially for head 1 since head 0 is helped a bit by gravity (although it too needs pressure from head 1 to keep any surface wobble from affecting head to media distance).

Deunan wrote on 2021-07-21, 09:21:

FYI it's not only the head position and azimuth - and keep in mind there's also a bit of side to side play, which might look like azimuth issue. There's also the pressure of the head, that's very important especially for head 1 since head 0 is helped a bit by gravity (although it too needs pressure from head 1 to keep any surface wobble from affecting head to media distance).

Managed to get it fixed up. One side of the bracket that holds the top head had snapped, so the head had to be soldered back onto its mount. It had to be soldered at JUST the right angle otherwise it'd read the outer tracks but fail to read the inner tracks due to the smaller radius of the tracks closer to the centre. Once that was done actually aligning the head assembly was a doddle in ImageDisk. Get it to read 18 passes on track 0, then make sure it reads 18 passes successfully and reads the data reliably at track 39 and test a few random tracks in between.

I managed to find a disk image of the original dead MASM 5.1 Editor disk and used a spare disk to make a copy and transfer the label. The original MASM 5.1 Editor disk was the issue, but it wasn't a virus. The surface of the disk had failed and was depositing itself on the drive head. This caused excess friction which caused one side of the head to snap off its mount.

All fixed up now though. New copy of the disk, heads fixed and aligned, drive reassembled and situated back in the 5160.

Good stuff. Well done.

Arkanix wrote on 2021-07-22, 07:01:

Managed to get it fixed up. One side of the bracket that holds the top head had snapped, so the head had to be soldered back onto its mount. It had to be soldered at JUST the right angle otherwise it'd read the outer tracks but fail to read the inner tracks due to the smaller radius of the tracks closer to the centre.

I never had to do a repair like this but I remember reading about this issue. I have some "suspect" second-hand floppies but they don't seem to be progressing into worse state, I do wonder if maybe some heads work better then others - due to shape maybe? Keeping them clean helps as well. Also there was a mention of unusual noise when floppy surface is peeling off and accumulating on the head - never witnessed that either. I have one noisy floppy but it looks fine, seems to be a problem of the envelope, not media.

I use a scope+IMD combo to align heads, first with old floppies that I won't miss much if they get damaged and when that works well I switch to original software - the assumption here is these floppies were written with near-perfect aligment so are good templates.

BTW I wanted to ask, did you install MASM to use it or just for the editor? I switched to NASM but the DOS version requires 386+ CPU (and it better be a fast one). Most often then not it's easier and faster to develop DOS apps in VM or using DosBox and only doing the low-level HW testing (and benchmarking) on actual DOS PC.

Yeah I saw a head rip off back in the day, that drive got tossed. Back then when I needed to align one though, I discovered that DR-DOS original disks were real easy to coarse tune to by fudge and feel. Then when it would read those near perfect, I'd stick a microsoft disk in and tweak for that. Then finally a generic ready formatted disk. What made me stop using floppies in contemporary at the time machines about a decade and a half ago, was that I had 3 machines with floppies, and at least 2 of them were drifting off, such that a disk one of them wrote, one of the others couldn't read it. But it was hard to tell which was the best one, and which was the worst one. There was only a few disks in use for sneakernet file transfers at the time. (It was that era where you couldn't trust routers not to leak internal network traffic, or my router was of that era) Anyhoo, it got to a point where I thought that whole bunch of disks was now suspect, and I disabled all three floppies before they did any more damage and made sure we had USB hooked up right for sneakernetting thumbdrives (then later got a better router). Anyway, just saying beware of that situation developing among your in use floppies and sanity check them once in a while.

I had a 2x speed 3.5" floppy drive that had one of the heads come off the mount. Wasn't worth my time to try to fix so I tossed and bought a new one.

5.25" floppy is a different story though.

Good job on the repair!

Arkanix wrote on 2021-07-22, 07:01:

The original MASM 5.1 Editor disk was the issue, but it wasn't a virus. The surface of the disk had failed and was depositing itself on the drive head. This caused excess friction which caused one side of the head to snap off its mount.

Hmm. I wonder if this will become a more common issue as our old disks age further.

The original IBM PC is 40 years old this year. I have some disks from that year. They still work fine -- for now! I wonder how long they can theoretically last.

keenmaster486 wrote on 2021-07-22, 15:17:

Arkanix wrote on 2021-07-22, 07:01:

The original MASM 5.1 Editor disk was the issue, but it wasn't a virus. The surface of the disk had failed and was depositing itself on the drive head. This caused excess friction which caused one side of the head to snap off its mount.

Hmm. I wonder if this will become a more common issue as our old disks age further.

The original IBM PC is 40 years old this year. I have some disks from that year. They still work fine -- for now! I wonder how long they can theoretically last.

The US air force only recently stopped using 8-inch floppies . https://www.popularmechanics.com/military/a29 … e-floppy-disks/ These were possibly close to a decade older than the oldest IBM PC disks ( until when here 8-inch floppies manufactured ? ) but lower density .

darry wrote on 2021-07-22, 16:51:

keenmaster486 wrote on 2021-07-22, 15:17:

Arkanix wrote on 2021-07-22, 07:01:

The original MASM 5.1 Editor disk was the issue, but it wasn't a virus. The surface of the disk had failed and was depositing itself on the drive head. This caused excess friction which caused one side of the head to snap off its mount.

Hmm. I wonder if this will become a more common issue as our old disks age further.

The original IBM PC is 40 years old this year. I have some disks from that year. They still work fine -- for now! I wonder how long they can theoretically last.

The US air force only recently stopped using 8-inch floppies . https://www.popularmechanics.com/military/a29 … e-floppy-disks/ These were possibly close to a decade older than the oldest IBM PC disks ( until when here 8-inch floppies manufactured ? ) but lower density .

8-inch disks were also by-in-large manufactured and QC'd to a higher standard than mini-floppies, due to there use being 90% business and industrial as opposed to home computers

maxtherabbit wrote on 2021-07-22, 16:58:

darry wrote on 2021-07-22, 16:51:

keenmaster486 wrote on 2021-07-22, 15:17:

Hmm. I wonder if this will become a more common issue as our old disks age further.

The original IBM PC is 40 years old this year. I have some disks from that year. They still work fine -- for now! I wonder how long they can theoretically last.

The US air force only recently stopped using 8-inch floppies . https://www.popularmechanics.com/military/a29 … e-floppy-disks/ These were possibly close to a decade older than the oldest IBM PC disks ( until when here 8-inch floppies manufactured ? ) but lower density .

8-inch disks were also by-in-large manufactured and QC'd to a higher standard than mini-floppies, due to there use being 90% business and industrial as opposed to home computers

Good point, though at least some 5 1/4 floppy disk manufacturers heavily advertised to business users in magazines such as Byte and PC Magazine during the 80s . I would imagine that these disks were probably better made and QC'd than more affordable disks meant for the price-sensitive home computer market . Then again, maybe this was mostly marketing .

Deunan wrote on 2021-07-22, 13:02:

I never had to do a repair like this but I remember reading about this issue. I have some "suspect" second-hand floppies but they don't seem to be progressing into worse state, I do wonder if maybe some heads work better then others - due to shape maybe? Keeping them clean helps as well. Also there was a mention of unusual noise when floppy surface is peeling off and accumulating on the head - never witnessed that either. I have one noisy floppy but it looks fine, seems to be a problem of the envelope, not media.

I use a scope+IMD combo to align heads, first with old floppies that I won't miss much if they get damaged and when that works well I switch to original software - the assumption here is these floppies were written with near-perfect aligment so are good templates.

BTW I wanted to ask, did you install MASM to use it or just for the editor? I switched to NASM but the DOS version requires 386+ CPU (and it better be a fast one). Most often then not it's easier and faster to develop DOS apps in VM or using DosBox and only doing the low-level HW testing (and benchmarking) on actual DOS PC.

I installed MASM 5.1 because I found a copy of it locally and I find it fun developing little bits of software on the host computer. Similar to using TMP on Commodore 64 or ASMOne on my Amiga 500.

I ended up getting the M editor (from the Editor install disk in the MASM 5.1 package) installed on the XT using a disk image and some imaging software to make a new disk. It seems to not be great in MDA mode, and I've just ended up using the MSDOS 5 EDIT program so far instead. The "Hello World" I typed in using EDIT compiled and linked just fine using MASM so I might just keep using that.

As for the disk drive, it's an original full-height IBM branded Tandon drive and they aren't easy to come by in Australia. I don't really have any other 5.25" drives I can drop in so it was 100% worth it to me to repair the drive. I'm having to regularly tinker with it to keep it aligned. I've got one of the half-height YE Data drives coming in from overseas with the plan of using the blank half-height bay for a 720k drive if I can find one of the original blue-button 720k drives IBM optionally shipped with the XT and XT286, similar to the AT and PC Portable 3.5" drives but with a textured black bezel for the XT and XT286. If this happens i'll take out the Tandon drive and use it for spares if I ever manage to get a 5150 for a decent price.

I'm not pulling the drive apart any time soon, but attached is an image of the head after an attempt to solder it back in place. This isn't how it looked in the end, as I ended up adding a very small piece of stripped magnet wire to the joint to sure it up and had to spend a few hours over several attempts getting the head Azimuth in the right ball park.