As for thin-film solar, stop dreaming. It ain't gonna happen. 1.367kW/m^2 is the maximum theoretical amount of energy we can get from the sun. Do the math on area required where we can power the 3.848 trillion kilowatt-hours the US consumed last year.
You end up with an area comparable to the size of the cities these people live in, and far, far less than the area of the farmland used to feed them.
Sounds reasonable to me.
What people tend not to realize is that 1 kW/m^2 is a _huge_ energy density.
Yes, because the US banning domestic enrichment for nuclear power has kept other countries from going nuclear (*cough* India, Pakistan, China, Israel, North Korea *cough*).
I didn't say it made sense; just that that was why the US wasn't doing it.
Up here in Canada, we're probably abstaining to avoid ticking off the Americans.
And handling concerns? How is keeping spent fuel in a pools, caskets, and eventual transport to remote Nevada mountains not already a handling concern?
Having fuel routinely moving around is more of a handling concern than keeping it in one place. Similarly, a reprocessing plant has far more things that can go wrong with it than a storage facility, so there's greater risk of contamination.
Whether the reduction in handling risks is worth throwing out about 98% of the usable fuel is another issue. The US has one opinion, France has another. Me, I think we'll end up with thin-film solar everywhere when it gets cheap enough.
Wrong way around. My point was the main disincentive for breeder reactors was not present in Canada, as Candus already produce Pu.
Any reactor burning uranium produces Pu in large quantities. The risk is not that they produce Pu, but that you have un-poisoned Pu around if you're reprocessing spent fuel.
As neither the US nor Canada reprocesses spent fuel, neither is exposed to that proliferation risk.
a quarter of a phone booth's worth of waste in volume
How much energy in burning Libraries of Congress could a phone booth of nuclear waste produce?
If we assume that only the books are burning, and that they weigh a couple of pounds each (say 1 kg), and that they give off the same energy from combustion that an equivalent weight of carbon would (very rough approximation), we can estimate the BLoC energy unit as about:
115M books * 1 kg/book * 390 kJ/mol CO2 / 0.012 mol C/kg...or on the order of 4 petajoules.
Let's assume the phone booth contains about 2 cubic metres of nuclear waste. Let's assume that it has a density of about 10 g/cm^3, as it's oxides, and that virtually all of this represents the weight of the heavy nuclei. We'll take a value of 10 MeV as the total decay energy of each heavy metal nucleus as it traverses the decay chain down to lead (or some other stable isotope, if it starts off lighter than lead, though most of the fuel rod will still be U238). We'll assume an atomic weight of 250 AMU for each nucleus, to make the math easier. As 1 AMU is approximately equivalent to 1 GeV (i.e. mass of a proton or neutron), we have a rest energy of each nucleus of 250 GeV, meaning 1/25000 of its rest mass is converted to released energy.
The phone booth contains 2 m^3 * 10000 kg/m^3 = 20000 kg of material. This has a rest energy of about 1.8e+21 J, meaning we get about 70 petajoules out if we wait long enough for all of its constituent elements to decay.
So, a phone booth full of nuclear waste could produce about 18 BLoCs worth of energy.
In practice, you'll only get around 1% of this out in any reasonable timeframe (short-lived isotopes, vs. the U238 that you'll have to wait a few billion years for unless you stick it back in a reactor).
Of course, it doesn't matter here in Canada, as we use Candu reactors.
Um, apples and oranges, here. Fuel enrichment != fuel reprocessing. Our (Canada's) reactors don't need enriched uranium to run, but our spent fuel contains just as much potentially useful material as the spent fuel from American reactors.
How about we refine the waste, make it further useful, and save on the amount of waste we create?
The US decided not to do this, as it presented a proliferation risk (the spent fuel contains significant amounts of plutonium, which was deemed a security problem after reprocessing stripped out the decay products poisoning it).
My understanding is that there was a fuel reprocessing plant online in the US at one point, and I believe the French nuclear power program does reprocess spent fuel. If you're doing fuel reprocessing, you can turn U238 and also thorium 232 into materials usable as fuels (plutonium 239 and, through relatively favourable neutron absorptions and beta decays, uranium 235, respectively).
If the waster is radioactive, it is inherently releasing energy. I have never understood why no one has tried to take advantage of this with some kind of "dirty" reactor.
