[sebhc] Z-37 Diag Disk

[Barry Watzman]

Re: " That would argue against a problem with the "data separator""

Not necessarily.

Typically, format first writes tracks and sectors, then at the very end
(after writing AND VERIFYING the tracks and sectors) writes an empty
directory.

What could be happening is that it's writing the tracks and sectors, but
when verification fails, the empty directory and operating system structures
(including the boot sector and it's disk parameter tables, for Heath
systems) never gets written.

-----Original Message-----
From: sebhc-bounces at sebhc.org [mailto:sebhc-bounces at sebhc.org] On Behalf Of
Dan Lanciani
Sent: Tuesday, March 22, 2005 10:42 PM
To: sebhc at sebhc.org
Subject: Re: [sebhc] Z-37 Diag Disk

|> |> What happens when you try to format?
|> |
|> |It goes through all the motions as if it worked; drive selects, motor
|> |runs, head load, head steps all the way thru the disk. When it finishes,
|> |it tries to verify and finds no data.
|> 
|> Does the disk so created read ok on another system?
|
|No, it does not.

That would argue against a problem with the "data separator" (really clock
recovery) circuit.  What about the master clock itself?  Were you able to
check or swap it?  I don't think I have either the unit or its schematics
anymore, but I still have the H8 versions.  Am I remembering correctly
that the H89 version was similar except that it did not need the HLT/NOP
hack for synchronization?

|However, it *is* writing something on the disk; it is
|no longer blank.

Analysis of the disk might prove instructive...

Did you try single as well as double density?

|So, I've concentrated more on getting it to read, since
|I can stick a disk in the drive, tell it to read, and hunt through it
|with a 'scope. Trouble is, it's some kind of timing abnormality, which I
|can't see with a simple scope.

Writing is actually simpler than reading (especially if you aren't doing
precomp) so you might want to concentrate on that for a while.  I'm having
trouble thinking of many problems (not already covered by swapping all the
chips) that would be common to the read and write paths on the disk
interface
side.  Almost makes me want to look at the bus side...

				Dan Lanciani
				ddl at danlan.*com
--
Delivered by the SEBHC Mailing List


--
Delivered by the SEBHC Mailing List