Tri Alpha Gets $50 million

[DeltaV]

Since the p-B11 fusion product alphas are expected to make about 1000 passes before exiting the magrid, coaxing most of them out through one weakened coil seems feasible.

It's my understanding that it's the hydrogen and boron-11 fusion fuel ions and NOT the He4 fusion product ions that are the ions that make the 1,000 semi-circular pases into and out of the core before thermalizing.

I'm fairly sure that Rick said it was the alphas that made ~1000 passes (with sufficient B field strength), as mentioned here:
viewtopic.php?p=43710&highlight=#43710
but I don't have a link to his original comment.
Anyone?

[MSimon]

I don't have a link either. But my memory is tolerably good: alphas.

Think about it: Polywell needs about 100,000 passes of electrons (probably a similar number for ions) to make the device deliver net energy. So from that alone the odds that "1,000" passes is about fuel is rather low.

[D Tibbets]

It is indeed the fusion alphas traveling at several million electron volts of kinetic energy that escape the magnetic confinement in ~ 1000 passes (based on what Nebel said). They do this because there is no effective potential well containment as they exceed the energy of the potential well by several orders of magnitude. I'm not certain how long long a fuel ion would last without a potential well. The electron last perhaps 10,000 passes (and perhaps 100,000 or more passes with recirculation).
But with a potential well, a non upscattered fuel ion would theoretically make ~ infinity passes without any chance of escape. The magnetic confinement is completely immaterial from the fuel ions perspective. Of course, in reality there are compromises and inefficiencies. Even with this, the escape of ions (mostly upscattered ions) is slow enough and with low enough residual energies that the energy cost is completely dwarfed by electron energy loss concerns. What is important is how fast these escaped ions (and any neutrals) fill up the space outside the magrid, as this leads to crippling arcing.

Dan Tibbets

[MSimon]

kurt9 said... at http://powerandcontrol.blogspot.com/201 ... -hope.html

I know that Tri-alpha hired a control system guy (PLC-based controllers) last fall. This alone tells me that they are "bending metal" as PLC-based controllers are used for high vacuum process equipment, which a plasma fusion system most certainly is.

[hanelyp]

The magnetic confinement is completely immaterial from the fuel ions perspective.

Perhaps not entirely. An up-scattered ion confined by the magnetic field would have many passes back through the plasma where it would have a chance to anneal.

[vahid]

What is this "Bussard/Ligon" effort cr@&?

By measure of effort, MSimon and a number of others here probably have me beat. My most recent contribution is a tee-shirt design and some new slides for my NASFic presentation in a couple of weeks.

NK and Iran getting in to the game are political moves, but government funding is a political problem. I'm more intrigued by the apparent interest at ONR, which previously had at least some individuals showing resistance. I expect changing attitudes there is more likely the result of real scientific evidence.

Of the two sources of pressure, politics probably trumps science, but maybe science benefits.

Hi Dear Tom Ligon.
When will ur experimental results of WB-8 be present????
Do You Hope 100MW polywell reactor still???
[/b]

[TheRadicalModerate]

I don't have a link either. But my memory is tolerably good: alphas.

Think about it: Polywell needs about 100,000 passes of electrons (probably a similar number for ions) to make the device deliver net energy. So from that alone the odds that "1,000" passes is about fuel is rather low.

This doesn't make any sense to me. Either the magrid B-field strength is low enough so that the high-energy alphas have an infinite gyroradius and they get through immediately, or it's high enough to reflect them, causing them to thermalize the fuel ions and electrons, and you wind up with a hot-ish, non-reactive mess. The highest probability of escape that fusion products will have is on their initial trip out from the reaction area. After that, it's all downhill.

What am I missing?

[WizWom]

I don't have a link either. But my memory is tolerably good: alphas.

Think about it: Polywell needs about 100,000 passes of electrons (probably a similar number for ions) to make the device deliver net energy. So from that alone the odds that "1,000" passes is about fuel is rather low.