The problem is that the fuel has been "poisoned" by decay products from previous reactions. Enough of these absorb neutrons that you can't sustain a critical fission reaction, and so you're left with sub-critical decay. This gives off energy, but far, far more slowly than a nuclear plant's active fuel bundles do. So you can't put them in a conventional reactor, and you can't get useful amounts of heat off them outside of one.
There are some types of reactor - actinide-burning fast-breeders - that have less trouble with these decay products than conventional slow-neutron reactors. These are widely viewed as one method of disposing of or at least reducing the amount of spent fuel waste. You can also chemically reprocess the fuel to remove the decay products (which are then disposed of as waste, but the majority of your "spent" fuel is reused). Neither of these solutions is allowed in the US, due to proliferation risks and handling concerns.
To allow particle accelerators you need to expand your parameters a bit to include natural objects accelerated by man.
How much tinkering needs to be done for an object to cease to be "natural"? We could accelerate ionized molecules with a molecular weight under about 100-200 using RHIC with no problems. That covers a wide variety of compounds that don't occur significantly in nature.
The fastest "man-made" objects - I would hazard a guess at probes sent from Earth to other planets. Voyager 1 travels at roughly 17.4 km/sec
I believe compressed gas guns can get faster muzzle velocities than that, per my previous posts. At minimum they can get in the same range.
This will all be put to shame by the Mini-Magnetospheric Plasma Propulsion system (M2P2)
Which would in turn be put to shame if anyone builds a maser-powered "starwhisp", or Orion, or any of a variety of other craft. If it hasn't been built yet, it doesn't count:).
A related question, how does this (and the speed of orbital rockets) compare to the fastest man-made object (whatever that may be)?
Particle accelerators accelerate anything from electrons and protons to ions close enough to C that the difference is academic.
For macroscopic objects, I believe compressed-gas guns used for simulating micrometeorite strikes and for producing shockwaves to study things like the metallic hydrogen phase transition accelerate projectiles to tens of km/sec, or larger than but of the same magnitude as orbital velocities.
Various other types of cannon (the so-called "ram accelerator", used to simulate scramjets, and various flavours of electromagnetic cannon) can also reach projectile speeds in the "greater than but still comparable to orbital" range.
Ever tried to modulate a high energy laser? I do not mean semiconductor toys.
You do realize that bar semiconductor lasers in the kilowatt range are available now, right?
You do realize that most forms of diode-_pumped_ laser can be modulated nearly as fast as you can modulate the pump diodes, right?
You do realize that GHz+ modulation is done on "do not look into fiber with remaining eye" class lasers for data transmission now, right?
You do realize that even a modulated laser pointer is likely to deliver more power to Earth than any kind of radio dish small enough for a space probe to carry, right? RTEGs and solar panels don't give you a multi-kW output.
In summary, you can make lasers competitive for high-speed interplanetary data transmission very easily.
I always wondered why they would want to use the visible spectrum...
We *CAN* make Laser-Radio waves! They go through atmosphere and trees and buildings....
The benefits of (visible) light are that it has much higher data capacity per unit time, and that it can be collimated very well with a much smaller dish (due to the factor of about 1000 difference in wavelength). The wavelength difference improves your channel gain for a given dish size by about 6 orders of magnitude. You want a filled aperture for this, not an array, as you're listening to a faint signal, here.
The drawbacks to visible light are that it's much more easily blocked than radio (as you point out), and that it costs about a thousand times as much energy per photon to produce, even assuming perfect production efficiency (as the photons are a thousand times as energetic). This means that, for scenarios where noise intrinsic to the beam (sqrt(photon count)) dominates over background noise, it'll be a thousand times as energy-expensive to send the same amount of data.
In summary, each method has its place. For interplanetary communications, using visible wavelengths is _really_ handy. For interstellar, if we ever get around to launching probes that go out that far, it's pretty much the _only_ practical approach (and even then, requires a light source strong enough that it has to be in the Sol system; too heavy to send one with the probe).
As I said, the mining and purification are the biggest difficulties with nuclear technology. A massive industrial base was required because of the time crunch imposed by the war. If we're talking smaller quantities and research performed over a few decades, then it is perfectly feasible for fissable materials to be produced in a smaller industrial base.
Where you get the idea that you'd need smaller quantities for a NERVA, I'm not sure.
Where you get the idea that the nuclear bomb would have been produced over the course of a century instead of a decade (or shorter) I'm not sure. There is no peacetime use for one, so Atlantis (the nation that allegedly dropped it) would already have been at war.
There's actually an article somewhere on the Internet that explains how to separate small quantities of U235 and U238 with a metal bucket and some muscle power.
