The report published a few days ago (also posted to slashdot) on orbital solar power proposed a ~10 MW power plant in space as a demonstration of feasibility, later to be expanded to ~10 TW. So, yes, they do plan to scale it up by 10,000x. Right now the problem is getting 'teh shit into space.
They fell into a pool of water. The drop was just to give it long enough to let surface tension form a sphere from the original blob of molten lead.
Not to say you're wrong about it being small shot, I wasn't particularly attentive in high school. But they definitely always called it the cannon tower, and I'm pretty sure I was told it was for cannon balls.
Neat! This is the same method that was used to make cannonballs during the US Civil War.
I can't find any references to cannon ball manufacture on Wikipedia, but my high school had a cannon forming tower (it was originally a civil war arsenal).
Outside of that, the more techniques the merrier! I'm somewhat curious how they create a PN junction out of a homogenous liquid of silicon, but I suppose that can be done afterwards. I'd also be a bit curious if it's single crystalline. I very much doubt it, as there is no seed crystal to nucleate on, so there should be a lot of independent surface nucleation sites (IAAMS).
I personally find the idea of a space based laser awesome (IANAPacifist), but the implication was that in order to create the microwave beam you need a very particular shape of antenna. I was left with the impression that in order to increase how focused it was, you'd need to physically change the shape of the emitter, which reasonably well jibes with my understanding of E&M.
That sounds very hard to me.
Maybe a jigawatt laser beam attached elsewhere? I wonder if it'd be possible to make that powerful of a laser beam without blowing out the laser crystal -- seems all that you might need is something very pure... but who knows?
They addressed this in the report. According to the paper, microwave radiation from that far away would be too low intensity to blow up shit. The intensity they anticipated was comparable to that from microwave oven leakage.
Basically, they were saying that microwave radiation at 1/6th the intensity of the sun on the earth at noon would be all the intensity needed to rectify out gigawatts of power due to the monochromaticity of the beam and extremely good efficiency rectifying a single frequency of electromagnetic radiation of human-scale wavelengths. Building a precise resonant antenna to rectify microwave radiation is a lot easier than building a solar cell.
This is just from the article though, I don't claim to have done the calculations myself.
In reading the report (which was fascinating), I kept wondering about something. It's not a huge issue, but when you focus the sunlight down to the 500+ suns that are going to be desirable for the solar panels, how do you cool the panels in space? I feel like these panels would melt, and radiative cooling only goes so far.
In any event, it sounds neat, and that this is the only unaddressed technical issue I see with the proposal is probably a very good sign (I am a mere materials scientist with a degree in physics on the side, so issues with launching and building aren't my forte' =)
talking about a unit of power (energy per unit time) per unit time is meaningless. it's like asking how many watts your computer uses per second, your computer doesn't use watts, it uses energy. the power is the amount of energy per unit time.
what's important is that you can, instead of paying 8 cents or whatnot for electricity per unit energy, you can instead pay an upfront cost of $1/watt and produce your own energy. the important question is the device lifetime
other people have done this calculation, I'll assume they did it correctly, and that their answer of about half a year of full utilization to break even and start producing "free energy" is fine.
Right now, the grid acts as a nearly perfect battery by distributing power around as needed. During the daytime, electricity use is far higher than during the night, so solar panels are really very nice in terms of when the provide power. The solar panels installed in houses would decrease daytime load on power plants, resulting in better efficiency throughout the system. Think of it as the solar panels working towards supplementing the grid with enough extra power to handle air conditioning and other day-time power use without running power plants at 100% of their rated output.
It lists the efficiency. The watts per square meter will depend on the amount of sunlight in your location. 13% is mid-range, people have made up to 60%, but those are state-of-the-art and expensive.
In system->administration->screens and graphics, there is an option for rotating or flipping your screen (or screens, it lets you set up multiple monitors there too -- finally something similar to the windows display dialog).
I suspect that flipping means left-to-right though instead of top-to-bottom. If you think a lot of people might want it it'd be worth filing a feature request or bug report.
Well, I've got one that's missing. I filed a bug about the calc solver crashing OOo every time I use it. I've not even seen a response, much less accepting that there's a bug, and trying to find a fix.
The solver was the only thing in OOo that really impressed me. It was actually easier to use than Matlab for the problem I was solving.
I dunno, I don't mind things being broken temporarily, but it's been broken in the developer releases too. If it wasn't fixed by the time they wanted to release it, I wish they had just reverted to an earlier version of the solver code.
Do most people not roll their own kernel for a new install?