This doesn't make any sense to me. Either the magrid B-field strength is low enough so that the high-energy alphas have an infinite gyroradius and they get through immediately, or it's high enough to reflect them, causing them to thermalize the fuel ions and electrons, and you wind up with a hot-ish, non-reactive mess. The highest probability of escape that fusion products will have is on their initial trip out from the reaction area. After that, it's all downhill.

What am I missing?

The fact is that the alpha particles, having a speed of 7.7x10^6 m/s for the slower ones, will not have ANY time to be affected by the magrid; it will go straight out without being turned significantly, reaching a 3m vacuum chamber wall in about 500 ns.

The whole "1000 passes" thing is nonsense. There's a lot of people on the board talking nonsense.

[D Tibbets]

WisWom, how do you justify this claim. In a non cusp system the ions will be contained till they can work their way through the magnetic field by several mechanisms (assuming the gyroradius is less than the distance to the magnet casing or first wall). In a cusp system, this 'baseline' leakage is dwarfed by charged particles exiting through cusps. Even without any shrinkage of cusp openings an ion may bounce around 5-10 times before it escapes through a cusp. I beleive this is called mirror confinement. With some closing of the cusp throats this number may increase to ~ 50-60 passes. This is called Cusp confinement by Bussard. I think this has been demonstrated in many different systems like Penning traps. Wiffleball traping at Beta=1 is an extension of cusp confinement and is claimed to be in the thousands.

Now enters the potential well set up by the excess energetic electrons. This potential well confines the ions for claimed times much above the Wiffleball traping factor. In an ideal system the ions would be contained indefinatly. But if the ion is upscattered it might gain sufficient energy to escape the potential well. So long as it is not subsequently downscattered (in speed) it will eventually find one of the cusps and escape. The fusion alphas (or other fusion charged particles are energetic enough that the potential well will only be perhaps 1% to 20% as strong as the KE of the fusion ion. Also, since these particles are at ~ several million eV, their coulomb crossection is much smaller, so they are much less likely to scatter with the slower fuel ions, so escape time is mostly dependant on the Wiffleball trapping (or cusp trapping if you doubt the validity of the Wiffleball.
Also, keep in mind that this is not a magnatized plasma, the charged particles spend most of their time flying around in the non magnetic central region of the machine. The ions and to a lesser extent the electrons(?) generally do not gyrate around magnetic field lines except when they are passing through a cusp (or rebounding off the Wiffleball border where they complete ~ 1/2 gyroradius orbit before rentering the magnatic free region and continueing on their merry way.

The same process applies to electrons, they are confined dependant on the relative size of the cusp holes as a percentage of the total surface area of the magnetic border.

If you hate the Polywell claimes, ignore it and consider gridded fusors. It is generally accepted that within the sloppy conditions due to the presence of neutrals, With a central cathode, any ions created or dropped into the field will oscillate back an forth until upscattering permits it to reach the wall, or until it hits the grid ( often ~ 90-95 passes are estimated for this to occur). Without the potential well the ions (electrons in this case) fly straight to the walls/ shell, and would complete at most one pass (ignoring scattering and charge exchange reactions). If 1/2 of the shell is magnetically shielded, the charged particle would make two passes on average before hitting the shell. If 90% of the shell is shielded, then ~ 10 passes, etc. This assumes the charged particle does not become traped on a magnetic field line, in which case it would reach the shell or it would bounce back and forth along the field line. This is what is attempted in some Penning trap designs. None of this is good enough of course. unless you can constrict the cusp opening enough , recirculate the escaping particle at low energy costs, or use other tricks to reach the same end.

There is a huge body of evidence describing cusped magnetic confinement of charged particles for many passes. The contention is the magnitude that is claimed for the Wiffleball effect.

