Generation of rectangular EMP impulses 30V 100MHz

Thread Starter

satcom

Joined Apr 1, 2015
65
I need generate rectangular EMP impulses with amplitude ~30V and frequency 100MHz, what would be the best way to accomplish this?
A block or schematic diagram would be very useful.
 

DickCappels

Joined Aug 21, 2008
10,153
upload_2017-9-4_22-29-17.png

upload_2017-9-4_22-29-33.png

http://www.aholme.co.uk/Avalanche/Avalanche.htm
Also, go to 5:06 on this video:
And the prequel for the above:

You can find slower transistors if you need a slower rise and fall time.

If you want to trigger it from an external source it is easy...
upload_2017-9-4_22-58-58.png

Avalanche action has been observed since the mid-1950's and work was done to improve the understanding of that mode of operation for the next 40 years.
 

Alec_t

Joined Sep 17, 2013
14,280
Generating pulses like that could severely affect VHF radio reception and would be anti-social if not illegal.
 

recklessrog

Joined May 23, 2013
985
Generating pulses like that could severely affect VHF radio reception and would be anti-social if not illegal.
As an amateur radio operator, I completely agree with you, and my post of the chopping block was meant to be a tongue in cheek poke at the idea. I guess it was a little too subtle :)
 

ian field

Joined Oct 27, 2012
6,536
As an amateur radio operator, I completely agree with you, and my post of the chopping block was meant to be a tongue in cheek poke at the idea. I guess it was a little too subtle :)
The accepted use of the term refers to the extremely rapid rise time of the broad spectrum radiation from a nuke detonation.

The American military have been tinkering with a tactical battlefield portable EMP generator - its something along the lines of a capacitor discharge system dumping into a very chunky antenna/coil. At the instant of CD, the coil is collapsed by a shaped charge explosive.

The whole point of EMP is not merely interfere - but to fry all the enemy's electronics. The shaped charge CD inductor is the only non nuke option that comes close.

The ZTX451 avalanche transistor might be more practical for the hobbyist. The pulse characteristics can be tweaked by using various lengths and types of co-ax tuning stub.

The 2N2369 also works in avalanche - but you have to select the best part from a quantity.
 

Thread Starter

satcom

Joined Apr 1, 2015
65
As an amateur radio operator, I completely agree with you, and my post of the chopping block was meant to be a tongue in cheek poke at the idea. I guess it was a little too subtle :)
Alec_t said:
Generating pulses like that could severely affect VHF radio reception and would be anti-social if not illegal.
The use is some experiments and research, study of the effect of EM pulses on certain hardware and the resulting effects.
No any harm will be done to humans or animals. And this will not affect VHF radio communications in any way, because it's not permanent generation, but a very short millisec impulse, and the area will be limited within radius 0.3-0.4m
 
Last edited:

Thread Starter

satcom

Joined Apr 1, 2015
65
View attachment 134296

View attachment 134297

http://www.aholme.co.uk/Avalanche/Avalanche.htm
Also, go to 5:06 on this video:
And the prequel for the above:

You can find slower transistors if you need a slower rise and fall time.

If you want to trigger it from an external source it is easy...
View attachment 134301

Avalanche action has been observed since the mid-1950's and work was done to improve the understanding of that mode of operation for the next 40 years.

That is helpful info, what is amplitude of the output impulse?
 

DickCappels

Joined Aug 21, 2008
10,153
The 2n3904 circuit typically makes a 35 volt pulse. You can change the amplitude by changing component values and the supply voltage.
 

Thread Starter

satcom

Joined Apr 1, 2015
65
The 2n3904 circuit typically makes a 35 volt pulse. You can change the amplitude by changing component values and the supply voltage.
There probably should be spark gap from tungsten rod where spark will be discharged, and also coil from 1/2" flat copper wire loop that will work as EMP emitter.
 
Last edited:

DickCappels

Joined Aug 21, 2008
10,153
You can do it with a spark gap too, but the required voltage will be higher. An avalanching transistor is roughly equivalent to a spark gap.
 

DickCappels

Joined Aug 21, 2008
10,153
This circuit operates from 10V DC which you can get from a benchtop power supply.

upload_2017-9-5_19-8-1.png

How much peak power are you looking for and what do you expect the pulse repetition rate to be?
 

Thread Starter

satcom

Joined Apr 1, 2015
65
Pulse energy approximately 10 Joule with a rep rate up to 20-30 pps. Pulse of power about ~10 megawatt.
The power need be high enough to produce a powerful standing wave of voltage capable access the circuitry, through the shielded enclosure.
.
 

shortbus

Joined Sep 30, 2009
10,045
I have a dumb question on this thread. Why is this one allowed to continue, with even a moderator participating, when all other EMP threads have been closed by said moderators???
 

Thread Starter

satcom

Joined Apr 1, 2015
65
I have a dumb question on this thread. Why is this one allowed to continue, with even a moderator participating, when all other EMP threads have been closed by said moderators???
I have question to you: what do you have against electromagnetic impulses? It's part of electronics working, and its a science.
 

DickCappels

Joined Aug 21, 2008
10,153
And an answer: There is no prohibition against EMP generation, but there are prohibitions against damaging people & property and jamming radio reception. This thread seems ok so far.
 

Janis59

Joined Aug 21, 2017
1,834
RE: pure technics
For the 100 MHz with a small + You have an ultrafast logics series capable to work near the 300 MHz, thus find the trigger and make a half chip in generator regime (demands two fast diodes[1n4148], two resistors[sic <<75k], and two small capacitors [few tens of pF], while another half is dividing frequency by 2 - that to get ultra-symmetrical signal. Circuit is more than well checked and works allways.
But the problem starts about 30V. I think the booster cascade is the correct way, however the common base would be faster some 100 times at least.... why to not use that?

Yet real headaches begins when You want a pico-second needles. As the institute where I am now working was one of the pioneers for excimer laser design in the Baltics, I have one and other places in shelves where residues of that machinery still lays, in-spite that was 50 years ago. It was revealed that there exists only one practical method how to get the thousand ampere few pico-second needle for multiple tens of kilovolts and hundreds of thousands Amperes. Its a patented non-inductive 50kV capacitor weighting some 300 kg and patented form of spark gap of finger size copper half-cones, where the angle of `scissors` of blades produce the time of discharge and gap width means a start and begin voltages. To ignite it there was used a more higher voltage radio-frequency streamer, but for cooling the water inside plus argon outside. Need to say that recharge the capacitor demanded an hours, but that pulse seriously disturbed aether in many tens of km, so we had a special permission of militarists to have those experiments when the impact is minimal, mainly at night-time.
However today, especially if one have no harsh currents, the small power 40 GHz what corresponds to 20 picosec transistor costs just few cents. Other way is to use a tunnel diode what routinely works up to several GHz, hovewer it will look more like a sinusoidal oscillator. If Your aim is Dirac Delta, then nothing is so good as elementary logics, fast logics or ESL logics. Always if swap the logics state but both inputs are slightly shifted one against other, in the output runs one sharp needle. Probably You want to make up the 2-input NOR by the discretes?
By the way, I remember how drastic peaks was generated just by prost thyristor feeding the TV ferrite HV transformer, and that thyristor was far from fast. Maybe You will want to experiment with neon bulb, as it is sure un-ordinary fast, so the speed is determined only by load and capacitor of PS, well wire inductance but that one may make an sendwich-panel.

RE:|| Pulse energy approximately 10 Joule with a rep rate up to 20-30 pps. Pulse of power about ~10 megawatt.||
Wow, N=I*U => 1E7=I^30 => I=300 000 Amp. Need to say its no a joke anymore and no capable for ANY imaginable kind of semiconductor up to now for sure.

RE:||The power need be high enough to produce a powerful standing wave of voltage capable access the circuitry, through the shielded enclosure.||
And here it is some bad coloured smell of evil genius plan behind it all. To have any more advices from me, the author of thread ought to explain more detailed the intention. I am never helping to ignite the nuclear bombs, nor kill the all electronics many kilometre radius, nor people as well (at such currents exists even such danger).
 
Last edited:
Top