You are an individual with competence in electronics, whose family member has been abducted. She is trapped in an underground server room via dead man's trigger connected to a substantial explosive device. Disabling the explosive is substantially more important than the integrity of the servers (that is, it doesn't matter if you break everything in the room, so long as the building doesn't explode, and neither do you or your family member).

In this situation, could you somehow generate an EMP capable of disabling the electronics in the bomb (it is electronically activated, not chemical) using only material at hand, in the server room? Ideally, you would do this very quickly, without spending much time having to manufacture something.

It seems to me that it might be possible to generate an ESD (electrostatic discharge) that would result in an EMP, or perhaps rapidly switching pulses in an electrical circuit to create an EMP. What I'm unsure about is whether such an EMP could be generated from materials in a server room (it's a fairly big server room), and whether that EMP would be strong enough to disable an explosive (without being so strong that it causes more problems than it solves).

  • 8
    $\begingroup$ As a cartoon villain myself, I'm all in favor of blowing up buildings and innocent people. However, there are simpler and less-risky ways to keep the hero busy while I carry out my nefarious plan. For example, I try to do many of my activities while the hero is on vacation out-of-state. Also, deadman switches are notoriously difficult to rig properly, and require an expert minion on-site to arm the explosive device without blowing themselves up. Finally, I have found that threatening the hero's loved ones tends to make them carry a grudge long past the current operation; rarely worthwhile. $\endgroup$
    – user535733
    Commented Oct 6, 2020 at 15:18
  • 9
    $\begingroup$ An EMP "disables" a circuit by generating a large surge of electricity in a circuit, a short-circuit bascially. Isn't this risky in that it could just trigger the bomb? $\endgroup$
    – kutschkem
    Commented Oct 7, 2020 at 13:38
  • 1
    $\begingroup$ I'll counter the nay-sayers here, and mention that it's plausible that a small EMP targeted at a microcontroller which is running the dead man's trigger might suffice to disable the circuit without setting off the bomb. If it's obvious that a physical attack (cutting wires) would be deadly, I might consider the EMP as a last resort... $\endgroup$
    – kgutwin
    Commented Oct 7, 2020 at 19:39
  • 5
    $\begingroup$ This answer depends very precisely on the construction of the bomb. Wouldn’t it be better to rephrase the question: “what types of bombs can be disabled by an EMP?” $\endgroup$
    – cms
    Commented Oct 7, 2020 at 20:24
  • 2
    $\begingroup$ I concur with the answer below that says EMP is simply not the way to go to disable a bomb. If your character is highly competent with electronics, he will know that any digitally actuated trigger could be set off just raising the line from low to high (which an EMP would be very likely to do), or dropping it from high to low (which destroying the digital switch would absolutely do, no matter what). Google "digital solenoid" to see what I mean. I think there are lots of ways this scenario could play out without using EMP per se, so I'd encourage you to keep at it and consider your [continued] $\endgroup$
    – automaton
    Commented Oct 7, 2020 at 22:23

4 Answers 4


Is the EMP necessary? There seems to be much easier ways to do this. Not to mention, the pulse from an EMP might induce a current on the dead-man's switch's wires that looks like the switch was triggered. The bomb could actually detonate before the EMP did enough damage to disable it. Besides, if your bomb was vulnerable to EMP, the bad guy would have spent a few bucks on a cheap Faraday cage.

A dead-man's switch detonates the bomb when the state of a switch changes. All you really have to do is to prevent the state of the switch from changing (or at least prevent the bomb from knowing that the switch changed).

Find the wires going between the switch and the bomb. Locate a non-contact voltage meter. They're common in places where work is done on electrical wiring, so they wouldn't be out of place in a server room. Clamp it around one of the wires between the switch and the bomb and see if it measures any electrical current flowing through the wire. Electrical current means that the switch is currently "closed" and the bomb will detonate when the switch opens. No current means the switch is currently "open" and the bomb will detonate when the switch closes.

The easy case is when the switch is currently open. Prevent it from ever closing by cutting one of the switch's wires and insulating the loose ends in electrical tape. The circuit is now permanently open, regardless of what happens to the switch.

If the switch is currently closed things get a bit more complicated. You'll want to bypass the switch and short the two switch wires together so that the circuit stays closed regardless of what happens to the switch. Carefully strip some insulation off of the middle of the two switch wires without cutting the wires (any sort of blade will work, but a lighter or soldering iron can melt off insulation without damaging the wire). Find a short section of scrap wire and strip all the insulation off (network cable works well for this) until all you have left is the bare copper. Press the two exposed sections of switch wire against each other, and wrap them up tightly with the scrap wire. Wrap with electrical tape to hold everything in place. You've now short-circuited the trigger mechanism, and the switch will always appear to be closed.

