Helium to be used in computer hard drives

[Tigerman]

If it is, its probably a good thing for divers though. If they all of a sudden have lots of helium plants and low demand, it might get cheaper

 

[nimoh]

If it is, its probably a good thing for divers though. If they all of a sudden have lots of helium plants and low demand, it might get cheaper

good point, although all of the ramped up helium production will be in China where the hard drives are made

 

[Tigerman]

Hmm.. lets rephraze that then - Itll be a big increase in asian tech divers

 

[Akimbo]

I think you lose the layer of gas that the heads float on if you use a vacuum.

Good point, forgot about Bernoulli. Air contamination of the Helium atmosphere around the platens and heads could really hose up the aerodynamics. Makes you wonder where hard disk cost curves will collide with solid state.

I wonder if anyone has ever investigated controlling head height magnetically in a vacuum? Interesting times.

 

[ianr33]

Even though the Helium molecule is about 2/3rds the size of a Nitrogen molecule, it is dramatically harder to contain than the size difference implies,

Helium is a monotomic gas though, and nitrogen goes around as N2 molecules. Nitrogen molecule is apparently around 3 times larger than a helium atom
Helium and Nitrogen Sizes

 

[Akimbo]

...Nitrogen molecule is apparently around 3 times larger than a helium atom
Helium and Nitrogen Sizes

My bad, correction made. I do need a copy editor. After re-reading my post, maybe this will help clarify the point I was attempting to make: Helium "seams" much more than ~3x more difficult to contain because it will slip by molecules in a lot of solids where larger gas molecules will not. For example, "air tight" systems at 3000 PSI will often leak like a sieve on Helium at 200 PSI.

Most forged and wrought metals are "usually" Helium-tight while very few plastics, glass, quartz, metal sand castings, or naturally occurring materials are. In practice, designs must accommodate “acceptable” Helium leak rates — the Helium relief valve in some diving watches being the most famous example.

 

[OkByMe]

Using inert gases is nothing new for disk drives. From about 1970.. Most of the larger platters in the 16"-20" diameter range had sealed enclosures and cycled nitrogen through the drive head assembly to keep it clean and cool.

 

[AfterDark]

I test hard drives and other electronics for data storage. The hot setup is going to be the solid state flash drives. The hard disk is going the way of the floppy. The price has been coming down and will continue that trend. We used to use just 5 SSD for the OP sys now we have "boxes" with only SSD drives from 15 to ?. How much storage do you need?

 

[miketsp]

..snip..
How much storage do you need?

Well my favorite movies collection occupies 5TB plus another TB of music and photos. That would work out pretty expensive with SSD.

 

[reefduffer]

There's a recurring theme in technology; older technologies have longer lifetimes than you might think. We're a long way from ssd being cheaper per bit than rotating hard disk. Sure, Moore's law says ssd gets ever cheaper, but so does rhd. Ssd has a number of advantages that offset the higher cost, but they aren't always compelling in every application. Ssd densities have already pushed the physics to the point where exotic data allocation strategies are needed to compensate for "wear-out" effects (yes, in solid state) so the slopes going forward won't be as different as you might think.

It's not the same technology, but the same concept: I'm reminded of the advice "Never underestimate the bandwidth of a station wagon full of magnetic tapes driving up the interstate". Fiber optics isn't necessarily faster for data transfer. Depends on how much data, and what your latency requirements are.

Do I want a rhd in my iPod? No, in fact I held off buying an iPod until flash versions became price competitive about 6 years ago. But as backup storage in a server room, even if it becomes another layer in the storage hierarchy with ssd "caches" in front of it, rhds will hang around longer than you think. The rhd manufacturers aren't ignorant of ssds, and when the business stops making sense, they'll stop making them. Most of them are, or can be, in the ssd business, they'll sell what they can make money selling, and the cost of entry to flash technology isn't that high for an existing rhd company. In a lot of ways, it's the same business.

But if they still think they can make a competitive product with acceptable sales volumes by pushing the areal density of rhds another 25% every couple of years, by doing things like using helium to reduce head gap height, then they'll keep pushing the technology.