Sciencemadness Discussion Board - A "new", old idea for a HV pulse generator.

A "new", old idea for a HV pulse generator.

jpsmith123 - 16-6-2006 at 04:45

This gadget, a "spiral pulse generator" or "vector inversion generator", may be of interest to some here.

I haven't tried to build one yet, but because of it's relative simplicity, low cost, compactness, and potential for high voltage, high power operation, it's something I'm thinking about doing one of these days.

Although the article implies new technology, the idea has actually been around for over forty years (see e.g., US Patent 3289015).

Basically, in its simplest form, it is a foil and film type capacitor made from two conducting strips and two insulating strips, rolled on a cylindrical form, thereby creating what are essentially two low impedance, parallel plate type transmission lines - an "active" and a "passive" line.

The "capacitor" is then charged up, like any other capacitor, to a moderately high voltage (Vc). When a switch (e.g., spark gap) subsequently shorts the input terminals together, an EM wave is launched on the active transmission line, ultimately resulting in a so-called "vector inversion" (i.e., the radial electric field vectors associated with the active and passive lines go from series-opposing to series-aiding, or something like that. This results in a triangular shaped high voltage pulse (Vo), and ideally Vo = 2*n*Vc, where n = number of turns (in practice there are some losses which result in lower Vo and depend on construction details, switch characteristics, etc.).

I'm presently trying to figure out how to do an EM simulation of this kind of thing...if i get any interesting results I'll post them here.

12AX7 - 16-6-2006 at 15:39

Never heard of it.

Sounds like a distributed pulse-forming network? That is to say, effectively a transmission line where the impedance increases in order to increase voltage (keeping energy of the pulse constant). Not sure if that actually works...

Tim

IrC - 16-6-2006 at 16:59

Also see 4,140,917 and 4,507,567. Forerunner of the vircator emp weapon.

woelen - 18-6-2006 at 22:48

Jpsmith, at which point can the high voltage be tapped? Is it at the same terminals, where the "capacitor" is charged and shorted again, or is it at the opposite side of the strips? The page you supplied is very brief.

jpsmith123 - 19-6-2006 at 02:42

Woelen, I think I would connect it as shown in figure 7, of US Patent #4507567. (In addition to the patent numbers already mentioned, there are also: 3322967, 4217468, 4717834 and 5567995).

BTW, another example of a "vector inversion" type high voltage pulse generator is the so-called "stacked blumlein" transmission line arrangement, such as shown in the attached paper.

Attachment: Impulse EMI Generator.pdf (367kB)
This file has been downloaded 2453 times

IrC - 19-6-2006 at 14:24

This thread is along the lines of my most favorite subject area in science, glad it was started. While the vircator I mentioned is usually a coaxial device as is mentioned in the below link, I have also seen vircator technology using spiral pulse and vector inversion schemes. Word of warning about the link below is they do tend towards some rather whacko subjects in other areas such as UFO's and the like, but I give the link soley due to the great writeup on the vircator emp devices.

http://www.abovetopsecret.com/pages/ebomb.html

Forgot to add: please all the best minds here get into this subject as I love nothing more than building experimental devices using absolute power in absolutely mad science ways and would love nothing more than to hear what some others here such as Marvin have to say on the subject!

[Edited on 19-6-2006 by IrC]

jpsmith123 - 19-6-2006 at 15:22

Attached is a pretty good paper regarding HV Blumlein Pulsers.

And here is a link where you can d/l a paper (Master's thesis), entitled "Compact, repetitive Marx generator and HPM generation with the Vircator".

Pulsed power is interesting to me in part because electrical breakdown is a time dependent process, and it seems you can therefore build things like multimegavolt particle accelerators in a much more compact size than would be the case for acceleration involving static high voltage or relatively low frequencies.

Attachment: 300KV_150AMP_Blumlein_Pulser.pdf (251kB)
This file has been downloaded 1923 times

12AX7 - 19-6-2006 at 16:15

I've been messing with diode reverse recovery pulsing, but common diodes aren't particularly impressive so far.

Tim

franklyn - 20-6-2006 at 05:23

Quote:

Originally posted by jpsmith123
This gadget, a "spiral pulse generator" or "vector inversion generator",

may be of interest to some here.

