From: Henry Spencer <henry@zoo.toronto.edu>
Newsgroups: sci.space.tech
Subject: Re: hard disk operation in vacuum
Date: Sun, 10 May 1998 04:23:52 GMT

In article <6j1mig$kqv$1@hermes.iol.it>,
giorgio perrotta <g.perrotta@iol.it> wrote:
>I'm considering operating a commercial hard disk in vacuum. Is anybody
>there knowing of previous experience  or of any  technical impediment
>(lubrication or like) that may cause difficulties or failures? for a
>limited time operation (say 6 -12 months ?)

All current hard disks fundamentally rely on having air inside the case,
because the heads fly a very small distance above the disk surface using
aerodynamic forces.

Since the case is not designed for large pressure differences, there is
typically a small pressure-equalization port with a fancy filter over it,
to keep the inside at the same pressure as the outside.

Much of the detailed design of the disk drive and its electronics also
assumes air cooling.  There doesn't have to be a lot of air circulation --
PCs are hopelessly bad thermal designs, with very little air flow in many
areas -- but there has to be some.

The lubricants likewise will not have been chosen to permit vacuum
operation, but that's a relatively minor problem by comparison.

If you read a hard-disk spec sheet carefully, you will find that the
specific limits of operating conditions include a maximum altitude, i.e. a
minimum air density.  It's typically something like 10,000ft.

There is no particular problem with using a commercial hard disk in orbit
(it's been done), but it absolutely must be kept pressurized.
--
Being the last man on the Moon                  |     Henry Spencer
is a very dubious honor. -- Gene Cernan         | henry@zoo.toronto.edu

From: Henry Spencer <henry@zoo.toronto.edu>
Newsgroups: sci.space.tech
Subject: Re: hard disk operation in vacuum
Date: Tue, 12 May 1998 03:19:36 GMT

In article <6j7k6f$ecr$1@hermes.iol.it>,
giorgio perrotta <g.perrotta@iol.it> wrote:
>Well, well... many many thanks for the very clear illlustration of the
>problems. I made up my mind  and the idea of putting the disk in an
>airtight (or slightly overpressurized) container sounds feasible
>enough.

Yes, this sort of thing is quite feasible.  Components in pressurized
housings are far from unknown, even in US spacecraft (despite their
reputation, unlike the Soviet/Russian ones, for running things in vacuum).
For example, the scan-platform actuator motors on the Voyagers had
pressurized housings, for the sake of the lubricants in their gearboxes.
(I ran into an old NASA paper discussing the details of the Voyager 2
actuator problem.)
--
Being the last man on the Moon                  |     Henry Spencer
is a very dubious honor. -- Gene Cernan         | henry@zoo.toronto.edu

From: jtkare@ibm.net (Jordin Kare)
Newsgroups: sci.space.tech
Subject: Re: hard disk operation in vacuum
Date: Thu, 14 May 1998 22:14:35 -0700

In article <EstqKp.8pp%spenford@zoo.toronto.edu>, Henry Spencer
<henry@zoo.toronto.edu> wrote:

> >the idea of putting the disk in an
> >airtight (or slightly overpressurized) container sounds feasible
> >enough.
>
> Yes, this sort of thing is quite feasible.  Components in pressurized
> housings are far from unknown, even in US spacecraft (despite their
> reputation, unlike the Soviet/Russian ones, for running things in vacuum).
> For example, the scan-platform actuator motors on the Voyagers had
> pressurized housings, for the sake of the lubricants in their gearboxes.
> (I ran into an old NASA paper discussing the details of the Voyager 2
> actuator problem.)

Spectrum Astro offers as a commercial product space-qualified disk storage
units based on commercial disk drives in pressurized boxes.  They were
tested on the MSTI missions, as I recall.  Capacities are something like 1
and 4 Gbytes.

Note that rotating disks used in space (as opposed to just in vacuum) must
typically be mounted in pairs, with opposed spins.  Otherwise they act as
gyros and are a real pain for the attitude control system.

I ran comparisons of solid state and disk storage for some remote sensing
satellites while I was still at LLNL.  At that time, RAM prices were at a
peak, and disks were a promising option, but the rapid fall in RAM prices
means that the breakeven point where cost of RAM exceeds the risks and
hassles of using disks has shifted up considerably.

For what it's worth, the breakeven point for that project was around half
a *terabit* (5 x 10^11 bits; about 60 gigabytes) of onboard storage on
each of several satellites, and we were trading solid state recorders
based on 64 Mbit RAM chips against 8 Gbyte 3.5" disk drives.


Jordin Kare
Kare Technical Consulting

Newsgroups: sci.space.history,sci.space.shuttle
From: henry@spsystems.net (Henry Spencer)
Subject: Re: computers (was Re: Could the shuttle be converted?)
Date: Fri, 3 Sep 1999 04:12:02 GMT

In article <MPG.12379207c8b79dc798968f@news.tiac.net>,
Alphageek <paulditt@NOSPAM.tiac.net> wrote:
>...Most modern hard
>drives are filled with inert nitrogen gas to a _slight_ overpressure and
>sealed.  That's why removing the seal, even if you don't open the case,
>is enough to void the warranty.

Unless this has changed since the last time I looked -- it's not a
technology I've kept up with the details on -- they are not literally
sealed, because the designers don't want significant pressure differences
developing across the case.  (Bear in mind that even at sea level, the
atmospheric pressure changes with weather... and most hard drives are
rated for use up to circa 10,000ft, where the atmospheric pressure is a
good deal less than at sea level.)

They are sealed *except* for a very small and very good filter whose sole
purpose is pressure equalization.  The point of the restrictions on
breaking the seal is not that they are literally airtight, but that any
unfiltered opening yields the possibility of contaminating the extremely
clean interior air with dust.
--
The good old days                   |  Henry Spencer   henry@spsystems.net
weren't.                            |      (aka henry@zoo.toronto.edu)