Homepage › Forums › Innovative Confinement Concepts (ICC) and others › My Fusion Sphere Idea
I have had a fusion idea for many years. I did not discuss it in open forums because I could not find a civil one where I felt I could introduce the idea and not be attacked by the conventional fusion crowd with their turf to protect.
I could have introduced this idea to some of the cold fusion groups but they have a turf of their own to defend and frankly most seemed so completely vested in their own projects they didn’t have the time or emotional bandwidth to explore ideas that were much different from their own.
And quite frankly the only other forums I could find to discuss this stuff were populated by hyper-dominant loons with frequent free energy proposals or with other people such as myself that would not be capable of bringing sufficient scientific rigor to the conversation.
But now I am here, and this group seems to be an entirely different animal.
So here goes the idea…
1) Build a large sphere, the larger the sphere the better (maybe even tens or hundreds of feet in diameter?)
2) Cover the sphere with a symmetric layer of material capable of making bulk expansion (piezoelectric, piezomagnetic, piezophotonic, etc)
3) Suspend the sphere in a manner designed to minimize the distortions due to gravity
4) Suspension for system is lightly coupled such that it will not over dampen mechanical vibrations
(Best case scenario the suspension is floating suspended in the micro-gravity of space)
5) Fill the sphere with a (solid, liquid, gas, or plasma)
Note: The type of material is “to be determined” but my
intuition says that a gas or a plasma will be preferred
although all phases hold some interesting possibilities
At least for the center section of the sphere the substance used
must contain some form of fusion fuel
6) The sphere should be built in a manner that the mechanical Q factor for
vibration is high, the higher the better as long as non radial vibrations
are minimized.
Having a high Q allows conservation of energy. Instead of
dissipating after a single “shot” it the energy pumped into the
system is used over and over again in the form of mechanical
vibration. The higher the Q the less energy it will take to drive
the system.
7) The sphere is broken into many independently controlled segments (think soccer ball segments for instance)
These sections would be slaved to a master oscillator and will move
in unison at the same phase & frequency (at the natural
resonant frequency of the sphere.) All over the sphere there would
also be local active servo control units that will ensure that all
segments move at the same frequency and the same amplitude.
8) As you drive the sphere you begin to develop a radial standing wave with a single maximal
pressure node right in the center.
9) During a pressure maximum the converging wave will induce phase changes to the material at the center.
You would have at stages solid, liquid, gas, and plasma as the pressure goes up and down and as material
near the center is alternately pushed violently inward and violently outward.
10) The assumption is that for a sufficiently large sphere and a sufficiently high pressure wave amplitude you could periodically obtain conditions that match the lawson criterion.
11) Some people may point out fairly that such a system could generate sufficient pressure at the very center and that it might match Lawson’s but the time that it spent at high enough pressure and temperature will be too low and so the number of fusion events would be quite small.
12) To them I would say that we don’t really need to know or even care about what is happening right at the center!
13) I believe the area of interest will be a varying diameter concentric two sphere volume near the center where the (density/temperature/time/n) distribution within that varying volume can support a reasonable amount of fusion.
14) Thus we do not need to know or really care what is happening at the mysterious asymptotic center.
15) What I have been thinking about lately is what happens to various species of materials (solids, liquids, metals, insulators, dense gasses, diffuse gases, dense plasmas, diffuse plasmas, asymmetrical plasma with nuclei at the center of sphere, asymmetrical plasma with nuclei at the edge of sphere) when they are subjected to these sorts of violent pure radial mode oscillations?
Note that multi-layer spheres may be beneficial to provide neutron slowing or to absorb photons and re-radiate energy back towards the center of the system.
Note that if you have a heterogeneous layer system you may really only need a small amount of fuel.
15) I know about the problem of conductive cooling so I suspect that a gas or plasma may be the best material at the center
16) Nevertheless if you imagine a large sphere you can see that the resonant frequency of the system decrease and the amount of system kinetic energy you are concentrating during each pressure wave at the center will increase, and the time that material spends at that higher pressure will increase as well. Of course that means more time to cool too! So there is a trade off.