I compiled my own kernels regularly with Debian, and am quite competent at it. However, I stopped about a year and a half ago when I found that the Dapper Drake Ubuntu default kernels were simply better than what I could do on my own without a lot of wrangling (and even then, the only benefit was having fewer modules and no initrd image -- unlikely to be more than an aesthetic improvement).
They do all the patching to get the proprietary drivers so that I don't have to. Plus, it gets automatically updated without me having to spend 20 minutes recompiling the system when there's a security update.
So... I would think that if I, as someone who has compiled scores of kernels and have a "pretty good" idea of how it works and what needs to be in it, no longer find it to offer an advantage over the precompiled default kernels, I doubt very much that a beginner should even need to go near it.
I'm not really sure if your post is implying that a PhD might teach poorly, but I had a PhD in physics as my high school physics teacher. I had never had another person with a PhD as a teacher before, and he was by *far* the best teacher I ever had. Pretty much exclusively due to his existence, I am now a fairly well published researcher getting my PhD in Materials Engineering from MIT. Granted, he's special in a lot of ways because he was willing to work as a teacher in an inner city high school despite being somewhat overqualified by our typical standards. However, I suspect that anyone who is able to get a PhD understands and is excited enough about their field so much that if they try at all they'll be able to generate many future PhDs who would never have thought about doing something more difficult than IT. Being Weird to an employer definitely does not imply that you are a bad physicist!
I plan to teach someday too, but currently I'm enjoying the heck out of myself doing actual research, so it'll probably be a few decades. =)
What struck me as most interesting is the "or teaching" part. The people who major in pure science, who can't find or don't want jobs in science, can't just immediately move into finance as I see many of my friends doing. Instead, they have to do *something*, and if that something involves providing a larger pool of qualified high school science teachers, then society wins. It's sort of like military service, they commit to either teaching, or actually doing work in the field, but either way, they *can't* flip burgers or go into finance without repaying all that tuition.
Back in 2003, my ethernet card (under debian) would *only* work if I was also playing music. Granted, that was because my ethernet card was broken and didn't properly send interrupts (so the sound card was sending them, and the ethernet driver was being activated when it noticed that it had an interrupt too), but it was still pretty awesome. Perhaps Vista has a similar problem... =)
The article isn't extremely clear, but these aren't made of just silicon. If that were the case, this would be much more awesome (and is why I was initially extremely excited). They split the light into high, medium, and low energy so that the most appropriate semiconductor can be used to absorb it. Silicon might be used for the medium or low energy light, because it has a band gap of 1.3eV. Something else would have to be used for the other two, running into the problem of gallium being unfortunately rare for widespread use.
Just like how they "deliberately omitted" the 5,673 other schools not in the list of 23 they didn't omit?
Seems strange to assume that the RIAA is scared just because they picked other targets. They're choices in every other instance seem completely random, why would this one be any different?
This is like saying that MIT is "conspicuously absent" and claiming it is because MIT refused to log traffic for the RIAA on their internal network because of the sheer technical insanity of the request. Correlation != causation.
Actually, typically people suggest research bias in the other direction. The US Department of Agriculture (I believe) put out the report that initially said the value of 1.3 for the energy content. The DOA is heavily invested in corn, so many people suggest that they would prefer a result that suggests increasing the amount of farming being done.
And yes, the necessity of liquid fuels is important. However, as you mention, it is not currently possible to derive most of the rest of the energy from stationary sources. For instance, it's extremely difficult to produce fertilizers from sunlight, or fuel a tractor with batteries charged from the grid. As long as you accept that fertilizers are made from oil, the suggestion put out is that you're better off just putting the oil in your car instead of losing 25% of it's energy content to convert it into ethanol.
However, if cellulosic ethanol has 13 times more energy that costs to produce it (which sounds extremely unlikely to me -- if anything, cellulose should be *harder* to convert than sugar, yes?), then you're all set, because you can use that ethanol to power your tractor, you can probably convert it into fertilizer (though I'm not sure how -- but at least it's got a carbon-carbon bond already), and then you've got a self-sustaining system.
Alternatively, yes, if you did things like use a solar collector to heat up your distillation system, you could easily save a lot of energy. If I recall correctly, distilling from 90% to 100% takes about 10% of the *total* energy put into making ethanol. Presumably, most of that energy is used in the form of heat, so it would be a particularly convenient place to increase that 1.3 (or whatever) value by 0.1.
I mean, it's bad news if the research is fraudulent and money that could be spent on better technologies is being diverted into shareholder pockets.
