They are not exactly transmitting the data by injecting it into the line - that method is limited by bandwidth, SNR and regulations to a few megabits per second. They are using a MASER to transmit a microwave RF signal that uses the magnetic field AROUND the powerline as a waveguide.
This is bogus.
Electromagnetic fields do not interact with each other. They couple with charged particles, which couple back to the field. A "magnetic field" wouldn't be acting as the waveguide - the wire would. Over a single wire with no interference, this would work quite well. However, the actual power grid does not meet this specification.
This will not work all the way down to your socket but powerlines hanging from pylons could be used by utilities to compete with fiber.
The problems with this include, but are not limited to:
Reflections from ends of wires on the grid.
Reflections from any split on the grid (place where one wire forks off from the primary wire).
Inductive coupling to... everything.
Noise injection from... everything.
Power networks were not designed with eliminating reflections in mind, and were not designed for noise rejection. Cable - which is shielded coaxial cable - _is_ designed for both of these, and so has much higher ultimate bandwidth limits. It too behaves like a waveguide at high enough frequencies, and it's shielded. Optical fiber can beat both hands-down - it's a waveguide for an EM signal with a frequency _many_ orders of magnitude higher than radio and microwave signals.
Summary: Bandwidth limits and signal quality problems make this a not-very-useful technology.
But what he is working on has more to do with the magnetic portion of the EM spectrum than the E part of it.
Click on "User Info", and read my previous post about this. Electric and magnetic fields are intimately connected - you can't have either alone, if one is changing.
Of course it was around the the conductor - all electronic signals are transmitted as changes in the magnetic/electric fields around a conductor - thats basic physics.....
Um, no.
Signals in most circuits are trasmitted as a flow of current _within_ wires, driven by an electric field gradient _within_ the wires (called "voltage"). Electric fields outside the wires try to move current between the wire and anything nearby, but this is an unwanted side effect, stopped by something called "insulation". However, the electric fields also result in capacitive coupling between nearby wires, which causes something called "capacitive cross-talk". This is minimized by keeping wires far apart and minimizing the amount of parallel surface area of conducting regions.
As a side effect of the current flow, a magnetic field is set up both around and within the wires. The current flowing within the wire and the magnetic fields around the wire are intimately connected; you can't have one without the other, and they interact very strongly with each other. You can't "transmit information in magnetic fields around the wire" without interacting with currents in the wire too - the magnetic field is _caused_ by local currents in the wire. In most systems, magnetic fields are an unwanted side effect. As there is mutual inductance between any two wires in a circuit, the magnetic fields caused by current in one wire will set up currents in other wires. This is called "inductive cross-talk". It is severe only for wires that are very close to each other, or that have a particularly vulnerable geometry.
For an excellent book on the basic physics involved, I recommend "Fundamentals of Physics, Fifth Edition", by Halliday, Resnick, and Walker. Another good reference is "Physics for Scientists and Engineers, Extended Version" by Tipler.
First, I see no mention of this using a multi-frequency tunable laser, and as near as I know no such animal exist (if it did the fiber optic companies would be all over it!)
Actually, most lasers are tunable to some degree; the emission bands that they use aren't perfectly sharp, and adjusting the geometry of the cavity and/or putting in a filter lets you select the final frequency that results. Lasers sold as "tunable" lasers are lasers that have very broad emission bands, with some piece of adjustable frequency-selective hardware (like a diffraction grating) to tune it with.
Dye lasers are the most significant example of this that comes to mind.
I've never heard of DVD drives using tunable lasers, so I'm not sure where the messages on the subject are coming from. If I understand correctly, they actually adjust the focal depth of the optical head (or just move the head up and down) to select the layer that they want. (DISCLAIMER: It's been a while since I read up on DVD technology).
What do we need ACs for? I mean, what amount of information about myself do I give away by creating an account? It's not that I need any ID or anything.
Actually, you have an email address stored in your user record (as the system needs to be able to mail you your password upon request). Granted, this could be faked or set up as a Hotmail account or the like, but it will still annoy many people who are sensitive about being "tracked".
As for why anonymous users are needed at all... Once, and only once (to date) have I seen an anonymous post that was very informative that could not have been posted if the user could be traced. The user posted (in general terms) about a lawsuit that his company had been involved in, as part of a debate about whether or not closed-source can be justified at all. This sparked a very extensive discussion thread.
I would like to be able to read posts like this. IMO, especially with moderation in place, it's worth tolerating the myriad of "First Post!" and "Slashdot Sucks!" messages if it means that we can get these messages too.
Not that I wouldn't love a way to regulate this sewage. I'd suggest a pattern-matching filter, but that would be too easy to get around. For now, thresholds will do.
Most of the solar approaches you propose are actually inefficient complications. Solar cell technology is making inroads in efficiency, cost, durability, etc. The FAQ for alt.solar.photovoltaics: http://www.means.net/~mschwarz/solar-faq.html
I'm afraid that neither your response nor your faq address the concerns I was addressing in my post.
The issues that I was addressing were 1) cost of maintainence of space-based and ground-based solar power plants, and 2) cost effectiveness of space-based solar power plants.
The primary issues for 1) are durability and cost of replacement. Once upon a time, solar cells were fragile. As you point out, this has changed. However, a solar "shingle" is a lot more expensive than a sheet of aluminum the same size, and not significantly more durable. Reflectors are still considerably more cost-effective per unit area per unit lifetime than any _presently_practial_ type of photovoltaic panel. Thus, concentrators are cost-effective to use.
Note that I say "presently practical". I'll get back to this below.
Additionally, _presently_available_ solar panels have rather horrid efficiency - about 15% for the best mass-produced panels in real-world conditions, and less than that for many of them (source: a recent overview of photovoltaic technologies in (if I remember correctly) Scientific American). Heat plants, on the other hand, can have efficiencies substantially higher. This makes concentrator/heat plant systems more efficient than _present_ photovoltaics, for many classes of system.
Note that I'm again stressing "present". Photovoltaics is indeed a fascinating field that continues to advance. The arguments that I present may well be invalidated when lighter, cheaper, and more efficient panels are produced. I'll come back to this below. However, I am discussing plants that can be built with today's technology, or with technology that is likely to be in mass-production within the next few years.
The second problem that I was addressing was cost concerns for a solar power satellite.
Micrometeorites are very destructive. They only damage tiny areas per strike, but will get through most coatings on solar panels, even ruggedized ones. Thus, the lifetime of panels (time before enough of their area is destroyed to make them useless) doesn't vary much with their durability, except for panels that are prohibitively heavy (cost of putting the satellite up is directly proportional to weight).
The natural solution to this is to use thin panels. However, reflectors (aluminized mylar) can be made much thinner than _presently_practical_ photovoltaic panels. This again favours a concentrator scheme, like the one I described. Cost per unit area is reduced, and lifetime isn't too strongly affected (possibly even increased, _if_ reflectors function better when badly damaged than photovoltaic panels).
Note the "presently practical" again. There is a panel technology that would invalidate all of my arguments above:
Thin-film amorphous or polycrystalline panels on mylar.
Unfortunately, while this technology is being poked at in the lab, cell efficiencies are at present miserable, even compared to other photovoltaic technologies. This will almost certainly improve with research, but IMO thin-film panels aren't likely to be competitive for at least several years, and probably longer.
These are the justifications for my arguments above. I am trying to describe solar plants that could be built (in production scale) with _today's_ technology that would be practical.
