Idea: X-Ray reflection

Discuss how polywell fusion works; share theoretical questions and answers.

Moderators: tonybarry, MSimon

KitemanSA
Posts: 6179
Joined: Sun Sep 28, 2008 3:05 pm
Location: OlyPen WA

Post by KitemanSA »

chrismb wrote:I've laid out perfectly reasonable statements for those looking to weight up whether or not Rossi's scheme is a scam (and polywell similarly). I've not said outright that it is, only that I see little that colours it as a bona-fide scheme.
Which means you think it at best specultive and at worst a scam. Goose!

chrismb
Posts: 3161
Joined: Sat Dec 13, 2008 6:00 pm

Post by chrismb »

KitemanSA wrote:
chrismb wrote:I've laid out perfectly reasonable statements for those looking to weight up whether or not Rossi's scheme is a scam (and polywell similarly). I've not said outright that it is, only that I see little that colours it as a bona-fide scheme.
Which means you think it at best specultive and at worst a scam. Goose!
I have not defined my conclusion. Could it be a scam? You'd better have that amongst your set of possible outcomes, else you may get very confused by the outcome.

I think the most likely possibility, amongst many, is that it is another case of a delusional quest for a piece of the modern alchemy. Being wrong doesn't mean you are a scammer or a liar.

Why are you drifting this thread?

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Disregarding, he said, she said back and forth sniping,; there are a couple of points that might be relevant. Radiation screening can be confusing. If you consider that a certain thickness of stainless steel (say 2 inches) stops 90% of the x-rays of a given energy. Then from a energy conversion stand point you have converted 90% of the x-ray energy has been converted to heat, which can be collected and converted by a coolant flow. The remaining 10% can be blocked/ absorbed to 90% efficiency by the next 2 inches of stainless steel or equivalent other material. That leaves 1 % of the x-rays free to proceed. 99% of the x-rays have been stopped and converted to heat. The remaining 1 % is trivial from any perspective of energy recovery. But from a safety perspective this level of x-rays may still be very dangerous. It would probably be necessary to have several additional equivalent walls to get the x-ray levels down to safe levels. But this would only represent a maximum of 1 % of the total x-ray heating power, and becomes log rhythmically less significant from a energy recovery stand point. Passive cooling would probably be completely adequate. These two points are important. Radiation shielding is not a linear process, This means that it might take more or less thickness to stop enough of the radiation than you might think when you are comparing two different radiation fluxes.

For instance, If 1 inch is enough to stop 90% of 1000 units of radiation, then 10 inches should be enough to stop 10,000 units to the same residual number. This is not the case. The first inch would stop 900 units, leaving 100, and 900 units ...... Forget it, I am confusing myself. Look up radiation shielding for an understandable description. Or just take my word for it that it is logarithmic.

The other point is the huge disconnect between the amount of shielding necessary to trap the vast majority of convertible heat, and the amount of shielding necessary for biologically safe fluxes.
The shielding between the level necessary to trap say 99% of the thermal equivalent , and that necessary to trap the remaining X-rays is considerable. But this outlaying shielding is only exposed to the remaining ~1 % of the heat equivalent. This larger volume of material has much more surface are to dissipate this small percentage of the thermal load. Air convection will probably easily handle this. All of the challenges will be in cooling that first few inches of steel. In a spaceship the picture is more complicated, mostly because there is not a large convenient heat sink available.

A direct x-ray conversion scheme at 80% efficiency verses the ~ 25% thermal conversion, may or may not be of great value to a terrestrial Polywell. If the net Q turns out to be ~ 2-5 it may be critical or just more profitable. If the Q is 20 (partly because Bremmstrulung is significantly suppressed compared to the fusion rate) then direct x-ray conversion might be reasonable from a heat management standpoit, but is not critical and may not be worth the cost.
In a spacecraft, because of the much greater difficulty of getting rid of waste heat (and the weight of a steam plant), the direct conversion of both the fusion ions and x-rays may give tremendous advantages. As far as biological x-ray and gamma ray shielding, you substitute stand off distance for much of the otherwise very heavy shielding that would be needed.

