16
$\begingroup$

A sufficiently strong electromagnetic pulse can/will destroy smartphones and computers.

I know somebody who went into MRI machine and forgot a Visa credit card in his pocket. The card was toast and he had to get a new one. A mobile phone in an MRI probably wouldn't fare better.

But a big part of the human body itself is based on electric signals. The brain and nervous system, including the heart, works on electric signals. And those signals have to go to very precise places. There is an area of brain processing vision, another is responsible for speech, etc. Also the heart function depends on precisely timed signals traveling very specific routes.

So it would seem that a trip to an MRI scan should totally fry anyone's possessing brain and heart.

Except it doesn't. An MRI scan is harmless (if you are not allergic to those injections they give).

Why?

And then there are those electromagnetic pulse devices they show in Hollywood movies. While totally trashing electronics of bad guys, fellow humans are always shown unharmed.

Again, why should the brain be different?

$\endgroup$
7
  • 29
    $\begingroup$ Because your brain is not a silicon integrated circuit. $\endgroup$
    – Jon Custer
    Commented May 5, 2023 at 18:49
  • 3
    $\begingroup$ An MRI is pretty much just a big static magnetic field. $\endgroup$
    – D Duck
    Commented May 6, 2023 at 9:53
  • 8
    $\begingroup$ I whipped my head around real fast when sat in an MRI, and saw goldfish. I moved more slowly after that. $\endgroup$
    – Neil_UK
    Commented May 6, 2023 at 14:51
  • 4
    $\begingroup$ An MRI isn't powerful enough to do that, but getting close enough to a magnetar definitely could $\endgroup$
    – eps
    Commented May 6, 2023 at 15:07
  • 5
    $\begingroup$ "MRI scan is harmless (if you are not allergic to those injections they give)." Just by the way, there are plenty of MRI sequences done without contrast. Also, do we really need a WP link to Visa? $\endgroup$ Commented May 7, 2023 at 0:30

9 Answers 9

26
$\begingroup$

Y'know that spark that jumps between your finger and a doorknob on a dry day in winter? That spark is enough to break down the "gate oxide" insulator between the gate electrode and the body of a microscopic field effect transistor. The insulator is so thin, that it only takes maybe a hundred volts or so to punch through it, ruining the transistor, and thereby ruining the integrated circuit chip of which the transistor is a critical part. There is nothing in your flesh that is so fragile or so critical. (Flesh heals, IC chips don't.)

Also, Y'know how there are wires all over the circuit boards in an electronic device? Those wires are like little antennas that can convert an electromagnetic pulse to an electrical pulse with enough voltage to toast one of those crucial transistors. There aren't any wires like that in our bodies.* The effects of electromagnetic radiation on our flesh is much more diffuse.


* Not true for everybody. My dad has an implanted cardiac pacemaker. He isn't allowed anywhere near an MRI scanner, and he might not fare as well as his neighbors if a nuclear weapon is exploded in orbit above the city where he lives.

$\endgroup$
10
  • 2
    $\begingroup$ "There aren't any wires like that in our bodies." this is not quite true. Our whole body is not a bad electric conductor. While humans have not been used as antennas, there have been experiments to use trees in a dense forest as HF, below 100kHz, antennas. It can work because the tree juices are not too bad conductors. You run a few loops around the tree as a kind of magnetic coupling to induce the currents. $\endgroup$
    – hyportnex
    Commented May 5, 2023 at 20:18
  • 24
    $\begingroup$ @hyportnex "humans have not been used as antennas". You must be too young to remember how the "rabbit ears" on VHF televisions sometimes worked much better with somebody touching one of them ツ $\endgroup$
    – John Doty
    Commented May 5, 2023 at 21:02
  • 4
    $\begingroup$ @JohnDoty actually I remember being the youngest in the family always to get that damned assignment to be holding on to the rabbit ears. Anyhow, I meant to say not any kind but specifically "transmitting antennas". I think you agree that kind of application would be a bit unusual for a human body... $\endgroup$
    – hyportnex
    Commented May 5, 2023 at 21:11
  • 6
    $\begingroup$ @JohnDoty That should have been the plot of the new Matrix.... Machines have run out of good antennas and now need humans to hold their rabbit ears. $\endgroup$
    – Therac
    Commented May 7, 2023 at 9:53
  • 3
    $\begingroup$ @ShoulO On a circuit board, the wires are surrounded by insulation. Nerves are surrounded by conductive salt water. $\endgroup$
    – John Doty
    Commented May 7, 2023 at 12:39
18
$\begingroup$

The trouble with circuits is that you have both highly conductive wires and highly resistive insulators, often very thin. When the magnetic field induces a voltage around a loop, the electric field winds up focused on the insulators. Very large fields may be induced. If the loop is closed, so no voltage is induced, the wire concentrates the induced current and may overheat if it's too thin.

