The state of the art in sapphire production: Kyocera - Single Crystal SapphireDr. Boaz Almog and Mishael Azoulay working in the group of Prof. Guy Deutscher at TAU's Raymond and Beverly Sackler School of Physics and Astronomy have developed superconducting wires using fibers made of single crystals of sapphire to be used in high powered cables. Factoring in temperature requirements, each tiny wire can carry approximately 40 times more electricity than a copper wire of the same size. They have the potential to revolutionize energy transfer, says Dr. Almog.
Sapphire fiber for superconductors
Sapphire fiber for superconductors
Innovative superconductor fibers carry 40 times more electricity
Last edited by DeltaV on Wed Sep 07, 2011 6:56 pm, edited 1 time in total.
Hmmm.
With B scaling directly with current capacity, this suggests that a machine the size of WB6 could put out 40^4 * .0006 W or ~1.5kW. Of course, no data on magnetic effects, so there might be other limits.
With the presumed 5X for the improved sphericity of different designs, this could be a home power unit with plenty to spare for your electric car!
Hmmm.
With B scaling directly with current capacity, this suggests that a machine the size of WB6 could put out 40^4 * .0006 W or ~1.5kW. Of course, no data on magnetic effects, so there might be other limits.
With the presumed 5X for the improved sphericity of different designs, this could be a home power unit with plenty to spare for your electric car!
Hmmm.
-
rjaypeters
- Posts: 869
- Joined: Fri Aug 20, 2010 2:04 pm
- Location: Summerville SC, USA
Re: Sapphire fiber for superconductors
Ouch! Dude, I was just Googling...DavidWillard wrote:You mean the same Kyocera that closed a perfectly good surface mount capacitor plant in Vancouver Washington and moved to China where they can dump their toxic waste cheaper for .2 cents a part production?
Oh, you mean the same plant that has the capacity to make such materials and treat enough heavy metal laden water for four states sitting idle?
Um. A nice power source if you ignore the the electron current drive energy. I believe this was ~ 480 KW ( ~ 40 Amps * 12,000 V) during the Beta= 1 operating condition in WB6. Still a ways to go. With some modest increase in size and higher drive voltage. I estimate that in a 30 cm diameter WB6 like machine, with a presumed 5X improvement in performance, and staying at 12,000 V drive voltage the breakeven B field would need to be ~ 10 Tesla. This ignores the inefficiencies in fusion power conversion into electrical output, thermal wall loadings, the volume inside the magrid that is dedicated to shielding and cooling (perhaps 60%), thermalizing issues , etc....KitemanSA wrote:Hmmm.
With B scaling directly with current capacity, this suggests that a machine the size of WB6 could put out 40^4 * .0006 W or ~1.5kW. Of course, no data on magnetic effects, so there might be other limits.
With the presumed 5X for the improved sphericity of different designs, this could be a home power unit with plenty to spare for your electric car!
Hmmm.
If the drive voltage is ~ 100,000 V the input power would be ~ 5 MW, the thermalization issues may be more tolerable, and the fusion rate may be ~ 100 times greater. This is getting closer, but the machine would probably melt into a pool of slag because of the thermal wall loading- perhaps in the neighborhood of 5-10 MW per square meter. P-11B fusion with associated direct conversion capturing and converting equipment might approach this at modest increased magrid size, and/ or B field strength may ease some of the thermal wall loading concerns. But, even Bussard, when pressed by an interviewer, limited very optimistic size estimates of an evolved P-11B size machine to that which might fit inside a semi truck engine compartment ( which would be extremely impressive).
PS: Actually the fusion output of WB 6 was closer to 1 mW. _ 500,000,000 neutrons per second, equates to ~ 1 billion fusions per second and 0.001 Watts. Your estimate is thus conservative. Of course the margin of error in the WB6 data is probably ~ +/- 30%.
Dan Tibbets
To error is human... and I'm very human.
DT wrote:PS: Actually the fusion output of WB 6 was closer to 1 mW. _ 500,000,000 neutrons per second, equates to ~ 1 billion fusions per second and 0.001 Watts. Your estimate is thus conservative. Of course the margin of error in the WB6 data is probably ~ +/- 30%.
Which number do you challange?Polywell-FAQ wrote:The power output of WB-6 was about 0.6 milliwatts.
