Helium breaks newer iPhones
By now many of you have heard about a case where helium was leaking during the installation of a MRI scanner at a hospital in Morris, Illinois. The helium was apparently low enough in concentration that no one noticed, i.e., their voices didn’t become higher pitch but it was enough to disable many newer iPhones.
There are some details in the story that don’t seem right. I’m not saying it’s wrong, I’m just wondering if they are off or being misreported.
FTA: “I discovered that the helium leakage occurred while the new magnet was being ramped [down to cool it]. Approximately 120 liters of liquid [helium] were vented over the course of 5 hours. There was a vent in place that was functioning, but there must have been a leak. The MRI room is not on an isolated HVAC loop, so it shares air with most or all of the facility. We do not know how much of the 120 liters ended up going outdoors and how much ended up inside. Helium expands about 750 times when it expands from a liquid to a gas, so that’s a lot of helium (90,000 L of gaseous He).”
First quibble is that you ramp up the voltage after the liquid helium is done filling, i.e., the magnet is already cold. I assume that the vent that they are talking about is a quench stack. A quench stack is a vital safety feature. In the event of an accident where the helium boils of rapidly, the gaseous helium needs to be vented out so that people do not suffocate. I wrote more about quenching here:
Quenching your thirst for MRI science
If there was a decent size leak in the quench stack, you would see the helium because it’s cold and if enough helium leaked, as suggested, the room would be 5 to 15 degrees F colder. I’m not arguing against the theory that helium was able to interfere with the MEMS (microelectromechanical systems) based clock in the newer iPhones. I’m doubtful about the concentration of helium that was present. The take home message for me is not that helium can stop the MEMS clock in an iPhone, it is that a very, very small amount of helium can turn your new iPhone into a brick (hopefully temporarily).
I’ve written about the importance of helium before:
Liquid helium is a lifeline for NMR and MRI research
https://plus.google.com/+ChadHaney/posts/99SpR28xBme
Helium Cache to the Bank?
https://plus.google.com/+ChadHaney/posts/PwWgiXXLamQ
https://ifixit.org/blog/11986/iphones-are-allergic-to-helium/
Interesting vulnerability. I’m puzzled about the concentration, however – why would it affect devices elsewhere in the building?
Wow, that is so bizarre! But really excellent write up on troubleshooting the problem.
I the weirdest problem that I saw back when I worked in electronics manufacturing as a technician, a batch of IC’s that had very slight radioactive contamination in their substrates.
I think it was beta radiation, but the result was random bit flipping. I wasn’t in on the diagnostic side of the problem, but it was certainly weird to watch it happen on a logic analyzer.
Bob Lai I think the effect wasn’t widespread.
Having worked in the diagnostic medical equipment installation industry for 10 years, this sounds like BS. While it’s completely normal for a brand new MRI to “boil off” helium, it doesn’t do it at a rate that would fill a room. Furthermore, you don’t ramp DOWN a brand new magnet. If you’re ramping down, you’re preparing to remove it. I have heard of defective safety valves letting out too much helium…and I’ve seen cases where a crew has taken too long to get the magnet connected to a cooling system, so the boil off can be significant by the time it’s done, but this sounds like an impossible thing.
Cheese Slap as I mentioned, I think they meant to say ramp up. I also think a lot about this doesn’t make sense.
I’ll believe the really massive electrical field of an MRI machine disabled the iPhone before I’ll believe it was the helium. How the f* does an inert gas disable electronics? That makes no sense at all.
John Poteet Android and older iPhones were not affected. The manufacturer of the MEMS based clock does mention helium can cause a problem. My biggest issues is that I don’t think that much helium was present. I wish they would have mentioned the spatial effect, i.e., were new iPhones affected that were on another floor in the hospital.