Date   : Mon, 29 Nov 2010 15:05:22 +0000
From   : jim@... (Jim Hearne)
Subject: Harddisc fakery using a microcontroller

  Changing the drive board on a IDE drive even with one from an 
identical drive doesn't work, there is too much calibration data for the 
drive held on the board.
We tried it quite a few times in the early IDE days with various faulty 
drives.

Sounds like you must have been hot swapping the power lead to kill the 
drive like that.
We did have a batch of 'Y' power cable spliters with some of the outputs 
wired up mirrored, cost us a few replacement drives when we sold some 
before we found out ourselves.
Luckily the drive manufactures didn't suss out the cause and replaced 
the drives that were under warranty.

Jim


On 29/11/2010 14:34, Rick Murray wrote:
> On 29/11/2010 12:48, Jim Hearne wrote:
>
>> The other problem is though that almost every MFM and RLL control card
>> disc format was incompatible with any other.
>> You had to have the same make and model of control card that the drive
>> was low level formated with, otherwise you couldn't read the drive.
> Not *so* different these days, if you consider the "controller card"
> sits directly under the drive in modern hardware. One moment of lack of
> concentration and a manufacturer who annoyingly puts the power socket a
> different way around to all your other kit, you could find you've
> delivered 12V into the logic before realising the plug won't fit.
>
> I only did this once, after switching IDEs for a disc that kept failing
> its format (so some consolation it was probably buggered anyway), but
> I've seen others toast drives this way a number of times.
>
> ...and if you can't find an *exact* same model of drive, you're stuck.
> The controller boards aren't interchangeable. Even then, I wonder if
> there isn't something held on flash to map peculiarities of the specific
> unit - never tried replacing a drive board, I don't have two identicals
> to try.
>
>
>
> Best wishes,
>
> Rick.
>