This value is determined as follows:
In his Valencia Paper , Dr. Bussard reported that the WB-6 produced 1e+9 D+D fusion events per second (estimated from a 250 microsecond pulse mode, 3 neutrons sampled) during the last experiment. From other reports, prior high power experiments showed consistent numbers.
Each D+D fusion event yields 3.66 MeV of energy.
There are 6.24E+18 electron volts in a joule.
1E+9{fusions/sec} * 3.66E+6 {eV/fusion} / 6.24E+18 {eV/Joule} = 5.86E-4 {joules/second} = 0.586 milliwatts.
I was under the impression that the high power input was an artifact of the pulse nature of the WB6. It was, in essence, what it took to "dig the well and fill it". If so, the steady state might be substantially lower. After all, there would be no magnet power losses except for cooling, and the electron recirculation back to the wall would recoup most of the drive energy. So if they have solved the "nub" problem...D Tibbets wrote: Um. A nice power source if you ignore the the electron current drive energy. I believe this was ~ 480 KW ( ~ 40 Amps * 12,000 V) during the Beta= 1 operating condition in WB6. Still a ways to go. With some modest increase in size and higher drive voltage. I estimate that in a 30 cm diameter WB6 like machine, with a presumed 5X improvement in performance, and staying at 12,000 V drive voltage the breakeven B field would need to be ~ 10 Tesla. This ignores the inefficiencies in fusion power conversion into electrical output, thermal wall loadings, the volume inside the magrid that is dedicated to shielding and cooling (perhaps 60%), thermalizing issues , etc....
If the drive voltage is ~ 100,000 V the input power would be ~ 5 MW, the thermalization issues may be more tolerable, and the fusion rate may be ~ 100 times greater. This is getting closer, but the machine would probably melt into a pool of slag because of the thermal wall loading- perhaps in the neighborhood of 5-10 MW per square meter. P-11B fusion with associated direct conversion capturing and converting equipment might approach this at modest increased magrid size, and/ or B field strength may ease some of the thermal wall loading concerns. But, even Bussard, when pressed by an interviewer, limited very optimistic size estimates of an evolved P-11B size machine to that which might fit inside a semi truck engine compartment ( which would be extremely impressive).
Much engineering to be do with no real data to use.
Obviously it is all a scam.
I wasn't challenging your numbers so much as pointing out that your initial estimate of WB6 output may have been low so the subsequent predictions are mildly pessimistic/ conservative from the most likely(?) numbers. And secondly, that fusion power output is next to meaningless without also considering the input costs and other losses like bremsstrulung.
My understanding of the electron energy input is that in WB 6 during the brief ~ 0.25 second interval where the Beta= one state was achieved required ~ 40-45 amps of electron current to maintain a steady state electron population within the magrid. This was "steady state" in Bussard's opinion because the dynamics of the electron behavior occurred in the microsecond time scales. The Wiffleball does not contain electrons perfectly, there is a leakage which needed to be made up with the ~ 40 amps of electron current from the electron guns. The electron lifetimes with primary magnetic confinement was ~ 0.2 to 0.3 ms, with recirculation this was increased to ~ 3 ms. Note this is dismal confinement compared to charged particle confinement in large Tokamaks, which might reach many hundreds of seconds. The important distinction is because of density differences, and to a debated extent the thermalizing effects, this confinement time is adequate for the fusion rate (square of the density) to surpass the energy loss rates. Also note that the more efficient electrostatic containment of the ions in comparison to the magnetic + recirculation confinement of the electrons results in minimal ion losses before most of them fuse. I have the impression that only ~ 20-30 ms ion confinement time is needed for high fuel burn up ratios (compared to upwards of ~ 1000 seconds need in a Tokamak).
In WB6 the electrons were traveling at an average speed of ~ 10 million M/s. With a (less than) 30 cm diameter, and a claimed lifetime of ~ 100,000 passes, this equates into ~ 100,000 passes * 0.30 M per pass = ~ 30,000 M traveled for each average electron before loss.
30,000 M / 10 million M/s = ~ 3 millisecond lifetimes for a single average electron with recirculation. If you multiply by the number of electrons in a coulomb of charge (1 amp) there are ~ 6 *10^19 electrons/M3 =~per amp. Divided by a lifetime of ~ 0.003 s gives an electron population of ~ 2 * 10^17 electrons per amp per cubic meter. Multiply by 40 amps and 0.03 M^3 and the numbers should work out to the reported electron density in WB6 of ~ 10^ 13 / cm ^3 or ~ 10^18 electrons / M^3. It all fits together (this assumes that I didn't screw up the math or assumptions too much).
