Small & Lightweight High Voltage Power Sources: Ideas?

[timelessmidgen]

<< Mentor Note -- thread moved to the Sci-Fi writing forum after starting in the technical forums >>

Hey folks,

I'm interested in the feasibility of providing high DC voltages (~kV) in a physically small and low mass package (~grams). The power source does not actually need to be very energy dense, since not a lot of work is required. If high voltages were not necessary, a small watch battery would be fine. I'm wondering if there's any good way of doing this. Normally for high voltages I would think of a Van de Graaff generator or Wimshurst machine, but I don't think these would scale down very well. Are there any exotic battery types that could do it? Could a normal battery with a good joule thief work? Anybody have a good idea for this?

ETA: Thanks for all the helpful suggestions and also the safety advice. I should mention that this is only a thought experiment, so don't worry I won't be electrocuting myself or x-raying my unwitting neighbors.

 

[berkeman]

Hey folks,

I'm interested in the feasibility of providing high DC voltages (~kV) in a physically small and low mass package (~grams). The power source does not actually need to be very energy dense, since not a lot of work is required. If high voltages were not necessary, a small watch battery would be fine. I'm wondering if there's any good way of doing this. Normally for high voltages I would think of a Van de Graaff generator or Wimshurst machine, but I don't think these would scale down very well. Are there any exotic battery types that could do it? Could a normal battery with a good joule thief work? Anybody have a good idea for this?

There are certainly ways to do it. What is the application? What voltage and at what power? How long does the battery need to last? Is the operation continuous or intermittent? If intermittent, how long is it on each time, and how long is it off?

 

[Nik_2213]

Given your restrictions, a fly-back or blocking oscillator may suit. Think 'Joule Thief'...
Rather than squeeze the bulk and mass of a tertiary winding onto your tiny core, tap the secondary for feedback...
But, DUE CARE PLEASE, no shocking pranks...

 

[timelessmidgen]

There are certainly ways to do it. What is the application? What voltage and at what power? How long does the battery need to last? Is the operation continuous or intermittent? If intermittent, how long is it on each time, and how long is it off?

The application is to accelerate particles (protons and heavier nuclei) to relatively high velocities (but not relativistic, maybe ~0.005c). A proton at 0.005c has kinetic energy of 12 keV, so 12 kV is of order the voltage needed. Please note that I'm not asking about the entire structure necessary for accelerating the particles - this would obviously be more than a few grams. I'm only worrying about the power source here. The operation is semi-continuous. It should be able to sit 'off' without losing power for ~years, then be switched on to supply a continuous voltage differential and run continuously for a similar amount of time. We don't need to push many particles so the power can be quite low, vanishingly small even. The higher the power the better as it would allow us to accelerate more particles per second, but it's not necessary (we can just make do with a lower particle flux). Power output would probably be the least important characteristic, to be maximized only after everything else is acceptable.

 

[timelessmidgen]

Given your restrictions, a fly-back or blocking oscillator may suit. Think 'Joule Thief'...
Rather than squeeze the bulk and mass of a tertiary winding onto your tiny core, tap the secondary for feedback...
But, DUE CARE PLEASE, no shocking pranks...

Haha, ok fine, no shocking pranks! Thanks, I'll take a look into those ideas.

 

[berkeman]

The application is to accelerate particles (protons and heavier nuclei) to relatively high velocities (but not relativistic, maybe ~0.005c). A proton at 0.005c has kinetic energy of 12 keV, so 12 kV is of order the voltage needed.

Can you say a few words about X-ray safety issues and high vacuum issues that may be involved in this project?

 

[phyzguy]

In principle, semiconductor based charge pumps could do this and are quite small. Each individual one would probably be limited to 50V or so, so you'd have to string together a whole lot of them in series, but I see no reason why this couldn't be done. Whatever method you choose, your ultimate limitation is going to be the dielectric strength of the insulators you use. Also, I think any method will involve oscillators that draw power, so the efficiency of your generator can be quite low. So even if you are drawing a small amount of power at 12 kV, you will probably be drawing many times more power from the battery.

 

[timelessmidgen]

Can you say a few words about X-ray safety issues and high vacuum issues that may be involved in this project?

I haven't thought too much about those issues. To be clear, this is only a thought experiment, so you don't need to worry about my irradiating my neighbors ;) What are you referring to regarding high vacuum issues? The fact that at such high voltages I run the risk of frying any dielectric material? That is certainly a concern. The actual bit that would do the particle acceleration would be in a vacuum, but indeed care would need to be taken in the power source itself, and in whatever voltage converter is used to make sure they don't get fried.

 

[Nik_2213]

A dozen kV ? You're into 'fly-back transformer' country for a big 'bottle' TV.

IIRC, the 'bottle' cover-glass was leaded to prevent low-energy X-rays escaping when those accelerated electrons were stopped by the phosphors etc.

May I suggest researching proton penetration and shielding ? You won't be zapping any 'Marshmallow Men', but better to err well on the side of caution, and very carefully document your safety precautions, logic, etc. Then get those formally reviewed by an appropriate inspector...