As far as cooling the reactor in a space plane / booster, a lot of coolent flow will be needed. A saving grace may be the amount of heat that each kg of coolent can absorb. In a chemical engine, the fuel may be heated to a few hundred degrees while cooling the engine, then combust to reach its final temperature. In a nuclear rocket, the fuel is inert and can absorb more heat as it cools the reactor, perhaps to several thousand degrees (as in a fission nuclear rocket). This increased efficiency is a gain. How far this can be carried would be the finale determinate in how much power/ thrust could be handled. ie: the weight and ISP performance my lag behind, or surpass to some minor to major degree the performance of a chemical rocket. The challenges for a booster is extremely more challenging than a space engine (provided there is not some burdensome minimum size to the reactor (like a Tokamak)).

As for the rapid increase in power and thus thermal wall loads as the Polywell scales up in only modest size scales, that is true if you keep the other conditions the same. But there might be considerable range aviable, ie: the machine is throatable. By varing drive voltage, B field strength, possible POPS effects, etc. the power output can be adjusted independantly of the size. There has been speculation, that at least for D-D fusion, the machine may need to be throttled back (larger machine to get the same power) so that aviable engineering wall thermal wall loadings are not exceeded (the limiting factor). I suspect this would be less applicable for P-B11 Polywells, due their predicted larger size and trade offs with direct conversion.

Bussard's prediction about a D-D Polywell being ~ 3 meters in magrid diameter and P-B11 Polywells being 4-5 meters in diameter may be inappropriate. The P-B11 machines may be in this range, but the D-D machines may also be in this larger size range due to thermal wall loading limitations. Direct conversion may have some compounding advantages, in the final output and thermal concerns (of both the magrids and vacuum vessel walls), that outweigh the raw fusion rates/ energy yield.

Dan Tibbets
To error is human... and I'm very human.

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

@Giorgio: I've just about had it with you. Unfortunately, while the post I was composing might actually have cleared up a good deal of this misunderstanding, I've started to lose my temper, and the post is suffering as a result. I've actually got some work to do right now, so I'll finish it and post it at a later time when I've calmed down some.

...

@D Tibbets: I did say half an inch would be good for 80-90% of the x-rays. There's room for interpretation, considering we don't actually know the spectrum, but I didn't pull that number out of thin air:

http://physics.nist.gov/PhysRefData/Xra ... b/z26.html

Giorgio
Posts: 3061
Joined: Wed Oct 07, 2009 6:15 pm
Location: China, Italy

Post by Giorgio »

93143 wrote:@Giorgio: I've just about had it with you. Unfortunately, while the post I was composing might actually have cleared up a good deal of this misunderstanding, I've started to lose my temper, and the post is suffering as a result. I've actually got some work to do right now, so I'll finish it and post it at a later time when I've calmed down some.
I don't see any reason to lose temper just to clear some misunderstanding.
Anyhow, I sent you a Private Message.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

93143 wrote:...
@D Tibbets: I did say half an inch would be good for 80-90% of the x-rays. There's room for interpretation, considering we don't actually know the spectrum, but I didn't pull that number out of thin air:

http://physics.nist.gov/PhysRefData/Xra ... b/z26.html
I looked at the graph and tried to decipher the jargon. I'm not sure what the end point target is, but it looks like 0.1 cm^2/ g is adequate for ~ 1 MeV photons. I think that would be ~ 10 g of iron to stop the radiation (90%?) With a density of ~ 8g/ cm^3, a thickness of a little over 1 cm would suffice (or ~ 1/2 inch) just like you said. It is interesting that not much more thickness would be needed for the 15 MeV gamma. That the levels actuall decrease some in regions above ~ 1 MeV is interesting, there must be some interesting physicis going on.
The actualthickness needed would probably be mildly less than this because the presumed stainless steel may contain portions of chromium, copper, molybdenum, etc.

My rough over estimate was that the thickness of 1-2 inches for the vacuum vessel walls would suffice was based on the assumption that this thickness would be the minimum for structural purposes. Add to that the water pipes, etc and the concern for x-ray deposited heat outside the vacuum vessel becomes increasingly trivial. Obvously, attention would have to be paid to ports into turbo molecular pumps, etc.