17) That is another interesting factor of this system, I believe that increasing the radius and mass for the system will raise the energy passing through the center in a cubic manner.
So has this idea been hashed to death somewhere? Or is it basically new?
If by some chance this idea does pan out and some of you get rich off it, or if you win the Nobel or something like that…
You owe me big time! And everyone here will know it. So just know I do expect to collect with a little Tahitian island
(one tall enough to deal with global warming) and with a few pretty cabana girls to peel my grapes.
And all the rest of you will be welcome visitors, but you will have to peel your own grapes!
-Steve
đź™‚
Hi Steve! Welcome to the forums. It’s good to hear new ideas, and this is a safe place to introduce them. Your idea sounds a lot like sonofusion, and a little like the project at General Fusion in Canada. Look those up, and then we’ll have more to discuss on your idea. Again, welcome.
Hi, and welcome to the forum!
Its good to think about novel ways of achieving the Lawson criteria. The trouble is the density, temperature & confinement time product is a very large number. In a spherical compressive model as you propose, the slight asymmetry would introduce Rayleigh-Talyor like instabilities way before the compressive wave could reach the centre. So peak density & temperature would be far lower than needed.
If it were that simple, the spherical imposive method of detonating plutonium bombs, would also work for fusion. To ignite a fusion burn wave at the centre of a spherically compressed volume you need the compression, heating and ignition to occur faster than any of the instabilities have time to develop. The instability grows exponentially in time, but is proportional to the size of the initial perturbation. Or in other words your large sphere being driven at the outside would have generate a spherical wavefront perfect to less than one atom width.
By increasing the size of the sphere to increase the ratio of driving oscillation to proposed peak oscillation in the centre, the wave takes longer to reach the centre and so reduces the tollerence on the sphereical accuracy. The exponential growth of the instablilty will always be faster then the cubic growth in geometric scaling.
On the other hand if by some miracle you could get it to work – having that much fusionable fuel in a sphere would make a very big bomb. The few mm size of the pellets, in a 10m radius vacuum chamber used in devices like NIF to a large extent due to the limited load the walls could take when it detonates. If as in your case the rest of the volume was full of ‘fuel’ then you would blow up the city!
“the slight asymmetry would introduce Rayleigh-Talyor like instabilities way before the compressive wave could reach the centre. So peak density & temperature would be far lower than needed.”
James, I did quite a bit of reading before on Rayleigh-Talyor instabilities (and I just did a quick review today on them) and whatever might be the limiting factor here I
really don’t think Rayleigh-Taylor would be the problem.
http://en.wikipedia.org/wiki/Rayleighâ€“Taylor_instability
I quote…
“Rayleighâ€“Taylor instability,is an instability of an interface between two fluids of different densities, which occurs when the lighter fluid is pushing the heavier “fluid.[1]”
You see in the case of a converging spherical shock wave the higher density material takes the form of an ever denser shell that is approaching a much less dense center. It is dense pushing sparse, not sparse pushing dense, hence it is not a problem (or at least it’s not a RT problem.)
Picture the situation some few pico-seconds before the shock-wave reaches the center…
Here I have made a formula as a sketch that explains the basic sort of density distribution I expect with respect to radius
This is a decent on line function plotter… http://rechneronline.de/function-graphs/
10*(1/(x-10)^2)*sin(3x)+3
With x-axis set to (-10 to 0) and with y-axis set to (-25 to 25)
I am assuming the following for the purposes of the illustration
1) Initial density of the material is “3”
Trace = density as a function of of distance from the center at r=0 some point in time after the system has already reached a steady oscillatory state.
So this is a snapshot of the approaching pressure wave/high density wave after the steady state oscillation has been established so the density at the center will rapidly swing back and forth between a very sparse and cold minimum and a very hot and dense maximum.