The yield will be low enough that you'd need thousands of times more _ore_, which removes the advantage of this operation.
You need a large (for NERVA) or at least modest-sized (for one (1) bomb) isotope separation plant, which means you need the manufacturing technology to build such a plant, which means you need an industrialized nation with associated medium- to high-tech infrastructure.
Why? How many traces currently exist of the V2 program? We certainly have the cultural aspect of the 50's sci-fi rockets resembling the V2. But could you produce a trace of a single V2 today? How about traces of a Saturn V? Lunar landers? Mercury Rockets?
The mines that supplied the materials, the railways that transported ore to refineries, the power plants that ran machine shops, the specialized tools used and the tools used to make those, are for the most part still here, and will be detectable by archaeologists for many centuries (and will still be found here and there for millenia).
It takes an industrial base to run a space program, or a technological war machine. Industrial infrastructures don't just vanish without a trace.
The first part is true, but the part about a massive infrastructure is not. Mining Uranium and separating the U235 from U238 are the most difficult parts. Neither one requires a very large infrastructure if you're looking for small quantities. In fact, the Manhattan Project was really one small research facility combined with many great minds. If there hadn't been such a time crunch, it is feasible that the study of nuclear power could have been carried out over a generation or two in smaller labs.
I seem to recall fairly well supported claims that a) the nuclear weapons program took a large fraction of the industrial base of the US to realize, and b) the Nazis didn't succeed in producing nuclear weapons largely because they didn't have a large enough industrual base.
Either way, if you're proposing NERVA-style engines for aircraft, as you'd mentioned in your previous post, and a space program, you're going to need a _large_ industrial base to support it. This would leave traces.
I can believe the ancients might have had hot air balloons. Much more, without traces of the support infrastructure, stretches credibility.
Nuclear thermal is limited by the need to keep the exhaust temperature to something the core materials can withstand. Ion drives have no such limitation, and can in principle achieve exhaust velocities close to C (though in practice you'd never want to build a drive that did this). This gives ion drives far higher Isp.
You can in principle build a magnetically confined nuclear thermal drive that holds a uranium plasma in a magnetic field and reaches very high temperatures, but this turns out to have many practical problems (not the least of which is a fairly large minimum size), so don't expect it to be done soon.
Nuclear-electric drives are the way of the future for craft outside the inner solar system. They use a small fission plant to generate electricity, which then drives an ion drive or hall effect thruster or plasma drive or what-have-you.
A ratio by itself is not a fractal. That ratio can play into a lot of fractals and is seen in many places, but by itself is just a number.
You can probably consider the shape made by continuing to divide a rectangle using the golden ratio a fractal, as it's definitely self-similar and based on an affine transformation.
You'd have to do a bit of sleight-of-hand defining the boundary for it to actually meet the definition, though. If you just count the lines added at each iteration, it has a fractal dimension of 1.
I wrote my first fractal in 8-bit color, sucker ! On a MacII no less.
As long as we're having a pissing contest, I have code for Mandelbrot rendering on a TI-81 calculator kicking around;). Took a couple of hours to render a 96x64 image taken to 32 iterations, if memory serves.
I've been meaning to dust off that calculator for quite a while, now. Main problem is that it eats batteries for breakfast.
Wavelet transforms involve expressing the input data as the sum of wavelet basis functions (much as a Fourier transform uses sine/cosine waves).
Fractal compression involves looking for self-similar features in the image itself, removing this redundancy by expressing it as a series of affine transformations, or something similar.
Frequency- and wavelet-transforms can make the search for self-similar structures easier, but they represent fundamentally different approaches (the best you can do to draw an analogy is to consider fractals to be a different type of parameterized basis function that you're doing a transform with).
I guess no one ever learned how to make a fractal equation that looked like a given image on the fly.
I may be mistaken, but I think somebody did, and called it JPEG.
JPEG and fractal compression are completely different, I'm afraid.
JPEG transforms blocks of the image from the spatial domain to the frequency domain, and keeps only the strongest spatial frequencies. To look at it another way, it tries to express each block as the sum of various functions that look like bands or ripple patterns.
Fractal compression tries to find similarities between different parts of the image, and to express the image as a bunch of these similarity relations (affine transforms, or different types of mapping).
There's more detail for each type of algorithm, but that's the basic approach for each. Some versions of fractal compression to a frequency transform of blocks during the compression stage, but that's just to make it easier to compare blocks to each other when sifting possibilities, as opposed to part of the mechanism of compression itself.
