Polywell Electrical System

[chrismb]

once the alphas are neutralized back to plain old helium, is there a planned mechanism for extracting them from the vaccuum chamber, or is one of the engineering hurdles that still needs to be investigated once they determine direct conversion of the alphas is even realistic?

You have hit what is likely an insurmountable problem for a multi MW polywell....

"Incidentally, I think this flux of 3E21 alpha particles would require a pumping speed of 120 billion litres per second to maintain a 1E-9 torr vacuum. That's about 50,000 olympic sized swimming pools of volume - per second. Those will sure be some cool vacuum pumps!!.... Where [on earth] does one buy billion litre/sec vacuum pumps?? Whoever makes them is gonna have a lot of business because we'll need over a hundred of them for each 500MW BFR!"

from;

viewtopic.php?p=18484#18484

[MSimon]

once the alphas are neutralized back to plain old helium, is there a planned mechanism for extracting them from the vaccuum chamber, or is one of the engineering hurdles that still needs to be investigated once they determine direct conversion of the alphas is even realistic?

You have hit what is likely an insurmountable problem for a multi MW polywell....

"Incidentally, I think this flux of 3E21 alpha particles would require a pumping speed of 120 billion litres per second to maintain a 1E-9 torr vacuum. That's about 50,000 olympic sized swimming pools of volume - per second. Those will sure be some cool vacuum pumps!!.... Where [on earth] does one buy billion litre/sec vacuum pumps?? Whoever makes them is gonna have a lot of business because we'll need over a hundred of them for each 500MW BFR!"

from;

viewtopic.php?p=18484#18484

Sure. Except that the vacuum only need be maintained in the 1E-6 to 1E-7 torr range which is tractable.

Of course if it has to be held to 1E-15 torr that will require 120E15 l/second. If it has to be held to 1E-21 torr that would be 120E21 l/s.

I think a vacuum pump as big as the sun will do it.

I like it though. You can make the scheme seem as impossible as you want if you only choose your numbers carefully.

Like maybe 20 million billion volts for particle acceleration. And magnets 120 km in diameter.

[chrismb]

Sure. Except that the vacuum only need be maintained in the 1E-6 to 1E-7 torr range which is tractable.

Because you'd then only need a 100 million litre/sec vac pump!?

[MSimon]

Sure. Except that the vacuum only need be maintained in the 1E-6 to 1E-7 torr range which is tractable.

Because you'd then only need a 100 million litre/sec vac pump!?

This has been discussed before. As I recall Dr. Nebel chimed in and shot you down as well. I believe I worked over those numbers before and came up with 1E6 l/s. IIRC. I'm not in the mood to do the calculation. But IIRC I estimated a vacuum pump with an inlet 3 m across would do the job. i.e on the same order as the size of the reaction space.

But here is one: a 3,000 l/s pump has an inlet 10 cm across (actually about 8 but I'm interested in ball parks). 1 m across would be 300,000 l/s. 3 m across would be 1 M l/s.

NASA has probably built pumps 10X as large for its very big vacuum chambers. Those could do in the range of 100 M l/s.

What you keep forgetting is that Dr. B was a top engineer. I doubt if he would have left vacuum pumps out of his calculations.

[chrismb]

As I recall Dr. Nebel chimed in and shot you down as well.

hah!? I don't recall being shot down. Me - I'm still flying high, fella.

Though looking back, I see rnebel didn't seem to be prepared to discuss it further, apart from making a hip-shot random comment about someone else's experiment, which seems to have no bearing on future MW prospects for his own. I'm sure he knows the score and doesn't need/desire to be involved at this level.

I've got my feet firmly planted in the clouds and a target roundall painted on my back - anyone wanna take a shot?

[KitemanSA]

I've read that electrons must first be fired with E guns into the polywell to create the wiffleball. The -ve electrons are trapped inside the polywell by the the +ve charge on the polywell and are prevented from hitting the polywell by the strong magnetic field around the polywell. The magnetic field also recirculates the electrons back inside the polywell if they do escape.

Fundamental error: The electrons are not "fired" and the magnets do not recirculate the electrons, it is the positive charge on the MaGrid that pulls the electrons into the core in the first place, and then pulls them back if they wriggle out a cusp. The "electron guns" on the WB6 and I suppose the WB7 were filiments from truck lamps without the glass. It is the difference between ground and the +12,000V MaGrid casing that accelerates the electons.

[zimdlg]

Thanks KitemanSA, glad to be corrected. I have to think carefully to keep all the interactions straight in my head.

KitemanSA

The "electron guns" on the WB6 and I suppose the WB7 were filiments from truck lamps without the glass

Is the electron source in WB8 so simple, if anyone outside EMC2 knows? Any speculation about something better for the net power polywell?

[MSimon]

Thanks KitemanSA, glad to be corrected. I have to think carefully to keep all the interactions straight in my head.

KitemanSA

The "electron guns" on the WB6 and I suppose the WB7 were filiments from truck lamps without the glass

Is the electron source in WB8 so simple, if anyone outside EMC2 knows? Any speculation about something better for the net power polywell?

Absolutely. Thoriated filaments.

[MSimon]

BTW the future is less than 1 nanosecond away.

[tombo]

a GE CF6-6 jet engine (boeing 747) is 2.19 m dia
pumps 590 kg/sec
air at stp is 1.2 kg/m^3
that gives 491,667 liters per second
scaling the area to 3.125 m dia would give 1e6 l/sec
This agrees with:

But here is one: a 3,000 l/s pump has an inlet 10 cm across (actually about 8 but I'm interested in ball parks). 1 m across would be 300,000 l/s. 3 m across would be 1 M l/s.

NASA has probably built pumps 10X as large for its very big vacuum chambers. Those could do in the range of 100 M l/s.

It is going to be an "interesting" engineering problem to fit 100 turbo pumps the size of 747 engines around a vacuum chamber.
Consider what you kid's 100 sided D&D die looks like.
And there are going to be megawatts expended just to overcome friction in that kind of vacuum system. (WAG)

OR

NASA Has built 30 M diameter vacuum pumps?
Really?
That is almost as wide as a football field. And it's going to spin HOW fast?
IIRC The 6" turbo pumps that I used turned 12,000 rpm.
I'd like to see that machine. It would be very impressive.
The final dynamic balancing of that shaft will be quite a trick too.

Vacuum loads will be aggravated by the out-gassing from the hot wall of the heat exchanger feeding the heat engine which will be needed to extract the bulk of the energy.

There had better be a way to finesse this one.

So:
If the alphas are channeled to strike the chamber walls in 14 spots 1/2 m diameter each and the chamber is 2.5 m radius (1 m more than the magrid). That gives a 24x improvement. ie the partial "pressure" of the particles is 24 times higher than if they were distributed over the whole vacuum chamber. So now we are down to 4 each 747-engine-sized vacuum pumps. That seems doable but still very expensive.
OR
Scaling the vacuum pump above the brings down the monster turbo pump's throat to a mere 12 meters or 6 meters each if we use 4 of them.
That's a lot better. But still way expensive.

This is going to cost on the order of the steam power generator that heretofore was the biggest cost item.

There had better be a better way to finesse this one.

[zimdlg]

If you want to be picky it's one planck time away

What I mean is "the future" as in strong AI, exploring the solar system, direct brain interface, insert favourite sci fi here, etc

this is WAY off the thread topic.

[D Tibbets]

a GE CF6-6 jet engine (boeing 747) is 2.19 m dia
pumps 590 kg/sec
air at stp is 1.2 kg/m^3
that gives 491,667 liters per second
scaling the area to 3.125 m dia would give 1e6 l/sec
This agrees with:

But here is one: a 3,000 l/s pump has an inlet 10 cm across (actually about 8 but I'm interested in ball parks). 1 m across would be 300,000 l/s. 3 m across would be 1 M l/s.

NASA has probably built pumps 10X as large for its very big vacuum chambers. Those could do in the range of 100 M l/s.

It is going to be an "interesting" engineering problem to fit 100 turbo pumps the size of 747 engines around a vacuum chamber.
Consider what you kid's 100 sided D&D die looks like.
And there are going to be megawatts expended just to overcome friction in that kind of vacuum system. (WAG)

OR

NASA Has built 30 M diameter vacuum pumps?
Really?
That is almost as wide as a football field. And it's going to spin HOW fast?
IIRC The 6" turbo pumps that I used turned 12,000 rpm.
I'd like to see that machine. It would be very impressive.
The final dynamic balancing of that shaft will be quite a trick too.

Vacuum loads will be aggravated by the out-gassing from the hot wall of the heat exchanger feeding the heat engine which will be needed to extract the bulk of the energy.

There had better be a way to finesse this one.

So:
If the alphas are channeled to strike the chamber walls in 14 spots 1/2 m diameter each and the chamber is 2.5 m radius (1 m more than the magrid). That gives a 24x improvement. ie the partial "pressure" of the particles is 24 times higher than if they were distributed over the whole vacuum chamber. So now we are down to 4 each 747-engine-sized vacuum pumps. That seems doable but still very expensive.
OR
Scaling the vacuum pump above the brings down the monster turbo pump's throat to a mere 12 meters or 6 meters each if we use 4 of them.
That's a lot better. But still way expensive.

This is going to cost on the order of the steam power generator that heretofore was the biggest cost item.

There had better be a better way to finesse this one.

It seems your numbers are based on a 100 million l/s need. If M. Simon's estimate of 1 million l/s is used the sizes are corresponding smaller. I like the idea of a 24 fold advantage gained by the alphas coming out the cusp, but the actual number would depend on how focused these beams actually are (from the narrow cusp exits, and any focusing from the direct energy converting grids and deflectors). They may be less or more focused. Also keep in mind that electrostaticlly charged plates might be placed in the throats of the turbomolecular pumps. These would help to collect decellerated but not yet neutralized alphas, effectively increasing the effective size of the pump openings (a further enhancement of focusing) at only a little cost in energy.

Dan Tibbets

[KitemanSA]

Is the electron source in WB8 so simple, if anyone outside EMC2 knows? Any speculation about something better for the net power polywell?

Absolutely. Thoriated filaments.

The electron source has never been an issue, I don't perceive it ever being one.

The ION source on the other hand remains an engineering topic. It has been opined that a neutral gas puff in a full size unit would work very well, but it leads to problems at small scale. Also, obtaining a neutral gas puff for Boron11 seems to require the use of one of the HydrogenBorides which tend to be quite toxic. What fun!

[D Tibbets]

This post should have been posted after
chrismb Posted: Thu Jul 16, 2009 4:23 pm
Please view it that way.

The inside of the polywell has to be negative otherwise the positive ions will not accelerate to the middle, collide and fuse. The electrons are fired out the E guns on the wall of the vacuum chamber towards the middle. They are attracted to the +ve magrid but are prevented from touching it by the magnetic fields around the magrid. The electrons are compelled to follow the magnetic field lines into the polywell even though it is negative inside. Once they are inside they are trapped by the wiffleball effect. I assume the +ve ions are fired from ion guns at the vacuum chamber walls with enough velocity and into a cusp to overcome the repulsion from the +ve magrid. Once past the magrid an ion is attracted to the -ve wiffleball and will continue oscillate about the middle of the wiffleball until it collides with another ion and fuses.

I was assuming that the ion guns would have to be located within the magrid so that the ions could be emitted at low velocities, so that they would not have enough kinetic energy to climb to deeply into the cusps on subsequent passes. But ions fired from an external gun would see both the more distant negative potential inside the machine and the positive potential on the magrid. So an ion gun voltage could be selected that allows the ion to enter the internal magrid volume with only a few residual eV of energy. There would not be any significant loss of energy as the decelerating ions would be transferring their energy to the magrid, in a sense helping to power the electron acceleration at other cusps.

In a perfect noncollisional system an ion that entered through one cusp with a given speed could exit through another with the same velocity, but in a real system the ion would lose some of it's radial velocity due to collisions, and so long as this process (thermalization) is not to fast you have effective confinement along with convergence (so long as the electron derived potential well can also be maintained) that allows the Polywell to be more efficient than thermonuclear systems. If my reasoning is sound (no guarantee) this would again represent the compromises in the Polywell concept that actually leads to it working.

[Enter here other concerns that contribute to the performance- potential well shape, quasineutrality and ambipolarity issues, sheath effects, annealing, electron confinement and recirculation issues, bremsstrulung, sputtering, magnet issues, vacuum maintainance issues, energy harvesting issues, radiation issues,etc., etc.]

Dan Tibbets

[EricF]

I like the idea of a 24 fold advantage gained by the alphas coming out the cusp, but the actual number would depend on how focused these beams actually are (from the narrow cusp exits, and any focusing from the direct energy converting grids and deflectors). They may be less or more focused. Also keep in mind that electrostaticlly charged plates might be placed in the throats of the turbomolecular pumps. These would help to collect decellerated but not yet neutralized alphas, effectively increasing the effective size of the pump openings (a further enhancement of focusing) at only a little cost in energy.

just spitting out ideas here,and assuming a good confined jet of alphas through the cusps, but would it be realistic to allow the alpha particle to travel far enough to unload its voltage into the collection grid, but neutralize the particle while it is still in motion with enough inertia left over to allow it to travel the remaining distance of the chamber and exit via a one way valve. That would be one small darn valve though. Carbon fiber nanovalve?