Since there really isn’t much material in the center there really isn’t much of anything to push back against the approaching high density shock wave.
You might say that if there isn’t anything there there will not be anything to fuse?
I concede that.
But I say the area that really should interest us is not the exact center of the sphere but rather that small bit of material on the leading edge of the traveling wave, like a surfer riding riding in on a giant wave that is forced through a narrowing inlet between cliffs and who is rising rapidly (and is speeding up dramatically) because the wave he is riding has no place to go but up.
You mention that just increasing the sphere’s diameter will not help presumably because it would make the system resonant frequency lower and thus there would be a slower change in pressure at the center allowing what instabilities there are more time to do their damage.
As I mentioned (I believe the wave velocity at the center is not a constant related to the diameter of the sphere) but instead is a variable and it effectively limited only to the amount of stored momentum in the system. So a small sphere with a very high mechanical Q could match a large one with a low Q but a large one with a large Q would be the most interesting case.
Using the case of sonoluminescence, you would have a converging wave in a liquid with a bubble of vapor in the center.
In my example you could have anything (solid, liquid, gas, plasma) but for ease just imagine it is a gas for the moment and then consider plasmas (both charged and
neutral ones next.)
Such a system would see a wave that would have made it a very thin gas at one instant just before the next approaching wave reaches the center. It’s like the
sonoluminescence case but I think much more stable because you get rid of that annoying dancing bubble of aspherical gas floating in the center trying to rise up
due to it’s buoyancy.
“jamesr” said…
“If it were that simple, the spherical implosive method of detonating plutonium bombs, would also work for fusion. To ignite a fusion burn wave at the centre of a
spherically compressed volume you need the compression, heating and ignition to occur faster than any of the instabilities have time to develop. The instability grows
exponentially in time, but is proportional to the size of the initial perturbation. Or in other words your large sphere being driven at the outside would have generate a
spherical wave front perfect to less than one atom width.”
James, I don’t think even the best H-Bombs achieved that level of precision. I don’t think you make a case that it is necessary.
“By increasing the size of the sphere to increase the ratio of driving oscillation to proposed peak oscillation in the centre, the wave takes longer to reach the centre and so reduces the tolerance on the spherical accuracy. The exponential growth of the instability will always be faster then the cubic growth in geometric scaling.”
As addressed earlier, I believe the bounding factor on center speed is not system diameter but rather system momentum. Spherical asymmetries are a real concern
but I think manageable ones. Trick is to drive the wall evenly!
“On the other hand if by some miracle you could get it to work – having that much fusionable fuel in a sphere would make a very big bomb.”
Now there you are touching on something I ponder upon with some real apprehension!
There probably is a maximum size beyond which this system might become a bomb which is why I propose that experts study it rather than simply having amateurs
build it. It might not be safe!!!
Beyond that I know it might seem ludicrous, but I am worried that it might even be possible that such a system (made suitably large and effective) could start producing little molecular sized chunks of neutronium or even black holes after each compression and that they might fall to the center of the earth and there begin
to coalescent into something of real concern.
Thanks – Steve
OK so maybe I didn’t explain myself very clearly… So here’s another way of looking at it:
If your oscillations in density & temperature grow in amplitude, at some point the radial gradient from a region of high to low density will grow beyond the point where the simple linear equations cease to fit, and non-linear interactions become important.
Generally there are only two classes of instability involving gradients: Rayleigh-Taylor; where the driving force is parallel to the gradient, and Kelvin-Helmholtz; where the driving force is perpendicular to the gradient (anything else is just a mix of the two). So although it is not exactly the same scenario as the classical Rayleigh-Taylor description, characteristics such as equation of the growth rate is the same.
If you want the peak density/temperature to be in the range for fusion then the gradient away from that peak must be very steep – too steep to be stable.
