[ale] OT: Erasing a toasted drive

[James P. Kinney III]

On Fri, 2005-10-28 at 11:00 -0400, Alan Dobkin wrote:
> On 10/28/2005 10:49 AM, James P. Kinney III wrote:
> > Get a bulk tape eraser that plugs into the AC wall socket. That will
> > ruin the hard drive platters and the heads right through the case. Give
> > it about 30 seconds of erasing.
> 
> Greg's earlier post contradicts this advice:

When I was doing my grad school work, we had numerous problems with high
magnetic fields and hard drives. Granted, we were working with some big
stuff (1-4 T) but we still had issues even 20 feet away where the field
strength was down to about that of a bulk tape erasure (.01T). We would
regularly have hard drive failures and damaged sectors even though the
drives were in some really heavy steel cases. Monitors fared even worse.

The NMR lab down the hall had 5T magnets and warning tape on the floor
for the danger zone for iron objects.

Eventually, we added some MU metal shielding to the pc station and the
hard drive failures abated somewhat. As I was finishing up, the lab was
looking at converting to netboot dumb terminals. :)

The casing on hard drives is pot metal and the covers are stainless.
Neither of those provide adequate shielding against magnetic flux
variations. 

Besides, all you are trying to do in that case is make the sector
boundaries blur into oblivion making it impossible for all but NSA-types
to put the platter(s) in a recovery system and successfully perform a
data extraction.

As a bulk tape erasure generates about 5kG at the head (which attenuates
in air to about 2kG at 10cm, the effective field through the cover and
across the 1.5cm distance to the platter should be on the order of at
least 1kG. The drive heads generate less that 150G so this should put
the erasing field about 8-9 times stronger. The AC character of the bulk
erasures also generate transient currents through the drive circuitry
that will have the potential to do some damage there as well.

The feds run their drives through shredders because A. It is REALLY
destroyed then, B. It's way cool, C. They were sold on the process by a
contractor. Realistically, AFM (atomic force microscopy) can distinguish
individual magnetic regions on a substrate. Big deal. After 3 passes
with random data followed by a couple of passes with the same value, it
won't be logically possible to associate a particular magnetic region
with a particular "layer" of data on the substrate. So what if you are
able to recover 011001000011011101111001001001001111011001011001001
and the only real point you KNOW is         ^
it's still useless.

You can always hang the drive by the nearest 100kW radio station antenna
and let the signal from that slowly heat the drive to death. You can
also pull off the electronics and bake the thing at 400F. You can inject
some corrosive salt solutions or an organic acid into the air vent
opening.

My favorite is to blast it with radiation from a nuclear reactor. About
20us from a 200MW nuclear reactor near the surface of the rods will do
it as well. I don't want the drive back after that, though :)
-- 
James P. Kinney III          \Changing the mobile computing world/
CEO & Director of Engineering \          one Linux user         /
Local Net Solutions,LLC        \           at a time.          /
770-493-8244                    \.___________________________./
http://www.localnetsolutions.com

GPG ID: 829C6CA7 James P. Kinney III (M.S. Physics)
<jkinney at localnetsolutions.com>
Fingerprint = 3C9E 6366 54FC A3FE BA4D 0659 6190 ADC3 829C 6CA7
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part