[mdlug] Hardware Issue: Hard drive power requirements

[Raymond McLaughlin]

Weird problem:
A guy wanted me to install a new hard drive (supplied by him)in his
computer. It was a 250 GB 7200 rpm ATA100 Seagate Barracuda drive. The
system worked fine before putting new drive in. As soon as new drive was
installed System would not power up. It would begin to power up for
about 2 seconds, then die. With a little trial and error I narrowed it
down to the new drive. This drive, all by itself, with nothing else
connected but a pin 14 jumper, was enough to shut down this PS.

I informed the guy that a new power supply was needed, he said he would
get one. While waiting I tried the drive in my portable USB enclosure.
It worked fine for about 2 minutes, then stopped. The power supply in
the enclosure never worked again. The drive reads and writes fine when
power is supplied power from a stand alone PS.

The guy brought the new PS, (same cheap brand as the first one) and
still no joy. This time all the drives spun up, but the motherboard (and
it turns out CPU) were dead, apparently due to under voltage during
earlier power up attempts.

I suspected that the drive was faulty, a short circuit or something, so
I hooked it up to an amp meter and tested it. The drive label lists:
   +5V 0.46A
  +12V 0.56A

So when I saw it spiking to 2.2A during power up, I was sure it was a
bad one. I checked the warranty status of the drive on the Seagate web
site, it is good. I then looked at the power specifications for the
drive expecting to make reference to it in an RMA request. I was shocked
to see in the drive's manual shows a peak current of up to 2.7A in the
first 10 sec after power up. (see pdf page 21 = print page 13)
<http://www.seagate.com/support/disc/manuals/ata/cuda_72008_pm.pdf>
So this huge draw that I was seeing, 2.2A, was just 80% of what the
manufacturer think is allowable for a hard drive! I tested several
other, older hard drives, and found none that pulled more than 1.5A,
even momentarily!

So it looks like I got burned by two cheap power supply units in a row,
first in the computer, then in the drive enclosure (it was a cheap one).
I'm posting this partly to caution any would be upgraders, and partly to
ask if anyone has any insight on the subject. I can't find any
documentation stating what is considered a permissible current draw on
an ATA hard disk. My guess is that it would be an IEEE standard or some
such, but I haven't been able to find any references.

Comments?

Raymond McLaughlin