WD Builds High-Capacity, Helium-Filled HDDs 356
Lucas123 writes "Western Digital subsidiary HGST today announced that after 10 years of development it is preparing to release 3.5-in data center-class HDDs that are hermetically sealed with helium inside. The helium reduces drag and wind turbulence created by the spinning platters, all but eliminating track misregistration that has become a major issue to increasing drive density in recent years. Because of that, HGST will be able to add two more platters along with increasing the tracks per inch, which results in a 40% capacity increase. The drives will also use 23% less power because of the reduction of friction on the spindle. HGST said the new seven-platter helium drives will weigh 29% less per terabyte of capacity that today's five-platter drives. In other words, a seven-platter helium disk will weigh 690 grams, the same as today's five-platter drives."
Why not a vacuum (Score:5, Informative)
Those of you wondering why they don't just use a vacuum inside the drive. Hard drive heads ride on a cusion of air (or in this case, a gas of some kind) so that they don't crash against the drive.
Re:This explains it! (Score:5, Informative)
Re:Disaster (Score:5, Informative)
Actually, you good-naturedly admitted your facepalm moment.
You may keep your card, good sir.
Re:But the cost? (Score:4, Informative)
Is this going to be cheaper than SSD?
Yes, of course. Only a tiny amount of helium is used.
Also, are these going to be significantly faster than the standard five platter density drives?
As usual for density increases, transfer rate goes up, seek time is unchanged. Moving disks even further into the role formerly occupied by tape. Maybe the reduced friction (= less heat) could make 10K drives more practical, improving seek time but probably also being a boutique product squeezed between SSD and 72k disks, and thus expensive.
Not the first (Score:5, Informative)
Re:Done 40 years ago (Score:4, Informative)
Good question. Helium atoms are so small that they can escape through tiny cracks between metal grain boundries in metals. Normal air does not. The only thing I can think is that they used some kind of penetrating sealant.
Re:This explains it! (Score:5, Informative)
Likely to be a permanent condition.
Helium is light enough that it doesn't persist very well in the atmosphere(unlike the heavier noble gasses, that you can just distill out if the price gets high enough to pay for the energy needed), and it is only replenished quite slowly by alpha decay of assorted radioactives in the crust.
The only significant source is natural gas wells in proximity to suitable minerals over geologic time and equipped to capture the helium when the product is brought to the surface.
Re:But the cost? (Score:2, Informative)
Weight matters in the datacenter too... Drives are probably the densest component in a rack. A fully populated rack can easily weigh over 2000lbs (900KG for people who use a sane system of measurement). A well managed datacenter (and competent sysadmins) should include consideration of the structural load (on the building), point loads (particularly on raised floors), and weight in transit. After you've done a few datacenter moves, you begin to think of things like the weight limit on elevators, truck lift-gates, ramps leading to a raised-floor facility, etc.
Re:Done 40 years ago (Score:4, Informative)
No, he's thinking of helium. Helium is monoatomic and hydrogen is diatomic. A single atom of helium is smaller and leaks more easily than a two-atom molecule of hydrogen. Helium is the most difficult gas to contain. It will seep through solid steel.
Re:Done 40 years ago (Score:3, Informative)
Gases will always flow from a region of higher partial pressure to one of lower pressure; the larger this difference, the faster the flow. [wikipedia.org]
In other words, the O2 and N2 will have no effect on the H2 diffusing out.
Secondary advantage to helium (Score:5, Informative)
Another advantage of using a drive filled with helium is better thermal conductivity than air (0.142 vs 0.024) . The heat generated by the inner workings of the drive will be conducted to the outer case, keeping the inside cooler.
Not necessarily (Score:4, Informative)
According to Wiki, for half a century the US Constitution was hermetically sealed inside a glass container with helium.
http://en.wikipedia.org/wiki/Hermetic_seal [wikipedia.org]
Re:This explains it! (Score:4, Informative)
"Likely to be a permanent condition."
Well, you can thank the U.S. government for that.
We used to have the world's largest helium supply, by far, in the U.S. Strategic Helium Reserve, until the government decided to do away with it just a few years ago.
And now there is a shortage. Imagine that.
Re:Disaster (Score:5, Informative)
Yes. In chemists' jargon, "oxidation" is the process of giving away electrons. The process is named after the most common electron recipient, oxygen, but lots of molecules can be oxidizers.
Or to put it another way, if I mix hydrogen with oxygen, and you mix it with chlorine, and we both add a spark, we'll get pretty much the same effect. Either way, there'll be a tremendous bang, and neither of us will have any eyebrows. Of course, after a few deep breaths, your nose might start bleeding as hydrochloric acid eats your mucous membranes, but that's a separate issue.
Re:Done 40 years ago (Score:3, Informative)
Pressure does not work that way. This is a standard thermodynamics brainteaser. The helium will diffuse out until the partial pressure inside equals the partial pressure outside (which is zero). The partial pressures of other gases on either side of the diffusion membrane are irrelevant.
Re:Low pressure to prevent leaks? (Score:5, Informative)
Look up partial pressure in some physics book. If you have gas in some leaky container at some pressure, it doesn't matter that there is another gas at another pressure outside of that container. Even if the different gases are at equal pressure, what will happen is that each will diffuse into the other through the aperture. That diffusion depends just on the respective partial pressures.
Helium at a lower pressure does not face an uphill battle due to the excess pressure outside of the container. It just diffuses through the apertures in the imperfect containers at a lower rate due to its lower pressure.
What would slow down the leak would be if there was helium on the outside at the same pressure, because then there would be as much helium diffusing into the container as diffusing out. But there is very little helium outside the hard drive, it being such a rare gas.
I think that if the apertures in the leaky container are such that helium can escape, but air cannot get in, then in fact the container will slowly evacuate. Pressure in that container will gradually drop as helium escapes, while hardly any finds its way back in.
Re:Why not a vacuum (Score:4, Informative)