 

[berkeman]

I haven't thought too much about those issues. To be clear, this is only a thought experiment, so you don't need to worry about my irradiating my neighbors ;) What are you referring to regarding high vacuum issues?

Even for a "thought experiment", this thread won't work for the PF. There are several problems:

** The application is not well-defined, so we are having to guess at what you may be wanting to discuss. That is a waste of electrons and PF members' time, which is not a good theme for a thread.

** Accelerating any charged particle to 12Kev and then hitting a target can generate enough X-ray radiation to be a concern. You say that your flux is very low, but we have no way of knowing that.

** Asking why a high vacuum is needed to accelerate charged particles to 12keV is very problematic for several reasons.

So, for now this thread needs to be closed. If you can send me a message (click on my avatar and Start a Conversation) giving me lots more details on this and on your background, I may be able to help you continue this thread. Thanks.

 

[berkeman]

Well, there's been an interesting turn of events. Turns out the OP is writing a Sci-Fi piece, and is looking for help making a communication device/channel more realistic. LOL. I will re-open the thread and move it to the Sci-Fi forum.

Yes, I can see how the application may have seemed not well defined. I'm writing a short scifi work, and there's an interplanetary civilization that communicates via particle beam. And so that's why I don't have such rigorously defined application parameters - they're flexible.

 

[RPinPA]

I assume you want your particles to be moving close to $c$ to maximize the speed of the communication. That is, you want them to be relativistic. So their kinetic energy needs to be on the order of their rest energy or larger. An electron is 1/2 an MeV, so even for electrons you're going to want millions of volts. For protons with a rest mass/energy of about a GeV, you'd need billions of volts.

I don't see that postulating the ability to do that is a big deal though. What are you worried about in constructing your world?

 

[timelessmidgen]

I assume you want your particles to be moving close to $c$ to maximize the speed of the communication. That is, you want them to be relativistic. So their kinetic energy needs to be on the order of their rest energy or larger. An electron is 1/2 an MeV, so even for electrons you're going to want millions of volts. For protons with a rest mass/energy of about a GeV, you'd need billions of volts.

I don't see that postulating the ability to do that is a big deal though. What are you worried about in constructing your world?

Thanks RPinPA! I prefer not to discuss many aspects of my world building while it's still in an unfinished state. Nonetheless, there is a plot element which requires far below light speed communication ( ~0.005c).

 

[timelessmidgen]

Great! I'm glad that my motivation for knowledge was deemed sufficiently worthy by berkeman to graciously allow this thread to continue. It is so upsetting when folks try to learn more for reasons that aren't deemed worthwhile. To clarify, I'm well aware that a high vacuum is needed for creating high energy particles. As I previously stated, "The actual bit that would do the particle acceleration would be in a vacuum." I'm not sure why berkeman decided it would be useful to misrepresent my statements and quote a badly-out-of context quote...

In any case, many thanks to those who have already responded with helpful suggestions. I believe a fly-back oscillator or general charge pumps will indeed be very useful for advancing the plot, and those suggestions have already given me an excellent place to continue my search. I certainly hope this question was not "a waste of time and electrons" as berkeman suggests. If any of the PF members who replied in our multiple back-and-forth exchanges found it to be a waste of time, you have my apologies. While I will, of course, keep an eye on this thread in case any further suggestions arise, for the moment my curiosity is satisfied and I consider this matter settled to my satisfaction.

 

[Vanadium 50]

and is looking for help making a communication device/channel more realistic

It will help if he explains what aspect he wants more realistic. 12 kV is well within a static electricity shock.

 

[berkeman]

As I previously stated, "The actual bit that would do the particle acceleration would be in a vacuum." I'm not sure why berkeman decided it would be useful to misrepresent my statements and quote a badly-out-of context quote...

I initially read your posts as asking about a small battery powered hand-held accelerator device. Hand-held in my mind precluded use in a vacuum. I probably skimmed your initial posts too quickly.

I certainly hope this question was not "a waste of time and electrons" as berkeman suggests.

Before your clarification to me in our PM exchange that this was for a Sci-Fi story, you had us wasting a lot of time trying to guess what you were doing. Please just open your kimono from your first post in your future threads here. The more information you can post, the better we can help you with our technical expertise. Thanks.

The application is to accelerate particles (protons and heavier nuclei) to relatively high velocities (but not relativistic, maybe ~0.005c). A proton at 0.005c has kinetic energy of 12 keV, so 12 kV is of order the voltage needed.

Why would using charged particles for long-distance communication be a really bad idea? (even in space)

 

[Rive]

Anybody have a good idea for this?

Cut it in two piece. One is the energy source, and one is the high voltage part. The energy source is quite trivial. For high voltage, the typical solution at the mentioned voltages are transformers (easy to scale up to ~2kV in small size if the actual power is low: you can find some examples in CCFL power supplies) with additional voltage multipliers (and those are just in love with high vacuum).

 

[mfb]

I'm writing a short scifi work, and there's an interplanetary civilization that communicates via particle beam.

Not exactly what you asked about, but: Forget it. Or at least accept that you are in the very soft science fiction range. But then you don't have to worry about details like the HV generation.