A body doesn't have that contrast: the conductivity isn't strongly variable from place to place, so it concentrates neither the electric field nor the current.

$\endgroup$
8
$\begingroup$

A strong electromagnetic pulse can mess with your brain. See this video https://youtu.be/AXxhX0Pmm8w about transcrainial magnetic stimulation.

The MRI is just a big magnet and because the field is static there is no induced currents in the brain.

The magnetic strip on a credit card is erased by the MRI magnet because the field has aligned all the magnetic domains in the strip.

$\endgroup$
20
  • 1
    $\begingroup$ The field may be static, but while you are being moved into the machine the field is time-varying, from your point of view. A time-varying magnetic field induces an electric current. Why aren't you electrocuted? $\endgroup$
    – d3jones
    Commented May 6, 2023 at 22:50
  • 4
    $\begingroup$ The main static field, $B_0$, is approximately 1 or 2 T, wheras $B_1$ is on the order of a mT. $\endgroup$
    – D Duck
    Commented May 7, 2023 at 9:14
  • 3
    $\begingroup$ @d3jones The movement of the bed in an mri scan is too slow to make induced currents and potentials relevant. The intensity of the induced fields are proportional to the relative speed between the static field and the moving object. That's why regenerative braking in EVs becomes irrelevant at very low speed and they still need mechanical brakes to bring the vehicle to a full stop or to keep it stationary. $\endgroup$ Commented May 7, 2023 at 10:49
  • 4
    $\begingroup$ "The MRI is just a big magnet and because the field is static there is no induced currents in the brain." That's not entirely correct: MRI scan works by irradiating you with EM high frequency radiation while you are subjected to a static magnetic field. This latter serves to align atomic dipoles so that when the tissues are irradiated the difference in the response to the variable EM field (the atomic resonance) will highlight differences in tissues composition and state. (Very rough explanation). $\endgroup$ Commented May 7, 2023 at 10:52
  • 2
    $\begingroup$ @Aron "High frequency" means different things to different people. In radio communication, where it has an internationally standardized definition, it means 3 MHz to 30 MHz. In other contexts, it's less well defined. $\endgroup$
    – John Doty
    Commented May 8, 2023 at 21:37
8
$\begingroup$

A general answer: Humans and computers are different in both their internals and protective envelope.

One could invent at least few different electromagnetic (and even more mechanical) actions that can damage a human and be safe for a smartphone. Or the other way round.

e.g. you can apply mains voltage (230V) at two points at the screen of the phone with no apparent ill effects. Don't try this on human skin, it is known to be deadly and explicitly forbidden in the electrical code. Most phones also survive accelerations of 100g (in fact quite often, if the owner is nor careful). Humans generally don't survive 10g.


In regard to the MRI scanner: Its main feature is the strong magnetic field.

Human bodies are impressively insensitive to quite strong magnetic fields. Because of some lucky coincidence, no substance in the human body is ferromagnetic.

Other living creatures may not be as much lucky.

Ferromagnetic materials (e.g. most steels and most iron-containing minerals) strongly interact with the magnetic field. The exerted forces can deform or break something.

Is is not that the human body is completely inert in regard to magnetic fields - e.g. water is diamagnetic and strong enough magnetic field (clearly way above the MRI scale) may stop one's blood from circulating. Diamagnetism is always a great deal weaker (orders of magnitude) compared to ferromagnetism.

On the other hand, the magnetic stripe in the credit card is especially made to react to magnetic fields, this is how it is programmed in the first place. What happened to your friend's Visa is that it simply got erased (in terms of magnetic stripe recording). The precise magnetic picture on the stripe got replaced with a meaningless single magnetic field. If the card had chip as well, the chip would have probably survived, but chip-only cards are somewhat usable only in parts of EU.


In regard to nuclear-scale EMP:

Believe it or not, it is also quite selective in what it damages. Things that are known to suffer from it are long (kilometer-scale and longer) wires and equipment directly connected to these wires.

Unfortunately, these are in the first place the power lines and the power transformers.

In the past, communication cables (phone lines and likes) were at risk, too, but today anything longer than 300-500m is invariably an optical cable - pretty much safe in this regard.