Also, you either completely skipped over or dismissed out of hand my methanol suggestion. Nothing beats a flammable liquid for energy storage density (or very few practial things), and we already have most of the infrastructure to deal with it.
Is there any good reason why fusion can't be achieved non-thermally, by bringing nuclei (or atoms) together with sufficiently high velocity ? For example, why not two beams of hydrogen ions meeting at sufficiently high energy ?
You could indeed produce fusion by firing two particle beams at each other. The problem is that particle accelerators are very inefficient; you'd wind up putting in far more energy to accelerate the particles than you'd get back from fusion events. Particle beams are also very sparse; you might only get a handful of nuclei even hitting each other, out of all of the nuclei in each of the intersecting beams. Lastly, when two nuclei collide, most of the time they bounce off of each other. Only a direct hit produces fusion (or other interactions, for higher-energy accelerators).
You can eliminate the tenuousness problems by firing a single beam at a target made of lithium or frozen hydrogen, but the other two problems remain.
Now, the fundamental issue - the "temperature" of something _is_ just the average energy of all of the particles in the object/gas/whatever. By accelerating particles to energies sufficient to overcome electrostatic repulsion... you've increased the temperature, as that's what temperature is a measure of.
As an interesting bit of trivia, fusion produced by a particle beam striking a target _is_ used in some applications. Certain types of nuclear warhead, for instance, use this type of device to produce a precisely-timed neutron pulse used to trigger fission. However, this is still done at a horrible energy cost (the object is to produce neutrons, not to get out more energy than they put in).
Just say energy gets abundant and cheap. And everybody goes zooming around with electric motors. I doubt you're going to strap on a fridge to your vehicle to pump heat out to space. So the resulting heat generated would be distributed to ambient. Not nicely organised heat.
Your proposal would mean airconditioning our environments, e.g cities, parks etc, leaving everything else rather warm.
The good news is that this would still eliminate _most_ heat pollution. By far the largest sources of heat are cities (consuming electrical and fossil-fuel energy), and power plants (producing electrical energy at imperfect efficiency). Both of these can be cooled without *too* much additional expenditure (electricity and fossil fuels double in price due to taxes for the increased infrastructure needed).
Build this into the infrastructure in smaller population centers like towns, and you're left with highways and small villages as heat sources. Somewhere between 90% and 99% of heat production is accounted for, which dramatically reduces the environmental impact of heat pollution.
From space, the 1300 degree radiators would look like yellow-white glowing patches that show up even during the day. The 230 degree patches wouldn't show in the visible spectrum (they'd glow quite brightly in the intermediate IR spectrum, though).
You raise several valid concerns; however, there are solutions to at least some of the problems you present.
Firstly, atmospheric attenuation isn't *that* much. Even a factor of two power loss still leaves 500 W/m^2, which is respectable.
Secondly, regarding storm damage and cost of solar power facilities. If a solar facility was built as an array of photovoltaic panels, then it would indeed be very expensive to produce and to maintain (at least until very good thin-film photovoltaics get here). However, there are a couple of workarounds.
Both workarounds involve using concentrators. The idea is that mirrors are quite cheap, especially if they don't have to be of laboratory quality. Run a set of metal troughs aligned to be parallel to the sun's course (on average). Make them out of aluminum so that they won't rust, and put drainage holes in them in case it rains. Run a strip of photovoltaic cells down the channel, suspended at the focus of the mirror. This reduces cost substantially, and may also increase efficiency (several types of cell work better in brighter light). A variant of this dispenses with photovoltaics and runs black pipes down the channels, again suspended at the focal point. Put a working fluid in the pipes, and run a conventional heat engine off of them. You need somewhere to dump the heat, but if you're near a body of water that's not a problem (heat pollution is another issue).
For space-based power, you can do much the same thing. Most of the area of your power satellite is made of cheap aluminized-mylar reflectors. Yes, they'll be shredded after a while, but you can replace them (or the entire satellite, if you want). At the foci of these reflectors are either photovoltaic panels or heat engines. These are small enough that you can put some micrometeorite shielding around them to extend their lifetimes.
This gives you a lighter satellite, that might even last longer, and is certainly much cheaper to replace.
The main problem with power satellites is that you have to beam the power back down to earth. This does not present a terrorism hazard - the most popular schemes use the receiving antenna array to control the phase of the beams produced from various parts of the satellite. Without this control, the satellite would emit in all directions, resulting in harmless power levels reaching the ground. This control takes a big array, and the resulting beam is targetted _at_ the array. So to fry a city, the terrorist group would first have to put an appropriate antenna on the roof of every home... Maybe offer to clean their eavestroughs while there...:)
The real problems with beaming power back are that your emitters have to be a fixed weight no matter what - which puts a lower limit on the mass of the satellite - and that you need substantial levels of microwave energy coming down on your receiving array (i.e. more than 1 kW/m^2). Any bird flying through this will get warm, and anybody camping out underneath will get warm also. They won't be flash-fried - but they will receive the equivalent of standing under light several times brighter than the sun.
Lastly, as this represents energy imported to earth, power satellites will contribute to the heat pollution problem. OTOH, the same applies to any other power source that releases energy that would otherwise not be released on earth.
The final way to (efficiently!) use solar power on earth is to grow crops and ferment them to produce methanol. This gives you a flammable liquid that can be used as fuel, mulch to supply the hydrocarbon/chemical industries, and takes carbon dioxide out of the atmosphere as fast as fuel-burning puts it into it.
The problem is that the crop method takes up a fair amount of space, and (like other forms of solare power) is limited by the amount of sunlight falling on to the collectors/fields.
Sure, if you make atoms move around really fast, some of them will fuse. There are other ways to create pressure among atoms, and they take a lot less energy than a giant donut shaped oven.
Um, no, there aren't. Not the required amount of pressure.
Even the sun doesn't produce enough pressure to physically force the nuclei together. The core of the sun is electron-degenerate - you have electrons in a strictly organized set of energy levels, with nuclei zipping around like gas particles within it.
Most of the nuclei, even in the sun's core, don't have enough energy to overcome the electrostatic repulsion between nuclei. A very few of them do - because the temperature of an object is only the _average_ temperature of the particles within it. It is very rare for a particle to gain this much energy, and so fusion reactions are extremely infrequent for a given particle, but the particles aren't going anywhere, and there are a lot of them. Thus, the sun produces a substantial amount of energy due to fusion.
On earth, there are very strong limits to the pressures we can produce stably. A steady-state system would therefore have to jack up the temperature to increase the reaction rate in order to generate more energy than it expends. You could argue that a steady-state system isn't necessary; well and good. However, when you produce a region of extreme pressure (as with inertial-confinement fusion), it doesn't last very long. You jack up the reaction rate, but the short confinement time brings your total power output back down again.
This covers all but two proposed methods of fusion.
One other method proposed is muon-catalyzed fusion. Click on "user info" above to find a message with more details about it. It uses a very sneaky method to remove most of the cost of bringing nuclei close enough together to fuse. However, the energy cost of the muons required for catalysis is far more than the energy produced by the catalyzed fusion.
The last method proposed of producing fusion is the "cold" fusion that caused such a media commotion recently.
People would very much like this style of fusion to work.
People claimed that they had gotten this style of fusion to work.
A number of countries invested and continue to invest substantial research energies into duplicating the original experiments that supposedly reliably produced energy by fusion by this method.
To date, they have failed. This, coupled with a number of extremely suspicious points in the presentations by the original discoverers, strongly suggests that the original researchers were either mistaken or actively defrauding the rest of the scientific community. Read the book entitled "Yes, we have no neutrons" for an in-depth discussion of this. Authour name escapes me at the moment.