No, it won't ignite. I've seen people try to do stuff like this before - not only in open areas, but in an enclosed chamber even with an actual spark gap. They did this on mythbusters, as just one example:)
All I can say is, relevant reference say that's a great way to autoDarwinate.
I'm using octane as a reasonable model for gasoline, though from a composition standpoint, heptane might be a better model (it's actually a witches' brew of hydrocarbons that amounts to isomorphs of octane and heptane with a few more volatile chemicals thrown in for grins).
The MSDS for octane says its flammable concentration range is 1.0% to 6.5% v/v. This corresponds to partial pressures of around 8 to 50 mmHg. Concidentally, this corresponds to the vapour pressure of octane over a temperature range of 15-50 degrees C. So, you have a very substantial ignition hazard over any open container of gasoline any time except the middle of winter.
Remind me in 6 months, and I'll be happy to do a controlled test outside, and film it.
For the comparable hydrogen situation, take a *pressurized* hydrogen container (we're not talking about a balloon of hydrogen here), and open it near that candle.
As others have pointed out, that'll get you what amounts to a hydrogen torch, and not an explosion. As _I've_ pointed out, you're going to get low pressure or nothing, because in addition to blowing out the candle, a high pressure hydrogen stream will make the broken canister take off like a rocket. The regulator knocks pressure down to something reasonable right at the tank nozzle, so the only case in which you'll get a high-pressure release is if someone rams a vehicle into the canister. This is a Bad Thing with any fuel (and arguably less catastrophic with a hydrogen tank than with a propane tank, due to buoyancy).
I have played extensively with small quantities of hydrogen...
under hundreds of atmospheres of pressure?
If you have hydrogen coming _out_ of the tank at hundreds of atmospheres, you have a lot bigger problems than a fire hazard. Compressed gas canisters are treated with respect for a reason.
Any leak that would occur under normal conditions would result in leaked hydrogen being at low pressure.
"It's a lot easier to get gasoline vapours... pooling where they can ignite"
When you get home today (if you're not already home), pour a basin full of gasoline. Put any sort of simple container you want around it. If you can get it to explode without extensive effort or a situation unlikely to occur in the real world, I'll bake you a dozen cookies and mail them.
You're going to be waiting about six months, because it's winter here, and I'm not trying this indoors.
Find a nice, open patio that's been receiving direct summer sunlight for a morning. This represents hot asphalt or concrete.
Put a lit candle or two near the edge of the experiment area. This represents the spark or other ignition source causing the accident.
Take a cup or two of gasoline and pour it thinly around the middle of the accident region. This represents a puddle of gasoline from the hypothetical leaky hose or what-have-you.
Stand well back.
Nice, thin film on a nice warm surface gives you lots of vapour.
Lots of vapour plus ignition source -> fireball on your patio.
You can do the same thing with a propane torch simulating a propane leak if you're feeling suicidal. This works best in _cooler_ weather.
You _cannot_ do the same thing with hydrogen, outdoors, at least. That is the point of this discussion.
Gasoline explosions in non-controlled circumstances are incredibly difficult to occur. Hydrogen explosions are not, by any stretch. That's the only thing that matters.
It's a lot easier to get gasoline vapours or leaked _propane_ pooling where it can ignite. Hydrogen goes straight up, _fast_, if the leak is outdoors. If the leak is indoors, it still dissipates quickly - most substances are permeable to hydrogen, to the point where it's hard to contain for any length of time when you _want_ to.
Realistically, a hydrogen leak would produce a torch at the leak point, and not much else. A propane leak would produce a ground-level sea of fire.
Checking my references, it looks like the frequency-doubled DUV sources were things like argon ion and copper lasers moved from visible to DUV, more like 5 years ago than 2-3. Live and learn.
Know your product before calling "bullshit", please.
I thought that pretty much all green pointers available today were still frequency-doubled DPSS, and that's why they're comparably so expensive?
My understanding is that pointers based on green laser diodes have been on the market for a year or two now, but I'd have to doublecheck to find out which models.
I was actually hunting for a blue pointer for a friend for whom a green pointer would be passe, but those still aren't obtainable (laboratory-grade blue diode lasers run around $2k or so, if memory serves).
Slashdot Mirror
As for thin-film solar, stop dreaming. It ain't gonna happen. 1.367kW/m^2 is the maximum theoretical amount of energy we can get from the sun. Do the math on area required where we can power the 3.848 trillion kilowatt-hours the US consumed last year.