But in order to make one really big breakthrough, an awful lot of money has to go into studying things that do not work and never will work. It looks like in this case, it's privately funded, which seems to frequently mean that it's more likely to be a commercial success (although still far from guaranteed... I'm comparing this to DARPA's perpetual motion machines).
It comes from a selection of five papers from the late nineties which did the calculation in a number of ways. Generally, they attempt to account for the entire manufacturing process, from energy in oil used in fertilizers to fuel for farm equipment, to transport of the ethanol or corn, to the refineries that distill out all the water. I do not believe they go so far as to account for feeding the farmer, but I honestly suspect that is a very minor correction, as much as I like farmers.
However, there is a fairly well known outlier which claimed to do a better job of accounting for processing costs. Pimentel and Patzek attributed what they claim are more accurate inputs to the agriculture, transport, industrial, and distribution components of the manufacturing process, giving the also oft-quoted value of around 25% energy *loss*. Ordinarily, people would probably dismiss that one given the seemingly overwhelming amount of contrary evidence, but Pimentel and Patzek are very well-respected scientists. It's difficult for me, as an energy researcher, to know who to believe. I suspect it's nigh impossible for people who only study this passingly.
Personally, I'm inclined to believe that even if Pimentel et al are wrong, 1.3 is just way, way too low to be reasonable. Improvements to technology (as this plant represents), are the only way that ethanol can ever be practical. We'll see soon enough if it's as good as they claim.
All I ask for is a way to send emails in the form of binary text files with no formatting attached to a blank email and still have Microsoft Word users able to open it.
Piezos are unlikely to work well by themselves because of their inability to source much current.
That said, devices that harvest energy from vibrations have been COMMERCIAL for a long time now. I've seen the durn things in person. They're incredible. Ferro Solutions has been doing this for at least a decade, and their products are commercially viable already. Oy.
DMF is poisonous. Granted, so is gasoline, but that's a big reason why we're focusing on ethanol. Much of the reason we aren't using ethanol based fuel cells is that the byproducts include acetaldehyde... which I don't believe is even much worse than DMF.
Slashdot Mirror
The report published a few days ago (also posted to slashdot) on orbital solar power proposed a ~10 MW power plant in space as a demonstration of feasibility, later to be expanded to ~10 TW. So, yes, they do plan to scale it up by 10,000x. Right now the problem is getting 'teh shit into space.
They fell into a pool of water. The drop was just to give it long enough to let surface tension form a sphere from the original blob of molten lead. Not to say you're wrong about it being small shot, I wasn't particularly attentive in high school. But they definitely always called it the cannon tower, and I'm pretty sure I was told it was for cannon balls.
Neat! This is the same method that was used to make cannonballs during the US Civil War.
I can't find any references to cannon ball manufacture on Wikipedia, but my high school had a cannon forming tower (it was originally a civil war arsenal).
Outside of that, the more techniques the merrier! I'm somewhat curious how they create a PN junction out of a homogenous liquid of silicon, but I suppose that can be done afterwards. I'd also be a bit curious if it's single crystalline. I very much doubt it, as there is no seed crystal to nucleate on, so there should be a lot of independent surface nucleation sites (IAAMS).
I personally find the idea of a space based laser awesome (IANAPacifist), but the implication was that in order to create the microwave beam you need a very particular shape of antenna. I was left with the impression that in order to increase how focused it was, you'd need to physically change the shape of the emitter, which reasonably well jibes with my understanding of E&M.
That sounds very hard to me.
Maybe a jigawatt laser beam attached elsewhere? I wonder if it'd be possible to make that powerful of a laser beam without blowing out the laser crystal -- seems all that you might need is something very pure... but who knows?
They addressed this in the report. According to the paper, microwave radiation from that far away would be too low intensity to blow up shit. The intensity they anticipated was comparable to that from microwave oven leakage.
Basically, they were saying that microwave radiation at 1/6th the intensity of the sun on the earth at noon would be all the intensity needed to rectify out gigawatts of power due to the monochromaticity of the beam and extremely good efficiency rectifying a single frequency of electromagnetic radiation of human-scale wavelengths. Building a precise resonant antenna to rectify microwave radiation is a lot easier than building a solar cell.
This is just from the article though, I don't claim to have done the calculations myself.
Hiya!
In reading the report (which was fascinating), I kept wondering about something. It's not a huge issue, but when you focus the sunlight down to the 500+ suns that are going to be desirable for the solar panels, how do you cool the panels in space? I feel like these panels would melt, and radiative cooling only goes so far.