Summary: Cold fusion appears to be bogus, despite much effort to duplicate it.
So, what method are you proposing to create the pressures necessary to produce fusion at room temperature?
I haven't really thought much about this, but I'm concerned that an unforeseen problem of fusion would be "heat pollution." Already this is a problem with today's nuclear reactors, which typically use lake or river water for cooling. Fusion offers vastly more energy than fission; will it make proportionally more heat pollution?
Heat pollution is indeed a problem, both at the reactor and in the cities where the power is used (most power eventually ends up being converted to heat). However, there is a safe place to dump the heat if it becomes a sufficiently bad problem - the sky.
Space has a black-body temperature of about 3 kelvin. Use heat pumps to concentrate your heat in a working fluid. Run the fluid through pipes. Put the pipes in mirrored channels, open at the top. Most of your heat goes into space, by blackbody emission from the pipes.
The problem with this is the rate at which heat is dumped. This is proportional to the fourth power of the temperature of your working fluid (and hence your pipes). The bad news is, you need a very hot fluid. The good news is that, because of the fourth order relation, your fluid doesn't have to be _insanely_ hot. The heat pump that performs the concentration will produce heat itself - you're working against entropy when concentrating heat like that. However, if all components are efficient enough, you should be able to dump this heat out with the rest.
This requires power. I am blithely assuming that if you have enough power plants that heat pollution is a problem big enough to do something about, you have enough power to run the heat pumps.
This doesn't work on cloudy days. I am blithely assuming that you have a heat sink large enough to store heat in until the next clear day. This would probably be a moderate-sized body of water.
Assuming that you can afford a one-hectare area per GW of heat for your emitter array (100m squared, or _roughly_ 100 yards by 100 yards), and that one quarter of this is actual emitting area (area of the pipes), your working fluid would have to be at a temperature of about 1600 kelvin, or about 1330 degrees centigrade. This is readily buildable. The only concern is the heat pump, which is working against a very strong temperature difference. Scale up the area - to, say, 1km by 1km (about 0.6 miles by 0.6 miles), and you need a temperature of about 515 kelvin, or about 242 degrees centigrade. Building the heat engine for this is much easier, though more of the resulting radiation will be absorbed by the atmosphere in that frequency band. Maintaining the array is more expensive in this scenario, too.
In conclusion, I think that heat pollution is a solvable problem. It just requires enough of an investment that we aren't likely to build heat-dumping plants until they become absolutely necessary.
Can anyone authoratively guestimate how much waste will be produced per energy unit? In other words, if the energy consumption changed the same, and all fission plants were replaced with fusion plants, would the waste production be higher or lower?
Sure.
In fact, it's easy to show that fission produces _less_ waste - the question is how much less.
DISCLAIMER: I AM MAKING LIBERAL USE OF "FERMI" ESTIMATES. NUMBERS QUOTED SHOULD BE WITHIN AN ORDER OF MAGNITUDE OF REALITY, BUT THAT'S ABOUT IT. THIS IS NOT A DETAILED ESTIMATE.
Both fission and fusion produce secondary waste, in the form of the reactor housing, which becomes radioactive. The housing won't last forever - pieces wear out, as with any device. Let's assume as a rough approximation that the entire thing has to be replaced once every 20 years, and that it is all filed as low-level radioactive waste.
Let's assume that we have a moderate-sized reactor core - a 10m cube (30 feet for the American audience). This isn't a solid structure; in a fission plant, it's a framework holding reactor rods, and either a containment vessel (American reactors) or a network of pipes (Canaadian reactors), with ample amounts of radiation shielding. Let's say it's equivalent to a shell 2m thick. This gives us a volume of about (10^3 - 6^3) = around 800 cubic metres. Assuming a density of around 5 (metal, concrete, and lead), we get about 4000 tonnes of waste every 20 years, or an average of 200 tonnes of waste per year (1 tonne = 1000 kg = about 2200 pounds, so about the same as a US ton, before you object to my spelling).
Now, a fission reactor also produces a fair bit of high-level waste in the form of spent fuel rods. This represents extra waste that a fusion reactor doesn't produce. We have to see how much it is compared to the (roughly) 200 tonnes produced on average per year by our hypothetical reactor.
Let's assume that a reactor with a core of the quoted size produces about 500 MW of power. Let's assume that this represents about 2% of the fuel in the reactor (typical of fission reactors that don't use reprocessed fuels). Let's assume that this fuel converts mass to energy at an efficiency of about 0.1% (typical for fission reactions, IIRC). 500 MW over one year gives us about 1.5e16 joules of energy produced per year. At 100% efficiency, this would correspond to about 0.17 kg of fuel. We're operating at 0.1% efficiency, which means about 170 kg of fuel burned. This represents about 2% of the total amount of fuel, which means about 8.5 tonnes of fuel burned in the reactor per year. This fuel is filed as high-level waste when its finished burning.
So, we find that, unless I'm off by _more_ than a factor of 10 in computing the _relative_ amounts of each type of waste, and assuming that you consider high-level and low-level waste equally bad, both fusion and fission produce about the same amount of waste, most of which is just parts of the reactor that wear out and have to be stored safely.
CAVEATS:
High level waste is actually much nastier than low-level waste, in that a leak causes much more harm. Per unit radiation produced by the waste, the low-level stuff is probably orders of magnitude safer.
A fusion reactor *might* not require as massive a core structure as a fission reactor, as it doesn't have to support several tonnes of fuel rods. On the other hand, it still has to deal with cooling pipes (you're generating power by using this as a heat source). It also has to deal with pressure on the field coils caused by magnetic forces (the coils want to violenly fly apart).
I'm lumping the first layer of core shielding in with the core structure. If you forgo the first level of shielding and just put a wall of lead bricks around the reactor, you don't have to count this. Lead does not become radioactive from exposure to radiation, for the most part (it converts neutron radiation to helium by a roundabout process, which is kind of nifty).
A few notes about fission, that might be confusing people:
Why I use the 2% figure. A conventional fission reactor has two fuel sources; U235, which is present from the start, and Pu239, which is bred from U238 within the reactor (any reactor containing U238 does this whether you want it to or not). Both of these fission to produce many light byproducts. These byproducts occasionally absorb neutrons, and almost never give them out. After you've left a rod in the reactor for a while, enough of these byproducts accumulate that reactions are not self-sustaining within the fuel rod anymore. It is absorbing neutrons without contributing much energy. This is after somewhere around 2% of its mass has been fissioned. At this point, you can either throw it away or send it to a reprocessing plant, which strips out the light, absorbing components so that you can use the rod again. Nowadays, we throw the rods away and lose 98% of its mass.
Why reprocessing isn't done any more. Reprocessing isn't done any more because it involves running what amounts to high-level waste through a chemical processing plant. Despite precautions, this resulted in unacceptably high exposure to radiation for people working in the plant. Neutron radiation from the waste also transmuted anything that the waste was in contact with (including whatever you dissolve it in for processing, and so forth), producing a lot of low-level waste. The result of all of this is that reprocessing fuel rods safely is a hassle of monumental proportions. There's enough uranium lying around that we can afford to throw away 98% of it, so we do.
> Second of all, if the reaction takes place in a > vacuum, then not as much heat is required to > sustain the reaction as stated in the article.