I can hear you asking "but what if it's a wireless trigger mechanism?" A server room is an absolute nightmare for wireless signals. Servers are full of circuits that switch at very high speeds, meaning they generate unreasonable amounts of RF noise (I've seen some that generate enough RF noise to be classified as a ham radio). Many datacenter products incorporate some sort of RF shielding into the chassis to block enough radiated noise to meet industry standards. Start packing running equipment around the bomb as tightly as you can. Remove every chassis and flat piece of sheet metal. Got WiFi routers? Use them too, and set them to broadcast on all channels at max power. Generate enough RF noise to completely drown out the trigger signal.

  • 2
    $\begingroup$ Nope, there's no reason for there to be appreciable current flow in the wire. There simply is voltage. $\endgroup$ Commented Oct 8, 2020 at 0:55


If you want to try this at home: you can just buy a kit that's so much safer!

If you want a passable action sequence that will put out a tiny bit of EMP but you can make in under 60 seconds - that's at the end.

If you actually want to disarm a bomb be aware depending on design it may blow the detonator as it fries the circuitry. I'm assuming your character has inspected the bomb and realised that an EMP is the right choice.

You can build an EMP gun from parts in a server power supply and wall socket cable.

This inside view of a power supply:
enter image description here

Those two blue capacitors tend to be quite high capacity and low inductance. Each server will have 2 power supplies minimum, and each power supply will have several large capacitors. Read the voltage on the side. Collect at least 250V worth, ideally 500V worth.

Grab some insulated, single core wire. Speaker cable is ideal, an extension cord will do. Cut both plugs off, strip the wires, and join all the wires together (positive / negative). Twist into a ring about 20cm in diamter, and then use duct tape or something to hold the ring.

Youre coil will look like this: enter image description here

Note both conductors are joined to make one big effective cable that loops.

Now build your capacitor chain. Read the voltage on the side, and wire them in series up to over 250 volts. (You may need to find a soldering iron in the server room somehow). If you have more than 250V, build two sets, and solder the strings of capaictors together in parallel.

Wrap with duct tape to stop them wiggling and breaking the soldering.

To build a charger, you need a bridge rectifier (or diodes) and more capacitors. The circuitry is very common; it can look like this:

enter image description here

Or you can get it prepackaged, it comes in dozens of forms:
enter image description here
Any of these will do so long as they're rated over 240V.

Grab a power cord and strip the connector off, leaving the wall plug and some cable.

Charging and use:

  • To charge - Connect bridge rectifier DC side to your capacitors (matching positive for positive). Then connect your stripped wall socket cable to the AC side of the bridge rectifier.

  • The plug it in to charge. Will take 1-2 seconds to charge tops.

  • Turn power off at wall socket, disconnect bridge rectifier from capacitor bank.

  • To fire, connect the capacitor bank to the antenna. It will fire through the ring one way (and you're probably going to fire it backwards first time.)

Too slow to assemble? There's a faster, but last realistic, way.

There is another way to do it - its much lower power and less likely to work, but you could make it under a minute.

  • Use scissors on an extension cord to make a loop as per above, ends stripped, with positive and negatives joined together (so they function as a single conductor).

  • Power supplies in servers are hot swappable. Rip a power supply out while its live (ignore the warnings saying don't do this).

  • Disconnect the power from the wall (leaving the capacitors charged).

  • Connect your antenna to the 12V and ground output of the power supply.

  • This uses the charge of the capacitors still inside the power supply. If you only need a tiny amount of EMP, this may be enough.

  • 2
    $\begingroup$ Why so careful? Just leave the power supply connected to the wall when shorting it. Many datacenters also have 48v DC in the wall - you could potentially short that for far higher currents. $\endgroup$ Commented Oct 7, 2020 at 5:53
  • 2
    $\begingroup$ @DanielVestøl Because that would make; in order; a weak electromagnetic, then a coil-shaped heater, and then a nice flaming coil. (Or it'll blow a fuse / circuit breaker). Current flowing through a coil isn't enough to make EMP - you need a rapid pulse in current. $\endgroup$
    – Ash
    Commented Oct 8, 2020 at 17:49
  • $\begingroup$ you will really struggle to create a greater current than from shorting a high power powersupply. We don't care about fuses, they will trip too late anyways. DC-DC power supplies have capacitors on the input already, essentially speeding up the harvesting process you proposed while allowing for using capacitors from dozens of PSUs at once. You are however correct in that there is no way this EMP would disturb anything except RF sensitive electronics, and probably not defuse the bomb. $\endgroup$ Commented Oct 9, 2020 at 21:13

An EMP affects only electronic circuits.