You end up with an area comparable to the size of the cities these people live in, and far, far less than the area of the farmland used to feed them.
Sounds reasonable to me.
What people tend not to realize is that 1 kW/m^2 is a _huge_ energy density.
Yes, because the US banning domestic enrichment for nuclear power has kept other countries from going nuclear (*cough* India, Pakistan, China, Israel, North Korea *cough*).
I didn't say it made sense; just that that was why the US wasn't doing it.
Up here in Canada, we're probably abstaining to avoid ticking off the Americans.
And handling concerns? How is keeping spent fuel in a pools, caskets, and eventual transport to remote Nevada mountains not already a handling concern?
Having fuel routinely moving around is more of a handling concern than keeping it in one place. Similarly, a reprocessing plant has far more things that can go wrong with it than a storage facility, so there's greater risk of contamination.
Whether the reduction in handling risks is worth throwing out about 98% of the usable fuel is another issue. The US has one opinion, France has another. Me, I think we'll end up with thin-film solar everywhere when it gets cheap enough.
Wrong way around. My point was the main disincentive for breeder reactors was not present in Canada, as Candus already produce Pu.
Any reactor burning uranium produces Pu in large quantities. The risk is not that they produce Pu, but that you have un-poisoned Pu around if you're reprocessing spent fuel.
As neither the US nor Canada reprocesses spent fuel, neither is exposed to that proliferation risk.
a quarter of a phone booth's worth of waste in volume
...or on the order of 4 petajoules.
How much energy in burning Libraries of Congress could a phone booth of nuclear waste produce?
If we assume that only the books are burning, and that they weigh a couple of pounds each (say 1 kg), and that they give off the same energy from combustion that an equivalent weight of carbon would (very rough approximation), we can estimate the BLoC energy unit as about:
115M books * 1 kg/book * 390 kJ/mol CO2 / 0.012 mol C/kg
Let's assume the phone booth contains about 2 cubic metres of nuclear waste. Let's assume that it has a density of about 10 g/cm^3, as it's oxides, and that virtually all of this represents the weight of the heavy nuclei. We'll take a value of 10 MeV as the total decay energy of each heavy metal nucleus as it traverses the decay chain down to lead (or some other stable isotope, if it starts off lighter than lead, though most of the fuel rod will still be U238). We'll assume an atomic weight of 250 AMU for each nucleus, to make the math easier. As 1 AMU is approximately equivalent to 1 GeV (i.e. mass of a proton or neutron), we have a rest energy of each nucleus of 250 GeV, meaning 1/25000 of its rest mass is converted to released energy.
The phone booth contains 2 m^3 * 10000 kg/m^3 = 20000 kg of material. This has a rest energy of about 1.8e+21 J, meaning we get about 70 petajoules out if we wait long enough for all of its constituent elements to decay.
So, a phone booth full of nuclear waste could produce about 18 BLoCs worth of energy.
In practice, you'll only get around 1% of this out in any reasonable timeframe (short-lived isotopes, vs. the U238 that you'll have to wait a few billion years for unless you stick it back in a reactor).
Of course, it doesn't matter here in Canada, as we use Candu reactors.
Um, apples and oranges, here. Fuel enrichment != fuel reprocessing. Our (Canada's) reactors don't need enriched uranium to run, but our spent fuel contains just as much potentially useful material as the spent fuel from American reactors.
How about we refine the waste, make it further useful, and save on the amount of waste we create?
The US decided not to do this, as it presented a proliferation risk (the spent fuel contains significant amounts of plutonium, which was deemed a security problem after reprocessing stripped out the decay products poisoning it).
My understanding is that there was a fuel reprocessing plant online in the US at one point, and I believe the French nuclear power program does reprocess spent fuel. If you're doing fuel reprocessing, you can turn U238 and also thorium 232 into materials usable as fuels (plutonium 239 and, through relatively favourable neutron absorptions and beta decays, uranium 235, respectively).
If the waster is radioactive, it is inherently releasing energy. I have never understood why no one has tried to take advantage of this with some kind of "dirty" reactor.
The problem is that the fuel has been "poisoned" by decay products from previous reactions. Enough of these absorb neutrons that you can't sustain a critical fission reaction, and so you're left with sub-critical decay. This gives off energy, but far, far more slowly than a nuclear plant's active fuel bundles do. So you can't put them in a conventional reactor, and you can't get useful amounts of heat off them outside of one.