4
jamesr wrote: OK so maybe I didn’t explain myself very clearly… So here’s another way of looking at it:
If your oscillations in density & temperature grow in amplitude, at some point the radial gradient from a region of high to low density will grow beyond the point where the simple linear equations cease to fit, and non-linear interactions become important.
Generally there are only two classes of instability involving gradients: Rayleigh-Taylor; where the driving force is parallel to the gradient, and Kelvin-Helmholtz; where the driving force is perpendicular to the gradient (anything else is just a mix of the two). So although it is not exactly the same scenario as the classical Rayleigh-Taylor description, characteristics such as equation of the growth rate is the same.
If you want the peak density/temperature to be in the range for fusion then the gradient away from that peak must be very steep – too steep to be stable.
James, I am really not trying to be difficult (I am trying to defend how this could work, not quite the same thing as being difficult I hope) but I looked at the Kelvin-Helmholtz equation as well and found this…
http://en.wikipedia.org/wiki/Kelvinâ€“Helmholtz_instability
“two fluids in parallel motion with different velocities and densities will yield an interface that is unstable for all speeds.”
In my case the motion appears to be all perpendicular, not parallel.
So the turbulence source if there is any would appear to me to a special case formula for RT turbulence used in cases of highly converging highly dense fluids pushing extremely light fluids at extreme speeds.
All the examples of Kelvin-Helmholtz I have seen apply to two fluids in parallel motion (like the initial parallel laminar flows of two fluids in contact say oil and water, one dense, and the other less so ) and with one of the layers traveling faster than the other.
Assuming forces involved are large enough that surface tension is insignificant, in such a case the flow is unstable for all speeds.
So things don’t look good (if there is a parallel flow in this system) and assuming that there is sufficient time available for that turbulence to propagate?
The thing is I still don’t see where there is a parallel flow of two different fluids with differing densities at different speeds in this system?
All I see is a system (at the encroaching wave front) where you have a very dense substance traveling inward (with all points along the imploding shell having the same velocity vector r and same density and all pointed directly towards the center)
And at the interface of contact on the inner wall you have (an inner sphere of extremely sparse gas that due to the recently passed adiabatic expansion in a near vacuum) is by definition very cold.
I think it safe to make the assumption of adiabatic expansion because of the short time scale of the expansion…
http://en.wikipedia.org/wiki/Adiabatic_process
“Such a process can occur if the container of the system has thermally-insulated walls or the process happens in an extremely short time, so that there is no opportunity for significant heat exchange[1].”
So whatever velocity vector that stuff in the center might have before it is effectively zero just prior to being struck by the incoming shock wave.
Since that wave is encroaching from all directions inward, the material inside the shell when struck will either become adhered to the wall in the manner of a classical inelastic collision or become part of much smaller mass that will be projected even faster inward to form a leading shock wave as some portion of the light mass inside it will be driven ahead in a m1*v1=m2*v2 manner of an elastic collision.
So that implies there will be not just one incoming shock wave for very long but instead there would be many onion skin layers of (at turns converging and diverging) lower energy but faster concentric spheres of waves that will lead the main incoming shock wave in and follow it out.
All of these concentric spherical waves arriving and leaving in extremely short succession.
These layers of masses in excited states and with various radii would create an interesting time variant trap for any light emitted at the center so when the primary mass arrives there will likely already be an intense flash of photons trapped there at the center perhaps even gamma radiation? Just the sort of thing you would want to trigger a fusion event.
I do see points of time when this system would be unconditionally unstable that is the time when the main wave has just passed center and is back on it’s way out. At that point (at some point some distance from the center) you would have a universally diverging vector. But it doesn’t matter then because hopefully by then you already got your tiny little H-Bomb to go off. And then you get to do it over and over again.
I think the existing form of fusion reactor that most closely resembles the system I am proposing is not the controversial method of sonoluminescence but rather the Fransworth Fusor and we know that device works!