In any case there needs to be a constant influx of new electrons to replace escaped electrons. Recirculation conserves electrons but only to a modest degree. What is important in this regard is that most of the electrons that do escape give up most of their energy to the magrid, so the energy losses are mitigated. This is an example of a direct conversion grid.
In the EMC2 patent application it is mentioned that electron losses through the magnetic fields may be ~ 0.01 to 0.1 times the rate compared to the cusp losses (with recirculation?). I have the indefensible impression that the cusp losses in WB6 with recirculation was ~100 X that of the cross field diffusion, and if recirculation can be improved enough (through elimination of nubs, etc) then this ratio may be reduced to ~ 10 or even to 1.
With the above assumptions, the absolute minimum electron losses would be that due to cross field transport and the cusp losses would be very close to zero (which might be bad if discarding up scattered electrons is important). Using the 100,000 passes before loss in WB6 could, with near 100% recirculation (as opposed to the ~ 90% efficient recirculation in WB6), result in the number of electron passes being upwards of ~ 1-10 million. This would reduce the replacement current necessary from ~ 40 amps to ~ 4 or even 0.4 amps. The effects this would have on the engineering aspects (Q and wall thermal loading) would be tremendous. If a WB100 operated at ~ 3 meter diameter and ~ 80,000 volts, and 10 Tesla) extrapolating from WB6 would imply a electron input power of > 300 MW. A ten fold improvement in electron containment and/or recirculation would reduce this to ~ 30 MW. The fusion output could be as much as ~ several GW (~ 100 billion scaling from the B field and volume, and perhaps 20 X scaling from elevated drive voltage). Why go to the larger drive voltages? There is an input energy and waste heat penalty while the fusion output might be too much to handle from a thermal wall loading perspective. On the plus side, these voltages may be needed in the larger (and higher ion density) machines to prevent ion thermalization, though there has even been some suggestion that ion thermalization is not a show stopper, at least for D-D fusion.
If the scaling holds the fusion capabilities of these machines could quickly outstrip the engineering constraints.
The 3-5 X improvement that Bussard suggested for higher order polyhedra, multiplied by any recirculation gains could easily result in a 10 fold or greater gain in the equivalent energy balance of a WB6 sized machine, even before the B^4 r^3 output scaling is applied.
This has two implications. There is some wiggle room in the predicted performance range that can result in positive Q's. And, also, it is somewhat scarey what these machines could do. Only engineering bottle necks limits the machines. Of course this requires that there are no fatal engineering constraints (like vacuum pumping), and that the system works at all...
PS: What will be (sort of ) free is the magnetic field energy input requirements if superconductors are used. Assuming 12 V batteries in parallel ( not banks hooked up in series) driving the copper wire magnets in WB 6, I guesstimate that ~ 20,000 Watts of power was used to generate the ~ 0.1 T fields in WB6. Scaling up to ~ 10 Tesla and ignoring other considerations would result in ~ 2 MW being needed to power the copper magnets. This increased cost can be mostly ignored with superconductors. This is how JET figured they reached ~ 80% of break even. They ignored the several hundred MW of power needed to maintain their copper electromagnets.
Dan Tibbets
My understanding of the electron energy input is that in WB 6 during the brief ~ 0.25 second interval where the Beta= one state was achieved required ~ 40-45 amps of electron current to maintain a steady state electron population within the magrid. This was "steady state" in Bussard's opinion because the dynamics of the electron behavior occurred in the microsecond time scales. The Wiffleball does not contain electrons perfectly, there is a leakage which needed to be made up with the ~ 40 amps of electron current from the electron guns. The electron lifetimes with primary magnetic confinement was ~ 0.2 to 0.3 ms, with recirculation this was increased to ~ 3 ms. Note this is dismal confinement compared to charged particle confinement in large Tokamaks, which might reach many hundreds of seconds. The important distinction is because of density differences, and to a debated extent the thermalizing effects, this confinement time is adequate for the fusion rate (square of the density) to surpass the energy loss rates. Also note that the more efficient electrostatic containment of the ions in comparison to the magnetic + recirculation confinement of the electrons results in minimal ion losses before most of them fuse. I have the impression that only ~ 20-30 ms ion confinement time is needed for high fuel burn up ratios (compared to upwards of ~ 1000 seconds need in a Tokamak).