There are some types of reactor - actinide-burning fast-breeders - that have less trouble with these decay products than conventional slow-neutron reactors. These are widely viewed as one method of disposing of or at least reducing the amount of spent fuel waste. You can also chemically reprocess the fuel to remove the decay products (which are then disposed of as waste, but the majority of your "spent" fuel is reused). Neither of these solutions is allowed in the US, due to proliferation risks and handling concerns.
To allow particle accelerators you need to expand your parameters a bit to include natural objects accelerated by man.
:).
How much tinkering needs to be done for an object to cease to be "natural"? We could accelerate ionized molecules with a molecular weight under about 100-200 using RHIC with no problems. That covers a wide variety of compounds that don't occur significantly in nature.
The fastest "man-made" objects - I would hazard a guess at probes sent from Earth to other planets. Voyager 1 travels at roughly 17.4 km/sec
I believe compressed gas guns can get faster muzzle velocities than that, per my previous posts. At minimum they can get in the same range.
This will all be put to shame by the Mini-Magnetospheric Plasma Propulsion system (M2P2)
Which would in turn be put to shame if anyone builds a maser-powered "starwhisp", or Orion, or any of a variety of other craft. If it hasn't been built yet, it doesn't count
A related question, how does this (and the speed of orbital rockets) compare to the fastest man-made object (whatever that may be)?
Particle accelerators accelerate anything from electrons and protons to ions close enough to C that the difference is academic.
For macroscopic objects, I believe compressed-gas guns used for simulating micrometeorite strikes and for producing shockwaves to study things like the metallic hydrogen phase transition accelerate projectiles to tens of km/sec, or larger than but of the same magnitude as orbital velocities.
Various other types of cannon (the so-called "ram accelerator", used to simulate scramjets, and various flavours of electromagnetic cannon) can also reach projectile speeds in the "greater than but still comparable to orbital" range.
Ever tried to modulate a high energy laser? I do not mean semiconductor toys.
You do realize that bar semiconductor lasers in the kilowatt range are available now, right?
You do realize that most forms of diode-_pumped_ laser can be modulated nearly as fast as you can modulate the pump diodes, right?
You do realize that GHz+ modulation is done on "do not look into fiber with remaining eye" class lasers for data transmission now, right?
You do realize that even a modulated laser pointer is likely to deliver more power to Earth than any kind of radio dish small enough for a space probe to carry, right? RTEGs and solar panels don't give you a multi-kW output.
In summary, you can make lasers competitive for high-speed interplanetary data transmission very easily.
I always wondered why they would want to use the visible spectrum...
We *CAN* make Laser-Radio waves! They go through atmosphere and trees and buildings....
The benefits of (visible) light are that it has much higher data capacity per unit time, and that it can be collimated very well with a much smaller dish (due to the factor of about 1000 difference in wavelength). The wavelength difference improves your channel gain for a given dish size by about 6 orders of magnitude. You want a filled aperture for this, not an array, as you're listening to a faint signal, here.
The drawbacks to visible light are that it's much more easily blocked than radio (as you point out), and that it costs about a thousand times as much energy per photon to produce, even assuming perfect production efficiency (as the photons are a thousand times as energetic). This means that, for scenarios where noise intrinsic to the beam (sqrt(photon count)) dominates over background noise, it'll be a thousand times as energy-expensive to send the same amount of data.
In summary, each method has its place. For interplanetary communications, using visible wavelengths is _really_ handy. For interstellar, if we ever get around to launching probes that go out that far, it's pretty much the _only_ practical approach (and even then, requires a light source strong enough that it has to be in the Sol system; too heavy to send one with the probe).
As I said, the mining and purification are the biggest difficulties with nuclear technology. A massive industrial base was required because of the time crunch imposed by the war. If we're talking smaller quantities and research performed over a few decades, then it is perfectly feasible for fissable materials to be produced in a smaller industrial base.
Where you get the idea that you'd need smaller quantities for a NERVA, I'm not sure.
Where you get the idea that the nuclear bomb would have been produced over the course of a century instead of a decade (or shorter) I'm not sure. There is no peacetime use for one, so Atlantis (the nation that allegedly dropped it) would already have been at war.
There's actually an article somewhere on the Internet that explains how to separate small quantities of U235 and U238 with a metal bucket and some muscle power.
The yield will be low enough that you'd need thousands of times more _ore_, which removes the advantage of this operation.