Um, not quite. It takes a very hefty amount of energy to heat the plasma to the required temperatures - about 1.0e8 degrees centigrade. At those temperatures, plasma cools quite quickly by radiative emission, regardless of whether it's in vacuum or not. This is one of the reasons that the reaction only lasts nanoseconds.
> Only about 100,000C is needed before the atoms > can become ionized and begin to transfer energy.
Um, not quite. The plasma doesn't just have to be plasma - it has to be plasma of high enough energy that the nuclei can overcome their mutual repulsion. Hence, the 1.0e8 degrees figure for most terrestrial reactors (stars use lower temperatures, higher densities, and longer confinement time).
> At this stage things get a little fuzzy, because > some of the normal laws of physics don't work > because the atoms are in a plasma state.
Actually, they work just fine. You're just dealing with a conducting gas.
> The good news is the are electrically charged > and can be controlled by magnets(as mentioned in > the article). This should help keep the reaction > in one spot.
Correct. This is the principle behind magnetic confinement fusion. Other approaches exist. Both the approach described in the article and laser fusion rely on inertial confinement instead - the particle is suddenly compressed, and by the time it disperses the reaction has finished.
For more information on fusion and many other interesting things, browse your local university's bookstore for physics texts. My personal favourite is "Physics for Scientists and Engineers, Extended Version", bt Paul A. Tipler.
> The simple fact is that the only known way to > produce fusion is at extremely high > temperatures. No one has suggested a meachanism > (that works!) that would allow otherwise, > possibly no one ever will.
Actually, muon-catalyzed fusion works at much lower temperatures (something like 800 degrees in the test rigs I heard about).
This is completely different from "cold" fusion. Completely different principles, well-understood, and well-tested.
The idea is that if you displace an electron in a hydrogen _molecule_ with a muon (its heavier cousin), the molecule will suddenly become much smaller, because the wavelength of a muon is much smaller than the wavelength of an electron. The hydrogen nuclei are now close enough to have a significant probability of fusion by tunnelling.
The problem with this is that muons are energy-expensive to produce, and don't stick around for very long (they decay into an electron, a muon neutrino, and an electron antineutrino). The energy released by the fusion reactions that the muons catalyze has so far been far, far less than the energy required to produce the muons in the first place.
> the "hot fusion" variation described in the > article seems like the only way to go right now. > Once they lick the symmetricity problem - > possibly by setting up some positive feedback in > the plasma to keep the reaction shape spherical > in the crucial nanoseconds - there should be > more progress.
It turns out that symmetry is extremely difficult to achieve.
This approach to fusion has been tried before - there is a whole class of fusion reactor that tries to induce fusion by running strong electric currents through a plasma (which is what this is - you just get your plasma by zapping a pellet).
From the first approximation, this looks very good - the magnetic fields created by the current flow act to compress the plasma, bringing it closer to satisfying the Lawson criterion. The problem is that variations in the current and in the density of the plasma/pellet cause assymetries, which tend to magnify themselves (the electric and magnetic effects of turbulence cause more turbulence). Because of the way that plasma behaves, there is no magical way to elminate this.
All fusion approaches encounter this problem. One approach to solutions - the one that it looks like they're using here - is brute force. It takes more time for turbulence to muck up a larger volume of plasma, which means that a larger device can come closer to satisfying the Lawson criterion. Also, dumping more energy into the plasma can help; fusion in conventional reactors occurs only among the hottest particles at the tail-end of the temperature distribution. Greater average energy means more particles of sufficient energy to cause fusion.
A second approach is to modify the design of the reactor to either reduce turbulence or to be less sensitive to turbulence. For magnetic confinement reactors, a lot of study has gone into different field geometries that stifle certain types of turbulence. For electrical discharge reactors - like the one described in the article - different electrode geometries have been tried. However, in the case of electrical reactors, this hasn't been enough to match the capabilities of magnetic confinement and inertial confinement (laser) fusion.
A hybrid reactor was recently built that used an electrical discharge to produce a hot plasma that was used as an X-ray source, which in turn was used to heat a fuel pellet in a chamber that reflected X-rays. This proved to be a very promising approach; for details, check back issues of (I think) Scientific American. This does not appear to be the approach described in this article, however.
Laser fusion's main problems aren't turbulence, which is why I haven't mentioned it much in this response.
> I think we'll need fusion before we see a base > on mars. How else would they get all the power? > AFAIK mars has quite a few clouds and storms > regularly, so that would rule out solar as a > sole source of power
AFAIK, weather on mars is mostly clear. That's why we get such nice views of the surface via telescopes. The atmosphere is much thinner than Earth's, which means that dense clouds would be difficult to produce. Also, Mars's ambient temperature is below the freezing point of water. Dust storms happen, but aren't terribly frequent in most locations AFAIK.
Solar power would work fine most of the time, and power storage in batteries or fuel cells would be adequate during the occasional dust storms.
> More seriously - since you can build H-bombs > that get a *very* high portion of their yield > from fusion, as opposed to fission, there > probably wouldn't be that much radiation to > worry about in the first place.
It turns out that there would be.
Fusion reactions, like most other nuclear reactions, produce copious quantities of neutron radiation. This would make anything nearby (in this case, the surrounding rock) radioactive by transmutation.
The main benefit of fusion power isn't that it "doesn't produce radiation" - it does. It's that fuel is virtually unlimited. You also don't produce any _primary_ waste - spent fuel rods and the like. Just secondary waste (reactor structures made radioactive by transmutation).
That is a pre-announcement typical of what Microsoft would do... if you don't have a product to match, simply announce a few months early.
Do you see GeForce boards on the shelves? Do you see Playstation 2s on the shelves?
They've been conversation topics for months, but all either has now is alpha test hardware. It becomes difficult to see what point you are trying to make, given that.
It says this here card can only do 15 million triangles per second. Playstation 2 can do 75 million.
Not if it can only transform 36 million polys per second, it can't (sustained transformation figure from an older slashdot article).
Based on all of the other numbers in that article, I suspect that they dropped a decimal point in the "75 million polys rendered" figure. That, or they're talking about flat-shaded untextured untransformed polygons.
Up till this point fill rate has been all that mattered for 3D chip makers, providing the capability of allowing higher resolutions, more rendering passes (for different visual effects) and higher frame rates. This is the first (consumer level?) chip to add transformation and lightning to the 3D chip, thus offloading these duties from the CPU.
"Consumer level" is correct. High-end graphics workstations have been doing this for several years; in fact, the entire OpenGL pipeline has been in hardware for quite a while. Check out 3dlab's high-end boards for examples, or take a look at their competitors. These tend to be 64-bit PCI boards in the $2,000-$3,000 range.
The consumer graphics manufacturers have been making noise about using geometry processing for a while now, but have only recently gotten around to it. In that market, yes it could be called revolutionary (in that it substantially changes game design).
This is indeed a nice chip; however, it has competition.
3dfx is rolling out another chip, as people have been talking about for a while. It is rumoured to be at 0.22 micron too, and will have hardware geometry processing.
S3 already rolled out a new chip - at 0.18 micron. It too has four texel engines and hardware geometry processing.
IMO, the S3 chip is actually the one to worry about. Architecture may or may not be great, but at 0.18 micron it may outperform nVidia and 3dfx's offerings just on linewidth. ATI did something similar when it rolled out the Rage 128, if you recall.
What I'm waiting for is the release of the GeForce or (insert name of 3dfx's offering here) at 0.18 micron. However, I'll probably be waiting a while.
The obvious question from one who doesn't follow the 3-D chipset world closely: what's 3dfx's answer to this chipset or has nvidia kicked there buts.
