Weird Floppy stuff

From: Spiro Trikaliotis (ml-cbmhackers_at_trikaliotis.net)
Date: 2005-05-16 19:22:48

Hello,

while working on some custom drive routines for the 1541/157x drives,
"we" (WoMo, Nicolas Welte and I) discovered a weird behaviour which
hopefully, we can discuss here.

This behaviour has to do with stopping writing to a track, and starting
reading from it.

Now, think about the following: You are writing a complete track to the
disk. After you are finished, you set the R/W head into read mode, and
start verifying the track by reading back each and every byte.

This program works flawlessly with a 1541 and 1541-II. It works
flawlessly with a 1570. But it does NOT work flawlessly with the 1571.
The drive reads back wrong bytes. Not all bytes are wrong, but there is
some difference now and then.

Interestingly, it seems to be highly dependable upon WHAT was written to
the drive before going into read mode!

For example: Writing a block with contents of $01 (that is, the same as
the 1541's block contents after formatting), this problem does not occur
in almost all cases. The very similar content of $00 (resulting in only
1 differing GCR bit per 10 GCR bits) causes this to occur some times for
some bytes (< 10 consecutive bytes after a SYNC). 
In contrast, writing $69 or even $00 causes this problem to occur *very*
frequently, with often over 256 byte "settle time" before the drive
reads correct bytes back. Sometimes, we must ignore two consecutive
sectors and start verifying with the 3rd sector to be able to
successfully verify!

Interestingly, the only tested 1571DCR does *not* show this behaviour,
as well as the 1570, which does not show this.

Changing some parts around, it seems it is the drive MECHANICS which is
the culprit. Even more interestingly, there are drives where using 
side 0 (lower R/W head) causes this behaviour, but using side 1 (upper
R/W head) does not cause this behaviour!

Here, we are stuck. Thus, I ask you:

1. Has anyone ever experienced something similar?

2. Has anyone an explanation for this? Especially the difference between
   different sides of the disk on the same drive seem to indicate that
   it is a problem with the electronics of the drive mechanics.

Regards,
   Spiro.

-- 
Spiro R. Trikaliotis
http://www.trikaliotis.net/
http://cbm4win.sf.net/

       Message was sent through the cbm-hackers mailing list

Archive generated by hypermail pre-2.1.8.