You need a large (for NERVA) or at least modest-sized (for one (1) bomb) isotope separation plant, which means you need the manufacturing technology to build such a plant, which means you need an industrialized nation with associated medium- to high-tech infrastructure.
Why? How many traces currently exist of the V2 program? We certainly have the cultural aspect of the 50's sci-fi rockets resembling the V2. But could you produce a trace of a single V2 today? How about traces of a Saturn V? Lunar landers? Mercury Rockets?
The mines that supplied the materials, the railways that transported ore to refineries, the power plants that ran machine shops, the specialized tools used and the tools used to make those, are for the most part still here, and will be detectable by archaeologists for many centuries (and will still be found here and there for millenia).
It takes an industrial base to run a space program, or a technological war machine. Industrial infrastructures don't just vanish without a trace.
The first part is true, but the part about a massive infrastructure is not. Mining Uranium and separating the U235 from U238 are the most difficult parts. Neither one requires a very large infrastructure if you're looking for small quantities. In fact, the Manhattan Project was really one small research facility combined with many great minds. If there hadn't been such a time crunch, it is feasible that the study of nuclear power could have been carried out over a generation or two in smaller labs.
I seem to recall fairly well supported claims that a) the nuclear weapons program took a large fraction of the industrial base of the US to realize, and b) the Nazis didn't succeed in producing nuclear weapons largely because they didn't have a large enough industrual base.
Either way, if you're proposing NERVA-style engines for aircraft, as you'd mentioned in your previous post, and a space program, you're going to need a _large_ industrial base to support it. This would leave traces.
I can believe the ancients might have had hot air balloons. Much more, without traces of the support infrastructure, stretches credibility.
More specifically, JPEGs are based on DCTs (Discrete Cosine Transforms), i.e. coefficients of various frequencies of cosine functions.
The DCT can be thought of as a Fourier transform that makes additional assumptions (input function is real and symmetrical).
Nuclear thermal is limited by the need to keep the exhaust temperature to something the core materials can withstand. Ion drives have no such limitation, and can in principle achieve exhaust velocities close to C (though in practice you'd never want to build a drive that did this). This gives ion drives far higher Isp.
You can in principle build a magnetically confined nuclear thermal drive that holds a uranium plasma in a magnetic field and reaches very high temperatures, but this turns out to have many practical problems (not the least of which is a fairly large minimum size), so don't expect it to be done soon.
Nuclear-electric drives are the way of the future for craft outside the inner solar system. They use a small fission plant to generate electricity, which then drives an ion drive or hall effect thruster or plasma drive or what-have-you.
A ratio by itself is not a fractal. That ratio can play into a lot of fractals and is seen in many places, but by itself is just a number.
You can probably consider the shape made by continuing to divide a rectangle using the golden ratio a fractal, as it's definitely self-similar and based on an affine transformation.
You'd have to do a bit of sleight-of-hand defining the boundary for it to actually meet the definition, though. If you just count the lines added at each iteration, it has a fractal dimension of 1.
I wrote my first fractal in 8-bit color, sucker ! On a MacII no less.
;). Took a couple of hours to render a 96x64 image taken to 32 iterations, if memory serves.
As long as we're having a pissing contest, I have code for Mandelbrot rendering on a TI-81 calculator kicking around
I've been meaning to dust off that calculator for quite a while, now. Main problem is that it eats batteries for breakfast.
It is called wavelets
Actually, no.
Wavelet transforms involve expressing the input data as the sum of wavelet basis functions (much as a Fourier transform uses sine/cosine waves).
Fractal compression involves looking for self-similar features in the image itself, removing this redundancy by expressing it as a series of affine transformations, or something similar.
Frequency- and wavelet-transforms can make the search for self-similar structures easier, but they represent fundamentally different approaches (the best you can do to draw an analogy is to consider fractals to be a different type of parameterized basis function that you're doing a transform with).
I guess no one ever learned how to make a fractal equation that looked like a given image on the fly.
I may be mistaken, but I think somebody did, and called it JPEG.
JPEG and fractal compression are completely different, I'm afraid.
JPEG transforms blocks of the image from the spatial domain to the frequency domain, and keeps only the strongest spatial frequencies. To look at it another way, it tries to express each block as the sum of various functions that look like bands or ripple patterns.
Fractal compression tries to find similarities between different parts of the image, and to express the image as a bunch of these similarity relations (affine transforms, or different types of mapping).
