Date   : Mon, 06 Mar 2006 16:32:17 +0100
From   : "W.Scholten" <whs@...>
Subject: Re: 3.5 / 5.25 floppy disk reliability

Jules Richardson wrote:

>> I started off using 5.25 disks, but had to clean the heads every week 
>> or so
>> with the brown gunk from the disks. Is this a common problem or do I just
>> have a poor batch of disks ?

> 
> Chances are you have a decaying batch of disks, or they're just 
> exceedingly dirty and leaving a lot of junk on the drive heads.

Brown gunk is usually the protective layer on top of the magnetic layer, 
that's gone bad. After that the magnetic layer itself will go. It 
happens with tapes too. As Jules, I would still recommend the 5.25 disks 
  over 3.5 if you can find a good batch.

Regarding decaying media, I've come across more EPROMs with unstable 
bits (bits that read as either 1 or 0 depending on what device I read it 
with). I've got 2 EPROM programmers (Morley and HCR), I usually use the 
Morley as it's faster/smaller but it doesn't handle some EPROMs and 
EPROM types (2532). IIRC Mitsubishi 27128/27256s are often problematic.
Anyway, there was an instance with a 2532 for an Atom programmed ca. 
1985 where I first saw this, and recently in BBC EPROMs. Reading them in 
in each programmer and  finally in a BBC micro gave 3 different results.
I suppose putting all the bits that are unstable as '0' will fix those, 
but that's no guarantee of some of the rest not being in much further 
state of unprogramming!

Anyway, I think this is going to happen a lot more, I read about the 
guaranteed duration for which EPROMs stay as programmed, I think it was 
in a datasheet, which mentioned 10 years... I'm definately seeing decays 
so everyone who has originals or copied EPROMs, read them in please so I 
can add them to the ROM archive.


-- 
Wouter
---
BBC micro | Calculators | Classic PC games: http://www.xs4all.nl/~swhs/whs/