A cell phone, a tablet or a computer are safe either - because of their size. They may only indirectly suffer because of their connection to the power grid. Anything smaller than a car or maybe a bus is absolutely safe.

$\endgroup$
7
$\begingroup$

EMP damages stuff by inducing voltage, according to Lenz's law.

Induced voltage is proportional to:

  • Rate of change of magnetic field, ie amplitude * frequency
  • Area of the loop the voltage is induced in, which is basically the size of the antenna

For example if lightning strikes close to your house, this will create a sharp magnetic field spike. The most likely victims are equipments that sit at the intersection of cable loops of large area, like phones, DSL routers, etc. A mobile phone sitting on a table has a very small loop area to act as an antenna, but a DSL modem which is connected to the grid, to a phone line, and to an ethernet LAN has many opportunities to form large loop area antennas from all this cables.

If a device is only connected to mains, the loop area would be the area between the conductors in the mains cable form the device to the nearest lightning arrestor. Since this area tends to be small, these devices are much less vulnerable, unless the lightning bolt directly hits aerial wiring, but in this case we're no longer talking about EMP and induced voltage, but the lightning current circulating directly in the wires.

MRI machines create a strong static magnetic field to align spins, on the order of 1 Tesla. On top of this, coils create a variable field gradient up to 50 mT/m which is not that much, and the HF field is much lower.

It will absolutely induce eddy currents into your body, but your body is not forming a loop. You can imagine many very small loops of tiny eddy currents induced everywhere into the patient, but there is no large loop, so there is no large voltage or current.

Normally during a MRI you will lay down with your arms at your sides. If you do form a loop with your arm by touching your thigh with a finger, then you'll definitely feel a tickle of electricity from the induced current. It's harmless.

Rings and other jewelry are also loop antennas.

What's dangerous in a MRI is not the induced current from variable magnetic fields, it's the static field which will grab anything ferromagnetic and suck it into the machine.

enter image description here

Oxygen cylinders, chairs, whatever, if it gets near the big magnet it'll accelerate like a rocket and shoot through the hole in the toroid magnet with enough force to crush or perforate whoever is inside. Ferromagnetic objects are most attracted to the center of the torus, which is where your head is if you're doing a head MRI, so it's important to be careful. If a steel pen gets in there, they'll need a winch to pull it out.

enter image description here

The problem with pacemakers is not the pacemaker being zapped by the field, it's the force pulling on it and causing it to move inside the patient and turn his insides into mush. Likewise if you go in a MRI with a cellphone it'll probably survive the magnetic field, but it'll end up inside your head.

$\endgroup$
6
  • 4
    $\begingroup$ I can confirm the bit about ferromagnetic objects and MRI magnets. When engineering an MRI machine, we got a fork-lift truck stuck to the magnet. Oooops! $\endgroup$
    – Neil_UK
    Commented May 6, 2023 at 14:53
  • $\begingroup$ "It will absolutely induce eddy currents into your body, but <...> there is no large voltage or current" - so what you are saying EMP will affect human, but it has to be strong one. $\endgroup$
    – ShoulO
    Commented May 6, 2023 at 20:30
  • 2
    $\begingroup$ Yes, the power level would have to be many orders of magnitude higher than a MRI. In an induction cooker, eddy currents are induced into a chunk of steel which has low resistance, so the current is huge which creates lots of heat. Resistance of flesh is much higher so even with the same high intensity AC magnetic field, induced current would be tiny. $\endgroup$
    – bobflux
    Commented May 7, 2023 at 8:15
  • 4
    $\begingroup$ Also EMP damages electronics by inducing voltage, and damage due to overvoltage doesn't require much energy at all (think puncturing 10 nanometer thick MOSFET gate oxide). On the other hand, because the entire human body is slightly conductive, most of the EMP's energy would turn into heat. Flesh is mostly water which has very high thermal capacity, so you'd need lots of energy to heat it enough to result in a wound. $\endgroup$
    – bobflux
    Commented May 7, 2023 at 8:20
  • 4
    $\begingroup$ @bobflux I would add that a single neuron dimensions are around 10-20 micrometers, which is thousands of times bigger than today's nanometer-scale features of microchips. So even if a neuron and a transistor were made of similar materials, the sheer dimensions would make the transistor thousands of times more easy to damage just because it would need thousands of time less thermal energy to do that (in a "spherical-cow" "only-heat-can-cause-damage" naive but very intuitive simplification). $\endgroup$ Commented May 7, 2023 at 11:01
2
$\begingroup$

Aside from conductors moving through the strong magnetic field and inducing a voltage (possibly resulting in a harmful current), a strong static magnetic field can saturate the (typically ferrite) cores used in switching power supplies. That will dramatically reduce the inductance and can cause off-line switching power supplies and DC-DC converters to fail from overcurrent.