Though ESD is an example of EMP, it won't have any disabling effect on an explosive, because the explosive acts only following chemical paths. On the contrary, generating sparks or currents close to an explosive is usually regarded as a very poor idea.

Since the activating circuit in a bomb is basically tasked of generating a pulse to initiate the explosion, I think the EMP is not the way to go to defuse the bomb.



The sort of components you would need to create a quick DIY EMP bomb is not equipment that would be needed/expected in a server room, but it's plausible that you might still find it in a server room. You will require:

  1. Lead acid battery bank. Most server rooms have uninterruptible power supplies, and most modern UPS systems use lithium-ion battery technology. If you try to pull the amount of current we're going to need through a LiOn system, you will disable the bomb but simultaneously set the batteries on fire, which is sub-optimal. A lead-acid system (effectively a large array of car batteries) is a sometimes-seen alternative, and has the advantage of being able to deliver a very large current if (im)properly connected - you can quite easily weld a spanner across the terminals of a large car battery if you short them, for instance. You need an array of at least 20 or so deep cycle traction batteries.
  2. A decent amount of chunky cable. None of the ordinary wire you're likely to find in a server room will be sufficient, but conveniently the hero finds a spool of steel-wire-armoured cable from an external cable run. Sod the copper cores inside, we'll be using the armour itself as a conductor.
  3. A Coil. The exact properties of the EMP pulse depend mostly on the size and shape of the coil, and there's a lot of maths and physics involved in getting it exactly right. Fortunately, your hero will just happen to find a spiral chair or similarly shaped radiator, or a roll of flexible copper pipe and a non-metallic cylinder to wrap it around.
  4. Actuator. This is a posh term for a switch, but one which can handle a very high current and close in a very short time. A not-particularly-plausible example would be a bear trap or other spring-loaded object with jaws. A mousetrap will not be large enough.
  5. Insulation. A large pile of plastic sheeting would be sufficient, or some polystyrene blocks.

The technique is relatively simple:

  1. Reconfigure the battery bank. It will probably be arranged with batteries in series stacks of 4-8 12V cells, for a working voltage of 48V-96V, arranged in parallel for greater capacity. Quickly disconnect each stack and reconnect in fewer, taller stacks. Ideally you would reach a working voltage of at least 1kV. With appropriate tools (ie a spanner) this should take no more than a few minutes. Risk of electrocution death rises exponentially as you go through this.
  2. Connect a length of SWA cable to the negative terminal. 'Connect' is a relative term mainly involving mashing the exposed metal onto the battery terminal and trying to keep it in place. If your hero can find a reel of solder or thinner wire, this could be used to 'tie' the cable in place; the discharge current will probably weld it on.
  3. Connect the other end of the negative cable to one end of the coil. Connect another piece of SWA between the other end of the coil and one side of the actuator, and then from the other side of the actuator back to the positive battery terminal. Try to minimise the area enclosed by these cables (ie run them next to each other as far as possible), but beware that shorts from the battery positive to ground can easily start fires or cause serious injuries at this point, so discretion may be the better part of valour. All components should be insulated from the ground on piles of plastic sheeting.
  4. Point the end of the coil at the bomb. The closer the better, but since the bomb is presumably earthed and can't be moved, stay well clear. It would be immensely preferable for the hero not to have to hold the coil, but if necessary, they can do so if they hold only one turn of the coil, and are standing on an insulating surface. If they're wearing any metallic items, remove them back before step 1. If they have a pacemaker, this is probably not the escape attempt for them.
  5. Trigger the actuator, preferably without touching it, certainly without touching it if also holding the coil.

The effect is essentially that the battery bank will short through the wires and coil, and produce a huge current which in turn produces a single huge magnetic field pulse as the coil acts as an inductor. In the process the wires will probably weld themselves together, and you'll be lucky not to start a fire. But you will probably also blow out every piece of electronics in the room, hopefully including the bomb.

  • $\begingroup$ Are you sure it's the end of the coil that should be pointed at the bomb? The magnetic field goes through the middle of the coil (where it's strongest) and then loops back around the outside. $\endgroup$ Commented Oct 7, 2020 at 9:55
  • $\begingroup$ Certainly putting the bomb inside the coil would be optimal, although probably impractical. AFAIK though EMP damage is done by the pulse - ie the change in magnetic field strength inducing eddy currents on the rising edge. I'm not sure whether the end or equator of the coil would be best, logic dictates that it must be one or the other. $\endgroup$
    – Stephen
    Commented Oct 7, 2020 at 10:27
  • $\begingroup$ Since the field is strongest in the middle, logically when it goes from 0 to maximum strength and back, the change is also strongest in the middle. $\endgroup$ Commented Oct 7, 2020 at 10:40

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .