Strange drive issue

[lucadip]

I searched for a similar issue in the forum but couldn't find anything like this.

Somehow my drive seems able to read and write properly but does not recognize any valid DOS-disk as such (DF0:NDOS).
The same disk in DF1 (or in another Amiga) is read perfectly. I can even format in DF0 and the disk is not recognized, if I put it in DF1 is fine.
If I extract an ADF image from DF0, the image is perfectly valid. If I write it back to DF0, same problem (and in DF1 is fine).

Has anybody already experienced that? Any help is very welcome!

[Daedalus]

So it seems to be able to write properly but cannot read. Strange... Have you checked that the cable's seated correctly and is undamaged? Does it do the same thing when DF1: isn't connected? Can you swap the DF0: and DF1: drives to see if the problem moves with the drive or stays with DF0:?

[lucadip]

Originally Posted by Daedalus

So it seems to be able to write properly but cannot read. Strange... Have you checked that the cable's seated correctly and is undamaged? Does it do the same thing when DF1: isn't connected? Can you swap the DF0: and DF1: drives to see if the problem moves with the drive or stays with DF0:?

Actually it can even read (well, raw read)!
Tested it when extracting an ADF-image of the inserted disk. I should add that my os (3.1) has no strange patches and that the problem also happens if I boot without startup-sequence.

DF1 is external, so I cannot swap the drives.

[thomas]

ADF read is not more raw than other reads. Which program did you use to create the ADF?

Have you checked for viruses? Can you try to boot from floppy (with the write protect switch on and after power-off of the Amiga)?

[lucadip]

Originally Posted by thomas

ADF read is not more raw than other reads. Which program did you use to create the ADF?

Have you checked for viruses? Can you try to boot from floppy (with the write protect switch on and after power-off of the Amiga)?

I've used TransADF to extract the disk image.
Did not try to boot from floppy though, I'll give it a try this afternoon. Thanks for the tip!

P.S.
Post #100 :-)

[Daedalus]

Ah, when you said extract, I thought you meant extract the ADF *to* the floppy. Very strange.

External floppies are just internal floppies with a little bit of interface logic. They're easy to open and swap drives with if you're comfortable with that sort of thing.

[lucadip]

Originally Posted by thomas

ADF read is not more raw than other reads. Which program did you use to create the ADF?

Have you checked for viruses? Can you try to boot from floppy (with the write protect switch on and after power-off of the Amiga)?

Checked. The amiga has no virus.
I can't boot the freshly started amiga from any DOS-Disk.
Using Image2Disk or TransADF I can perfectly read and write the whole disk. Using X-Copy-Pro I can even read the directory of the disk, checkdisk shows no error.

I start to believe that there is a problem with the alignment of the heads. After the disk is inserted the heads start moving back and forward and can't seem to find the right sector. Why on earth there is no problem when reading the tracks "directly" (for example with TransADF or X-Copy) drives me crazy....

[ReadOnlyCat]

Originally Posted by lucadip

Checked. The amiga has no virus.
I can't boot the freshly started amiga from any DOS-Disk.
Using Image2Disk or TransADF I can perfectly read and write the whole disk. Using X-Copy-Pro I can even read the directory of the disk, checkdisk shows no error.

I start to believe that there is a problem with the alignment of the heads. After the disk is inserted the heads start moving back and forward and can't seem to find the right sector. Why on earth there is no problem when reading the tracks "directly" (for example with TransADF or X-Copy) drives me crazy....

This indeed looks like an alignment problem, TransADF and X-Copy probably position the heads in a sequence which allows proper alignment by reaching some physical stop of some kind while the regular OS stepping does not follow it thus leaving the heads improperly aligned.

Are there programs which allow to freely step the disks's head in both directions? If so you could 1) disable OS head stepping when no disk is present, 2) manually step the head to full 80 (disk center), 3) insert a DOS disk and see if it is recognized. Then retry with track 0 (disk outer track)

If this works, then it's definitely an alignment issue. If it does not work then we haven't learned anything.

[Daedalus]

Yes, good point. I had ruled out an alignment issue as the disks it was writing were readable on another drive, but if the heads only move in one direction (as is the case with the type of reads and writes being carried out), any play or hysteresis in the head position is irrelevant, whereas the OS will move the heads forwards and backwards, meaning any looseness will cause tracks to not be read.

[hooverphonique]

That's not how it works.. the 'physical' stop is always used to reference cylinder zero, whether it's dos or x-copy. Now, you could have a borderline alignment problem that will show itself when switching the stepping direction, or a mechanical problem that causes misalignment in the same situation. Under those circumstances, ROC is right in the sense that reading the whole disk from cylinder zero to 79 might work, while random access won't.

[Daedalus]

Yep, well aware of how it works, and that's pretty much what I said too.

[ReadOnlyCat]

Originally Posted by hooverphonique

ROC is right in the sense that reading the whole disk from cylinder zero to 80 might work, while random access won't.

Kittens are always right!

\^^/

[lucadip]

First, thanks for all the info.

I went on and wrote a small piece of code which just reads track 0, track 40 and again track 0 and track 40. Nothing else.
Guess what? After the code ran a few times (properly reading the tracks!), the drive is now happily recognizing all the disks again.
I don't believe that I coded a kind of magic program that can fix misaligned drives. Probably the truth is that the drive wasn't used for such a long time, that after being stressed for a while it found back its alignment?!

BTW, I know that this kind of documentation is available electronically and is even online, but having the "RKRM Devices" opened next to the keyboard and close to the DevPac manual gave me a nice, warm feeling! :-)

[lucadip]

Originally Posted by lucadip

I went on and wrote a small piece of code which just reads track 0, track 40 and again track 0 and track 40. Nothing else.
Guess what? After the code ran a few times (properly reading the tracks!), the drive is now happily recognizing all the disks again.
I don't believe that I coded a kind of magic program that can fix misaligned drives. Probably the truth is that the drive wasn't used for such a long time, that after being stressed for a while it found back its alignment?!

Sorry for reviving this old thread of mine, but the problem with my drive arose again, and another drive had the same issue, which I could mitigate likewise.

I thought for a while about a valid explanation, and came to the idea that since both drives have never been opened for inspection, the grease on the shaft might have dried up, after more than 3 decades.

So I opened up the drives, cleaned up the dust and the shafts, which were very dry indeed, put some new grease (especially made for bearings) on it, and tested again.
The drives worked flawlessly afterwards! This information might sound trivial to many of you, but I wanted to share it anyway. Maybe somebody could find it useful in thirty years from now!

[rjd324]

It's seems too under estimated how important reviving an old thread with some sort of solution is. People don't do it enough. I am always in the habit of trying to conclude a thread/topic and if I own the thread I will edit it to say (solved).

Thanks for the update, although it is not relevant to me. People should try to follow this procedure, always.