The Voodoo 4 will be coming out around Christmas, and it will have hardware geometry as well. Rumour has it that it too will be at 0.22 micron instead of 0.18. I don't remember the name of the chipset off-hand.
S3 has also rolled out a new chip, with four pipelines and hardware geometry, at 0.18 micron. Check Sharkey Extreme for details.
Also, I've heard some reports that the PlayStation II will beat the living daylights out of PIII's loaded with then recent and most modern 3d accels. Even with this kind of chip, and most likely other chips to follow from nVidia's competitors, does this still hold true? Will the PlayStation II live up to the hype?
No, but it won't sink either. See my previous response on this subject (check my user info to find the post).
This just makes me sit back and wonder - is the Playstation 2 now history? Yeah, I know, diffy platform and all that, but still. Leaving out the Emotion Engine or whatever it is called, it seems to me the PS2 is now a relic in terms of what it is delivering for graphics. Granted we are not seeing any numbers, but still.
The Playstation II has a modified R10000 processor with very hefty floating point extensions - it won't have much of a problem doing geometry transformations. IMO, it will probably be about on par with the graphics cards floating around at the time of its release. It won't leave them in the dust, but neither will it be left in the dust.
OTOH, a friend in the gaming industry says that the Playstation II has architectural problems that might degrade performance (low system bus bandwidth, among other things). We'll see what happens when it ships.
"hands-on tests" means prototype hardware already available, so this should be out fairly soon. None too soon, as S3 pulled ATI's trick and is coming out early with a chip at 0.18.
Slashdot Mirror
This is bogus.
Electromagnetic fields do not interact with each other. They couple with charged particles, which couple back to the field. A "magnetic field" wouldn't be acting as the waveguide - the wire would. Over a single wire with no interference, this would work quite well. However, the actual power grid does not meet this specification.
This will not work all the way down to your socket but powerlines hanging from pylons could be used by utilities to compete with fiber.
The problems with this include, but are not limited to:
Power networks were not designed with eliminating reflections in mind, and were not designed for noise rejection. Cable - which is shielded coaxial cable - _is_ designed for both of these, and so has much higher ultimate bandwidth limits. It too behaves like a waveguide at high enough frequencies, and it's shielded. Optical fiber can beat both hands-down - it's a waveguide for an EM signal with a frequency _many_ orders of magnitude higher than radio and microwave signals.
Summary: Bandwidth limits and signal quality problems make this a not-very-useful technology.
But what he is working on has more to do with the magnetic portion of the EM spectrum than the E part of it.
Click on "User Info", and read my previous post about this. Electric and magnetic fields are intimately connected - you can't have either alone, if one is changing.
Of course it was around the the conductor - all electronic signals are transmitted as changes in the magnetic/electric fields around a conductor - thats basic physics.....
Um, no.
Signals in most circuits are trasmitted as a flow of current _within_ wires, driven by an electric field gradient _within_ the wires (called "voltage"). Electric fields outside the wires try to move current between the wire and anything nearby, but this is an unwanted side effect, stopped by something called "insulation". However, the electric fields also result in capacitive coupling between nearby wires, which causes something called "capacitive cross-talk". This is minimized by keeping wires far apart and minimizing the amount of parallel surface area of conducting regions.
As a side effect of the current flow, a magnetic field is set up both around and within the wires. The current flowing within the wire and the magnetic fields around the wire are intimately connected; you can't have one without the other, and they interact very strongly with each other. You can't "transmit information in magnetic fields around the wire" without interacting with currents in the wire too - the magnetic field is _caused_ by local currents in the wire. In most systems, magnetic fields are an unwanted side effect. As there is mutual inductance between any two wires in a circuit, the magnetic fields caused by current in one wire will set up currents in other wires. This is called "inductive cross-talk". It is severe only for wires that are very close to each other, or that have a particularly vulnerable geometry.
For an excellent book on the basic physics involved, I recommend "Fundamentals of Physics, Fifth Edition", by Halliday, Resnick, and Walker. Another good reference is "Physics for Scientists and Engineers, Extended Version" by Tipler.
Actually, most lasers are tunable to some degree; the emission bands that they use aren't perfectly sharp, and adjusting the geometry of the cavity and/or putting in a filter lets you select the final frequency that results. Lasers sold as "tunable" lasers are lasers that have very broad emission bands, with some piece of adjustable frequency-selective hardware (like a diffraction grating) to tune it with.
Dye lasers are the most significant example of this that comes to mind.
I've never heard of DVD drives using tunable lasers, so I'm not sure where the messages on the subject are coming from. If I understand correctly, they actually adjust the focal depth of the optical head (or just move the head up and down) to select the layer that they want. (DISCLAIMER: It's been a while since I read up on DVD technology).
What do we need ACs for? I mean, what amount of information about myself do I give away by creating an account? It's not that I need any ID or anything.
Actually, you have an email address stored in your user record (as the system needs to be able to mail you your password upon request). Granted, this could be faked or set up as a Hotmail account or the like, but it will still annoy many people who are sensitive about being "tracked".
As for why anonymous users are needed at all... Once, and only once (to date) have I seen an anonymous post that was very informative that could not have been posted if the user could be traced. The user posted (in general terms) about a lawsuit that his company had been involved in, as part of a debate about whether or not closed-source can be justified at all. This sparked a very extensive discussion thread.
I would like to be able to read posts like this. IMO, especially with moderation in place, it's worth tolerating the myriad of "First Post!" and "Slashdot Sucks!" messages if it means that we can get these messages too.
Not that I wouldn't love a way to regulate this sewage. I'd suggest a pattern-matching filter, but that would be too easy to get around. For now, thresholds will do.
The FAQ for alt.solar.photovoltaics: http://www.means.net/~mschwarz/solar-faq.html
I'm afraid that neither your response nor your faq address the concerns I was addressing in my post.
The issues that I was addressing were 1) cost of maintainence of space-based and ground-based solar power plants, and 2) cost effectiveness of space-based solar power plants.
The primary issues for 1) are durability and cost of replacement. Once upon a time, solar cells were fragile. As you point out, this has changed. However, a solar "shingle" is a lot more expensive than a sheet of aluminum the same size, and not significantly more durable. Reflectors are still considerably more cost-effective per unit area per unit lifetime than any _presently_practial_ type of photovoltaic panel. Thus, concentrators are cost-effective to use.
Note that I say "presently practical". I'll get back to this below.
Additionally, _presently_available_ solar panels have rather horrid efficiency - about 15% for the best mass-produced panels in real-world conditions, and less than that for many of them (source: a recent overview of photovoltaic technologies in (if I remember correctly) Scientific American). Heat plants, on the other hand, can have efficiencies substantially higher. This makes concentrator/heat plant systems more efficient than _present_ photovoltaics, for many classes of system.
Note that I'm again stressing "present". Photovoltaics is indeed a fascinating field that continues to advance. The arguments that I present may well be invalidated when lighter, cheaper, and more efficient panels are produced. I'll come back to this below. However, I am discussing plants that can be built with today's technology, or with technology that is likely to be in mass-production within the next few years.
The second problem that I was addressing was cost concerns for a solar power satellite.
Micrometeorites are very destructive. They only damage tiny areas per strike, but will get through most coatings on solar panels, even ruggedized ones. Thus, the lifetime of panels (time before enough of their area is destroyed to make them useless) doesn't vary much with their durability, except for panels that are prohibitively heavy (cost of putting the satellite up is directly proportional to weight).