The only advantage this capacitor ( what it really is ) has is

that it is a compact single device that can substitute for others

comprised of many dozens of discrete components. Its ability to

deliver pulses of comparatively modest power for a short duty cycle

still leaves it wanting as a supply source. Better means are

available for pulsed power. Samarium Cobalt and newer Neodymium

magnets require fields that can only be achieved by pulsed selonoids.

An excerpt and practical how to is here ( click image to enlarge ) ->

http://img11.imagevenue.com/img.php?loc=loc284&image=057...

Some useful sites _

http://home.earthlink.net/%7Ejimlux/hv/hvmain.htm

http://www.kronjaeger.com/hv

http://www.electricstuff.co.uk

http://www.teslamania.com

Quote:

Originally posted by IrC
This thread is along the lines of my most favorite subject area in science,
glad it was started. While the vircator I mentioned is usually a coaxial
device as is mentioned in the below link, I have also seen vircator
technology using spiral pulse and vector inversion schemes.

That article has been shopped around some, here is a better link ->

http://www.airpower.maxwell.af.mil/airchronicles/kopp/apjemp...

This is an article I saved , a light overview of EMP ->

http://www.keepmyfile.com/download/f4d7b9822994

Directed energy as a concept goes back at least to Archimedes who

as the legend goes , played at it using mirrors to focus sunlight

to burn Roman galleys, he was slain by a roman soldier at the battle

of Syracuse. More recently Gerald Bull the developer of the Supergun

and otherwise conventional artillery was assasinated it is supposed

by the Israeli Mossad because extreme long range gunnery was deemed

a destabilizing threat in the Middle Eastern stalemate. In 1903

Professor Philipov worked to transmit wireless signals generated

with explosions. Czarist Russia was not known for progressive

thinking and it's secret police the Okhrana stopped him - permanently.

The heat ray fictionally described by Heubert George Wells in The

War of the Worlds finally materialized when the laser was invented

in 1960. Masers operating at microwave frequencies had already

been developed by that time. Even chemical lasers have only paltry

efficiencies and require ungainly size to muster significant beam

intensity for any sustained duration. The only types that can cause

meaningful damage explode to produce a single shot such as the

nuclear pumped x-ray laser. Russians Pavlovskii and Saharov who was

disaffected with his work on thermonuclear bombs applied himself

in the late 1950's to the development of what have become known as

magnetic cumulation flux compression generators or explosive

magnetos. Transient field strengths of up to a Tesla squared ( 100

Megagauss ) have been obtained. A well engineered one can capture

perhaps 20 % of the energy of the explosion and channel it as an

electrical surge. That's a lot of juice from a few kilograms of

high explosive, which coupled to an emitter and waveguide, ( essentialy

a maser ) can aim a microwave pulse that is strong enough to fry aircraft

electronics and even knock out the pilot. This is the sort of technology

about which those who know aren't talking and those who talk don't know.

Such is the nature of scuttlebutt.

See -> http://www.sciencemadness.org/lanl1_a/lib-www/la-pubs/002028...

get the same here ->

http://www.fas.org/sgp/othergov/doe/lanl/lib-www/la-pubs/002...

Sciencemadness administration may want to add this one to the LA library

-> http://www.osti.gov/energycitations/servlets/purl/10124948-7...

Every few years there an international symposium published

International Conference on Megagauss Magnetic Field Generation and related Topics

http://megagaussx.physik.hu-berlin.de/workshop/Contributed-A...

.

[Edited on 20-6-2006 by franklyn]

unionised - 20-6-2006 at 10:23

Are we talking about this sort of thing?
http://fusor.net/board/view.php?site=fusor&bn=fusor_hvpo...

jpsmith123 - 20-6-2006 at 18:20

Quote:

The only advantage this capacitor ( what it really is ) has

Well in the original patent the inventor refers to it as a "capacitor", but I would speculate that the idea evolved from a parallel plate blumlein transmission line. I would bet money that the inventor was trying to come up with a way to make a blumlein more compact, and ended up discovering a new device.

Quote:

is that it is a compact single device that can substitute for others comprised of many dozens of discrete components.

And that would be a pretty big advantage, no? One charging resistor vs. many, one spark gap vs. many, thus smaller size, lower cost, higher reliability, etc.

Quote:

Its ability to deliver pulses of comparatively modest power for a short duty cycle still leaves it wanting as a supply source. Better means are available for pulsed power.

Obviously, whether or not "better means are available for pulsed power", depends on your specific requirements.

When it comes to delivering hundreds of joules to a few KJ, at perhaps hundreds of KV to a MV or so, in tens to hundreds of nanoseconds, your choices are limited.

IOW I wouldn't power a toaster with it, but for things like portable flash xray systems, EMP generators, HV testing, certain types of particle accelerators, etc. (essentially many of the things you would use a conventional Marx generator for), it may be the ideal device.

[Edited on 21-6-2006 by jpsmith123]

jpsmith123 - 20-6-2006 at 20:39

Quote:

Originally posted by unionised
Are we talking about this sort of thing?
http://fusor.net/board/view.php?site=fusor&bn=fusor_hvpo...

That's exactly the device! Thanks for that link!

There are some interesting comments there, especially this one:

"It's interesting that the side protrusion is indeed a spark gap, since it only makes a faint clicking during discharge (provided the sparking from the output terminal is suppressed).

I also found out the hard way that the closed outer loop of copper foil actually has a vital function. Removing it and firing the impulse generator caused EMP which destroyed a $15 multimeter nearby, and damaged the display circuit on my regulated DC power supply (supply still works, but does not display the correct voltage). My computer, 16 ft. away, rebooted itself and failed to show anything on screen after Windows loaded. I determined after much nervous sweating that only the AGP graphics card was damaged and replaced it today with a new one. The old card provided an output only in "safe mode", and I think that may actually be VGA from the motherboard itself. Anyway, moral of the story is that I need to restore that copper strap!

When Richard gets back I'd be interested to hear anything you can remember about the internals in the generator you sacked such as the type of dielectric, the type of liquid inside, etc. The electrical circuit for the spiral line generator is pretty simple and it should be possible to homebrew these things.

Dave, the phenolic was from the scrap pile at a local plastics shop and I have a lot of it. I do have 1" solid acrylic rod but no way to drill a clean 3/8" hole down axis through an 10" long piece. I think I'll just have to mount the generator in a custom acrylic tank and fill with oil asI have done various other transformers.

The supply should be very useful for intense pulsed x-ray work or even neutron tubes".

If just removing some shielding did that kind of damage to nearby electronics, I shudder to think what the outcome may have been if the pulsed HV output were directly fed into a properly designed and matched wideband antenna, like a TEM horn antenna.

Twospoons - 21-6-2006 at 14:09

There's a 2Mb tech brief about these things on the NASA website here (scroll down to August 2004). I can't get at it as I'm not a US citizen. Could someone who is get a copy and post it here, please?

jpsmith123 - 21-6-2006 at 16:35

Twospoons,

The only file I could see there is the attached one, which is only about 32 kb.

Attachment: MFS31870-1.pdf (32kB)
This file has been downloaded 1364 times

Twospoons - 21-6-2006 at 17:04

Um. What the hell was I looking at then? Must've had a brain fart. Thanks anyway. All the other info i've spotted that looked worthwhile is locked behind pay sites like iop.org and ieee.org.

Shame - seems like an amusing sort of device to play with.

franklyn - 26-6-2006 at 11:44

Quote:

Originally posted by jpsmith123
Obviously, whether or not "better means are available for pulsed power", depends on your specific requirements.

When it comes to delivering hundreds of joules to a few KJ, at perhaps hundreds of KV to a MV or so, in tens to hundreds of nanoseconds, your choices are limited.

for things like portable flash xray systems, EMP generators, HV testing, certain types of particle accelerators, etc. (essentially many of the things you would use a conventional Marx generator for), it may be the ideal device.

It worked for Franklin didn't it ?

Direct Quote from the following website :

http://repairfaq.ece.drexel.edu/sam/lasercn2.htm

I have an idea for building a Gigajoule laser out of simple materials.
The power source might seem outlandish at first but very viable in the
end. This laser would be based upon the nitrogen laser. It would be
constructed out of a glass or PVC pipe 4 inches x 100 feet evacuated
to 25 Torr filled with nitrogen gas from probably LN or a decomposition
reaction and the electrodes will be made out of band saw material
(because it stands heat real well). The laser would be operated in a
thunderstorm. You already know how to attract lightning using wires
attached to rockets. I estimate its power to be equal to about 14.8
Gigajoules based on the charge and voltage of your typical lightning
bolt and the expected (very precisely computed) efficiency of a
nitrogen laser:

LBC = One standard Lighting Bolt Charge: 10,000 Coulombs.
LBV = One standard Lighting Bolt Voltage: 200,000,000 Volts.
NLE = Nitrogen Laser Efficiency: .00074 (.074 percent).

Then, output energy will be: LBC * LBV * NLE = 14.8 Gigajoules. If
anybody is foolish enough to try this just imagine what you could do
with that burst of power ionize the upper layer of atmosphere and
create a blackout or pump a dye laser and drill yourself a hole in a
mountain. DON'T think about all of the fun you would have with this
and DO think about the real possibility of being fried to a crisp.

______________________________________________________________

A more modest proposal _

http://spt06.chez-alice.fr/00/air.htm

More advanced _

http://spt06.chez-alice.fr/00/lasers.htm

.

[Edited on 26-6-2006 by franklyn]

12AX7 - 26-6-2006 at 14:19

I wish people would remember the essence of a voltage divider.

Just how much voltage does the lightning bolt itself consume? Well, near a gigavolt...
Just how much voltage can you develop across your load? 100V? 1kV? 100kV?! Not even NEAR a hundredth of the total voltage!

The best you can hope is to utilize as much voltage as possible, using it as essentially a pulsed constant current source in the 10kA range. You would of course need a low inductance, high voltage, high energy capacitor bank, and a high voltage input, pulse-hardened inverter to step it down to reasonable grid voltages (240V or some industrial kilovolt line if you have a rather large output).

Say you get 100 lightning strikes in an hour storm (of a size that might pass ten times each year, mind you), and all deliver 5,000C (5kA * 1s, or 10kA * 0.5s, or more or less). To charge the capacitors to 100kV from a single strike, you need Q/V = C = 50mF = 50,000µF (storing E = 0.5 * C * V^2 = 250MJ, or about 70kWh, incidentially), which would probably be the size of a factory, and cost significantly more. (Let's see, at $0.50 (SWAG) for a single 1µF, 600V film capacitor in quantity, you would need series strings of 167 to handle the voltage, and 8.35 million strings paralleled to reach 50mF, for a total of about 1.4 billion pieces, or a cost of 697 million dollars. Yeah, not real pretty. It would make every Tesla Coiler splooge his pants, though.) During the storm, 100 strikes/hour gives 70kWh * 100/hr or 7MW average power, enough to power a city I suppose.

I suppose it is concievable that, somewhere in Florida or Kansas (wherever the lightning capitol is at), such a building, with antenna, could be located. Cheaper capacitors could be used, for instance the parts count could be reduced by 565x by using 1000uF 450V electrolytics (223 per string, 11k strings paralleled, 2.48M count). Price would drop to about $62M (figure maybe $25 each). The output could be filtered carefully (to remove obvious transients) and DC sent to inverters or storage systems (batteries, or other peak demand storage methods), or if immediate demand is high enough, it could be inverted locally.

Tim

franklyn - 26-6-2006 at 21:56

Quote:

Originally posted by 12AX7
During the storm, 100 strikes/hour gives 70kWh * 100/hr or 7MW average power, enough to power a city I suppose.
The output could be filtered carefully (to remove obvious transients) and DC sent to inverters or storage systems (batteries, or other peak demand storage methods), or if immediate demand is high enough, it could be inverted locally.

Calm down no one is suggesting that a strike can be stored.
The essence of this proposal is that it could be harnessed.

P.S.
Related to my reference earlier to explosive magnetos, it is
established that during the Soviet era work by them was
underway to develop a directed energy beam weapon powered by
an explosive magneto fired by a small nuclear explosion.
Reportedly this device also consisted of a condensor. I'll
let you figure the price.

.

unionised - 27-6-2006 at 10:55

I think that the intrinsic problem (never mind the practical ones) is that lightning isn't fast enough to run a N2 laser. So you would need to store it in a cap bank, then fire it into the tube with a sub microsecond time constant.
There's a given charge in a lightning bolt- say its 10000C.
The capacitor has to store that or flash over. You could store it in a 10000F capacitor (preposterous but less so than the ones 12AX7 was talking about; it would only be about $10000). Your cap would end up charged to 1 whole volt and practically all the energy would be wasted.
You could use a smaller cap and a higher voltage. That would store more of the energy. Unfortunately the size and therfore the price, of a capacitor is roughly proportional to the energy its stores.
You might want to trap just part of the energy at some very high voltage - say 100KV. The problem there is you need a big enough cap. If you try to use a smaller high voltage cap it will break down because the clouds will keep pumping the whole 10000C into it. You can't do it unless you have a suitable switch and I don't see that working properly.
BTW, getting electrical power from lightning is a pipe dream.
If you had 10000 strikes a year each delivering 2E12 J then you have a mean power of about 60MW. With the rate I pay for electricity that's roughly £60000 a year.
(2E12 J
2E9 KJ
2E9 KW sec
33E6 KW minutes
555000 KW Hr
at about £0.1/KW Hr).
Near enough to $100000 for the electrical energy.
If you bought the cap bank that 12AX7 was on about, the interest at 5% would be about $3.1M a year.
Never mind the idea that you would need other equipment. Never mind the poor efficiency that you would get, or even the fact that bulk purchases of energy would be cheaper. You simply couldn't ( by a factor of thirty or so) sell the electricity for enough money to finance the kit to harvest it.

12AX7 - 27-6-2006 at 14:19

Yeah, went off on a tangent there, but sometimes it's fun to indulge in totally absurdly scaled engineering projects.

Oh, and with some SWAGs, you can figure such a cap bank, wired loosely (i.e., no specific transmission lines, for instance), would discharge in maybe a few miliseconds (with a few thousand amps of course- a pretty hot spark for just a few feet (100kV) in length; it would be interesting to discharge it through deuterium and/or tritium and observe the effects!), nowhere near the speed required for an N2 laser, even one under optimal conditions (tmax ~ 50ns, IIRC).

Peak power would be in the TW range, but only because energy is in the near-GW range. High capacity transmission lines are the way to go (I wonder what kind of peak power a dielectric ceramic TL could deliver?).

Oh, and here's a SWAG for you- figure a ceramic disk capacitor of 2cm diameter (1cm radius) and capacity 0.1uF, charges to 1kV before breaking over. Assume the dielectric fails in the center of the disk (riiiight..), and the discharge is essentially instantaneous (meh...ought to be). The distances are pretty short, so all the energy (50mJ) discharges in about the time the wave takes to travel from the center to the edges and back, or 2cm round trip. Guesstimate velocity factor as 0.5 (that is, 1.5 x 10^8 m/s = 1.5 x 10^10 cm/s). Pulse length, then, is 0.13 nanoseconds, or 133 picoseconds. Pretty damn fast. Peak power is then roughly 50mJ / 133ps = 0.375 GW. Not bad for a cheezy bit of silvered ceramic.

Tim

[Edited on 6-27-2006 by 12AX7]

unionised - 28-6-2006 at 11:54

Don't forget to take account of exactly god-knows-how-much inductance. That might limit the current/power.
Also I'm pretty sure that the high dielectric constant of teh ceramicmeans that the speed of the discharge will be lower. (Might be the Clausius-Mossotti equation, but I'm not sure)

[Edited on 28-6-2006 by unionised]

12AX7 - 28-6-2006 at 13:58

Yeah, every TL I've measured by TDR (Time Domain Reflectometry) comes to v = 0.67c, those TLs landing in the 30 to 100 ohm range. Nothing particularly exotic though. As I mentioned, I figured half (v = 0.5c), but even if it's a tenth (unlikely, even with as much capacitance as there is), the pulse is still a nanosecond, quite fast. Inductance isn't an issue because the disc is 1cm radius (although a nonzero seperation) and there are no ferromagnetics around. Remember this is a discharge through the axis, from the ceramic breaking down. Not a discharge through the leads, which would likely have a shorter period (acting more like a lumped-constant system of perhaps 10 or 20nH and the capacitor, thus discharging (1/4 wave) in a much slower 50ns or so).

Tim

JohnWW - 29-6-2006 at 00:40

I have an important book called "Electrostatics & Its Applications", A.D. Moore (editor), Wiley 1973. In view of its importance, including in chemical processing, I think I will have to scan it to image files (about 500 pages including some grayscale photos, and notes in margins by me and a previous owner) and upload it sometime. It has a section on static electricity in the atmosphere, chapter 17. According to this, as regards the possibility of getting electric power out of atmospheric static electricity, there is a weak electric field of the order of 100 V/m at Earth's surface, negative on Earth's surface and positive in the atmosphere (mostly in the ionosphere), and which integrated over the entire surface indicates a total charge of about 1,000,000 coulombs. The lower atmosphere serves as a dielectric separating Earth's surface and the ionosphere, and if the potential gradients typically measured through the lower atmosphere are summed, the total potential difference between the Earth's surface and ionosphere is of the order of 300 KV. This implies fine-weather capacitance C = Q/V = 1,000,000/300KV = 3.3 farads.
Measurements, according to this book, show that the fine-weather conduction current through the atmosphere is of the order of picoamperes per square meters. When summed over Earth's entire surface, this conduction current is of the order of 1,000 amps. Given the potential difference of 300 KV, this implies DC electric power available (if the whole Earth's surface could be tapped) of 300 MW. While this is a useful amount of power in many small countries, it is a mere "drop in the ocean" compared to total world electric power demand.

[Edited on 29-6-2006 by JohnWW]

franklyn - 29-6-2006 at 07:58

Quote:

Originally posted by JohnWW
as regards the possibility of getting electric power out of atmospheric static electricity,
there is a weak electric field of the order of 100 V/m at Earth's surface, negative on
Earth's surface and positive in the atmosphere

This topic has been lightly covered and experimentally demonstrated by Oleg Jefimenko

http://www.as.wvu.edu/coll03/phys/www/OJ/JEFIMENK.HTM

A short summary appears in his book Electrostatic Motors 1973

download a pdf of those five pages here _

http://www.keepmyfile.com/download/77714c1335901

As a useful power source this ranks about equal to barometric pressure differential

still used in self winding clocks.

.

[Edited on 18-1-2007 by franklyn]

franklyn - 17-4-2010 at 12:47

Update
Oleg Jefimenko died just last year.
http://en.wikipedia.org/wiki/Oleg_D._Jefimenko
The article on his atmospheric electrostatic motor can be viewed here _
http://books.google.com/books?id=kwAAAAAAMBAJ&pg=PA80&am...

.

Sedit - 17-4-2010 at 13:38

Im not sure I fully understand what being said since I read it really quick and I have to be somewhere soon but the idea sound simular to how a TEA laser works. In the laser setup there is a common foil plate under an insulator and there is two plates seperated on the top with a small gap with a resistor connecting the two to ensure they are both charged. When the spark gap is triggered it causes a rapid change in current across the two plates and in an instant an extremely high voltage cascade is observed across the gap between the two. If the impedence is kept low and the discharge is able to occure in under 2.5 nano seconds then it will stimulate the N2 in the air causing lasing to occure in the UV range.

I may be way off base like I said im in a rush but is this a simular circuit?

Power from the ether

franklyn - 27-3-2012 at 11:39

www.youtube.com/watch?v=hNjmjMNqLCo
www.pennystockwizard.com/stock-report/sefe-01
US7855476
US8102078
US8102082
US8102083
http://www.marketwatch.com/investing/stock/SEFE
http://www.google.com/finance?q=OTCBB%3ASEFE

Related posts in this thread
http://www.sciencemadness.org/talk/viewthread.php?tid=6032#p...
and the two following
http://www.sciencemadness.org/talk/viewthread.php?tid=6032#p...
http://www.sciencemadness.org/talk/viewthread.php?tid=6032#p...

http://en.wikipedia.org/wiki/Sky_voltage
Oleg Jefimenko
See pg 80
http://books.google.com/books?id=kwAAAAAAMBAJ&pg=PA80&am...
Same article in pdf here _
www.only1egg-productions.org%2FAltSci%2FElectrostaticMotors%2FImages%2FAmazing_Motor_That_Draws_Power_From_The_Air_PS_Apr_1971.PDF
Another article see pg 612
http://books.google.com/books?id=qLDJukH3hPgC&pg=PA612&a...
Electrostatic Motors
www.scribd.com/doc/78393253/Oleg-Jefimenko-Electrostatic-Motors-FULL-BOOK-75p

- C L I C K IMAGE to VIEW MOVIE CLIP -