• Please review our updated Terms and Rules here

3.5" Floppy trouble

The Underlord

Member
Joined
Nov 8, 2009
Messages
14
Location
Omaha, NE
I'm having a little trouble with a Mitsubishi MF355C-88UF 3.5" floppy drive. I consider it quite an interesting design, given that I hadn't done much digging inside of computers until fairly recently.

Anyway, the trouble I'm having is that the read/write head doesn't want to move when it's time to read or write something from a floppy. The head does move during what I would consider a start-up test when the machine is turned on.

At first I simply thought the stepper motor was bad, but after shipping the drive halfway across the country to have it fixed, it worked fine. And it worked when tested upon its return, but after some time of sitting it again refused to operate when I wanted to use a floppy.

The question is: would the cause of this be a failure in the drive somewhere, or could it be a failure of the controller?

I'll admit that I don't know a whole lot about the controller, other than it is on a 16-Bit ISA expansion slot and controls both the floppy drives and the hard drive.


I suppose I could produce a picture or two if anyone thinks it might help.
(I know someone out there just wants to look for the sake of looking ;) )
 
If it works on another system, I would tend to expect the problem to be with either the controller or the interface cable. When the drive works right after installing but fails a little later, I tend to find that means the cable is working loose. Try a different cable if you can.
 
Thanks for the reply krebizfan, but I did manage to get it to work again, after poking at it a little bit. After taking the opportunity to copy the old files that were the primary reason that I wanted to get that drive working, only having one failure during the process, I wonder if there is a problem with the read/write heads themselves. Either in possibly the distance from the the heads to the disk surface or some other problem. I'll have to experiment with it a little bit more.

(hmm, apparently the file format for MS Works v1.05 is a bit of a mystery today. I guess I'll be busy converting files and saving them again. I'll make an update if I figure anything else out with the drive)
 
Last edited:
The stepper motor seems to be operating fine, so my only guess would be a problem with the heads themselves.

Since I didn't think my first post completely through, I'll mention now that a 5.25" FDD attached to the same controller (with the same cable) works just fine.
 
For those of you that have been following along with/paying a smidge of attention to my half-baked attempt at floppy drive trouble shooting, and even if you weren't, I'll drop an update in my escapades. Using a Shareware version of FormatMaster software I've determined that the drive itself seems to be working for the most part. In the attempt to format a disk it worked through most of the process and turned up with the error "Address mark not found" and during a verification, (in which it attempted to read it as a 720K rather than a 1.44MB) it gave the error "Unable to read boot sector"

Preliminary research seems to yield the "toss it and get a new one they're cheap as dirt" answer to the problem, but since this is Vintage Computing, I would imagine that a few of you share the sentiment of not throwing away a component unless it's compltetely destroyed. (Specially if the component would be darn hard to replace.)

I'll try digging a little more to try and figure out the cause of the error, and maybe find a solution, but if anyone has an idea I'm always happy to listen. (I tend to lurk too much :roll:)
 
For those of you that have been following along with/paying a smidge of attention to my half-baked attempt at floppy drive trouble shooting, and even if you weren't, I'll drop an update in my escapades. Using a Shareware version of FormatMaster software I've determined that the drive itself seems to be working for the most part. In the attempt to format a disk it worked through most of the process and turned up with the error "Address mark not found" and during a verification, (in which it attempted to read it as a 720K rather than a 1.44MB) it gave the error "Unable to read boot sector"

Preliminary research seems to yield the "toss it and get a new one they're cheap as dirt" answer to the problem, but since this is Vintage Computing, I would imagine that a few of you share the sentiment of not throwing away a component unless it's compltetely destroyed. (Specially if the component would be darn hard to replace.)

I'll try digging a little more to try and figure out the cause of the error, and maybe find a solution, but if anyone has an idea I'm always happy to listen. (I tend to lurk too much :roll:)

My guess would be the floppy cable in the system, especially if the drive works elsewhere and on this system it seems to identify as a lower density format...

Those IDC cables/connectors can become damaged, and only affect one drive...

It's an easy swap at least...
 
My guess would be the floppy cable in the system, especially if the drive works elsewhere and on this system it seems to identify as a lower density format...

Those IDC cables/connectors can become damaged, and only affect one drive...

It's an easy swap at least...

Hey thanks for the idea! :D

Gave it a shot with the only cable I have around that made a suitable replacement (5.25" connectors) And there was no difference in performance. I even gave running the drive without its built-in 5.25" to 3.5" style plug converter, and there was still no change.

I'm still a bit mystified as to why it would be unable to read the boot sector. But anyway, thanks for your suggestion.

(arg, apparently it's been a long day, managed to close the window with my initial response before posting.)
 
I'm still a bit mystified as to why it would be unable to read the boot sector. But anyway, thanks for your suggestion.

The boot sector, since it has the volume ID is always the first sector read when a new disk is inserted. So, chances are, you can't read any other sector either.

Can you?
 
The boot sector, since it has the volume ID is always the first sector read when a new disk is inserted. So, chances are, you can't read any other sector either.

Can you?
Indeed it does, under normal operation the drive simply returns a general failure error. But when I use the FormatMaster (v3.36) program it saves the boot sector for last during the process of formatting. After it attempts to write the boot sector, it delivers the Address mark not found error (or, not being 100% certain how the program works, at least when it tries to finish). The disk is then essentially no longer formatted, even if it had been before, as it will not read on a different machine where it had before.

Long story short, the drive is capable of writing to the disk, at least enough to change any information that was there into something of an unformatted nature.
 
I've never used FormMaster, so I don't know if it verifies on the fly or makes a separate pass. Do you know?

Regardless, play around with the drive and a formatted diskette using AnaDisk--you can selectively read, write or even format any track on the disk. A few minutes of playing around should tell you what's going on.

If FormMaster, for example, doesn't verify until after it's formatted the whole disk, then the WRITE DATA signal path might be goofed up. So when WRITE GATE is asserted, you get what amounts to a DC erase of the track and everything is wiped out.
 
FormatMaster does its verification in a separate pass, verification is selectable as an option. It just didn't manage to make it that far.

I did manage to track down a version of AnaDisk, transferring it to the machine in question was a whole other adventure.

AnaDisk works fine on this computer, but it identified the disk as blank or unreadable. (disk isn't actually blank) I couldn't find anything at individual tracks either.

Makes me want to think that the heads are misaligned, but then that wouldn't explain why I can't format anything. (I suppose unless it's a really big problem) Still that wouldn't explain why it would actually read on rare occasion.

So I'm at a loss for the moment.
 
AnaDisk gripes if the FDC sees no information at all. As little as one good address mark will produce a finding of something.

Your drive could be misaligned or there could be something else going on, such as a misplaced jumper. It might even be that the head isn't writing anything at all as well as not reading.

I've had that happen where the mechanism gets bollixed up and the head doesn't even contact the media because of some gorilla forcing things.
 
Thanks for your input. The only thing that I could imagine could've caused the heads to move out of alignment in the first place would be having moved the computer at one point.

As far as the jumpers go, I do not believe that they have been changed from the last known working configuration.

Also, as far as I can tell the heads are still in contact with the media, so I suppose I'll have to track down some way of checking the alignment.
 
Back
Top