7 Nov
2001
7 Nov
'01
8:49 a.m.
----- Original Message -----
From: "Eric J. Korpela" <korpela(a)ssl.berkeley.edu>
To: <classiccmp(a)classiccmp.org>
Sent: Wednesday, November 07, 2001 11:07 AM
Subject: Re: Apple Floppy Drives (was: More Apple Pimpers)
The mechanism taken by itself may have been reliable enough, BUT, since there
was no track-zero sensor, (I think that's the reason) the "recal" operation
rams
the head assembly into the outside stops multiple times each time it is
performed, and that's going to harm the mechanism. Do that enough times and the
system loses alignment, which makes it prone to failure. As the drive changes
in radial alignment, the data written with it becomes "off-track" so it will be
difficult to read when the drive is realigned or when the diskette is put in a
properly aligned drive. The consequences of poor alignment is not an Apple
problem, though the Apple way of using the drives causes misalignment more
quickly than with drives that sense when track zero has been reached.
Keep in mind that Apple left out the track zero sensor and index sensor, not so
much because the sensor was costly or difficult to manage, but because it
required that each drive have them manually adjusted by a human
operator/technician. This was costly and, while most drives were done in
exactly that way, the Apple process skipped those steps, saving tons of money.
>> the Apple floppy disk subsystem
wasn't as solid as one might have
>> hoped.
>
> ??? whaddya mean not solid? The ONLY problem I've seen is issues related to
> when one drive is out of alignment and as a result, may or may not be able
to
read disks
from another drive. The disk ][ was simple, clever and RELIABLE.
Please quantify your statement!
I have to agree with Dick. Simple and clever, yes, but Disk ][ was not all
that bullet proof. The mechanism was reliable. I rarely recall failure of
the mechanism. But I certainly recall disks being rendered unreadable by
leaving a disk in the drive when powering down. Lifetimes of frequently used
disks were on order of months, even if the disks were write protected.
Later Disk ][ drives didn't seem to have as many problems. I assume there
was some modification to the circuitry to prevent stray currents in the heads
and to clamp surges on power down. The problems resurfaced in some //c
internal
drives. I don't recall having any problems with
the Apple 5.25 (plastic case,
D connector) drives.
The matter with the drive turning on its heads during the power on/power off
transient is something that one would expect with any drive, hence the drive
should NEVER be closed with a diskette in it during the power-up or power-down
process. There's always a risk that the write-gate will be perceived by the
drive as being "ON" as the supply becomes unstabile, and that can cause
problems. That's not an Apple-specific problem but exists wherever there are
disk/tape drives. I'd classify diskette damage caused by this bugaboo as
procedural error.
I don't recall having as many problems with drives on other systems with the
exception of a Kaypro that would wipe most any disk over the course of a
couple hours. I only had access to the one Kaypro, so I don't know if the
problem was widespread.
Eric