Helium leak from MRI disabled all Apple devices in the facility

old.reddit.com

25 points·fanf2··5 comments

How does one of the most inert gasses in the universe 'interact with microcontrollers' ? The only reason it can kill humans is because it displaces a critical gas (oxygen) if breathed in massive quantities. Is it displacing some other gas that is required for these microcontrollers to work? (seems highly unlikely)

TL;DR Possible fault in an onboard MEMS oscillator from helium exposure.

https://file.scirp.org/pdf/JST_2013122009560886.pdf

> Exposure of absolute pressure sensors, resonant microtube density, binary concentration sensors and chip-scale vacuum packaged pirani gauges to room temperature helium resulted in a gradual drift in sensor output. No effect was found for differential pressure sensors and pirani gauges vacuum packaged with ceramic or metal packages. The observed results apply to other vacuum packaged MEMS devices such as gyroscopes, voltage controlled oscillators, infrared and Coriolis mass flow sensors. Potential causes for this loss of hermeticity are discussed as well as application limitations for MEMS sensors.

I love how sure everyone was that it's EMI.

Now I'm thinking of some crazy scenario where someone saves the day by putting an iPhone in a helium balloon to reversibly disable it.

One of the comments is this:

"Try one of two things:

1) Hold power and home button for 10 seconds and see if the device reboots.

2) Degauss the device then try step 1.

Source: Had an arc welder in proximity lock my 4s up once."

Looks like the salient comment:

Helium affecting Apple device's MEMS clock source and IP67 not helium-proof

https://old.reddit.com/r/sysadmin/comments/9mk2o7/mri_disabl...

Reply on news.ycombinator.com