Helium filled harddisk to 7 TB

..

Hmmm, I wonder if it's a bit over atmospheric pressure, or if they have a partial vacuum... seems like less gas would reduce the drag.... but maybe not, gas dynamics is weird.

What he said. It's especially weird at < 100 angstrom dimensions.

The dynamic viscosity of an ideal gas is pressure-independent--see e.g.

. It also isn't a very strong function of composition--helium and air are pretty close together, according to Sutherland. Dynamic viscosity is what you think of when comparing different grades of oil, for instance. It's the drag force per unit area for a given amount of shear.

The kinematic viscosity has a factor of 1/density in it, though, which is what you want when considering the motion of the fluid itself. Perhaps the main effect is to increase the Reynolds number and improve the operation of the air bearing. (I haven't thought about self-acting air bearings a whole lot, but I guess that a low kinematic viscosity might help.) Flying heights have gotten really small--when I lost track of HDD development around 2005, the head-disc clearance was only about

100 angstroms, so it must be less now.

There aren't too many air molecules in a space that small, so many of the usual assumptions about fluid behaviour start to break down--the mean free path of an air molecule (i.e. how far it goes between collisions) is about 100 angstroms at 1 Atm, so with helium, the bearing is getting into the collisionless regime, which is pretty interesting.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

It is also the case that helium offers the fastest transfer of heat. (which might or might not be relevant in this case) ISTR 5-6x better

Maintaining better internal thermal equilibrium might be the reason.

--
Regards,
Martin Brown

That might be part of it, but a 100-angstrom layer even of air is equivalent to (1E-8 m)*(400 W/m/K)/(0.026 W/m/K) ~= 150 microns of copper, not counting the shear contribution. The head would have to be dissipating a lot of heat for that to matter much.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

What could go wrong?

- Helium diffuses through just about anything. This would obviously require a sealed drive. I don't know if the silicon rubber seals are good enough. The drive could have excess helium under pressure to compensate for the loss, but that would create a finite lifetime for the drive based on the diffusion rate.

- With a sealed drive, changes in atmospheric pressure will cause the cover to bulge in and out. While the cantilevered motor assembly will not be directly affected, changes in pressure might change the head flying height enough to produce some errors.

- Helium embrittlement of stainless steel is not much of a concern, but might result in small dimensional changes.

- Helium causes the drive to float away and stick to the ceiling.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

I knew you would find the only negative with this! :)

(...)

Most engineers spend their careers designing, building, and supporting things. I spent about half my career doing that, and then switched to fixing other peoples designs, damage control, repair, and science fiction (i.e. reviewing business plans). I don't have much talent for original designs, but I'm really good at breaking things. There are people out there that pay well to know why something will NOT work.

I don't seriously expect to see a plague of lighter than air disk drives floating around the room. Yet, it's the first image that comes to mind when I read about a helium filled disk drive. Perhaps that should be the new Hitachi logo... a disk drive floating through the air supported by a helium gas bag. Never mind... People will think that their data will also float away.

I just thought of another problem. Helium is less dense than air and causes the vocal chords to resonate at a higher pitch. If my digital music library decides to play in a higher key, the helium must be the cause.

--
Jeff Liebermann     jeffl@cruzio.com
150 Felker St #D    http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann     AE6KS    831-336-2558

The heads fly aerodynamically over the platters. Partial vacuum with He would not give much aerodynamic lift. He has a lot less viscosity than air, so maybe they spin it at very high speed, while the He reduces aerodynamic heating. He will also carry away the heat better.

Jon

It's not just the head, it is all the gas whirling around at 7000+ RPM.

Jon

Other folks' 10 krpm drives don't need helium, though, so it seems likely that it's connected with the high capacity, which requires very low flying height.

Magnetostatics obeys Laplace's equation, so the resolution can't be better than the flying height, because the field goes off in all directions. Gravity gizmos have the same problem, for the same reason.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Only if you rub it on your hair first. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

-dr...

tp://mac6.ma.psu.edu/stirling/simulations/DHT/ViscosityTemperatureS...

Interesting... thanks

k

The article said it was the reduced drag that was the important part. (not that everything you read on the web is correct.)

Hmm doesn't the mean free path go as the square root of the mass? Oops, no!

y

I once had this bizzare'o' behavior in a Rb vacuum cell that must have had the smallest trace of Hydrogen in it. The cells were used to do saturated absoprtion spectroscopy, which is very sensitive to any trace amounts of gas. The 'reputable' supplier of the cells could determine that there was less than some small amount of gas in the cells by sticking them in some RF coils and looking for a discharge (Pres >10^-7 torr or something like that, I'd have to look up the exact number.) Anyway sticking that pressure into the 'classical' gas scattering theory I couldn't make the numbers work. My conclusion at the time was that I must be seeing this weird QM enhanced small forward angle scattering. (See for instance Ramsey's "Molecular Beams" chapter 2.) After a few weeks the trace gas was gone, gettered away by the Rb presumably.

George H.

-dr...

r y

ng

ack

g

Ahh now that I can get right. The velocity does go as the square root of the mass! So sqrt 28(N2)/4(He) only about 2.5?? Maybe there is the cross section factor that I don't know.

George H.

e.g.http://mac6.ma.psu.edu/stirling/simulations/DHT/ViscosityTemperatureS...

Yeah, free path is like molecular whack-a-mole.

Great story.

I've had worldwide fame stolen from me by similar misfortunes, one time by a fairly elementary math error that had me convinced I could cancel the group delay of a filter in a PLL (IOW, as I belatedly realized, build a time machine). Probably all for the best. ;)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

So it was you that invented the way-back machine.

Nice.

Don't you know, weather affects your data in "The Cloud"? ;-)

Tim

--
Deep Friar: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

Since helium diffuses a lot faster than air, atmospheric pressure and temperature cycles would, over time, pump helium out but not let air in. It's a gaseous diffusion diode.

--

John Larkin         Highland Technology, Inc

jlarkin at highlandtechnology dot com
http://www.highlandtechnology.com

Precision electronic instrumentation
Picosecond-resolution Digital Delay and Pulse generators
Custom laser drivers and controllers
Photonics and fiberoptic TTL data links
VME thermocouple, LVDT, synchro   acquisition and simulation

sk-dr...

..

g.

..

air

ity

h e

ting

track

t

ing

.

Yeah, duh N2 has two atoms so it's about twice as big.

George H.

. y

My embarrassing diffusion story involves CO2 laser gas. (stop me if you've heard this before.) In grad school I knew of this unused high pressure gauge and regulator. I decided the lab could save money by mixing up our own CO2 gas. (recipe was something like one part CO2, 3 parts N2, and 10 parts He...) The department had a helium liquefier, and added several 2000 psi bottles every week. So I could fill co2 gas bottles to 1000 psi at the cost of the helium. (the most expensive gas)

I mixed up a batch and it was a total failure. I tried another smaller batch with a baked out storage bottle. Same thing. The failed tank sat strapped to the wall for over a month, till late one night a grad student ran out of CO2 gas, eyed the tank on the wall, and gave it a try. Worked like a charm. It takes a long time for gas to diffuse at 1000 psi.

George H.