There's more detail for each type of algorithm, but that's the basic approach for each. Some versions of fractal compression to a frequency transform of blocks during the compression stage, but that's just to make it easier to compare blocks to each other when sifting possibilities, as opposed to part of the mechanism of compression itself.
No, it won't ignite. I've seen people try to do stuff like this before - not only in open areas, but in an enclosed chamber even with an actual spark gap. They did this on mythbusters, as just one example :)
All I can say is, relevant reference say that's a great way to autoDarwinate.
I'm using octane as a reasonable model for gasoline, though from a composition standpoint, heptane might be a better model (it's actually a witches' brew of hydrocarbons that amounts to isomorphs of octane and heptane with a few more volatile chemicals thrown in for grins).
The MSDS for octane says its flammable concentration range is 1.0% to 6.5% v/v. This corresponds to partial pressures of around 8 to 50 mmHg. Concidentally, this corresponds to the vapour pressure of octane over a temperature range of 15-50 degrees C. So, you have a very substantial ignition hazard over any open container of gasoline any time except the middle of winter.
Remind me in 6 months, and I'll be happy to do a controlled test outside, and film it.
For the comparable hydrogen situation, take a *pressurized* hydrogen container (we're not talking about a balloon of hydrogen here), and open it near that candle.
As others have pointed out, that'll get you what amounts to a hydrogen torch, and not an explosion. As _I've_ pointed out, you're going to get low pressure or nothing, because in addition to blowing out the candle, a high pressure hydrogen stream will make the broken canister take off like a rocket. The regulator knocks pressure down to something reasonable right at the tank nozzle, so the only case in which you'll get a high-pressure release is if someone rams a vehicle into the canister. This is a Bad Thing with any fuel (and arguably less catastrophic with a hydrogen tank than with a propane tank, due to buoyancy).
I have played extensively with small quantities of hydrogen ...
under hundreds of atmospheres of pressure?
If you have hydrogen coming _out_ of the tank at hundreds of atmospheres, you have a lot bigger problems than a fire hazard. Compressed gas canisters are treated with respect for a reason.
Any leak that would occur under normal conditions would result in leaked hydrogen being at low pressure.
"It's a lot easier to get gasoline vapours ... pooling where they can ignite"
When you get home today (if you're not already home), pour a basin full of gasoline. Put any sort of simple container you want around it. If you can get it to explode without extensive effort or a situation unlikely to occur in the real world, I'll bake you a dozen cookies and mail them.
You're going to be waiting about six months, because it's winter here, and I'm not trying this indoors.
Find a nice, open patio that's been receiving direct summer sunlight for a morning. This represents hot asphalt or concrete.
Put a lit candle or two near the edge of the experiment area. This represents the spark or other ignition source causing the accident.
Take a cup or two of gasoline and pour it thinly around the middle of the accident region. This represents a puddle of gasoline from the hypothetical leaky hose or what-have-you.
Stand well back.
Nice, thin film on a nice warm surface gives you lots of vapour.
Lots of vapour plus ignition source -> fireball on your patio.
You can do the same thing with a propane torch simulating a propane leak if you're feeling suicidal. This works best in _cooler_ weather.
You _cannot_ do the same thing with hydrogen, outdoors, at least. That is the point of this discussion.
Gasoline explosions in non-controlled circumstances are incredibly difficult to occur. Hydrogen explosions are not, by any stretch. That's the only thing that matters.
It's a lot easier to get gasoline vapours or leaked _propane_ pooling where it can ignite. Hydrogen goes straight up, _fast_, if the leak is outdoors. If the leak is indoors, it still dissipates quickly - most substances are permeable to hydrogen, to the point where it's hard to contain for any length of time when you _want_ to.
Realistically, a hydrogen leak would produce a torch at the leak point, and not much else. A propane leak would produce a ground-level sea of fire.
Cymer employee says, "BULLSH*T!"
Well, let's see...
Check.
Checking my references, it looks like the frequency-doubled DUV sources were things like argon ion and copper lasers moved from visible to DUV, more like 5 years ago than 2-3. Live and learn.
Know your product before calling "bullshit", please.
I thought that pretty much all green pointers available today were still frequency-doubled DPSS, and that's why they're comparably so expensive?
My understanding is that pointers based on green laser diodes have been on the market for a year or two now, but I'd have to doublecheck to find out which models.
I was actually hunting for a blue pointer for a friend for whom a green pointer would be passe, but those still aren't obtainable (laboratory-grade blue diode lasers run around $2k or so, if memory serves).