It's very hard to keep helium in anything. Hard drives are not sealed, they have (filtered) holes to allow for expansion / contraction of the internal atmosphere caused by heating / cooling cycles.
Helium is not a renewable resource, and we're running out.
I keep on thinking that we've reached the limits of traditional drives, and hoping that we dump a lot of investment into SSDs and new SSD architecture. It just feels weird to have something so mechanical in a computer.
Theoretically we're running out of Helium, just like we're theoretically running out of oil, but actually we've still got plenty left. Because of the US dumping their stocks the price has been artificially low making it unprofitable to extract from natural gas which contains a lot, and it has been just released in the atmosphere because of that instead of contained. As soon as prices start to reflect the extraction price there won't be a problem anymore.
This makes me wonder if these new hard drives will have a limited shelf life. I know they have spent so much research on keeping the He in, but it's more likely that they have just slowed down the leaking, so will this hard drive still be spinning in ten years? Some hard drives are expected to keep running for at least that long.
Also, I worry about air pressure changes. If a particularly nasty weather system came along with a very low pressure centre, and the hard drive is completely sealed, then it will have to either flex to accommodate the He expansion or be very rigid indeed. If it flexes, does that de-calibrate the head alignment? Is the case strong enough to deal with 200N force pulling it apart?
I would imagine they include a bladder of some sort to handle expansion and contraction due to temperature changes while the drive is operating, in order to keep the internal pressure fairly constant. That's probably a wider range of pressure change handling than weather systems would require. (There was a time when I knew how to do the math for that, but I haven't used my MechEng degree in a long, long time.)
I don't think it took them 10 years to figure it out. It's probably only now that the cost is justifiable when compared to other techniques that increase capacity.
