Mass comparison of Focus Fusion and Polywell systems

[Aero]

I was blathering over in the NEWS forum, Lawrenceville Plasma Physics thread, about comparing Focus Fusion Power to Polywell power for flight vehicles, but chrismb pointed out that we were off-topic. Here is where the drift got serious, but hopefully this thread will stand alone.
viewtopic.php?t=2596&postdays=0&postorder=asc&start=30

These are some assumptions:

• A two meter diameter Polywell generates 100 MW,
  Polywell Magrid shadow factor is 20%. (Magrid shades 20 % of the interior wall of the vacuum chamber, (spherical chamber))
  Polywell Magrid density ranges from 3 mtons/cubic meter to 5 mtons/cubic meter,
  A Focus Fusion thruster generates in the range of 10MW to 45 MW,
  An FF thruster masses from 50 kg to 500 kg.
  FF thrusters can be ganged in large numbers,
  Balance of Plant mass and volume are not considered.
  Other?

Power Plant applications are less sensitive to mass and volume than flight vehicle applications, but, for mature systems, construction cost can be estimated roughly by system mass. Lets look at the mass of 100 MW systems.

A 100 MW Focus Fusion system mass ranges from 150 kg to 5000 kg under the above assumptions.

Polywell Magrid mass can be estimated with the above assumptions by calculating its volume and multiplying by density. For radius = 2 meters, the total length of all of the coils is 37.7 meters. That is, 6 x pi x D. The total area of the "sky" at the Magrid distance from the center of the Polywell core is just the surface area of a sphere with 2 meter radius. That is, 50.3 square meters, and 20 % of that is 10.05 square meters. Dividing the shadowed area by the total length of the 6 coils gives the coil minor diameter of .27 meters. (That's really a chord, but I'll call it the diameter). The volume of the Magrid then is the cross sectional area of a coil times the total length of the coils, that is 0.056 square meters times 37.7 meters, volume = 2.11 cubic meters.
A 100 MW Magrid mass ranges from 6.3 mtons to 10.5 mtons. I have:

A 100 MW Focus Fusion system masses from 150 kg to 5000 kg.
A 100 MW Polywell Magrid masses from 6,300 kg to 10,500 kg.

My conclusion is: If the mass of the Focus Fusion system comes in toward the low end of the range, then chances are good that mature FF systems will cost less than Polywell systems.

I need better assumptions to estimate the mass of the FF system. Would someone who has a better knowledge than I please comment? My concern is that FF system mass may be so low that FF systems out perform Polywell throughout the useful range of power. However, spaceships can utilize larger power-plants than is common for ground based, so I'm looking for the size where both technologies mass the same. Larger than that the Polywell should have the advantage. I would also like to estimate volume of the systems, as frontal area is significant in airplanes and spaceships, but have no data.

[Tom Ligon]

We're lacking some critical data here, namely which, if either, will actually work!

That said, it has been clear to me for some time that the basic Focus Fusion device can be more compact than Dr. Bussard's net power Polywell projections. That argues in favor of it at least for smaller vehicles. My running gag is that the Ford Focus Fusion (ain't it handy how they already own both names) might use a piston engine with some really fancy plasmoid-firing spark plugs.

I'm worried I'll need a lot of lead in the firewall, though. Brem, even if tamed to powerplant standards, would still be a bear in a passenger car. So there are limits.

[Aero]

Tom, the critical assumption here is that they both do work...

... it has been clear to me for some time that the basic Focus Fusion device can be more compact than Dr. Bussard's net power Polywell projections.

That's clear, the question is a mater of degree. For example, use the mid-range numbers, 27.5 MW at 275 kg for the Focus Fusion device and for the Magrid, 8,400 kg. That is the mass of 30 Focus Fusion devices which would produce 840 MW. Compare that to 100 MW for the Polywell, the FF/Polywell power ratio = 8.4.

Its true that Polywell power increases dramatically with radius. For example, a Magrid with a 4 meter radius would produce 3.2 GW, as I scale things. But the mass goes up big time, to 150 mtons. Now, 150 mtons of FF devices at 275 kg each would produce 6.7 GW under those assumptions. But Polywell has gained. The FF/Polywell power ratio is now only about 2..
OK, 6 meter Polywell, 24.3 GW, 227 mtons = 827 FF devices, 22.7 GW. Now the FF/Polywell power ratio advantage swings to favor the Polywell at 0.97.

But I am thinking that the mass of a Focus Fusion device might be down in the sub one hundred kg range. Use 100 kg per FF device and you get 2.75 times more power from the Focus Fusion devices for the masses listed above.

Now for the question, "Where does one use a 24 GW power plant?" The only answer I can think of is, "In spaceships."

[TallDave]

Too much sky, not enough pie. I don't think we know enough about the parameters of an operational machine in either case. For instance, how much more density could we get from a Polywell with POPS?

I think we can usefully say only this: we know there's a good chance working Polywell and FF reactors could both be small enough to be economical, assuming they work.

