h a l f b a k e r yNot so much a thought experiment as a single neuron misfire.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Anyone who has played with a gyroscope knows that once you get the thing spinning, it resists any effort to change the orientation of its rotational axis. You can force such a change, of course, but this always causes some extra wear-and-tear of the gyroscope bearings.
Well, now, think about the
hard disk drive in the average personal computer. When the thing spins at 7500 or 10,000 or even 15,000 RPM, it most certainly will behave much like a gyroscope. And as a result, you should not change the orientation of the computer case while the drive (or drives) spin. That is, you shouldn't do that unless you want to shorten the lifespan of the spindle bearings of the drive(s). I'm fairly sure that this type of wear is the most common mechanical cause of eventual drive failure.
Finally, now think about the fact that many people leave their computers on all the time, and especially for businesses, the hard drives of those computers may be active almost all the time (Web servers, database "farms", etc). Well, WHILE those disk drives are running almost all the time, they are physically located at the surface of the Earth, which is rotating.
So, the axial orientations of almost all those hard disk drives are slowly being forced to re-orient, inexorably, every hour of every day. WITH associated extra wear on the spindle bearings. The only way to prevent this is to align the hard-drive axes with the North-South axis of Planet Earth.
Therefore, what we need is a special "hard drive cage" that is "gimballed", so that after you are satisfied with the orientation of the computer case in your room (or in a rack of a server farm), you can orient the hard drive cage, inside the case, so that the axes of the hard drives are parallel with the planetary axis. Businesses, of course, won't worry about how unsightly it might be, for all the racks of computers to be oriented at a strange-looking angle in the server room (and such would cost less than buying a bunch of gimballed drive mounts). My suggested inside-case solution is for the average person who worries about external appearances.
Doing this orientation will require knowing your planetary latitude, and knowing which direction is true North or true South (because the magnetic poles are not exactly at the planet's axial poles). But after you have done the alignment, you can rest easier; your hard disk drives should now have a longer useful life.
Egocentric_20Clock
[FlyingToaster, Feb 04 2011]
Axial Precession
http://en.wikipedia...ki/Axial_precession
[bs0u0155, Nov 05 2014]
Flywheel Basics
http://home.earthli...fradella/basics.htm
[xaviergisz, Nov 06 2014]
[link]
|
| |
What is the lateral force required to rotate the axis of a hard
drive over 24 hours in this way? |
|
| |
All that really matters is that if the axial reorientation force is greater than zero, it will be detrimental to the lifespan of the axial bearings. |
|
| |
... or you could make use of it <link> |
|
| |
The key question is what this does to the survivability of the electronic connections to the drive as opposed to the bearings. I suspect the decreased life of the former would more than offset any gain in the latter. |
|
| |
This would not need a gimballed drive: the entire computer casing could be mounted at an angle. If I had a desktop / tower machine instead of a collection of various laptops, I would be trying this out right now using blocks of wood to prop the thing up! |
|
| |
//asymmetrical gravity load// The bearings "fight" the weight of the platters no matter what. |
|
| |
The obvious solution here are air bearings. |
|
| |
Does this introduce another problem when gravity is
considered? Ignoring the gyroscopic effect, the load
on the bearing is still downward.
[+] for Foucault's gyroscope |
|
| |
You could buy the drives at pre-set angles, and have the housings
rotatable at 90 degree angles. I mean, if you are at 39 north, and your
server stack faces east, you buy the appropriate set and screw it in
right. I don't see needing a full gimbal housing for each drive. |
|
| |
Even if this weren't really a solution to a problem, it would be a
marketing gimmick. [+] |
|
| |
No-one responded to [lurch]'s gravity point. |
|
| |
Perhaps the only solution is to put all the server farms at the poles. |
|
| |
//Perhaps the only solution is to put all the server farms
at the poles.// |
|
| |
That only mitigates the vast majority of the problem. The
Earths axis isn't totally stable, it is also subject to axial
precession. <link> |
|
| |
Obviously we need to get the hard drives off-world. |
|
| |
Wait, what's the wear on the bearings due to the rotation
around the galactic centre? |
|
| |
From the 'Flywheel Basics' link: |
|
| |
//Precession torque (tending to tilt the flywheel spin axis), for a stationary flywheel battery application (subjected to earth rotation of one revolution per day), does not present serious problems, but is not negligible. It can be computed from: |
|
| |
Precession Torque = (Moment of Inertia) (Spin Rate) (Precession Rate) |
|
| |
For a vertical spin axis, at the earth's equator, the precession rate is due to earth rotation, and is 0.00069 rpm// |
|
| |
OK, who wants to do the calculations for a HDD? |
|
| |
the RPM for the galaxy is a rather zippy 8.5X10-15 rpm.
Can't find that one on my record player. |
|
| |
I've calculated 17E-7 Nm of precession torque for a HDD (r=0.04m, m=0.03kg, spin=10,000rpm). |
|
| |
Can someone figure out the asymmetrical gravity load? |
|
| |
//hard disk drives...they are physically located at the surface of the Earth, which is rotating. |
|
| |
Can't ever say I've ever noticed a problem with all the bits sloshing over to one side of the disk, but now you mention it.. |
|
| |
kid with dads pw enters server room. next scene... |
|
| |
(cue chanting) " The Ringworld is unstable, the Ringworld is unstable..." |
|
| |