Some numbers based on the assumption that 90% of the incident x rays are absorbed by each 1/2 inch length of steel or equivalent, and a flux of 1 MW of x-rays/ M^2 at the vacuum vessel wall (or cooling pipes if they are on the inside surface).

First 1/2 inch- 900 KW of heat /M^2 absorbed , 100KW penetrates
Second 1/2 inch - 90 KW/M^2 absorbed , 10KW penetrates
Third 1/2 inch - 9 KW/M^2 absorbed , 1 KW penetrates
Forth 1/2 inch - 900 W absorbed , 100 Watts penitrates.

So after 2 inches of equivalent iron that is actively cooled, there would be ~100 watts of X ray heating left / M^2. This is comparable to a ~ 100 Watt incandesent light bulb shining all its light on a 1 M^2 section of wall. A wall exposed to this flux would probably only heat enough to feel mildly warn to the touch, unless it was well insulated against conduction and air convection heat losses.

Heat management will be trivial outside the vacuum vessel/ active cooling system envelope. Biological concerns will require additional considerations.

Inside the vacuum vessel the problem is reversed. The magrids may be exposed to 2-4 times the x-ray flux compared to the walls. Cooling efforts per unite of surface area will be 2-4 times more difficult. Two approaches - high transparency to x-rays (this is something Focus Fusion plans for the DPF)- by using structural materials like beryllium. Problem with this is that the superconductor may have heavy metals like copper in it, so it would absorbs a lot of x-ray heat. Stopping most of the x-rays before reaching the superconductors may be better. Very dense metals may help here, though based on the above information, the space gains may be trivial over regular stainless steel. Pushing enough coolant through the Magrid may be the limiting consideration. That is why I mentioned in a previous post that once a certain D-D or D-P11 fusion efficiency is passed, the thermal considerations may dictate the minimum size of a steady state productive reactor.

Dan Tibbets
To error is human... and I'm very human.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Considering the x-ray absorbing efficiency of the vacuum vessel wall and internal structures, if 80->90% percent of the available x-ray heat is absorbed by the walls, there is little advantage in trying to direct convert the remainder. Consideration of lower X-ray opaque materials such a beryllium, lithium aluminum alloy(?) for the vacuum vessel wall may help some, but it would be a problem for a Polywell. A DPF with its polar pulsed fusion ion flows would be much more amenable to X-ray direct conversion efforts.

Dan Tibbets
To error is human... and I'm very human.

93143
Posts: 1142
Joined: Fri Oct 19, 2007 7:51 pm

Post by 93143 »

D Tibbets wrote:I'm not sure what the end point target is, but it looks like 0.1 cm^2/ g is adequate for ~ 1 MeV photons. I think that would be ~ 10 g of iron to stop the radiation (90%?)
63%. It's a decaying exponential. But the bremsstrahlung won't be at 1 MeV either, since the plasma isn't that hot.
That the levels actuall decrease some in regions above ~ 1 MeV is interesting, there must be some interesting physicis going on.
Pair production.

Giorgio
Posts: 3061
Joined: Wed Oct 07, 2009 6:15 pm
Location: China, Italy

Post by Giorgio »

D Tibbets wrote:My rough over estimate was that the thickness of 1-2 inches for the vacuum vessel walls would suffice was based on the assumption that this thickness would be the minimum for structural purposes. Add to that the water pipes, etc and the concern for x-ray deposited heat outside the vacuum vessel becomes increasingly trivial. Obvously, attention would have to be paid to ports into turbo molecular pumps, etc.
I think that in case of an high percentage of bremsstrahlung one of the most important point will be to understand where they will be produced and where they will actually impact the vacuum vessel.
I doubt we will have an uniform distribution as it has been implied till now.

This (and the Alpha distribution) will probably drive the whole design process for the vacuum vessel and heat exchange loops.