In any event, it sounds neat, and that this is the only unaddressed technical issue I see with the proposal is probably a very good sign (I am a mere materials scientist with a degree in physics on the side, so issues with launching and building aren't my forte' =)
Who knows? I'm looking at it sideways...
talking about a unit of power (energy per unit time) per unit time is meaningless. it's like asking how many watts your computer uses per second, your computer doesn't use watts, it uses energy. the power is the amount of energy per unit time.
what's important is that you can, instead of paying 8 cents or whatnot for electricity per unit energy, you can instead pay an upfront cost of $1/watt and produce your own energy. the important question is the device lifetime
other people have done this calculation, I'll assume they did it correctly, and that their answer of about half a year of full utilization to break even and start producing "free energy" is fine.
Right now, the grid acts as a nearly perfect battery by distributing power around as needed. During the daytime, electricity use is far higher than during the night, so solar panels are really very nice in terms of when the provide power. The solar panels installed in houses would decrease daytime load on power plants, resulting in better efficiency throughout the system. Think of it as the solar panels working towards supplementing the grid with enough extra power to handle air conditioning and other day-time power use without running power plants at 100% of their rated output.
It lists the efficiency. The watts per square meter will depend on the amount of sunlight in your location. 13% is mid-range, people have made up to 60%, but those are state-of-the-art and expensive.
In system->administration->screens and graphics, there is an option for rotating or flipping your screen (or screens, it lets you set up multiple monitors there too -- finally something similar to the windows display dialog).
I suspect that flipping means left-to-right though instead of top-to-bottom. If you think a lot of people might want it it'd be worth filing a feature request or bug report.
Well, I've got one that's missing. I filed a bug about the calc solver crashing OOo every time I use it. I've not even seen a response, much less accepting that there's a bug, and trying to find a fix.
The solver was the only thing in OOo that really impressed me. It was actually easier to use than Matlab for the problem I was solving.
I dunno, I don't mind things being broken temporarily, but it's been broken in the developer releases too. If it wasn't fixed by the time they wanted to release it, I wish they had just reverted to an earlier version of the solver code.
Do most people not roll their own kernel for a new install?
I compiled my own kernels regularly with Debian, and am quite competent at it. However, I stopped about a year and a half ago when I found that the Dapper Drake Ubuntu default kernels were simply better than what I could do on my own without a lot of wrangling (and even then, the only benefit was having fewer modules and no initrd image -- unlikely to be more than an aesthetic improvement).
They do all the patching to get the proprietary drivers so that I don't have to. Plus, it gets automatically updated without me having to spend 20 minutes recompiling the system when there's a security update.
So... I would think that if I, as someone who has compiled scores of kernels and have a "pretty good" idea of how it works and what needs to be in it, no longer find it to offer an advantage over the precompiled default kernels, I doubt very much that a beginner should even need to go near it.
I'm not really sure if your post is implying that a PhD might teach poorly, but I had a PhD in physics as my high school physics teacher. I had never had another person with a PhD as a teacher before, and he was by *far* the best teacher I ever had. Pretty much exclusively due to his existence, I am now a fairly well published researcher getting my PhD in Materials Engineering from MIT. Granted, he's special in a lot of ways because he was willing to work as a teacher in an inner city high school despite being somewhat overqualified by our typical standards. However, I suspect that anyone who is able to get a PhD understands and is excited enough about their field so much that if they try at all they'll be able to generate many future PhDs who would never have thought about doing something more difficult than IT. Being Weird to an employer definitely does not imply that you are a bad physicist!
I plan to teach someday too, but currently I'm enjoying the heck out of myself doing actual research, so it'll probably be a few decades. =)
What struck me as most interesting is the "or teaching" part. The people who major in pure science, who can't find or don't want jobs in science, can't just immediately move into finance as I see many of my friends doing. Instead, they have to do *something*, and if that something involves providing a larger pool of qualified high school science teachers, then society wins. It's sort of like military service, they commit to either teaching, or actually doing work in the field, but either way, they *can't* flip burgers or go into finance without repaying all that tuition.