[Aero]

Too much sky, not enough pie. I don't think we know enough about the parameters of an operational machine in either case. For instance, how much more density could we get from a Polywell with POPS?

Well that's true but we're unlikely to get more and it hasn't stopped our speculating in the past. Are we going to turn over a new leaf now?

I think we can usefully say only this: we know there's a good chance working Polywell and FF reactors could both be small enough to be economical, assuming they work.

We can say a lot more than that. We can develop good estimates for the mass of a Polywell given a radius. Its harder to estimate the power but that hasn't stopped us in the past. We can make WAGs at the power and mass of a Focus Fusion device, too, but I'd like to have a narrower range than the range I have stated above. And as above, the assumption that they both work is the critical assumption in this thread.

[WizWom]

A 100 MW Focus Fusion system masses from 150 kg to 5000 kg.
A 100 MW Polywell Magrid masses from 6,300 kg to 10,500 kg.

All well and good, but I doubt the FF will have the scaling factor of polywell. Polywell will very quickly surpass FF at higher energy needs.

And we're talking Terahjoules delivered at Gigawatt rates for getting to orbit.

[Tom Ligon]

All I can say is that if we get to the point where we're debating if Polywell or Focus Fusion works best, and both work, that would be one helluva nice predicament to have!

Unless, of course, you work for ITER or NIF, in which case you'd have some 'splaining to do.

Regarding 24 GW powerplants, I recently looked up the capacity of the Mount Storm coal fired plant in West-By-God Virginia. It is 1.6 GW. I was interested in comparing it to the entire wind generation output of AES, who is installing turbines on our mountain soon (to considerable displeasure of many neighbors). They cite 1.3 GW "of wind capacity" in the US, Europe, and China (and don't say if that is installed power rating or production with a utilization factor probably in the 0.3 range).

Considering Mount Storm is Dominion Electric's largest plant, I'd say over an order of magnitude larger would indeed beg the question of justifiable economic scale. But that could be another good probem to have. It could open up industrial opportunities we just cannot imagine now. For instance, recycling obsolete wind turbines.

[Aero]

A 100 MW Focus Fusion system masses from 150 kg to 5000 kg.
A 100 MW Polywell Magrid masses from 6,300 kg to 10,500 kg.

All well and good, but I doubt the FF will have the scaling factor of polywell. Polywell will very quickly surpass FF at higher energy needs.

And we're talking Terahjoules delivered at Gigawatt rates for getting to orbit.

My point is that the Magrid is so massive and the mass grows so quickly with radius that gangs of FF devices will produce more power than the Polywell over a wide range of Polywell scaling. In the initial post, I calculate that a 6 meter radius, 24 GW polywell is the smallest size that gives as much power as ganged FF devices of equal mass.

[Tom Ligon]

Do we know that the mass of a magrid really scales that way? That will depend on the superconductor technology, and particularly the maximum magnetic field the superconductor can take before it goes resistive.

The problem with small superconducting liquid helium magrids is the cryo jacket system is fat relative to the superconductor. You have superconductor in liquid helium, a vacuum break, liquid nitrogen, another vacuum break, and maybe a conventional coolant outside that. Larger radius magnets need not have any thicker jacketing system. Magrids could have the same thickness as they get bigger. Circumfrence of magnet will thus scale linearly with radius, and area and mass with circumfrence. Shadowing also drops with increasing size, although this loss is small enough the gain does not matter much.

So if the volume of a Polywell goes up by the cube of radius, and magrid mass and area linearly with radius, that does not say to me that the mass goes up "rapidly". Power is supposed to scale as B^4R^3, so a great deal will depend on the superconductor performance.

That's the neat thing about presuming zero resistance. Your conductor does not necessarily have to get thicker to carry more current, at least up to some critical threshhold.

[D Tibbets]

45 MW for a FF reactor sounds very optimistic. I've heard numbers of 5-10 MW net power as being good. At a Q of three, that is ~ 10 MW of useful power and ~ 3 MW of waste heat. In a small vessel, that will be hard to handle. There will need to be a lot of cooling of the electrodes, and vessel wall. Also, the required bremsstrulung X-ray energy recovery looks like it will take up significantly more space and possible more mass than the reactor itself.
Admittedly a 2 meter Polywell producing 100 MW of power with P-B11 fuel is also optimistic.

I suspect a FF may provide more compact and lighter power/ thrust source, but the breakeven point between the two will be much earlier than the gigawatt range.
Keep in mind that Bussard has toyed with the application of a Polywell to power a semi truck. I don't know what optimizations he was considering, other than he said "second generation". Perhaps he was envisioning ridiculously strong magnets (like 40-50 Tesla).

Dan Tibbets

[TallDave]

Well that's true but we're unlikely to get more and it hasn't stopped our speculating in the past.

True enough, but if there's a 10% chance our estimates of either tech is correct, there's only a 1% chance that our estimates of both are correct. It's still sort of fun, though.

As Tom points out the SC performance is an important variable, so we should probably hope Johan was on to something.