The natural solution to this is to use thin panels. However, reflectors (aluminized mylar) can be made much thinner than _presently_practical_ photovoltaic panels. This again favours a concentrator scheme, like the one I described. Cost per unit area is reduced, and lifetime isn't too strongly affected (possibly even increased, _if_ reflectors function better when badly damaged than photovoltaic panels).
Note the "presently practical" again. There is a panel technology that would invalidate all of my arguments above:
Thin-film amorphous or polycrystalline panels on mylar.
Unfortunately, while this technology is being poked at in the lab, cell efficiencies are at present miserable, even compared to other photovoltaic technologies. This will almost certainly improve with research, but IMO thin-film panels aren't likely to be competitive for at least several years, and probably longer.
These are the justifications for my arguments above. I am trying to describe solar plants that could be built (in production scale) with _today's_ technology that would be practical.
Also, you either completely skipped over or dismissed out of hand my methanol suggestion. Nothing beats a flammable liquid for energy storage density (or very few practial things), and we already have most of the infrastructure to deal with it.
Is there any good reason why fusion can't be achieved non-thermally, by bringing nuclei (or atoms) together with sufficiently high velocity ? For example, why not two beams of hydrogen ions meeting at sufficiently high energy ?
You could indeed produce fusion by firing two particle beams at each other. The problem is that particle accelerators are very inefficient; you'd wind up putting in far more energy to accelerate the particles than you'd get back from fusion events. Particle beams are also very sparse; you might only get a handful of nuclei even hitting each other, out of all of the nuclei in each of the intersecting beams. Lastly, when two nuclei collide, most of the time they bounce off of each other. Only a direct hit produces fusion (or other interactions, for higher-energy accelerators).
You can eliminate the tenuousness problems by firing a single beam at a target made of lithium or frozen hydrogen, but the other two problems remain.
Now, the fundamental issue - the "temperature" of something _is_ just the average energy of all of the particles in the object/gas/whatever. By accelerating particles to energies sufficient to overcome electrostatic repulsion... you've increased the temperature, as that's what temperature is a measure of.
As an interesting bit of trivia, fusion produced by a particle beam striking a target _is_ used in some applications. Certain types of nuclear warhead, for instance, use this type of device to produce a precisely-timed neutron pulse used to trigger fission. However, this is still done at a horrible energy cost (the object is to produce neutrons, not to get out more energy than they put in).
Just say energy gets abundant and cheap. And everybody goes zooming around with electric motors. I doubt you're going to strap on a fridge to your vehicle to pump heat out to space. So the resulting heat generated would be distributed to ambient. Not nicely organised heat.
Your proposal would mean airconditioning our environments, e.g cities, parks etc, leaving everything else rather warm.
The good news is that this would still eliminate _most_ heat pollution. By far the largest sources of heat are cities (consuming electrical and fossil-fuel energy), and power plants (producing electrical energy at imperfect efficiency). Both of these can be cooled without *too* much additional expenditure (electricity and fossil fuels double in price due to taxes for the increased infrastructure needed).
Build this into the infrastructure in smaller population centers like towns, and you're left with highways and small villages as heat sources. Somewhere between 90% and 99% of heat production is accounted for, which dramatically reduces the environmental impact of heat pollution.
From space, the 1300 degree radiators would look like yellow-white glowing patches that show up even during the day. The 230 degree patches wouldn't show in the visible spectrum (they'd glow quite brightly in the intermediate IR spectrum, though).
Firstly, atmospheric attenuation isn't *that* much. Even a factor of two power loss still leaves 500 W/m^2, which is respectable.
Secondly, regarding storm damage and cost of solar power facilities. If a solar facility was built as an array of photovoltaic panels, then it would indeed be very expensive to produce and to maintain (at least until very good thin-film photovoltaics get here). However, there are a couple of workarounds.
Both workarounds involve using concentrators. The idea is that mirrors are quite cheap, especially if they don't have to be of laboratory quality. Run a set of metal troughs aligned to be parallel to the sun's course (on average). Make them out of aluminum so that they won't rust, and put drainage holes in them in case it rains. Run a strip of photovoltaic cells down the channel, suspended at the focus of the mirror. This reduces cost substantially, and may also increase efficiency (several types of cell work better in brighter light).
A variant of this dispenses with photovoltaics and runs black pipes down the channels, again suspended at the focal point. Put a working fluid in the pipes, and run a conventional heat engine off of them. You need somewhere to dump the heat, but if you're near a body of water that's not a problem (heat pollution is another issue).
For space-based power, you can do much the same thing. Most of the area of your power satellite is made of cheap aluminized-mylar reflectors. Yes, they'll be shredded after a while, but you can replace them (or the entire satellite, if you want). At the foci of these reflectors are either photovoltaic panels or heat engines. These are small enough that you can put some micrometeorite shielding around them to extend their lifetimes.
This gives you a lighter satellite, that might even last longer, and is certainly much cheaper to replace.
The main problem with power satellites is that you have to beam the power back down to earth. This does not present a terrorism hazard - the most popular schemes use the receiving antenna array to control the phase of the beams produced from various parts of the satellite. Without this control, the satellite would emit in all directions, resulting in harmless power levels reaching the ground. This control takes a big array, and the resulting beam is targetted _at_ the array. So to fry a city, the terrorist group would first have to put an appropriate antenna on the roof of every home... Maybe offer to clean their eavestroughs while there...
The real problems with beaming power back are that your emitters have to be a fixed weight no matter what - which puts a lower limit on the mass of the satellite - and that you need substantial levels of microwave energy coming down on your receiving array (i.e. more than 1 kW/m^2). Any bird flying through this will get warm, and anybody camping out underneath will get warm also. They won't be flash-fried - but they will receive the equivalent of standing under light several times brighter than the sun.
Lastly, as this represents energy imported to earth, power satellites will contribute to the heat pollution problem. OTOH, the same applies to any other power source that releases energy that would otherwise not be released on earth.
The final way to (efficiently!) use solar power on earth is to grow crops and ferment them to produce methanol. This gives you a flammable liquid that can be used as fuel, mulch to supply the hydrocarbon/chemical industries, and takes carbon dioxide out of the atmosphere as fast as fuel-burning puts it into it.
The problem is that the crop method takes up a fair amount of space, and (like other forms of solare power) is limited by the amount of sunlight falling on to the collectors/fields.
Um, no, there aren't.
Not the required amount of pressure.
Even the sun doesn't produce enough pressure to physically force the nuclei together. The core of the sun is electron-degenerate - you have electrons in a strictly organized set of energy levels, with nuclei zipping around like gas particles within it.
Most of the nuclei, even in the sun's core, don't have enough energy to overcome the electrostatic repulsion between nuclei. A very few of them do - because the temperature of an object is only the _average_ temperature of the particles within it. It is very rare for a particle to gain this much energy, and so fusion reactions are extremely infrequent for a given particle, but the particles aren't going anywhere, and there are a lot of them. Thus, the sun produces a substantial amount of energy due to fusion.
On earth, there are very strong limits to the pressures we can produce stably. A steady-state system would therefore have to jack up the temperature to increase the reaction rate in order to generate more energy than it expends. You could argue that a steady-state system isn't necessary; well and good. However, when you produce a region of extreme pressure (as with inertial-confinement fusion), it doesn't last very long. You jack up the reaction rate, but the short confinement time brings your total power output back down again.
This covers all but two proposed methods of fusion.
