Marx generator and EMP / Electromagnetic Radiation / Forums

Forums

4hv.org :: Forums :: Electromagnetic Radiation

« Previous topic | Next topic »

Marx generator and EMP

1 2 next

Move Thread

LAN_403

empdude

Wed Apr 21 2010, 05:05PM

Registered Member #2732 Joined: Thu Mar 11 2010, 07:48PM
Location:
Posts: 4

Hi, we are trying to use the marx generator we built in order to create an EMP. We tried dischargind the marx into a solenoid but we haven't seen any movement with our galvanometer. Is it possible that the pulse is too fast to change anything or its something more like the power of our marx that is not big enough?

However, when our marx is not grounded, it seems to be a small iduced current in the galvanometer. Why?

By the way, the marx makes sparks that are a little bigger than 1cm.

We included pictures of our project in the MarxGen.zip file.

We know its a lot of questions at the same time.
Thank you for your answers
]marxgen.zip[/file]

Kipmans

Wed Apr 21 2010, 05:47PM

Registered Member #91 Joined: Thu Feb 09 2006, 03:03PM
Location: The Netherlands
Posts: 45

wrote ...

Is it possible that the pulse is too fast to change anything or its something more like the power of our marx that is not big enough?

Perhaps. Maybe you can try making a simplified model of the (discharging) Marx and the galvanometer setup to see what kind of pulses you are dealing with.

wrote ...

However, when our marx is not grounded, it seems to be a small iduced current in the galvanometer. Why?

Probably because the discharge from the marx creates a strong and fast changing electromagnetic field which, in turn, induces a current in your galvanometer.

Proud Mary

Wed Apr 21 2010, 06:44PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

A Marx generator of whatever size only works at its best when its output impedance matches the input impedance of the load.

The rate at which the magnetic field can build in a solenoid is governed by its time constant t = L/R. which will make a solenoid a very bad mismatch with the usually quite low output impedance of a Marx generator.

Marx derived EMP has been radiated by so-called 'impulse radiating antennas' with moderate efficiency using a bow-tie type of radiator in UWB pulse experiments, but even here, big termination resistors to earth from both ends of the sheet-metal 'bow tie' have been found necessary to stop flash-over.

Google the names Mayes and Altgilbers together with Marx to learn more!

empdude

Thu Apr 22 2010, 12:09AM

Registered Member #2732 Joined: Thu Mar 11 2010, 07:48PM
Location:
Posts: 4

Thanks a lot for the replies.

So the best antenna seems to be a TEM horn. However, it seems quite complicated to build. Since we are a little short in time because it is a school project, could you recommend the simplest antenna that would do a reasonable job?

Also, in one of the documents you recommended, it said that the impedance of the Marx is Z = sqrt(Leq/Ceq). How can we know the inductance of the marx? And for the capacitance, do we calculate it when the capacitors are in parallel, or in series?

Is there a way to measure the strength of the EMP effect other than with a galvanometer?

Thank you, and sorry for our insufficient understanding of the subject.

Proud Mary

Thu Apr 22 2010, 08:33AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

At a basic level, a Marx works by charging up a number of capacitors in parallel through a charging resistor. When the voltage across the caps has risen to a point sufficient to bridge the lowest spark gap, then that gaps fires connecting the first two caps in series. The second gap now has double voltage across it, and fires next, the third gap three times the voltage and so on.

Now put out of your mind everything to do with spark gaps, and think in terms of switches, and you will have a better idea of how your contraption works. Nowadays, the designer of smaller Marx impuse generators where a very fast rise time rather than a very high voltage is what is looked for, may decide to use solid state devices such as avalanche transistors, IGBTs, or thyristors as switches, rather than the less reliable and ever troublesome gaps.

Now, consider all your gaps as switches, and you'll see that when all switches are closed, the total capacitance (not including stray capacitance which is very important in some advanced Marx designs) is simply 1/C = 1/C1 + 1/C2 + 1/C3 etc as you know. It is this figure, the total erected capacitance, that you should use when calculating the output impedance and the energy storage of your Marx.

As for inductance, even a short, straight piece of wire has it (but not much!) so the biggest source of inductance in your Marx is likely to be in your capacitors, especially if they are of the wound variety, rather than the flat plate type. Look up the inductance of your caps in their data sheet, and if you can't find one for them, look to the data sheet of something similar.

Carey, Mayes and Altgilbers are the gurus of contemporary Marx design, and I suggest you try to learn what you can from

MARX GENERATOR DESIGN AND PERFORMANCE by W. J. Carey and J. R. Mayes, which you'll find here:

And now for the practical bit: your best chance of radiating enough energy to produce intriguing effects elsewhere is to go back in time to the very earliest days of radio transmission.

