Building an EMP Generator

[jk96]

Any one have circuits or ideas to build a Battery operated EMP Generator
I saw this generators on youtube lighting Fluorescent lamp and destroying calculators
I tried google and there is not much i can find about Coils ect..

[WhiteKnight]

That stuff depends on what kind of equipment you are going to influence. Can you describe your task in more details?

[Gallymimu]

Sounds like this battery powered EMP that can light bulbs and destroy calculators is probably fake.  Do you have a link to the video?

[solutions]

It's just a high frequency stepup transformer and antenna...baby Tesla coil.

Oscillator:high_turns_flyback:charge_pump_voltage_doubler:helical-antenna

See attached schematic

[optikon]

I think the circuits & effects are real. Ionizing the gas in a flourescent bulb doesnt require much energy, as would be expected from something running from 9 volt batteries. You can take that bulb and stand under some outdoor high tension lines and get the same glowing effect. Coupling through the air is about the worst way to transfer energy. Yawn.

[Gallymimu]

oh, I didn't realize we were holding the thing right up against the lightbulb.  I had this vision of a handheld device lighting up a small office full of fluorescent and trashing calculators on all the desks.  I feel dumb

[optikon]

EMP from Nuclear Bomb blast = damage to nearby electronics that are not shielded and fairly nearby.. minor damage too. This was studied back in the nuclear hayday.
EMP from a 9 volt battery = joke is on you!.. oh, maybe a glowing light bulb.

[solutions]

It is a pulse, and it is electromagnetic....but it's not "EMP"
 
The toy here, aka "Slot machine Jammer", runs from 27V, not 9V

There was an article in Pop Mechanics a few years ago that was about a non-nuke EMP device. I'm off to go request it in the right section of this site and have them drop it in this thread.

[Magnox]

The article (not the whole magazine) is here:

http://prepbug.com/files/EMP/E-Bombs And Terrorists.pdf

There's also a document here, which shows a good description of FCGs:

aoc-gardenstate.org/lunch_flyers/april%20/EMP_AOC2april.pptx

Flux Compression-type EMP generators work on a simple principle of putting a high current though a conductor which is coiled around another (the coil former) conductor. As the current peaks, an explosive inside the coil former is detonated at one end.

This expands the former and causes it to short the coiled conductor, with the short moving along as the explosive expansion propagates. The magnetic field is compressed and moved along with it, and the inductance of the coil is decreased leading to an increase in coil current and hence more magnetic field.

Added:

I vaguely recall seeing a device demonstrated that was used by the police to disable a moving car. It was a small box with a long cable attached, and was placed so that the car would drive over the box. Whatever was in the box disabled the engine management computer, stopping the car. I don't recall if they claimed it was an actual EMP device, probably (much more likely) it was just a powerful transmitter like the "Slot machine Jammer". I have never seen them since, probably because those spiky snakes they use are cheaper and more reliable.

[WhiteKnight]

If talking about fun - then your choice is Tesla gun http://www.gizmag.com/battery-operated-tesla-gun/22570/
Quite simple design.

Actually, any pulse generator connected to antenna (or even cable) will radiate EMI noise. If talk about serious approach - it is necessary to know the target's response at different bandwidth (say, 100 MHz, 500MHz, 1GHZ, etc). After that it is necessary to choose a generator/antenna capable of producing the required electrical field.

[Magnox]

Personally, I favour a microwave oven magnetron fed into a metal plate lens antenna. Good for short-range directional nuking! It won't run from a 9V PP3 battery though...

Even better, how about a bigger version:

http://typnet.net/Articles/WeedKiller.pdf

Just put a big metal plate lens in front of that.

[solutions]

Actually, any pulse generator connected to antenna (or even cable) will radiate EMI noise. If talk about serious approach - it is necessary to know the target's response at different bandwidth (say, 100 MHz, 500MHz, 1GHZ, etc). After that it is necessary to choose a generator/antenna capable of producing the required electrical field.

EMP is not noise. It's an extremely high field strength, wideband signal, and has rise/fall times that are very fast.



I think you are talking about a jammer, from the sounds of it. EMP not only jams, it toasts (breakfast puns intended) 

[WhiteKnight]

Solutions - Depends on power. Absolutely correct about rise / fall times, they are indeed important. Once you increase peak power to several GW and reprate to kHz you also shall damage parts. UWB signals may cause damage in much higher bandwidth than just 100 MHz. A lot of cool nonlinear effects appear.

So, it is necessary to define "extremely high field strength" depending on PRF/bandwidth

[jpsmith123]

One approach would be something called a "switched oscillator"; which is basically just an LC tank circuit connected to an antenna.

The LC tank circuit can be made from lumped elements, or it can be a piece of transmission line. It can be made in a balanced or an unbalanced configuration.