A 12 keV proton needs a magnetic field below 0.015 nT to have a useful range of 1 million km (that is its curvature radius at this field strength). Most of our solar field has stronger magnetic fields, and 1 million km is not an interplanetary distance.

Let's replace your 12 keV proton source by the LHC and all its pre-accelerator structure. Because we can. At 6.5 TeV and a requirement of 1 billion km curvature radius the maximal field increases to 20 nT. Much better. You still have to measure the field distribution and aim accordingly, but at least you can find a way to make protons arrive somewhere in the target area. Can you use it for communication? Well...
The LHC beams have an excellent focusing. Their normalized emittance is 3.75 μm mrad at a gamma factor of 7000 for bunches with 110 billion particles. Let's give the LHC a system that widens the beam over an unprecedented 1 m while preserving this emittance. Then its angular spread is 3.75 nrad/7000. If it would stay that way then the beam still has a width of about 1 m after traveling 100 million km. Unfortunately all the protons repel each other. At 10¹¹ protons in a bunch and ~0.5 m separation they experience accelerations of the order of 5*10¹⁰ m/s² in their reference frame. The beam quickly spreads out. A numerical simulation suggests a final velocity of just 330 km/s and a spread of 16 km after 100 million kilometers - roughly linear with the distance. Focusing in beam direction will almost certainly be worse but difficult to calculate - let's be optimistic and say 16 km as well. To keep our pulses separate we can only use one per ~30 km, which happens to be a single bunch per revolution for the LHC, limiting our transmission to 10,000 pulses per second. A 100 m² detector will receive 40,000 pulses in a bunch. Add a lot of shielding against low energy particles and the background should be tolerable, although it would still be a challenge to detect the beam. I guess I wouldn't try to transmit more than one bit per pulse (just pulse there or not), but we can allow the beam to get wider, maybe 100 km.

An optimistic 100 billion km curvature radius means the beam still deviates by O(100,000 km) from its straight line. To hit the detector over 100 million km you have to know the magnetic field with ~0.1% uncertainty.

There are a couple more problems with this setup, but I could imagine that you can communicate over interplanetary distances with the LHC beam. With an optimistic rate of 10 kbit/s if the distance is short.

Not really practical.

 

[sophiecentaur]

The problem with trying to send a beam of charged particles is surely their mutual repulsion. Any charged particle beam will diverge eventually. This problem is in addition to the straight problems of using EM waves that behave more reasonably. Did you consider a Neutron Beam?

 

[mfb]

You can't accelerate neutrons and you can't make a well-collimated beam out of them either.

You can neutralize the overall beam by injecting electrons. Over time magnetic fields will make the two charge types diverge and the overall beam will still blow up. A narrower beam also makes aiming more difficult again.

 

[phyzguy]

You can't accelerate neutrons and you can't make a well-collimated beam out of them either.
You can neutralize the overall beam by injecting electrons. Over time magnetic fields will make the two charge types diverge and the overall beam will still blow up. A narrower beam also makes aiming more difficult again.

You can accelerate neutral beams. They are used for heating and re-fueling fusion plasmas. You take a beam of protons (or deuterons in the case of fusion plasmas), accelerate it to some energy, typically 10's-100's of keV for fusion beams, then pass it through a charge exchange cell. The charge exchange cell is just a region of low pressure neutral gas. Some of the accelerated protons will pick up an electron from the gas atoms and you have an accelerated neutral atom. With proper design, I think you can neutralize as high as 60% of the incoming protons. Maybe this would work for the OP's communication device.

 

[mfb]

Well, neutral gas doesn't work at large gamma factors, the protons won't pick up electrons from that and you just get more scattering. Injecting electrons into the beam is something I proposed. Matching the proton gamma factor would need quite a significant accelerator on its own.

 

[sophiecentaur]

You can't accelerate neutrons and you can't make a well-collimated beam out of them either.

You can neutralize the overall beam by injecting electrons. Over time magnetic fields will make the two charge types diverge and the overall beam will still blow up. A narrower beam also makes aiming more difficult again.

My neutron beam suggestion was a bit fatuous, true.
But what about recombination between ions and electrons? Wouldn't that be relevant? Wouldn't the low mass electrons be more affected by the charges on the massive positive ions than the mutual repulsion of the ions. The particles would need to be traveling at the same speed, in any case.

 

[mfb]

I don't see how you would get relevant recombination rates at the energies considered here.

 

[Rive]

The particles would need to be traveling at the same speed, in any case.

Even so any magnetic field will send them on different ways due the difference in mass.
This method is actually a big mess. It would be better to keep the realistic part at the minimum and pick a particle which is less sensitive. With enough technobabble it's still acceptable for sci-fi. Coherent neutrino beams or whatever.

 

[sophiecentaur]

I don't see how you would get relevant recombination rates at the energies considered here.

Why would you say that? Their relative velocities would have to be small in order for the particles to keep each other together. So their relative Energies would need to be around zero.
But I agree with @Rive that it's all too fanciful. And I see not possible advantages over EM signalling - except for the benefits of a whacky Sci Fi scenario.