Here is one of the very earliest transmitters of electromagnetic energy, invented by the German physicist Heinrich Hertz (1857 â€“ 1894). Discharge your Marx across the first spark gap, marked sender in the diagram.

Afterthought added later: Whilst the term EMP means literally, an Electro-Magnetic Pulse, it is most often used to mean the very special case of EMP associated with a nuclear explosion, such as the code-named Starfish Prime (Google this!) test of 1962, which interfered with the Earth's magnetosphere all the way round the planet.

empdude

Thu Apr 22 2010, 11:56PM

Registered Member #2732 Joined: Thu Mar 11 2010, 07:48PM
Location:
Posts: 4

So the antenna is basically a non grounded park gap parallel with a coil? We are experiencing difficulties in understanding how to connect the antenna to the marx generator. Do we just but the antenna at the end of the condensator chain? Also, which parts of the montage should be grounded?

We found no information about the inductance of our capacitors. We will probably talk briefly about nuclear EMP in the theory portion of our report.

Thanks a lot

Proud Mary

Fri Apr 23 2010, 09:03AM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

empdude wrote ...

So the antenna is basically a non grounded park gap parallel with a coil? We are experiencing difficulties in understanding how to connect the antenna to the marx generator. Do we just but the antenna at the end of the condensator chain? Also, which parts of the montage should be grounded?

We found no information about the inductance of our capacitors. We will probably talk briefly about nuclear EMP in the theory portion of our report.

Thanks a lot

Ground/Earth one side of the spark gap, and connect the output terminal of your Marx to the other. These designs of Hertz are what are called lumped constants tuned circuits. The spheres or plates at the end of each line are like the top loading of a transmitter antenna. The metal plates are certainly easier to make than the spheres. You could use aluminium foil on a simple frame, for example.

You may find it will work better without grounding one side of the spark gap, and simply connect your Marx output to one of the terminals. An oscillatory voltage can then flow back and forth across the gap radiating like mad.

Be very careful around the high voltage in this contraption.

wylie

Sat May 01 2010, 03:52PM

Registered Member #882 Joined: Sat Jul 07 2007, 04:32AM
Location:
Posts: 103

A Marx generator of whatever size only works at its best when its output impedance matches the input impedance of the load.

Well, that's pretty much true of any circuit, right?

the designer of smaller Marx impuse generators where a very fast rise time rather than a very high voltage is what is looked for,may decide to use solid state devices

That's when i start imagining the day that some sort of transistor comes into play with 10kv+ blocking and less than 0.1ohm RdsON. In hefty, brick packages. C'mon nanotech, you can do it!

Proud Mary

Sat May 01 2010, 05:59PM

Registered Member #543 Joined: Tue Feb 20 2007, 04:26PM
Location: UK
Posts: 4992

wylie wrote ...

A Marx generator of whatever size only works at its best when its output impedance matches the input impedance of the load.

Well, that's pretty much true of any circuit, right?

It may seem self-evident to you, but if you look through this site, and other similar amateur sites, you will not find much discussion of Marx output impedance matching.

Mattski

Sat May 01 2010, 07:24PM

Registered Member #1792 Joined: Fri Oct 31 2008, 08:12PM
Location: University of California
Posts: 527

wylie wrote ...

A Marx generator of whatever size only works at its best when its output impedance matches the input impedance of the load.

Well, that's pretty much true of any circuit, right?

It's actually not universally true. Impedance matching allows the maximum amount of power to be transferred from a linear source into a linear load. When you are transmitting something in a transmission line where high frequency effects need to be accounted for, reflections will occur at any interface with an impedance mismatch, which can be a good thing or a bad thing.

However by necessity half of the total power is dissipated in the source impedance and half in the load. The voltage on the load is half of the open circuit voltage. If your load resistance is much higher than source resistance then you can have a very efficient transfer of power, even though it will not be near maximum transfer. Imagine plugging in an impedance matched load to regular AC mains. It would mean that half of all generated power is lost in transmission to the load. Having a higher load resistance will maximizes the voltage on the load, which can be useful for example in amplifiers, where a higher load resistance can give a larger voltage gain.

But in the case of an EMP you want to couple the maximum amount of power into the radiated EM pulse because that will give you the highest field intensities. The ratio of electric to magnetic field is fixed by the free space impedance, and their product gives the power, so to increase either field intensity you must increase both, and you can only do so by increasing power.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob