Is there an order in which different kinds of (unshielded) electric devices go out in the event of a high altitude EMP? For example, first small gadgets, then electric vehicles, then grids. Or do all unshielded devices go out pretty much at the same time, seemingly instantaneously, with no specific order or delays depending on the type of device?

  • 15
    $\begingroup$ Worth noting: any time you take an electrical component out of spec (such as zapping it with an EMP), there is a finite possibility of a latent failure. I don't know what you need for your story in particular, but I can say that if you got hit by an EMP and your hardware "seemed fine," that's no guarantee it wont start failing later. $\endgroup$ – Cort Ammon Jan 30 '17 at 17:32
  • 13
    $\begingroup$ futurescience.com/emp/EMP-myths.html $\endgroup$ – Matthew Whited Jan 30 '17 at 18:19
  • 3
    $\begingroup$ If your plot requires some devices to remain functioning after an EMP blast, you could try using a Faraday cage $\endgroup$ – Shokhet Jan 30 '17 at 18:23
  • 5
    $\begingroup$ Ville can you frame your question out a bit? As it stands this is simply a question on existing technology which isn't really what we do here on WB. Check out the help center for more details on writing questions and feel free to visit Worldbuilding Chat and discuss $\endgroup$ – James Jan 30 '17 at 20:56
  • 8
    $\begingroup$ I feel like this question would have been perfect suffixed with ", I need to know by Friday" $\endgroup$ – cloakedninjas Jan 31 '17 at 17:07

10 Answers 10


Last summer, a lightning bolt struck a tree right outside my apartment building (so close that I hit the floor). As you may be aware, lightning creates a small EMP blast. The power went out momentarily and when it came back on, my desktop PC would not boot and my TV no longer had sound (yes, everything was plugged in to a very nice APC UPS). The PC turned out to be a fried motherboard. A week or so later I noticed that my Wi-Fi was flaky. It was constantly dropping connections. That could be unrelated, but I threw it on the insurance claim as well. I wasn't the only one affected. My upstairs neighbor swears he saw a small ball of lightning in his bedroom, and he lost his TV, cable box, and cable modem. Another neighbor lost the DSL modem and a TV as well, but the worst (IMO) was the car that was parked directly under the tree. I don't know what fried on it, but it wouldn't start after that and had to be towed to a mechanic.

So to actually answer your question; anything that is going to be affected is immediately affected (well, immediately as in tens of microseconds). However, a device may sort of function after it's been affected and it may take a while to notice that it has failed (like my Wi-Fi router). So, an example might be a radio that is no longer able to receive but can still transmit, or maybe that radio can only send/receive on certain wavelengths.

If you had several circuits carrying current to a device, and only some of those circuits were blown out by the EMP blast, the remaining circuits could fail if the device they were powering drew too much current for the remaining circuits to handle. I can't speak for military or "hardened" devices, but I know consumer and professional electronics just aren't designed that way. In a consumer device, there is one power circuit, and if it doesn't go in the EMP pulse, it's 100% fine afterwards. In a commercial device (like the APC switchable UPS's I use at work), there are redundant power supplies. If one goes, the other is designed to handle 100% of the load. Designing something to only work if both PSU's are running would be silly (because it would be less dependable, not more).

