WD Builds High-Capacity, Helium-Filled HDDs | Hacker News Reader

WD Builds High-Capacity, Helium-Filled HDDs

computerworld.com

52 pointsXERQ30 comments

Oh this will be fun, the only gas harder to keep inside is hydrogen. While its absolutely true that you get a huge benefit in head flying by switching to helium it also means that there is pretty much a guaranteed lifetime cap on the usefulness of the drives. As someone who tries to run drives 24/7 for 3 years before tossing them this probably makes my life harder. Especially if the 'failure' mode is that they just get slower and slower as the heads have a harder and harder time staying on track as other gases migrate inside.

ha! that makes me think of failure modes of SSD's.. just getting slower (but for a different reason). Imagine if the florist who does helium party balloons starts offering the service to refurbish your HD.

Ordinary magnetic platter HDs slow down with age too, though I'm not sure why.

theory: your outer tracks are filled with data which is older and that you presumably read less

Simply swap the air in your facility with helium. As a bonus, corrosion and fires will be a thing of the past.

Sadly the heat conduction properties of helium would not make that really practical. One could consider submerging the data center under water (great heat conduction, relatively easy to keep it from migrating into a helium filled cavity) but finding data center techs with scuba experience, very tough.

What about water vapor? It has a lot of heat capacity, and should have about half the momentum exchange with the hard drive head assembly. Preventing condensation would be the trick.

Filling HDDs with Helium seems like a bad idea since the Earth is running out of Helium.[1]

Increasing prices could very well make the technology infeasible before it even comes to market.

[1]http://digitaljournal.com/article/321439

As long as we have enough to fill it into toy balloons en masse and drive airships with it we'll probably have enough to fill up those drives.

On the other hand you're right, Helium is a precious resource - one of many that could be feasibly mined elsewhere in the solar system (in the hopefully not-so-far future).

As long as we're thinking about that timescale, it's one of the easier elements to fuse from "raw ingredients".

And it's the only compressed gas that gets down to -270C for cooling MRI machines, hydrogen only gets to -253C and is much more dangerous.

... and they use it in NMR. I think the one at the local university takes several dewars of helium just to bring it online.

The cost of helium will be insignificant. He has a molecular mass of 4, so 4 grams gives you 22.4 liters at standard temperature and pressure. So 1 gram of helium should give you about 50 drives (at 100cc each).

Helium is like, 25 cents a gram. 2 cents a hard drive. If it goes up 100 times in price, it'll still be cheap compared to the cost of the drive ($2 a drive). It'll probably still be cheap, compared to the added cost of sealing the drive.

I wanted to say "Eh, we can always fuse some more", but running low by 2025? That's pretty damn soon.

Yeah, we all know that commercial nuclear fusion reactors are 50 years in the future.

Commercial fusion reactors that don't generate energy? Totally.

I was hoping to take some technical diving courses but it sounds like the raw materials price is going to go up. Helium is often mixed with nitrogen and oxygen to provide gas mixtures to reduce the narcotic impact of nitrogen. I think there's been some experimentation with hydrogen/oxygen mixtures but they're a relatively low concentration of oxygen (4%), not to mention that it scares the hell out of me.

I wonder what the effect on divers, recreational or professional, will be if/when helium is either too expensive to use for deep diving or just not available? Perhaps neon/oxygen, though that would also be very expensive.

Please stop spreading this "running out of heilium" nonsense. Natural gas contains plenty of helium, and making LNG produces helium as a byproduct. As soon as the US Government stops liquidating it's Cold War reserve of helium at super low prices, every LNG facility will compress the helium and sell it. Thanks to fracking, and a return of LNG production in the US, there will be plenty of helium for the foreseeable future.

Natural gas has single digits percentages of Helium. It's still a finite resource on sensible timescales (yes radioactive decay replenishes supply eventually). It has many important uses. There are good reasons to object to it being squandered in children's party balloons.

I seriously doubt that even at higher prices it would not be commercially viable to extract Helium from Natural Gas reserves.

The tiny amount of Helium contained in Natural Gas probably isn't worth the investment to separate out, especially since the company would probably have to separate it from Nitrogen or another non-reactive gas (liquify air => expensive).

In which case, they'd just release it into the atmosphere.

So why not a vacuum? Vacuumed KERS systems rotate an order of magnitude faster. I get that a KERS system can afford heavier casing than a HD but still, if your cases can hold helium you ought to be able to hold a reasonable vacuum too.

My understanding is that hard drive heads float over the platters on a cushion of air. That's why high altitude applications need special sealed hard drives.

Specifically, in a vacuum the heads crash into the platter, damaging both.

Something doesn't make sense here.

Turbulence is related to viscosity as far as I know.

Viscosity of air: 18.6 micropascal seconds.

Viscosity of helium: 20.0.

For comparison hydrogen is 9.0 and argon is 22.9.

Actually - maybe a higher viscosity is better (it causes the turbulence to die out)?

But in that case why not use argon? It's cheaper and a LOT easier to contain.

Yet the article talks about light atoms?

The turbulance is related to viscosity, but the force exerted on the drive head is related to turbulance and mass. The density of helium is an order of magnitude lower than the density of air.

The link goes to the second page of a two-page article, here's a better link (all on one page):

http://www.computerworld.com/s/article/print/9231220/Helium_...

So what is the capacity of these new drives? Nowhere in the article it seems to be mentioned.

This is great.. but there is an international shortage of hydrogen, and unfortunately, I think MRI scanners should have first dibs. Having said this, it will be great when nuclear fusion is cracked and there is an abundance of the stuff.

You mean helium, I presume.

Yes, I am a tool.

Reply on news.ycombinator.com