[Aero]

Do we know that the mass of a magrid really scales that way? That will depend on the superconductor technology, and particularly the maximum magnetic field the superconductor can take before it goes resistive.

The problem with small superconducting liquid helium magrids is the cryo jacket system is fat relative to the superconductor. You have superconductor in liquid helium, a vacuum break, liquid nitrogen, another vacuum break, and maybe a conventional coolant outside that. Larger radius magnets need not have any thicker jacketing system. Magrids could have the same thickness as they get bigger. Circumfrence of magnet will thus scale linearly with radius, and area and mass with circumfrence. Shadowing also drops with increasing size, although this loss is small enough the gain does not matter much.

So if the volume of a Polywell goes up by the cube of radius, and magrid mass and area linearly with radius, that does not say to me that the mass goes up "rapidly". Power is supposed to scale as B^4R^3, so a great deal will depend on the superconductor performance.

That's the neat thing about presuming zero resistance. Your conductor does not necessarily have to get thicker to carry more current, at least up to some critical threshhold.

Thanks for that information.
Have you considered the difficulty of surrounding the actual coil in a vacuum jacket? I wonder if it is possible, but it will certainly take a lot of mass if it is, and the more powerful the magnet, the more structural mass needed. Has any other project ever taken even two magnet coils in the Tesla strength range, placed them face to face (like poles facing), separated by their diameter, then turned them on?

I did a force calculation for a pair of 2 meter radius, 4 Tesla coils, facing each other, 4 meters apart. The magnitude of the force is an astounding 43 million newtons. That's about 4.4 million kg-f . Of course we realize that this force is reacted on the superconducting wire itself so the wire is what must be supported. If my numbers are right, the force on each meter of coil is 3.4 million newtons, or 274 mton-force. That's 274 kg-f/mm

That's ridiculous, I must have used the wrong equations, someone please check my calculations,

Edit : Check, because these numbers are for WB-100, they don't take into account the other 4 coils of the Magrid or the forces holding the wiffleball and they are still large enough to blow the fusor apart.

[Aero]

I went through the calculations using WB-6 as a check. Its an Excel spread sheet so I just plugged in WB-6 numbers. I calculated total repulsive force as follows:

Code: Select all

    F =     107.421875    Newtons                
        1.07E+02    Newtons        1 newton =    0.101971621    kilogram-force
        1.10E+01     kilogram-force                

Coil Length, pi() x d =        0.942477796    meters                

    F_rate =     1.14E+02    Newtons/meter                
        1.16E+01    kg-force/meter                

These numbers look reasonable to me for WB-6.

[GIThruster]

Now for the question, "Where does one use a 24 GW power plant?" The only answer I can think of is, "In spaceships."

Good question. I went looking for the specs on the new A1B reactors planned for the new Ford class supercarriers. I checked almost 20 places on the web and they all say the Ford class will have between 1.5 and 3X the electrical power out as the Nimitz they're replacing. Trouble is, no one says how much the reactors put out.

The Ford class has EM catapults and is designed to be upgraded with things like lasers and rail guns--all very electric power intensive, so I reasoned perhaps maybe the Ford starts to approach the need for 24 GW. After all, you're driving a city sized ship at 33 kts. I did not however, find the specs on the new reactors. Disappointing. Guess it's classified, but that's the only application other than a spacecraft I can see for a 24 GW reactor.

[GIThruster]

Do we know that the mass of a magrid really scales that way? That will depend on the superconductor technology, and particularly the maximum magnetic field the superconductor can take before it goes resistive.

The problem with small superconducting liquid helium magrids is the cryo jacket system is fat relative to the superconductor. You have superconductor in liquid helium, a vacuum break, liquid nitrogen, another vacuum break, and maybe a conventional coolant outside that. Larger radius magnets need not have any thicker jacketing system. Magrids could have the same thickness as they get bigger. Circumfrence of magnet will thus scale linearly with radius, and area and mass with circumfrence. Shadowing also drops with increasing size, although this loss is small enough the gain does not matter much.

So if the volume of a Polywell goes up by the cube of radius, and magrid mass and area linearly with radius, that does not say to me that the mass goes up "rapidly". Power is supposed to scale as B^4R^3, so a great deal will depend on the superconductor performance.

That's the neat thing about presuming zero resistance. Your conductor does not necessarily have to get thicker to carry more current, at least up to some critical threshhold.

Second Gen high temperature superconductors like YBCO don't require liquid helium. The jacket for the liquid nitrogen is quite small, and total it is much smaller than copper able to carry the same current. All the motors and generators being built for large duty on Naval ships are now using YBCO as the HTSC systems are all much smaller and obviously, more efficient than their copper ancestors. Almost all the work in ultra-high mag fields is being done with YBCO as well.

Here's an idea of the numbers you're looking at for current density:

http://www.superpower-inc.com/system/fi ... n_0809.pdf

I think the Poly is going to scale better than some suspect, based entirely on this YBCO tape, presuming it is not fragile towards things like X-Rays. Would be good to find out.