One other method proposed is muon-catalyzed fusion. Click on "user info" above to find a message with more details about it. It uses a very sneaky method to remove most of the cost of bringing nuclei close enough together to fuse. However, the energy cost of the muons required for catalysis is far more than the energy produced by the catalyzed fusion.
The last method proposed of producing fusion is the "cold" fusion that caused such a media commotion recently.
People would very much like this style of fusion to work.
People claimed that they had gotten this style of fusion to work.
A number of countries invested and continue to invest substantial research energies into duplicating the original experiments that supposedly reliably produced energy by fusion by this method.
To date, they have failed. This, coupled with a number of extremely suspicious points in the presentations by the original discoverers, strongly suggests that the original researchers were either mistaken or actively defrauding the rest of the scientific community. Read the book entitled "Yes, we have no neutrons" for an in-depth discussion of this. Authour name escapes me at the moment.
Summary: Cold fusion appears to be bogus, despite much effort to duplicate it.
So, what method are you proposing to create the pressures necessary to produce fusion at room temperature?
Heat pollution is indeed a problem, both at the reactor and in the cities where the power is used (most power eventually ends up being converted to heat). However, there is a safe place to dump the heat if it becomes a sufficiently bad problem - the sky.
Space has a black-body temperature of about 3 kelvin. Use heat pumps to concentrate your heat in a working fluid. Run the fluid through pipes. Put the pipes in mirrored channels, open at the top. Most of your heat goes into space, by blackbody emission from the pipes.
The problem with this is the rate at which heat is dumped. This is proportional to the fourth power of the temperature of your working fluid (and hence your pipes). The bad news is, you need a very hot fluid. The good news is that, because of the fourth order relation, your fluid doesn't have to be _insanely_ hot. The heat pump that performs the concentration will produce heat itself - you're working against entropy when concentrating heat like that. However, if all components are efficient enough, you should be able to dump this heat out with the rest.
This requires power. I am blithely assuming that if you have enough power plants that heat pollution is a problem big enough to do something about, you have enough power to run the heat pumps.
This doesn't work on cloudy days. I am blithely assuming that you have a heat sink large enough to store heat in until the next clear day. This would probably be a moderate-sized body of water.
Assuming that you can afford a one-hectare area per GW of heat for your emitter array (100m squared, or _roughly_ 100 yards by 100 yards), and that one quarter of this is actual emitting area (area of the pipes), your working fluid would have to be at a temperature of about 1600 kelvin, or about 1330 degrees centigrade. This is readily buildable. The only concern is the heat pump, which is working against a very strong temperature difference. Scale up the area - to, say, 1km by 1km (about 0.6 miles by 0.6 miles), and you need a temperature of about 515 kelvin, or about 242 degrees centigrade. Building the heat engine for this is much easier, though more of the resulting radiation will be absorbed by the atmosphere in that frequency band. Maintaining the array is more expensive in this scenario, too.
In conclusion, I think that heat pollution is a solvable problem. It just requires enough of an investment that we aren't likely to build heat-dumping plants until they become absolutely necessary.
Horrid typo.
That should read, "fusion produces _less_ waste".
plants, would the waste production be higher or lower?
Sure.
In fact, it's easy to show that fission produces _less_ waste - the question is how much less.
DISCLAIMER: I AM MAKING LIBERAL USE OF "FERMI" ESTIMATES. NUMBERS QUOTED SHOULD BE WITHIN AN ORDER OF MAGNITUDE OF REALITY, BUT THAT'S ABOUT IT. THIS IS NOT A DETAILED ESTIMATE.
Both fission and fusion produce secondary waste, in the form of the reactor housing, which becomes radioactive. The housing won't last forever - pieces wear out, as with any device. Let's assume as a rough approximation that the entire thing has to be replaced once every 20 years, and that it is all filed as low-level radioactive waste.
Let's assume that we have a moderate-sized reactor core - a 10m cube (30 feet for the American audience). This isn't a solid structure; in a fission plant, it's a framework holding reactor rods, and either a containment vessel (American reactors) or a network of pipes (Canaadian reactors), with ample amounts of radiation shielding. Let's say it's equivalent to a shell 2m thick. This gives us a volume of about (10^3 - 6^3) = around 800 cubic metres. Assuming a density of around 5 (metal, concrete, and lead), we get about 4000 tonnes of waste every 20 years, or an average of 200 tonnes of waste per year (1 tonne = 1000 kg = about 2200 pounds, so about the same as a US ton, before you object to my spelling).
Now, a fission reactor also produces a fair bit of high-level waste in the form of spent fuel rods. This represents extra waste that a fusion reactor doesn't produce. We have to see how much it is compared to the (roughly) 200 tonnes produced on average per year by our hypothetical reactor.
Let's assume that a reactor with a core of the quoted size produces about 500 MW of power. Let's assume that this represents about 2% of the fuel in the reactor (typical of fission reactors that don't use reprocessed fuels). Let's assume that this fuel converts mass to energy at an efficiency of about 0.1% (typical for fission reactions, IIRC). 500 MW over one year gives us about 1.5e16 joules of energy produced per year. At 100% efficiency, this would correspond to about 0.17 kg of fuel. We're operating at 0.1% efficiency, which means about 170 kg of fuel burned. This represents about 2% of the total amount of fuel, which means about 8.5 tonnes of fuel burned in the reactor per year. This fuel is filed as high-level waste when its finished burning.
So, we find that, unless I'm off by _more_ than a factor of 10 in computing the _relative_ amounts of each type of waste, and assuming that you consider high-level and low-level waste equally bad, both fusion and fission produce about the same amount of waste, most of which is just parts of the reactor that wear out and have to be stored safely.
CAVEATS:
A few notes about fission, that might be confusing people:
A conventional fission reactor has two fuel sources; U235, which is present from the start, and Pu239, which is bred from U238 within the reactor (any reactor containing U238 does this whether you want it to or not). Both of these fission to produce many light byproducts. These byproducts occasionally absorb neutrons, and almost never give them out. After you've left a rod in the reactor for a while, enough of these byproducts accumulate that reactions are not self-sustaining within the fuel rod anymore. It is absorbing neutrons without contributing much energy. This is after somewhere around 2% of its mass has been fissioned. At this point, you can either throw it away or send it to a reprocessing plant, which strips out the light, absorbing components so that you can use the rod again. Nowadays, we throw the rods away and lose 98% of its mass.
Reprocessing isn't done any more because it involves running what amounts to high-level waste through a chemical processing plant. Despite precautions, this resulted in unacceptably high exposure to radiation for people working in the plant. Neutron radiation from the waste also transmuted anything that the waste was in contact with (including whatever you dissolve it in for processing, and so forth), producing a lot of low-level waste. The result of all of this is that reprocessing fuel rods safely is a hassle of monumental proportions. There's enough uranium lying around that we can afford to throw away 98% of it, so we do.
> vacuum, then not as much heat is required to
> sustain the reaction as stated in the article.
Um, not quite.
It takes a very hefty amount of energy to heat the plasma to the required temperatures - about 1.0e8 degrees centigrade. At those temperatures, plasma cools quite quickly by radiative emission, regardless of whether it's in vacuum or not. This is one of the reasons that the reaction only lasts nanoseconds.
> Only about 100,000C is needed before the atoms
> can become ionized and begin to transfer energy.
Um, not quite.
The plasma doesn't just have to be plasma - it has to be plasma of high enough energy that the nuclei can overcome their mutual repulsion. Hence, the 1.0e8 degrees figure for most terrestrial reactors (stars use lower temperatures, higher densities, and longer confinement time).