It’s just that the Fusor isn’t very efficient due to the collisions of the ions with the grids used to create the radial oscillations in the plasma. BTW, we know those grids are not even remotely close to a true sphere! Yet they seem to work.
So I would think that if we restricted the consideration of the system to a plasma instead of a gas that the potential for this system should seem obvious, it just isn’t all that different from a system we already know works. Well with the single exception that the driving method I propose completely eliminates the primary loss mechanism of the Fusor.
Thank You Again for your time – Steve
Maybe someone else can help me to see the source of turbulence James talked about?
I say if you want a symmetric compression, drive the system symmetrically, most systems don’t really try.
Have a look at these simulations:
http://www.astro.virginia.edu/VITA/ATHENA/implosion.html
http://www.astro.virginia.edu/VITA/ATHENA/blast.html
(click on the right hand images to play the animations)
They are not exactly the situation you describe (given the simulations are in a rectangular box), but they do show how interactions between initially smooth compression waves lead to instabilities and turbulent mixing.
What I think would happen in your scenario is that the slight imperfections in the waves travelling towards and away from the centre would interact and become unstable.
Further, this instability and subsequent mixing would reduce the peak denisty capable to well below fusion requirements.
Hey, I was searching terms related to a similar idea and came across this post.
Steve, I think you are on to something here. I had a similar concept (though smaller in scale) that I have been toying with for the past ten years. It is nice to see someone else exploring the concept of a spherical compression wave collapsing towards the centroid.
A couple points though…
A smaller sphere will have a higher resonant frequency, which means a higher pressure wave amplitude from the initial displacement.
Using the Ideal Gas Law (specifically the molecular adaptation) you can calculate the temperature from isentropic compression at the peak pressure as the wave travels towards the centroid. When that temperature passes the threshold for fusion, that is the active region of the device. The confinement time can be taken as the amount of time a particle with that much kinetic energy will take to cross the active region. Density can be cross-checked at the same time with the IGL.
One thing to remember is that we cannot just assume that the macroscopic image of a wavefront is accurate, we need to consider the brownian motion of the particles themselves, traveling in a direction generally towards the centroid, but still bouncing against each other as the wave collapses.
The actual resonant frequency of the sphere will have to be experimentally determined, as the conditions at the pressure wave vary so greatly as the temperature and density rises that calculating it precisely will be difficult.
I did the relevant math at one time to calculate the active region for a small sphere with a pure hydrogen fuel gas, but due to a computer error I lost the final version. One of these days I will revisit the draft I have and try to re-locate the mathematic errors I know are there and re-write the whole thing.
Maybe by discussing our 2 concepts and their similarities and differences we can get a better mental handle on the physics involved.
Merlyn_x
Magical Engineer and Technical Metaphysicist
While it is always good to test ideas with a simple model like the ideal gas law, to get rough ballpark figures. You should also be aware of under what conditions they are a good approximation and when they will break-down, (and they do break down at any fusion capable densities/temperature/timescales) .
In plasma physics it is common to order each term in the equations in a characteristic time and length scale (or characteristic speed). So for example the conservation of energy gives you the rate of change of pressure, which in turn gives you the characteristic speed(s) of a pressure fluctuation, which for a normal gas is just the speed of sound. In a plasma however there is not just one solution to the equation, there are many.
If the local fluid velocities are small in comparison to the speed of sound then the adiabatic term is the only one you need. However, as the temperature & pressure gradients increase the velocities of the individual ions/electrons become further from Maxwellian and you need to take into account the other terms such as the heat flux.
At that point, if you want to stay with the fluid model, you can either just keep adding on the extra equations of the higher order moments of the velocity distribution function with a closure at a higher order. Or move to a full kinetic model and use a PIC (particle in cell) approach.
If you’re interested in looking at the maths the full fluid equations are described here http://farside.ph.utexas.edu/teaching/plasma/lectures/node32.html
Nifty.
I’ve been needing to get more in depth with the plasma physics.
Now if I can just find some free time to parse it into my understanding.