Wiping data from a hard drive.

[Colin Law]

On 25 August 2010 11:41, Jordon Bedwell <jordon at envygeeks.com> wrote:
>  On 8/25/2010 4:51 AM, Basil Chupin wrote:
>> A few days ago I 'lost an HD which is 11 months old - and it still has 4
>> years of warranty remaining.
>>
>> However, the HD contains personal information which means that I will
>> simply put the sledge hammer to it - unless I can wipe the HD clean
>> before I take it back for replacement.
>>
>> Since the HD doesn't spin-up, I cannot use the normal data destroying
>> programs to wipe the data.
>>
>> When we were still using 3 1/4 and 5 1/2 inch floppies I had a powerful
>> magnet thingie which would wipe a floppy clean (but I don't have that
>> 'thingie' anymore, in anycase it would probably be too weak for a 500GB HD).
>>
>> I was wondering if anyone has any ideas on how I can wipe the data off
>> the HD?
>>
>> Would taking it to a wrecker's yard and putting it under the magnet they
>> use to lift car bodies do the trick, for example?
>>
>> Thanks for any advice.
>>
>> BC
>>
> Your best bet is to simply pull apart the HD and destroy everything.
> Bending is also your best bet during destruction.  You can always raise
> it above it's curie point, but considering each HD manufacture is some
> what unique in it's materials it's not feasible to calculate the curie
> point unless you are destroying truly sensitive data.  You could degauss
> it, but using the wrong degauss could leave magnetic remnants which you
> obviously don't want.  Since I don't know how the magnets at a wreckers
> yard works and if you even have control over that magnet, I can only say
> it could leave some remnants, but it could just be strong enough to mess
> up the alignment especially if you have some control over the magnet.
> You could always play manly and just put it through a tree shredder.   I
> guess, by  theory, if you could find a company in your area that has an
> oven that can go above 1200 degrees you might be able to reach the curie
> point, depending on what metals it has; for example some cobalt metal
> mixtures would need above 1200 degrees bulk curie but some cobalt metal
> mixtures only need as little as 140 degrees.

I think those suggestions might invalidate the warranty.

Colin