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

I used 1 MeV photons as a convenient round up number representing an overestimate of the energies involved, and 90% absorption efficiencies, again for convenience.
Where within the Wiffleball most of the P-B11 fusion will occur is an open question. It may be anywhere from diffuse production, to a tightly focusesd core region. 90% or more of the fusions may (or may not ) occur in a small core. In either case, there will be a variable shadow region behind the magrids dependant on this and the cross section thickness of the magrids. Lots of magnetic cusp optimizations (central point cusps), spacing, cusp losses verses cross field transport losses, heat loads, protection of direct conversion grids by placing them in the magrid shadow, port placements,etc that will evolve into the best balance between power density, size, configuration, etc. A virtual smorgasbord for engineers. Of course all of this assumes P-B11 fusion actually works at an economical level.

[EDIT] Irregardless of where the fusion occurs, I should point out that the Bremsstrulung should originate mostly in a central core region because this is where the ions are expected to be at the highest density if there is any convergence. Because the electrons are slowest here the amount of Bremsstrulung will be greatest here in an occurrence perspective, but the energy per photon will be less than the less frequent ion electron collisions further from the center. There is a trade off between a low energy high intensity core and a high energy but low intensity periphery. How it would balance is interesting from both energy loss; and thermal loading , penetration and shadowing perspectives.

Dan Tibbets
To error is human... and I'm very human.

NewtonPulsifer
Posts: 3
Joined: Thu May 05, 2011 4:19 am
Location: California

Post by NewtonPulsifer »

For neutronic and/or very high percentage bremsstrahlung fusion it seems using aluminum would be a good (maybe the only feasible) material choice for engineering a polywell design that lets neutrons and x-rays just "pass through" to a moderator layer/fluid for thermalizing the neutrons via boiling pressure, or an x-ray absorbing material that heats up to generate a similar effect (or both).

From my reading aluminum is relatively transparent to X-ray photons 50kev and above, and along with zirconium one of the two elements most transparent to neutrons.

So the magnetic coils could be made from aluminum instead of copper with a somewhat reduced magnetic field strength, perhaps beefier to compensate.

Not exactly the "lightweight and clean" aneutronic fusion we all are hoping for but still a reactor design that might be able to compete quite favorably with current nuclear fission power station designs?

Reading:

X-ray and neutron transparent pulse magnets (using aluminum)

http://iopscience.iop.org/1742-6596/51/ ... 51_113.pdf

Radiation Penetration: Half Value Layer for Aluminum
http://www.sprawls.org/ppmi2/RADPEN/#HA ... UE%20LAYER

D Tibbets
Posts: 2775
Joined: Thu Jun 26, 2008 6:52 am

Post by D Tibbets »

Making the magnet wires more transparent to neutrons and/ or x-rays is reasonable within limits. In a 3 meter diameter magrid, the non superconducting magnet wire bundle may be ~ 30-40 cm wide (assumes a magrid minor diameter of ~ 25% of the magrid diameter, and a ~ 40% fill factor). If copper would absorb most of the x-rays and/ or at least partially moderated neutrons, and a lighter metal like aluminum would only absorb perhaps 50%, then the heat load on the windings would be ~ 1/2. This advantage would be somewhat less as thicker wires would be needed. This may be a significant engineering advantage. An increased allowable packing fraction would possibly allow for increased magnetic fields despite the less conductive and thicker wires. Aluminum's electrical conductivity is only slightly lower than copper. The resultant Ohmic heat load would need to be handled by faster coolant flow and/ or thicker wires. Considering the megawatts of fusion neutrons or X-rays the resultant gain may be worthwhile.

Aluminum is not the only possible substitute for copper. Lithium, magnesium and beryllium are other candidates. Beryllium is very transparent to x-rays and yet is a strong, high melting point material. It has an electrical conductivity ~ 2/3rds of copper. Beryllium probably would not be a good candidate for D-D fusion because of all the neutrons, but it might be attractive for P-B11 fusion.
Another question is how the conductivity changes with temperature. Would the other metals have the same improvement in conductivity at liquid nitrogen temperatures?

Then, there is the graphene possibilities. Significantly higher electrical conductivities may be obtained. The X-ray transparency is possibly good. The tolerance and transparency to neutrons is also a consideration.

Dan Tibbets
To error is human... and I'm very human.

Post Reply