  • 29
    $\begingroup$ Your equipment was not damaged by an EMP. It was damaged by an overvoltage surge caused by a lightning strike. $\endgroup$ – Matthew Whited Jan 30 '17 at 18:07
  • 33
    $\begingroup$ @MatthewWhited But that's exactly what an EMP causes. The EMP induces a current in conductive materials leading to overvoltage surges. Except that in the case of a proper EMP, there won't be any localized source that the current has to propagate too (why surge protectors won't help much) the current will be induced even in things that are not plugged in. $\endgroup$ – Shufflepants Jan 30 '17 at 18:32
  • 5
    $\begingroup$ @MatthewWhited I'm not sure if you're trying to say that EMP is harmless and its really the overvoltage that is bad, that EMP was not involved in damaging my equipment at all, or if you've confused an EM field with an EM Pulse. 1- My equipment was not directly struck by the lightning. Ergo, it must have been a radiant emission. 2- Lightning creates an EMP. Faraday proved induction. My UPS did not blow its fuse (but I did have a couple of breakers in my apartment that tripped). Ergo, no overvoltage on the power line reached my equipment (or I woulda been on the phone my APC rep first thing). $\endgroup$ – Tim Jan 30 '17 at 18:53
  • 17
    $\begingroup$ So, the electric flow from the lightning bolt that blew the tree apart went through the brick apartment wall, insulation and sheet rock, into my computer, TV, and router, and then into the ground without upsetting my overkill of a rackmounted UPS? Would it also have flowed through my apartment and across the hallway to the neighbor's to get their TV and DSL modem? $\endgroup$ – Tim Jan 30 '17 at 19:36
  • 14
    $\begingroup$ It's quite possible, likely even, that it induced a current inside other wires apart from power, e.g. network cabling. As an added bonus, the circuits these cables are plugged into tend to be far more delicate than the great big beefy traces and components in your power supply designed to deal with dozens of amps and hundreds of volts (incl. transients). And that's ignoring the more direct currents on the low-voltage traces themselves. $\endgroup$ – Bob Jan 31 '17 at 6:51

All direct effects happen at practically the same time (light speed from the EMP source still apply).

That said, if you will burn longer tracks on PCB, tracks in complicated devices would get burnt immediately - because these are thin and delicate. On the other hand, tracks on PCB in an old car might get only some damage, and electronic connected to them might be hardy enough not to fail immediately. In such case, a damaged diode or track might function correctly for some time, it's condition worsening with use, until complete fail.


EMPs work by creating rapidly changing electrical and magnetic fields. When these come into contact with electronics they induce a power surge that may fry electrical components. All modern electronics have weak/sensitive components that will fry instantly.

EMPs are also quick so if an electronic component doesn't instantly fry from it then it will most likely keep working, unless it is badly damaged and over a few seconds it keeps getting worse until it dies out.

Note: Weaker EMPs may simply force devices to restart since they won't fry the components but simply mess with the bits making the device run into a "Blue Screen of Death" or similar.

Edit: Since Tim added some real life experience I thought I could add some experiences I've had with lightning

One day when I was walking back home while listening to music a lightning hit kinda close and I heard some interference before finally hearing the thunder. I guess the current induced by the EMP wasn't strong enough to fry anything but only to create interference in the wires.

I've also had experience with lightning affecting devices connected to the grid.
The first time this happened my TV got fried and after that I got a UPS.
Now that I have a UPS whenever a lightning strikes, my UPS just shuts off, protecting my devices, and I have to restart it. The UPS has a battery so the devices can keep running with it powered off.

  • 1
    $\begingroup$ te anything powered off will not be bothered, but keep in mind things like televisions and often computers are not really powered off unless unplugged. $\endgroup$ – John Jan 30 '17 at 15:28
  • 3
    $\begingroup$ @John Can I have some source on that? Because I am sure the devices will still get hit by the EMP, just not by the power surge captured by the power grid and transferred to the device. $\endgroup$ – user31746 Jan 30 '17 at 15:53
  • 3
    $\begingroup$ @Masterzagh From what I undersatnd, there's a few effects. The EMP can cause direct damage by breaking down insulators inside the chip. The EMP can cause failures by flipping bits but not damaging hardware. The EMP can create a large potential across the computer by using the power lines as an antenna. Direct damage happens if your'e plugged in or not. Bit flipping is only an issue if you're powered on. Large power line voltage spikes can occur if you're powered off but plugged in. $\endgroup$ – Cort Ammon Jan 30 '17 at 17:20
  • 1
    $\begingroup$ It's a near certainty if it is a big enough EMP. Being powered up and being connected to any long wires both increase the chance of damage. Being disconnected, powered off, and inside a metal box = Faraday cage greatly decreases the chance of damage. If it is made of thermionic valves instead of semiconductors it is almost immune. $\endgroup$ – nigel222 Jan 30 '17 at 18:44
  • 2
    $\begingroup$ @Masterzagh That's my bad I should have specified I was referring to mostly disconnected devices, such as hand held devices or automobiles. You are correct that anything attached to the grid or long wires will be affected since those wires act as antenna. On the other hand a device being off is not a guarantee, but it is much harder to damage a device if it powered off an not attached to a giant antenna. futurescience.com/emp/EMP-myths.html $\endgroup$ – John Jan 30 '17 at 21:35

An Electromagnetic Pulse is a brief magnetic field.

When electrically conductive materials (like wires) pass through a magnetic field it causes electricity to flow through the materials. This is how a generator creates electricity.

In an EMP attack the idea is to create a very strong magnetic field that causes electricity to flow in wires and circuitry. The attacker is hoping to cause enough power to flow to burn out components. It doesn't matter if the device is plugged in or not, though long wires like power lines can build up more of a power surge if they are lined up right with the magnetic field, which can send a surge through an outlet.

Magnetic fields travel at the speed of light, so from an observer's perspective it will appear to affect everything at the same time.

Devices won't go down at the same time though, it will depend on how well they handle the pulse. If they're magnetically shielded the pulse might not be strong enough to affect it, or the pulse won't create enough power flow in some devices to hurt it. A weaker pulse can just cause the device to crash or reset.

  • $\begingroup$ Welcome to the site Jandsm, good answer. Check out the help center when you have a chance and let us know if you have any questions. $\endgroup$ – James Jan 31 '17 at 17:43
  • 2
    $\begingroup$ Just an FYI - We don't theorize that magnetic fields travel at basically the speed of light, we know they travel at the speed of light. It is not theory. It is a measurement. ;) $\endgroup$ – Tim Jan 31 '17 at 17:46
  • $\begingroup$ Thanks for pointing that out. When answering I just did a quick google and didn't see any definite commends on how fast a magnetic field travels. I didn't look long enough to find a definite answer so thanks for pointing that out. I've corrected it now. $\endgroup$ – jandsm5321 Jan 31 '17 at 18:13

EMP won't kill everything

It'll be totally hit-and-miss what gets nailed by EMP. A row of police cars, all same make/model, might all get their PCM (engine) computer killed except for one, while two more parked at a different angle might work fine except for power windows and radio. Boeing aircraft might be flyable (with the DirecTV and FMC not working) while 75% of Airbus aircraft are heavily damaged.

The power companies may find many of their Eaton controls took a hit but very few of their GE's.

A bunch of railroad lines, using older tech, may be totally unaffected, while the railroads are unable to control trains from their central dispatching centers and must go back to "old-school".

In case you catch what I'm implying, engineering mentality and build quality matters. I expect Boeing to be thinking hard about this, since a lot of their business is warplanes or military derivative of commercial airframes, like Air Force One and other VIPs, P-8 Poseidon, KC-767, etc. Eaton vs GE, no idea, it's just a made-up example.

You just don't know until you pop a nuke over one, and you're not going to do that, unless you take a serious disliking to the Test Ban Treaty.

  • $\begingroup$ There are conventional explosive EMP weapons. I'm very surprised we haven't seen them used yet. $\endgroup$ – Blackbeagle Feb 2 '17 at 2:32

If what you have is a single, strong pulse, then the timing differences are going to be based on physical distance from the source and the propagation of the wave. Since the wave travels at pretty near the speed of light, for practical purposes it's instantaneous unless the distances involved are quite large. Depending on their sensitivity to EMPs and the strength of the pulse, devices might catch fire, suffer component damage, suffer transitory malfunctions, or be unaffected. Devices may be affected differently depending on whether they are running or not at the time the pulse happens.

If, on the other hand, you have a series of pulses, then all of the above applies with the addition that devices that suffer only minor damage or transitory effects may have those effects stack up over time to cause actual failures. In this case, the less shielded/more susceptible devices will likely fail first, followed by incremental failure of the tougher devices as the pulses continue.

How "tough" a device is can be roughly determined from three factors: How easy it is to damage the components, how much wire is oriented in such a way that it will pick up voltage as the EMP goes past, and what percentage of the EMP can reach the device (faraday cages and distance can make a big difference.)


  • The power grid is made of exceptionally tough components, but it has miles of wire that act like giant antennas and pick up massive amounts of induced current. One, extra-large solar flare in 1859 plus miles of wire resulted in some telegraph offices on half the globe literally catching fire. (And, apparently, some of the ones that survived being able to send messages for a while without the power supply being hooked up.)
  • Your phone has much less wire and won't pick up nearly so much stray power, but its components are much more delicate and it doesn't take as much to damage them.
  • Put your phone inside an unplugged microwave, and the faraday cage that normally keeps the microwaves in will, instead, keep the EMP out (unless it's really, really powerful, but past a certain point it becomes more death-ray than EMP)
  • A series of light pulses could fry the phone without causing any significant trouble to the grid.
  • One, big pulse tuned to probably something in the ELF frequency range could toast the power grid without the phone even noticing. (Unless it was plugged into said grid at the time, obviously.)
  • $\begingroup$ Not as much as you'd think. Look close, power lines are actually twisted pair (well, twisted triples). The idea is each wire inducts equally, and cancel each other out. Doeesn'tbquite make them EMP proof, but it helps. $\endgroup$ – Harper - Reinstate Monica Feb 1 '17 at 7:29
  • 1
    $\begingroup$ "One, extra-large solar flare in the late 1800s" I believe you are thinking of the Carrington event, but that happened in 1859. Related reading: en.wikipedia.org/wiki/Solar_storm_of_1859 $\endgroup$ – a CVn Feb 1 '17 at 14:34
  • $\begingroup$ @Harper The lines from the transformer to my house are twisted around a steel cable so they don't get broken as easily. The HV lines going down the street run parallel and aren't even insulated, and the long-distance transmission lines that feed the substation are the same, just further apart and thicker cable. $\endgroup$ – Perkins Feb 1 '17 at 22:18
  • $\begingroup$ @MichaelKjörling Yep, that was it. I suppose it depends on where you draw the line between "early" and "late" 1800s. $\endgroup$ – Perkins Feb 1 '17 at 22:22
  • $\begingroup$ @Perkins correct, but that's to carry the conductors, not for EMP. The EMP twists on high-tension lines happen either subtly between poles (when spaced in a triangle, or particularly when a flat layout goes to a triangle for a couple of poles)... or with extra insulators at certain poles. Can also be seen on railroad "telegraph" wires. If you really look for it, it's quite conspicuous. Not every power company does it on every line. $\endgroup$ – Harper - Reinstate Monica Feb 1 '17 at 22:47

As the other noted, the EMP blast propagates from origin with the speed of light and its intensity decreases with a square of radius.

Also, some components are more vulnereable to the overload - microprocessor on the weak part, thick copper wire on the tough side.

The mode of failure matters.

  1. Direct failure: The working part is the one that was EMPed. Such devices, no matter their size, comlexity or age, will fail almost immediately the shockwave hits them. Fine electronics will fail instantly, tougher ones will last fractures of a second.
  2. Delayed failure: The EMPed part is a component of a more complex device. In this case the device will fail when the EMPed part is needed for the first time. Say there is really old car and EMP fried its alternator only. The car is working until it drains all the power from its battery.
  3. Failure initiation: The EMP overloaded some parts and their lifetime was significantly reduce. Or the EMP killed some part that was backed up by different part which is now constantly overloaded. After some time the now-overloaded part fails. Say there is powerline divided into several parallel and insulated lines and because of the overload part of it was overheated, several wires were cut and welded together. In the wounded part the current is higher than in the untouched wires and is heating up. At some point it will break.

tl;dr: The EMP damage front is moving with a speed of light. Before and after, no damage is dealt. The effect of the damage, on the other hand, may be dealyed with the respect to what was damaged and how serious the damage was.


For the most part EMP would fry anything solid state whether plugged in or not. For complete EMP immunity go old school technology- Vacuum Tube.. they are completely immune.

EMP is seen as a secondary effect of all types of nuclear weapons, and was a problem in the very first nuclear weapons test in July, 1945. The official technical history for that first nuclear test in 1945 states, "We can understand the difficulty of transmitting signals during the explosion when we consider that the gamma rays from the reaction will ionize the air and other material within hundreds of yards. Fermi has calculated that the ensuing removal of the natural electrical potential gradient in the atmosphere will be equivalent to a large bolt of lightning striking that vicinity. . . . All signal lines were completely shielded, in many cases doubly shielded. In spite of this many records were lost because of spurious pickup at the time of the explosion that paralyzed the recording equipment."

As I stated on another page on this site, consider this Cold War era quotation from a widely-read and highly-respected publication: "The United States is frequently crossed by picture-taking Cosmos series satellites that orbit at a height of 200 to 450 kilometers above the earth. Just one of these satellites, carrying a few pounds of enriched plutonium instead of a camera, might touch off instant coast-to-coast pandemonium: the U.S. power grid going out, all electrical appliances without a separate power supply (televisions, radios, computers, traffic lights) shutting down, commercial telephone lines going dead, special military channels barely working or quickly going silent." -- from "Nuclear Pulse (III): Playing a Wild Card" by William J. Broad in Science magazine, pages 1248-1251, June 12, 1981.

The reason that EMP is getting more attention now is that the critical infrastructure that sustains our lives is becoming increasingly sensitive to the effects of EMP. This is happening because electronics equipment is becoming more sensitive to EMP all the time, and critical infrastructures are becoming increasingly dependent upon electronics.

The next myth is what I call the either-or myth with regard to electromagnetic shielding. The myth is the belief that either an electronic device is resistant to EMP or that it must be enclosed in a military-grade faraday cage. This myth is so bizarre that I heard it for years before I realized that many people actually believe it. Unfortunately, many of the "professionals" working in the EMP field seem to be responsible for spreading this myth through careless comments.

Of course, it is always better to have as much shielding protection as possible, and the means to obtain a high level of electromagnetic shielding are well known. The problem is that a maximally effective shield is often simply neither affordable nor practical.

If a device has a damage threshold of an EMP field of 20,000 volts per meter, then reducing the electromagnetic field by a factor of 3 or 4 will be enough to protect it from known weapons, and shielding it by a factor of 10 will protect it from the super-EMP weapons that are believed by many to exist. A very efficient 80 db. faraday cage would reduce the EMP by a factor of 10,000. In other words, it would reduce a 20,000 volts per meter EMP field to 2 volts per meter. This high level of shielding is necessary for some applications, but not for the average consumer (except for the most critical electronics such as an emergency radio receiver). For many applications, an imperfect shield is quite helpful and may be all that is necessary. (In some cases, though, such as an expensive solar panel system, it makes sense to try to get as close as possible to military grade protection since a functioning solar power system may determine whether you have electricity or not.)


When EMP (or lightning) hits the ground, the currents tend to spread out horizontally. These ground currents can do great damage, especially to underground cables of all kinds. Metal conduits are of little help, and may actually make the situation worse by providing a path for underground currents which can, in turn, induce large voltage spikes on the underground lines inside of the conduits. A large amount of damage has actually occurred due to these underground currents, due to both lightning strikes and nuclear EMP. This is one reason that so much of the information on the internet on grounding and on underground cabling is pure nonsense. Large variations in soil conductivity makes the ground current situation even more complex.

Source: Jerry Emanuelson, B.S.E.E. Futurescience, LLC

  • 1
    $\begingroup$ Hi, @resident_heretic, welcome to WorldBuilding! I want to let you know your answer is likely to be removed, but you shouldn't feel bad about that. Yours is a particularly short answer with no citations. In rare cases, that's ok, but on this question, there are lots of other more detailed answers. As such, a short answer like this will probably be cleaned up as cruft. In the future, try to add a bit more meat to your answers to make them more useful. In this case, links to the science behind vacuum tubes and why they're immune would be a good addition, for example. $\endgroup$ – SRM Feb 1 '17 at 6:26
  • 2
    $\begingroup$ @SRM and resident_heretic: Removing this answer would IMHO substantially decrease it's usefulness. Heretic is correct, but hasn't cited anything; so here:en.wikipedia.org/wiki/… FYI, I write as somebody who has both designed and worked with prototype EMP-hardened avionics. $\endgroup$ – Catalyst Feb 1 '17 at 13:52
  • 3
    $\begingroup$ futurescience.com/emp/emp-protection.html has some points on vacuum tube technology; search that page for "vacuum tube". To state that vacuum tube technology is "completely immune" to an electormagnetic pulse is a very bold statement that should be backed up by clear references substantiating it. @Catalyst Even the Wikipedia article you link to only speaks in terms of "generally much less vulnerable", and even then the claim is tagged "dubious". The next paragraph says "Vacuum tube equipment was damaged in the 1962 testing.", with a genuine-looking citation. $\endgroup$ – a CVn Feb 1 '17 at 14:39
  • $\begingroup$ @catalyst Good. Heretic, you can take the help that Catalyst and Michael have offered and edit your answer to incorporate the new info. This is how comments help improve answers, typically. (This could turn into a good training answer for new users.) $\endgroup$ – SRM Feb 1 '17 at 14:41
  • $\begingroup$ Did you just copy a huge chunk plus a smaller chunk from futurescience.com/emp/EMP-myths.html into your answer? (I edited to use quotation blocks.) $\endgroup$ – a CVn Feb 1 '17 at 17:26