Back in 2003, my ethernet card (under debian) would *only* work if I was also playing music. Granted, that was because my ethernet card was broken and didn't properly send interrupts (so the sound card was sending them, and the ethernet driver was being activated when it noticed that it had an interrupt too), but it was still pretty awesome. Perhaps Vista has a similar problem... =)
The article isn't extremely clear, but these aren't made of just silicon. If that were the case, this would be much more awesome (and is why I was initially extremely excited). They split the light into high, medium, and low energy so that the most appropriate semiconductor can be used to absorb it. Silicon might be used for the medium or low energy light, because it has a band gap of 1.3eV. Something else would have to be used for the other two, running into the problem of gallium being unfortunately rare for widespread use.
Just like how they "deliberately omitted" the 5,673 other schools not in the list of 23 they didn't omit?
Seems strange to assume that the RIAA is scared just because they picked other targets. They're choices in every other instance seem completely random, why would this one be any different?
This is like saying that MIT is "conspicuously absent" and claiming it is because MIT refused to log traffic for the RIAA on their internal network because of the sheer technical insanity of the request. Correlation != causation.
Right, because we'd know if they could. And China would certainly let everyone know.
Actually, typically people suggest research bias in the other direction. The US Department of Agriculture (I believe) put out the report that initially said the value of 1.3 for the energy content. The DOA is heavily invested in corn, so many people suggest that they would prefer a result that suggests increasing the amount of farming being done.
And yes, the necessity of liquid fuels is important. However, as you mention, it is not currently possible to derive most of the rest of the energy from stationary sources. For instance, it's extremely difficult to produce fertilizers from sunlight, or fuel a tractor with batteries charged from the grid. As long as you accept that fertilizers are made from oil, the suggestion put out is that you're better off just putting the oil in your car instead of losing 25% of it's energy content to convert it into ethanol.
However, if cellulosic ethanol has 13 times more energy that costs to produce it (which sounds extremely unlikely to me -- if anything, cellulose should be *harder* to convert than sugar, yes?), then you're all set, because you can use that ethanol to power your tractor, you can probably convert it into fertilizer (though I'm not sure how -- but at least it's got a carbon-carbon bond already), and then you've got a self-sustaining system.
Alternatively, yes, if you did things like use a solar collector to heat up your distillation system, you could easily save a lot of energy. If I recall correctly, distilling from 90% to 100% takes about 10% of the *total* energy put into making ethanol. Presumably, most of that energy is used in the form of heat, so it would be a particularly convenient place to increase that 1.3 (or whatever) value by 0.1.
I mean, it's bad news if the research is fraudulent and money that could be spent on better technologies is being diverted into shareholder pockets.
But in order to make one really big breakthrough, an awful lot of money has to go into studying things that do not work and never will work. It looks like in this case, it's privately funded, which seems to frequently mean that it's more likely to be a commercial success (although still far from guaranteed... I'm comparing this to DARPA's perpetual motion machines).
It comes from a selection of five papers from the late nineties which did the calculation in a number of ways. Generally, they attempt to account for the entire manufacturing process, from energy in oil used in fertilizers to fuel for farm equipment, to transport of the ethanol or corn, to the refineries that distill out all the water. I do not believe they go so far as to account for feeding the farmer, but I honestly suspect that is a very minor correction, as much as I like farmers.
However, there is a fairly well known outlier which claimed to do a better job of accounting for processing costs. Pimentel and Patzek attributed what they claim are more accurate inputs to the agriculture, transport, industrial, and distribution components of the manufacturing process, giving the also oft-quoted value of around 25% energy *loss*. Ordinarily, people would probably dismiss that one given the seemingly overwhelming amount of contrary evidence, but Pimentel and Patzek are very well-respected scientists. It's difficult for me, as an energy researcher, to know who to believe. I suspect it's nigh impossible for people who only study this passingly.
Personally, I'm inclined to believe that even if Pimentel et al are wrong, 1.3 is just way, way too low to be reasonable. Improvements to technology (as this plant represents), are the only way that ethanol can ever be practical. We'll see soon enough if it's as good as they claim.
http://www.news.cornell.edu/stories/July05/ethanol .toocostly.ssl.html has a summary of the debate.
All I ask for is a way to send emails in the form of binary text files with no formatting attached to a blank email and still have Microsoft Word users able to open it.
Piezos are unlikely to work well by themselves because of their inability to source much current.
That said, devices that harvest energy from vibrations have been COMMERCIAL for a long time now. I've seen the durn things in person. They're incredible. Ferro Solutions has been doing this for at least a decade, and their products are commercially viable already. Oy.
DMF is poisonous. Granted, so is gasoline, but that's a big reason why we're focusing on ethanol. Much of the reason we aren't using ethanol based fuel cells is that the byproducts include acetaldehyde... which I don't believe is even much worse than DMF.