PS: No time to be effected by the magnetic field? So long as the particle is traveling slower than the speed of light, it will be turned by a magnetic field if the magnetic field is strong enough that it can accomidate the gyro radius of that charged particle at that energy. If you don't beleive that, then explain how the LHC manages to steer and control particles at several thousand GeV.

Dan Tibbets

[WizWom]

WisWorm, how do you justify this claim. In a non cusp system the ions will be contained till they can work their way through the magnetic field by several mechanisms (assuming the gyroradius is less than the distance to the magnet casing or first wall). In a cusp system, this 'baseline' leakage is dwarfed by charged particles exiting through cusps. Even without any shrinkage of cusp openings an ion may bounce around 5-10 times before it escapes through a cusp. I beleive this is called mirror confinement. With some closing of the cusp throats this number may increase to ~ 50-60 passes. This is called Cusp confinement by Bussard. I think this has been demonstrated in many different systems like Penning traps. Wiffleball traping at Beta=1 is an extension of cusp confinement and is claimed to be in the thousands.

Dan, I'm not going to bother correcting your basic errors in physics.
Simply: if the magnetic field is in the same direction as a particle is moving, their is no "gyro radius". that only happens when the ion is travelling perpendicular.

The magrid is a cube of magnetic fields facing in; these fields will NOT AFFECT any ion going exactly toward them, except to slow them. The well depth won't make an appreciable dent in the energy of a 2.46MeV alpha particle. The source ions, however, are not traveling that fast at all, they have energies in the 200 keV range.

Now enters the potential well set up by the excess energetic electrons. This potential well confines the ions for claimed times much above the Wiffleball traping factor. In an ideal system the ions would be contained indefinatly. But if the ion is upscattered it might gain sufficient energy to escape the potential well.

OK, so you're saying an ion pushed somewhat above 200keV will escape, but don't think an ion with 10x that energy will? Do you have any idea what you are typing, or do you just spout for amusement?

[quote="D Tibbets"Also, keep in mind that this is not a magnatized plasma, the charged particles spend most of their time flying around in the non magnetic central region of the machine. The ions and to a lesser extent the electrons(?) generally do not gyrate around magnetic field lines except when they are passing through a cusp (or rebounding off the Wiffleball border where they complete ~ 1/2 gyroradius orbit before rentering the magnatic free region and continueing on their merry way.[/quote]OK, add more reasons why the Alphas don't recirculate. Good.

The same process applies to electrons, they are confined dependant on the relative size of the cusp holes as a percentage of the total surface area of the magnetic border.

The electrons are trapped better because they are lighter.

If you hate the Polywell claimes

Where did you get that idea? I hate stupid people who don't know physics and spout off about things that just WON'T HAPPEN as if they are gospel.

There is a huge body of evidence describing cusped magnetic confinement of charged particles for many passes. The contention is the magnitude that is claimed for the Wiffleball effect./quote]Yes, and, generally, it's expected that your confinement is limited by the energy of the ions verses the magnetic field. An ion going against a magnetic field loses energy. But nowhere near enough with a 10T field to stop a >2MeV alpha.

PS: No time to be effected by the magnetic field? So long as the particle is traveling slower than the speed of light, it will be turned by a magnetic field if the magnetic field is strong enough that it can accomidate the gyro radius of that charged particle at that energy. If you don't beleive that, then explain how the LHC manages to steer and control particles at several thousand GeV.

The speed of light has nothing to do with it; if you could charge a photon, it would get turned by a perpendicular magnetic field just the same.

Perpendicular. Parallel fields either add or subtract energy from the particle.

I take it no one has actually run the numbers for a polywell in steady-state run conditions, to see just what the alpha particles typical speeds will be going out the faces and out through the funny and line cusps. Oh, right. No one has even managed a crude simulation of the charge density inside, you're thinking the device will have "charge shells" or some other insanity.

Why not just shut up until you've got something like a clue?

[krenshala]

Its the theory section of the forums that has all the numbers on that, right?

[D Tibbets]

WisWom,
My communication skills are not the best and I tend to ramble, but you are way off in left field. If the charged particle has a greater kinetic energy than the electrostatic potential well, of course it will escape that well. This has nothing to do with the separate issue of magnetic confinement (or as Bussard referred to it, magnetic constraint).

And magnetic fields do not slow or decelerate (accelerate) charged particles. They can only only change their direction. If the magnetic field gradient is steep enough a charged partical can complete ~ 1/2 of a gyroradius orbit and gob back ~ in the direction it came from. This is supposed to be a major process in the Polywell. If the magnetic field gradient is much flatter than the gyroradius (like the Earth's magnetic field) then it will capture a charged particle and confine it to a magnetic field line where it will oscillate back and forth till it hits something or it works its way through the field by several transport mechanisms. If a charged particle approaches a magnetic field parallel or nearly parallel to the magnetic field lines its gross direction may not change much, but it will spiral around a field line. If there are two opposing magnetic fields there will be an infinitely small area where there is no field and a charged particle could pass straight through that cusp. Cusps are regions where the opposing field lines are nearly parallel to the core of the machine and charged particles escape here because when they bounce they hit the opposite side of the cusp and continue on like a rock rattling down a drain pipe. It is a matter of geometry.

I don't recall the exact numbers, but Nebel mentioned the alphas don't hit the magrids provided the size of the machine is great enough and the magnetic fields are great enough. I think the numbers was somewhere near ~ 2.5 Tesla in an ~ 1 meter machine (at those alpha energies and B field strength the gyroradius (Lamar radius) is ~ 10-15 cm (?).

Either we are having a failure to communicate, or you should really look up charged particle behavior in magnetic fields.

Dan Tibbets

[KitemanSA]

I don't recall the exact numbers, but Nebel mentioned the alphas don't hit the magrids provided the size of the machine is great enough and the magnetic fields are great enough. I think the numbers was somewhere near ~ 2.5 Tesla in an ~ 1 meter machine (at those alpha energies and B field strength the gyroradius (Lamar radius) is ~ 10-15 cm (?).

Either we are having a failure to communicate, or you should really look up charged particle behavior in magnetic fields.

Dan,
You are basically right, he is wrong. He seems to have a totally invalid mental picture of the mag field of the MaGrid. He has them pointing in radially all around which is nonsense. They go in at a point (small area), spread laterally under the influence of the electron pressure at Beta=1, and go out at other points. Those points are called cusps. The overall configuration of lateral fields and holes is called the wiffleball.

By the way, Dr. N's B field for his statement of ~1000 passes was 10T which AFAIK is taken to be the field measured at the centerline of the coil. The field right near the coil itself, under the influence of the electron pressure can be higher, no?

[TallDave]

What Dan and Kite said (except I still believe upscatter is a "red herring" as Rick tells us Luis Chacon put it).

Actually, WizWom, we mostly thought the same about alphas (in fact, we thought alpha sputtering would be a major problem) until Rick shared this:

The alphas make about 1000 passes before they exit through the cusps. They leave at essentially full energy.

viewtopic.php?p=18432&highlight=alphas#18432

Don't take it too hard, we've all had to adjust our thinking a lot over the years here at T-P.

[TheRadicalModerate]

Ah, got it. I was thinking that the B-field was a sheath, but it's not. It's diamagnetically rejected from the center of the machine, but the mostly transverse B-field gets stronger and stronger the closer it gets to the magrid coils. So the real question is what the condition is to avoid the fusion products from hitting the coils, and the answer is that the radial distance from the coils to the diamagnetic boundary has to be greater than the gyroradius of the alpha fusion product when it's subjected to some average of the B-field strength between the boundary and the coils.

If that condition is met, then the alpha particle will be mirrored back into the center and it will bounce around until it finds its way out of a cusp. If it's not met, its trajectory will get bent transversely, but it will escape and either hit the coil or the outer vessel wall.