Most of the answers dance around the real issue, I think. If I break "a component" of a computer, when will the computer stop working? Well, obviously it might stop immediately, it might stop after a couple of minutes, or it might not ever fail. It depends on the component, right? So, if instead of me breaking it, it "fries" due to EMP, do you think the possible range of behaviors are any different? As said, the currents created in conductors due to an EMP are immediate and quickly over. The damage may be so severe that the circuit is then shorted or it may have damaged the circuit so that it will soon fail or the damage may be marginal so that, for example, instead of a mean time between failures of 100,000 duty cycles, it's been reduced to 1,000. Or perhaps its been damaged so that the next time it sends current to a particular device, it shorts. Keep in mind that anything which absorbs the EMP (which has to happen to do damage) will also (to some extent) shield other devices in its "shadow". Perhaps you've seen the photographs of Hiroshima, where people's silhouettes have been burned onto walls and roadways? Those same pictures show some buildings still standing when most everything else was flattened. My point here is it's neither all-or-nothing, nor is it gradual. To some extent it will be random (depending on what the device is doing at the instant it happens and where exactly it is (as well as what is around it).) Consider it to be a statistical thing. Let's say cell phones. There will be some distance inside of which everything will surely fry. Beyond that there will be distances with decreasing probability of failure (average failure rate will decrease with distance as 1/r²) until at some distance there's a low probability of damage. A very important qualification: Most nuclear weapon EMP scenarios include the interaction between the initial EMP and the Earth's ionosphere which can distribute the EMP over a very wide area. In that scenario, the damage would be literally hit and miss and hit and miss...


Capacitors could soften the blow of an EMP strike. They store extra charge, devices with lots of strong/solid caps might survive an EMP blast. But they might fail tomorrow due to wear of rapid and high voltage charge. Microcontrollers, if detached from any other circuitry, will probably survive due to the small area they have. Also if the frequency is kind of low, very thin lines in the microcontrollers may not get charged at all (not 100% sure about this part). Due to these conditions a device can fail tomorrow after surviving an EMP strike. But it should be relatively easy to replace capacitors to get it to working condition.

  • $\begingroup$ A capacitor might fix your immediate power problem, but it will not save you from the rest of the EMP-induced problems. Saying it will 'soften the blow' could be misleading. $\endgroup$ – Mast Jan 31 '17 at 20:27
  • $\begingroup$ There's even a big chance of the capacitors blowing up at the same time as the rest of the electronics, rendering them null and void in this proposed solution. $\endgroup$ – Mast Jan 31 '17 at 20:29
  • $\begingroup$ And even if they don't blow up, even a capacitor rated for 5000 V isn't going to do you much good when there's a couple of megavolts or more on the input feed. Note: Most capacitors in everyday circuitry are rated for far less than this, 50 V being a reasonably common value in low-voltage digital electronics, and maybe 500 V in ordinary consumer power supplies which regularly deal with about 250 V AC. $\endgroup$ – a CVn Feb 1 '17 at 14:50
  • $\begingroup$ I am not quite sure if the EMP can reach megavolts on the target. It sure reaches those values at the source. A microcontroller can easily fry when exposed to 24V 0.01A I have seen it happen. In these cases, caps can save the device as the duration of the surge may not be enough to fill the capacitor. $\endgroup$ – Cem Kalyoncu Feb 2 '17 at 5:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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