> At this stage things get a little fuzzy, because
> some of the normal laws of physics don't work
> because the atoms are in a plasma state.
Actually, they work just fine. You're just dealing with a conducting gas.
> The good news is the are electrically charged
> and can be controlled by magnets(as mentioned in
> the article). This should help keep the reaction
> in one spot.
Correct. This is the principle behind magnetic confinement fusion.
Other approaches exist. Both the approach described in the article and laser fusion rely on inertial confinement instead - the particle is suddenly compressed, and by the time it disperses the reaction has finished.
For more information on fusion and many other interesting things, browse your local university's bookstore for physics texts. My personal favourite is "Physics for Scientists and Engineers, Extended Version", bt Paul A. Tipler.
> produce fusion is at extremely high
> temperatures. No one has suggested a meachanism
> (that works!) that would allow otherwise,
> possibly no one ever will.
Actually, muon-catalyzed fusion works at much lower temperatures (something like 800 degrees in the test rigs I heard about).
This is completely different from "cold" fusion. Completely different principles, well-understood, and well-tested.
The idea is that if you displace an electron in a hydrogen _molecule_ with a muon (its heavier cousin), the molecule will suddenly become much smaller, because the wavelength of a muon is much smaller than the wavelength of an electron. The hydrogen nuclei are now close enough to have a significant probability of fusion by tunnelling.
The problem with this is that muons are energy-expensive to produce, and don't stick around for very long (they decay into an electron, a muon neutrino, and an electron antineutrino). The energy released by the fusion reactions that the muons catalyze has so far been far, far less than the energy required to produce the muons in the first place.
> article seems like the only way to go right now.
> Once they lick the symmetricity problem -
> possibly by setting up some positive feedback in
> the plasma to keep the reaction shape spherical
> in the crucial nanoseconds - there should be
> more progress.
It turns out that symmetry is extremely difficult to achieve.
This approach to fusion has been tried before - there is a whole class of fusion reactor that tries to induce fusion by running strong electric currents through a plasma (which is what this is - you just get your plasma by zapping a pellet).
From the first approximation, this looks very good - the magnetic fields created by the current flow act to compress the plasma, bringing it closer to satisfying the Lawson criterion. The problem is that variations in the current and in the density of the plasma/pellet cause assymetries, which tend to magnify themselves (the electric and magnetic effects of turbulence cause more turbulence). Because of the way that plasma behaves, there is no magical way to elminate this.
All fusion approaches encounter this problem. One approach to solutions - the one that it looks like they're using here - is brute force. It takes more time for turbulence to muck up a larger volume of plasma, which means that a larger device can come closer to satisfying the Lawson criterion. Also, dumping more energy into the plasma can help; fusion in conventional reactors occurs only among the hottest particles at the tail-end of the temperature distribution. Greater average energy means more particles of sufficient energy to cause fusion.
A second approach is to modify the design of the reactor to either reduce turbulence or to be less sensitive to turbulence. For magnetic confinement reactors, a lot of study has gone into different field geometries that stifle certain types of turbulence. For electrical discharge reactors - like the one described in the article - different electrode geometries have been tried. However, in the case of electrical reactors, this hasn't been enough to match the capabilities of magnetic confinement and inertial confinement (laser) fusion.
A hybrid reactor was recently built that used an electrical discharge to produce a hot plasma that was used as an X-ray source, which in turn was used to heat a fuel pellet in a chamber that reflected X-rays. This proved to be a very promising approach; for details, check back issues of (I think) Scientific American. This does not appear to be the approach described in this article, however.
Laser fusion's main problems aren't turbulence, which is why I haven't mentioned it much in this response.
> on mars. How else would they get all the power?
> AFAIK mars has quite a few clouds and storms
> regularly, so that would rule out solar as a
> sole source of power
AFAIK, weather on mars is mostly clear. That's why we get such nice views of the surface via telescopes. The atmosphere is much thinner than Earth's, which means that dense clouds would be difficult to produce. Also, Mars's ambient temperature is below the freezing point of water. Dust storms happen, but aren't terribly frequent in most locations AFAIK.
Solar power would work fine most of the time, and power storage in batteries or fuel cells would be adequate during the occasional dust storms.
> that get a *very* high portion of their yield
> from fusion, as opposed to fission, there
> probably wouldn't be that much radiation to
> worry about in the first place.
It turns out that there would be.
Fusion reactions, like most other nuclear reactions, produce copious quantities of neutron radiation. This would make anything nearby (in this case, the surrounding rock) radioactive by transmutation.
The main benefit of fusion power isn't that it "doesn't produce radiation" - it does. It's that fuel is virtually unlimited. You also don't produce any _primary_ waste - spent fuel rods and the like. Just secondary waste (reactor structures made radioactive by transmutation).
Do you see GeForce boards on the shelves?
Do you see Playstation 2s on the shelves?
They've been conversation topics for months, but all either has now is alpha test hardware. It becomes difficult to see what point you are trying to make, given that.
Not if it can only transform 36 million polys per second, it can't (sustained transformation figure from an older slashdot article).
Based on all of the other numbers in that article, I suspect that they dropped a decimal point in the "75 million polys rendered" figure. That, or they're talking about flat-shaded untextured untransformed polygons.
"Consumer level" is correct. High-end graphics workstations have been doing this for several years; in fact, the entire OpenGL pipeline has been in hardware for quite a while. Check out 3dlab's high-end boards for examples, or take a look at their competitors. These tend to be 64-bit PCI boards in the $2,000-$3,000 range.
The consumer graphics manufacturers have been making noise about using geometry processing for a while now, but have only recently gotten around to it. In that market, yes it could be called revolutionary (in that it substantially changes game design).
3dfx is rolling out another chip, as people have been talking about for a while. It is rumoured to be at 0.22 micron too, and will have hardware geometry processing.
S3 already rolled out a new chip - at 0.18 micron. It too has four texel engines and hardware geometry processing.
IMO, the S3 chip is actually the one to worry about. Architecture may or may not be great, but at 0.18 micron it may outperform nVidia and 3dfx's offerings just on linewidth. ATI did something similar when it rolled out the Rage 128, if you recall.
What I'm waiting for is the release of the GeForce or (insert name of 3dfx's offering here) at 0.18 micron. However, I'll probably be waiting a while.
The Voodoo 4 will be coming out around Christmas, and it will have hardware geometry as well. Rumour has it that it too will be at 0.22 micron instead of 0.18. I don't remember the name of the chipset off-hand.
S3 has also rolled out a new chip, with four pipelines and hardware geometry, at 0.18 micron. Check Sharkey Extreme for details.
Also, I've heard some reports that the PlayStation II will beat the living daylights out of PIII's loaded with then recent and most modern 3d accels. Even with this kind of chip, and most likely other chips to follow from nVidia's competitors, does this still hold true? Will the PlayStation II live up to the hype?
No, but it won't sink either. See my previous response on this subject (check my user info to find the post).
The Playstation II has a modified R10000 processor with very hefty floating point extensions - it won't have much of a problem doing geometry transformations. IMO, it will probably be about on par with the graphics cards floating around at the time of its release. It won't leave them in the dust, but neither will it be left in the dust.
OTOH, a friend in the gaming industry says that the Playstation II has architectural problems that might degrade performance (low system bus bandwidth, among other things). We'll see what happens when it ships.
"hands-on tests" means prototype hardware already available, so this should be out fairly soon. None too soon, as S3 pulled ATI's trick and is coming out early with a chip at 0.18.