The main idea is to charge the "capacitor" to a high voltage and then a spark gap switch fires and completes the circuit. A switched oscillator can generate very high peak power, but the bandwith isn't very much.

http://www.ursi.org/proceedings/procGA08/papers/E01p7.pdf

http://www.ursi.org/proceedings/procGA11/ursi/E03-5.pdf

There is also a similar concept for something called a "frozen wave generator" which also uses stored charge and spark gaps to generate high power RF pulses.

I remember seeing some patents for some fairly simple high power spark gap based designs.

######################################################

Edit: Here's an interesting paper:

https://www.ncjrs.gov/pdffiles1/nij/grants/236756.pdf

[FTL]

I guess it all depends on scale. I can see a solution with a dry day, nylon carpet and rubber soled shoes. Rub you feet on the carpet for a bit, and then touch the flourscent light or calculator. I'm thinking you really want a bit more range though.

I used this technique a few years ago when my neighbour has a 4 year old laptop from his job and they wouldn't give him a new one as long as the old one still worked. All it took was a cold winter day with no humidy and a couple of minutes on the carpet. It did take a few tries, and I suggested he do a backup first.

[solutions]

This is not EMP



(though she likely is an IMP)

The electrostatics you've described are not analogous to electromagnetics...there's no B-field for starters.

Then, as I said, there's the risetime of 5-10 NANOseconds....it wreaks as much havoc as the e-field strength itself and makes protection a real challenge because everything is a huge impedance at those risetimes (like a lightning rod, conductor, and ground stake would be)



At 50,000 V/m and 5 nanoseconds is how many volts/second of slew rate (dV/dt) is that?

Current is 1/C dv/dt...so how many amps do you get for anything that's, say, a nF (make it 5nF for easy math) in capacitance between two conductors seeing that e-field of 50kV/m? Like a long power transmission line spaced 1m (for easy math)

bzzzt!

[Parmin]

If access are possible, maybe a stungun rewired to a lower voltage transformer would do it.

[2N5109]

Do a search on frozen wave generator.  I have generated megawatts peak power with a 10kv spark gap and short length of transmission line. 

--2N5109

[Gallymimu]

Do a search on frozen wave generator.  I have generated megawatts peak power with a 10kv spark gap and short length of transmission line. 

--2N5109

related patent... pretty cool
http://www.google.com/patents/US4491842

[solutions]

This one is a bit more dramatic (it references 4491842).

If you don't stand a chance of blowing your fingers or hand off, it can't be as much fun to build or observe, can it?

http://www.google.com/patents/US7218016

LoL....here's the FFT (look at the Y axis units  )

[Gallymimu]

This one is a bit more dramatic (it references 4491842).

If you don't stand a chance of blowing your fingers or hand off, it can't be as much fun to build or observe, can it?

http://www.google.com/patents/US7218016

LoL....here's the FFT (look at the Y axis units  )

NICE!  Now wait for someone to complain that it isn't broad spectrum enough

[2N5109]

...
Current is 1/C dv/dt...so how many amps do you get for anything that's, say, a nF (make it 5nF for easy math) in capacitance between two conductors seeing that e-field of 50kV/m? Like a long power transmission line spaced 1m (for easy math)

bzzzt!

Isn't current i = dq/dt = d/dt (Cv) = C dv/dt not 1/C ??

[solutions]

yup....brain fart...which then messes up my analysis...no PhD for me. Time to start a new thesis 

[2N5109]

with a FWG, whatever voltage you charge up to is discharged into a Zo ohm load when the spark gap fires, where Zo can be 50 ohms.  So for a 10kV gap (and 50 ohm trans. line) you have 10000/50 or 200 amps.  Power is Vp^2/(2R) in the megawatts peak 10e8/100.  May be a factor of 2 in there somewhere and there are losses with the gap.  You get a monocycle square wave out, rep rate depends on how fast you charge up line and de-ionize gap. 

[govtcontact]

When you are dealing with fast/short pulses, especially HV, the load from a dc perspective is usually 0 Ohms, but you want to make the 'active portion' of the circuit have the highest impedance.  Like stated above, at these frequencies, everything looks like a large impedance, so keep your transmission lines a thin/flat and WIDE as possible to lower your inductance or your pulse gets drug out in time.

The last project I was working on involved charging ~2.7nF to various voltages (15kV to 78kV), connected to a triggered spark-gap next to a single turn loop around a transformer core to ground.  Around 35kV on up, I had no problem knocking routers out at work, destroying multiple timing/trigger generators (the newer FPGA based units) and even a o-scope from the EMP.  (Yes, power was all filtered, units all shielded, cables triple shielded, and most of the equipment a min of 50ft away [two rooms]).  Granted, the discharge was only 70 - 120ns wide and at ~ 3 kA, its not hard to make an EMP

[HINT:  Think in the RF realm]