In WB6 the electrons were traveling at an average speed of ~ 10 million M/s. With a (less than) 30 cm diameter, and a claimed lifetime of ~ 100,000 passes, this equates into ~ 100,000 passes * 0.30 M per pass = ~ 30,000 M traveled for each average electron before loss.
30,000 M / 10 million M/s = ~ 3 millisecond lifetimes for a single average electron with recirculation. If you multiply by the number of electrons in a coulomb of charge (1 amp) there are ~ 6 *10^19 electrons/M3 =~per amp. Divided by a lifetime of ~ 0.003 s gives an electron population of ~ 2 * 10^17 electrons per amp per cubic meter. Multiply by 40 amps and 0.03 M^3 and the numbers should work out to the reported electron density in WB6 of ~ 10^ 13 / cm ^3 or ~ 10^18 electrons / M^3. It all fits together (this assumes that I didn't screw up the math or assumptions too much).
In any case there needs to be a constant influx of new electrons to replace escaped electrons. Recirculation conserves electrons but only to a modest degree. What is important in this regard is that most of the electrons that do escape give up most of their energy to the magrid, so the energy losses are mitigated. This is an example of a direct conversion grid.
In the EMC2 patent application it is mentioned that electron losses through the magnetic fields may be ~ 0.01 to 0.1 times the rate compared to the cusp losses (with recirculation?). I have the indefensible impression that the cusp losses in WB6 with recirculation was ~100 X that of the cross field diffusion, and if recirculation can be improved enough (through elimination of nubs, etc) then this ratio may be reduced to ~ 10 or even to 1.
With the above assumptions, the absolute minimum electron losses would be that due to cross field transport and the cusp losses would be very close to zero (which might be bad if discarding up scattered electrons is important). Using the 100,000 passes before loss in WB6 could, with near 100% recirculation (as opposed to the ~ 90% efficient recirculation in WB6), result in the number of electron passes being upwards of ~ 1-10 million. This would reduce the replacement current necessary from ~ 40 amps to ~ 4 or even 0.4 amps. The effects this would have on the engineering aspects (Q and wall thermal loading) would be tremendous. If a WB100 operated at ~ 3 meter diameter and ~ 80,000 volts, and 10 Tesla) extrapolating from WB6 would imply a electron input power of > 300 MW. A ten fold improvement in electron containment and/or recirculation would reduce this to ~ 30 MW. The fusion output could be as much as ~ several GW (~ 100 billion scaling from the B field and volume, and perhaps 20 X scaling from elevated drive voltage). Why go to the larger drive voltages? There is an input energy and waste heat penalty while the fusion output might be too much to handle from a thermal wall loading perspective. On the plus side, these voltages may be needed in the larger (and higher ion density) machines to prevent ion thermalization, though there has even been some suggestion that ion thermalization is not a show stopper, at least for D-D fusion.
If the scaling holds the fusion capabilities of these machines could quickly outstrip the engineering constraints.
The 3-5 X improvement that Bussard suggested for higher order polyhedra, multiplied by any recirculation gains could easily result in a 10 fold or greater gain in the equivalent energy balance of a WB6 sized machine, even before the B^4 r^3 output scaling is applied.
This has two implications. There is some wiggle room in the predicted performance range that can result in positive Q's. And, also, it is somewhat scarey what these machines could do. Only engineering bottle necks limits the machines. Of course this requires that there are no fatal engineering constraints (like vacuum pumping), and that the system works at all...
PS: What will be (sort of ) free is the magnetic field energy input requirements if superconductors are used. Assuming 12 V batteries in parallel ( not banks hooked up in series) driving the copper wire magnets in WB 6, I guesstimate that ~ 20,000 Watts of power was used to generate the ~ 0.1 T fields in WB6. Scaling up to ~ 10 Tesla and ignoring other considerations would result in ~ 2 MW being needed to power the copper magnets. This increased cost can be mostly ignored with superconductors. This is how JET figured they reached ~ 80% of break even. They ignored the several hundred MW of power needed to maintain their copper electromagnets.
Dan Tibbets
To error is human... and I'm very human.
Thin SC layer on sapphire disc, demonstrating quantum trapping:
http://www.quantumlevitation.com/levita ... ction.html
http://www.quantumlevitation.com/levita ... ction.html