Humans don't have anything like that going on internally (well, unless we added it, in which case it might be vulnerable).

$\endgroup$
2
$\begingroup$

Why does an MRI machine or other EMP generating machine not damage humans, but it will fry computers?

An MRI machine does not generate EMPs of strength high enough to mess anything up in most computers that don't have ferromagnetic storage media. I have a mostly non-ferromagnetic electronic watch and its tiny battery gets a tug when coming near the scanner, but otherwise nothing bad happens to it. Don't try to replicate this experiment, but that's just to indicate that an MRI will not generally "fry" electronics at all.

You just won't be allowed to come anywhere near an MRI magnet with most computers, since they got lots of ferromagnetic elements in them, and those will really want to swing into the middle of the bore. And, due to liability reasons, nobody will let you come near an MRI machine with anything other than clothes on you. So it's not like you can just walk up to one and test it yourself :)

If you had a computer made entirely out of non-ferromagnetic components, most likely nothing much would happen if you brought it into a magnet of an operating MRI. Maybe it would reset due to RAM corruption, but it wouldn't "fry".

$\endgroup$
4
  • $\begingroup$ I'd be careful with your watch example, to avoid giving people a false-sense of security. Maybe your digital watch had no moving parts and its case was mostly plastic or a non-ferromagnetic alloy, but other watches (e.g. battery-powered analog watches) may be more susceptible. I had a university fellow who worked with MRI for liquid crystal characterization. His apparatus was no bigger than a cylinder of 1m height by 50cm diameter (so much less powerful than a MRI scan machine for people). ... $\endgroup$ Commented May 7, 2023 at 11:17
  • 2
    $\begingroup$ ... In the lab the "absolutely keep away" red line on the floor was just about a meter away from that cylinder and there was a yellow "danger" line a meter further away. He told me that if you wore a watch in the red area the magnetic field was enough to bend the hands of a watch and could even make the glass of the watch pop out. He lost two cheap "Swatch" watches this way and a female colleague of him risked an earring to be torn away from her earlobe! $\endgroup$ Commented May 7, 2023 at 11:18
  • 2
    $\begingroup$ BTW, yep, that was a quite risky setup, with the control PC placed at 3m away with no further protection, but that was the situation with underfunded Italian universities labs in the 90s * sigh! *. The whole thing was placed in an ex-storeroom in the basement of the faculty. The room was no bigger than 4x4 meters. $\endgroup$ Commented May 7, 2023 at 11:19
  • $\begingroup$ @LorenzoDonatisupportUkraine The watch I had was polycarbonate with a tiny PCB and the battery contacts were conductive plastic. The only ferromagnetic alloys were the case of the battery. And it gave a good tug for sure. I don't recall what brand of a watch it was - it was one of those minimalistic designs without even a backlight. The LCD survived all that abuse. It does get uncomfortable when one's wrist is being tugged at though. $\endgroup$ Commented May 9, 2023 at 13:02
1
$\begingroup$

The magnetic flux density of an MRI is not enough to disrupt biological systems, because carbon based life-forms don't coalesce massive pieces of nickel, cobalt and iron which can conduct current and cause disruptions.

The pure metals in electronics can conduct high currents induced by pulsed magnetic fields, in similitude to an AC adaptor does from the magnet of an alternating current source.

$\endgroup$
-1
$\begingroup$

Why does the lack of breathable oxygen not damage computers, but it slays humans within minutes? Same reason.

MRI machine deletes the data contained in computers, but it doesn't delete the data from humans. Use ionizing radiation, and you could delete the "data" from humans as well, specifically the data contained within DNA. This is called radiation sickness and sufficient "data" deletion is fatal. MRI machine cannot delete "data" contained in human cells.

The story saying that "Visa card was toast" seems to be heavily colorized and ridiculous. The card certainly had the data on its magnetic strip erased, but it is impossible that it has been actually "toasted", as the magnetic strip is made of ferromagnetic metal oxides which are nowhere near the electrical conductivity of metals. Visa card is mostly made of plastic. MRI cannot in any reasonable way raise the temperature of such card enough for it to burn.

$\endgroup$
1
  • 3
    $\begingroup$ Visa card was no longer recognized by the ATM. Word 'toast' here indicated that ir was unusable anymore $\endgroup$
    – ShoulO
    Commented May 6, 2023 at 20:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.